"Mark Ridley’s Evolution has become the premier undergraduate text in the study of evolution," according to the back cover of the third edition.1 The book is published by Blackwell, a prestigious academic publishing house, and Riley himself currently works in the Department of Zoology at Oxford, having also held positions at Cambridge and Emory.
In this textbook he has a section on "The Evidence for Evolution." Presumably there is no better place to go to find the best available evidence for evolution—up to the time of publication (2004).
Under the first heading, We distinguish three possible theories of the history of life, he says:
For purposes of argument it is useful to have some articulate alternatives to argue between. We can discuss three theories (Figure 3.1): (a) evolution; (b) "transformism"…and (c) separate creation, in which species originated separately and remain fixed (44).The problem with this introductory statement is that it misrepresents creationism in two key respects:
In separate creation, species have separate origins and do not change; each [c-e] are different versions of the theory of separate creation that might be proposed to explain extinct fossil forms, and they do not differ in their two essential features (species have separate origins and do not change) (44, figure 3.1).
Whether species have separate origins, and whether they change after their origin, are two distinct questions; some kinds of evidence, therefore, may bear upon one of question [sic.] but not the other (44).
The existence of fossil species unlike anything alive today, however, does not distinguish between the three theories of life in Figure 3.1. An extinct species could just as well have been separately created as any modern species. The theory of separate creation can easily be modified to account for extinct forms. Either there was one period in which all species separately originated and some have subsequently gone extinct (Figure 3.1d) or there were rounds of extinction followed by round of creation (3.1.e). All three versions of separate creation (Figure 3.1 c-e) share the key features that species have separate origins and do not change in form after their origin (44-45).
We concentrate here on evidence that can be used to test between the three theories in Figure 3.1 [evolution, transformism, and creationism] (45).
i) If by "separate creation," we mean the Biblical doctrine of special creation, then the fundamental unit of separate creation is not the species, but the natural kind. A natural kind is a broader category than a species.
The Bible doesn’t operate with a modern taxonomy. So it would be anachronistic to equate separate creation with scientific taxa.
ii) It is unclear what Ridley means when he says that according to "separate creation," a species does not "change," or change in "form," but remains "fixed."
OT Jews were certainly aware of natural variations. They were acquainted with racial diversity. They knew the difference between wild animals and domestic livestock. They practiced such forms of selective breeding as animal husbandry and viticulture.
The next section falls under the heading of: On a small scale, evolution can be observed in action
He cites drug-resistant strains of HIV to illustrate this principle. But that would be a case of microevolution. Creationism doesn’t deny microevolution. This is not inconsistent with the Biblical doctrine of special (or "separate") creation.
He also mentions the famous example of evolution in the peppered moth (46).
i) To begin with, he doesn’t explain how that’s an example of evolution. Rather, that would seem to be an example of natural selection. Some moths with some markings survive, while other moths with other markings do not. That would not be a case of one moth evolving into another moth. It would merely affect the relative frequency of certain preexisting species.2
ii) Moreover, the development of specific camouflage, even if that were in view, would be an instance if microevolution rather than macroevolution, would it not?
iii) Furthermore, special creation doesn’t rule out speciation, for—as already noted—special creation has reference to natural kinds rather than species.
We look at changes in the average beak size of a population of a finch species in the Galapagos islands…we look at geographic variation in the house sparrow (46).But, once again,
i) These are examples of microevolution rather than macroevolution.
ii) They operate at the level of species rather than natural kinds.
The next section falls under the heading of Evolution can also be produced experimentally
This has reference to artificial selection, viz. rats could be successfully selected to grow better or worse teeth. Evolutionary change can therefore be generated artificially.
i) Artificial selection is fundamentally different from natural selection, for artificial selection requires a rational, goal-oriented agent—the human breeder.
ii) This would be yet another instance of microevolution rather than macroevolution.
iii) As already noted, Ancient Near Easterners were quite familiar with such forms of artificial selection as animal husbandry and horticulture. Gen 1-2 is not opposed to such phenomena.
