Psychic immunity and uncomprehended pain: what Maimonides can tell us about the problem of suffering

Abstract

Using Moses Maimonides’ theodicy to respond to contemporary formulations of the problem of evil initially seems unpromising. Maimonides is committed to claims that make the task harder rather than easier. Chief among them is his belief that all suffering is deserved by the sufferer. But Maimonides is often misinterpreted: he does not hold that innocent people are never subject to bodily harm, but that it is possible to achieve a kind of ‘psychic immunity’ from suffering via intellectual enlightenment, and that failure to do so is blameworthy. I argue that while the Maimonidean psychic immunity theodicy has some attractive features, it struggles to explain ‘inculpably incomprehensible’ suffering: that of infants and people with serious cognitive disabilities. I propose two responses: defending Maimonides’ intellectual elitism using work on moral status from Singer and McMahan; and defending a more limited version of the theodicy grounded on ‘sceptical’ readings of Maimonides that emphasize the limitations of human knowledge. I conclude that the second is more promising, and that the limits of Maimonides’ theodicy point to more general limits on theodicies that insist on what I call ‘first-personal adequacy’ – the requirement that a theodicy provide a satisfying explanation of suffering to sufferers themselves.

The medieval approach to the problem of evil differs in many ways from that of most contemporary philosophers. Like Epicurus and Hume before them, contemporary writers tend to frame it as a problem about the existence of evil or suffering that threatens the rationality of belief in God. The medievals are different. They do not try to prosecute or defend God for allowing evil or suffering: it's generally taken as given that God exists and could have no case to answer to human beings. Rather, they aim to explain the value and purpose of evil and suffering, to make sense of them within God's created order. Perhaps in part because of this, they are more inclined to what I'll call third-person theodicies than first-person ones. A first-person theodicy, is, roughly, a justification or satisfactory explanation for the suffering that an individual person is experiencing, an explanation that the sufferer could themselves accept. A first-person theodicy is a response to the question ‘why am I suffering?.’ A third-person theodicy, on the other hand, is a justification for the suffering of others, or for the existence of suffering itself. The same theodicy can play both a first-personal and third-personal role, but it is possible for a theodicy to be the latter without being the former. Contemporary theodicies tend to start from the perspective of the sufferer, and regard first-person adequacy for all people undergoing serious suffering as a condition of a successful theodicy.

These considerations might make us sceptical about turning to the medievals for help solving our problem of evil. Moses Maimonides might seem a particularly unpromising candidate. Other medievals – Thomas Aquinas, Saadya Gaon, Ibn Sina – share this much in common with contemporary philosophers: they hold that terrible suffering can be experienced by people who do not deserve it. Maimonides on the other hand denies that there is such a thing as the suffering of the righteous, or of the innocent. He thinks that any suffering must be deserved by the sufferer. Maimonides’ rationale for this is summed up in his discussion of the book of Job in his Guide of the Perplexed (III.24).¹ He holds that Job's sufferings would not have been as devastating to him – would perhaps not have been true suffering at all – if he had possessed a certain kind of knowledge, knowledge that he gains by the end of the book. Maimonides proposes that obtaining this knowledge enables a person to develop, to borrow a term from David Shatz, ‘psychic immunity’ to suffering. Further, gaining the relevant knowledge and the psychic immunity that comes with it is a consequence of morally and intellectually perfecting oneself. If someone fails to perfect themself in this way, they are not truly righteous and deserve the suffering from which they have failed to protect themself.

This article argues that Maimonides’ psychic immunity theodicy fails to address what I call the ‘inculpably incomprehensible’ suffering of very young children, infants, and people with radical cognitive impairments.² However, the limits of Maimonides’ theodicy point to something of broader relevance. This is that any theodicy that successfully justifies or explains the suffering of people with very limited cognitive faculties will be unable to do so by explaining it to the sufferers, at least in this life. Any successful theodicy will need a third-person element that is not first-personally adequate: a way of explaining the suffering of people who cannot understand it themselves. This is, I argue, something we can learn from the medieval approach to the problem of evil. What's more, even if Maimonides’ theodicy is not successful, something like the concept of ‘psychic immunity’ can do some theodic work.

I'll proceed as follows. First, I outline a standard contemporary version of the problem of evil or suffering and lay some of the groundwork for Maimonides’ response to his version of the problem. I then describe his psychic immunity theodicy. Although many of Maimonides’ theodic arguments take a ‘third-personal’ form, I argue that Maimonides’ idea of psychic immunity can do ‘first-personal’ work in a way that could make his arguments relevant to a contemporary version of the problem. Next, I argue that the psychic immunity theodicy has little to say about the suffering of anyone inculpably incapable of gaining this immunizing knowledge. This places hard limits on the scope of the theodicy's first-personal adequacy. In the penultimate section I explore two ways of modifying the theodicy to address this problem. The first draws on work on moral status from Peter Singer and Jeff McMahan, embraces Maimonides’ intellectual elitism, and reduces the number of individuals whose suffering really requires first-personal justification. The second leans on so-called ‘sceptical’ interpretations of Maimonides’ thought to ground a more limited and modest role for the psychic immunity theodicy. I argue that the sceptical option is more promising. The limitations it leaves the psychic immunity theodicy with, I conclude, shed light on the limitations of any first-personal theodicy.

The suffering of the righteous

A modern problem

Here is one way of framing the problem of evil.

1. (1) God is omniscient, omnipotent, and perfectly good.

2. (2) If God (as described in 1) existed, God would not allow innocent people to suffer terribly.

