The fine-tuning argument (as told by William Lane Craig)

The next argument Craig presents in Reasonable Faith (after his versions of the cosmological argument) is the fine-tuning argument, which is a version of the argument from design or teleological argument. I’ve previously discussed problems with other versions of this argument here and here, those posts will be useful background to this one. Craig’s version is a tidy deductive argument:

  1. The fine-tuning of the universe is due to either physical necessity, chance, or design.
  2. It is not due to physical necessity or chance.
  3. Therefore, it is due to design.

This argument fails at every step. First, Craig seems to have gotten the idea for setting the argument up as a deductive argument (rather than a probabilistic argument, the way Robin Collins does) from anti-evolutionist and “Intelligent Design theorist” William Dembski, who claims to have come up with a general procedure for demonstrating that something is designed.

Dembski’s ideas, however, are not generally accepted in the scientific community. He has claimed his ideas describe how scientists in a huge range of fields, including forensics and archaeology, work. However, many scientists, including researchers at SETI (the Search for Extraterrestrial Intelligence, which Dembski likes to invoke) have explicitly rejected those claims.

Craig and Dembski are also fellows at the same anti-evolution think tank, the Discovery Institute. Yet Craig generally avoids talking about evolution. All this is worth mentioning because I think much of Craig’s appeal is not that his positions are more reasonable than those of other fundamentalists, but because he’s good at avoiding certain topics when talking about them would be inconvenient (and evolution isn’t even the worst example of this).

Now in Reasonable Faith, Craig also justifies premise (1) of his argument by claiming it “seems to exhaust all the alternatives.” It’s easy to poke holes in that claim, though. For one thing, religious believers do not generally think God is due to either physical necessity, chance, or design. Putting a requirement on the laws of physics but not God, without giving any reason for doing this, is a straightforward example privileging the hypothesis.

It’s also not clear if Craig has bothered to consider explanations that involve physical laws without being a simple matter of “it’s a physical necessity.” For example, the laws of physics help explain why the Earth orbits the Sun, but it would be odd to say it’s a physical necessity that the Earth orbits the Sun. Or: evolution by natural selection relies on both random processes (mutation) and fairly predictable processes (selection). Failing to understand that is a common mistake of creationists.

Craig’s arguments for premise (2) are just as bad. Craig dismisses the possibility of a physical explanation for alleged fine-tuning, saying it “requires strong proof. But there is none” (p. 162). But as I already pointed out when discussing Craig’s Kalam argument, it makes no sense to dismiss alternatives to God for lack of evidence, and then conclude God must exist. It would make just as much sense to dismiss God for lack of evidence and then conclude some alternative theory must be right.

Craig discusses the many worlds hypothesis, but the discussion badly misrepresents the relevant science (in this case, I’m not sure whether through ignorance or dishonesty). Craig gives the impression that the many worlds hypothesis is something atheistic scientists made up to avoid having to admit God exists, which simply isn’t true.

It’s possible Craig disagrees with the reasons physicists give for entertaining the many worlds hypothesis, but it’s wrong for him to give the impression they have no other reasons for their views. I’ve looked at the main book Craig cites as a source for the many worlds hypothesis, Alexander Vilenkin’s Many Worlds in One, and Vilenkin doesn’t seem too worried about the possibility that “God did it!” is the right explanation. He spends as much time on Hindu mythology as any supernatural hypothesis.

This, by the way, is yet another example of privileging the hypothesis: assume all of Professor McInfidel’s work is driven by a hatred of God, so if you show he’s wrong about anything, it’s a victory for your religion, even if the two issues have nothing logically to do with each other.

Craig has a second criticism of the many worlds hypothesis: “if the Many Worlds Hypothesis is to commend itself as a plausible hypothesis, then some plausible mechanism for generating the many worlds needs to be explained.” To which I reply, “If the God Hypothesis is to commend itself as a plausible hypothesis, then some plausible mechanism for generating the god must be explained.” (Okay, that’s somewhat tongue-in-cheek, but you should be able to fill in the rest by now.)

Oh, and finally, like so many defenders of the argument from design, Craig doesn’t even try to give a reason why the designer must be God.

(Note: I’ve aimed to be concise in this post, though it’s based on a longer post I wrote at my old blog. You can go there if you want a somewhat more detailed version of what I’ve said here.)

Kris Komarnitsky’s Doubting Jesus’ Resurrection
Slavery abolition and animal rights: the biggest problem
Harry Potter and the problem with genre deconstructions
Why I’ve decided to start deleting jerky comments more often
  • mikmik

    Craig has a second criticism of the many worlds hypothesis: “if the Many Worlds Hypothesis is to commend itself as a plausible hypothesis, then some plausible mechanism for generating the many worlds needs to be explained.”

