Let’s begin reviewing the logical form of the argument, as described in Part 1 of this series.

(1) Evolution is antecedently much more probable on the assumption that naturalism is true than on the assumption that theism is true.

(2) The statement that pain and pleasure systematically connected to reproductive success is antecedently much more probable on the assumption that evolutionary naturalism is true than on the assumption that evolutionary theism is true.

(3) Therefore, evolution conjoined with this statement about pain and pleasure is antecedently very much more probable on the assumption that naturalism is true than on the assumption that theism is true. [From (1) and (2)]

(4) Naturalism is at least as plausible as theism.

(5) Therefore, other evidence held equal, naturalism is very much more probable than theism. [From (3) and (4)]

(6) Naturalism entails that theism is false.

(7) Therefore, other evidence held equal, it is highly probable that theism is false. [From (5) and (6)]

One objection to the evidential argument from biological evolution (ABE) is that premise (1) is false; evolution is evidence for theism. William Lane Craig, among others, has used precisely this objection in his debates with Frank Zindler, Massimo Pigliucci, and Paul Draper, among others. In his words:

In fact, that leads me to his other argument, concerning biological evolution. And I’m going to suggest that the idea that evolution could have occurred without an intelligent Designer is so improbable as to be fantastic. This has been demonstrated by Barrowand Tipler in their book The Anthropic Cosmological Principle. In this book, they list ten steps in the course of human evolution, each of which is so improbable that before it would have occurred the sun would have ceased to be a main sequence star and would have burned up the earth. They estimate the odds of the evolution of the human genome by chance to be on the order of 4-360 (110,000), a number which is so huge that to call it astronomical would be a wild understatement. In other words, if evolution did occur, it would have been a miracle, so that evolution is actually evidence for the existence of God![1]

Craig takes this to be an effective refutation of ABE.

Craig’s Objection Isn’t Even a Prima Facie Objection to ABE

But how? Look again at Draper’s formulation of the argument. If Craig’s objection is to be even relevant to ABE, it needs to be an objection to premise (1). For convenience, here is that premise again:

(1) Evolution is antecedently much more probable on the assumption that naturalism is true than on the assumption that theism is true.

But, taken at face value, Craig’s argument is not even a prima facie good undercutting defeater for (1), much less the rebutting defeater he makes it out to be. Remember that we have stipulatively defined “evolution” as the hypothesis that (a) complex life evolved from simple life; and (b) all evolutionary change in populations of complex organisms is or is the result of trans-generational genetic change. In the passage just quoted, however, Craig, doesn’t even present an argument against evolution so defined. Rather, he objects that unguided evolution (in his words, “evolution … without an intelligent Designer”) is false. Thus, he is presenting an argument against Darwinism, a hypothesis which entails but is not entailed by evolution. But Draper, in his written work, explicitly states that ABE is an evidential argument from evolution, not an evidential argument from evolution conjoined with Darwinism. So Craig’s objection isn’t even a rebutting defeater for ABE.

Draper (or any other proponent of ABE) could, for the sake of argument, consistently agree with both ABE and Craig’s argument. In other words, one could consistently believe both “The fact of evolution is more probable on naturalism than on theism” and “Given that evolution is true, the fact that intelligent life evolved on Earth is more probable on theism than on naturalism.” So, again, Craig’s objection is, at best, an independent argument for theism, not a refutation of ABE.

Is There a Good Theistic Evidential Argument from Evolution?

Furthermore, a careful reading of Craig’s comments reveals that he isn’t arguing that “evolution” so defined is evidence for God’s existence. Rather, he argues the fact that humans evolved is evidence favoring theism over naturalism. I’m going to quote his comments again, this time emphasizing the references to human evolution.

