Thursday, December 15, 2022

The Failure of Theodicy

 Note; Once again, this was originally a reddit post. A relatively recent one. I do some edits, and add 2 new sections.

Introduction


Greetings, viewers of my blog. Today, I return with some reflections on a few relatively recent theodicies, I've come across that I think are difficult to respond to. What is a theodicy? To understand this, we will first need an understanding of the problem of evil.

The problem of evil is an argument for the non-existence of God ('God' here is understood as a being with omni-attributes, omnipotence, omniscience, omnibenevolence) which starts with some facts or observations, of evil, or suffering we see in the world. Be it the occurrence of evil and suffering generally, or more concrete horrific examples of evil or suffering such as a Fawn burning to death in a forest fire (Rowe 1979). There is taken to be two kinds of problem of evil. An evidential, and a logical version. A logical problem of evil, argues that theism and the occurrence of evil are jointly inconsistent. The basic form of such an argument would be as follows.

1. If evil exists, then it's not the case that God exist
2. Evil exists
3. Therefore it's not the case that God exists


An evidential problem of evil on the other hand, argues that the occurrence of the evil we see (Usually, in particular, horrendous, seemingly gratuitous suffering) is highly unlikely on theism. That is, that given the truth of theism, antecedently, it would be very surprising if it turned out there were evils of the sort we observe in this world. There are many ways to formulate such an argument that atheist philosophers have come up with, there are inductive formulations, intuition-based formulations, Bayesian formulations etc. Here is a basic evidential argument from evil.

1. If God exists, then gratuitous evil does not exist
2. Probably, gratuitous evil exists
3. Therefore, probably, God does not exist

Where gratuitous evil is taken to be evil with no morally justifying reason for it's existence. Or evil which does not entail a greater good.

There are two ways theists tend to respond to the problem of evil. One would be a defense, another would be a theodicy. A defense is a response to a logical problem of evil, it is a story one can tell, which, when conjoined with theism, makes theism and the existence of evil compossible. The most famous example is Alvin Plantinga's free will defense (Plantinga 1978). It's important to note that, a defense, does not need to be remotely likely, or plausible, it only needs to be true for all one knows and not entail a logical contradiction.

A theodicy, on the other hand, is a response to the evidential problem of evil. It is a story which is taken to be both plausible given the truth of theism, and when added to theism, makes the occurrence of the evils we see unsurprising from a probabilistic standpoint. Unlike a defense, a theodicy must be a story which is plausibly actually true, not just true for all one knows. To be a successful theodicy, much more is needed then mere logical coherence since what needs to be shown is not merely that theism is consistent with evil, but that there is no significant evidential tension between theism and observations of evil. A standard example of a theodicy is John Hick's soul-making theodicy. I will be taking a look at a more recent kind of soul-making theodicy in the final section. But, for the most part, I will be addressing more recent, non-traditional theodicies, that have been discussed on the philosophically sophisticated theist YouTube-Sphere. It is to this, that I now turn.


The Non-Identity Theodicy


The first theodicy I will examine is the non-identity theodicy. Before reading this section, I highly recommend watching Apologetic Squared's short 7 minute video on this theodicy or reading the original paper on it. I will just give a brief sketch of it.

The basic idea is this; what makes me, me, and you, you, is our essential properties. And, the story goes, our essential properties include (and are possibly exhausted by) facts about our origins, indeed very specific facts about our origins, including the total causal history which has resulted in ones coming-to-be, our ancestral history, and evolutionary history. But this history includes a lot of suffering. So, God permits this suffering to bring about you, and me, that is you and me in particular. If God did not allow such suffering, then you wouldn't exist, rather someone else would, with some other set of essential properties. God's intention is to bring about us in particular, you, me, Jim and Bob across the street, etc. He views bringing about us in particular, as a valuable end in itself, such that the means he uses to achieve such an end, which includes vast amounts of horrific suffering, makes it all worthwhile! This theodicy is a fascinating one, but one I also think is highly implausible and faces some, by my lights, powerful objections. We will now turn our attention to those.

Objection 1

The non-identity theodicy requires the truth of origin essentialism. This seems reasonably plausible on it's own. It seems like, at least partly, what would make me numerically distinct from an identical clone of myself would be my origins. Perhaps my clone was created in a test-tube using my DNA, whereas I was created through regular human reproduction etc. So my origins are an essential property of me, if those weren't my origins, I wouldn't be me. The problem is, the theodicy is more radical, it requires that the entire causal history prior to my coming-to-be, including facts about suffering that do not seem to have any relevant causal connection to my origins, are essential to me. This strikes me as unintuitive in the extreme. While an object like you are me, could not have a entirely different origin, it seems like we could have a slightly different origin. That is, it seems like some degree of modal tolerance is permissable. Suppose I didn't believe my great grandfather fought in the first world war, but then my father tells me he did. It doesn't seem, in the slightest, that I'm learning about who I am, that is, learning about my essential characteristics. Or that, if it later turned out my grandfather did not fight in WW1, I would be finding out that I'm a different person than I previously thought I was, that seems utterly bizarre. On the other hand, if I found out that I am a brain in a vat, or a clone created in a test-tube, it seems like I would indeed be shocked to find out that I am not who I thought I was! But then, if not every fact about our origins is essential, what facts are and aren't? On reflection, one answer with some plausibility, is the essential part of my origin is the event of my father's sperm fertilizing my mother's egg, and the set of pre-natal embryonic states which followed etc. That is, what is relevent to an object or persons origins, is the causal history of the material out of which it was originally constituted. What makes the desk on which I am typing, *this* desk, is the history of the wood and nails which constitutes it. A view such as this is often used in tandum with Kripke's causal theory of reference. For a technical paper on this see here.

But suppose we even grant (as I would not) that every state of affairs which forms part of the causal chain ultimately leading to my birth, stretching back even to specific facts related to our evolutionary history, are an essential part of my origins as the argument needs. There is now a further question. Why then, couldn't the events which feature creaturely suffering, have radically different psychophysical connections? Were there a God, He could have easily set it up so that basically all the physical events in the causal chain leading up to my and your creation are the same as they are in this world, but the sensations of pain experienced internally by the creatures and persons were not horrifically agonizing, or at least greatly less so. So your origin and my origin is surely the same or same enough, our existence/essential properties would be preserved. Yet, the extremely undesirable experiences felt by persons and sentient non-persons leading up to us, would be greatly diminished. Surely God could have, indeed, would have, actualized that world, and as far as this theodicy goes, it appears to be left a mystery as to why He did not.

The theodicist might conceivably respond, that actually if God were to change the psychophysical laws, He would have to change the physical events, since those phenomenal states, necessarily supervene on those physical states. But, for one this is highly implausible on the theists own view, since presumably God has phenomenal states prior to the creation of any physical states, and for those who maintain an afterlife, humans as well can have phenomenal states despite no longer having a physical body. So, it would seem phenomenal content does not necessarily depend on certain physical states, you can have different phenomenal contents despite changes in physical states. Further, it seems like, in principle, an omnipotent God should be understood as having complete control over the causal laws factive to His creation. Finally, even if a change in the specific phenomenal contents like pain, entails a change in physical states, it seems like only minor changes, utterly irrelevant to our material origins are required. The theist would need to motivate the prima facie implausible claim that changes relevant to our material origins are required, that isn't just an ad hoc move to save the theodicy.

Scott Hill has a paper on the non-identity theodicy, where he attempts to motivate a radical origin essentialism. I will address this in turn.