3. (3) However, innocent people do suffer terribly.

4. (4) Therefore God (as described in 1) does not exist.

Let me note a few features of this argument. First, I follow Eleonore Stump in talking about suffering rather than evil.³ I do this mostly for reasons of scope: even if Stump is wrong to think that the problem of evil is essentially a problem about the existence of suffering, suffering is at least among the most significant evils to be explained. Forms of moral evil that involve no suffering, if there are any, would raise different issues from those I want to consider in this article. Second, by ‘innocent’ I mean people who have not committed any bad actions which would render them deserving of terrible suffering. Of course, it might be that there are few or no such actions, and that no-one or very few people deserve such suffering, but the use of the word ‘innocent’ is also meant to focus our attention on the people who seem least plausibly to deserve it: people who have lived good and virtuous lives, or children, for example. The terrible suffering of such people is a way in which the problem of evil presents itself particularly vividly. The reference to ‘terrible’ suffering is meant both to highlight the worst cases of such suffering, and to clearly exclude some cases of pain that might not count as suffering at all, such as the muscular pain an athlete might feel at the end of a long run.

Finally, a little more about the difference between first-personal and third-personal theodicies and how they interact with this ‘standard’ problem. A theodicy is first-personally adequate for person P if it provides a justification or satisfactory explanation for P's own suffering that P can endorse or accept. It will not necessarily be an emotionally satisfying answer to a question posed by a sufferer in the midst of their suffering: it is not a condition of first-personal adequacy that the theodicy in question is satisfying in the torture room. Rather, it is supposed to be intellectually satisfying to a person reflecting on or anticipating their own suffering in a cool hour. Nor does the theodicy have to be comprehensible to the sufferer in its most complex form. A sufferer unfamiliar with analytic philosophy might be unable to understand a theodicy formulated using lots of technical jargon. However, so long as it is possible to present the theodicy in a way that the sufferer could endorse or accept, the theodicy can pass the test of first-person adequacy for them.⁴

A third-personal theodicy is one that justifies or explains the existence of person P's suffering, or of many people's suffering, or of suffering in general in a way that is satisfying to whoever is looking for the explanation, but not necessarily to the sufferers themselves. The same theodicy can fulfil a first-personal and a third-personal role, but it's conceptually possible for a theodicy to be third-personally adequate without being first-personally so. Consider as an example Pope Gregory the Great's claim (endorsed by Thomas Aquinas) that the sufferings of the damned will in part be justified by the benefits that accrue to the saved.

The wicked are all condemned to eternal punishment, and are punished for their own wickedness. Yet they will burn to some purpose, namely that the just may all both see in God the joys they receive, and perceive in them the torments they have escaped: for which reason they will acknowledge themselves for ever the debtors of Divine grace the more that they will see how the evils which they overcame by its assistance are punished eternally.⁵

This part of the theodicy makes no attempt to be first-personally adequate for the damned: their suffering is of no benefit to them, and its benefits to others are not willed or endorsed by them. But for both Gregory and Aquinas, this fact does not weaken its adequacy as a partial explanation or justification for their suffering. They take it to be third-personally adequate regardless of its first-person adequacy. As we will see below, Maimonides too doesn't take first-person adequacy to be a necessary part of many of his theodic arguments.

Contemporary theologians and philosophers of religion tend to be much less sympathetic to these kinds of theodic arguments. In what follows I will be concerned with how well Maimonides’ ‘psychic immunity’ theodicy meets a constraint of first-person adequacy. This constraint is that for a theodicy to be third-personally adequate to explain innocent person P's terrible suffering, it must also be first-personally adequate for P. As such, to meet this constraint a successful theodicy must be first-personally adequate for each innocent or righteous sufferer. I take this to be one way of articulating a commitment often implicit in contemporary discussions of the problem of evil.⁶

Note that while the ‘standard argument’ above is framed in a third-personal way, it could still be the case that any successful theodic response to it would have to meet the constraint of first-personal adequacy. For example, the second premise could be challenged by the claim that God has a good reason to allow innocents to suffer, because doing so helps other people to better appreciate his mercy. This theodic argument could be rejected as failing the test of first-personal adequacy for the innocents in question.

Maimonides’ problem

As David Shatz points out, the extent to which any ‘standard’ problem of evil arises for Maimonides is highly debatable.⁷ For Maimonides, our words refer to divine attributes in a highly equivocal way: when a predicate is applied both to God and to creatures, the terms are essentially homonyms. This means that it is hard for arguments like the above to find a grip within his framework. If God's goodness is sufficiently different from our goodness, any premise like ‘a being with a very high level of human goodness would wish to prevent x’ will necessarily not apply to God, and any inference making use of such a premise will be invalid, committing a fallacy of equivocation.

Maimonides, however, does spend a substantial amount of time trying to understand the suffering of the innocent or righteous and its purpose within the operation of God's providence. My aim here is not to reconstruct Maimonides’ problem and assess his different solutions. Rather I want to focus on one kind of theodicy that Maimonides develops to solve his problem, and its potential to respond to a more contemporary formulation like the one above. In this section I will explore elements of Maimonides’ thought that lay the groundwork for the ‘psychic immunity’ theodicy, before explaining the theodicy itself in the next section. Maimonides has a number of different theodic arguments about suffering in general, so a fortiori the suffering of the innocent. He returns repeatedly to the evils that arise from matter, from our status as material creatures. Perhaps his key theodic claim is one about metaphysical necessity: that, as there is no form without matter, God needed to make us out of matter in order for us to have intellects at all (III.10, III.8 482–3).⁸ But – also as a matter of metaphysical necessity – matter ‘never ceases to be joined to privation’ (III.8, 430–431) and is thus subject to corruption, death, and disease. As Maimonides writes, ‘He who wishes to be endowed with flesh and bones and at the same time . . . not to be attained by any of the concomitants of matter merely wishes, without being aware of it, to combine two contraries’ (III.12, 444). Even God cannot do the metaphysically impossible.

