As has been mentioned, the general consensus is that Plantinga’s critique of Mackie is successful not only in directly answering Mackie’s challenge but also in undermining all possible logical formulations of the problem of evil. Plantinga has shown that there can be no essential inconsistency in the theist’s set of beliefs, at least when it comes to the problem of evil. The only option that remains for the problem of evil is to shift to evidential formulations. I think that is a mistake.
I do not dispute that Plantinga’s critique successfully answers Mackie’s challenge. What I dispute is that it thereby fatally undermines all logical formulations of the problem of evil, necessitating the shift to evidential formulations. In previous work, I have noted that there is some ambiguity about what people mean by ‘logical’ or ‘evidential’ formulations and the difference between the two, that our terminology here is very vague and poorly-argued and likely isn’t fit for purpose. I’ve also argued that logical formulations are preferable in all cases. But I won’t repeat myself here because it’s not strictly necessary for my point. It’s old news, anyway.
Let’s settle for a simple and hopefully uncontroversial starting point: the ‘logical’ formulation of the problem of evil claims that there is an inconsistency in the three propositions: God is good, God is powerful, evil exists. The evidential formulation, by contrast, does not claim there is any inconsistency here, only that it is unlikely, on the evidence available to us, that all three things are true. For the logical formulation, if any two of the three propositions are true then the third must be false; for the evidential formulation, if any two of the three propositions are true then the third could be true but is more likely to be false.
The crucial difference between the two, for my purposes, is the presence or absence of a logical ‘must’. It’s the presence of this logical ‘must’ that compels the theist, with all the power of logic, to offer a solution to the problem of evil. Only the logical formulation of the problem of evil remains logically binding in this way, which is what’s needed for the first premise in my argument: ‘A consistent set of theistic beliefs requires a solution to the logical problem of morally-impossible evil.’ A logical problem requires a solution; an evidential problem does not. If a solution is not required, then there is no need to reach for one. And since the available solutions are morally problematic, the best response might be to say nothing and remain with your innocence presumed. But can you remain there, in your silence, in such a way that you are consistent with yourself?
I think the shift from logical formulations to evidential formulations was a mistake, but the truth of that is not necessary for my argument. All that is necessary for my argument is that logical formulations of the problem of evil remain possible after Plantinga. To this end, I will take up the gauntlet laid down by Plantinga’s critique of Mackie.
Plantinga’s Critique
Why is Plantinga’s critique of Mackie so successful? Why does it work? A brief summary of it will show.
According to Plantinga, Mackie claims this set of beliefs is inconsistent: God is wholly good, God is omnipotent, and evil exists. But logical inconsistency is demonstrated by pointing to a contradiction, and Plantinga asks: where is the contradiction in this set? There is clearly no ‘explicit’ contradiction, of the form ‘God is good and God is not good’, or ‘evil exists and evil does not exist’. So this cannot be what Mackie meant to suggest in claiming the set was inconsistent. Did he mean to suggest that a contradiction could be generated simply by the application of logical rules? Like this set: All men are mortal; Socrates is a man; Socrates is immortal. There is no ‘explicit’ contradiction here, but we can easily generate one if we apply basic logical rules and use the first two statements to infer an additional (unstated yet necessarily following) statement: If all men are mortal and Socrates is a man then Socrates is mortal. The claim that Socrates is mortal necessarily follows from the first two statements in the initial set and yet is clearly contradictory with the third. Plantinga labels this type of contradiction a ‘formal’ contradiction. It’s not clear that Mackie’s set can do this either: there is nothing in the three propositions, as stated, that would allow us to generate a contradictory statement merely by the application of basic logical rules.
But if there is no explicit or formal contradiction then Mackie must have meant to suggest an ‘implicit’ contradiction: a contradiction that can be generated via the application of logical rules, but only if we first acknowledge some additional statements. These statements serve the purpose of clarifying what we mean by certain words or concepts. As such, they take the form of necessary truths. For example, there might be something implicitly contradictory about this set: Mike is a bachelor; Sarah is Mike’s wife. Clearly, if Mike is a bachelor then he has no wife, so the set cannot be consistent. But there is no explicit nor formal contradiction, as it stands; nothing clearly stating ‘Mike is married and Mike is not married’, for example.
What is needed are some other propositions to add to the set that will clarify what we mean by the terms ‘bachelor’ and ‘Mike’s wife’: a ‘bachelor’ is an unmarried man, and ‘Mike’s wife’ is a woman who is married to Mike. These are necessarily true statements; they cannot be false; they simply capture and clarify what we mean by our words or concepts. In being necessarily true, it cannot do any harm to add them to any set of beliefs. They cannot change anything because, in being necessarily true, they were true all along. If we add them to our set about Mike and Sarah then we can work towards generating an explicit contradiction: Mike is a bachelor; Sarah is Mike’s wife; a bachelor is an unmarried man; ‘Mike’s wife’ is a woman who is married to Mike. If we put the first and third statements together we can generate a new statement: Mike is an unmarried man. If we put the second and fourth statements together we can generate another new statement: Sarah is married to Mike. A new set of propositions emerges following this clarificatory work: Mike is an unmarried man; Sarah is married to Mike. Note, these are essentially no different from the first incarnation. Nothing has changed because we haven’t changed anything: we’ve only clarified what we meant all along.