Scott characterizes the origin essentialism he accepts as follows.
"Assembly Origin Essentialism: If the materials from which a creature originated were assembled by a process that was too different, then that creature would not have existed."
At first blush, there doesn't seem to be an issue. But, it is left ambiguous what is precisely meant by "too different". Scott and I, are going to disagree on the threshold. Be that as it is, Scott gives 3 justifications for why an Assembly Origin Essentialism should be strong enough to motivate the non-identity theodicy.
"My first argument is that intervening in the world in the ways described in Waiting and Painless Evolution constitutes changes that are too big. At most a proponent of Assembly Origin Essentialism should allow a few minor changes to the process by which the materials from which I arose are assembled such as the one in Forbes' example. On the other hand, the miracles required in Waiting and Painless Evolution require big changes to the process. So my origin is not preserved in such cases."
What Scott seems to be imagining here is God constantly making significant interventions, and performing miracles. This, however, is not a requirement. As discussed above, God does not have to make significant interventions, indeed, it doesn't seem like God has to intervene at all, He could have set it up so, prior to the initiation of the given causal chain, the events in the causal chain qua creaturely suffering, are linked up with less horrifically painful phenomenal states, or there are different psychophysical laws of nature. Otherwise, God could make small adjustments, so that the history of the materials leading to me, or you, are the same or similar enough, but there isn't such a frequency, intensity, and distribution, of horrific suffering among persons and non-persons. Afterall, it doesn't seem like a T-Rex mauling a Brochosaurus alive, or even my mother's pre-birth agony, is absolutely necessary for me, or Bob's, or anybody elses origins. The material out of which we originated could have an identical, or, intuitively, an identical enough history, such that I am still me and you are still you, sans such horrors. But even if we grant that God needs to make really major changes or interventions, it just needs to be logically actualizable that, in light of such major changes which are minimally sufficient to prevent horrific suffering, the history of our material origins could still be similar enough so that we would still exist. This seems logically possible, and since it seems logically possible, it seems like a state of affairs which God can actualize. So there are further burden-shifting grounds to reject Scott's claim.
"My second argument is that accepting such a strong variant of origin essentialism yields a gain in explanatory power... adopting a version of origin essentialism that rules out Waiting and Painless Evolution forms part of a theodicy that explains why God allows evil. That is a significant gain in explanatory power for the theist."
I just have a couple points here. For one, it's debatable whether the non-identity theodicy does have such explanatory power, I've already offered a couple reasons to think it does not, and I will provide more reasons. For two, even if adopting such a view does explain the occurrence of evils under the truth of theism, as a non-theist, I do not grant that this is a gain in explanatory power tout court.
"My third argument is that adopting such a strong version of origin essentialism eliminates vagueness... There seems to be no principled way to say when an origin becomes so different that it fails to preserve the existence of an organism other than never or always. And if we are antecedently attracted to Assembly Origin Essentialism, it isn’t plausible to hold that a change in origin never yields a different organism. So we should say that any change at all in the process by which the materials of my origin were assembled yields a different mere duplicate of me rather than me."
This argument has more bite than the other two, but I think it is defeasible. I agree that if you hold that any change whatsoever in the causal history of ones origins is a change in the essential properties of some object or organism then that does indeed eliminate vagueness and arbitrariness, and, prima facie, this does seem to be a problem for views which are restrictive. But, so too would positing that having any strands of hair, constitutes non-baldness, or positing that any collection of 2 or more grains of sand constitutes a pile. We have an intuitive, conventional understanding of what a pile is, and what non-baldness is, and these views, while having the virtue of eliminating vagueness, simply do not track our concepts. This is analagous to radical origin essentialism, it simply doesn't track my, and I think our shared intuitive understanding of what makes us, us, and I take myself to have illustrated why this is so, above. If vagueness is a cost of maintaining such a view, then it's one I gladly pay, with the hopes that one day we do discover a restrictive criterion for essentiality of origins, which avoids issues of vague predicates.

Objection 2

Suppose we forgo the first objection, and agree that a very radical kind of origin essentialism is true. Even so, I believe the plausibility of the theodicy is parasitic upon a subtle sort of equivocation. It rests on the intuition that the existence of you, and you in particular, your mother and father, your best friends and the love of your life, and those people in the actual world who satisfy those descriptions in particular are valuable in themselves, for their own sake. But suppose I asked you what makes them valuable, or at least, why you value them so dearly. Plausibly, you'd list out the set of things they've done for you, the bad times they were there to help you get through, the virtues and other admirable qualities they embody, possibly some physical features if you know what I mean. Among other things in that ballpark. It would be very odd for you to list off a specific set of essential properties related to their origins like that a great ape was eaten alive by an alligator some millions of years ago. In fact, if you were to do that, I'd probably try to contact a mental health clinician. What this tells me is, while it can be granted that there are essential properties related to my origins that make me, me, rather than someone else. These essential features are, in fact, accidental if we zoom out and ask what stands in the right sort of making-relation in terms of my value properties. If God, in creating, was aiming towards some highly specific set of essential characteristics and in doing so actualizing states of affairs that are technically necessary to make us who we are, rather than the second order goods that plausibly make us valuable, then it seems like God's intentions are not contrastively, aimed at the right sort of end.

A proponent of the theodicy might be tempted to argue, that actually, there is nothing further that makes me valuable, it is irreducible. I'm not valuable in virtue of exemplifying certain goods, I'm just valuable as such. Even this, however, won't do. If the value is irreducible, then God could have actualized such irreducible value properties without actualizing my origins. What's needed, is that there are irreducible value properties which somehow necessarily supervene on states of affairs related to my origins. So what makes me valuable, is my essential properties qua my origins. This again, just doesn't appear plausible on reflection and would need to be motivated.

But even if it is motivated, there is of course an issue of whether the value of contrastively actualizing me, rather than a qualitatively identical counterpart with a different origin that doesn't involve as much horrific suffering, is so great that it outweighs the aforementioned horrific suffering which was permitted. This, of course, is even more implausible. Granted, Vince Vitale, and Scott Hill, for this reason seem to grant that this theodicy doesn't do the work on it's own, but rather only when supplemented with the fact that creaturely evils will be defeated in an afterlife. More on this later.

Objection 3

The last objection I will present will be an undercutting one, from Kantian (or other deontological) duties. The theodicy seems to be motivated by a kind of deontology, wherein God is permitting creaturely suffering for the sake of creating me, and you, whom are ends in themselves, in the Kantian sense. In actuality though, this theodicy is the furthest thing from an endorsement of Kantian ethics, and is in fact radically consequentialist. What God is doing, according to the non-identity theodicy, is permitting the agonizing suffering, death, violation of autonomy and destruction of faculties, of billions of rational agents, and non-rational sentient agents, as a mere means to a heteronomous (another Kantian term) end. That is, he treats rational agents as a mere means to bring about other rational agents with certain essential traits, He fails to treat rational agents as ends in themselves. No contemporary Kantian in the known universe would find such actions tolerable, and Kant himself would most likely be rolling in his grave. As a related point, even the vast majority of consequentialists would find the kind of normative framework this theodicy requires implausible in the extreme. It's not like God is maximizing pleasure, or happiness, or any metric of utility consequentialists would find plausible, rather he permits horrors just to bring about specific persons with specific essential characteristics.

In light of these objections, while this theodicy is an interesting and unique piece of philosophy, I cannot find it plausible, as even a partial justification of the evil we see. That's enough of the non-identity theodicy. Next, we will look at John Buck's Participation Theodicy.


The Participation Theodicy


The next theodicy I will be examining is one created by a catholic self-proclaimed twitter apologist John Buck. Do not get the wrong impression though, he is not a pushover like his description implies and his participation theodicy is a clever one. Here is a paper he wrote on the theodicy, and here is a recent discussion he had on it with an atheist Youtuber.

To briefly summarize, atheists will often present the problem of evil by arguing that God would create a perfect, one might say heavenly world, and we are not in such a world. The participation theodicy starts by granting that the best thing God can do is actualize an ideal world. But, it would be better for God, to use Alvin Plantinga's jargon, not to strongly actualize such a world, that is actualize it with His own power, from the get-go. But to weakly actualize it, by allowing that free creatures participate in the actualization of an ideal world. The catch being, for free creatures to participate in the creation of the ideal world, the world must start out non-ideal, indeed it might start out in the very notably far from ideal situation we find ourselves, with great amounts of creaturely suffering. But God is justified in permitting such suffering, because it is an all-things-considered good that an ideal world is brought about, by virtue of the creative participation of free agents, rather than an act of God's will alone.

Before we start with objections a few things are worth noting regarding the theodicy. The first is that it avoids a lot of the pitfalls of most other responses to the problem of evil, such as moral paralysis, reliance on controversial thesis such as libertarian free will, explanations for natural evils or animal suffering, why a better state of affairs was not actualized etc. It is not without issues, but it is I think a quite impressive feature that it is able to sidestep many of the traditional problems that other theodicies run afoul of. The next is that, the theodicy has a lot in common with the free will defense and free will theodicies. While, it's admittedly better than any free will theodicy I've encountered, and doesn't suffer from some of the same issues as the free will defense, I think it does inherit some of it's follies. The last is, while it doesn't suffer from many of the same issues of other theodicies, it does suffer from a common issue that I call the issue of weighty evils. The participation theodicy requires the intuition that God letting creatures actualize an ideal world is a greater good than God actualizing it Himself from the get-go, this intuition is not widely held. But even if you do share the intuition, it's surely not so strong that God's contrastively actualizing the world where creatures participate in the creation of the ideal world rather than God Himself creating the ideal world, is enough to outweigh the horrors we see. This is especially clear when we reflect on particularly horrific evils like a baby's being born with butterfly disease to live a short earthly life of nothing but pure agony. The intuition pump John uses, to motivate the theodicy, wrt a husband allowing his children to participate in making the wifes breakfast (See his paper for more details) simply does not work for horrendous suffering. However, it may still be that the participation theodicy can be a partial justification and form part of a cumulative case against the problem of evil in conjunction with other partial justifications. My objections will attempt to undermine the participation theodicy as a partial justification.