So far, so third-personal: this is an explanation for the existence of suffering, but not one that seems able to provide a satisfying answer to each sufferer's first-person questions about their own terrible suffering. One reason for this is that suffering seems unequally distributed. Some people suffer much more terribly than others, and Maimonides’ argument about metaphysical necessity doesn't explain this. Specifically, the argument from matter fails to answer a terrible sufferer's first-personal questions about the extent of their own particular suffering. A person undergoing terrible suffering might look at another person who seems to have lived out their whole life experiencing much milder suffering. They might ask ‘why am I, in particular, suffering so much?’ and this question would not be answered by an appeal to metaphysical necessity. The person they are comparing themselves with, after all, is also composed of matter. ‘If God could allow that person to suffer so much less without doing anything metaphysically impossible’ our sufferer might justifiably wonder, ‘why not me?’ For this reason, the argument from matter does not seem to meet the first-person adequacy constraint.⁹

Many of Maimonides’ arguments are equally third-personal in their theodic character. When he considers the suffering inflicted by natural disasters (one example of the apparently arbitrary distribution of suffering) he makes the point that such disasters are relatively rare (III.12, 444). Modern readers will be unimpressed: the rarity of natural disasters is not particularly relevant to the person hit by the tsunami. Even if such answers work as responses to some third-personal questions like ‘is the world that God created generally good?’, they lack first-person theodic adequacy.

Maimonides’ most plausible candidate for a theodic claim that aspires to first-person adequacy is on its face one of his most unpleasant. He is committed to the claim that suffering is always deserved by the sufferer, a view that he thinks is supported both by rabbinic tradition (what he calls ‘our opinion’) and the Hebrew Bible (what he calls ‘my opinion’). He rejects the opinion of the sages and the Mu'tazila that sufferings are sometimes permitted by God so that sufferers will receive a greater reward in a future life (III.17, 470–471).¹⁰ Instead he thinks that sufferings are allotted to humans, and only humans, by ‘divine will in accordance with the deserts of those people as determined in His judgments, the rule of which cannot be attained by our intellects’ (III.17, 472). Divine providence might protect a human from suffering but ‘does not watch in an equal manner over all the individuals of the human species, but providence is graded as their human perfection is graded’, and thus ‘His providence that watches over excellent and righteous men is proportionate to their excellence and righteousness’, while the ‘ignorant and disobedient’ are in a state that is ‘despicable’ in proportion to their lack of providential care (III.18, 475). He cites scripture repeatedly in defence of this claim, and calls this one of the ‘fundamental principles of the law’ (III.18, 476).

In other words, where many theodicies challenge the second premise of the ‘standard’ argument above, by providing reasons why God would allow innocent people to suffer, Maimonides instead wants to challenge the third premise.¹¹ He denies that there are any innocent sufferers. People suffer in direct proportion to their lack of righteousness.

Note that this claim, if it were true, would be a first-personally adequate answer to the problem of the suffering of the innocent. A person's question ‘why do I suffer when I am innocent?’ would be met with the answer ‘if you are suffering, you are not innocent’. The problem would be answered by being dissolved.¹² But it seems very implausible that Maimonides’ claim is true. To bring out the implausibility, we can examine an apparent tension between different claims in Maimonides’ account. He reads the song of mishaps from the Psalms as describing God's ‘great providence and the safeguard and the protection from all bodily ills, both the general ones and those that concern one individual rather than another, so that neither those that are consequent upon the nature of being nor those that are due to the plotting of man would occur’ (III.51, 626–627). He seems to be saying that divine providence offers protection both from what moderns call ‘natural evils’ or sources of suffering, and from ‘moral evils’ or suffering caused by other agents. Given Maimonides’ belief that providence watches over people in proportion to their righteousness, this seems very strange – in reality, the righteous certainly do not seem to be protected from pestilence, hails of arrows, or physical suffering generally.

How to make sense of this? Maimonides could rely here on his claim that ‘the rule of [God's judgments] cannot be attained by our intellects’ (III.17) and respond that apparently righteous sufferers are in fact all secret wrongdoers. But this would be extremely implausible, and would lend support to a pernicious sort of victim-blaming. Maimonides, at any rate, doesn't rely on this sort of claim. He says something quite different in his treatment of the three different kinds of evil or suffering that befall humans in Guide III.12. The three kinds of evil are ‘that which befalls man . . . because of his being endowed with matter’; ‘those [evils] that men inflict upon one another’; and ‘those that are inflicted upon any individual among us by his own actions’ (III.12, 443–445). I want to focus here on what Maimonides says about the second kind of suffering: that which results from human action for which an agent other than the sufferer is responsible. Maimonides draws a contrast between these evils and the third, self-inflicted kind: unlike self-inflicted evils, ‘the wronged man [has] no defence against them’ (III.12, 444). Here we come to the tension. If physical suffering befalls a man in proportion to his wrongdoing, then surely there is a defence against being wronged by others – simply refrain from doing evil oneself. If God defends righteous men from all physical harm, they would have to be immune from this second kind of evil. But Maimonides claims they have no defence.

