Where the Conflict Really Lies: Science, Religion, and Naturalism (29 page)

What this arms race problem really shows is that the FTA is pretty anemic if stated in the likelihood version. Sadly enough, something similar holds for the Bayes theoretic version. Here, of course, we do take the antecedent probabilities into account. But how do we figure out the antecedent probability of theism? What is the probability of theism, prior to the consideration of the evidence, if any, afforded by fine-tuning? There are serious problems in determining this. First, there is the problem with the modal status of theism. According to classical theism, God is a necessary being; he exists in every possible world; the probability of theism is therefore 1, and 1 on any evidence. Of course the atheist will think the probability of classical theism is 0;
if the proposition that there is such a person as God, conceived as a necessary being, is false, then it is necessarily false, in which case its probability is 0, and 0 on any evidence.

Suppose we turn from objective probability to epistemic probability, or from classical theism to a version in which theism is, if true, contingently true: we still have problems. The theist will assign a high probability here; the atheist a low probability; others a probability somewhere between these extremes; still others won’t find themselves able to assign any probability at all to the proposition. So to whom is this argument addressed? It may possibly add a bit of confirmation for the theist; for the atheist it may slightly raise the probability of theism.
³⁹
Perhaps the argument can carry the most weight with those who are prepared to assign a probability to theism, but one that is neither extremely high nor extremely low (provided they see their way past the many-worlds objection); perhaps it can sensibly be expected to cause them to modify their estimates of the probability of theism. It is fairly clear, however, that FTA, taken this way—that is, taken as involving antecedent probabilities—offers at best modest support for theism. Of course modest support is still support, and it is not to be disdained.

The third way in which the FTA can be construed is as an inference to the best explanation. We should note that inference to the best explanation isn’t really
inference
: you aren’t compelled by some rule of inference to accept a bad explanation of some phenomenon, even if that explanation is the best one you can think of. Suppose
there are six candidates; suppose the most probable among them has a probability of .2. Even if that explanation is the best one, you will quite properly refuse to accept it as the truth of the matter. And this points to a problem with the FTA construed as something like an inference to the best explanation; substantially the same problem that afflicts it construed Bayesianly. Part of what makes an explanation good or bad is its probability. My car won’t start. A Brazilian tribesman might think it quite likely that it is inhabited by a malicious spirit who is out to give me trouble (or at least minor annoyances). That explanation doesn’t commend itself to me, however—mainly because I think it very unlikely. Upon walking past the Notre Dame Stadium on a Saturday afternoon, we hear a roar from the crowd. I suggest the explanation is that there is a football game there with thousands of fans present and a small plane has just flown over, dropping many $1000 bills. That would certainly explain the roar, all right, but you very sensibly refuse to accept it because there are other, more probable explanations. So we are back at the antecedent probability of theism: whether theism is a good explanation of the phenomena depends in part on the antecedent probability of theism. And this will be no easier to fix in this case than in the case of the FTA taken Bayesianly.

The right conclusion, I think, is that the FTA offers some slight support for theism. It does offer support, but only mild support. Granted: this is not a very exciting conclusion, not nearly as exciting as the conclusion that the argument is extremely powerful, or the conclusion that it is wholly worthless. It does, however, have the virtue of being correct. In the next chapter we’ll consider a different and more promising approach to the phenomena to which the FTA appeals.

In the last chapter, we thought about cosmological fine-tuning arguments for the conclusion that our world has been designed.

There are biological arguments for the same conclusion. Chief among these are Michael Behe’s. In
Darwin’s Black Box
he created quite a stir by arguing that there are a number of structures and phenomena at the molecular level that display “irreducible complexity.”
¹
Behe then describes a number of molecular structures that, he says, are “irreducibly complex” in his sense: among them are the bacterial flagellum, the cilia employed by several kinds of cells for motion and other functions, the incredibly complex cascade of biochemical reactions and events that occur in vision, blood clotting, the transport of materials within cells, and the immune system. He doesn’t say merely that there are such systems: he explains them in considerable and illuminating detail.

What exactly is irreducible complexity? According to Behe, “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.”
²

As an example, consider his account of the biochemistry of human vision, in particular the astounding series of events that take place at the molecular level when a photon strikes the retina. What happens is a stunning, multifarious concatenation of interconnected biochemical reactions and events. These are related in such a way that if any of them doesn’t occur, or doesn’t function properly, vision won’t happen.
³

According to Behe, there is no way in which this whole complex system could have evolved (from simpler systems, by way of the small, incremental steps required by a Darwinian explanation). That is because the system is irreducibly complex. And the significance of irreducible complexity is this:

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.
⁴

Irreducibly complex structures and phenomena, therefore, can’t have come to be, he says, by gradual, step-by step Darwinian evolution. These systems, then, present what Behe calls a Lilliputian challenge to (unguided) Darwinism; if he is right, he might also have said they present a Gargantuan challenge.