Objection 1

The first objection I will present is I think probably the most common sense one, and one that Buck is familiar with. In order for the theodicy to be successful, it must be true, or at least plausible, that the actions of free creatures are tending toward the actualization of an ideal world. The problem being, this just doesn't seem to be plausible based on the evidence we have. In some ways, we have made substantive moral progress, technological progress, medical progress etc. but in many other ways things seem to be getting worse. Shootings in the US as of late are at an all-time high, global starvation is (almost) as bad as it's ever been, as is the horrors of factory farming and global predation, a certain party in the US is increasingly more deranged, we have the invasion of Ukraine, and the Isreali-Palestine conflict neither of which seem to be generally improving etc. We don't seem to be seeing a general trajectory towards an ideal state of affairs.

Further, given our limitations, it doesn't seem likely at all, perhaps not even possible, that without any aid from God we could actualize an absolutely perfect world. Even if we do somehow manage to solve every single issue on the planet, and create a harmonious utopia where we live blissful lives, which is already unlikely in the extreme, it seems to me. There are plausibly goods that are beyond the capacities of creatures like us, which only God could access. So, without aid, we will always fall short of an ideal world.

Lastly, these are just issues on our tiny little spec of a planet, there is also, of course, the possibility of horrific suffering (or other related horrors we can imagine, such as disasters on a galactic scale) on other solar systems in the trillions upon trillions of Galaxies which exist in our universe (possibly multiverse). Additionally, there is an expectation of the heat death of the universe. Which means not only are we on a time limit to create an ideal world, but once we do participate in this theater of growth, learning, and ultimately creation of an ideal world, all biological creatures, and everything else in God's physical creation will be destroyed. That is, unless of course God chimes in to prevent it. In which case, it's hard to see what the principled difference would be between God's intervention to prevent the natural death of all, and God's intervention to prevent horrific suffering.

Buck responds to this concern in his paper.
"The theodicy can grant that most creatures are incapable of significantly contributing towards the idealization of the world, but recall that the theodicy suggests we find ourselves in the midst of the process of idealization, not necessarily near its apex. For all we know there could be a billion more years before our world achieves idealization, well past the extinction of the human race. However, the very best things humans could be doing while we do exist would be to contribute what we can towards the eventual idealization of the world, even if that means simply doing what we can to prepare the next generation to be better at their responses to the hardships they will face."

This response, however, is unsatisfying. Recall in the introduction that the job of a theodicy, unlike a defense, is to tell a story that is plausible in order to remove the evidential tension evil creates for theism, not merely true for-all-one knows. I am happy to grant that the story John wants to endorse, that if not humans, some superior species or other will causally contribute to an ideal world possibly billions of years from now, is true for all we know. But, antecedently, it seems to me this story is no more probable than a set of mutually exclusive, jointly exhaustive and equally intrinsically simple stories I can tell which fail to include the actualization of an ideal world. Such as that humans will be the last intelligent species to exist, or that there will be other species that exist but they will all fail utterly at creating an ideal world, or that instead of realization of an ideal world the actions of free agents will ultimately create a hell world (that is the worst world possible) etc. If John does not give a reason to think his story is antecedently more probable than a set of incompatible stories like these, that is importantly, independent from the truth of the theistic hypothesis. Then the probabilities balance each other out, and we are then left with our initial judgement that, all other things equal, we do not seem to be tending towards an ideal world.

Objection 2

Another problem is that, insofar as the participation theodicy is used to explain the existence of seemingly gratuitous evils, it appears to succumb to a serious lack of imagination. There seems to be numerous ways God could have set it up so that the world starts out non-ideal, so creatures can participate in bringing about an ideal world, without having the worst natural and moral evils we see. For instance, God could have created an aesthetically imperfect world with no evil, so creatures can contribute alongside God to improving the world's aesthetic beauty. Otherwise, God could have created a world with moral imperfections but not ones so severe as the overwhelmingly horrific, heart-breaking evils we see in our world.

Objection 3

As I said previously, this theodicy, I think runs afoul of some of the same issues as the free will theodicy and defense. This theodicy relies on a distinction between God actualizing an ideal world, and free creatures being placed in a set of circumstances and participating in the actualizing of an ideal world. In other words, a distinction between God's strong actualization of an ideal world, and weak actualization of an ideal world. I think there are reasons to think the distinction collapses in this case, however.

Following the late and brilliant atheist philosopher J.L. Mackie. It would seem omniscience in conjunction with omnipotence jointly entails, or at least strongly implies, omnifiscence. That is, God is responsible for the strong actualization of every event, including the actions of His created free creatures. Here is an argument for that.
  1. If God is omnipotent, then all events are maximally within His power.
  2. If God is omniscient, then all events are maximally within His awareness.
  3. If all events are maximally within God's awareness and power, then God is maximally responsible for all events.
  4. Therefore God is maximally responsible for all events.
The intuition here is that, the common-sense distinction between doing something, and allowing something to happen does not hold for a being of unlimited power, and unlimited knowledge. Mackie argues that this everyday distinction is motivated by two considerations

A) If we bring something about, there is effort we exert, but if we allow something to happen there is not. The less effort we must exert to prevent the given state of affairs, the more responsible we are.
B) Letting something happen is often associated with some degree of inadvertedness, while bringing something about requires great attentiveness. The more attentive one is, the greater their responsibility is.

But A, and B become inapplicable as power and knowledge increases without limit. It would seem that for a being with unlimited power and unlimited vision, this distinction would not hold at all. So a being that is omniscient and omnipotent does everything. Note this does not mean that free creatures do nothing, but as Mackie I think cogently argues, God is in complete control of people's free choices.

Once we realize the distinction between God actualizing an ideal world, and allowing creatures to participate in the actualization of an ideal world collapses due to God's omnifiscence, Buck is once again open to the question; Why did God not actualize an ideal world from the get-go? If not, at the very least, why did God permit all this horrific suffering? It cannot be because God wanted to allow His creatures to participate in the creation of an ideal world, because if the above argument is right, then all the same, their participation just is His participation.

Even if Mackie's argument is unsound, nonetheless I think we have good reason to think the distinction the theodicy rests on does not do the work Buck requires. We can ask the obvious question, even if free creatures participating in the creation of an ideal world is all-things-considered better than the world where God alone participates. Why then, did God not actualize the world where free creatures never make decisions which hinder the creation of an ideal world, free creatures who always freely choose to do good and who's actions are always directed at the creation of an ideal world, never comiting horrific moral evils. Prima facie, that state of affairs seems to be better than the state of affairs where free creatures often do evil, which we find in the actual world. Perhaps Buck can respond, if God did that, then His creatures wouldn't truly be free, and they would not truly be participating in the creation of an ideal world without the aid of God. But this would be to remove intrinsic goods that would otherwise exist.

This won't do though for a few reasons. First, consider that it seems God could actualize the truth-conditions for counterfactuals of creaturely freedom. A counterfactual of creaturely freedom would be a conditional statement that has the following form:

C*) "If S were to obtain, Smith would do A"

Where S is a placeholder for all relevant states of affairs prior to Smith's choice, and A is, for our purposes, a freely chosen action that has the property of being morally right. It seems God could actualize truth-makers for counterfactuals where free creatures, in the circumstances God actualizes, always do good, and always contribute to the creation of the ideal world. Why might it be the case that God cannot make such counterfactuals true? Afterall, such counterfactuals are contingent truths, and God is omnipotent. Buck might argue that if God, prior to creation, actualizes the truth-conditions of such counterfactuals, then agents would not truly be free, and God would simply be causing agents to do the right thing. This, it seems to me, is not right. For one, God making C* true would not entail that there is a necessitating causal relation between the antecedent and the consequent of the conditional. Yes, God would cause the proposition C* to be true, but this does not entail that the action included in C* is itself causally determined, it could still be freely chosen. For two, even if God did not actualize the truth-conditions for C*, it would still seem that C* or some other mutually exclusive counterfactual would be true even if by happenstance, and so God by actualizing S or any set of circumstances that would constitute the antecedent of the given counterfactual, would nonetheless make it true that Smith does A or A* (whatever actions those specify).