This is a crucial hint that Maimonides’ interpretation of the song of mishaps at III.51 should not be read as claiming that the righteous are protected by providence from physical harm. Any apparent inconsistency can be resolved by reinterpreting what ‘protection from bodily ills’ amounts to for Maimonides. It's not that God's providence blocks the physical ‘evil’ from occurring to a righteous would-be sufferer. Rather, God's providential protection prevents bodily ills from impacting their fundamental happiness. Something about such a person's mental state is such that they do not suffer, or at least suffer very little. This is entirely consistent with Maimonides’ discussion of the Book of Job, which argues that the something in question is a kind of knowledge. This discussion concludes by describing the insight that Job gained during his dialogue with God, an insight that he initially lacked:

This is the object of the Book of Job as a whole . . . that you should not fall into error and seek to affirm in your imagination that His knowledge is like our knowledge or that His purpose and His providence and His governance are like our purpose and our providence and our governance. If man knows this, every misfortune will be borne lightly by him. And misfortunes will not add to his doubts regarding the deity and whether He does or does not know and whether He exercises providence or manifests neglect, but will, on the contrary, add to his love, as is said in the conclusion of the prophetic revelation in question: Wherefore I abhor myself, and repent of dust and ashes. (III.24, 497; bold emphasis mine)

This idea is the basis of the theodicy that is the subject of this article. It offers the prospect of what Shatz calls ‘psychic immunity’ from suffering.¹³ Even those physical ills which are unavoidable due to metaphysical necessity or the actions of others can be, for the person who has the right kind of knowledge, ‘borne lightly’.¹⁴ Note that what Maimonides describes here is Job receiving a first-personally adequate answer to his questions about his suffering! The knowledge Job gains leaves him in a situation where suffering does not cause him to doubt God, his omniscience, or his providence.

So does the possibility of psychic immunity from suffering generate a first-personally adequate theodicy for the suffering of the innocent? To find out, we must look more closely at the details of how psychic immunity works.

The psychic immunity theodicy

The axiological shift

For Maimonides, gaining psychic immunity is clearly not just a matter of gaining another piece of empirical knowledge: what Job achieves by the end of the book is a certain kind of intellectual enlightenment or perfection. What form this intellectual enlightenment takes is a point of dispute in Maimonides scholarship, and I will conclude this section by discussing one major divide and its implications for the idea of psychic immunity – that between ‘dogmatic’ and ‘sceptical’ readers of Maimonides. When it comes to their understanding of how psychic immunity works, though, these readings have substantial common ground.

Maimonides believes that ‘providence is consequent on the intellect and attached to it’ (III.17, 474). He thinks that divine providence is a function of the divine intellect (what Maimonides calls the ‘Active Intellect’) ‘overflowing’ into the being that is providentially watched over. The individual ‘with which this intellectual overflow is united . . . is the one accompanied by divine providence, which appraises all its actions from the point of view of reward and punishment’ (III.17, 471–472). The talk of the ‘point of view of reward and punishment’ is just Maimonides reiterating his claim that providence is a matter of desert. The language of ‘overflow’ is meant to capture the explanatory relations at play here: being watched over by providence is a function of the perfecting of the human intellect, which is itself explained by the power of the Active Intellect. This is something that admits of degrees: ‘everyone with whom something of this overflow is united, will be reached by providence to the extent to which he is reached by the intellect’ (III.17, 474). Thus divine providence will more greatly protect the enlightened: ‘when any human individual has obtained, because of the disposition of his matter and his training, a greater portion of this overflow than others, providence will of necessity watch more carefully over him than over others’ (III.18, 474–475).

This helps us better understand the claim that God's providence protects people according to their deserts. Specifically, it tells us what the relevant sort of desert is, in proportion to which providence protects a person. Maimonides, in his discussion of four different forms of human wisdom at the end of the Guide, writes that the highest form of human perfection is not ‘the perfection of the moral virtues’, which is ‘preparation for something else and not an end in itself’. Rather intellectual perfection is the ‘ultimate end’, the ‘true perfection’. This perfection is a matter of ‘the acquisition of the rational virtues – I refer to the conception of intelligibles, which teach true opinions concerning the divine things’ (III.54, 635).¹⁵ This point is significant to the way Maimonides thinks about innocence: a morally virtuous person who is not intellectually perfected is not really living righteously. Only the intellectually perfected person is (III.11, 440).

How then does intellectual perfection and the consequent protection of divine providence free a person from suffering? Here I follow David Shatz: Maimonides seems to think it's a matter of a certain kind of attention or focus on God and the divine things, a focus that takes a person out of the tendency to focus on his own material well-being or that of the human species.¹⁶ ‘If a man's thought is free from distraction, if he apprehends Him, may He be exalted, in the right way and rejoices in what he apprehends, that individual can never be afflicted with evil of any kind’ (III.51, 625). This state of undistracted contemplation of God is to be cultivated through training of the mind, prayer, and study. It may even be possible to reach a state where one is always contemplating God in one's heart, even as one goes about one's life – Maimonides suggests that Moses and the Patriarchs may have been in such a state (III.51, 623). As to how this state of contemplation is supposed to protect against suffering, there is some interpretative work to do: Maimonides sometimes sounds as though he is taking the position that the previous section said he rejected, that God will literally prevent bodily evils from happening to a person in this state of enlightenment (cf. III.51, 624–626).¹⁷ I maintain though that this is not the best way to read Maimonides. Rather he thinks that bodily harms or evils will not be perceived or judged to be suffering by the intellectually perfect. He writes:

Yet in the measure in which the faculties of the body are weakened and the fire of the desires is quenched, the intellect is strengthened, its lights achieve a wider extension, its apprehension is purified, and it rejoices in what it apprehends. The result is that when a perfect man is stricken with years and approaches death, this apprehension increases very powerfully, joy over this apprehension and a great love for the object of apprehension become stronger, until the soul is separated from the body at that moment in this state of pleasure. (III.51, 627)

Here the intellectually perfect person does not seem to be free of bodily ills but rather to transcend them, psychologically escaping the consequences of embodied being by perfecting their intellect. Maimonides’ description of Job's perplexity reinforces the point:

But when he knew God with a certain knowledge, he admitted that true happiness, which is the knowledge of the deity, is guaranteed to all who know Him and that a human being cannot be troubled in it by any of all the misfortunes in question. While he had known God only through the traditional stories and not by the way of speculation, Job had imagined that the things thought to be happiness, such as health, wealth, and children, are the ultimate goal. For this reason he fell into such perplexity and said such things as he did. (III.23, 492–493; bold emphasis mine)

This passage further explains the sense in which God's providence protects the righteous from suffering. Intellectual perfection, not any kind of worldly doing well, is true happiness, and worldly ills are powerless to take it away. When a person comes to know God as perfectly as possible, they will reject egocentrism and anthropocentrism and stop believing that either themselves or the human species is the most important thing in the universe (III.12, 211; III.13, 491). But precisely by ceasing to worry about their own flourishing, and adopting a perspective closer (in certain limited respects) to God's, they will flourish. This is what Shatz calls Maimonides’ ‘axiological shift’ and by performing it the enlightened will achieve psychic immunity from suffering.¹⁸

Here is how that idea seems to do theodic work for Maimonides. People who do not perform the axiological shift are not truly righteous. Rather they are still trapped within their egocentrism and anthropocentrism, ‘false pictures’ of what truly matters.¹⁹ They value and prize the wrong things. Any suffering they experience is fundamentally a result of this bad mental state that they are in, not the external evils that plague them. Once they come to understand what is truly valuable, they can undergo precisely the same bodily ills without truly suffering.

We now have a better picture of Maimonides’ ‘psychic immunity’ theodicy, and how it might provide a first-personally adequate answer to the questions of sufferers like Job. The next section will evaluate the prospects for this theodicy – or one along these lines – as a response to the modern problem of the suffering of the innocent outlined above.

Dogmatism and scepticism

First, however, I want briefly to introduce two different ways of interpreting Maimonides’ account of intellectual enlightenment, as the difference between the two will be relevant to the following section. Following Josef Stern, I will call the first sort of interpretation ‘dogmatic’ and the second ‘sceptical’.²⁰ Dogmatic interpreters argue that metaphysical knowledge – what Maimonides thinks of as the most noble or perfect kind of knowledge – is possible for human beings. The subjects of metaphysical knowledge include God and the fundamental metaphysical principles of reality.²¹ Intellectual enlightenment consists in the achievement of such knowledge, which allows a person to perform the axiological shift, reorienting their values and providing them with psychic immunity. A person with metaphysical knowledge knows the reasons why people suffer, because they know all the most important and fundamental truths about reality.²²

Sceptical readers, by contrast, do not think that Maimonides believes metaphysical knowledge is really humanly possible. Instead, the most a human intellect can achieve is the complete knowledge of natural physics and (as part of one's knowledge of oneself as a hylomorphic substance in the natural world), a recognition of its own inability truly to know God or metaphysical principles. It's still the case that the fully enlightened person's values and priorities change such that they can perform the axiological shift. But the sceptical version of the shift does not result in complete understanding of the reasons for suffering, but rather a sort of peace of mind that results from surrendering both the drive to achieve this and the consequent anxiety of attempting to satisfy unsatisfiable epistemic desires. The enlightened sceptic knows that he lacks no knowledge that is possible for him: his intellect is now in a state that is, in Stern's words, ‘perfected if not perfect’.²³ He can take solace in the fact that God has reasons for allowing the existence of suffering, in whatever sense God ‘has reasons’ – though he does not know what these reasons are or what the relevant sense is (III.19, 308–310, III.20–21). This state is the source of ‘psychic immunity’ for the sceptic. The enlightened person can still reject anthropocentrism and egocentrism and achieve a more accurate picture of reality, but their rejection is not based on the fully explicit and worked out discursive propositions about God and metaphysics that one would find in a science. Importantly, the resulting ‘immunity’ is substantially weaker: the person does not escape suffering but finds a way to ‘bear it lightly’ by humbly and happily accepting the limits of possible understanding.²⁴

The challenges to the psychic immunity theodicy that the next section raises apply to both dogmatic and sceptical interpretations of the Guide. But the sceptical version may be better equipped to respond to them.

The problem of inculpably incomprehensible suffering

What, then, of those challenges? It seems that the theodicy faces many. Let's recall our standard or modern problem of the suffering of the innocent:

1. (1) God is omniscient, omnipotent, and perfectly good.

2. (2) If God (as described in 1) existed, God would not allow innocent people to suffer terribly.

3. (3) However, innocent people do suffer terribly.

4. (4) Therefore God (as described in 1) does not exist.

