Kahane's argument

Here is the standardized version of Kahane's argument:

1. (1) If God had existed, this would necessarily make things better in important respects.

2. (2) If God had existed, this would necessarily make things worse in other important respects.

3. (3) All (or at least most) of the benefits under [1] could be had without God.

   Therefore,

4. (4) There are possible atheist worlds that offer all (or most) of the benefits of God's existence but without the costs.

   Therefore,

5. (5) Some atheist worlds are overall the best, or among the best, and are superior to all theist alternatives.Footnote ⁵

   Therefore,

6. (6) The worlds we should most prefer are atheist worlds (Kahane (Reference Kahane and Kraay2018), 103; premise numbers mine).

Kahane says that ‘[t]he basic idea is simple. Notice that while the benefits mentioned in [worlds where God exists] are expected upshots of God's existence, they do not inherently require God's existence’ (ibid., 102). In my earlier work in expanding Kahane's argument I sought to detail how some of the advantages of theism could obtain in atheist worlds. I explained that I hoped ‘to show the method that can be applied to each good in order to discover how such goods can obtain without God. On the assumption the goods I examine are representative of the theistic goods in general, then this method can be applied other theistic goods that I don't examine’ (Lougheed (Reference Lougheed2020), 180). I then argued that theistic advantages such as cosmic justice, a good afterlife, divine intervention, no gratuitous evil, relationship with a maximal being, and salvation can all obtain in atheist worlds (ibid., 181–186).

Notice, however, that neither I nor Kahane mention anything about the disadvantages of theism not obtaining in the relevant atheist worlds. But in order for (4) to follow from (1)–(3) it also has to be the case that:

1. (7) All (or at least most) of the costs of theism can be avoided without God.

I didn't think to defend (7) in my original analyses and Kahane never considers it either. Perhaps it seems rather obvious or intuitive that (7) didn't really need to be included in Kahane's original argument. For how could the costs of theism possibly obtain in a world without God? But this reasoning could be applied precisely to the other question: how could the benefits of theism possibly obtain in the world without God? Kahane and I have tried to answer the latter question but not the former. However, we never offered a principled reason for thinking that the one, but not the other, needed defending. Consider, for example, that I have argued that part of the dignity harm of God's existence involves being necessarily subservient to a more powerful being. But if a really powerful (albeit less than maximal) supernatural being is necessary to have some of the advantages of theism, might not we also be necessarily subservient to that being? It is this sort of question which must be answered in order to complete Kahane's argument. In other words, (7) needs to be defended.

Before proceeding to my criticisms of (7) it's important to be aware of three additional caveats. First, there is sometimes a difference between the experience of a particular advantage or disadvantage and such advantages or disadvantages actually existing. For example, I have explored the possibility that God might hide in order to allow humans to experience some of the advantages of atheism (e.g. privacy) while providing most of the benefits (Lougheed, Reference Lougheed2018a). However, I have also suggested that this argument is unsuccessful at worst or needs more work at best because theistic benefits might well obtain in a world where God doesn't exist (Hendricks & Lougheed, Reference Hendricks and Lougheed2019).Footnote ⁶ In light of this stalemate, it's currently unclear whether appealing to the experience of advantages or disadvantages lends more support to either the pro-theist or anti-theist. I therefore won't make use of this distinction in my criticisms of (7). Second, notice that if Kahane's argument is successful it establishes that all versions of pro-theism, narrow/wide/personal/impersonal are false. At least this is so if all (or at least most) of the benefits of theism can be had in atheist worlds without also the costs. Third, Kahane says that:

Now, while Kahane is right that this cannot be ruled out in advance, I have suggested that it's quite doubtful the best atheist worlds can be naturalistic given the nature of some of the advantages and disadvantages in question (Lougheed (Reference Lougheed2020), 180). While in what follows I won't explicitly defend the claim that the very best worlds are supernatural, I will flag points in the discussion below that support this claim.

Problems for premises (3) and (7)

In what follows I'm going problematize both premises (3) and (7). I do this by showing that it's difficult to get the benefits of theism without the associated costs. Some of this discussion will target (3) or (7), though it's important to understand that part of my worry is that the benefits and costs of theism are inextricably interwoven together. There is thus an important sense in which the following discussion necessarily targets both (3) and (7). Finally, keep in mind that I am not evaluating the following considerations in support of anti-theism. For the sake of analyzing Kahane's argument I'm simply going to assume that they're correct.