Another reason this won't seem to do, is that it seems, by the theists own lights, God is an essentially morally perfect entity and yet always does the right action. So, it's on the participation theodicist to show that God could not actualize beings who, like Him, are both essentially morally perfect and completely free, in which case God could have actualized a world, where there are none but perfect creatures like Him participating in the actualization of an ideal world.

Finally, it seems like the nature created beings are endowed with, explain why creatures make the choices they do. Since no creature is responsible for their nature, not even God (This is easily proved, if God were to choose to have a perfectly good nature, then God would first have to have such a nature that is motivated to make such a choice. Thus God's nature is logically prior to His choices) then that creatures are, beforehand, actualized with the right sort of natures does not at all take away their responsibility for their actions. So, it's unclear why God could not actualize creatures with natures that are morally perfect, or at least far less disposed towards moral evils in the process of contributing to the ideal world.

Objection 4

The final objection I will present, is that, even if Buck is able to successfully answer the previous objections, the participation theodicy is still, I believe, impotent against the best formulations of the problem of evil which is inference to the best explanation/Bayesian arguments. The one I will be focusing on here, will be Draper's argument from pain and pleasure (Draper, 1989). The basic argument goes as follows;

On naturalism or a hypothesis of indifference, given our background information of evolution by natural selection and that organisms are goal-directed biological systems. Antecedently, what we'd expect is that, generally all functional traits of organisms including pleasure and pain, contribute to the biological goals of reproduction or survival, and the times they don't is explained by the imperfect tuning of our biological processes (E.g burning to death doesn't need to be that painful for the biological goals of an organism to be satisfied). On theism, the inference from the background information of natural selection, to pleasure and pain serving biological functions is undercut, since it needs to also be true that God's desires, or moral reasons, happen to align with pleasure and pain happening to align with the biological goals of survival and reproductive success. Antecedently, we have some reason to expect God to treat pleasure and pain differently then other functional properties, since pleasure and pain have moral value, and God being morally perfect, ceteris paribus would be expected to have compassion for His creatures and not permit their horrific suffering unless He has a morally justifying reason to do so. The theist can tell an auxiliary story to explain why, on theism God permits horrific pain and allows that pleasure and pain happen to serve biological functions, but of course this story will inevitably be much less virtuous then the naturalists, more straightforward explanation.

The participation theodicy is utterly powerless against this argument because it only explains general facts, such as why we are not in an ideal world, why there are horrific instances of suffering. It gives us no reason to antecedently expect specific facts about pain and pleasure being tied to the goals of organisms etc. While it can explain the general fact of why there is horrific evils, it provides no explanation for such specific facts. Thus, it fails to address Draper's argument.

Interestingly, Buck, in his conversation with Emerson Green, responds to Draper's argument. I'll address a couple of his responses.

Buck argues the theodicy will grant that pain and pleasure being morally correlated would be an ideal world, but on the theodicy we will start out with a non-ideal world, so it's fine that we start out with pain and pleasure being biologically and non-morally correlated. Granted, on Buck's theodicy it could be that we start out with a world where pains and pleasures happen to serve biological rather than moral roles, but all the same this gives us no reason, antecedently to expect such a link between pain and pleasure and these biological functions. God could have actualized a world where pain and pleasure contribute to moral goals rather than biological goals, are causally related to any number of other physical states or functions then the ones we see, or are not caused by physical states at all. When God decided to create creatures who will contribute to the ideal world, why is it that God's desires, goals, or moral reasons just so happen to align with actualizing creatures with the precise psychophysical connections we see? In-fact, I contend that Buck's theodicy, if anything, makes this data even more surprising on theism, since if God's goal was to create agents so that they can participate in causally contributing to the creation of an ideal world, then we'd expect God to make pain and pleasure to serve roles which make agents grow mentally, and morally and cause them to cultivate behaviors or dispositions which are conducive to their contributing to create an ideal world. We do not see this. On naturalism, given natural selection is a part of our background knowledge, we do in fact have such an antecedent reason to expect these psychophysical connections since, once again, given that biological organisms are goal-directed systems developed through the process of evolution via natural selection which tends towards producing traits that service biological goals such as survival and reproduction, we'd expect to see states such as pain, and pleasure align with physical states with causal roles which serve such biological goals.

He also argues that because epiphenomenalism is consistent with naturalism and evolution via natural selection, this undercuts the inference we would be able to make on naturalism and evolution that pain and pleasure will have these correlations. I think this both misunderstands Draper's argument, and epiphenomenalism. The argument holds, not that it is necessary that pain and pleasure will have such biological functions and be correlated with certain physical states, but that we can inductively reason from our background information that, since all these other functions (including mental properties) are correlated with biological goals, we have good antecedent reason to expect pain and pleasure to work the same way. Further, epiphenomenalism holds that mental states themselves do not have causal powers, not that they cannot causally supervene on physical properties, all our evidence suggests that they do so causally supervene.

The Multiverse Theodicy


Another theodicy I want to look at is the Multiverse theodicy. The basic idea of such a theodicy is that God created not just our universe, but infinitely many universes all of which are better to create than not. So, the answer to "Why didn't God create a better universe then ours" is that, He did, but He also created our universe, because, though it contains horrendous evils, it was still "good enough" so that it is a universe worth making. Here is a good video on it. I will lay out some quick objections.

Objection 1

While our universe may be worth creating despite the evil contained within it. It nonetheless seems contrary to God's perfect benevolence to passively allow creatures to suffer terribly, for seemingly no good reason. From a deontic perspective, it seems like he would be wronging those creatures, treating them as mere means. Further, if God's goal is to maximize the good, it seems like, rather than ever creating this universe, God would create infinite duplicates of universes otherwise qualitatively identical to, or similar-enough to ours, but the counterpart sentient beings in said universes do not undergo horrific suffering (Monton 2010). The multiverse theodicist might reply that, God did create universes-like-ours but with no horrific suffering, all the same the world would be better if it contained our universe as well. While it's true that God could add to the value of the world by creating our universe, He could also do an even better job by instead creating another duplicate or near-duplicate of an otherwise qualitatively identical universe-like-ours with no horrific suffering. Any time God chooses to create a universe, He has a choice between actualizing our universe, or a duplicate or near-duplicate of a universe otherwise qualitatively identical in all or most respects, but which does not contain horrific suffering. Since the latter option will always seem to be the better state of affairs, God will never have reason to create our universe, which includes horrific suffering. From both a deontological and consequentialist perspective the multiverse theodicy does not seem like a satisfying answer to the problem of evil.

Objection 2

Another problem with the multiverse theodicy is that it is not possible for God to actualize all good possibilities since they are not compossible. There are many good ways my life could have been at this time. I could have chosen to be a doctor at this point in my life, or I could have chosen to be a humanitarian activist and help refugee's, or war victims in the Middle East at this point in my life. Both of these cannot be true. God could make counterparts of me in some other universes, where I choose both lives at this point in time. But that person couldn't actually be me, since right now I'm here in this universe, and I unfortunately chose neither. So, since it isn't anyways possible for God to create all good possibilities, on theism, it seems like what we'd expect is God to give everyone the best kind of lives, or at least, very good lives devoid of undesirable hardships, and horrific pain at any time, and any universe they exist. As we look around, we see this is not the case. Credit to Michael Huemer for making this point in his book Knowledge, Reality and Value.

Objection 3

The last objection I will consider, is that even if we forgo the first 2 objections, and we grant that the multiverse theodicy shows that theism and horrendous evils are consistent. It nonetheless does nothing to resolve the evidential problem of evil. It might be the case that given theism there is bound to be a universe with evil in the multiverse, because some universes worth creating contain evil, we might even grant that the chance that evil obtains and is observed in some universe or other is 1. Still, the probability of being in a universe where evil is observed would be extremely low on theism, since the vast majority of good universes would be universes without evil. Even lower would be the probability of observing particularly horrendous evils like the holocaust. If we think, as I believe is plausible, that God creating a universe with horrific suffering is antecedently very unlikely, while it might be the case that God creates some universe which has horrific suffering, it's very unlikely that we would be in that universe. Consider this analogy; there are 3 houses on a block. House A is a giant house that will hold a thousand people, House B will hold five, and House C will hold one. You (and 1006 other people) are randomly assigned to houses. Which house should you bet on being assigned to? The answer is obviously House A, there's over a 99% chance you will be one of the thousand people assigned to House A. Analogously, since God will be expected to create good universes, and the wide majority of good universes will have no horrendous evil, you should expect on theism to end up in a world with no horrendous evil. Therefore, even granting the multiverse theodicy, the observation of evil, and in particular, horrendous evil, is very improbable on theism. Thus the theist hasn't resolved the evidential problem.