We can keep working to expose the implicit contradiction, adding further necessarily true clarifications until the contradiction becomes explicit: if Mike is an unmarried man, then Mike is not married; if Sarah is married to Mike, then Mike is married. Again, these are just clarifications of what we meant all along. With these further additions we can now apply basic logical rules and generate an explicit contradiction from our original set: Mike is a bachelor; Sarah is Mike’s wife; if Mike is a bachelor then Mike is an unmarried man; if Mike is an unmarried man then Mike is not married; if Sarah is Mike’s wife then Sarah is married to Mike; if Sarah is married to Mike then Mike is married. So Mike is married and Mike is not married and we have made our implicit contradiction explicit. All it took was a little clarificatory work to uncover what was there all along.
Is this what Mackie meant to do with his logical formulation? It certainly seems so. There is nothing explicitly contradictory about the set: God is wholly good, God is omnipotent, evil exists. And neither can we easily generate a contradiction simply by the application of logical rules. What is needed is further work to dig down and clarify what we mean by ‘wholly good’, ‘omnipotent’, and ‘evil’. Mackie offers his suggestions: ‘good is opposed to evil in such a way that a good thing always eliminates evil as far as it can’, and ‘there are no limits to what an omnipotent thing can do’. (It’s worth noting that he does not clarify what we mean by ‘evil’, other than that it is opposed to good. That lack of clarification could be seen as underselling the importance of the role that our understanding of ‘evil’ plays in the problem of evil.) These ‘quasi-logical rules’ would need to be necessarily true if they are to do the work of merely clarifying what we mean when we say ‘wholly good’ or ‘omnipotent’.
Plantinga’s first step is to take issue with these clarifications. They are not necessarily true, he says. There are some limits to what an omnipotent thing can do, even if only the limits of logic. And it is not clear that a good thing must always eliminate evil as far as it can: sometimes a good thing might only be able to eliminate one of two evils, and so whilst it ‘can’ eliminate either, it cannot eliminate all; sometimes a good thing might allow a little evil if that meant avoiding something worse; sometimes a good thing might allow a little evil if it meant attaining something better.
Mackie’s clarifications will not cut it. Plantinga pushes and prods and makes his best attempt at helping Mackie out, but none of the statements that result will serve Mackie’s purposes. Try as he might, Plantinga cannot find any additions that will reveal an implicit contradiction in the set: God is wholly good, God is omnipotent, evil exists. He concludes there can be no such additions, because the set is not, in fact, inconsistent.
The consensus unites in agreement with Plantinga. As a result, few are willing to defend the logical formulation of the problem of evil. Any who wish to discuss the problem of evil should now do so on an ‘evidential’ basis only.
The Return of the Inconsistent Triad
Is it possible to construct a logically-binding formulation of the problem of evil on the basis of the inconsistent triad? I think it is, but only if we acknowledge the fundamental role that moral beliefs play in the problem of evil. Up to now, it seems as if atheologians have tried to construct versions of the problem of evil that question the consistency of theistic belief. They have spoken about the problem of evil as if it were not an ethical problem at all. Is it any wonder, then, that they find insufficient means to ground an incompatibility between God’s goodness and the evils of the world?
For me, the problem of evil is an ethical problem. It is a problem about goodness, value, and how we are to orient ourselves in a world of suffering. As such, any inconsistency will not be found in ‘beliefs’ but in ‘values’; or rather, not in beliefs as such but in specifically ethical beliefs about what is good, what is evil, and what is the appropriate attitude to take to a world of suffering. If we are to find an inconsistency that grounds a logically-binding formulation of the problem of evil, we should look to find an inconsistency in values.
Let’s start again from the beginning. Let’s work away at it, in the same way we worked away at our earlier example of ‘Mike is a bachelor; Sarah is Mike’s wife’. Let’s dig down and uncover what we mean by ‘God is wholly good’, ‘God is all-powerful’, and ‘evil exists’ in order to expose an implicit contradiction. I think if we do this then we can construct a version of the problem of evil that is logically binding. I call it the logical problem of morally-impossible evil. It comes from our capacity to recognise a difference between what is morally possible and what is morally impossible, between the morally contingent and the morally necessary; in short, this version of the problem of evil is a product of our understanding of moral modalities. (And this, of course, is the purpose of investigating the problem: to reflect on, understand, and improve our moral understanding.) But since the product of this investigation can only be a clarification of ‘what we mean when we say’, it is not a matter of establishing a fact about the external world but only of understanding a truth about ourselves. The question is not ‘are there moral necessities?’ but only ‘do you recognise moral necessities?’ And who can answer that question, other than you? (This will be discussed in more depth in chapter five: ‘The Problem of Evil as an Ethical Problem’)
What do I understand, when I investigate my own understanding? I find that I do recognise moral necessities, and that recognition sends me on the way towards a viable formulation of the logical problem of evil.