    That’s my thoughts on the ontological argument where he goes from “if all worlds are possible” to “god exists in one possible world.”
    Same thing, where did God come from? Further to that, he goes from possibly exists, to DOES exist in a possible world. Just imagining God could possibly exist is the same as saying He absolutely has existence. No mention of everything else that might possibly exist, like you say above.

    PS This is an excellent idea, and series of articles. I’m living in a deeply Christian environment and they think Craig is the end all to the question of God. Funny thing, they want to debate me, as if that is what determines the facts by whoever wins. Another example of privileging the hypothesis. If they ‘win’, it proves they are right, if I win, they didn’t present all the relevant arguments.

    • Chris Hallquist

      Well, not exactly. Unless you’re David Lewis, possible worlds (whatever they may be) aren’t anything like the worlds being discussed in reference to the fine-tuning argument.

  • MNb0

    “Dembski’s ideas are not generally accepted in the scientific community.”
    That’s the understatement of the year.

    You neglected one problem with fine-tuning, even though you hinted at it. The argument assumes that the physical constants are adapted to human intelligence. That’s a travesty of causality, as I pointed out before. It assumes teleology, something modern science has thrown out of the window.
    Maybe you could write a bit more about teleology.

  • Amadan

    On what basis do people maintain that the Universe is “fine-tuned”?

    Given that n=1, how exactly have they calculated the probabilities?

    • sqlrob

      There was an article recently (Stenger? Not sure) that no, the universe is not fine tuned. There are other, more optimal choices for the constants.

  • James Sweet

    I’m not sure if this is Craig’s error or yours (or maybe my misunderstanding), but it seems you are using “Many worlds” where “multiverse” would be more appropriate. It’s true that some physicists are now saying that they stem from the same underlying physical phenomenon, but the concepts are distinct and we don’t know for sure if they can be reduced to the same thing yet.

    Since you don’t explain what you/Craig mean by Many Worlds, so it’s possible the confusion is mine… but generally, the multiverse hypothesis is the one that is thought to have the most utility in addressing the fine-tuning “problem”. (Scare quotes because it’s only an apparent problem; we don’t really know yet if it even needs solving, or if it’s a case of mere puddle logic or something like that)

    • Chris Hallquist

      That crossed my mind, sort of. I was calling it “multiverse” as I typed the post, then I noticed Craig was calling it “many worlds” so I switched my usage to match his.

      Meh. Probably I should fix it when I compile these posts into The Book.

      • James Sweet

        Okay, yeah, I think Craig is the one who has made the error there, and I even think I remember reading someone else point it out at some point in the past.

        • Chris Hallquist

          Not in the places I’ve seen, and even if he did slip up once, it’s a case where he obviously meant to say something else.

  • James Sweet

    Craig has a second criticism of the many worlds hypothesis: “if the Many Worlds Hypothesis is to commend itself as a plausible hypothesis, then some plausible mechanism for generating the many worlds needs to be explained.” To which I reply, “If the God Hypothesis is to commend itself as a plausible hypothesis, then some plausible mechanism for generating the god must be explained.”

    It’s worse than that, because there are many plausible hypotheses for what mechanism might generate multiple universes. We don’t know which one (if any) is correct, but it’s disingenuous to the extreme to pretend that physicists don’t have any plausible ideas here. I alreadly linked to Sean Carroll once this morning, so I may as well mention the excellent From Eternity to Here, where in the final chapters he describes in layman-accessible terms a plausible mechanism that would not only explain the existence of many universes, but would also go a long ways towards addressing Craig’s shenanigans about time/eternity/etc. Carroll’s model allows for an eternal effectively-timeless* multiverse in which new bubble universes are continuously springing into existence in a very low state of entropy.

    * What I mean here by “timeless” is that time is not a separate phenomenon requiring explanation, but rather a natural consequence of the laws of statistical mechanics.

    • miller

      Yeah, it’s ridiculous how wrong Craig got that one. Physicists are not proposing that there is a multiverse, and then struggling to come up with mechanisms. Rather, physicists analyze other theories, and sometimes multiverses naturally pop out as interesting consequences of these theories. Every multiverse theory has a mechanism associated with it, because the mechanism came first.

  • J. Quinton
  • tonylloyd

    The first premise begs big time: what “fine tuning”?.

    For anything being just right for anything else to amount to fine tuning the anything else must be intended. Shuffle a pack of cards and deal a poker hand out. Whatever hand appears, say As 10d 7d 7c 6d, the pack would have had to have been shuffled just right for exactly that hand, which is very, very improbable. But we know we’re going to get a hand, if we deal one. So we know we’re going to get a very, very improbable result.

