The doing/allowing distinction in the divine context

Abstract

The theist needs a conception of the distinction between doing and allowing because much of the literature focused on the problem of evil attempts to justify (via theodicy) or defend (via defence) God's allowing evil to occur. I present a counterfactual account of the doing/allowing distinction in the divine context and argue that, even if there are compelling objections to counterfactual accounts of the distinction in the human context, they do not work against such an account in the divine context. The counterfactual analysis to follow will allow the theist to plausibly claim that God does not ever bring about evil, which is crucial to some defences against the problem of evil. I conclude by defending my account against possible objections.

Introduction

Some moral theories assert that there is some morally significant difference between bringing about and allowing harm. There are two relevant questions here: (1) what is the distinction between bringing about and allowing harm and (2) is this distinction morally relevant? I will focus on (1) in this article with regard to divine bringing about and allowing. I present a counterfactual account of the doing/allowing distinction in the divine context and argue that, even if there are compelling objections to counterfactual accounts of the distinction in the context of human agency, they do not work against such an account in the divine context (e.g. Kagan (1989); Howard-Snyder (2002)). The counterfactual analysis to follow will allow the theist to plausibly claim that God does not ever bring about evil, which is crucial to some defences against the problem of evil (e.g. Shea and Ragland (2018)).

The theist needs a conception of the distinction between doing and allowing because much of the literature focused on the problem of evil attempts to justify (via theodicy) or defend (via defence) God's allowing evil to occur. I take it that this use is morally significant. Take, as an example, a recent use of the distinction between bringing about and allowing in Shea and Ragland (2018). They write: ‘The occasionalist is unable to claim that God truly permits suffering for morally sufficient reasons, which is the standard theistic response to the argument from evil. Instead, God immediately produces suffering, a position that is much harder to reconcile with God's perfect goodness’ (Shea and Ragland (2018), 276). An argument along these lines requires a coherent distinction between bringing about and allowing suffering. If there is no such distinction, there would not be much sense in emphasizing that God's allowing suffering is easier to justify than his bringing about suffering. Without a clear distinction, there is no reason to say that God allowed one agent's action but brought about another's. Yet, the force of the positions like the one outlined in Shea and Ragland (2018) comes from the intuitive idea that it is easier to justify God's allowing some agent to bring about harm – for example, for all the goods that would come from free choice – than God bringing about that harm. In order to make such a claim, the theist needs a clear conception of the distinction between bringing about/allowing.¹

I suggest that a counterfactual account of the distinction is plausible in the divine context. Some authors (Lim (2017); Sen (2022)) have recently dismissed such an account far too quickly. It is therefore the goal of this article to provide a counterfactual framework for the bringing about/allowing distinction in the divine context.

A counterfactual analysis of divine bringing about/allowing

For divine bringing about, a first approximation of God's bringing about would just be the simple counterfactual analysis: God's action counts as bringing about event e iff his action stands in the following counterfactual dependence relation to e: had x not taken place, e would not have occurred.² Take, for instance, the parting of the Red Sea. God's action, x, counts as bringing about the parting of the Red Sea iff, (1) God performed x and the Red Sea parted, and (2) had x not occurred, then the Red Sea would not have parted. Thus, a first pass at defining the bringing about/allowing distinction in the divine context can be outlined as follows:

Divine Bringing About: God brings about some effect, e, via action(s) x⊆ X = df. (i) x and e occur and (ii) had no member of X occurred, e would not have occurred.

Divine Allowing: God allows some effect, e, via possible action(s) x⊆ X = df. (i) e occurs and no member of X occurs (ii) had x occurred, then e would not have occurred.³

where X is the set of divine actions which are sufficient for e to occur (or, in Divine Allowing, sufficient for e not to occur) and x is any subset of that set, each of which is sufficient for e to occur (or, in Divine Allowing, sufficient for e not to occur).

So, God brings about someone being hit with a rock iff (i) God performs some action (or set of actions) which is (are) sufficient for bringing about that event and (ii) had God not performed any action which was sufficient for this event to come about, it would not have occurred. Likewise, God allows someone to get hit by a rock iff (i) there is a possible action or set of possible actions which God does not perform and (ii) had God performed any of those actions, the event would not have occurred.