Step One: Some People Believe
Some people believe that God exists. Some people believe that God is good. Some people believe that God is powerful. Some people believe that God created the world; some people further believe that God can and perhaps does intervene in the world, either by miraculous or more ordinary means. Some people recognise that there is evil in the world. Some people seem to believe all of these things. Of the great and complicated Venn diagram that could be constructed for these beliefs, we will find a certain subset of people that will occupy the intersection of all these beliefs. Some people believe that God exists and is good and powerful, and that that God created this world and can sometimes intervene in it, and that evil exists in this world. So far, so uncontroversially good, no?
The claim of the logical formulation of the problem of evil is that no one can consistently hold all of these beliefs, meaning that some of these beliefs must be rejected or modified in order for rational consistency to be saved. That ‘must’ carries the force of a logical ‘must’, mandated by the law of non-contradiction. It is a powerful mandate, not lightly cast aside.
Where is the inconsistency? Clearly, it’s not clear. More needs to be done to unpack quite how or why it is not possible to believe all these things at the same time. Intuition suggests there is some kind of inconsistency, lurking beneath the surface, but we’re not sure. We need to do some digging to unearth the inconsistency, if it’s there.
Where should we start digging? Where is the inconsistency likely to be found? Some areas are more promising than others. There seems to be little point trying to find an inconsistency between God’s creative act and God’s power, for example; or between God’s power and God’s goodness; or between God’s existence and God’s ability to miraculously intervene in the world. There may be paradoxes to be found there, but they are of no great concern for the problem of evil (not yet, at least). Clearly, the inconsistency we are interested in, when it comes to the problem of evil, is between some or all of the other beliefs and the recognition that there is evil in the world. The clue is in the name: it is the problem of evil, after all.
Which of the other beliefs in the set are understood to be inconsistent with this recognition that there is evil in the world? It’s unhelpful to say ‘all of them’. We want to narrow it down so that we can find somewhere promising to dig. Which of the other beliefs are most likely to be inconsistent with the recognition that there is evil in the world? Or, to put it another way, which of the other beliefs are most obviously inconsistent with a recognition that there is evil in the world? The natural answer, I suggest, is the belief that God is good. There is an intuitive clue in the natural opposition of the words used. Power is intuitively opposed to weakness or ‘impotence’, and there is no mention of weakness here. Existence is intuitively opposed to non-existence, and there is no mention of non-existence here. Creation is intuitively opposed to destruction, and there is no mention of destruction here. Goodness is intuitively opposed to evil, and there is mention of both here. That suggests we should start digging there.
What is it that could be inconsistent about believing that God is good and recognising that there is evil in the world? It will depend on what you mean by ‘God is good’ and ‘there is evil in the world’.
It would be difficult to find a more obvious statement than this: that any inconsistency in the problem of evil will depend on your values. And yet, it is an issue that is often glossed over in any discussion of the problem of evil. Sometimes it is neglected entirely. This can be for superficially virtuous reasons. Often the philosophy of religion, in an attempt to speak as objectively as possible, tries to speak of the ethical (or religious) world as if from no place within it. But we are not interested in what an object might think about the problem of evil; we are interested in what a particular human being believes. Even if that particular human being is represented in a way that makes it objectively representative of a human being ‘as such’, it still requires an ethical perspective on the world in order to be representative. It’s that ethical perspective that will ground any inconsistency in the problem of evil, because it is only an ethical perspective that enables you to recognise good or evil.
Step Two: Moral Meaning
What is it to say that evil exists in this world? It is to make a moral or value judgement. It is to identify something as bad, not to be preferred, perhaps even forbidden. It is to say that it ought not to be. If you say ‘it ought to be’, then what reason would you have to label it ‘evil’? If you didn’t think it was all that bad, wouldn’t you label it something else, such as ‘regrettable, but necessary’? Or ‘tough, but on balance worthwhile’? When I think about the pain I experience when I exercise, I’d hesitate to label it ‘evil’. Uncomfortable, maybe, but not evil. Let’s avoid making a straw man of the evils of the world and focus only on the bad things that can do some heavy lifting in the problem of evil.
When we recognise the existence of evil in the world, we are not interested in the bad things that we suspect are justified by good reason, or even the bad things that we do not know whether or not they are justified by good reason; we are interested in the bad things that happen for no good reason, or at least seem to us to happen for no good reason. We are interested in the bad things that it seems to us could not have a good justifying reason. This lack of good justifying reason causes us to recognise these bad things as things that, in our judgement, ‘ought not to be’. And in most cases, we’re not just talking about bad things but very bad things, the worst kinds of things. And this is what I think people mean when they say ‘evil exists’, in the context of the problem of evil. They mean that very bad things happen that really ought not to happen. A recognition of their badness forces us to see these things as something that we judge ‘ought not to be’.