I argued earlier that Maimonides’ response can be read as putting pressure on the third premise by denying that there are innocent sufferers. Does Maimonides’ argument succeed in doing so in a first-personally adequate way? I wrote above that to read Maimonides as claiming that only moral wrongdoers are vulnerable to physical sources of suffering is to read him as endorsing an obviously false claim. It is clearly empirically false that people face disaster, illness, or attack in proportion to their moral wrongdoing. The psychic immunity theodicy makes a different claim: that while anyone could find themselves in a situation in which they have ‘no defence’ against bodily harm, the enlightened or truly righteous will be able to either (on the dogmatic view) entirely avoid experiencing these harms as suffering, or (on the sceptical view) tranquilly endure sufferings in a way that doesn't threaten their faith in God. This second claim is the more plausible of the two. Some people do seem to have achieved a state resembling Maimonidean psychic immunity, bearing even terrible trials lightly. The Catholic Priest Maximilian Kolbe led other victims in prayers and hymns as he starved to death in a Nazi concentration camp, having taken the place of another prisoner. Survivors report that he remained completely calm and tranquil as he wasted away and was eventually murdered by lethal injection.²⁵ The Tibetan Buddhist Lama Phakyab Rinpoche was imprisoned and brutally tortured by the Chinese Communist Party for three months in 2003. He sustained serious injuries and nearly died of gangrene in his leg. As he tells it, he felt more pity and compassion for his torturers during this ordeal than he did for himself, describing the idea that he would hate them as ‘absurd’. ‘I . . . feel gratitude toward those who tortured me’, he writes. ‘They taught me patience, unconditional compassion, and impartiality, more than have any of my masters. Every day, I express my wishes for them and offer them my prayers so that they may free themselves from mental states upset by hatred and anger.’²⁶

So something like psychic immunity seems to be possible even in the worst circumstances. The people who exhibit it are certainly extraordinarily righteous and admirable. What's more, this kind of psychological state seems to have a kind of first-personal theodic adequacy built into it. Kolbe's horrific circumstances did not threaten his faith in God. One could perhaps push the point further. If a psychically immune person knows that anyone who truly deserves to will be able to escape or peacefully endure terrible suffering, this fact might be in itself a source of consolation. The possibility of psychic immunity for each other suffering person would contribute to one's own psychic immunity. Thus, if the potential scope of psychic immunity were universal, there would exist the sort of connection between the first-personal and third-personal aspects of this theodicy that could satisfy a contemporary philosopher.

Even so, we can immediately raise a serious question: even if psychic immunity is achievable, does it follow that people who do not achieve it are not truly innocent sufferers? For Maimonides the answer is yes. If someone chooses to pursue material or bodily satisfaction rather than focusing on God and on the perfection of her intellect and suffers for lack of bodily satisfaction, then she is blameworthy for her choices and their consequences (Guide III.12, 445–447). A number of objections could be raised to this, but I will focus on one. This argument about blameworthiness assumes that people who do not achieve psychic immunity are choosing not to: they have an opportunity that they voluntarily forgo. The problem is that this does not seem to be the case for some paradigmatic cases of innocence. Infants and the radically cognitively impaired seem to lack the ability to develop psychic immunity through no fault of their own.²⁷

Maimonides himself is firmly of the view that a certain amount of natural ability is absolutely necessary for the achievement of intellectual perfection. He writes that ‘There are . . . many people who have received from their first natural disposition a complexion of temperament [or a temperamental disposition] with which perfection is in no way compatible’ (I.34, 77). A man who is naturally too hot-tempered cannot be expected to refrain from anger ‘even if he subject his soul to very stringent training’ (I.34, 77). He will thus be incapable of achieving the kind of calm and detachment necessary for intellectual perfection. Maimonides goes on to list various other congenital problems of temperament which make intellectual perfection impossible for certain people, and writes that ‘to make an effort for their benefit in this matter is pure ignorance on the part of him who makes the effort’ (I.34, 77).

Lest we think that Maimonides believes that differences in temperament are the only ones which might make intellectual perfection impossible, he also clearly states that differences in raw intellectual ability can too. Discussing the fact that the differences in human capacity when it comes to the senses and other bodily faculties is widely acknowledged, he writes that ‘The identical rule obtains with regard to human intellectual apprehensions’ (I.31, 65). Maimonides continues:

There are great differences in capacity between the individuals of the species . . . It may thus happen that whereas one individual discovers a certain notion by himself through his speculation, another individual is not able ever to understand that notion. Even if it were explained to him for a very long time by means of every sort of expression and parable, his mind would not penetrate to it in any way . . . (I.31, 65–66)²⁸

Given this, it seems clear that Maimonides’ theodicy will not in fact have the universal character that I suggested above that it could have – at least not according to him. If this is so, the psychic immunity theodicy will look much less appealing to modern readers, because it will fail the test of first-person adequacy for anyone who is incapable of performing the axiological shift.²⁹

One potential response is to excise Maimonides’ elitism from the theodicy and proceed with the rest of it.³⁰ The problem with this is that Maimonides is at least partially correct. It is an inescapable fact that humans have differences in intellectual ability. In particular, infants and people with radical cognitive disabilities (like adult survivors of the genetic condition Trisomy 18 or Edwards' Syndrome) seem to lack the cognitive capacities necessary to achieve the kind of enlightenment about which Maimonides writes. Whether one is a dogmatic or a sceptical reader of Maimonides, there is no doubt that achieving the ‘perfected’ state of the intellect requires a high level of intellectual sophistication. And this seems to chime with our actual experiences. Cases of people enduring terrible physical suffering with anything approaching full or perfect psychic immunity are extremely rare. Putative cases almost invariably involve people who have spent a lot of time in the dedicated practice of a particular religious or spiritual tradition, people able to situate and contextualize their own pain in a broader narrative. Maximilien Kolbe was a Catholic priest who believed in the redemptive power of suffering united to the cross of Christ. Phakyab Rinpoche wrote that when he was being tortured by the Chinese regime he perceived that:

What was happening to me was only the result, the consequence, of a negative spirit and negative thoughts that in previous lives had led me to injure and cause pain to other beings, both human and nonhuman. My torturers were not my enemies. The real enemy is not outside of us.³¹

This kind of sophisticated narrative-building could not be achieved by an infant or a person with a radical cognitive disability.³²

It could be objected that cognitive impairment sometimes itself grants something like psychic immunity from suffering. Depression and anxiety themselves require a certain degree of cognitive ability, and a young child or a radically cognitively disabled person might avoid them in situations where an adult without a cognitive disability would not. This would be very different from Maimonidean psychic immunity – almost its opposite – but might mean that the people unable to achieve the latter wouldn't need it to avoid suffering. This objection faces two serious problems. First, not all forms of terrible suffering depend on cognitive sophistication, severe physical pain being one obvious case. Second, if avoidance of suffering is a result of an inability to properly understand some situations, it could be argued that this inability is itself bad for the person. Certainly Maimonides would think so. Though such a situation would fall outside of the scope of the problem of suffering that I am addressing in this article, it's plausible that it would require theodic justification in a broader consideration of the problem of evil.³³

It seems then that Maimonidean psychic immunity cannot be achieved by some of the very people whose suffering seems most badly in need of justification. The uncomprehending, helpless suffering of a young child or a person with a cognitive disability strikes us as among the most terrible and pointless kinds of suffering. It also seems entirely wrong to blame these people for their lack of ability to achieve Maimonidean intellectual perfection: as Maimonides points out, children are simply not mentally developed enough to pursue it fruitfully (I.34, 77).³⁴ It can hardly be their fault if they encounter terrible suffering before reaching maturity. Similarly, no-one chooses to have a radical cognitive disability. We seem to be dealing with inculpably incomprehensible suffering.

This seems to be a serious deficiency of the psychic immunity theodicy as a response to the problem of the suffering of the innocent. What can be said in its defence? I will consider two potential avenues: to bite the bullet and accept Maimonides’ intellectual elitism, or to accept a more limited role for the theodicy. The second option is, I think, more compelling than the first, though it sits uneasily with a dogmatic interpretation of the Guide.

Responses

Modern intellectual elitism

There are modern writers whose views on moral status give us reasons to think that the suffering of infants or the cognitively disabled is not as morally significant as we might believe. Jeff McMahan and Peter Singer argue that humans who do not possess ‘sophisticated cognitive capacities’ ought not to be classified as ‘persons’ with the ‘full moral status’ possessed by cognitively normal adults. This has led Singer to write that ‘killing a newborn baby is never equivalent to killing a person, that is, a being who wants to go on living’ and to argue that infanticide is primarily a wrong against a child's parents and is sometimes permissible.³⁵ McMahan has stated that ‘If we think that a person is a higher form of being than a dog, it is difficult to avoid the conclusion that a person is a higher form of being than a human being with psychological capacities and potential comparable to those of a dog.’³⁶ These views have some commonalities with some of Maimonides’ claims – though whether Maimonides meant them sincerely or ironically is disputed.³⁷ For instance, he writes: ‘as for the ignorant and disobedient, their state is despicable proportionately to their lack of [the divine] overflow, and they have been relegated to the rank of the individuals of all the other species of animals . . . For this reason it is a light thing to kill them’ (III.18, 47). The reference to the divine overflow is explicitly meant to contrast such people with the intellectually enlightened who partake in it.

One option, then, for rescuing Maimonides’ psychic immunity theodicy from the objection raised in the previous section would be simply to bite the bullet. One could argue that its silence about the suffering of infants and the cognitively disabled is a virtue rather than a vice – or at least not a flaw. On this view what we ought to be most worried about is not ‘human suffering’ but ‘the suffering of rational agents’. The psychic immunity theodicy might only work for addressing the suffering of some humans, but those are the only really morally important cases of suffering anyway. In other words, we could reduce the number of individuals for whom a theodicy would have to be first-personally adequate by reducing the number of individuals who count as persons.

This defence of the theodicy suffers from serious flaws. First, ‘unpersoning’ some of the most vulnerable human beings is a very large bullet to bite. If one is independently persuaded by Singer and McMahan's arguments about moral status this will of course not be a concern. But many of us are not so persuaded. Second, neither Singer nor McMahan think that the suffering of infants or of the cognitively disabled is deserved, nor that it is morally irrelevant. They are both consequentialists who think we should aim to reduce the suffering of all sentient creatures. Here, a defence of the Maimonidean theodicy would need a compelling argument that combined Singer and McMahan's ‘unpersoning’ of infants and the cognitively disabled with reasons for indifference to the individual sufferings of non-persons.

This would be far too high a price to pay for the sake of saving the psychic immunity theodicy. This approach would require abandoning not just the particularly modern commitments previously discussed, but the standard ethical teachings of the Abrahamic faiths on human dignity, caring for the weak, and relieving suffering.

