A Fine-Tuning Argument For Atheism

Why our natural laws point away from a designer, not towards one

Pictured: A universe fine-tuned for good stock photos

One of the most common arguments for the existence of God is the fine-tuning argument. Proponents of this argument point to the unimaginably large number of ways our natural laws could have been and argue that theism is the best explanation for why the natural laws we actually have happen to fall inside the extremely narrow range that would allow for life to develop. There are plenty of atheists who think this (apparent) fine-tuning is nothing to be concerned about, but I disagree; I think modern physics has given us a legitimately puzzling picture of reality over the last few decades, and that atheists should engage with it seriously. At the same time, the fact that apparent fine-tuning presents a meaningful scientific and philosophical conundrum doesn’t imply that fine-tuning arguments are necessarily successful. Here, I want to build on Neil Sinhababu’s “electrons in love” objection to explain why, even as someone who who has real questions about why our natural laws are the way they are, traditional fine-tuning arguments for theism don’t move me - and why I think this apparent fine-tuning might actually end up providing strong evidence against theism, rather than for it.

To start, I’ll briefly summarize the main thrust of the “electrons in love” objection, first raised by Sinhababu in his 2016 paper “Divine Fine-Tuning Vs. Electrons in Love.” (You can, and should, read the entire paper here.) It starts by recognizing that, in the same way God is able to fix the particular structure of our natural laws, he should also be able to fix the particular structure of the psychophysical laws that govern how and when morally valuable minds come into being. The psychophysical laws in our actual universe happen to be extremely “mind-unfriendly,” in that they allow for morally valuable minds to exist only in rare cases where specific forms of advanced biological life develop. But these psychophysical laws are by no means necessary. As Sinhababu says:

Under more mind-friendly psychophysical laws, protons and electrons themselves could have minds like ours. These laws could dictate that these particles have sensory experiences of all the forces other particles exert on them, with the forces most strongly affecting them giving rise to the psychology of belief and then intentional action. Protons and electrons could yearn to be together, feeling delight at the presence of the other as their opposite charges drew them closer. When they formed a hydrogen atom, they could fall in love. Whenever two electrons were a prime number of centimeters apart, they could have the mental states involved in heartfelt communication about their histories. Every subsequent time they were a whole number of meters apart, they could fondly remember each other. The remaining strong nuclear force between protons, despite being too weak to bring them into one atomic nucleus, could at its moments of greatest intensity realize a tantalizing but forbidden attraction. When any six particles formed a regular hexagon, they could share awe at the grandeur of the universe.

If psychophysical laws like these are truly possible, then traditional fine-tuning arguments for theism - which rely on the premise that morally valuable minds require the natural laws governing our universe to exist within a very narrow range - would be significantly weakened. Of course, it’s true that the natural laws governing our universe must exist within a very narrow range for morally valuable minds to exist if the actual psychophysical laws of our universe are fixed; if you absolutely need advanced biological life in order to have morally valuable minds, then you’ll absolutely need natural laws that allow for molecules to hold together and planets to form and metabolism to function and so on. But given the sufficiently wide range of possible psychophysical laws that God should be able to bring about, any set of natural laws, no matter how unfriendly to biological life, could have corresponding psychophysical laws that allow for morally valuable minds to exist inside whatever physical universe the natural laws themselves produce.¹ Therefore, theists have no grounds for asserting that any particular set of natural laws is, by itself, any more or less conducive to the existence of morally valuable minds.

Further, if we do fix the actual psychophysical laws of our universe and consider only the likelihood of natural laws that allow for morally valuable minds to arise in conjunction with them - that is, natural laws that allow for a specific sort of advanced biological life - then we lose any grounds for thinking the total state of affairs in question is more likely under theism than under atheism. This is because, on the assumption that theism is true, the prior probability of mind-unfriendly psychophysical laws is already extremely low, and therefore the conjunction of those laws with any set of natural laws will be unlikely as well. To quote Sinhababu again:

God’s goodness gives him no motivation to create a world with mind-unfriendly laws, as the mind-unfriendliness of a world’s laws is a very implausible candidate for realizing moral value. While many candidates for what makes states of affairs good have been suggested in the history of moral philosophy, nobody has suggested the mind-unfriendliness of psychophysical laws. A loving God wouldn’t have any obvious reason to situate creatures in a world where it was hard to have minds. … In fact, if quantitative judgments can be made about the likelihood of creating particular universes as the finetuning argument requires, God would be much more likely to create a universe with mind-friendly psychophysical laws. Those universes are much more likely to instantiate intelligent life.