What is it to say that God is good? It is to say that God has a moral capacity and to determine what that capacity is. It is to say that God has moral judgement, and it is to make a moral judgement about God’s moral judgement. Is God a good God, or an evil God? Not an evil God, surely. Is God a good God, or an indifferent God? Not an indifferent God, surely. What does it mean to say that God is good, morally good, and not indifferent or evil?
It’s a tricky question. There are various ways we use the term ‘good’. Only some of these are distinctively moral. In the context of the problem of evil, when we say ‘God is good’ we tend to mean a distinctively moral kind of goodness. We do not mean ‘God is good’ in the same way as we mean ‘this car is good’ or ‘this food is good’, etc. We do not mean that God is useful, or that God is good at being God, or that God is good for you, etc. We mean God is just ‘good’, morally good. But this raises immediate problems, because the only version of moral goodness that we commonly ascribe to things is a distinctively human concept of moral goodness. We are normally fairly restrictive on where and to what we apply this distinctively human concept. We apply it to persons, and only to certain types of persons. It’s obviously inappropriate to apply a moral concept to non-persons. I might shout and swear at my car if it does not start, even give it a damn good thrashing, but it would be stupid to hold my car morally accountable for its failings. I cannot demand an apology, for example. Even in more liminal cases, it’s clear we can’t straightforwardly apply the same human moral concept to non-humans: we do not mean the same thing when we describe a dog as ‘good’, for example, even though many dogs I’ve known have been better people than many people I’ve known. A distinctively human concept of moral goodness requires that one can be responsive to moral reasons, but since a non-human object or subject cannot respond to those moral reasons, there can be no sensible obligation for them to do so. And therefore we don’t include non-human objects or subjects within the conceptual space of being responsive to moral reasons. On that basis, we hesitate to assign them a moral capacity.
But that means that if we are to speak meaningfully of God being morally good and assign God a moral capacity, we need to include God in that conceptual space. We do this intuitively when we describe God as ‘good’ and certainly not ‘evil’, for example. This is to make a moral judgement about God’s moral capacity, which implies that God has a moral capacity. And that implies that God must be responsive to moral reasons.
What is it to be responsive to moral reasons? Another tricky question. I tentatively suggest, in simple, circular, and evidently self-serving terms, it is an ability to recognise and be responsive to our judgements about what ‘ought’ and ‘ought not’ to be. If our judgement recognises something as ‘good’ and ‘ought to be’, then that recognition in itself gives you a reason to do or promote or favour it. If our judgement recognises something as bad or ‘ought not to be’, then that recognition gives you a reason not to do or promote or favour it; it gives you a reason to change it to what ‘ought to be’, or at least something that is not ‘ought not to be’, or else prevent it entirely, if you can. Perhaps these are over-simplifications too far, but that you probably believe them to be true can be shown by imagining the counterexample: what would it mean to recognise that something ‘ought not to be’, yet see no reason not to do or promote or favour it? Such as saying and sincerely believing that ‘you ought not to beat children’, yet seeing no reason not to beat children. That seems to me to be contradictory; I wouldn’t know what to make of such a person; they seem confused. If you think you have no reason not to do what you know ought not to be done, you only show that you are lacking a moral responsiveness.
But if God is wholly or perfectly good, then God cannot be lacking in goodness. And since goodness requires that you have sufficient moral responsiveness, this means that God cannot be significantly lacking in moral responsiveness. That God is good implies that He is responsive to moral reasons, and that He is wholly or perfectly (or even just highly) good implies that He cannot significantly lack in responsiveness to those reasons.
The contradictory tension within the problem of evil is now becoming clearer. When we say ‘evil exists’, we mean that there are some things in this world that, in our judgement, ‘ought not to be’. Our judgement that something ‘ought not to be’ gives us a moral reason not to do or promote or favour it, and to change it or prevent it if we can. When we say ‘God is good’, we mean that God is responsive to moral reasons. It follows that God would seem to have a moral reason not to do or promote or favour those things that we judge ‘ought not to be’, and to change or prevent them if He can. A good God does not want evil to exist; a wholly good God cannot allow what ought not to be.
Whilst this might be enough to establish this statement as true, for you, this is not yet a necessarily true statement; but it is getting there. More digging is required to make it clearer.
Step Three: Moral Necessity
To say that ‘God is good’ is to say that God has a moral reason to prevent the occurrence of what ought not to be. To say that ‘evil exists’ is to say that something is that ought not to be. It follows that God has a moral reason to prevent the occurrence of evil. But perhaps this is only a prima facie or pro tanto reason? Perhaps this moral reason can be overridden or overruled? Only, perhaps, under certain ethical perspectives. But we should allow the possibility of those ethical perspectives. How can we firm up our moral reason to make it something that cannot be overridden or overruled?