The next section falls under the heading of Interbreeding and phenotypic similarity provide to concepts of species
Most of the evidence so far has been for small-scale change within a species. The amounts of artificially selected change in pigeons and other domestic animals borders on the species level, but to decide whether the species barrier has been crossed we need a concept of what a biological species is…What does it mean to say a new species has evolved? The question unfortunately lacks a simple answer that would satisfy all biologists…there are several concepts of species (48).But as I already said, Biblical creationism doesn’t rise or fall on the question of whether "new species have evolved," or whether the "species barrier has been crossed," according to a narrowly technical and equivocal definition of what constitutes a "species."
The question, rather, is whether life occurred the way the Bible describes it or the way Ridley describes it.
Museum experts often have to classify birds from dead specimens, of unknown reproductive habits, and they make use of phenotypic characters of the bones, beak, and feathers (50).This exposes the precarious nature of some classification schemes.
The next section falls under the heading of Ring "species" show that variation within a species can be extensive enough to produce new species
Ring species can provide important evidence for evolution, because they show that intraspecific differences can be large enough to produce an interspecies difference…At least some species, therefore, have arisen without separate creation…to deny it would require an arbitrary decision about where evolution stopped and separate creation started…The idea that nature comes in discrete groups, with no variation between, is a naïve perception (52-53).Unfortunately for him, Ridley is the one who is guilty of drawing arbitrary distinctions and erecting false dichotomies. He keeps flailing away at a straw man argument. None of this makes a dent in Biblical creationism.
Scripture never says there can be no "variation" between "separately" created organisms. Indeed, Scripture allows for natural variation.
The next sections falls under the heading of New, reproductively distinct species can be produced experimentally
The species barrier can be broken by experiment too. The varieties of artificially produced domestic animals and plants can differ in appearance at least as much as natural species; but they may also be able to interbreed…In conclusion, it is possible to make new, reproductively isolated species, using a method that ha been highly important in the origin of new natural species (53-54).Once more, Ridley is simply repeating himself by restating the same fallacious argument in slightly different ways, with different illustrations. He is both persistent and consistent. Unfortunately for him, he is aiming at the wrong target, so even if he never misses the target, his back is to the true target.
The next section falls under the heading of Small-scale observations can be extrapolated over the long term
The reasoning principle here is called uniformitarianism…it also refers to the more controversial claim that processes operating in the present can account, by extrapolation over long periods, for the evolution of Earth and life…This principle is not peculiar to evolution. It is used in all historic geology (54).The problem, here, is that Ridley is attacking a theological position (creationism) while remaining ignorant of the underlying theology. Christian theology distinguishes between creation, miracle, and providence.
Someone who permits uniformitarian extrapolation only up to a certain point in this continuum will inevitably be making an arbitrary decision (54).
His appeal to uniformitarian processes roughly corresponds to ordinary providence. And a Christian has no problem with a doctrine of providence.
However, providence allows for miraculous or creative events which fall outside the scope of providence, and there is nothing arbitrary about that discontinuity. For one thing, there could be no providence apart from creation.
For another thing, providence is not some impersonal process that merely operates according to inanimate forces and mechanical causes. Rather, there is a rational agent who lies behind the course of providence. Means are adjusted to ends. There is room for rational discretion.
Further study erodes the impression away. The fossil record contains a continuous set of intermediates between the mammals and reptiles, and these fossils destroy the impression that "mammals" are a discrete type. Archaeopteryx des the same for the bird type, and there are many further example (55).i) Archaeopteryx is another controversial example.3
ii) Whether there is a continuous set of intermediates is a contentious claim. Ridley is substituting a question-begging assertion for an argument:
a) Are these evolutionary intermediates or ecological intermediates?
b) How does one establish lines of descent? Consider some of Ridley’s damning admissions:
The phenetic and phylogenetic principles are the two fundamental types of biological classification, but three schools of though exist about how classification should be carried out (474)How can one establish the existence of an evolutionary intermediate if you can’t agree on their lineage?
The modern forms of phenetic classification are numerical and multivariate, and they were developed in reaction to the uncertainties and imprecision of evolutionary classification…a classification of a group based on its phylogeny is liable to be unstable…For many groups of living things, hardly anything is known about phylogeny, and a "phylogenetic" classification of such a group will inevitably be poorly supported by evidence (476).
This is a universal problem, not just a peculiar problem in this example. A taxonomist working with one sample of characters will often produced a different classification from another taxonomist working with a different set of characters (476).
So there is a degree of subjectivity in the phenetic philosophy…Moreover, the choice of cluster statistic is not the only subjective choice in phenetic classification. The measurement of distance poses an analogous problem (478).