Meaning in life

In some earlier work, Kahane suggests that for certain individuals their lives may lose meaning if it turns out that God exists. This is because the life pursuits that are important to them might be strongly tied to the fact that God does not exist (Kahane, Reference Kahane2011). For example, such pursuits might be interwoven with privacy, understanding, and independence.

In further developing the argument for anti-theism from meaning, I've noted that the argument is consistent with any theory of the meaning in life except supernatural theories which say God's existence is necessary in order to have a meaningful life.Footnote ⁷ This observation shows that if God does not exist, then losing meaning because of God is impossible.Footnote ⁸ So Kahane's initial suggestion might be best understood as one about naturalistic atheist worlds, not supernaturalist atheist worlds. On this interpretation, then, the worry is that if (certain versions) of supernaturalism turn out to be true, then certain individuals will be unable to have meaningful lives. The problem, however, is that the proponent of this argument isn't entitled to say all that's needed is the naturalist atheist world to avoid this cost of theism since the world in question is also one where the benefits of theism obtain and hence must be supernatural. To see this worry more vividly consider the following case study:

• MEANING: Troy's life will lose meaning if it turns out that God exists. He highly values privacy, understanding, dignity, and autonomy. Troy is a scientist and wants to be able to understand the world through the normal methods of human inquiry. He wants his thoughts to remain private and he doesn't want to be watched when he is at home. Additionally, Troy doesn't want to be subservient to a more powerful being. These things are so dear to Troy that his life would become meaningless were he to lose them. And, as with most everyone else, a meaningful life is incredibly important to Troy. Discovering that God exists would ruin Troy's life. His friend Tabitha, however, assures Troy that God doesn't in fact exist (and let's say she's somehow able to show him decisive evidence for this conclusion). Troy is happy to discover that God doesn't exist because he can live a fulfilled and meaningful life. Tabitha even tells Troy more good news: the world they inhabit is one in which they also get most of the benefits of God's existence. Troy is puzzled by this and asks Tabitha for an explanation. She proceeds to tells him that there are ten lesser gods, instead of just one maximally powerful God. Some of these gods are really close to perfectly loving and forgiving, while others are very powerful, and still others are incredibly knowledgeable. But no god has the maximal (or infinite) degree of any such property. While these ten gods came together to create our universe it's unclear whether they exist necessarily or came from elsewhere. But this explanation bothers Troy. If there really are these ten supernatural beings who created the world, it's doubtful that the entire universe can be understood completely through the normal means of inquiry. He also worries that the very knowledgeable gods are spying on him to add to their knowledge. And finally, he wonders whether he is now subservient to ten lesser gods instead of just one maximal God. Troy realizes that even though God doesn't exist, he's still at risk of having a meaningless life.

This is just one example among numerous possibilities that might have been devised in order to show that it is difficult to get the benefits of theism in an atheist world without the also the harms. Hence, the truth of premise (7) is in question. If there is a way to show that such a world is possible it's up to proponents of Kahane's argument to detail how this is so. The challenge to Kahane's argument should now be coming into sharper focus for the reader. In what follows I examine some of the specific disadvantages of theism in order to show how it is hard to avoid them while simultaneously reaping the benefits of theism.

Privacy

Privacy is a disadvantage of theism often mentioned in literature. I have argued extensively that anti-theism can be supported based on various considerations about privacy (Lougheed (Reference Lougheed2020), ch. 3). Here's the basic idea of the argument: God is all-powerful and all-knowing, so God knows the whereabouts of every human at all times. Furthermore, God knows the thoughts of every person and even what it is like, phenomenologically, to be each and every person (since if God did not know this then she would lack knowledge of certain facts which is impossible). At least for those individuals who value privacy, God's existence makes things worse with respect to privacy. Hence, this argument easily establishes narrow personal anti-theism. Here's an example to help show in an atheist world where privacy is possible, many important advantages of theism may be lost:

