Everyone’s heard about the traditional fine-tuning agument for God. And for good reason! There are a number of physical constants which fall in a very narrow range needed for life. The cosmological constant has odds by chance of being life-permitting on the order of one in 10^120. Even atheists often regard this argument as formidable.
But there’s another potential fine-tuning argument which you’ve probably never heard of. It was formulated by Robin Collins, though apparently a number of physicists have double-checked his calculations. Because this version is newer, the physics might turn out to be wrong. But if it’s not—if the physics checks out—then it would almost single-handedly eliminate naturalism as a viable hypothesis. This new kind of fine-tuning has the potential to be much more evidentially weighty than the traditional kind.
The original fine-tuning argument is about fine-tuning for life. Absent certain physical parameters falling in a narrow range, no life could ever arise. This version concerns something quite different: fine-tuning for discoverability. Certain parameters in physics are finely-tuned for the universe being discoverable. If they didn’t fall in that narrow range, life could still exist, but it wouldn’t be able to do science anywhere near as well.
Our ability to do scientific discovery is a very good thing and is broadly continuous with a number of other surprising respects in which the universe is discoverable. So if there’s a God, it’s not that improbable that the constants would be fine-tuned to allow us to discover the universe. I don’t think this is guaranteed, but it’s at least not supremely unlikely.
If there’s no God and the physics checks out, the odds of fine-tuning for discoverability are infinitesimal. Robin has suggested the odds are on the order of one in a trillion trillion. And while sometimes you should think improbable things happened, you should never, never, never assume a one in a trillion trillion coincidence happened when there’s an alternative viable theory believed independently by a number of smart people. That is about the odds that if a present human was selected at random, you’d be selected, then selected again, then that you’d enter a lottery where your odds of winning were one in a hundred thousand and win that too.
Here are some of the most important cases Collins discusses (though he thinks the particle physics examples are by far the strongest):
Cosmic microwave background radiation is the faint, nearly uniform glow of microwave light permeating all of space. It is extremely useful for doing physics and was integral in discovering that the big bang had happened. The higher its intensity, the better for discovery. The CMBR is determined by the ratio of baryons (normal matter particles) to photons (light particles). The ratio is about one to a billion. This ratio happens to be ~exactly ideal for maximizing the intensity of the CMBR. Odds of this by chance are about one in a billion. Collins double checked this calculation with three physicists.
(This is the weakest example as Collins agrees) The fine-structure constant governs the strength of the electromagnetic force. If it were larger, the electromagnetic force would be stronger, while if it were weaker, the electromagnetic force would be weaker. If it were greater, then wood fires wouldn’t work, and civilization wouldn’t have domesticated fire. If it were smaller, then fires would burn roughly continuously, meaning forest fires would burn through wood. This would make it impossible for early humans to light firesa. The other effect of it being smaller would be “the resolving power of a light microscope decreases, electric transformers and motors rapidly become less efficient, and the usability of the magnetic compass and paleomagnetic dating rapidly diminishes.”
Certain parameters in particle physics fall in a tiny range in terms of their mass that is ideal for us being able to discover them. If they were weaker or stronger, they would be harder to discover. The muon is a good example of this.
Fine-tuning for discoverability dodges a lot of the big objections to fine-tuning for life.
One objection to fine-tuning for life goes: of course the world had to be fine-tuned because otherwise we wouldn’t be here. Now, let’s ignore for a second that this response is demonstrably wrong (as roughly all philosophers who have thought about it have recognized). This objection doesn’t even arise with fine-tuning for discoverability because we could have existed in a less discoverable universe.
I think the best response to fine-tuning is to posit a multiverse. If there are a bunch of universes with different constants, then it isn’t at all surprising that *some *universes would be habitable. And it’s not surprising that *we *find ourselves in the habitable universes, because we couldn’t find ourselves anywhere else. The uninhabitable universes by definition have no life.