Phylogenetic classification groups species solely according to recency of common ancestry (479).4
Cladistic classification has the advantage of objectivity…In practice, the inference of ancestral relations (that is, the phylogeny in figure 16.3a) can be difficult, and cladistics classification can be uncertain (480).
The cladistics classification of the tetrapods can seem odd. The Reptilia were recognized in almost every formal classification before cladism; but cladism rules them out…the category "fish" (containing the lungfish and salmon, but excluding the cow) does not exist in a cladistic classification (482).
The beauty of a purely phylogenetic classification is that there can be no doubt what the branching relations of the classificatory groups are (Figure 16.3). But if taxonomists defined some relations phonetically and others phylogenetically, it is no longer possible to say what any particular relation means. The branching relations are obscured and lost (483).
The main advantage of phylogenetic classification is theoretical. It can run into many problems in practice. One is ignorance. We do not know the phylogenetic relations of many living creatures, and cannot classify them phylogenetically. Another problem is instability (483).
Evolutionary taxonomists disagree with phenetic classification for much the same reasons we discussed above, though they express the argument differently (485).
If we take any two living species, they will show some similarities in appearance. Here we need to distinguish two sorts of similarities: homologous and analogous similarity (55).And there’s no reason why they shouldn’t. If five digits get the job done, then why should we expect a three- or seven- or 12-digited limb? The argument cuts both ways. Since they don’t need any particular number of digits, they don’t need a fixed number, but by the same token they don’t need a variable number.
The non-evolutionary usage is needed here in order to avoid circular argument: evolutionary concepts cannot be used as evidence for evolution (55n1).
An analogous similarity, in this non-evolutionary, pre-Darwinian sense, is one that can be explained by a shared way of life. Sharks, dolphins, and whales all have a hydrodynamic shape which can be explained by their habitat of swimming through water. Their similar shape is analogous; it is a functional requirement. Likewise, the wigs of insects, birds, and bats are all needed for flying: they too are analogous structures (55).
Other similarities between species are less easily explained by functional needs. The pentadactyl (five digit) limb of tetrapods is a classic example…Tetrapods occupy a wide variety of environments, and use their limbs for many differing functions. There is no clear functional or environmental reason why all of them should need a five-digit, rather than a three- or seven- or 12-digit limb (55).
Something doesn’t need to be a functional requirement to be functional. And as long as it is functional, there is likewise no requirement that it be diverse for diversity’s sake. It is unnecessary to either use the same design or use a different design.
Living creatures show similarities that would not be expected if they had independent origins (55).This is ambiguous. According to creationism, the natural kinds originated independently of one another, but they share a common Creator. As such, it comes as no surprise if they also share a common design—as long as that design is functional.
And yet they all do; or, rather, all modern tetrapods do—fossil tetrapods are known from the time in the Devonian when tetrapods were evolving from fish that have six-, seven-, and eight-digited limbs (55).i) But this argument undercuts the previous argument.
ii) Moreover, if you’re going to invoke common ancestry, then wouldn’t we expect our extant, pentadactylic tetrapods to descend from a five-digited ancestor? His argument is tugging in opposite directions.
iii) Furthermore, he’s done nothing here to establish the evolution of tetrapods (i.e. amphibians, reptiles, birds, and mammals) from fish.
Some modern tetrapods, in the adult form, do not appear to have five-digit limbs (Figure 3.6). The wings of birds and bats are in different ways supported by less than five digits, and the limbs of horses and of some lizards also have less than five digits (55).Once again, this undercuts the original argument. He initially argued for common descent from the fact that all tetrapods are pentadactylic—even though that is not a design requirement. Yet this is the second time he has had to mention a major exception to his argument. These qualifications undermine the force of the original argument.
Why does he say all when he only means some? Is this a tactic on his part? Lead with an overstatement that makes your case look stronger than it is. Hoping that first impressions will be more memorable than the qualifications you throw in further down the line?