• PRIVACY: Ahmed is an introvert and highly values his privacy. Moreover, Ahmed is a philosopher who conducts research about quite controversial topics including the legitimacy of transrace, whether transwomen should be allowed in certain ‘women only’ spaces, and the ethics of violent pornography. Ahmed rarely, if ever, presents his work at conferences or shares drafts with colleagues. He only wants his well-thought-out published views to be publicly accessible. In trying to understand the objections to his views Ahmed sometimes has to take on the thought processes of his opponents. In other words, he sometimes has to imagine what it is like to believe what he thinks are morally heinous positions. Ahmed needs privacy to conduct this research and from his perspective it would be better if no one had access to these exercises, let alone even knew that this is sometimes the method he uses. It's therefore better for Ahmed, at least with respect to privacy, if God does not exist (even if it would be worse for his life overall). Suppose Ahmed's friend, Sally, reassures Ahmed that God doesn't exist and so his privacy is fully intact (let's say she's somehow able to show him decisive evidence for this conclusion). Sally tells Ahmed the same good news that Tabitha gave to Troy: there are ten really powerful (though not maximal) gods that allow us to have the benefits of theism without the costs. For instance, some of these gods possess a great deal of knowledge, though not the maximal (or infinite) amount of knowledge. At first Ahmed is relieved. He discovers that the gods in question know where everyone is located in space and time but can only read thoughts inasmuch as they can perfectly read human body language and other external factors. These gods can't literally read his mind, nor do they have direct access to his phenomenological experiences. This makes the world with ten lesser gods much better with respect to privacy than if there was just one maximal God. Ahmed soon realizes, however, that some of the advantages typically associated with God's existence are actually impossible in a world with just these ten lesser gods and not God herself. For example, Ahmed worries whether the gods could be a perfectly fair judge since knowing our thoughts and experiences seems essential to being a perfectly fair moral judge. Additionally, Ahmed worries that the gods might not be able to ensure just compensation for victims of evil in this world because to truly know the harm someone has gone through, knowledge of their experience is required. Thus, while a loss of privacy is avoided in the world with ten lesser gods, some other advantages of theism are also lost.

Unlike the example with Troy, the problem isn't that the disadvantage actually still occurs in the atheist world. Rather, the problem is that some of the purported advantages of theism are lost in the atheist world. Remember, for Kahane's claim to succeed it has to be the case that both the advantages of theism obtain and the disadvantages of theism are avoided in the atheist world in order to say definitively that it's axiologically superior to a theistic world. This above example is thus a reason to think that premise (3) is false.

Privacy and cosmic justice

Now, the claim being made above implicitly ties together a personal judge and cosmic justice. To see this, consider a standardized version of this suggestion:Footnote ⁹

• The Incompatibility of Cosmic Justice and Complete Privacy Argument

  1. (8) In order for the theistic advantage of cosmic justice to obtain there needs to an all-powerful, all-knowing, and all-good being (i.e. God) who is able to monitor both the outer and inner lives of all persons (and any other moral agents).

  2. (9) If (8) is true, then a world that realizes cosmic justice would lack complete privacy.

     Therefore,

  3. (10) The theistic advantage of cosmic justice cannot obtain atheist worlds.

Of course, the key premise to defend here is (8) and the examples in the previous section give us some (thought not decisive) reason to accept it. Notice too that this is not a claim that there cannot be objective morality without God. There could be a realist non-theistic (and naturalistic) morality, but this is consistent with there not being cosmic justice. Let's grant too, that perfect knowledge, goodness, and power are necessary for cosmic justice. After all, even slight deficiencies, say, in knowing an agent's intentions could lead to inaccurate judgments and hence punishments.Footnote ¹⁰ Instead, a controversial assumption in this argument is that the mechanism for cosmic justice has to be personal. (8) therefore requires the following:

1. (11) The mechanism of cosmic justice must be personal.

I think that (11) is true and that it can be defended but it's significant that obstacles to it are immediately found in the axiology of theism literature. For instance, in discussing how to delineate the relevant comparison class in the axiology of theism, Michael Tooley notes that atheist worlds include ‘also worlds of the sort that the Buddha believed in, where there are immaterial souls, along with reincarnation and karma, so that good deeds and evil deeds affect how good one's state is in the next life’ (Tooley (Reference Tooley and Kraay2018), 48). Additionally, Toby Betenson writes that:

In listing examples of what atheist worlds include Kahane notes that such worlds can have ‘the metaphysics of Karma and rebirth’ (Kahane (Reference Kahane and Kraay2018), 98). And finally, he says of cosmic justice that:

So, in figuring out whether (11) is true we need to assess whether cosmic justice can obtain via an impersonal mechanism such as Karma. It's impossible to offer a complete survey of the relevant issues here, which include offering further details about Karma itself, but here are some brief reasons for thinking that karma cannot ensure cosmic justice. Personal identity does not obtain in reincarnation.Footnote ¹¹ The reason why a person has a terrible life right now is because of things that individual did in previous lives. However, personal identity does not carry across such lives and as such it's difficult to see how they really are the same person and, more importantly, how this constitutes a form of justice. More remains to be said, but I'm doubtful Karma is a plausible atheistic candidate that could ensure cosmic justice.

