DOES EVOLUTION EVEN HAVE A MECHANISM?
[Paper presented 23 April 2002 at the American Museum of Natural History]
By William A. Dembski

Talk delivered at the American Museum of Natural History, 23 April 2002 at 
a discussion titled "Evolution or Intelligent Design?" The 
participants included ID proponents William A. Dembski and Michael J. Behe 
as well as evolutionists Kenneth R. Miller and Robert T. Pennock. Eugenie 
C. Scott moderated the discussion.


Evolutionary biology teaches that all biological complexity is the result 
of material mechanisms. These include principally the Darwinian mechanism 
of natural selection and random variation, but also include other 
mechanisms (symbiosis, gene transfer, genetic drift, the action of 
regulatory genes in development, self-organizational processes, etc.). 
These mechanisms are just that: mindless material mechanisms that do what 
they do irrespective of intelligence. To be sure, mechanisms can be 
programmed by an intelligence. But any such intelligent programming of 
evolutionary mechanisms is not properly part of evolutionary biology.

Intelligent design, by contrast, teaches that biological complexity is not 
exclusively the result of material mechanisms but also requires 
intelligence, where the intelligence in question is not reducible to such 
mechanisms. The central issue, therefore, is not the relatedness of all 
organisms, or what typically is called common descent. Indeed, intelligent 
design is perfectly compatible with common descent. Rather, the central 
issue is how biological complexity emerged and whether intelligence played 
a pivotal role in its emergence.

Suppose, therefore, for the sake of argument that intelligence--one 
irreducible to material mechanisms--actually did play a decisive role in 
the emergence of life's complexity and diversity. How could we know it? To 
answer this question, let's run a thought experiment. Imagine that Alice is 
sending Bob encrypted messages over a communication channel and that Eve is 
eavesdropping. For simplicity let's assume all the signals are bit strings. 
How could Eve know that Alice is not merely sending Bob random coin flips 
but meaningful messages?

To answer this question, Eve will require two things: First, the bit 
strings sent across the communication channel need to be reasonably 
long--in other words, they need to be complex. If not, chance can readily 
account for them. Just as there's no way to reconstruct a piece of music 
given just one note, so there is no way to preclude chance for a bit string 
that consists of only a few bits. For instance, there are only eight 
strings consisting of three bits, and chance readily accounts for any of them.

There's a second requirement for Eve to know that Alice is not sending Bob 
random gibberish: Eve needs to observe a suitable pattern in the signal 
Alice sends Bob. Even if the signal is complex, it may exhibit no pattern 
characteristic of intelligence. Flip a coin enough times, and you'll 
observe a complex sequence of coin tosses. But that sequence will exhibit 
no pattern characteristic of intelligence. For cryptanalysts like Eve, 
observing a pattern suitable for identifying intelligence amounts to 
finding a cryptographic key that deciphers the message. Patterns suitable 
for identifying intelligence I call specifications.

In sum, Eve requires both complexity and specification to infer 
intelligence in the signals Alice is sending to Bob. This combination of 
complexity and specification, or specified complexity as I call it, is the 
basis for design inferences across numerous special sciences, including 
archaeology, cryptography, forensics, and the Search for Extraterrestrial 
Intelligence (SETI). I detail this in my book The Design Inference, a 
peer-reviewed statistical monograph that appeared with Cambridge University 
Press in 1998.

So, what's all the fuss about specified complexity? The actual term 
specified complexity is not original with me. It first occurs in the 
origin-of-life literature, where Leslie Orgel used it to describe what he 
regards as the essence of life. That was thirty years ago. More recently, 
in 1999, surveying the state of origin-of-life research, Paul Davies 
remarked: "Living organisms are mysterious not for their complexity per se, 
but for their tightly specified complexity" (The Fifth Miracle, p. 112).