However, all these limbs develop embryologically from five-digited precursor stages, showing that they are fundamentally pentadactyl (55).But the principle of recapitulation is dubious.5
What could be more curious than that the hand of man formed for grasping, that of a mole, for digging, the leg of a horse, the paddle of a porpoise and the wing of a bat, should be all constructed on the same pattern and should include similar bones and in the same relative positions? (56).i) How is that curious? Why isn’t such unified versatility a mark of elegant, economical design? A triumph of the one over many—where just one homologous plan can be deployed to solve so many different problems. A splendid piece of cost-effective engineering. Easily reproducible in a wide variety of organic forms.
ii) Notice that Ridley isn’t offering us a working model of a more efficient limb structure in any particular case. He has nothing better, or even as good, to offer us.
iii) Suppose this were not the case? Suppose that every natural kind had a completely different design?
It’s easy to imagine a Darwinian cite that as evidence against common design. He would exclaim that each species was the fortuitous and idiosyncratic result of its immediate circumstances, with no overarching intelligence to coordinate or consolidate the outcome.
If species have descended from common ancestors, homologies make sense; but if all species originated separately, it is difficult to understand why they should share homologous similarities (57).Except that he tries to play both sides of this argument, For, as we already saw, he also says that tetrapods are descended from ancestors with five-, six-, seven-, and eight-digited limbs.
So we end up with a tautology: homologies prove common descent—except when they don’t.
Without evolution, there is nothing forcing tetrapods all to have pentadactyl limbs (57).i) Once again, we see him appeal to all tetrapods, even though he admitted that this is a sweeping overstatement.
ii) Why must something "force" all tetrapods to be pentadactylic? From a theistic standpoint, which is what he is opposing, the choice is not between necessity and fortuity, but between teleology and dysteleology.
Why should the genetic code be universal? Two explanations are possible: that the universality results from a chemical constraint, or that the code is a historic accident (57).i) Once more, he is foisting a false dichotomy on the reader. There is another explanation: simplicity. Doesn’t science favor parsimonious solutions, such as the least action principle?
ii) Another problem running throughout this discussion is his tacit appeal to teleological explanations, such as whether a particular pattern is "functionally necessary."
But that is quite anthropomorphic from a Darwinian perspective. Naturalistic evolution has no room for teleological explanations. The evolutionary mechanisms were never goal-oriented in the first place. Ridley is sneaking criteria into his evaluation of the evolutionary process that are forbidden by the evolutionary process. Ironically, Ridley is having to assume a God’s-eye view in the premise to deny a God’s-eye view in the conclusion.
iii) Of course, we’re assuming, for the sake of argument, that the genetic code is, indeed, universal. But is that the case?
Biologists have known for years that some bacteria, algae and single-celled animals do not have the same genetic code as most other organisms. Darwinists claim that the exceptions are unimportant, since they "know" that the aberrant organisms are descended from organisms that had the standard code. But the code itself was supposed to be the primary evidence for such descent, and no comparable evidence is offered to replace it. Clearly, the Darwinists' "knowledge" in this case is philosophical rather than empirical.6
An organ that is described as vestigial may not be functionless…fossil whales called Basilosaurus, living 40 million years ago, had functional pelvic bones (Gingerich et al. 1990) and may have used them when copulating; and the vestigial pelvis of modern whales arguably is still needed to support the reproductive organs. However, that possibility does not count against the argument from homology; why, if whales originated independently of other tetrapods, should whales use bones that are adapted for limb articulation in order to support their reproductive organs? If they were truly independent, some other support would likely be used (60).This inference suffers from the same fallacies we’ve already drawn attention to.
i)"Independent" in relation to what? To each other? Yes. In relation to God? No.
All creatures are dependent on God for their common origin. And that is consistent with a common plan.
Remember that at the very outset of his discussion, Ridley said he was going to provide evidence distinguishing Darwinism from creationism. But all he’s supplied us with is evidence which is, at best, consistent with either position.
ii) He even admits that these "vestigial" organs are functional.
Incidentally, creationism doesn’t even deny that some organs may degenerate.7
iii) He also assumes what he needs to prove when he says that whales use bones that are adapted for limb articulation in order to support their reproductive organs.
Observe that this is not, in fact, evidence for evolution. To the contrary, it takes for granted an evolutionary explanation of what these bones were originally for ("limb articulation"), and then assumes that they were later co-opted to perform a different task.
The next section goes under the heading of Different homologies are correlated, and can be hierarchically classified
His discussion of amino acid sequencing seems to be a variant on his prior discussion of DNA. Been there, done that.