However, notice that Kahane also says that cosmic justice could be had through ‘some kind of purely naturalistic process’. Maybe all Kahane needs here is for it to be merely logically possible that cosmic justice obtains through a purely naturalistic process. But is it possible? A better candidate than Karma might be to stipulate that cosmic justice is a brute fact of the universe. Maybe it is just a brute fact that there is a blissful afterlife and everyone who suffers during their earthly lives will receive just compensation, and so on. If it is just a brute fact, then there is no personal force that has to violate our privacy in order to ensure cosmic justice. Questions remain: does there have to be a ‘knower’ who possesses the relevant information in order for there to be cosmic justice? Does such a ‘knower’ have to be personal? The way in which these questions are answered will dictate how one assesses the Incompatibility of Cosmic Justice and Complete Privacy Argument. At this stage I think the burden is on Kahane to provide more details about just how cosmic justice could be ensured by some impersonal force or mechanism.

Dignity

Another disadvantage of theism has to do with autonomy and independence. I've argued that the best way to understand these concerns is that they're fundamentally about dignity (Lougheed (Reference Lougheed2020), ch. 4). The worry is that it's disrespectful to be created by another being to fulfil ends other than your own. For instance, parents who have children for a particular reason and force that reason on their children (e.g. for more help on the farm, or to be a professional musician), undermine the dignity of their children (Metz (Reference Metz2013), 103). By analogy, it's disrespectful for God to create persons to fulfil ends other than their own. The problem is that many conceptions of theism say that the sole end of humans is to worship and glorify God, or perhaps to be in relationship with God. Furthermore, even if God created humans to fulfil their own purposes, that would still ultimately be what God intended for them (Lougheed (Reference Lougheed2020), 125–126). Finally, even if this cannot be concluded from bare theism, Kahane holds (tentatively) that if God exists then we have a reason to worship God. On this view, the problem arises because of the inherent hierarchy that God's existence entails.

Furthermore, a relationship with God is said to be a very great good. Indeed, the fact that such a relationship isn't (apparently) available to people who genuinely seek it has been levelled as reason against the existence of God (e.g. Schellenberg, Reference Schellenberg2015). A relationship with the lesser ten gods might be possible but there's little reason to think that it will be qualitatively similar to a relationship with God.Footnote ¹² So even if the dignity harm is avoided in the atheist some benefits of theism are immediately lost.

One might object that as limited humans there would not be any qualitative difference between a relationship with an almost perfect being versus a perfect being. Additionally, a relationship with ten lesser gods will be different from a relationship with God, but this needn't entail that it is worse.Footnote ¹³ However, part of what makes a relationship with God uniquely valuable (at least according to the theist) is that there can be perfect trust between humans and God. And part of what makes perfect trust available is that God is a perfect judge, she is perfectly fair. But as I've shown earlier, God needs to be all-knowing, all-good, and all-powerful in order to be a guarantor of justice. So this type of perfect trust is not possible to have with any other being than God. A reflective person will realize this and hence to them a relationship with an almost perfect being will indeed feel differently than would a relationship with God (i.e. there will also be a slight reason to doubt such a being).

Another objection is to question the claim that having a relationship with God is valuable in the first place.Footnote ¹⁴ Schellenberg's hiddenness argument establishes that if God exists then it would be bad if she deprived us of a relationship with herself. It doesn't show that lacking such a relationship in a world where God doesn't exist is itself bad. Perhaps the incredibly asymmetrical nature of such a relationship is bad. However, this is not what's at stake here. At the outset I stated that I was simply going to assume that the goods typically associated with atheism are indeed goods. Likewise, I'm also assuming that the goods typically associated with theism are goods. I'm focusing on whether these goods can obtain all at once in an atheist world, not whether they are in fact goods. Those questions are quite important and much of my work elsewhere addresses them, but that's not what's in question here.

