This is the main optional reading for our discussion group this Sunday. It was written by philosopher Michael Ruse in 2003. He's generally against "intelligent design," though he tries to objectively and fairly evaluate the arguments presented in support of it. In case you want to read the original article, it's from one of my favorite websites, The Stanford Encyclopedia of Philosophy. Here's the link: plato.stanford.edu/entries/creationism/
The entire article isn't directly relevant to our topic, so I've pasted below only the relevant parts of it, sections 6-9.
-------------------------------------------

6. Intelligent Design
Let us move on now from the more philosophical sorts of issues. Complementing Johnson today, there is a group of people who are trying to offer an alternative to evolution. These are the enthusiasts for so-called ‘Intelligent Design.’ Supporters of this position think that Darwinism is ineffective, at least inasmuch as it claims to make superfluous or unnecessary a direct appeal to a designer of some sort. These are people who think that a full understanding of the organic world demands the invocation of some force beyond nature, a force which is purposeful or at least purpose creating.

There are two parts to this approach: an empirical and a philosophical. Let us take them in turn, beginning with he who has most fully articulated the empirical case for a designer, the Lehigh University biochemist Michael Behe. Focusing on something which he calls ‘irreducible complexity,’ Behe writes:

By irreducibly complex I mean a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning. An irreducibly complex system cannot be produced directly (that is, by continuously improving the initial function, which continues to work by the same mechanism) by slight, successive modifications of a precursor system, because any precursor to an irreducibly complex system that is missing a part is by definition nonfunctional. (Behe 1996, 39)
Behe adds, surely truly, that any

irreducibly complex biological system, if there is such a thing, would be a powerful challenge to Darwinian evolution. Since natural selection can only choose systems that are already working, then if a biological system cannot be produced gradually it would have to arise as an integrated unit, in one fell swoop, for natural selection to have anything to act on (p. 39).
Now turn to the world of biology, and in particular turn to the micro-world of the cell and of mechanisms (or ‘mechanisms’) that we find at that level. Take bacteria which use a flagellum, driven by a kind of rotary motor, to move around. Every part is incredibly complex, and so are the various parts, combined. The external filament of the flagellum (called a ‘flagellin’), for instance, is a single protein that makes a kind of paddle surface contacting the liquid during swimming. Near the surface of the cell, just as needed is a thickening, so that the filament can be connected to the rotor drive. This naturally requires a connector, known as a ‘hook protein.’ There is no motor in the filament, so that has to be somewhere else. ‘Experiments have demonstrated that it is located at the base of the flagellum, where electron microscopy shows several ring structures occur’ (p. 70). All, way too complex to have come into being in a gradual fashion. Only a one-step process will do, and this one-step process must involve some sort of designing cause. Behe is careful not to identify this designer with the Christian God, but the implication is that it is a force from without the normal course of nature. Irreducible complexity spells design.

7. Is Complexity Irreducible?
Irreducible complexity is supposedly something which could not have come through unbroken law, and especially not through the agency of natural selection. Critics claim that Behe shows a misunderstanding of the very nature and workings of natural selection. No one is denying that in natural processes there may well be parts which, if removed, would lead at once to the non-functioning of the systems in which they occur. The point however is not whether the parts now in place could not be removed without collapse, but whether they could have been put in place by natural selection. Consider an arched bridge, made from cut stone, without cement, held in place only by the force of the stones against each other. If you tried to build the bridge from scratch, upwards and then inwards, you would fail — the stones would keep falling to the ground, as indeed the whole bridge now would collapse were you to remove the center keystone or any surrounding it. Rather, what you must do is first build a supporting structure (possibly an earthen embankment), on which you will lay the stones of the bridge, until they are all in place. At which point you can remove the structure for it is no longer needed, and in fact is in the way. Likewise, one can imagine a biochemical sequential process with several stages, on the parts of which other processes piggyback as it were. Then the hitherto non-sequential parasitic processes link up and start functioning independently, the original sequence finally being removed by natural selection as redundant or inconveniently draining of resources.

Of course, this is all pretend. But Darwinian evolutionists have hardly ignored the matter of complex processes. Indeed, it is discussed in detail by Darwin in the Origin, where he refers to that most puzzling of all adaptations, the eye. At the biochemical level, today's Darwinians have many examples of the most complex of processes that have been put in place by selection. Take that staple of the body's biochemistry, the process where energy from food is converted into a form which can be used by the cells. Rightly does a standard textbook refer to this vital organic system, the so-called ‘Krebs cycle,’ as something which ‘undergoes a very complicated series of reactions’ (Hollum 1987, 408). This process, which occurs in the cell parts known as mitochondria, involves the production of ATP (adenosine triphosphate): a complex molecule which is energy rich and which is degraded by the body as needed (say in muscle action) into another less rich molecule ADP (adenosine diphosphate). The Krebs cycle remakes ATP from other energy sources — an adult human male needs to make nearly 200 Kg a day — and by any measure, the cycle is enormously involved and intricate. For a start, nearly a dozen enzymes (substances which facilitate chemical processes) are required, as one sub-process leads on to another.