Or perhaps our judgements about what ‘ought’ and ‘ought not’ to be are different from God’s? And so whilst we recognise various reasons to do or not do certain things, those reasons are not recognised by God because He does not share our judgement. Value judgements are notoriously difficult to universalise, after all. How can we firm up our moral reason to make it something that applies universally?
I think these questions suggest an obvious concept: the morally necessary, or (more accurately in the case of evil) the morally impossible. There are many versions or expressions of this moral modality, and clearly any acknowledgement of it will depend on your ethical and metaethical perspective. Some of these ethical perspectives will be more hospitable to the concept than others. All that matters is that you recognise some moral reasons to be undeniable and incapable of being set aside; they necessitate a moral responsiveness. There are some things that categorically ought not to be; no ifs, no buts, no obfuscations, no excuses, and no escape from the guilt and remorse that necessarily follows if it is done, promoted, favoured, unprevented, or unchanged. (This is not to say that these things cannot happen, of course, because moral necessity is not alethic. But that is not relevant here because we are looking for an inconsistency amongst values, not between values and contingent states of affairs in the world.) I will discuss this concept, and the role it might play in the problem of evil, in more detail in chapter four: ‘Moral Modalities and the Limits of Moral Thinking.’ All that is necessary to establish for now is that a) there are such things as moral modalities, and b) those concepts can be applied here, in some way, to the problem of evil. I am aiming for relatively minimal claims here, only establishing enough to ground the inconsistency in the problem of evil; I will leave the more expansive claims for later.
I think most people would happily concede that there are moral modalities, that there are lines to be drawn between the morally possible and the morally impossible, even if they might disagree wildly about where the lines would be drawn and why. For this reason, perhaps it’s best to avoid any detailed line-drawing at all and remain at the level of loose intuition; you can decide for yourself where you draw the moral lines, if you want, all I ask is that you accept that there are lines to be drawn.
For example: Let’s say that it’s true that you ought to give to charity or that you ought not to lie to your friend, and let’s say that it’s true that you ought not to commit murder or rape. Clearly these types of obligations would be true in different ways. You ought to give to charity, perhaps, but it’s not fatally undermining of your moral status if you don’t; it’s morally possible to not give to charity and remain a tolerably good person. You do not need to give to charity. There might also be contingent states of affairs that affect your ability to fulfil this obligation, such as that you are poor and lack the surplus means to give to charity. If this is the case, then you’re off the hook. It is morally possible to justify your lack of charitable giving. Likewise, you ought not to lie to your friend, but it’s not unthinkable that you should do so. Perhaps you would rather tell a kind lie than a cruel truth. It is not necessary that you don’t lie to them. Perhaps it might even, under certain circumstances, be better to lie to your friend; it might seem like the right thing to do. These are all within the realms of moral possibility.
But if you commit a murder or rape, that is not the same thing. To commit a murder or rape is to step into a radically different moral status, destroying any moral status you once had. It is morally impossible to do these things and remain the same person you once morally were. There might be contingent states of affairs that affect your ability to fulfil this obligation – or ‘extenuating circumstances’, as we might say – but these are not sufficient to absolve you of the guilt of murder or rape. (Note I say ‘murder’ rather than ‘killing’, such as in self-defence. That we recognise a difference between the two – that you can have ‘killed’ but not committed ‘murder’ – is another example of the meaning of our concepts being defined by the limits of justification via a morally-sufficient reason. We recognise no comparable possibility of justification in the case of rape.)
It is morally necessary that you do not commit murder or rape. We recognise this in our moral responsiveness, when we might ask ‘How could you do such a thing?!’, rather than ‘I think you could have done better’. We would say you have done the morally impossible, rather than opted for the wrong choice amongst a range of moral possibilities. We capture this difference when we say that it is unthinkable that we could commit murder or rape, that murder or rape is not an option, even if it could somehow be a solution to our problems; but presumably it is ‘thinkable’ and an open option to solve our problems by lying to our friends or not giving to charity. And yet you would still say that it’s true that you ought not to commit murder or rape, and that it’s true that you ought to give to charity and not lie to your friends. The two types of moral obligation are true in different ways. One is expressive of a moral necessity, the other of moral contingency.
To be clear, I am not saying what is or must be considered to be morally impossible, or even that anything in particular is morally impossible; I’m not even asking that you agree with my example and say that murder or rape is morally impossible (though I reserve the right to look at you funny if you disagree…). And it is not relevant that some types of moral violation are more destructive than others, or that some are obligations and others prohibitions, or that these things would be seen differently in the eyes of the law. All the examples are intended to show is that there is a conceptual difference between the morally possible and the morally impossible; that some things are not right, but could be, and other things are not right and can never be. Do you recognise this difference? If you do, then I suggest that you recognise that there are such things as moral modalities, in some form. The question that remains is whether and to what extent this applies to the problem of evil.