Orgel and Davies used specified complexity loosely. In my own research I've 
formalized it as a statistical criterion for identifying the effects of 
intelligence. For identifying the effects of animal, human, and 
extraterrestrial intelligence the criterion works just fine. Yet when 
anyone attempts to apply the criterion to biological systems, all hell 
breaks loose. Let's consider why.

Evolutionary biologists claim to have demonstrated that design is 
superfluous for understanding biological complexity. The only way to 
actually demonstrate this, however, is to exhibit material mechanisms that 
account for the various forms of biological complexity out there. Now, if 
for every instance of biological complexity some mechanism could readily be 
produced that accounts for it, intelligent design would drop out of 
scientific discussion. Occam's razor, by proscribing superfluous causes, 
would in this instance finish off intelligent design quite nicely.

But that hasn't happened. Why not? The reason is that there are plenty of 
complex biological systems for which no biologist has a clue how they 
emerged. I'm not talking about handwaving just-so stories. Biologists have 
plenty of those. I'm talking about detailed testable accounts of how such 
systems could have emerged. To see what's at stake, consider how biologists 
propose to explain the emergence of the bacterial flagellum, a molecular 
machine that has become the mascot of the intelligent design movement.

Howard Berg at Harvard calls the bacterial flagellum the most efficient 
machine in the universe. The flagellum is a nano-engineered outboard rotary 
motor on the backs of certain bacteria. It spins at tens of thousands of 
rpm, can change direction in a quarter turn, and propels a bacterium 
through its watery environment. According to evolutionary biology it had to 
emerge via some material mechanism. Fine, but how?

The usual story is that the flagellum is composed of parts that previously 
were targeted for different uses and that natural selection then co-opted 
to form a flagellum. This seems reasonable until we try to fill in the 
details. The only well-documented examples that we have of successful 
co-optation come from human engineering. For instance, an electrical 
engineer might co-opt components from a microwave oven, a radio, and a 
computer screen to form a working television. But in that case, we have an 
intelligent agent who knows all about electrical gadgets and about 
televisions in particular.

But natural selection doesn't know a thing about bacterial flagella. So how 
is natural selection going to take extant protein parts and co-opt them to 
form a flagellum? The problem is that natural selection can only select for 
pre-existing function. It can, for instance, select for larger finch beaks 
when the available nuts are harder to open. Here the finch beak is already 
in place and natural selection merely enhances its present functionality. 
Natural selection might even adapt a pre-existing structure to a new 
function; for example, it might start with finch beaks adapted to opening 
nuts and end with beaks adapted to eating insects.

But for co-optation to result in a structure like the bacterial flagellum, 
we are not talking about enhancing the function of an existing structure or 
reassigning an existing structure to a different function, but reassigning 
multiple structures previously targeted for different functions to a novel 
structure exhibiting a novel function. The bacterial flagellum requires 
around fifty proteins for its assembly and structure. All these proteins 
are necessary in the sense that lacking any of them, a working flagellum 
does not result.

The only way for natural selection to form such a structure by co-optation, 
then, is for natural selection gradually to enfold existing protein parts 
into evolving structures whose functions co-evolve with the structures. We 
might, for instance, imagine a five-part mousetrap consisting of a 
platform, spring, hammer, holding bar, and catch evolving as follows: It 
starts as a doorstop (thus consisting merely of the platform), then evolves 
into a tie-clip (by attaching the spring and hammer to the platform), and 
finally becomes a full mousetrap (by also including the holding bar and 
catch).

Ken Miller finds such scenarios not only completely plausible but also 
deeply relevant to biology (in fact, he regularly sports a modified 
mousetrap cum tie-clip). Intelligent design proponents, by contrast, regard 
such scenarios as rubbish. Here's why. First, in such scenarios the hand of 
human design and intention meddles everywhere. Evolutionary biologists 
assure us that eventually they will discover just how the evolutionary 
process can take the right and needed steps without the meddling hand of 
design.