In fact, that leads me to his other argument, concerning biological evolution. And I’m going to suggest that the idea that evolution could have occurred without an intelligent Designer is so improbable as to be fantastic. This has been demonstrated by Barrowand Tipler in their book The Anthropic Cosmological Principle. In this book, they list ten steps in the course of human evolution, each of which is so improbable that before it would have occurred the sun would have ceased to be a main sequence star and would have burned up the earth. They estimate the odds of the evolution of the human genome by chance to be on the order of 4-360 (110,000), a number which is so huge that to call it astronomical would be a wild understatement. In other words, if evolution did occur, it would have been a miracle, so that evolution is actually evidence for the existence of God![1]

So, despite his claims to the contrary, Craig’s theistic argument here is not an argument from evolution to theism. Rather, it’s an argument from human evolution to theism.

This raises an interesting possibility. Regular readers of this blog are familiar with the fallacy of understated evidence; Craig could argue that the naturalistic evidential argument from evolution commits the fallacy of understated evidence. It successfully identifies some fact (evolution) which is evidence favoring naturalism over theism, but ignores other, more specific evidence (human evolution) which favors theism over naturalism. Of course, that reply would require him to admit that evolution is evidence favoring naturalism. I will leave it as an exercise for the reader to determine if “human evolution” constitutes “understated evidence” which favors theism.

Is There a Good Theistic Evidential Argument from Human Evolution?

But does Craig’s theistic argument from human evolution even work? I believe the most charitable formulation of Craig’s argument begins with an inductive argument from authority.

(8) The vast majority of statements made by Barrow and Tipler concerning the probability of the Darwinist evolution of Homo sapiens on Earth are true.

(9) “Darwinist evolution of Homo sapiens on Earth is fantastically improbable” is a statement made by Barrow and Tipler.

(10) [probable] Therefore, “Darwinist evolution of Homo sapiens on Earth is fantastically improbable” is true. [by Statistical Syllogism]

An inductive argument from authority, however, is evidentially worthless if (a) equally qualified  authorities disagree; or (b) the authorities mentioned in the argument do not actually make the statement attributed to them. As I shall argue below, Craig’s argument from authority violates condition (b), viz., Barrow and Tipler (B&T) do not claim that the Darwinist evolution of Homo sapiens on Earth is fantastically improbable.

Barrow and Tipler’s “Ten Steps”

B&T’s discussion discussion is very sophisticated; it would probably take a 25- to 50-page essay to fully summarize and assess the intricacies of their sensitivity analysis and corresponding mathematical model. But, for our purposes, that isn’t necessary. Let’s begin by investigating Craig’s reference to B&T’s “ten steps in the course of human evolution.” Turning to their book, they describe those ten steps as “crucial steps,” which they summarize as follows.

Crucial Step #1: The development of the DNA-based genetic code.

Crucial Step #2: The invention of aerobic respiration.

Crucial Step #3: The invention of glucose fermentation to pyruvic acid is unique seme which evolved in bacteria and remained unmodified in all eukaryotes.

Crucial Step #4: The origin of autotropic photosynthesis (oxygenic photosynthesis).

Crucial Step #5: The origin of mitochondria: these are the bodies in the cytoplasm of eukaryotes wherein the energy molecule ATP is synthesized.

Crucial Step #6: The formation of the centriole/kinetosome/undulipodia complex; such an event was essential to the evolution of the reproductive system of eukaryotes and of nerve cells.

Crucial Step #7: The evolution of an eye precursor.

Crucial Step #8: The development of an endoskeleton.

Crucial Step #9: The development of chordates.

Crucial Step #10: The evolution of Homo Sapiens in the chordate lineage.[2]

B&T’s discussion is very nuanced; they are very careful to state that the “arguments for the above 10 steps as being crucial … are not conclusive by any means; they are offered as suggestions only…” [3] But I don’t propose holding that against them. For the record, as a non-biologist, I’m fine with granting the assumption that these ten steps are crucial in the sense they have in mind. Now what?