But positing a multiverse won’t help at all with fine-tuning for discoverability. We could have found ourselves in a less discoverable universe. So while the multiverse can maybe explain why some universes are inhabitable, it does absolutely nothing to explain why *we *find ourselves in the habitable universes.
As an analogy, imagine that there are two theories. One theory says that everyone in the world is born with a birthmark. The other one says that only one person has a birth-mark. If you have a birth-mark, that gives you evidence for the first theory. While both theories explain why *someone *has a birthmark, the second one better predicts that *you *have a birthmark. Analogously, a multiverse explains why some people are in discoverable worlds, but it doesn’t explain why we are.
This is pretty bad news for atheism. Maybe there are some things that can be said that chip away at the fine-tuning argument to some degree. For example, atheists often argue that theism doesn’t super neatly predict fine-tuning. But none of the arguments are anywhere near good enough to license accepting that a one in a trillion trillion coincidence simply happened. So if the multiverse doesn’t help with fine-tuning for discoverability then atheism is in rotten shape.
You might object that fine-tuning for discoverability is subject to similar anthropic selection effects. If there wasn’t fine-tuning for discoverability, then we wouldn’t know about it. But this is wrong in two ways:
Collins argues that even if the parameters in question weren’t finely-tuned, we’d still know about them. We’d still know about the baryon to photon ratio and the CMBR. We would be able to figure out that they weren’t ideal.
This is bad reasoning. The fact that there not being a discoverable universe would have left us unaware of that fact doesn’t mean there being a discoverable universe isn’t evidence for theism. Return to the birthmark analogy, where there are two theories: one says everyone has a birthmark, while the other says just one person has a birthmark. If you have a birthmark, that is obviously evidence for theory one. And that’s true even though if you hadn’t had a birthmark, you probably wouldn’t know that birthmarks existed.
1
Now, one last argument that people give against fine-tuning is the electrons in love argument. The argument goes: God would be able to make a supremely valuable universe no matter what the constants were by direct creation. So the odds of there being a finely-tuned universe are still minuscule given theism.
I don’t buy this objection. But let’s say you do as your main objection. Even there, I think fine-tuning for discoverability seriously undermines atheism. If we’re really generous to the electrons in love point, maybe let’s say that fine-tuning comes out a tie, neither favoring theism nor atheism.
But after we’ve considered that, then fine-tuning for discoverability comes as a shock to the atheist but not the theist. The theist should think that conditional on God going through the bother of fine-tuning a universe, it isn’t terribly unlikely he’d also make it ideal for science. Maybe it’s not guaranteed, but it’s not hugely unlikely.
For the atheist, the story is very different. There’s no bias in the laws for us to be able to do science. The laws weren’t written with us in mind, so it would be a bit coincidence if they were ideal for our doing science. The odds remain low even after conditioning on fine-tuning for life.
In other words, fine-tuning for life is a double-whammy. Maybe electrons in love can cancel out fine-tuning for life. But once it’s done that, it’s unable to block the fine-tuning for discoverability argument. Conditioning on fine-tuning for life, fine-tuning for discoverability is still shocking given atheism, while a lot less shocking given theism.
Fine-tuning for discoverability doesn’t dodge every objection to normal fine-tuning. A few of the weak ones still apply here (see this article for my explanation of why I don’t buy most of the objections and here for why I don’t buy the measure problem—the last remaining formidable objection to fine-tuning for discoverability). But I basically think that if fine-tuning for discoverability holds up, naturalism has no remotely reasonable way of explaining it. Absent any explanation of the one in a trillion trillion coincidence, naturalism is toast. If Robin Collins didn’t calculate wrong, then naturalism is almost definitely false.
To walk through the analogy in more detail: theism is analogous to the theory that everyone has a birthmark, atheism the theory that only one person does, and you having a birthmark is analogous to the universe being discoverable. Even if you not having a birthmark would have left you unaware of the evidence, you having a birthmark is still evidence that everyone has a birthmark. In the same way, even if most atheistic worlds would have left people unaware of the suboptimal discoverability of the world, the optimal discoverability of the world is still evidence for theism.