But all such assurances presuppose that intelligence is dispensable in 
explaining biological complexity. The only evidence we have of successful 
co-optation, however, comes from engineering and confirms that intelligence 
is indispensable in explaining complex structures like the mousetrap and by 
implication the flagellum. Intelligence is known to have the causal power 
to produce such structures. We're still waiting for the promised material 
mechanisms.

The other reason design theorists are less than impressed with co-optation 
concerns an inherent limitation of the Darwinian mechanism. The whole point 
of the Darwinian selection mechanism is that you can get from anywhere in 
configuration space to anywhere else provided you can take small steps. How 
small? Small enough that they are reasonably probable. But what guarantee 
do you have that a sequence of baby-steps connects any two points in 
configuration space?

Richard Dawkins compares the emergence of biological complexity to climbing 
a mountain--Mount Improbable, as he calls it. According to him, Mount 
Improbable always has a gradual serpentine path leading to the top that can 
be traversed in baby-steps. But that's hardly an empirical claim. Indeed, 
the claim is entirely gratuitous. It might be a fact about nature that 
Mount Improbable is sheer on all sides and getting to the top from the 
bottom via baby-steps is effectively impossible. A gap like that would 
reside in nature herself and not in our knowledge of nature (it would not, 
in other words, constitute a god-of-the-gaps).

The problem is worse yet. For the Darwinian selection mechanism to connect 
point A to point B in configuration space, it is not enough that there 
merely exist a sequence of baby-steps connecting the two. In addition, each 
baby-step needs in some sense to be "successful." In biological terms, each 
step requires an increase in fitness as measured in terms of survival and 
reproduction. Natural selection, after all, is the motive force behind each 
baby-step, and selection only selects what is advantageous to the organism. 
Thus, for the Darwinian mechanism to connect two organisms, there must be a 
sequence of successful baby-steps connecting the two.

Again, it is not enough merely to presuppose this--it must be demonstrated. 
For instance, it is not enough to point out that some genes for the 
bacterial flagellum are the same as those for a type III secretory system 
(a type of pump) and then handwave that one was co-opted from the other. 
Anybody can arrange complex systems in a series. But such series do nothing 
to establish whether the end evolved in a Darwinian fashion from the 
beginning unless the probability of each step in the series can be 
quantified, the probability at each step turns out to be reasonably large, 
and each step constitutes an advantage to the organism (in particular, 
viability of the whole organism must at all times be preserved).

Convinced that the Darwinian mechanism must be capable of doing such 
evolutionary design work, evolutionary biologists rarely ask whether such a 
sequence of successful baby-steps even exists; much less do they attempt to 
quantify the probabilities involved. I attempt that in chapter 5 of my most 
recent book No Free Lunch. There I lay out techniques for assessing the 
probabilistic hurdles that the Darwinian mechanism faces in trying to 
account for complex biological structures like the bacterial flagellum. The 
probabilities I calculate--and I try to be conservative--are horrendous and 
render natural selection entirely implausible as a mechanism for generating 
the flagellum and structures like it.

If I'm right and the probabilities really are horrendous, then the 
bacterial flagellum exhibits specified complexity. Furthermore, if 
specified complexity is a reliable marker of intelligent agency, then 
systems like the bacterial flagellum bespeak intelligent design and are not 
solely the effect of material mechanisms.

It's here that critics of intelligent design raise the 
argument-from-ignorance objection. For something to exhibit specified 
complexity entails that no known material mechanism operating in known ways 
is able to account for it. But that leaves unknown material mechanisms. It 
also leaves known material mechanisms operating in unknown ways. Isn't 
arguing for design on the basis of specified complexity therefore merely an 
argument from ignorance?