As stated, however, the account entails that God brings about everything, even the harm brought about by created agents (Sen (2022)). To see why, suppose that Suzy harms Billy by throwing a rock at him. Suzy's throwing the rock counterfactually depends on God's action of creating the universe; had God not created the universe, Suzy would not have thrown the rock. The account therefore leads to the undesirable result that all events are instances of God's bringing about. I take it that many theists want to avoid this conclusion; if we do, then we have the following desideratum for an adequate account of divine bringing about/allowing: an account of divine bring about/allowing must avoid making it an analytic truth that God brings about all events. This article presents one attempt to meet this desideratum in counterfactual terms.

This worry can be addressed as follows: imagine that some version of a free will theodicy is true. In such a theodicy, the free will of creatures is necessary to obtain some good that could not be obtained otherwise. The existence of creatures with free will and all their action counterfactually depends on God's divine act of creation. The evil free acts of agents were a consequence of God's creating agents with free will.⁴ There exists counterfactual dependence between the agents' actions and God's divine act of creation. Yet, theists do not claim that God, by bringing about the existence of agents with free actions, thereby brings about the evils resulting from their free choices. Many theists will not claim that God brings about these evils because, assuming that God is omnibenevolent, he does not bring about the evil actions performed by free creatures' actions.

I shall propose a nuance within the analysis of bringing about/allowing that will not entail that God brings about every human action because Bringing About (suitably understood) must not entail that God brings about the evil actions of free agents. In particular, I suggest that a necessary condition of God's bringing about some effect is God's intending that effect as a result of a divine action.⁵ This is a plausible condition of God's bringing about because, necessarily, if God brings about e, then God intends that e occur as a result of divine action. In this analysis of divine bringing about and allowing, the addition of this clause will render the intuitive results. God will not count as bringing about evil human actions so long as he does not intend them as a result of his creating free creatures, and this tracks what the theist means when she claims that God does not bring about evil actions. This proposal makes sense of Shea and Ragland's claim that God does not bring about the harm caused by free agents. On the present proposal, God creates the world and knows that free agents will perform terrible deeds. But God never intended that they perform such deeds. As a result, the theist can say that God creates free creatures without thereby bringing about the harm of their actions. For instance, if God never intended that Suzy throw the rock at Billy, then God cannot be said to have harmed Billy.

Divine Bringing About: God brings about some effect, e, via action(s) x⊆ X = df. (i) x and e occur and (ii) had no member of X occurred, e would not have occurred, and (iii) God intends that e occur.

Divine Allowing: God allows some effect, e, via possible action(s) x⊆ X = df. (i) e occurs and no member of X occurs (ii) had x occurred, then e would not have occurred and (iii) God does not intend that e occur.

where X is the set of divine actions which are sufficient for e to occur (or, in the case of Divine Allowing, sufficient for e not to occur) and x is any subset of that set, each of which is sufficient for e to occur (or, in the case of Divine Allowing, sufficient for e not to occur).

Kamm's ‘because of/in order to’ distinction

There is, however, a clarification I must make to be more precise about what I have in mind when I say God intends some effect occur. There are two senses in which someone might intend an effect.⁶ Kamm (2007) provides a useful distinction between types of actions.⁷ Kamm applies the following distinction to actions: one could act in order to produce an event or one could act because an event would be produced by one's action. To state the distinction precisely, one performs an action in order to produce a certain event, when that event occurring is the (primary) reason for performing that particular action. On the other hand, one performs an action because some effect occurs which one takes advantage of, but need not intend.

Kamm presents an example to help clarify the distinction which she calls the Party Case:

I intend to give a party in order for me and my friends to have fun. However, I foresee that this will leave a big mess, and I do not want to have a party if I will be left to clean it up. I also foresee a further effect of the party: if my friends have fun, they will feel indebted to me and help me clean up. These expectations are conditions of my action. I would not act unless I had them. The fact that they will feel indebted is a reason for my acting. But I do not give the party even in part in order to make my friends feel indebted nor in order to not have a mess. To be more precise, it is not a goal of my action of giving the party to do either of these things. (Kamm (2007), 95)

This example helps clarify the distinction. It is clear in this example that the host does not throw the party in order to make a big mess or cause her friends' feeling of indebtedness; that is not the primary reason for throwing the party. On the other hand, this host throws the party because throwing the party results in both the mess and her friends' feeling of indebtedness. The mess would defeat the reason of the party, which is to have fun. However, since the action results in her friends' feeling of indebtedness, which in turn takes care of the mess of the party, the host holds the party. Yet, it is not the case that the host holds the party in order that the mess and the feeling of indebtedness arise. Rather, she can take advantage of the undesirable consequences of throwing the party, and throws the party in part because she can take advantage of those consequences. So, it is possible to intend a particular goal without intending the means to that goal.

