I recently had my first debate in many years on a new website called DebateCraft. DebateCraft was put together by some of the old members of Debate.org where I used to debate and Debateart.com, where I rarely participated.
The person I debated wanted to defend the fine-tuning argument, and since I've become an advocate of this argument, I wanted to play devil's advocate. You see, most of the objections raised against the fine tuning argument are pretty terrible, so I wanted to test the argument against a handful of what I thought were the strongest objections.
My opponent was apparently playing devil's advocate, too, because he conceded the debate in the third round saying, "I never really thought this was a decent argument."
In spite of not completing the debate, I did learn something. I learned that whether the multiverse objection to fine-tuning is a good objection depends on how the fine-tuning argument is framed.
If you try to characterize the fine-tuning argument as an inference to the best explanation, then the debate comes down to whether God or the multiverse best explains fine-tuning. A person raising the multiverse objection would be obliged in that case to actually defend the existence of a mulitiverse or say why it's a better explanation than God.
But if you characterize the fine-tuning argument as a disjunctive argument then the objector doesn't have to defend the actual existence of the multiverse. They only have to raise the multiverse as a possibility.
William Lane Craig's version of the argument is disjunctive. The way a disjunctive argument works is that you start with all the possibilities. Then you rule out all but one, and that’s the conclusion. In Craig’s case, the options are chance, law, and design. He wants to argue for design, so he needs to rule out chance and law. This puts the burden of proof on him to rule out chance. All the multiverse objector has to do is keep chance as a live option and show that Craig has not ruled it out.
The multiverse objector does not have the burden of proving that there is a multiverse. All he has to do is explain why it’s a possibility. As long as it’s a possibility, chance has not been ruled out, and Craig’s disjunctive argument collapses.
It is going to be very hard to rule out a multiverse because we simply don’t know what’s beyond the event horizon of our region of space. We have no idea what lies beyond the observable universe, and that unknown prevents us from ruling out a multiverse which, in turn, prevents us from ruling out chance + the observer selection effect as the reason for why we find ourselves in a life-permitting universe.
Craig's attempts at responding to the multiverse objection miss the mark. One argument he likes to use is his card game illustration. If somebody were accused of cheating because every time they dealt, they got all aces, they might defend themselves by saying, "Well, in this great big multiverse, there are universes in which I do get all aces when I deal." That arugment fails because there's no observer selection effect. Presumably, he'd exist in all the univeres in which he was playing poker, so the multiverse wouldn't explain why he happens to be in a universe in which he gets all aces when he deals. It's a bad illustration.
Craig's other attempt at a response is to say the multiverse raises the Boltzmann brain problem. That might work as a response to somebody who was actually trying to defend the existence of a multiverse. Or it could work as a response to somebody who claimed fine-tuning was evidence for a multiverse. But it doesn't do much to answer somebody who is only raising the multiverse as a possibility that can't be ruled out. After all, they could simply raise, as a possibility, the existence of a multiverse that explains fine-tuning without generating Boltzmann brains.
The Boltzmann brain problem doesn't come up in every multiverse scenario, so it has limited application. The Boltzmann brain problem is a real issue for any multiverse model in which universes are explained as being random low entropy fluctuations in a larger space that is, on average, in thermodynamic equilibrium. But it's not as much of an issue in a scenario in which bubble universes emerge as local areas of stability within a larger inflaton field. And it only takes a bit of imagination to cook up multiverse scenarios that can explain fine-tuning without generating Boltzmann brains. String theory/M-Theory could serve as such a model, but that's debatable.
As a side note, Boltzmann brains do not emerge in the many worlds interpretation of quantum physics, but then again, the many worlds interpretation would do nothing to explain fine-tuning either, so it's really irrelevant to the whole debate.
After thinking about the multiverse objection, how I deployed it in the debate, and how my opponent might've responded had he not conceded, and how I would've defended it, I've decided the disjunctive version of the fine-tuning argument is not the best way to defend it. I think a better way to defend the fine-tuning argument is to say that God (or a cosmic engineer) is the best explanation for fine-tuning. To refute the argument, somebody would have to either deny the univierse is fine-tuned at all, or offer a better explanation than God for the fine-tuning. If they offer the multiverse as an explanation, it won't be enough to offer it as a mere possibility. They'll have to defend its actual existence. Unless it exists, it can't be the explanation for fine-tuning.