We can make this point more explicit by considering a simplified probability space with 100 possible sets of natural laws, one of which allows for biological life under our actual psychophysical laws. (Of course the true ratio between the sets of natural laws that do allow for biological life and those that don’t is unimaginably larger, but the point being made scales perfectly regardless of the actual distribution.) If a naturalist observes that this one biological-life-permitting set is actual, they should naturally wonder why it is that any of the other 99 weren’t actualized instead. However, since God would be capable of bringing about life under any of the 100 possible sets of natural laws by picking various corresponding psychophysical laws, and since he has no reason to actualize any one set of psychophysical laws over the other, the fact that he chose to actualize the biological-life-permitting set is equally mysterious; in both cases, there’s a brute 1% chance that the observed biological life would develop. So while naturalists are confronted with the question of why they exist in this one world, as opposed to not existing at all, theists are confronted with the equally pressing question of why they exist in this one world as opposed to any of the 99 others. Therefore, the mere existence of natural laws that permit the development of biological life is no evidence for theism.

As an analogy, you can imagine having a seizure and waking up in the hospital, wondering which of two ambulance services were called to transport you there. The first service has a single yellow ambulance for the whole city, which means they have to ignore 99% of the calls they receive. On the other hand, the second service has a fleet of 100 ambulances that can fully meet demand, but only one of them is yellow. If a nurse lets you know that you were brought to the hospital by a yellow ambulance, what should that tell you about which service was called? For the first service, there was a 99% chance you wouldn’t arrive at the hospital at all, but given that you did arrive, the fact that you arrived in a yellow ambulance specifically is guaranteed. With the second service, you have the exact inverse situation: It was guaranteed that you would arrive at the hospital, but with a 99% chance that you would do so specifically in a non-yellow ambulance. This means the conjunction of the two observations - that you are at the hospital at all, and that you got there in a yellow ambulance - would have a probability of (1)(0.1) = 0.1 regardless of which service was called, and therefore you have no grounds for deciding either way which one was more likely. In the same way, the probability that we exist in a universe finely-tuned for biological life is identical regardless of whether naturalism or theism is true, with the only distinction being which aspect is guaranteed and which is vastly improbable.

To summarize these two points: If we allow the psychophysical laws to vary, then theists have no grounds for asserting that the particular natural laws we see in this universe are “fine-tuned” for the emergence of morally valuable minds (because any set of natural laws will be capable of producing such minds given the right corresponding psychophysical laws); and if we fix the psychophysical laws to what they actually are in this universe, then theism is automatically an unappealing explanation for the overall state of affairs (because the probability that God would bring about our particular set of psychophysical laws is extremely low regardless of what natural laws he was then required to pair them with).

This “electrons in love” objection is meant as an undercutting defeater for traditional fine-tuning arguments, in that it attempts to show the biological-life-permitting natural laws we see in our universe don’t constitute evidence for theism over atheism. However, it seems likely to me that these same sorts of observations can also be developed to make positive arguments against theism and for atheism as well. Before I sketch out what these arguments might be, I first want to make an important rhetorical point about the way apparent fine-tuning is often represented by advocates of fine-tuning arguments. It’s common for laypersons, apologists, and even philosophers to regularly frame our natural laws as a form of subtle revelation, as though their ability to permit life as we know it - that is, biological life - is like a signature God has left tucked into the corner of reality; many go so far as to directly compare the apparent fine-tuning we see with specific sentences or messages like Made By God appearing in the sky or in certain subatomic particles. But the considerations raised by the “electrons in love” objection show this is absolutely not the case.

To see why this is, we can imagine what life would be like for conscious particles in the mind-friendly universe previously described by Sinhababu. These particles would presumably have the ability to engage in certain forms of scientific and philosophical investigation, which would include considering the psychophysical and natural laws they were subject to. And if those particles did engage in that sort of investigation, they would very likely come to the conclusion that God exists, on the grounds that no naturalistic explanation of their experiences would be even remotely plausible. They most likely couldn’t even imagine the possibility that their mental lives were reducible in some way to purely physical qualities of certain particles, or that these perfectly mind-friendly psychophysical laws were the result of random chance. More generally, the vast majority of universes wherein morally valuable minds arise from something other than biological life would be ones in which naturalistic (and therefore atheistic) explanations for their existence would be radically implausible, given that almost all forms of non-biological life are incompatible with naturalism.²

These sorts of universes - universes hostile to biological life that nonetheless give rise to morally valuable minds in ways that make naturalism extremely implausible - must constitute the vast majority of the probability space for fine-tuning arguments to succeed. Therefore, advocates of the fine-tuning argument are forced to concede that all universes fine-tuned for biological life are drawn from an infinitesimally small subset of possible universes wherein God’s authorship could plausibly be denied. Or, to put it another way: To the degree that our universe’s natural laws really were fine-tuned by God, they were fine-tuned in such a way as to give morally valuable minds the greatest possible rational grounds for denying his existence. This is the exact opposite of the traditional framing given by advocates of the fine-tuning argument, where natural laws fine-tuned for biological life are regularly compared to explicit, if hidden, declarations of God’s authorship or providence. In reality, to the degree that these laws would act as evidence for God’s existence at all, they would be less like a message written out in the sky and more like a small seam or zipper that a costume designer was unable to fully hide.