A- and B-intending

I propose that this distinction, originally used by Kamm to distinguish between types of action, will be useful in delineating two different types of intending. Suppose that God performs some action, x, in order to obtain some good, e. An instance of this would be in the case of the plagues brought about on Pharaoh. God sent the plagues, intending that the plagues bring pain to the Egyptians. In order to deliver the Israelites from slavery, God intended that the plagues brought on Egypt be a means by which this deliverance was to come about. God intending the plagues as a means of deliverance is an instance of the first sort of intending, which I will call a-intending:

God a-intends that e occur = df. God performs some action, x, in order to obtain e.

However, this is not the only way that one can intend e. It may be the case that, in the words of Kamm (2007), one can perform an act which results in a state of affairs ‘only because [she] can also take advantage of one of its bad effects’. In other words, God might perform x in order to achieve some good, even though x results in some undesirable side effect, because he can also take advantage of that side effect to achieve a further good. And, as demonstrated by the Party Case, the undesirable effect need not be intended by the agent. The general idea here is that, as Murphy (2017) phrases it, it is possible that God performs some action in part because he can make use of undesirable side effects of that action.

Take the healing of the paralytic at the pool of Bethesda as recorded in the New Testament. Suppose that God had not acted in order that this man be paralysed. Rather, God might have acted in order to obtain some other good, namely, the glorification of God, without intending that the man be paralysed. We can imagine that, in this case, God performed some action which caused the man to be paralysed, such as establishing the laws of nature which result in paralysis. Suppose that, for whatever justifying reason God has for establishing the laws of nature as he did, this man's paralysis is an undesirable consequence. Yet, God foresaw, also as a result of his action, that Jesus would heal the paralysed man. God also saw that the man's being healed would defeat the undesirable effect. But, just as in the Party Case, God need not intend the undesirable effect. It is possible that God does not act with the intention that the man be paralysed. In other words, it is possible that that he did not establish the laws of nature in order to paralyse the man. Rather, it may be the case that God establishes the laws of nature because he intends to make use of the man's paralysis. This example shows that it's possible for God to intend that some effect follow from his action in a second sense, which I will call b-intending:

God b-intends that e occur = df. (i) God performs some action, x, to make use of e and (ii) God does not a-intend e.

With this clarification made, I can now disambiguate the sense of ‘intend’ that I use in the third clause of Bringing About. Given this distinction, I suggest that God brings about some effect only if he a-intends that effect. This addition is important since, if God's intending is not limited to a-intending, then all events will count as having been brought about by God. This result follows from the fact that every event counterfactually depends on God's willing that the universe exist. So, saying that God b-intends at least some events avoids the conclusion that God brings about every event, which was the original motivation for introducing divine intentions. Thus, clause (iii) in Bringing About should be read as: God a-intends that e occur.

Objections

In this last section, I show how objections to the counterfactual analysis presented above are unsuccessful.

Divine intentions

The addition of intentions might seem odd: how exactly are intentions relevant to determining a case of divine bringing about? Certainly, intentions are not relevant in an evaluation of human bringing about. Imagine, for instance, that the US military targets a terrorist group. There is a small chance that the drone targeting the group accidentally kills civilians nearby. If the drone does in fact kill the civilians, it is clear that the US military has brought about their death. However, in this example, they certainly did not intend to bring about the deaths of those civilians. Thus, it seems that, in any other agent's case, it is not true that, necessarily, if agent A brings about some event e, then A intends that e occur.

However, there is good motivation for thinking that necessarily, if God brings about some event e, then God a-intends that e occur. Suppose that this statement is false of God. If this statement is false, then it is possible that God brings about things which he does not a-intend. But suppose that we assume that God does not a-intend evil. Given these assumptions – it is possible that God can bring about things which he does not a-intend and that God does not a-intend evil – it is possible that God brings about evil. If we want to preclude the possibility that God bring about evil, then the theist has good reason to accept God's a-intending e as a necessary condition for his bringing about e. While this response is not meant to prove that God's a-intending e is necessary for his bringing about e, it is, I believe, sufficiently strong to motivate the account. Thus, in the absence of other strong objections, the theist has good reason to accept the present proposal if she wants to exclude the possibility of God bringing about evil.