I think it does apply to the problem of evil. In fact, I think it is absolutely fundamental to our understanding of the problem of evil. That is why I will discuss it in more detail later. For now, it will suffice to say that if there is any meaning to be given to the morally impossible, at all, then it must apply to the problem of evil, because the ‘evil’ that the problem of evil considers is not just a particular instance or type of evil, such as murder or the refusal to give to charity, but any evil that has ever occurred. As such, if the morally impossible has ever happened, then the problem of evil will consider that as an instance that might be incompatible with the existence of a good and powerful God.
Has the morally impossible ever happened? This depends on where you draw the line on what is or is not morally impossible. We all know that truly terrible things have happened. Did any of them violate a moral necessity? Are any of them such that we would judge that they categorically ought not to be?
I suggest, minimally, that some of them could be. I suggest, minimally, that we should not exclude that possibility. I suggest, minimally, that it is possible that the morally impossible has happened.
For me, I believe that the morally impossible has happened and does happen, a lot. I think it happens all the time, so often that it is practically routine. That I recognise such an intolerable frequency of violations of moral necessity is likely what makes me side with Ivan Karamazov and reject any story of an ‘eternal harmony’ that is achieved on that morally impossible basis, even if I am not right. But these are more expansive claims and can be left for later chapters.
All that is necessary for my argument is the extremely minimal claim that it is possible that the morally impossible has happened. It’s possible that some of the terrible things that have happened have violated a moral necessity. This claim is defensible by argument and not only intuition. I suggest that if there is any meaning whatsoever to the concept of the ‘morally impossible’, then that concept cannot be necessarily empty: that is, it cannot refer only to things that have never and could never happen. If it were such an empty concept, it would remain only ever hypothetical, an intellectual construction that can have no real reality (like a square circle), a figment of the imagination; that would seem to undersell it, and it would then be difficult to see what role the concept could play in our down-to-earth moral thinking. Because we do recognise the conceptual difference, do we not, between the morally possible act of lying to your friends and the morally impossible act of raping them? Or the morally possible act of an interventionist war (however ill-judged) and the morally impossible act of genocide? But if we insist that the morally impossible cannot happen, then were we to actually do the morally impossible, we would thereby show it not to be so. And that cannot be right.
We can say, then, that it is possible that the morally impossible does happen. Some of us will think this possibility is actual, others will disagree. Universal agreement is not necessary for my argument.
Step Four: Moral Tragedy
Let’s recap: For the purposes of finding an inconsistency between the belief that God is good and the belief that evil exists, we found: First, that ‘evil’ represents things that we judge ‘ought not to be’, and that this gives us a moral reason not to do or promote or favour these things, and to change or prevent them if we can. Second, that God’s ‘goodness’ entails that God must be responsive to moral reasons, and therefore a good God does not want evil to exist and a wholly good God cannot allow what ought not to be. But it is possible that these moral reasons could be overruled or overridden, or else that God does not judge as we judge and so share our moral reasons. To avoid falling into these blind alleys, we can restrict our talk to only those moral reasons that cannot be overruled or overridden or opted out of: the morally impossible. With this amendment in place, ‘evil exists’ is taken to mean not only that things are that ought not to be, but that things are that categorically ought not to be.
No one is under obligation to believe that these morally impossible things have happened, of course; that is something you will need to decide for yourself. But if you think that the morally impossible has happened, then you will find that belief to be inconsistent with a belief in the existence of a wholly good God. Because such a God would be morally necessitated to prevent those things from happening, if He can.
This phrase ‘if He can’ suggests that there is one final gap that remains to be plugged: God’s ability to prevent what categorically ought not to be. I don’t think this divine attribute needs to be overstated or overcomplicated, in this context. The amount of power required to change these categorically awful things would not be all that much, in itself. It is the kind of power that even lowly human beings have. We can cure children of cancer, sometimes; we can prevent rapists from perpetrating, sometimes; we can prevent genocidal atrocities, sometimes. Terrible things can be prevented by happy accident. It’s not that uncommon for cars to randomly break down in the middle of nowhere; we’ve all been there. Is it too much to ask that the terrorist’s car breaks down on the way to their target? And as they sit there on the side of the road, waiting for the AA, they get cold and tired and rethink their life choices. Is it too much to ask that the rapist slips and breaks their ankle, and cries in pain for help, never appreciating the irony of choosing somewhere quiet and isolated to pursue their nefarious purposes? We’re in danger of slipping into slapstick trivialisations. But the point remains that the level of intervention required is extraordinarily minimal; it is entirely ordinary, in fact. There is nothing that would be obviously supernatural in these examples, allowing God to retain His epistemic distance. There is nothing here to suggest that God could not physically prevent what categorically ought not to be.