Saint-making and The Defeat of Evil


Last but not least, we come to Trent Dougherty's Saint-Making Theodicy expanded on in his book The Problem of Animal Pain. The theodicy comes in two "phases" the first is that evil is defeated or "defeasible". Marilyn Mccord Adams defines a defeat of evil, where E is some evil;
"E is defeated within the context of the individual’s life if the individual’s life is a good whole to which e bears the relevant organic unity ... by being relevantly integrated into x’s relation to a great enough good."
Additionally, the evil is such that the individual endorses the life they lived and does not wish away their experience of the evil. As Adams states;
"Retrospectively, I believe, from the vantage point of heavenly beatitude, the victims of horrors will recognize those experiences as points of identification with the crucified God, and not wish them away from their life histories."
The second phase of the theodicy, is basically a soul-making theodicy but one which is taken to include animal universalism as well. The story of our lives both humans and non-human animals, will continue on in heaven, in which all will become saints. The evil states of affairs we observe, are such that they fall into a certain range which allows all creatures to experience moral growth and develop virtues, including the highest virtues exemplified by the saints, such as; courage, compassion, kindness, generosity, benevolence, mercy, magnanimity, tolerance, honor, truthfulness, trustworthiness, responsibility, friendship, cooperation, diligence, discipline, helpfulness, gratitude, empathy, forgiveness and self-sacrificial love. Even for those particularly horrific evil states of affairs which we observe on earth, which do not seem conducive to virtue-building and character growth, rest assured in the next world a saint-making process will take place. This is possible so long as the evil isn't 'overwhelmingly soul-crushing'. Thus, what's supposed to follow from this, is that the evil states of affairs we observe do not create an evidential problem for theism.

Objection 1

The first issue I will point out is that there are some reasons to doubt that the theodicy successfully answers the problem of evil.

For phase one, it seems false that some of the most horrific evils we see are defeasible, for the same reason it seems to be the case that some evils are gratuitous. In fact, it's not clear that moving the dialectic to "indefeasible" rather than "gratuitous" evils is doing any work. All the same, upon reflection there doesn't seem to be a whole which, when the evil of, say, the torture and rape of children committed by Peter Scully, is added to it, it is pro-toto good, such that no compossible whole which did not include the torture, would have been all-things-considered better. At least, as far as I'm aware. Even if I'm wrong in my intuitive judgement of some particular prima facie evil, that it is indefeasible, all that needs to be true is that there is at-least one indefeasible evil. Consider the vast swaths of evil including extended torture and mutilation of infants and children, the pain of being burned or skinned alive, the multi-million year process of evolution which includes the immense suffering and languishing of billions, possibly as high as quintillions of sentient creatures, state sanctioned genocides such as the holocaust, agonizing and deadly diseases, the uneven and seemingly morally random distribution of such horrors across the global population. It may be true that there are unknown reasons or factors which make the evil justified, or "defeated". But by that same argument, there might be unknown reasons that the prima facie evils in the world are even worse than they appear to be all-things-considered. So there are four possibilities (Tooley 2019): The evils in the world are exactly as evil/wrong as they appear to be (no hidden relevant-facts); they're worse and not better in some other way (hidden wrong-making factors); they're better and not worse in some other way (hidden right-making factors); and they're worse in some ways and better in some ways, such that the unknown right-making and wrong-making factors are counterbalanced all-things-considered. But only one of those four possibilities allows God to permit those evils. So without reason to think any of the mutually exclusive, jointly exhaustive possibilities is more likely antecedently, it's unlikely that the situation has occurred in which those evils are permitted. At least, the probability that at-least one evil of the vast array of horrendous evils and facts about their frequency, intensity and distribution, satisfies any three of the four possibilities whereby all-things-considered, it would be wrong to permit seems overwhelming.


Objection 1.2

For phase two, you have many of the same issues that plague regular soul-making theodicies. Firstly, much of the most excessive suffering, torture, and abuse we see does not at all seem to be conducive for saint-making. Reportedly, John Wayne Gacy, and Ed Gein were abused as children and rather than benefiting their moral development, this played a part in warping them into the infamous serial killers they are now known as. There is evidence that past trauma increases disposition for mental illness' such as PTSD, depression, anxiety, as well as substance abuse so it seems empirically false that suffering and abuse one experiences in the past generally leads to strength of character and virtue-cultivation. Further, many cases of suffering seem to afflict non-moral agents such as non-human animals, toddlers, and babies and it's difficult to see how such non-moral agents could effectively develop virtues from their suffering. Dougherty is likely going to want to say that their existence will continue in the afterlife in which it will turn out that their suffering prior to becoming moral agents was somehow logically necessary for their saint-fostering in heaven. While this is certainly a story one can tell, as far as I can tell, no reason has been offered to think this story is remotely plausible given the evidence we have, or antecedently likely on theism. Unless independent motivation is given, it seems to be a just-so story, so it will do the theist no good, and tacking it on will only serve to greatly lower the prior probability of the explanation they proffer.

Objection 1.3

Secondly, it seems like God could have created agents with a disposition for virtues, or virtues-from-the-get-go, that do not require being built, and therefore do not require great trials and suffering to respond to. The theist might insist that a world with built virtues is more valuable than a world with virtues-from-the-get-go or a mere disposition for virtues. However, I think the motivation for this is undercut on the theists view. For plausibly, if we are to suppose it is true that built virtues are more valuable then virtues from the-get-go, then what makes built virtues more valuable than virtues-from-the-get-go is that virtues which are built through a process of learning, and growing from ones hardships and pains is more morally praiseworthy then virtues from-the-get-go which are not earned. Yet, the theist is plausibly committed to God both being maximally morally praiseworthy and having virtues-from-the-get-go prior to creation, it seems to fall out from the concept of God as a perfect being, or maximally great being. So, given the theists own commitments, maximal moral praiseworthiness does not entail built-virtues rather than virtues from-the-get-go. Thus, the objection is undercut, and it looks like it would be better for God to create creatures with virtues-from-the-get-go. For a good extended discussion and defense of this point see (Woods 2014).

Objection 1.4

Secondly, it seems virtues such as courage, tolerance, sympathy, and forgiveness are only valuable, for a world which happens to be fraught with hardships and the threat of suffering. In a world without such evils, those virtues are not valuable. Having a world which includes virtues doesn't seem to be intrinsically good for it's own sake, but only contingently valuable on the condition that we already live in a world with suffering. So the value of these virtues given that there are these evils, it would seem, cannot explain why God permits a world containing such evils in the first place when He did not have to. (Maitzen, 2019).

Objection 1.5

Finally, there seems to be a lack of imagination on the part of the theodicist. It seems flatly false that the existence of the most horrific, morally damaging evils we see are absolutely necessary for building virtues, and that we can imagine many ways the world could have been actualized so that virtues are built all the same, sans horrendous evils. For instance, courage can be manifested, both physically by engaging in athletic activities such as sports, and intellectually by pursuing academic inquiry (Kane, 1975). Generosity can be manifested, without poverty and the squalor it entails, teachers and professors can be generous with their time and knowledge, and parents with their care. A mother can be just as kind to her infant child without their child having a disease, illness or injury (Martin, 1992). Sympathy and compassion can be manifested through reading, and being passionately immersed into novels, movies, TV shows, or other fictional works, and feeling sympathy for the hardships of the characters in the story. Moral learning, could be built through being motivated, and inspired by the acts of heroism, and the hearts of justice shown by those of the protagonists, and being disgusted by the vile acts of villainy committed by the antagonist in the story. Even forgiveness, which plausibly entails that you were wronged in some way, can be manifested without horrific torture. You might forgive someone who has lied to you, or who has broken your promise, who has insulted you etc. It might be objected that while these virtues can exist without horrendous suffering, the greatest acts of kindness, compassion, generosity, courage and forgiveness, can only occur in relation to horrendous suffering. This doesn't seem to be the case, are we meant to believe that a mothers kindness towards her infant daughter is made less because her daughter is not ill or injured? Must it be the case that a rising star in football making a comeback in the face of overwhelming odds is building less courage and perseverance than a soldiers charging towards battle? Even if this claim is accepted, it doesn't seem logically necessary, and so it seems God could have adjusted the requisite features of human psychology so that the greatest manifestations of courage, kindness, and generosity and so on, can be manifested in response to situations which do not include horrific suffering. Even if we assume that it is logically necessary, there's still the issue of whether God actualizing a world with the greatest manifestations of virtues, rather than a world with those same virtues manifested to a lesser degree, is such that it outweighs horrendous evils such as baby torture and the holocaust. It seems there are many hurdles for this sort of theodicy.