But perhaps there are cases of God bringing about some effect without intending that effect. An anonymous reviewer raises the possibility that God brings some event about without intending that it come about. This reviewer writes:

Suppose God a-intends to make some mountains but doesn't really care how tall they are. God foresees that if God triggered a certain tectonic shift, a mountain with height h would be formed. God triggers the shift, a-intending to form the mountain, but merely foreseeing and not a-intending that the mountain be height h. It seems to me that God has brought about the mountain's being height h.

In this case, there seem to be a range of options which would satisfy God's intentions, but it is not true of any member of that range that God intends this option.

One plausible response is that God can intend not only specific effects, but also a disjunction of effects where, were one of the disjuncts to obtain, God's intention is satisfied. To use the reviewer's example, a mountain must fall within a certain height range. And the structure's reaching any of these heights satisfies God's intention to create a mountain. So, we might say that God brings about the height of a mountain so long as that height falls within the range necessary to satisfy the intention of creating a mountain.⁸ God intends a certain disjunction of effects: mountain with h₁ or mountain with h₂ or . . . mountain with h_n. While it is true that God does not intend that the mountain be height h, he does intend that one of the above disjuncts is true. So, it's plausible to think that God brings about the mountain of height h so long as h is a member of the set of disjuncts God intends.

Divine inaction

Lim (2017) suggests that one cannot conceive of divine allowing as divine inaction, and that this consideration renders any counterfactual account of the bringing about/allowing distinction in the divine context implausible. He states that:

There are serious challenges facing the no-action counterfactual analysis of the doing-allowing distinction. These challenges are only exacerbated when applied to God. For what could it mean to imagine a situation where God refrains from acting? . . . Outside of imagining God's being asleep or being in a trance the only way to make sense of God's not performing a given action, it seems, is to imagine God performing a different action. So it seems the counterfactual analysis fails to provide a compelling analysis of the doing allowing-distinction in the divine context. (Lim (2017), 280)

In short, God cannot be inactive, and the only alternative to his inaction would be performing some other action. Further, Lim seems to be implying that, since God's allowing can only be imagined as performing some other action, we cannot analyse God's allowing some event to occur in terms of his not performing an action.

This argument is not very compelling for two reasons. First, one need not imagine God's inaction as his ‘being asleep’ or ‘in a trance’. It is clear enough to say that God does not perform a certain action. Lim provides no compelling reason to suggest that God cannot be merely observing an event take place.

Second, it is not true that the only alternative to understanding God's not performing some action is God's performing another action. It could very well be the case that God's not performing some action is just a possible divine action. In other words, to say that ‘God does not perform action x’ is just to say that ‘God possibly performs action x but does not.’ So, to say that God does not create the Big Bang does not necessarily imply that God performs some other action; it may very well just imply that God could possibly have created the Big Bang but does not. As a result, I have built this suggestion into the present proposal, but that is not to suggest that there cannot be other ways of conceptualizing God's inaction.

Inaction and counterfactual dependence

Kagan (1989) considers a counterfactual analysis of the bringing about/allowing distinction based on a notion of interference. Roughly, the idea is that, one counts as bringing about a certain state of affairs when one interferes in the course of a sequence of events. An agent, further, counts as interfering if, had she not reacted in a particular manor, then the effect would not have occurred. Kagan specifically uses the term ‘reaction’ to refer to an agent's response to any particular situation. What is crucial to his objection to the analysis of the present counterfactual analysis of the bringing about/allowing distinction he considers in his book is the fact that an agent's reaction might be action or inaction. He gives the following example: ‘in failing to send money to famine relief, I interfere with the life of stranger who dies of starvation, for had my reaction not occurred – had I sent money – someone's life would have been saved’ (Kagan (1989), 98). There is counterfactual dependence between one person's inaction and the starvation of the other. Therefore, the person who does not send money would count as having brought another's starvation when, intuitively, that does not seem to be the case.

The worry can be applied to the present analysis. Let's say that God's possibly, but not actually, bringing about some effect counts as a divine inaction. It seems there would then be a counterfactual dependence between God's failing to bring about world peace and all the deaths that result from war. An analysis similar to what Kagan considers would then count God as having brought about those deaths because there would exist a counterfactual dependence between his inaction and their dying. Any plausible analysis of divine Bringing About will exclude this result.