His next section goes under the heading of Fossil evidence exists for the transformation of species
The diatoms [singled-celled, photosynthetic organisms that float in the plankton] in Figure 3.11 show that the fossil record can be complete enough to reveal the origin of new species; but examples as good as this are rare. In other cases, the fossil record is less complete and there are large gaps between successive samples (Section 21.4, p602). There is then only less direct evidence of smooth transitions between species. The gaps are usually long, however (maybe 25,000 years in a good case, and millions of years in less complete records) (64).i) How does he define a new species? We’ve already seen that the definition of what constitutes a species is a controverted point in contemporary biology.
ii) It is oxymoronic to speak of "less direct evidence of smooth transitions between species" when "the gaps are usually long." You can’t have evidence for smooth transitions if there are long gaps in the evidence. And absent such evidence, why assume that there are transitional forms in the first place?
His next section goes under the heading of The order of the main groups in the fossil record suggests they have evolutionary relationships
The deduction follows from the observation that an amphibian, such as a frog, or a reptile, such as an alligator, is intermediate in form between a fish and a mammal. Amphibians, for instance have fills as fish do, but have four legs, like reptiles and mammals, and not fins…The forms of modern vertebrates alone, therefore, enable us to deduce the order in which they evolved (65).Is that a fact? Isn’t this a rather obvious case of ecological intermediates rather than evolutionary intermediates? They share some features in common with aquatic species as well as terrestrial species because they are semiaquatic species. They were designed to survive and flourish in that particular habitat.
Ridley drew this very distinction only ten pages earlier:
An analogous similarity, in this non-evolutionary, pre-Darwinian sense, is one that can be explained by a shared way of life. Sharks, dolphins, and whales all have a hydrodynamic shape which can be explained by their habitat of swimming through water. Their similar shape is analogous; it is a functional requirement. Likewise, the wigs of insects, birds, and bats are all needed for flying: they too are analogous structures (55).So why is he deducing evolution from features which are arguably analogous rather than homologous? Can’t he remember his own argument?
The inference, from the modern forms, can be tested against the fossil record. The fossil record supports it: fish, amphibians, reptiles, and mammals appear in the fossil record in the same order as they should have evolved (Figure 3.12b). The fit is good evidence for evolution, because if fish, amphibians, reptiles, and mammals had been separately created, we should not expect them to appear in the fossil record in the exact order of their apparent evolution. Fish, frogs, lizards, and rats would probably appear as fossils in some order, if they did not appear at the same time; but there is no reason to suppose they would appear in one order rather than another (66).i) But is this an evolutionary order, or an ecological order? If you were to suddenly fossilize a cross section of the ecosystem, land animals would be on top, fish would be on the bottom, and semiaquatic species would be sandwiched in the middle.
I’m not saying that that’s what happened. I’m merely saying that there’s nothing apparently evolutionary in the fossil sequence, as he describes it. He will need a subsidiary argument to turn the fossil sequence into an evolutionary sequence.
ii) Let us also remember what he said about the cladist taxonomy on the classification of fish and reptiles (see above).
His next section goes under the heading of Creationism offers no explanation of adaptation
Living things are well designed, in innumerable respects, for life in their natural environments. They have sensory systems to find their way around, feeding systems to catch and digest food, and nervous systems to coordinate their actions (67).i) This is very commonsensical. Unfortunately for him, Ridley is once again resorting to teleological explanations. And he is using this appeal to disprove creationism.
Ironically, creationism allows for teleology while naturalistic evolution disallows teleology. If teleology is true, then naturalistic evolution is false.8
Ridley is pilfering a category from creationism to disprove creationism. The fact that he cannot explain adaptation without invoking teleological criteria is self-refuting.
ii) He also fails to explain how biological organisms could survive and reproduce at the very time that they are in process of adapting to their natural surroundings.
The theory of evolution has a mechanical, scientific theory for adaptation: natural selection (67).To my knowledge, creationism doesn’t deny natural selection. What it denies, rather, is that natural selection is a mechanism for macroevolution.
Creationism, by contrast, has no explanation for adaptation. When each species originated, it must have already been equipped with adaptations for life, because the theory holds that species were fixed in form after their origin (67).i) He keeps acting as if special creation implies the static character of natural kinds.
ii) Creationism allows natural kinds to adjust to new habitat. They have a certain built-in adaptability. For example, the Bible believes in the common ancestry of all men from Adam. At the same time, it is always aware of racial diversity, which is a climatic adaptation.