Objection 2

Even if the theodicy is capable of overcoming said hurdles. There is a worry that it can only do so on pain of a great theoretical cost for theism qua explanatory hypothesis. Allow me to explain.

The theistic hypothesis posits a maximally powerful and knowledgeable agent, this on it's own does not generate any predictive expectations, the only information we have is that the agent can actualize any state of affairs that is logically possible. Since we do not know the agents desires, for all we know the agent can desire anything, so the hypothesis as currently stated is equally consistent with any logically possible states of affairs we can observe. So, in order to generate predictions, theism needs a constraint on the kinds of desires God would have, and thus on the kinds of states of affairs consistent with theism. The only resource theism has for such a constraint, is God's moral perfection, perfect goodness and perhaps perfect lovingness depending on who you ask. So theism generates predictions by ruling in those states of affairs which are morally right to allow, which are intrinsically good, and which are sensitive to the good of creatures. Which entails ruling out states of affairs which are morally wrong to allow, intrinsically bad, and overall bad for creatures. This is where the force of the problem of evil comes, the suffering we observe seems to square better with a world where fundamental reality is indifferent to the plights of conscious creatures, then a world where fundamental reality is morally perfect, or perfectly loving. So it seems the theistic hypothesis has gone wrong with respect to the predictions it makes.

The theodicist wants to say that theism did not go wrong with respect to the predictions theism makes, because the suffering we observe is indefeasible. The problem is, it's not clear they've got access to a way of distinguishing those evils which are defeasible from those which are indefeasible in a way that both answers the difficulties I proffered in the first objection, and allows that there is some state of affairs we can observe which would count as indefeasible. If there is a way of observing some state of affairs, such that we can know that it is indefeasibly evil, it seems to me there are no more clear examples I can hope for than those of the holocaust, brutal baby torture, the prolonged rape and mutilation of a child, evolutionary animal suffering, and agonizing diseases. If, on the other hand, for any evil I point to, even the most horrific kinds, the theist can always appeal to an afterlife where the evil will be defeated and that is enough to completely dissolve the evidential problem. Then it seems there isn't some state of affairs we can observe that would count as indefeasible evil, and it seems theism does not make any predictions. For recall that the only resource we have for generating predictions on theism is the predictive content of God's moral perfection, and yet if there is no state of affairs we can observe which theism rules out in virtue of God's moral constraints, then we are back to the drawing board and every state of affairs we observe would be equally consistent with theism. Even for the most horrific evils we can imagine, as long as it is possible that it could be overcome by the sufferer in heaven, ex hypothesi it is completely expected on theism.

The theist might be tempted to say, that on their view of defeat; the eternal destruction of a soul, or eternal isolation of a soul, or pain that is permanently morally devastating are inconsistent with theism. The problem is, those are infinite states of being and cannot be verified or falsified by our observations, any state of affairs we can observe at any point in time is consistent with a soul being isolated only from times t to t* (any arbitrarily large time interval), or pain which is morally devastating to the victim only at times t to t* and later overcome at some point in the future, or outside of our observation. So while, granted, there are descriptions of states of affairs which would be inconsistent with theism, all the same there is in principle no observation we can have which could count as disconfirmation for theism. Thus, the point remains, on the view outlined, theism lacks empirical content and is untestable. As a result, arguments from the existence of biological organisms, fine-tuning arguments, arguments from psychophysical harmony and other such arguments which pick out particular facts about the actual world which are taken to be evidence for theism are faced with a stumbling block. It looks like theism can no longer predict those particular facts in virtue of their goodness, because the predictive content of God's goodness is now such that there is no observable state of affairs which theism does not predict. The explanatory resource of theism has been effectively stripped. I take it this is a disaster for an evidentialist theist like Trent Dougherty, or any other theist who is interested in arguments for theism which appeal to inference to the best explanation.

This objection is tentative. Perhaps there is a principled basis by which we can be justified in believing a given observable state of affairs is indefeasibly evil, and is such that it excludes the horrors we see, and so the theist gets the best of both worlds. But I'm very skeptical of this.

Objection 3

Another indirect attack on the theodicy, is a familiar one, and that is it seems to succumb to a problem of moral paralysis. If every evil we see is such that it will be defeated in the afterlife, such that it will be a net benefit to the sufferer by fostering their development towards sainthood, and such that they will endorse their life and not wish the suffering away from their life histories. We no longer seem to have a good reason to intervene to prevent suffering. We would not intervene in a suffering child's dental operation, because we know it will be a net benefit for them in the long-term. Similarly, it seems reasons we would otherwise have to save a child from drowning disappear when we realize their drowning is a net benefit for them. This suggests the following plausible principle;

Condition of obligation: We are morally obligated to intervene to prevent suffering, only if we are justified in believing the suffering will not be a net benefit for the sufferer.

Indeed, the more horrific the suffering the more of a reason we have to believe it will be ultimately defeated and a great benefit for the sufferer later on in the afterlife. There need not entail much "defeat" and benefits one gets in heaven for a minor cold to be justified, but certainly there does need to be great defeat and benefits accrued for the sufferer in heaven for the overwhelming torture of burning to death to be justified. So the more horrendous the evil, the more of a reason we have to believe it will be justified and that the sufferer will be benefited, and therefore, it seems, the less reason we have to intervene. Thus, we have another way in which theism, in conjunction with saint-making and defeat, is completely backwards when it comes to ordinary morality (Maitzen 2009).

The theist might argue that theism and defeat only implies we do not have moral obligations to prevent suffering if we assume consequentialism. This is mistaken. It would only assume consequentialism if we were permitting the suffering of an individual, not necessarily for the good of the sufferer, but for greater consequences in general. The defeat condition on theodicy entails that the suffering will be good for the sufferer, so allowing their suffering is not wronging them for the sake of bringing about unrelated greater goods. Additionally, even if it is correct that the paralysis worry assumes consequentialism it doesn't stave off the objection. For then, if consequentialism is false then this implies God would be acting wrongly by permitting suffering for greater consequences, and if consequentialism is true, then surely it's not a problem that we assume it.

A common objection the theist may levy is to draw a distinction between human reasons and divine reasons. Humans have different obligations in virtue of their limited perspective, it would be good for us to act on prima facie moral reasons we have. God, on the other hand, is in an entirely unique moral position where He has access to reasons, we do not have, and bears relations to the situation which we do not share. However, I think this will not do to absolve the theist of moral paralysis. Plausibly, God has different moral reasons then humans in virtue of knowing relevant morally justifying facts. Those facts being that the evil will be defeated, such that the sufferer will accrue great benefits and develop sainthood in heaven. It seems if a human came to knew these relevant facts, for instance by knowing the propositions "God exists", and "God will ensure that horrific suffering will be defeated in the afterlife", then it seems that humans, just like God, would no longer be obligated to intervene in the suffering. This is just an entailment from the condition of obligation I outlined above.

Conclusion


I've comprehensively gone through 4 unique theodicies, and have determined that they do not seem to satisfactorily answer the evidential problem of evil and each appear to have serious issues. This is not a surprising result for me. Very few things in philosophy are more obvious, by my lights, than that a world which contains horrendous evils like the Holocaust, the rape of Nanjing, a fawn burning to death, a child being immolated, a child being raped and abused over long periods of time, hundreds of millions of years of seemingly pointless suffering and brutal predation of sentient creatures, is not the kind of world we would expect to see on the view that reality was created by a being that is all powerful, loving. morally perfect and desires the best for creatures. The failure of theodicy is something which secures this belief even further for me.

