The anthropic argument against the existence of God.

Our existence may be one of the most compelling arguments against the existence of God—the anthropic argument claims that, given a God that is morally perfect and all-powerful, our existence as a species shouldn’t be possible, since such a God wouldn’t focus his infinite amount of processing capability on materializing mortal, finite beings that are capable of suffering.

Instead, the morally perfect decision would be for such a God to create other morally perfect deities that experience an infinite amount of happiness.

A steelmanned version of this argument runs like this:

If God were morally perfect and all-powerful, he would perform the morally best action in the context of creation.
Creating the morally best beings possible within a perfect God’s capacity for creation would entail only creating other morally perfect beings.
Humans are not the morally best beings possible within a perfect God’s capacity for creation.
If a morally perfect, all-powerful God existed, he would not create humans.
Humans were created.

Conclusion: A morally perfect and all-powerful God does not exist.

It does seem like it’s in the best interest of the omnipotent God to create deities that are morally perfect that can infinitely increase the amount of happiness in the universe; this implies that the causal mechanism behind our existence is probabilistically restricted to either a God that is not all-powerful and morally perfect, or, alternatively, purely unguided naturalism. Why would an infinitely flawless being choose to create such morally inferior beings?

The argument, however, ultimately falls short of proving that a morally perfect God would refrain from creating morally imperfect beings.

Firstly, a morally perfect God would aim to maximize the total amount of utility within his domain. Such a God would not be restrained by any Deontological bars, so his moral actions should entail creating the most ‘goodness’, or utility, possible.

This is since:

A morally perfect God has perfect goodness.
Perfect goodness entails materializing all potential for net contributions to the amount of value in the universe.
Utility is the same as the total amount of value in the universe.
A morally perfect God would seek to maximize utility.

Even abstract concepts like diversity hold value in their ability to increase utility; everything holding value—a trait that the perfect God would want to maximize—collapses into utility.

Under utilitarianism, it’s in such a God’s best interest to create an infinite amount of beings that are morally infinite, infinitely happy, and morally perfect; a group of perfect deities would be realistic. The question lies in whether or not a morally infinite being has a reason to create any morally finite, mortal beings. In this case, whether or not the morally perfect God operates under total utilitarianism or average utilitarianism determines whether or not we can expect to exist under such a god or not. We can show the clash in terms of some simple utility math.

Now, we’ll represent the infinite happiness experienced by God’s necessary deity creations as ω (a hyperreal infinity that we can perform utility calculations on), the total number of deities as k, the utility experienced by human beings as u, and the number of human beings as n. Since we can perform calculations on ω, it must be established that:

ω + u > ω, and total utility = ω*k + u*n, where k and n both have to scale up to infinity to maximize happiness.

ω + u > ω is intuitively true; having an infinitely happy being and a happy human chilling on an Eden-like planet is better than having only an infinitely happy being on the planet. Adding u to ω creates a greater set than if we only worked with ω.

But if k and n both are forced to scale up to infinity, the average utility must be represented as

(ω*k + u*n) / (k + n), while if God only created the infinite beings, the average would be (ω*k) / k, or just ω. Since ω >> u, adding finite beings reduces the average utility in every conceivable scenario! That means if the perfect God were an average utilitarian, the anthropic argument still works.

Average utilitarianism is implausible since it leads to a reverse Repugnant Conclusion—that the perfect God shouldn’t create a maximally happy finite being, since it would still bring the average down. The absurdity of this conclusion is proportional to the number of happy beings created. If God had the option to create aleph null (ℵ₀) happy finite beings, all living on lush, Eden-like planets, average utilitarianism would mean that God would refrain from doing so since it brings down the average.

Calculating utility as an average is simply incompatible with a morally perfect God’s obligation to create an infinite number of infinitely happy beings.

Proponents of the anthropic argument do have a decent objection to this, though. If humans add a tiny bit of happiness into the universe, they can take up space that could have been filled by more infinite utility beings. If you have infinite money, you still wouldn’t spend $10 to give a person a Popeyes sandwich when you could’ve spent that $10 (effort of creation is the same) to give someone an entire luxury mansion on a utopian planet surrounded by utopian beings.

Such a frontline is promising—if an infinity like ω is just an infinitely continuous set, having one term in that set assigned to a morally inferior being is still going to be worse than if that being were swapped by a morally perfect being. If a god continuously creates beings that have ω utility, replacing even 1 of those beings with a being that has u utility decreases the amount of total utility.

Consider 2 worlds:

World A has k beings, each at value ω, along with n beings at value u.
World B has k + n beings, each at value ω.

For every world A you can conceive of, there can always be a world B that has the exact same set of beings, except the ones experiencing finite happiness get replaced by deities experiencing infinite happiness. World B is obviously better than world A under any utility calculus.

Even if we assume that a God can always either create:

k beings at value ω OR
k beings at value ω and n beings at value u

This still doesn’t give us an escape. Option 2 is obviously better, but an atheist can simply point out that you can swap the n beings at value u for beings at value ω—k + n beings at value ω, which is then better than both options.

An objection like this still falls apart; it simply proves too much. The swap argument assumes that the set of all possible beings is inexhaustible, even with infinite swaps! We could simply make up a world C with k + 2n infinitely happy deities, or a world D with k + 3n infinitely happy deities, or a world E with k + 4n of those deities… meaning that every single world you’re looking at has to be suboptimal, since you can always infinitely optimize the infinite set of infinite beings. In this case, the God wouldn’t be able to act at all, proving their argument false by reductio ad absurdum.

But if we simply took our original scenario with k beings with utility ω and n beings with utility u, both k and n scaling up to infinity would make for an infinitely happy world. If a God weighed his own options here, he would obviously proceed with creating an infinite amount of both finite and infinite beings rather than become trapped in a loop where committing to a decision is impossible—the latter option leaves 0 happy creations in the entire universe.

Since the swap argument eliminates every possible scenario as suboptimal, it fails to take out our original total utilitarianism example. We now have a proof demonstrating that a morally perfect God has a moral reason to create finite beings like humans, all of which have relatively small amounts of happiness and are prone to suffering:

A morally perfect God needs to maximize the value of the universe.
The value of the universe is dictated by its utility.
Therefore, a morally perfect God would maximize the utility of the universe.
It must be true that a morally perfect God creates k beings experiencing ω utility, with k being forced to scale up to infinity.
That God must create an infinite number of infinitely happy beings.
Such a God can’t be an average utilitarian, since that would mean refusing to actualize potential happiness experienced by finite beings.

The swap argument is the best response against total utilitarianism, but it still fails.

No optimizations based on replacing finite beings work since it leads to an infinite regress, causing no decision to be made and thereby creating no happiness.
The only coherent utilitarian principle remaining is total utility maximization under hyperreal arithmetic.
ω + u > ω, so adding a finitely happy being still is able to raise the total utility of the universe when u > 0.
A morally perfect God has a positive obligation to create finite beings alongside infinite ones, since leaving u*n unactualized means refraining from creating positive utility.
Since humans are finite beings with u > 0 positive utility, a morally perfect God has a positive obligation to create humans.

Our existence now seems quite likely under a morally perfect God!

The anthropic argument against the existence of God falls short of proving that a God’s obligation to create an infinite number of infinitely happy beings is mutually exclusive with the obligation to create an infinite number of finitely happy beings. In fact, we should come to expect our existence more under a morally perfect, utility-maximizing God.