Yet the cycle did not come out of nowhere. It was cobbled together out of other cellular processes which do other things. It was a ‘bricolage.’ Each one of the bits and pieces of the cycle exists for other purposes and has been coopted for the new end. The scientists who have made this connection could not have made a stronger case against Behe's irreducible complexity than if they had had him in mind from the first. In fact, they set up the problem virtually in Behe's terms: ‘The Krebs cycle has been frequently quoted as a key problem in the evolution of living cells, hard to explain by Darwin's natural selection: How could natural selection explain the building of a complicated structure in toto, when the intermediate stages have no obvious fitness functionality?’ (Meléndez-Hervia et al 1996, 302). What these workers do not offer is a Behe-type answer. First, they brush away a false lead. Could it be that we have something like the evolution of the mammalian eye, where primitive existent eyes in other organisms suggest that selection can and does work on proto models (as it were), refining features which have the same function if not as efficient as more sophisticated models? Probably not, for there is no evidence of anything like this. But then we are put on a more promising track.

In the Krebs cycle problem the intermediary stages were also useful, but for different purposes, and, therefore, its complete design was a very clear case of opportunism. The building of the eye was really a creative process in order to make a new thing specifically, but the Krebs cycle was built through the process that Jacob (1977) called ‘evolution by molecular tinkering,’ stating that evolution does not produce novelties from scratch: It works on what already exists. The most novel result of our analysis is seeing how, with minimal new material, evolution created the most important pathway of metabolism, achieving the best chemically possible design. In this case, a chemical engineer who was looking for the best design of the process could not have found a better design than the cycle which works in living cells. (p. 302)

Rounding off the response to Behe, let us note that, if his arguments are well-taken, then in respects we are in bigger problems than otherwise! His position seems simply not viable given what we know of the nature of mutation and the stability of biological systems over time. When exactly is the intelligent designer supposed to strike and to do its work? In his major work, Darwin's Black Box, Behe suggests that everything might have been done long ago and then left to its own devices. ‘The irreducibly complex biochemical systems that I have discussed... did not have to be produced recently. It is entirely possible, based simply on an examination of the systems themselves, that they were designed billions of years ago and that they have been passed down to the present by the normal processes of cellular reproduction’ (Behe 1996, 227-8).

This is not a satisfactory response. We cannot ignore the history of the preformed genes from the point between their origin (when they would not have been needed) and today when they are in full use. In the words of Brown biochemist Kenneth Miller: ‘As any student of biology will tell you, because those genes are not expressed, natural selection would not be able to weed out genetic mistakes. Mutations would accumulate in these genes at breathtaking rates, rendering them hopelessly changed and inoperative hundreds of millions of years before Behe says that they will be needed.’ There are masses of experimental evidence showing that this is the case. Behe's idea of designer doing everything back then and then leaving matters to their natural fate is ‘pure and simple fantasy’ (Miller 1999, 162-3).

What is the alternative strategy that Behe must take? Presumably that the designer is at work all of the time, producing mechanisms as and when needed. So, if we are lucky, we might expect to see some produced in our lifetime. Indeed, there must be a sense of disappointment among biologists that no such creative acts have so far been reported. More than this, as we turn from science towards theology, there even greater disappointments. Most obviously, what about malmutations? If the designer is needed and available for complex engineering problems, why could not the designer take some time on the simple matters, specifically those simple matters which if unfixed lead to absolutely horrendous problems. Some of the worst genetic diseases are caused by one little alteration in one little part of the DNA. If the designer is able and willing to do the very complex because it is very good, why does it not do the very simple because the alternative is very bad? Behe speaks of this as being part of the problem of evil, which is true, but not very helpful. Given that the opportunity and ability to do good was so obvious and yet not taken, we need to know the reason why.

8. The Explanatory Filter
Behe is in need of help. This supposed comes from a conceptual argument in favor of Intelligent Design due to the philosopher-mathematician William Dembski (1998a, b). Let us first look at his argument, and then see how it helps Behe.

Dembski's aim is two-fold. First, to give us the criteria by which we distinguish something that we would label ‘designed’ rather than otherwise. Second, to put this into context, and show how we distinguish design from something produced naturally by law or something we would put down to chance. As far as inferring design is concerned, there are three notions of importance: contingency, complexity, and specification. Design has to be something which is not contingent. The example that Dembski uses is the message from outer space received in the movie Contact. The series of dots and dashes, zeros and ones, could not be deduced from the laws of physics. But do they show evidence of design? Suppose we can interpret the series in a binary fashion, and the initial yield is the number group, 2, 3 , 5. As it happens, these are the beginning of the prime-number series, but with so small a yield no one is going to get very excited. It could just be chance. So no one is going to insist on design yet. But suppose now you keep going on the series, and it turns out that it yields in exact and precise order the prime numbers up to 101. Now you will start to think that something is up, because the situation seems just too complex to be mere chance. It is highly improbable. ‘Complexity as I am describing it here is a form of probability....’ (Dembski 2000, 27).

