#import "@preview/unequivocal-ams:0.1.1": ams-article, theorem, proof #import "@preview/wordometer:0.1.3": word-count, total-words #show: ams-article.with( title: [The Argument for Betting on God and the Possibility of Infinite Suffering], bibliography: bibliography("refs.bib"), ) #set cite(style: "institute-of-electrical-and-electronics-engineers") #set text(fractions: true) #show: word-count.with(exclude: (heading, , table)) #align( center, table( columns: (auto, auto), [ Perm: A2V4847 ], [ Word Count: #total-words #footnote[Figure computed programmatically during document compilation. Discounts content in tables and the AI contribution statement.] ], ), ) = Introduction The argument for Betting on God says that you should believe in God, regardless of other evidence, purely out of self-interest. In this paper, I challenge this argument by assessing the premise that believing in a particular God always guarantees the greatest expected utility. The author's argument for belief in God #cite(supplement: [p. 38], ) goes as follows: #pad( x: 16pt, [ #set par(first-line-indent: 0pt) #smallcaps[The Argument for Betting on God]: #pad( y: -2pt, [(BG1) One should always choose the option with the greatest expected utility ], ) #pad( y: -2pt, [(BG2) Believing in God has a greater expected utility than not believing in God ], ) #pad( y: -2pt, [(BG3) So, you should believe in God ], ) ], ) BG1 should be uncontroversial. If you expect an action to bring you the most utility (i.e. be the most useful), why wouldn't you choose to do it? To justify BG2, the author uses a so-called "decision matrix" to compute the expected utility of each combination of action and possible outcome. The possible actions are placed on the rows, and the possible outcomes are placed on the columns, except for the last column, which is the calculated expected utility. At each intersection of a row and column, we place the utility we gain from that combination of action and outcome. The expected utility for a given action is computed by multiplying the utility of each action-outcome pair in that row by the probability of the corresponding outcome occurring, and summing up all of those values. Here is the decision matrix the author proposes on #cite(supplement: [p. 38], ) which gives the expected utility for believing or not believing in God. #show table.cell.where(x: 0): strong #show table.cell.where(y: 0): strong #figure( caption: [Author's decision matrix], align( center, table( columns: (auto, auto, auto, auto), table.header( [], [God exists ($50%$)], [God doesn't exist ($50%$)], [Expected utility], ), [ Believe in God ], [$infinity$], [2], [$infinity$], [ Don't believe in God ], [1], [3], [2], ), ), ) Note that the numerical utility values themselves have no meaning, and they are meant to be viewed relative to each other. Utility doesn't literally provide an empirical measure of "usefulness" or "happiness." We assign the various finite utilities as we see fit, based on how much each scenario benefits us. In the case where God does exist, and you believed in God, then you are rewarded with an eternal afterlife of bliss and pleasure in heaven. This reward is infinitely greater than any possible reward on earth, so it has a utility of $infinity$. So, the expected utility for not believing is $0.5 times 1 + 0.5 times 3 = 2$, and the expected utility is $0.5 times infinity + 0.5 times 2 = infinity$. If, according to BG1, you should pick the option with greatest expected utility, clearly you should choose to believe in God, because the expected utility is $infinity$. The exact utilities don't matter much, since any finite utility you could gain for atheism cannot possibly be greater than the infinite expected utility of believing in God. Also, as the author points out on #cite(, supplement: [p. 40]), the exact probabilities don't matter either since multiplying them by $infinity$ still results in the expected utility of $infinity$. I will show that the Argument for Betting on God fails because BG2 fails. In section 2, I argue you cannot determine whether or not believing in God has the greatest expected utility because the decision matrix approach fails when possible outcomes involving infinitely negative utilities are introduced. In section 3, I address a possible response to this objection. = Possibility of Infinite Suffering It is possible there are more gods than just the one that sends you to an eternal afterlife for believing? The author partially addresses this in #cite(, supplement: [pp. 43-44]), using the example of Zeus. Zeus will only reward those who believe in him with an eternal afterlife of pleasure. So, if you believe in the wrong god, you don't go to the afterlife. The author concludes either believing in Zeus or the Christian God still has expected utilities of $infinity$, while being an atheist does has a finite expected utility. Therefore, it is still preferable to believe in _some_ god that may grant you an eternal afterlife, although no argument is made for _which_ god. However, this leaves out the possibility of gods who punish you for some reason. For instance, suppose there exists an _Evil God_ who sends anyone who believes in any god to hell for eternity, and does nothing to atheists. Let us modify our decision matrix to model an outcome where the Evil God exists. #pagebreak() #[ #set figure() #figure( caption: [Possibility of an Evil God], table( columns: (auto, auto, auto, auto, auto), align: center, table.header( [], [Correct god exists ($33.3%$)], [No god exists ($33.3%$)], [Evil God exists ($33.3%$)], [E.U.], ), [ Believe in some God ], [$infinity$], [1], [$-infinity$], [$?$], [ Don't believe in any God ], [2], [3], [4], [4.5], ), ) ] We've added the new option to our matrix. For the sake of argument, let's say each option has an equally likely outcome. Again, the exact probabilities don't really matter when we're multiplying them by infinity. The utilities are mostly the same as before. Not believing in any god and the Evil God existing is now the best case for the atheist since they avoided infinite suffering. However, the theist now faces the possibility of the worst case of all: eternal punishment for believing in the wrong god. If eternal bliss in heaven has a utility of $infinity$, then it follows that we should represent eternal punishment in hell with a utility of $-infinity$. There is a problem: how do we calculate the expected utility of believing in god? We have $0.333 times infinity + 0.333 times 1 + 0.333 times -infinity$. What is $infinity - infinity$? A naive answer might be 0, but infinity is not a number in the traditional sense. It makes no sense to add or subtract infinite values. For instance, try and subtract the total amount of integers ($infinity$) from the total amount of real numbers (also $infinity$) #footnote[Famously, this infinity is "larger" than the infinite number of integers in the sense of cardinality (G. Cantor). But subtracting them still makes no mathematical or physical sense.]