    To classify as “fine tuning” the result needs to be, somehow “fixed” ahead of time. The way we do this with a pack of cards is to announce “I’m know going to deal a Royal Flush in spades”. Announce that, do it and it’s pretty conclusively proven that the pack was “fine tuned” (“stacked”).

    How can we “fix” the “way that the universe is”? Obviously that God intended it that way. And how do we know that God intended it that way? Because that’s the way it is. The argument becomes:

    1. This is how the universe is
    2. So this is how God intended it to be
    3. What God wants, God gets
    4. So this is how the universe is

  • Reginald Selkirk

    Yet Craig generally avoids talking about evolution.

    Generally, but he occasionally drops a stupid-bomb of a bad probability argument. Reference to a probability argument by Barrow and Tipler

  • Iain Walker

    “if the Many Worlds Hypothesis is to commend itself as a plausible hypothesis, then some plausible mechanism for generating the many worlds needs to be explained.” To which I reply, “If the God Hypothesis is to commend itself as a plausible hypothesis, then some plausible mechanism for generating the god must be explained.”

    Or, just to be going on with, how about a plausible mechanism by which God supposedly created the universe? “Free choice” isn’t a mechanism.

  • Bronze Dog

    I might be misremembering this, but several years ago on a science show about alternate universes, one physicist made the argument that a multiverse model is more parsimonious than a one universe model because the one universe model would have to posit some additional mechanism or law to restrict existence to just one universe.

    Just because we’ve only seen it once doesn’t mean it couldn’t have happened before, elsewhere, or can’t happen again.

    • MNb0

      Problem with the multiverse is that we can’t falsify the hypothesis, which means it’s unscientific.

      • miller

        I’m going to quote a Sciam article:

        Remember: Parallel universes are not a theory—they are predictions of certain theories.

        For a theory to be falsifiable, we need not be able to observe and test all its predictions, merely at least one of them.

  • josh

    I think one of the big, damning flaws of these kinds of arguments, although it’s not one that everyone wants to get into when there are other glaring errors, is the notion that ‘design’ is inherently distinct from physical necessity (or chance). Basically, everything we know about every mind we can examine indicates that a ‘design’ is a mental process that ultimately is identical with a necessary physical process (usual caveats about QM here).

    Really, how could it be otherwise? Your mind has a goal (which has to come from somewhere, cue evolution and psychology fundamentally built on physics), to accomplish that goal it designs something (a process dependent on the capabilities and materials the mind has as limitations, its not clear what meaning could attach to ‘design’ for an omnipotent, omniscient mind, but putting that aside…) which, if it is to accomplish that goal has to be logically related to it in various ways, that logic is founded upon the logic of physical laws and the physical arrangement of your brain. Even if we suppose that God’s mind is some totally other substance than our familiar electrons and whatnot, it would still have to be a logical mechanism to design anything and I can’t see any reason we would privilege it over the logical mechanisms of KNOWN, EVIDENCED physics that Craig wants to dismiss.

  • Kevin

    Craig has made another probability gaffe. He says that the chance is so low that it is not by chance. This should give you pause. If you buy a lottery ticket and end up winning, the explanation is not that the odds of you winning it fairly are low, therefore you cheated. Because of this, premise two is left unsupported.

  • eric

    The habitable volume of the solar system is about 1E-28%*. Put another way, 0.0000000000000000000000000001% of the solar system is tuned for life.

    If that’s “fine tuned,” I’ll eat my hat.

    If the universe is fine tuned at all, its tuned for critters that must love a temperature of 4K, pressure of 10E-11 Pa, and a pretty big flux of cosmic rays.

    [*back of the envelope calculation, arbitrarily going out to Pluto's orbit. As you consider more and more of the universe, the number gets smaller, even if you were to consider every solar system to have a planet in the habitable zone]

  • Skepticali

    Couldn’t premise 1 also be said to suffer from a false choice? I seem to perceive that a lot in arguments of any sort where someone wants to steer the conclusion in a pre-selected direction.

  • Annatar

    Things are the way they are.

    If they were different, that would be bad.

    Therefore, God exists.

  • bad Jim

    Why does he think the universe is finely tuned? Because it produced us, which is clearly the point of the entire universe. That’s what the argument boils down to. We exist, therefore God exists, because God’s job is to create us.

    This universe includes all sorts of shiny sparkly things like supernovas; why not concluded that God created the universe as it is because he likes the fireworks? But no, it’s always about us.

  • Robert Oerter

    Can’t resist pointing out that the fine-tuning argument against God is just as good (i.e. just as bad) as the fine-tuning argument for God.