This point is important because it bears on how we interpret God’s actions as an agent. Generally, we assume that someone who wants to send some message will, when faced with a wide range of otherwise equally appealing options, choose the option that best communicates that message, or if nothing else, they will certainly avoid choosing the one option that makes their message as easy as possible to miss. Therefore, when we encounter a possible message presented in that optimally ambiguous form, that should significantly lower the probability that it actually does encode a message, on the grounds that the inherent probability of the ambiguous form is so low given the motivations of the sender we have in mind. You can imagine my wife leaving work one day and finding a red rose by the tire of the car she parked outside. In this case, even if my wife knew it was rare for red roses to blow onto tires by accident, the fact that I could easily have placed the red rose in dozens of other places should give her reason to doubt that I placed it there, assuming she’s confident that I want her to know the rose is from me. Similarly, God’s purported desire to make himself known (or at least avoid making atheism unnecessarily reasonable) gives us reason to lower the prior probability of him bringing about morally valuable minds through biological life, even beyond the low intrinsic probability of him choosing any particular conjunction of natural and psychophysical laws in the first place. This plausibly makes the natural laws we observe less likely given God exists, which in turn makes apparent fine-tuning evidence for atheism.

There are also other reasons to see apparent fine-tuning as evidence against theism that don’t involve reference to God’s desires or motivations, and instead rely on plausible (but admittedly controversial) assumptions about anthropic reasoning and probabilistic updating. Up to this point, I’ve been making arguments that assume both our existence as biological life and our existence, period are both observations to update on. But if we hold the mere fact of our existence in the background and consider only the observation that we exist specifically as biological life, then things play out very differently. We can then ask: Which theory, naturalism or theism, would predict this single observation? Obviously, naturalism essentially guarantees that, given we exist, we exist as biological life. The “electrons in love” objection, however, shows us that we should at best consider all possible sets of natural laws to be equally likely under theism, in which case theism would predict with overwhelmingly high confidence that we exist in those universes where another sort of life other than the biological kind exists. Therefore, when we observe that we are in fact biological life, we observe something guaranteed by naturalism but radically improbable given theism, which again shows that our existence as biological creatures in a universe with natural laws apparently fine-tuned for our emergence is strong evidence against the claim that God exists.

We can update the analogy I gave earlier about the competing ambulance services to accommodate this shift. Here, instead of making our existence at the hospital hinge on the ambulance arriving, we can pick a new scenario where both services are guaranteed. Imagine you’ve recovered from your stroke and are now being discharged from the hospital, and the concierge randomly chooses one of two taxi service for you to take home. The first service has only one car, which is painted yellow, while the second has 100 cars of various colors, only one of which is yellow. If you step outside and find a yellow car waiting for you, then you should obviously assume the concierge called the first service, on the grounds that her calling that service would give you a 100% chance to see a yellow car while her calling the second service would give you a 1% chance to make the same observation. And if the second fleet were to have trillions upon trillions of different cars, only one of which was yellow, then seeing a yellow car would be so vanishingly improbable you could be essentially certain the concierge had called the first service instead. By the same reasoning, so long as we fix the mere fact of our existence as background evidence, finding out that we exist specifically in a universe apparently fine-tuned for biological life is incredibly strong evidence for naturalism (and thus atheism), given that such an observation is all but guaranteed on naturalism while being unimaginably unlikely given theism.

Together, I think these two arguments - one based on very plausible premises regarding agents’ motivations and another based on very plausible premises regarding anthropic reasoning and probabilistic updating - should give anyone who accepts the “electrons in love” objection good reason to not only reject fine-tuning arguments, but also to see apparent fine-tuning as strong positive evidence against the existence of God. This is not to say that either argument resolves the question of how it is that we did come to exist in a universe with laws that permit life like ours. On that, I’m still completely baffled. Nonetheless, whatever plausible options for explanation there might be, I’m confident theism isn’t on the list.

1. Some theists accept this point, but use it to bolster a different sort of fine-tuning argument based on the apparent harmony between psychophysical laws and natural laws. I’ve written a response to this argument as well.

2. Some may disagree here, saying that denizens of these worlds would only have reason to reject naturalism and not atheism. They could, for example, be drawn to a form of atheistic non-naturalist axiarchism as a way of explaining their extremely mind-friendly psychophysical laws. I think there are good reasons to doubt this, but if you disagree, then we could still consider the sorts of universes wherein God sustains life directly through constant, unambiguous miraculous action rather than any particular psychophysical laws at all. In these universes - which also take up a much larger share of the probability space than those with biological life - denizens would clearly be inclined to accept theism, and the same point can be made by contrasting those possible universes with ours. More generally, anyone who is skeptical of psychophysical laws as a concept could also reframe the broader “electrons in love” objection in terms of miraculous sustaining.