Perhaps you will object that there might be something logically preventing God from intervening. A creature’s free will, perhaps, or the prevention of one morally impossible event only by the permission of another. If this were the case, then God’s inability to prevent the morally impossible would seem to undermine any sense that God has an obligation to prevent what categorically ought not to be. Ought implies can, after all, so God cannot ‘ought’ to do what God cannot do. Ordinarily, this point would be telling, especially if we were talking about prima facie or pro tanto obligations. But it won’t change anything in the face of moral necessity.
With moral contingencies, the ‘ought implies can’ principle holds very clearly. If you have failed to do something that you feel you ought to have done, but then discover that you could not have done differently, it gets you off the hook. The inability to fulfil an obligation can affect and possibly eliminate your moral responsiveness to that obligation, and there is nothing obviously wrong with it doing so. But it does not seem to work in the same way with moral necessities. If you violate a moral necessity, your guilt might be lessened by learning that you could not have done differently, but it is not eliminated. The clearest examples here are ‘Sophie’s choice’-style scenarios in which you have a forced choice between two morally impossible options. That you have no choice but to do the morally impossible does not affect your moral responsiveness to those obligations, let alone eliminate it. If someone has suffered such a moral tragedy, we might encourage them to not feel too bad about what they’ve done, but we wouldn’t expect them to feel indifferent to what they have done, safe in the knowledge that they had no choice. We would expect some grief, some remorse, some recognition of the meaning of what they have had to do. The morally necessary nature of the wrong that has been done necessitates some moral responsiveness. If someone were to feel genuinely indifferent to the participation in a moral tragedy such as this, we would likely question their moral responsiveness, in the same way that we might question the moral responsiveness of someone who orders the detonation of an atomic bomb over civilian populations, twice, exclaims ‘this is the greatest thing in history!’, and loses no sleep over that decision, even if they felt they had to do it. We might rather empathise with the bomber’s copilot’s reaction: ‘My God, what have we done?’
We don’t ordinarily incinerate the children of our enemies in an attempt to preserve the lives of our soldiers. Whether or not it is sometimes right or necessary to do the morally impossible, we are free to question whether one ought to feel right about it. We might still think that participation in the morally impossible, even if impossible to avoid, has an inescapable impact on your moral status. That you are damned if you do and damned if you don’t is what makes it a moral tragedy, but you are damned either way. If you lacked a moral responsiveness in such a morally tragic situation, feeling yourself absolved because you had no choice, we might question whether you recognise the morally impossible nature of what you have done. When we discover that we have done the morally impossible, we feel remorse, we seek forgiveness, we do what we can to atone, but we do not justify ourselves because to do so would be to deny the nature of what we have done: it would be to say that we ought to have done what categorically ought not to have been done. That is at the very least inconsistent.
There is a difference between doing and letting be done, of course. When it comes to the problem of evil, it’s rare to suggest that God is perpetrating the morally impossible. Ordinarily we say only that He looks on from the heavens and will not take His part. But whilst there is a difference between doing and letting be done, when it comes to the morally impossible, letting be done is sufficient to incur unavoidable guilt. The distinction between moral necessity and moral contingency remains even at this distance of ‘omission’ because the power of necessitated moral responsiveness remains. You can stand by and watch one friend lie to another, without intervening with the truth, and nothing compels you to do otherwise. You can recount this story later, without expression of remorse, and we probably wouldn’t think any less of you. But to stand by and watch one friend rape another, without intervening, and consider yourself free from guilt on account of the fact that it was not you that did the deed? Could you tell that story later, without expression of remorse, and expect people to see you guiltless? ‘How could you stand by and do nothing?!’, they might say. They would be incredulous that you were not more morally responsive in the face of the morally impossible.
Can it make a difference, then, if God has no choice but to do or permit the morally impossible? If we are to stay true to our ideas about necessitated moral responsiveness in the face of moral necessities, it cannot. God might have to do or permit the morally impossible, but God cannot escape the impact that this has on His moral status. If He is responsive to moral reasons, He will know this as much as we do. He becomes a grief-stricken and guilt-ridden God, a damned God, a God who knows He has done wrong, even if He had no other option. He would not offer excuses or justifications, and He would not ask or allow anyone to do so on His behalf. He would seek our forgiveness and look to atone. If He does not have this moral responsiveness then He would be callous and insensitive. Not even God would call Himself ‘good’ under either circumstance. He would say ‘Oh Me, what have I done?’ and not ‘this is the greatest thing in history!’ He would ask His children to forgive Him, for He knows what He has done.
It’s possible I have over-egged this pudding. It’s possible I have put more emphasis in this argument than is needed. I think it is enough to say that the morally impossible has happened, that a good God could not let the morally impossible happen, and that if He did then He would incur guilt for doing so, even if He had no other option.
The Logical Problem of Morally-Impossible Evil
I think we can find an inconsistency in the problem of evil if we look to our values and to the concept of the morally impossible. Such a problem of evil would look like this:
1. God is good (by which we mean wholly good).
2. God is powerful (by which we mean all-powerful).
3. Evil exists (where ‘evil’ is the morally impossible, what ‘categorically ought not to be’).
4. ‘God is good’ means that God is responsive to moral reasons and so must be responsive to moral necessity.