Sacrifice

The final disadvantage of theism I'm going to discuss is about sacrifice. I've suggested that if theism is true, then genuine sacrifices aren't possible (Lougheed (Reference Lougheed2020), 144–154). This is because for an individual to make a genuine sacrifice for someone else it has to be the case that she is irrevocably harmed by the sacrificial action. If she isn't, it's not really a sacrifice. But on some conceptions of theism God's existence is thought to entail that there are no instances of gratuitous evil. This in turn might entail that genuine sacrifices are impossible. But that the possibility of genuine sacrifice is lost on theism is a disadvantage because sacrifices represent great acts of love. Furthermore, many conceptions of theism hold that there is a good afterlife. If the person making the sacrifice gets to experience a good afterlife, then no matter what sacrifices she makes in her early life they're not really irrevocable. For eventually she's going to receive the compensation of a good afterlife.

Again, it's unclear that this purported disadvantage of theism can be avoided in the atheist world with ten lesser gods. If some of the lesser gods are close to all-powerful and all-loving, then just by virtue of their natures they will be inclined to prevent instances of gratuitous evil. Of course, this is a good thing in itself, but it implies that genuine sacrifice might not be possible. If the lesser gods are not quite powerful, loving, or knowledgeable enough to prevent every instance of gratuitous maybe there would be some room for genuine sacrifice in their world. Of course, in a naturalistic atheist world there could be genuine sacrifices on the assumption that there is no afterlife. But then any of advantages tied to the ten lesser gods (i.e. the theistic advantages) are lost.

Finally, God's existence on its own does not entail that there is a good afterlife. The same is true of the ten lesser gods. So there is the possibility that a good afterlife cannot be appealed to as a way to show that genuine sacrifice is impossible in the world with ten lesser gods. While the proponent of Kahane's argument might say this is one area where the atheist world is clearly better, the theist can also simply deny that a good afterlife is entail by theism. No progress has been made in defending (3) or (7).

The methodological challenge

The above discussion isn't an exhaustive survey in a number of different ways. First, I don't examine all the disadvantages of theism. Still, I hope my survey is representative in that the reader can see how similar worries could be applied to any (or most) purported disadvantages. Second, the reader may wonder why I have referred to the same supernatural atheist world (i.e. the world with ten lesser gods) throughout my survey. An exhaustive survey (or at least a more exhaustive one) would examine a variety of different atheist worlds in trying to discover whether theistic advantages can be gained while the disadvantages are simultaneously avoided. But the simultaneously part is important here. For a defence of the argument to be successful it has to be the case that the same atheist world could avoid all (or most) of the costs of theism while also providing all (or most) of the benefits. This is because the argument needs it to be the case that all (or most) of the benefits obtain while all (or most) of the costs are avoided in the same atheist world. So the proponent of the argument can't respond to my challenge here by describing a number of different worlds where individual advantages obtain and disadvantages are avoided. It would be a feat of conceptual engineering to offer a detailed description of one atheist world where the theistic benefits Kahane wants obtain, while the costs are avoided. Part of what I've tried to show is that this is likely to be more much difficult to accomplish than it might seem at first glance. Indeed, I've offered some reason to think it's not possible, though I'm curious to see whether someone in the future can accomplish it.

One way one to accomplish this task without all of the conceptual engineering that I suggest above is required is to simply say that the theistic goods in question could be a brute fact of some atheist world. I've offered reasons in the previous section to doubt that an impersonal force or mechanism could be responsible for certain theistic goods (e.g. cosmic justice), but the appeal to brute fact is different. It doesn't necessarily imply any kind of impersonal force or mechanism. Maybe the universe just so happens to exist in such a way that all of the relevant goods in question obtain. This is, I think, a promising and currently underexplored avenue. Kahane clearly gestures at these considerations but he never really develops them. Suppose I am right that the impersonal mechanism option is unpalatable so Kahane (or someone trying to defend him) needs to appeal to brute facts instead. Well, what is a brute fact? Are they even possible? If they are, can certain brute facts be incompatible with others such that they can't coexist in the same world? These and related questions need to be answered to fill out Kahane's account and avoid the criticisms I've raised throughout this article. Just as I'm curious to see whether the more complicated conceptual engineering route could be successful, I'm equally curious to see whether an appeal to brute facts could be used in order to get of the theistic goods into an atheist world. In this sense the reader need not take my project here as an attempt to offer decisive criticisms of Kahane, but rather as a call for additional information and arguments.