The present analysis is not prey to the same worry. God counts as bringing about some effect only when he a-intends to bring about that result. Even though there may exist a counterfactual dependence between God's inaction and some effects, he would only count as bringing about those effects if he a-intended them to occur.

Pre-emption

The final objection I will consider pertains to pre-emptive counterexamples. Howard-Snyder (2002) leverages pre-emption problems to provide an objection to the counterfactual analysis.⁹ She argues as follows:

suppose an SS officer, Franz, tortures someone to death. But this is standard practice in the Gestapo. If Franz had stayed home with a sore throat, or if Franz had never existed, his pal Hans would have done the torturing, in the same way, at the same time Franz did. If the counterfactual account is correct, then Franz is negatively relevant to the victim's death by torture. That is, Franz merely allowed the death to occur.

The simple counterfactual analysis she is considering is similar to the one examined by Kagan. She considers the following view: someone counts allowing some event to occur iff, (i) some event occurs and (ii) that event would have occured regardless of one's presence or actions. In the example she gives, if Franz does not kill his victim, then some other SS officer would have. Since the effect would have occurred whether or not Franz did the killing, Franz counts as merely allowing his victim to die, which is the wrong result.

While this objection does not apply directly to the present analysis, the pre-emption scenario can be reimagined to apply to an analysis outside the divine context which resembles my proposal. The analysis might run as follows: (i) e occurs (ii) if x occurs, then e would not occur and (iii) x does not occur. In other words, an agent counts as allowing some event iff, (i) some event occurs and (ii) had the agent acted to prevent the effect, that effect would not have occurred. The previous example can now be modified. Imagine that Franz has a sudden conviction that what the Nazis are doing is immoral. However, his cowardice restricts him from stopping his pal Hans from brutally assaulting a Jewish prisoner. He would count as allowing his friend Hans to beat the prisoner iff, (i) Hans beats the prisoner and (ii) had he acted to stop Hans, Hans would not have beaten the prisoner. Yet, even if Franz attempted to stop Hans, some other Nazi could have stopped Franz's assault before he stopped Hans, and he would still be able to beat the prisoner as a result. Therefore, Franz would not count as having allowed the prisoner to be beaten when he stands by and watches, which seems to be the wrong result.

Even if these pre-emption scenarios provide counterexamples to the bringing about/allowing distinction in the human context, I will argue in the next section that it does not do so in the divine context.

Pre-emption and Bringing About

In previous work, I (2021) argued that pre-emption problems can be avoided for a counterfactual account of divine causation; I will briefly apply this argument to the present analysis of divine bringing about. In short, if some event is not counterfactually dependent on God's act – that is, God does not stand in the Bringing About relation to some event – then that event is counterfactually dependent on God's allowing it to occur. This shifts the burden of the objection. In other words, if counterfactual dependence does not hold between God's action and some event, then counterfactual dependence holds between God's allowing and that event's occurring. As a result, in order for there to exist an instance of pre-emption, it is counterfactual dependence between God's allowing and that event's occurring that must be broken; something else would have to allow that event to occur if God did not allow it to occur. Yet, if God does not allow an event to occur, it is impossible for that event to occur. Therefore, there is no example of pre-emption against Bringing About.

Pre-emption and allowing

However, even if this line of thought works for Bringing About, I have not yet argued that Allowing can also avoid cases of pre-emption. In order to advance this analysis of the bringing about/allowing distinction in the divine context, I must demonstrate that cases of pre-emption are not counterexamples to Allowing.

Some form of pre-emption would pose a problem for Allowing as I have characterized it above if there could be some case in which God clearly allows some event but the counterfactual dependence does not hold. Recall the analysis: God allows some effect, e, via possible action(s) x⊆ X = df. (i) e occurs and no member of X occurs (ii) had x occurred, then e would not have occurred and (iii) God does not a-intend that e occur. In other words, if some divine act occurs, then some effect would not have occurred. For example, say God allows some tree to fall in the forest. If God allows this event, then it should be the case that, had God acted to prevent the event, then the event would not have occurred. In this example, in order for pre-emption to be a concern, it must be the case that this counterfactual statement is made false, and it is difficult to see how this could be done. If one understands God as omnipotent, where omnipotence is classically defined as God's ability to bring about anything logically possible, then it does not seem possible for this statement to be made false. If God acts to prevent the tree from falling, the only way the relevant counterfactual is made false is if the tree falls even though an omnipotent being is acting to prevent its falling. If God is omnipotent, and if God acts to prevent the falling of the tree, then it is impossible for the tree to fall. Since it is not possible for the relevant counterfactual statements to be made false by some other event, it does not appear that pre-emption is an issue for this analysis of the bringing about/allowing distinction.