Two comments to this objection: First, the great promise of Darwinian and 
other naturalistic accounts of evolution was precisely to show how known 
material mechanisms operating in known ways could produce all of biological 
complexity. So at the very least, specified complexity is showing that 
problems claimed to be solved by naturalistic means have not been solved. 
Second, the argument from ignorance objection could in principle be raised 
for any design inference that employs specified complexity, including those 
where humans are implicated in constructing artifacts. An unknown material 
mechanism might explain the origin of the Mona Lisa in the Louvre, or the 
Louvre itself, or Stonehenge, or how two students wrote exactly the same 
essay. But no one is looking for such mechanisms. It would be madness even 
to try. Intelligent design caused these objects to exist, and we know that 
because of their specified complexity.

Specified complexity, by being defined relative to known material 
mechanisms operating in known ways, might always be defeated by showing 
that some relevant mechanism was omitted. That's always a possibility 
(though as with the plagiarism example and with many other cases, we don't 
take it seriously). As William James put it, there are live possibilities 
and then again there are bare possibilities. There are many design 
inferences which, to question or doubt, require invoking a bare 
possibility. Such bare possibilities, if realized, would defeat specified 
complexity. But defeat specified complexity in what way? Not by rendering 
the concept incoherent but by dissolving it.

In fact, that is how Darwinists, complexity theorists, and anyone intent on 
defeating specified complexity as a marker of intelligence usually attempts 
it, namely, by showing that it dissolves once we have a better 
understanding of the underlying material mechanisms that render the object 
in question reasonably probable. By contrast, design theorists argue that 
specified complexity in biology is real: that any attempt to palliate the 
complexities or improbabilities by invoking as yet unknown mechanisms or 
known mechanisms operating in unknown ways is destined to fail. This can in 
some cases be argued convincingly, as with Michael Behe's irreducibly 
complex biochemical machines and with biological structures whose geometry 
allows complete freedom in possible arrangements of parts.

Consider, for instance, a configuration space comprising all possible 
character sequences from a fixed alphabet (such spaces model not only 
written texts but also polymers like DNA, RNA, and proteins). Configuration 
spaces like this are perfectly homogeneous, with one character string 
geometrically interchangeable with the next. The geometry therefore 
precludes any underlying mechanisms from distinguishing or preferring some 
character strings over others. Not material mechanisms but external 
semantic information (in the case of written texts) or functional 
information (in the case of polymers) is needed to generate specified 
complexity in these instances. To argue that this semantic or functional 
information reduces to material mechanisms is like arguing that Scrabble 
pieces have inherent in them preferential ways they like to be sequenced. 
They don't. Michael Polanyi offered such arguments for biological design in 
the 1960s.

In summary, evolutionary biology contends that material mechanisms are 
capable of accounting for all of biological complexity. Yet for biological 
systems that exhibit specified complexity, these mechanisms provide no 
explanation of how they were produced. Moreover, in contexts where the 
causal history is independently verifiable, specified complexity is 
reliably correlated with intelligence. At a minimum, biology should 
therefore allow the possibility of design in cases of biological specified 
complexity. But that's not the case.

Evolutionary biology allows only one line of criticism, namely, to show 
that a complex specified biological structure could not have evolved via 
any material mechanism. In other words, so long as some unknown material 
mechanism might have evolved the structure in question, intelligent design 
is proscribed. This renders evolutionary theory immune to disconfirmation 
in principle, because the universe of unknown material mechanisms can never 
be exhausted. Furthermore, the evolutionist has no burden of evidence. 
Instead, the burden of evidence is shifted entirely to the evolution 
skeptic. And what is required of the skeptic? The skeptic must prove 
nothing less than a universal negative. That is not how science is supposed 
to work.

Science is supposed to pursue the full range of possible explanations. 
Evolutionary biology, by limiting itself to material mechanisms, has 
settled in advance which biological explanations are true apart from any 
consideration of empirical evidence. This is arm-chair philosophy. 
Intelligent design may not be correct. But the only way we could discover 
that is by admitting design as a real possibility, not ruling it out a 
priori. Darwin himself agreed. In the Origin of Species he wrote: "A fair 
result can be obtained only by fully stating and balancing the facts and 
arguments on both sides of each question."

POWRÓT