As a sort of warm-up for my return to the world of analytical atheology, I have been looking for something, relevant to the atheism-Christianity dialogue/debate, something that I could rip into shreds. I typed “atheism” into Google, and the first thing that came to me was the description blurb for CARM’s page on atheism: “Proves that atheism is invalid. Also provides tips on how atheists and Christians can communicate.” Note to Matt Slick: if you’re going to lead off with the former, then I wouldn’t exactly qualify you as an expert on the latter. Especially if you suck at the former as much as the writings on your page suggest — but that’s a topic for another article, as your page is too much to take on in one go.
In this article I shall focus on Robin Collins’ recent defense of that old apologetic canard, the fine-tuning argument. Collins derives his conclusion, that theism is more probable than atheism (the single-world version of it — he addresses the many-worlds hypothesis elsewhere but I won’t go there right now) from the following two premises:
1. The fine-tuning’s existence is not improbable under theism.
2. The fine-tuning’s existence is improbable under atheism.
The first thing that occurred to me is that Collins forgot a premise, one more basic than, and certainly essential to, the above two:
0. The fine-tuning exists.
If by “fine-tuning” Collins means the state of the universe having been fine-tuned like a series of radio dials to achieve exactly the right settings, then this premise begs the question in a number of ways, not the least of which it assumes the existence of a fine tuner. However, if by “fine-tuning” Collins simply means the variables that allow us to live our lives without blowing up like a balloon, or whatever, then this premise is defensible, and even obvious enough to go unstated. But in that case, to avoid misleading the reader, it should be phrased in such a way that does not imply the question-begging meaning. For example,
0′. The universal variables are life-friendly.
Which means that the other premises would read,
1′. The universal variables being life-friendly is not improbable under theism.
2′. The universal variables being life-friendly is improbable under atheism.
This seems to me to be a clearer wording, and does not significantly alter the analytic force of the argument, although it does rob it of some of its rhetorical force. Since Collins has anticipated most of the standard objections to the fine-tuning argument, including the ones that immediately came to my mind, I shall here focus on his responses to three of those objections, the ones I personally use most often:
The variables could be the result of a more fundamental law, or a logical necessity. Collins responds that such a fundamental law is entirely speculative, and only pushes back the problem since the more fundamental law would itself be improbable, even if it were regarded as logically necessary.
With respect to the first response, Collins appears to have not done his homework. He seems unaware of the scientific work on the fundamental law theory — particularly Quentin Smith’s commentaries on the Hartle-Hawking theory, which shows that this theory, though unproven, is certainly much better grounded, scientifically, than the phrase “entirely speculative” would imply.
With respect to the second, I submit that Collins, like myself, is not in a position to make any determination regarding the logical necessity of the fundamental constant(s) of the universe. I cannot stay with any authority or knowledge that it is probable, Collins cannot say that it is improbable. Any assignation we make to the probability would be as good as any other. Although this places us in the same epistemological boat, I am not the one making an argument whose premises depend on the probability tipping toward one end of the spectrum rather than the other. Collins is. Unless he can resolve this problem — and there is no reason to think that the problem is resolvable, even in principle — premise 2 will have a universe-sized hole in it.
Other forms of life than human are possible. Collins responds that the fine-tuning proponent doesn’t necessarily assume otherwise, and that this objection misses the point; the point is that “the life-permitting range for the values of the constants of physics (such as gravity) is small compared to the surrounding range of non-life-permitting values.”
Let’s assume for a moment that Collins has managed to figure out exactly what the life-permitting range for the physical constant values is. The question is, how does he know what the surrounding range is? For example, let’s say the life-permitting range for gravity is 0.8g to 1.2g, where g is the present value. What is the surrounding range? Collins would no doubt like for it to be zero to (infinity)g, but why should we assume that? Just because that’s the range of real numbers?