5. ‘Evil’ here means what is morally necessary to prevent or change, such that you incur guilt if you do not.
6. ‘God is powerful’ here means that God is powerful enough to prevent the morally impossible in all cases except morally tragic cases where the permission of the morally impossible is logically unavoidable.
From this it follows that:
7. The only evils that can exist in the world, consistent with this set of beliefs, are those of a morally tragic nature.
But:
8. It is the nature of a moral tragedy that those who do or permit the morally impossible recognise that they have done wrong, even if they had no other option, and that that wrongdoing prevents them from being considered wholly good.
From this it follows that:
9. If God permits the morally impossible in a case of moral tragedy, God cannot be considered wholly good.
From this, we should be able to generate an inconsistency. If, in permitting the morally impossible in the case of a moral tragedy, God cannot be considered wholly good, then in order for Him to be considered wholly good, it is necessary that God does not permit the morally impossible in the case of a moral tragedy. But if this is true, and God does not permit the morally impossible except in cases of moral tragedy, then God ought not permit the morally impossible at all. And yet the morally impossible is permitted by Him when it happens. The set is inconsistent.
The only options that remain are to deny one of the constituent propositions and say that the morally impossible does not happen, or that God is not wholly good, or that God is not all-powerful (and given the relatively low level of power required here, really this is to say that God is not at all powerful). Alternatively, you could reject the ethical perspective underlying all this talk of necessitated moral responsiveness. These are open options and philosophers can and do occupy these positions: You could say that moral goodness does not necessitate responsiveness to moral reasons; you could reject the concept of the morally necessary; you could reject that doing the morally impossible in the case of a moral tragedy prevents you from being considered wholly good. Each of these comes at a cost. But either way, that philosophers can and do occupy these positions does not change the inconsistent nature of the problem of evil any more than that some philosophers have been willing to say that God is not wholly good, or not all-powerful, or that evil does not exist.
The original three propositions of ‘God is good’, ‘God is powerful’, and ‘evil exists’ have been shown to be inconsistent with one another. The additions that reveal this inconsistency, which take the form of clarifications of ‘what we mean when we say’, are necessarily true. Not true of metaphysical necessity, perhaps, nor logical necessity, nor even physical necessity, but true of moral necessity. You cannot say that the morally impossible ought to happen or that you can be indifferent to it happening. If you recognise the concept and agree that it does sometimes happen, then you will face the challenge of clarifying and if possible reconciling that belief with your beliefs about God’s goodness and power. That is the logical problem of morally-impossible evil.
The inconsistency is at the level of values only. It should not be considered any less necessary for that. In fact, I think focussing on values makes it easier to find the kind of necessity that we’re looking for. Rather than trying to generate an essential inconsistency between contingent states of affairs, or between some contingent states of affairs and some other conceptual/analytic claims, I say ‘necessarily this or that value is inconsistent with this other value’: your value judgement relating to the morally impossible is inconsistent with your value judgement about God’s goodness and moral responsiveness. This can be necessarily true, even when we are talking about contingent states of affairs, and even when the value judgements themselves are not necessarily ‘true’ claims. What matters is only that these ideas are related to each other in a necessary way.
A good thing cannot permit the morally impossible without incurring guilt. A guilty thing cannot be considered wholly good. Therefore, there cannot be a wholly good thing that permits the morally impossible. This statement is necessarily true and so it can establish an inconsistency in any set of beliefs that includes the claims that there is a wholly good thing that permits the morally impossible.
That this is a logical problem entails that the theist wishing to retain their set of beliefs must offer a solution to it, in the form of a theodicy, or else reject the ethical perspective that underpins all talk of moral necessity. In the next chapter, I explore the first option. The prospects are not promising, I think, because all solutions to the problem of evil can be shown to be morally problematic, especially when the problem of evil is understood in this morally necessary way. If this is true, then only the other option would remain for the theist, which is to reject the ethical perspective that gives rise to any talk of moral necessity. That comes with its own morally-problematic costs but it remains a bullet to be bit. I suspect it is the more preferable of the two options. This serves to expose a key point of difference between the theist and the atheist on the matter of the problem of evil, understood as an ethical problem: the theist must believe that everything that has happened could be justified by a morally-sufficient reason, whereas the atheist need not believe such a thing. The disagreement here is one of moral modalities. The theist cannot believe that the morally impossible has happened, but the atheist can. They inhabit different moral worlds: in one world there is no conceptual space for the reality of the morally impossible; in the other, there is. Depending on your moral beliefs, this might give you some reason to swing towards atheism over theism. That reason would be a moral reason: a recognition of the reality of the morally impossible and an incapacity to remain indifferent to it.
tobybetenson.com 