But perhaps another sort of pre-emptive scenario is problematic for the account: say that God allows the tree to fall by not sending rain; as a result of the lack of rain, the tree dries up and falls. Yet, we might also imagine that, had God sent rain, he would have intervened to bring about the tree's falling some other way. So, God allows the tree to fall but it seems that the counterfactual, had God sent rain, the tree would not have fallen, is false.¹⁰

The distinction between a- and b-intending will help answer this objection. Let's assume Lewisian (1973b) truth conditions for counterfactuals. On this assumption, the counterfactual, ‘had x occurred (i.e. a member of X which is sufficient for e's failing to occur), e would not have occurred’ is true iff, in every nearest antecedent permitting world, the conditional ‘if x occurs, e does not occur’ is true. In the present case, the antecedent permitting worlds will be ones in which God sends rain. Now, part of the scenario described above is that God actually allows the tree to fall, which means, given the definition I've provided, that God does not a-intend that the tree fall in the actual world. We can divide the nearby worlds into two types of antecedent permitting worlds: worlds in which God sends rain and a-intends that the tree fall and worlds in which God sends rain and does not a-intend that the tree fall. Given that God does not a-intend that the tree fall in the actual world, the rain-permitting worlds closest to the actual are the ones in which God does not a-intend that the tree fall.

Recall, now, that the objection requires the following two claims to be true: (1) God allows the tree to fall and (2) had God sent rain, he would have intervened to bring about the trees falling some other way. Yet, given that God does not a-intend the tree to fall in the nearest worlds (i.e. (1) is true), (2) turns out false. If, as I have argued, it is false to say that God a-intends that the tree fall in the nearest antecedent permitting worlds, it follows that God does not intervene to bring about the falling of the tree in any of these nearest worlds. Why? Because he does not a-intend them in these worlds; and a-intending some effect is a necessary condition of God's bringing about that effect. Therefore, since the objection needs both (1) and (2) to be true, the objection fails.

There remains, however, one more possible pre-emption counterexample to my account of divine allowing. Take the famous story of Augustine stealing from the pear tree. On the present analysis, God allows Augustine to steal the pears if and only if, had God acted to prevent him from stealing the pears, he would not have stolen the pears. But suppose that God had acted to prevent him from stealing the pears by appearing to him in a dream and commanding him not to steal the pears. Yet, barring theological determinism, it seems plausible that Augustine would steal the pears anyway, regardless of God's command. Suppose that he resolves to steal despite God's command. This example seems to break the counterfactual dependence relation in Divine Allowing; it is not true that, had God acted to prevent Augustine from stealing the pears that he would not have stolen the pears.¹¹

I will offer one plausible response to this worry. Suppose that God has ‘middle knowledge’, i.e. counterfactual knowledge. In brief, this amounts to the claim that God knows what agents would freely choose to do given certain conditions. If one grants that God has this sort of knowledge, then one can claim that God knew, if he appeared to Augustine in a dream and commanded him not to steal the pears, that Augustine would still choose to steal the pears. I propose that God counts as acting to prevent an event only if he knows that some event would not occur given his divine action, which, in this case, would be God speaking to Augustine in a dream. This suggestion means that God would not count as having acted to prevent Augustine from stealing the pears. Perhaps God counts as revealing sinful intentions in such a case, or perhaps he counts as telling Augustine what he ought to do. But it's plausible that God does not count as preventing in that case, since he knows that Augustine will steal the pears even if he were to command otherwise. As a result, this objection is not a counterexample to divine Allowing. I readily grant that it is controversial whether or not God has knowledge of such counterfactuals. I only intend this to be one response to the worry, although there may be others.

Conclusion

While I believe the present analysis provides a workable bringing about/allowing distinction in the divine context, there is still much work left to be done to determine whether or not this distinction is morally significant. I have not tried to answer the further question of whether or not it is true that it is easier to justify God's allowing harm than bringing it about, but trust that this analysis provides a beneficial and rigorous conceptual analysis of a concept which has not yet been explicitly defined even though it is implicit or explicitly used in recent discussions (e.g. Lim (2017); Loke (2018); Shea and Ragland (2018); Sen (2022)).