Perhaps (and here I am getting into a different objection framed by Collins, but they are similar enough to consider together) it has more in common with a randomized throw of a pair of dice, the range of the results of which is between two and twelve pips, not one and infinity, and in which one particular result, seven, is more likely than all the others. I have no reason to think that that is the case, but Collins has given no reason to think otherwise other than the probabilistic “principle of indifference,” under which when we have no reason to prefer one parameter over another we should assign them equal probabilities. But again, Collins hasn’t done his homework; as pointed out above, we do have reason to favor this set of parameters over another.
One more point: if Collins assumes that the surrounding range of possible physical constants is infinite — which, given his defense of the probabilistic principle of indifference, I believe he must — then he faces this dilemma: the life-permitting range is either infinite or finite. If infinite, then there is no possible set of physical values that prohibits life, and premise 2 is false. If finite, then the point that the life-permitting range is smaller than the surrounding range is tautologically true, since any finite range is necessarily smaller than infinity. Non-tautologies cannot be derived from tautologies. Therefore, the point that Collins makes cannot support premise 2, and that premise is unsupported.
God is more complex than the universe, so who designed God? Collins responds that the fine-tuning argument still disproves atheism even if this objection is granted: his argument, he states, “only required that the fine-tuning be more probable under theism than under the atheistic single-universe hypothesis. But this requirement is still met even if God exhibits tremendous internal complexity, far exceeding that of the universe. Thus, even if we were to grant the atheist assumption that the designer of an artifact must be as complex as the artifact, the fine-tuning would still give us strong reasons to prefer theism over the atheistic single-universe hypothesis.”
First of all, it bears noting that the force of this objection is not necessarily related to complexity. Rather, it is related to the fact that, whatever quality is possessed by intelligent life in the universe that calls for an explanation for intelligent life in the universe — whether it’s complexity, or self-awareness, or whatever; call it the E factor — God has it too. Since God has the E factor, God calls for an explanation to the same extent as intelligent life in the universe.
Collins may be correct that the requirement of the fine-tuning argument, that the fine-tuning is more likely under theism than atheism, is still met if God possesses the E factor, but as it turns out, it doesn’t matter, because conceding that God has the E factor — and I can think of no definition of the E factor such that intelligent life would have it but God would not — defeats the purpose of the argument.
This is true for two reasons. First if theism is true, then the theistic God exists. If the theistic God exists, then God cannot have an explanation outside of himself. If God cannot have an explanation outside of himself, then God cannot possess the E factor. Therefore, to concede that God has the E factor is to set of a string of modus tolens conclusions, like lighting a string of firecrackers, at the end of which the fine tuning argument explodes. Theism cannot be more probable than atheism if the probability of theism is zero.
Second, if God is the best explanation for intelligent life in the universe, and God calls for an explanation, then whatever entity explains God would most likely have the E factor itself, and would therefore itself call for an explanation, and so on. Thus we have an infinite regress, and the prior probability of theism is again much lower.
As if realizing these problems, Collins quickly brings up another response: that God isn’t actually complex, and that Medieval theologians were actually correct when they said that God was absolutely simple. Even if complexity is indeed the E factor, this response is just plain silly; an entity having set S of characteristics plus the added characteristic of being concerned about every human’s sex life is obviously more complex than an entity having set S alone.
In addition to the three objections above, there is one objection I make which Collins failed to address: although the universal constants do allow intelligent life, an objective observer looking at the entire universe could hardly conclude that humans are the intended product of any fine-tuning. Given the many orders of magnitude by which humans are outnumbered by stars, or asteroids, or even bacteria, one could easily assume that the universe was fine-tuned for them to flourish, in which case humans would be an unintended by-product. Therefore, one cannot infer that the fine tuner has the concern for humans that the theistic God must have, and premise 1, that the fine-tuning is not improbable under theism, loses any support it might have had.
As I stated in the outset, this paper is a sort of warm-up for my return to the realm of analytic philosophy of religion; a shaking off of the dust and clearing of the cobwebs. Accordingly, this work might be sub-standard. Any constructive criticism or comments would be most welcome.