As I was reading B&T’s discussion, I expected them to estimate the probability of each step conditional upon the prior step. Assuming these ten steps are statistically independent (which seems right to a non-biologist such as myself), I then expected them to apply the chain rule in an equation like this:

Pr(evolution of Homo Sapiens) = Pr(Step #1) x Pr(Step #2 | Step #1) x Pr (Step #3 | Step #1 & Step #2) + ….

Instead,they take the discussion in a totally different direction and use the minimum number of “crucial steps” to estimate “an upper bound for the length of time the biosphere can continue in the future.”

So where does the probability estimate quoted by Craig come from? B&T attempt to estimate the probability of human evolution by focusing on the evolution of the unique set of proteins coded by the human genome. They do that by using (a) an estimate of the number of unique genes; and (b) an estimate for the “odds for assembling a single gene.” Regarding (a), Dobzhansky et al estimate that the human genome has approximately 110,000 different genes. As for (b), they cite DeLey, who estimated the probability to be between 4^(-180) and 4^(-360). B&T use (a) and (b) to arrive at the following result.

The odds against assembling the human genome spontaneously is even more enormous: the probability of assembling it is between 4^(-180)^110,000 … and (4^360)^110,000. These numbers give some feel for the unlikelihood of the species Homo sapiens.[4]

If you compare this quotation to the Craig quotation at the beginning of this post, however, you’ll notice that Craig mentions only the lower bound of this range without mentioning the upper bound or the fact that they estimated a range rather than a single number. But let that pass.

The important point is this. The word “spontaneously” is key; it reveals that B&T are not estimating the probability of the Darwinist evolution of human beings. Rather, they are estimating the probability of the “spontaneous” assembly of the human genome. As Craig himself admits, they are estimating the probability of the human genome arising by chance.  But evolution and especially Darwinist evolution denies–i.e., is logically incompatible with–the spontaneous emergence of the human genome. That entails, however, that premise (9) is false.

Furthermore, on the next page Barrow and Tipler seem to caution against precisely the sort of intelligent design inference which Craig makes.

We should emphasize again that the enormous improbability of the evolution of intelligent life in general and Homo sapiens in particular at any randomly chosen point in space-time does not mean we should be amazed we in particular exist here. This would make as much sense as the Elizabeth II being amazed she is Queen of England. Even though the probability of a given Briton being monarch is about 10^-8, someone must be. Only for the person who is monarch is it possible to ask, ‘how improbable is it that I should be monarch?’ Similarly, only if an intelligent species of a particular kind does evolve in a given space-time location is it possible for its members to ask how probable it was intelligent life of some form to evolve there. And, provided that the Universe is of a sort in which intelligent life is likely to arise somewhere, then both cases are examples of WAP self-selection in action.[5]

If Craig has an objection to this paragraph and specifically B&T’s use of the WAP (Weak Anthropic Principle), he doesn’t tell us what it is. But rather than pursue that line of thought, instead I want to close with this observation. It seems rather one-sided for Craig to appeal to part of B&T’s probability estimate, which supports his point, without also mentioning that B&T apparently endorse WAP, which undermines his point. The upshot is that Craig’s argument from authority fails and therefore his evidential argument from human evolution likewise fails. It’s possible that there may be a logically correct inductive argument from human evolution to theism, but this one clearly isn’t it.

Notes

[1] See, e.g., William Lane Craig, “First Rebutal,” Does God Exist? William Lane Craig vs. Massimo Pigliucci (1998), http://www.reasonablefaith.org/does-god-exist-the-craig-pigliucci-debate#section_3. Craig’s reference to Barrow and Tipler is John Barrow and Frank Tipler, The Anthropic Cosmological Principle (Oxford: Clarendon Press, 1986), 561-565.

[2] Barrow and Tipler 1986, 561-562.

[3] Barrow and Tipler 1986, 564.

[4] Barrow and Tipler 1986, 565.

[5] Barrow and Tipler 1986, 566, my emphasis.