His final section goes under the heading of Modern "scientific creationism" is scientifically untenable
Scientific arguments only employ observations that anybody can make, as distinct from private revelations, and consider only natural, as distinct from supernatural causes (68).i) The Bible is a public revelation, not a private revelation. Anyone can read the Bible.
ii) Science is by no means limited to "observational" data—especially regarding the origin of world or life on earth. Science theorizes about things that are too small to observe, too distant in space to observe, and too distant in time to observe.
Indeed, two good criteria to distinguish scientific from religious arguments are whether the theory invokes only natural causes, or needs supernatural causes too, and whether the evidence is publicly observable or requires some sort of faith (68).i) But suppose that God really is responsible for the origin of the world, as well as life on earth. If one stipulates in advance that only natural causes are permissible, then the "scientific" explanation will be dead wrong. What’s the value of a scientific explanation if you summarily exclude what may be the only right answer?
The method is getting ahead of the subject matter and prejudging the answer. Is that supposed to be scientific? If so, then so much the worse for science.
Is the scientific method an end in itself, or a quest for the truth? Are we naturalistic for the sake of naturalism? But what if naturalism is false?
How can science correct itself if science immunizes itself to contrary evidence? If science is unfalsifiable, because a naturalistic explanation cannot be falsified by a supernatural fact, then science is an inward-looking rather than outward-looking discipline. A self-contained discipline, like a fictional story.
ii) It also depends on what he means by "faith." Science relies on a number of metascientific assumptions about the world. What conceptual scheme is underwriting these assumptions?
Without these two conditions, there are no constraints on the argument. It is, in the end, impossible to show that species were not created by God and have remained fixed in form, because to God (as a supernatural agent) everything is permitted (68).i) Ridley seems to be operating with some form of theological voluntarism. Why does he identify creationism with voluntarism?
ii) Suppose, once more, that God is, indeed, the Creator of the world? That he did, in fact, make the natural kinds by his immediate, creative fiat.
Why is Ridley ruling out an explanation for the origin of life that may be the true explanation for the origin of life? He doesn’t seem to think that this explanation is impossible, or even improbable.
If anything, his objection appears to be that it cannot be quantified at all. But he doesn’t say why he thinks that. Why could there be no evidence for divine creation?
iii) Suppose we have a revelation from the Creator of the world in which he tells us what he did. In that event, we are not left guessing.
It cannot be shown that the building (or garden) you are in, and the chair you are sitting on, were not created supernaturally by God 10 seconds ago from nothing—at the time, He would also have to have adjusted your memory and those of all other observes, but a supernatural agent can do that. That is why supernatural agents have no place in science (68).If he wants to play that game, then it’s child’s play to construct naturalist versions of global illusions and delusions. Instead of Cartesian demons, we have alien telepaths.
A final problem with this chapter is that Ridley only gives one half of the story. He gives his arguments for evolution, and his arguments opposing creationism. But he doesn’t give the arguments for creationism, or the arguments opposing evolution.
So the reader is left with a very skewed presentation, as if the best a creationist could hope for is to come up with ad hoc explanations that deflect his criticisms and thereby show that creationism is merely consistent with the same data.
It puts the creationist on the defensive from start to finish, as if the onus is on him to answer all these charges, with no positive evidence for his own position, or positive evidence against the opposing position.
But let’s recall how it was that Ridley chose to frame the issue:
We concentrate here on evidence that can be used to test between the three theories in Figure 3.1 [evolution, transformism, and creationism] (45).Has Ridley succeeding in discharging his own burden of proof? Not in the slightest. Creationism emerges without a scratch—in large part because he mischaracterized creationism.
1 M. Ridley, Evolution: Third Edition (Blackwell 2004).
2 I’d add that this textbook example has come under fire. Cf. J. Wells, Icons of Evolution (Regnery 2000).
4 Notice how this takes macroevolution for granted. Common ancestry is a presupposition of phylogenetic classification.
5 Cf. S. Gould, Ontogeny and Phylogeny (Belknap 1977), 501; http://www.arn.org/docs/nelson/pn_darwinianparadigm061593.htm
8 "Teleological Explanation," W. H. Newton-Smith, ed. A Companion to the Philosophy of Science (Blackwell 2001), 492-494.