Epistemic humility

Another response to the objection from inculpably incomprehensible suffering would be to accept that some people cannot achieve psychic immunity, and that the psychic immunity theodicy thus cannot act as a complete explanation or way of making sense of innocent suffering that is first-personally adequate for all innocent sufferers. But the theodicy would show that God has given at least some people an opportunity to escape or reduce the severity of terrible suffering.³⁸ Achieving psychic immunity could still be worthwhile for those who can achieve it. Other justifications would still be required for the suffering of infants and people with cognitive disabilities, and psychic immunity would function not as a complete theodicy but rather as a theodic strategy, doing some work as part of a broader theodicy. This move seems much more promising a response to the objection from inculpably incomprehensible suffering than the attempt to embrace Maimonides’ elitism. However, it's hard to reconcile with dogmatic interpretations of the Guide. On dogmatic readings, an intellectually perfect human being ought to have a complete third-personal understanding of how each case of suffering can be explained either by natural science or by metaphysics. Thus, if the psychic immunity theodicy only made sense of some suffering, a dogmatic reader would need a different theodic strategy to account for the suffering of people who are not covered by it. Without this full understanding or knowledge, they won't achieve the requisite intellectual perfection for their own first-person psychic immunity. The success of the dogmatic version of the psychic immunity strategy would be entirely dependent on that of another theodicy. This would make the role of the psychic immunity strategy very limited indeed.

Sceptical readings of the Guide fit much better with a limited version of the psychic immunity theodic strategy. The kind of peace or tranquillity that creates psychic immunity on the sceptical reading is, in part, an awareness of the limits of one's own knowledge – one that is compatible with trust in God. As such one could achieve it without having a full understanding of the justifications of all suffering, and the theodic strategy could still do some work without requiring that the other theodic strategies be in place. This version of the theodic strategy could be useful even if one thought that the justifications for some people's suffering was beyond the capabilities of the human intellect. The sceptic might be able to achieve a state of tranquillity in part by coming to believe that God has sufficient reasons for allowing suffering, in whatever equivocal sense God ‘has reasons’. In this way the limited version of the Maimonidean psychic immunity theodicy could provide an interesting complement to modern ‘sceptical theist’ arguments.³⁹ The sceptic, after all, does not know the reasons for their own suffering, and so the fact that they do not know the reasons for the suffering of infants or people with radical cognitive disabilities will not jeopardize the tranquillity of their psychic immunity in the way it would for dogmatists. The first-person adequacy of the theodicy for the enlightened person would not be undermined by its lack of first-person adequacy for people who are unable to achieve enlightenment. Much more would need to be said about this, but this version of the theodicy seems to me to be the most promising candidate for surviving the objection from inculpably incomprehensible suffering.

The limits of first-person adequacy

To bring my argument to an end, I want to return to the difference between first- and third-person theodicies. I have argued that the Maimonidean psychic immunity theodicy cannot achieve first-personal adequacy for all innocent sufferers, because it cannot do so for infants and people with serious cognitive disabilities. Despite the many peculiarities of Maimonides’ theodicy, however, this problem is not unique to it. Compare another medieval theodicy, that of Thomas Aquinas. Eleonore Stump interprets Aquinas' theodicy as arguing that ‘warding off a greater harm for a person or providing a greater good for her justifies God in allowing suffering’ for people without and with Christian faith, respectively.⁴⁰ Stump argues that Aquinas' theodicy is committed also to a constraint: if ‘a good God allows suffering, it has to be for the sake of a benefit that outweighs the suffering, and that benefit has to be one that, in the circumstances, cannot be gotten just as well without the suffering; the benefit has to defeat the suffering’.⁴¹ This is not the same as my idea of first-person adequacy. It's about objective benefits rather than a sufferer having a satisfying explanation for – or way of making sense of – their suffering. (There is, however, a close connection between the two ideas. If on a given theodicy someone didn't receive any benefit from their suffering this would plausibly make the theodicy first-personally inadequate for them.)

Stump writes that Aquinas pays ‘little to no attention’ to the suffering of human beings who are ‘not mentally fully functioning adults’, though she thinks a broadly Thomistic theodicy could cover their suffering.⁴² But notice that however well Aquinas's theodicy meets Stump's constraint in the case of the suffering of infant children and people with radical cognitive disabilities, it cannot do so by being first-personally adequate for them, at least not in this life. Not, at any rate, in all cases of such suffering. It's possible that a young child could suffer terribly before attaining adult mental functioning, and then later come to have a first-personally adequate answer to the question ‘why did I suffer as a child?’ But this will not be possible for children who suffer and die in infancy, or for radically cognitively disabled people who never achieve a level of cognitive ability which would allow them to understand the reasons for their suffering. This point generalizes: no theodicy can achieve first-person adequacy in the cases of inculpably incomprehensible suffering that I have discussed, because the very conditions that make the suffering inculpably incomprehensible make this impossible. These cases – some of the cases of suffering most in need of theodic justification or explanation – cannot be justified or explained in a way that is first-personally adequate for the sufferers while they are still alive.

What should we make of this? If it is correct, it implies that any successful theodicy will need some kind of ‘merely’ third-person component that explains, justifies, or makes intelligible the suffering of some people in a way that they themselves cannot access in this life. I think this should make contemporary philosophers of religion and theologians more sympathetic to the medieval approach to theodicy with its greater emphasis on third-person accounts. If there is to be any justification for the suffering of people who cannot understand the reasons for their own suffering, it will have to take a third-person form. If the sceptical readers of Maimonides are right, then infants and the cognitively disabled are not the only people in such a predicament.