References

Internet encyclopedia of philosophy; Logical Problem of Evil; https://iep.utm.edu/evil-log/
Internet encyclopedia of philosophy; Evidential Problem of Evil; https://iep.utm.edu/evil-evi/
Rowe, William L. (1979). The Problem of Evil and Some Varieties of Atheism. American Philosophical Quarterly 16 (4):335 - 341.
Plantinga, Alvin (1978). God, Freedom, and Evil. Eerdmans.
Hick, John. (1966). Evil and the God of Love. Macmillan.
Vitale, Vince. (2017). ‘Non-Identity Theodicy’ Philosophia Christi 19: 269-290 
Salmon, Nathan. (1979). Journal of Philosophy 76 (12):703-725
Scott Hill. The Non-Identity Theodicy. 
John Buck. The Participation Theodicy.
Mackie, J.L. (1982). The Miracle of Theism: Arguments for and Against the Existence of God. Oxford University Press
Draper, Paul. (1989). Pain and pleasure: An evidential problem for theists. Noûs 23 (3):331-350.
Monton, Bradley. (2010). Against Multiverse Theodicies. Philo 13 (2):113-135.
Huemer, Michael. (2021). Knowledge, Reality and Value. 
Dougherty, Trent. (2014). The Problem of Animal Pain: A Theodicy For All Creatures Great And Small.  Palgrave Frontiers in Philosophy of Religion.
Adams, Marilyn McCord. 1989. Horrendous Evils and the Goodness of God. Cornell University Press.
Tooley, Michael. (2019). The Problem of Evil. Cambridge University Press.
Woods, Amanda. (2014). What's evil got to do with it? A Thesis on William Rowe's Argument from Evil and John Hick's Soul-Making Theodicy. 
Maitzen, Steven. (2019). Normative Objections to Theism. Ch. 14. A Companion to Atheism and Philosophy. 
Kane, G. S. (1975). The failure of soul-making theodicy. International Journal for Philosophy of Religion, 6(1), 1–22. 
Martin, Michael. (1992). Atheism: A Philosophical Justification. Temple University Press. 
Maitzen, Stephen. (2009). Ordinary Morality Implies Atheism. European Journal for Philosophy of Religion 1 (2):107 - 126.

History of edits

December 17-20. Correction of typos and syntax errors and formatting issues. Added a new objection to the Participation Theodicy

December 24. I added a whole new objection to the Saint-Making Theodicy (objection 1.3 now) and a new citation to go with it.
December 25. Reworded some stuff, fixed typos. Extended objection 1.3. 

Wednesday, December 7, 2022

Some Thoughts on Apologetics Squared's Pascalian-style arguments

Note; Once again this was originally a reddit post. I did some minor edits on it.

I will be examining two arguments from the YouTuber, Apologetics squared (Henceforth AS). The first, will be a Pascalian argument for the claim that Atheists ought to delete their YouTube channels. The second, will be a more recent argument presented on Capturing Christianity, for the claim that one cannot deconvert from Christian Theism on pain of irrationality. It is recommended that you watch both video's before reading this post. So, let's get started!

Squared's Wager

In the first video linked above, AS presents an argument for the claim that atheists should delete their channels. In the description, he gives the following formulation;

1. If something changes the probability that a person will receive X utility from p_1 to p_2, then the average amount of utility that thing gives them is (p_2-p_1)(X)
2. Def; The probability that a Christian will receive infinite utility is p_C.
3. Def; The probability that an Atheist will receive infinite utility is p_A.
4. p_C is greater than p_A by a non-infinitesimal amount.
5. Therefore, if something changes a person from a Christian to an Atheist, then the average amount of utility it gives them is negative infinity.
6. Counter-apologetics YouTube channels tend to change people from Christians to Atheists.
7. If a person is able to destroy something which tends to give people an average of negative infinite utility, then they are obligated to destroy it.
8. Counter-apologists can destroy their Counter-apologetics YouTube channels by deleting them.
9. Therefore, counter-apologists are obligated to delete their YouTube channels.

What should we make of this argument? There are several possible objections.

We might object to premise 4 on the grounds that for atheists, even for agnostics, the probability of Christianity is 0 or atleast so infinitesimally small that it cannot have an advantage over alternative views. Why? For one, Christianity entails Anselmian theism. The objective probability of Anselmian theism is either going to be 1 or 0, since God is either metaphysically necessary or impossible. So on atheism then, the objective probability of God existing is 0. Now, at this point AS might claim that he is talking about epistemic probability, not objective probability. Still, it seems to follow that the degree to which it seems to us possible that God does not exist, which by the atheists lights should be extraordinarily high indeed - is going to be identical to the degree to which we are warranted in believing that, necessarily, God does not exist. Even if the theist wants to take a much less popular view and (following, say, Richard Swinburne) take God to be a contingent being, an atheist might be convinced by a-priori arguments for the impossibility of such a theism, such as the ones I've defended here and here and many others. If even one of those arguments are sound, we can be close enough to certain that theism qua omni-properties is false. (Related, here is leading atheist philosopher Graham Oppy on why the epistemic probability of theism approximates to 0)

For two, we can look at some teachings which are specific to Christianity, which appears to lay out a very radical anthropology and epistemology. For instance, on Christianity;

i. The Bible teaches no falsehoods.
ii. The universe sufficiently evinces God in His artistic beauty to all mankind. (Psalm 19)
iii. All men truly know God in their hearts and suppress the truth. (Romans 1: 18)
iv. Life is meaningless outside of faith in God. (Ecclesiastes 1-2)

In other words, if Christianity is true, then it's doctrine of revelation is true, which entails that everyone knows God necessarily, that the universe and all of our experiences are God-revealing, and that all of our joy, desires, passions, and love is pointless in the absence of faith. So if you, say, genuinely take yourself to be agnostic - to not know that God exists, then it seems by your lights Christianity must be false. If you reflect on your experiences and genuinely take them as empty of the presence of God or empty of any recognizable divine presence then this entails, by your lights, that Christianity is false. If you think your life, and your practices have value despite you not having faith in God, then once again, by your lights, Christianity is false. None of these points will be convincing to the Christian, of course, but it does mean that, practically speaking, if you are an atheist, or even an agnostic - it seems that the epistemic probability of Christianity is 0 or at least very very close to 0. Thus, premise 4 is probably false.

There is yet another way to attack premise 4. We might say that the probability of;

a. There is a God who rewards believers in Christianity with eternal bliss and punishes those who fail to believe in Christianity with eternal torment.

Relative to a set of alternatives such as;

b. There is a God who already decided prior to creation who He wants to reward or punish.
c. There is a Christian-hating God who punishes all believers in Christianity with eternal torment and rewards everyone else with eternal bliss.
d. There is a universalist God who will reward everyone with eternal bliss no matter what.
e. There is a Devil who will punish everyone with eternal torment no matter what.

-is completely inscrutable. So that we simply cannot know whether it is more probable that deconverting Christians take their chances to achieve infinite utility and avoid infinite disutility away or whether it is in-fact increasing their chance to achieve infinite utility and avoid infinite disutility. Or, whether our actions don't matter at all with respect to who will be receiving infinite utility or infinite disutility, which means we might as well encourage people to cultivate intellectual virtues and believe what we think is true in this life. At the very least, AS has given no argument why we shouldn't think the probability that (a) is true relative to the probability that one of the conjuncts of (b-e) is true is inscrutable.  Indeed by my lights, these possibilities do indeed seem equiprobable if not completely inscrutable. Thus, premise 4 is unjustified and likely false.

The argument can also be attacked at premise 7. Premise 7 assumes that we always ought to maximize expected utility and minimize expected disutility. However, there are cases where this turns out to violate our intuitions. For example, suppose we have to go through one of two doors d1 and d2. If I go through d1 there is a 99.99999% probability I will be viciously tortured in horrific ways for a very large but finite amount of time say, googolplex years, and a 0.000001% probability I will experience infinite pleasure. On the other hand, if I go through d2 there is a 0.000001% chance I will be tortured for an infinite amount of time, and a 99.99999% chance I will experience extreme pleasure for a very large albeit finite amount of time. It seems overwhelmingly obvious that I should pick d2, but on the principle that we ought to maximize expected utility and minimize expected disutility, I ought to pick d1, as the potential utility I on average get is infinite, whereas with d2 the potential disutility I on average get is infinite. Thus, premise 7 is false.

AS has made a video response to Rationality Rules where he attempts to address this objection. His response is that given some evolutionary debunking argument, we have a reason to think our intuition in this case is unreliable. There are a couple points to make here. For one, we can use evolutionary debunking arguments to undermine the reliability of our intuitions in general, but the justification of all arguments seems to ultimately harken back to intuition, so there might be a looming global skepticism worry. For two, he gives no reason to think that such evolutionary debunking considerations do not just as well count against the Pascalian intuition which motivates the argument - that we should always maximize expected utility. Since after-all, both intuitions regard what we ought to do in cases involving infinite utility. AS needs to provide a principled reason why we should accept the principle of maximizing expected value in this case, where it seems to clearly get it wrong. 