. Clearly, this notion is meaningless and we cannot obtain a solution. So, we consider $infinity - infinity$ an _indeterminate form_. So, the expected utility is now _undefined_. Consider the following argument: #pad( x: 16pt, [ #set par(first-line-indent: 0pt) #smallcaps[The Indeterminate Utilities argument]: #pad( y: -1pt, [(IU1) If the expected utility of believing in god is undefined, then we cannot compare the expected utilities of believing in god or not believing in god.], ) #pad( y: -1pt, [(IU2) The expected utility of believing in god is undefined.], ) #pad( y: -1pt, [(IU3) So, we cannot compare the expected utilities of believing in god or not believing in god. ], ) #pad( y: -1pt, [(IU4) If we cannot compare the expected utilities of believing in god or not believing in god, then we cannot determine if believing in god has a higher expected utility than not believing in god. ], ) #pad( y: -1pt, [(IU5) So, we cannot determine if believing in god has a higher expected utility than not believing in god. ], )], ) We just showed why the premise IU2 is true, and the conclusion IU5 is in direct contradiction with BG2. So, if IU5 holds, then BG2 must fail. It's important to note that the Indeterminate Utilities argument doesn't say that the _opposite_ of BG2 is true. It doesn't argue that the expected utility of being an atheist is greater. In fact, it doesn't say anything about the expected utilities, except that they cannot all be compared. If they cannot all be calculated, then they cannot be compared. So, we can't say whether believing in god has a greater expected utility or not. = Addressing Objections // == Believing in a god is still preferable to atheism // // One might argue that believing in a god that rewards believers is always // preferable to atheism since you at least have the _opportunity_ to receive // eternity in heaven. Perhaps there exists a god who punishes non-believers with // eternal damnation. Then, even without the exact expected utility calculation, // it's clear that the expected utility of believing in some god must be higher // than believing in none as you stand to gain more. Either as a theist or // atheist, you run the risk of eternal punishment, but you only have the // opportunity to go to heaven by believing in some god rather than none. // // Fair, the possibility that you are punished for believing in the wrong god // doesn't imply that you should be an atheist either. Indeed, there may be a god // that punishes atheists. However, there could also exist a god who sends // everyone to heaven regardless. Or perhaps they only send atheists to heaven. // Either way, there is also the possibility of attaining the infinite afterlife // in heaven by being an atheist, so it's still impossible to say that the // expected utility of believing in god is must be higher. == The Evil God is not plausible One might argue that it is not plausible there is an Evil God who punishes all theists, including their own believers. Many religions present a god that rewards believers and at most punishes disbelievers. None of the major world religions propose an Evil God who punishes all believers. It's much more likely that a benevolent god exists than an evil one. I contend that it doesn't matter whether or not the Evil God is less plausible than a benevolent god. Surely, if a rational atheist who is unconvinced by all the world's scriptures can still concede that there is at least a non-zero chance that some god exists, the rational theist should also concede that there is a non-zero chance that the Evil God exists. All it takes is that non-zero chance, no matter how small, because multiplying it by $-infinity$ still results in the undefined expected utility. == Finite utilities One might argue that we can avoid using $infinity$ to ensure that all expected utility calculations are defined. Instead, suppose that the utility of going to heaven is just an immensely large finite number. The utility of going to hell is likewise a very negative number. All of our expected utility calculations will be defined, and given sufficiently large utilities, we should be able to make a similar argument for believing in god. // The problem with this argument is that we now open our expected utilities up to // individual subjective determination. A core feature of the previous argument // involving infinite utilities is that they can effectively bypass numerical // comparison. If, instead, finite utilities were used, then each person may // assign different utilities to each possible outcome based on their own beliefs. // Also, the probabilities are no longer irrelevant, so they must be analyzed as // well. This greatly complicates the decision matrix. The problem with this argument is that infinity has a special property argument relies on. Namely, any number multiplied by $infinity$ is still $infinity$, so the exact probabilities we set for the existence of God don't matter. This is important for defending against the objection the author mentions on #cite(, supplement: [p. 40]), that the probabilities are possibly incorrect, since the numbers don't matter anyways. If, instead, only finite utilities were used, then the theist must contend with the concern that the probabilities in the matrix are wrong. There could conceivably exist a matrix with probabilities for a benevolent god and an Evil God such that the expected utility of atheism is actually higher. The issue is we cannot say for sure what the probabilities of the benevolent god and the Evil God existing are. If we cannot know what the actual probabilities are, then we cannot know the final outcome of our matrix. So, without knowing the final outcome of the matrix, we still cannot determine whether or not believing in god has greater expected utility, and BG2 still fails. #pagebreak() #[ = AI Contribution Statement #quote[I did not use AI whatsoever in the writing of this paper.] ]