But although you are probably happy now to conclude (on the basis of the prime-number sequence) that there are extraterrestrials out there, in fact there is another thing needed. ‘If I flip a coin 1000 times, I will participate in a highly complex (that is, highly improbable) event.... This sequence of coin tosses will not, however, trigger a design inference. Though complex, this sequence will not exhibit a suitable pattern.’ Here, we have a contrast with the prime-number sequence from 2 to 101. ‘Not only is this sequence complex, but it also embodies a suitable pattern. The SETI researcher who in the movie Contact discovered this sequence put it this way: "This isn't noise, this has structure"’ (pp. 27-8). What is going on here? You recognize in design something which is not just arbitrary or chance or which is given status only after the experiment or discovery, but rather something that was or could be in some way specified, insisted upon, before you set out. You know or could work out the sequence of prime numbers at any time before or after the contact from space. The random sequence of penny tosses will come only after the event. ‘The key concept is that of "independence". I define a specification as a match between an event and an independently given pattern. Events that are both highly complex and specified (that is, that match an independently given pattern) indicate design.’

Dembski is now in a position to move on to the second part of his argument where we actually detect design. Here we have what he calls an ‘Explanatory Filter’ (Dembski 1998a, b). We have a particular phenomenon. The question is, what caused it? Is it something which might not have happened, given the laws of nature? Is it contingent? Or was it necessitated? The moon goes endlessly round the earth. We know that it does this because of Newton's laws. End of discussion. No design here. However, now we have some rather strange new phenomenon, the causal origin of which is a puzzle. Suppose we have a mutation, where although we can quantify over large numbers we cannot predict at an individual level. There is no immediate subsumption beneath law, and therefore there is no reason to think that at this level it was necessary. Let us say, as supposedly happened in the extended royal family of Europe, there was a mutation to a gene responsible for hemophilia. Is it complex? Obviously not, for it leads to breakdown rather than otherwise. Hence it is appropriate to talk now of chance. There is no design. The hemophilia mutation was just an accident.

Suppose now that we do have complexity. A rather intricate mineral pattern in the rocks might qualify here. Suppose we have veins of precious metals set in other materials, the whole being intricate and varied — certainly not a pattern you could simply deduce from the laws of physics or chemistry or geology or whatever. Nor would one think of it as being a breakdown mess, as one might a malmutation. Is this now design? Almost certainly not, for there is no way that one might pre-specify such a pattern. It is all a bit ad hoc, and not something which comes across as the result of conscious intention. And then finally there are phenomena which are complex and specified. One presumes that the microscopical biological apparatuses and processes discussed by Behe would qualify here. They are contingent, for they are irreducibly complex. They are design-like for they do what is needed for the organism in which they are to be found. That is to say they are of pre-specified form. And so, having survived the explanatory filter, they are properly considered the product of real design.

Now, with the conceptual argument laid out in full, we are in a position to turn back to Behe and to see how Dembski's explanatory filter is supposed to let Behe's god off the hook with respect to the problem of evil. Given the explanatory filter, a malmutation would surely get caught by the filter half-way down. It would be siphoned off to the side as chance, if not indeed simply put down as necessity. It certainly would not pass the specification test. This would mean that a dreadful genetic disease would not be the fault of the designer, whereas successful complex mechanisms would be to the designer's credit. Dembski stresses that these are mutually exclusive alternatives. ‘To attribute an event to design is to say that it cannot plausibly be referred to either law or chance. In characterizing design as the set-theoretic complement of the disjunction law-or-chance, one therefore guarantees that these three modes of explanation will be mutually exclusive and exhaustive’ (Dembski 1998b, 98).

9. Mutually exclusive?
The key assumption being made by Dembski is that design and law and chance are mutually exclusive. This is the very essence of the explanatory filter. But in real life does one want to make this assumption? Suppose that something is put down to chance. Does this mean that law is ruled out? Surely not! If one argues that a Mendelian mutation is chance, what one means is with respect to that particular theory it is chance, but one may well believe that the mutation came about by normal regular causes and that if these were all known, then it would not longer be chance at all but necessity. The point is that chance is a confession of ignorance not (as one might well think the case in the quantum world) an assertion about the way that things are. That is, claims about chance are not ontological assertions, as presumably claims about designers must be.

More than this, one might well argue that the designer always works through law. This may be deism and hence no true Christianity — some Christians would insist that God does sometimes intervene in the Creation. But truly Christian or not, a deity who always works through law is certainly not inconsistent with the hypothesis of a designing intelligence. The designer may prefer to have things put in motion in such a way that his/her/its intentions unfurl and reveal themselves as time goes by. The pattern in a piece of cloth made by machine is as much an object of design as the pattern from cloth produced by a hand loom. In other words, in a sense that would conform to the normal usage of the terms, one might want to say of something that it is produced by laws, is chance with respect to our knowledge or theory, and fits into an overall context of design by the great orderer or creator of things. In short, Dembski's filter does not let Behe's designer off the hook.

If the designer can make — and rightfully takes credit for — the very complex and good, then the designer could prevent — and by its failure is properly criticized for — the very simple and awful. The problems in theology are as grim as are those in science. (The intelligent design theorists have provided work for many philosophers eager to refute them. Pennock 1988 and Sober 2000 are good places to start.)