I conclude that Squared's Wager is unsound.

Is deconversion from Christianity irrational?

In the second video, AS presents an argument for the claim that Christians can never rationally deconvert from Christianity because given that you accept Christianity, you must accept that your greatest obligation entails always believing Christianity is true. As far as I can tell, there is no precise formulation of the argument to be found, so I made my best attempt at a reconstruction of what I take to be his argument (The formal structure may not be perfect, but leave me alone);

1. There are categorical moral and epistemic norms.
2. There are cases where our categorical epistemic norms and our categorical moral norms conflict, and one will outweigh the other.
3. If (1) and (2) then we always ought to obey what is our outweighing norm.
4. Therefore we always ought to obey what is our outweighing norm.
5. There is a greatest norm that outweighs all other epistemic and moral norms.
6. If you accept Christianity, then the greatest norm that outweighs all other epistemic and moral norms is to fulfill your love to God, which entails having the belief that Christianity is true.
7. Therefore if you accept Christianity, you always ought to believe Christianity is true

Before I respond to this argument. It is important that I clarify a couple distinctions. First, moral norms vs epistemic norms.

What is a norm? A norm is a rule you ought to obey.

A moral norm then, is going to be a rule guiding how you ought to act, or what you ought to do in a given situation. Some easy examples are.
"You ought not torture babies for fun"
"You ought to save a drowning child if you are able with no great risk"

An epistemic norm is a rule guiding your acquisition of beliefs. Here are some examples.
Consistency; You cannot believe p and not-p.
Evidence; If some belief p concerns a fact about which it is easy to acquire evidence, believe p iff one has stronger evidence for p than for not-p.

Next, hypothetical norms vs categorical norms.

A hypothetical norm is a norm that is conditional upon a desire, or a goal. For example, given that it is your goal to lose weight, then you ought to go on a diet or exercise.

A categorical norm, by contrast, is a norm that is not conditional upon a goal or a desire, but is rather reason-giving in itself. For example, we might think "You ought not murder" is a categorical norm, it holds regardless of whatever goals or desires you have.

So, why do I think AS's argument needs premise 1? The answer is I take it that his argument cannot be using mere hypothetical norms, otherwise all he's argued is that it would be irrational for Christians to deconvert given that it is their strongest desire/goal to remain a Christian or something such like. But that's not interesting at all, it certainly does not entail that you can't rationally deconvert as a Christian if you change your mind and begin to lack that desire/goal, so clearly that can't be what he's saying.

But now, we come to the arguments first obstacle, the moral anti-realist and epistemic anti-realist is going to reject premise 1. An error theorist for example is going to think all categorical moral/epistemic norms are false, an expressivist will think that moral/epistemic norms express non-cognitivist attitudes which are not truth-apt. So, on their view there is no categorical normativity. Sadly, AS provides no argument against moral or epistemic anti-realism, thus his argument as it stands simply begs the question against those views.

But let's now assume epistemic realism and moral realism are true. Does premise 2 fare any better? Recall that an epistemic norm is strictly belief-guiding, and a moral norm is strictly action-guiding. How then, can a non-belief guiding norm count against a belief-guiding norm or vise versa? AS runs a thought experiment, which he got from the contemporary Pascalian philosopher Liz Jackson. It goes like this; Suppose there is a truth-hating mafia who kidnaps your family, and they tell you that they'll torture and kill your family unless you take a pill which will force you to believe the millionth digit of pi is 7 (It's not). Thus, we have a case where epistemic norms and moral norms weigh against eachother. But is that really so?

Consider; there are 2 separate oughts here, how you ought to respond to the situation with your direct actions, and what beliefs you ought to have. How you ought to respond to the situation with your direct actions will be 'covered' so to speak, by moral facts, and what your beliefs ought to be will be covered by epistemic facts. In this case, you both ought to take the pill given that is what moral norms command of you, and you ought not believe the millionth digit of pi is 7 given that is what epistemic norms command of you. The moral norm does not weigh against the epistemic norm, your moral duty to take the pill does not change what beliefs you ought to have all-things-considered. All that's happening is that you would be succeeding with respect to your moral duties, and failing with respect to your epistemic duties (because of the pill). Further, the epistemic norm itself does not in any way weigh against you taking the pill, epistemic norms are silent on what actions you ought to take. As they only tell you what your beliefs, and belief-forming methods should be. In-fact, if it is a true moral fact that you ought to take the pill, then according to epistemic norms you ought to believe you ought to take the pill. At best we might say that there is a pro-tanto reason to act in such a way to maximize your disposition for true beliefs and minimize your disposition for false beliefs. But that is not a conflict between bare epistemic norms and moral norms, it should be no surprise that moral norms override weaker pro-tanto reasons.

Moving on to premises 5-7. AS provides another thought experiment to illustrate what he means here. Suppose we have a person, Jim, and he believes that it is his greatest obligation to believe what is written on a piece of paper. You can give Jim a bunch of arguments and evidence to think it is false that what's written on the paper is his greatest obligation, but Jim cannot rationally be compelled by those reasons, since you would just be appealing to lesser epistemic norms. But, his greatest obligation which is to believe what is written on the paper, outweighs them. AS calls beliefs of this kind "anchor beliefs" and, he goes on to say, Christianity is such an anchor belief. Thus, Christians cannot be rational in deconverting from Christianity even on the basis of irrefutable evidence and argumentation that Christianity is false.

There are a few things that can be said here. The first is that there doesn't seem to be any reason to accept premise 5, even if we grant all the previous premises. It may be that there are infinitely many ways categorical epistemic and moral norms can outweigh eachother, and yet there is no single greatest norm which outweighs all other norms. AS has made no attempt to rule such a possibility out after-all. However, the problem gets much deeper. It seems premise 5 is both incoherent given a fairly standard conceptual analysis of categorical epistemic norms, as well as self-undermining. It is incoherent because, by definition, categorical epistemic norms just are those norms which are inherently overriding when it comes to belief-forming. They cannot be outweighed, to say they can is just to misunderstand the overriding normative force that is constitutive of them. AS tries to motivate the claim that epistemic norms can be overridden with Dr. Jackson's truth-hating mob thought experiment, but as we have seen, the motivation doesn't work. To see why it is self-undermining let's look back at AS's 'Jim' thought experiment. Jim believes that what is written on the paper is his greatest obligation. Presumably, this is so because he takes it to be true that his greatest obligation is to believe what's written on the paper. But wait, why does Jim think that it being true that his greatest obligation is to believe what is written on the paper, is a good reason to believe what is written on the paper? The answer it seems to me, is because Jim acknowledges that it is an epistemic norm that we should believe what we think is true. So Jim himself must acknowledge that he is justified in believing his greatest obligation is to believe what's written on the paper, to the same degree that he is justified in holding that it is an overriding epistemic norm that he should believe what is true, and that it is in-fact true that he ought to believe what's written on the paper. Or at the very least, that he has good reasons to think it is true. So, if it turns out there aren't any good reasons to think it is true that his greatest obligation is to believe what is written on the paper and there are good reasons to think it is false, he must acknowledge that he should correct his beliefs.

Another example, how did Jim come to the conclusion that he ought to believe his obligation to believe what is written on the paper outweighs all his epistemic obligations? Perhaps he reasoned like this;

1.My greatest obligation is to believe what is written on this paper
2. Since it is my greatest obligation it must override all other obligations
3. Epistemic obligations are a type of obligation
4. Therefore my greatest obligation to believe what is written on this paper overrides epistemic obligations.

But, why does he believe he ought to think that the premises in the above argument entail the conclusion? Presumably, it is because he takes it to be an epistemic norm that we ought to follow deductive validity, and proper rules of inference. But then, it turns out he is justified in believing his greatest obligation outweighs all epistemic norms, only to the degree that he presupposes the epistemic norm that he ought to obey proper inference rules. So, if he now wants to say that he can ignore his epistemic norms because his greatest obligation overrides them, he would be throwing away the proverbial ladder he climbed to get where he is in the first place. Thus, for the above reasons, premise 5 is incoherent, and self-undermining.

I conclude that AS's argument against rational deconversion from Christianity, is unsound.