What is peculiarly interesting about his account of the biochemistry of vision is that the adequacy of Darwinian explanations of the eye has been a subject of dispute and controversy ever since Darwin’s day. A great deal was known about the anatomy of the eye in the
nineteenth century. It was known that the pupil is an adjustable shutter that closes down to let in the right amount of light to permit vision in bright sunlight, but also opens up in dim light. The lens focuses light in a sharp image on the retina; the lens is also adjustable; the ciliary muscle causes it to relax or contract in order to accommodate both near and distant vision; and the density of the lens changes over its surface to correct for chromatic aberration, thus permitting color vision. St. George Mivart, for example, pointed out as early as 1871 that there is a problem here for Darwinian explanations. Such explanations proceed in terms of natural selection operating on a source of genetic variation such as so-called random genetic variation. Obviously the various components of the eye must all work together to enable vision; so all these different parts would have to evolve in tandem. But an improvement to the lens (a step on the way to the present function of the lens) wouldn’t automatically permit better vision, and in fact may interfere with it.

The problem, then, is to conceive of a series of steps through “design space” where (1) the first point is occupied by a design with no more than a light sensitive spot, as we find with certain relatively primitive animals; (2) each point (except the first) represents a design arising by way of heritable genetic variation (the main candidate is random genetic mutation) from the previous point; (3) divine or other guidance or causality is not involved in the transition from any point to the next; (4) each point is an adaptive step forward with respect to the previous point, or else a consequence, by way of spandrel or pliotropy, of a design that is such a step forward; (5) each point is not overwhelmingly improbable with respect to the previous point; and (6) the last point is occupied by (correlated with) the design of the human eye.
⁵

Naturally it isn’t required that the Darwinist come up with the actual sequence of design plans here, or a sequence which could have been that actual sequence; what is required for a satisfying evolutionary account of the eye is perhaps more like a reasonably detailed specification of some important stages along the way. Darwin attempted to meet the challenge by pointing to a series of other animal eyes varying in complexity from a simple light sensitive spot, a patch of photoreceptors as in certain jellyfish, through cupped eyes (that is, a patch of photoreceptors at the bottom of a kind of cup, thus registering the direction from which the incident light is coming) as in marine limpets, and primitive eyes with a lens, as in marine snails. It’s hard to say what exactly
is
required, and it is equally hard to say whether Darwinists have discharged their obligation to make the requisite specifications. Due to long familiarity, however, most people seem to have become accustomed to the problem, and tend for the most part to ignore it. (Of course similar problems beset proposed Darwinian accounts of other anatomical features and structures.)

Behe claims something stronger than just that Darwinian accounts of the structures he describes—vision, cilia, flagella, cellular transport, and so on—are extremely
difficult
to find. Sometimes, at least, he claims that there
aren’t
any Darwinian accounts of any of these structures, and, by virtue of their irreducible complexity, there
won’t be
any; such accounts can’t possibly be given. What he proposes as an alternative, notoriously, is intelligent design.
⁶

Not everyone is pleased. We are in the neighborhood of cultural conflicts (“culture wars”) where feelings run high; the level of vitriol, vituperation and contempt heaped on Behe’s unsuspecting head is really quite remarkable. There are screams of hysterical anguish, frenzied denunciations, accusations of treason (how could an actual scientist say things like this?), charges of deceit, duplicity, deviousness,
tergiversation, pusillanimity, and other indications of less than total agreement.
⁷
One is reminded of the medieval philosopher Peter Damian, who said that those who held a certain position (oddly enough, one different from his own) are contemptible, not worthy of a reply, and should instead be branded. Many of those who comment on Behe seem to think along similar lines. These screeds are not of course the sort of thing to which one can give an argumentative reply: they aren’t so much arguments as brickbats.

Fortunately there are less hysterical replies to Behe’s arguments. One of the best is by Paul Draper.
⁸
According to Draper, Behe fails to show that the systems he says are irreducibly complex are in fact irreducibly complex (that is, such that if they missed any of their parts, they couldn’t function at all); some biochemists, Draper points out, have argued that they are not. Perhaps some of these systems could function, even if not as well, in the absence of one or more of their parts. There is also difficulty, here, about what constitutes a
part
: couldn’t some of these systems function perfectly well in the absence of a molecule or two? So Behe’s book doesn’t really deliver what he says it will—at any rate he doesn’t show that these systems are irreducibly complex in the sense he specifies. “A careful reading of the second part of his book,” says Draper, “reveals that he rarely even claims (let alone proves) that all of the parts of the systems he discusses are required for those systems to function.”
⁹
In a subsequent
paper, Draper reports, Behe changes his definition of irreducible complexity to accommodate the above difficulties:

This response presupposes a new definition of irreducible complexity, one that implies that a system is irreducibly complex even if it has working parts that are not essential for it to function, so long as it has (at least two) interacting and closely matched parts which are essential. Given this definition, it would seem that the biochemical systems Behe discusses are indeed irreducibly complex.
¹⁰

Behe argues that there are no direct Darwinian routes to these complex phenomena. Draper points out however, that even if there aren’t any direct Darwinian routes to these systems, there may still be various
indirect
Darwinian routes to them:

The sort of route I have in mind occurs when an irreducibly complex and irreducibly specific system S that serves function F evolves from a precursor S* that shares many of S’s parts but serves a different function F*. Notice that parts that S and S* share and that are required for F need not be required for F* even if they contribute to F*, and parts that are irreducibly specific relative to F may be only reducibly specific relative to F*. Thus, both the parts of S* and their specificity may have been gradually produced by a direct evolutionary path. Then one or more additional parts are added to S*, resulting in a change of function from F* to F. And relative to F, the parts and their specificity, which had not been essential for F*, are now essential.
¹¹