Here is the draft of my paper which I read at Rethinking Hell Conference 2015 in Pasadena this past weekend.
Christian Apokatastasis contra Crisp and Walls
Steven Nemes, Fuller Theological Seminary Arizona
Athanasius once argued that it would be unfitting and unworthy of God’s goodness that he permit human beings created in his image to be destroyed, whether deservedly or because of the deception of demons (On the Incarnation 6). Such thinking motivates a doctrine of universal salvation while challenging traditionalist or conditionalist alternatives. Now theologians of various stripes might respond in different ways to the Athanasian challenge. Oliver Crisp and Jerry Walls are two renowned contemporary scholars who have offered cogent and impressive defenses of their traditionalist conviction drawing from two very different theological traditions, though their arguments could just as well be accommodated to a conditionalist framework. Crisp argues against universalism and for particularism on broadly Augustinian grounds, appealing to the nature of divine justice and goodness. Walls argues against universalism and for particularism on broadly Wesleyan grounds, insisting on the inviolability of human freedom in salvation. But in addition to these traditions, there is also something of a classical universalist tradition, with its own unique understanding of the relevant theological, metaphysical, anthropological and axiological questions. Ilaria Ramelli has described the development of this tradition through the first millennium of the church in great detail in her recent tome, The Christian Doctrine of Apokatastasis (Brill, 2013). Drawing from her work, my intention is to elaborate upon this tradition and to support it with varied arguments when it contradicts those aspects of Augustinian or Wesleyan tradition motivating the objections of Crisp and Walls. In this way, I want to neutralize potential conditionalist adaptations of their arguments and in general to uphold the superiority of the universalist tradition relative to its competitors.
Crisp and Walls: two objections
Some time ago, Oliver Crisp (2003) argued provocatively that an Augustinian could affirm the doctrine of universalism in good faith. Now essential to the Augustinian theological framework are “the doctrines of election and theological determinism” (2003a, 127). Thus if a human being is to be saved at all, she must be elected by God for salvation. In fact every happening in history has been antecedently preordained by God, including (in Calvin’s words) “what [God] willed to become of each man” (in 2003a, 128). Furthermore, Augustinians hold that God must show his justice as well as his mercy, the whole divine character, in the created order. As Jonathan Edwards said, God is a communicative being who communicates his whole self to others to be understood and enjoyed by them for his greater glory (in Crisp 2014, 125; cf. 2003a, 130-1). Typically, then, Augustinians would appeal to this distinction between God’s justice and mercy, together with the principles of election and determinism, to justify their particularist conception of hell: “the elect” are the objects of God’s mercy, and they are saved; “the reprobate” are the objects of God’s justice, and they are rightly damned for eternity (2003a, 130; 2014, 104-5).
Crisp’s suggestion was that God can accomplish the expression of both his mercy and his justice without damning everyone. Crisp enumerates the three salient principles of Augustinian soteriology to show their compatibility with universalism:
(a) The need for infinite retributive punishment to be meted out for the sin of those human agents who are fallen.
(b) The need for sin to be atoned for (in Christ’s death on the cross) for sinful human agents to be counted among the elect.
(c) The need for the display of both God’s grace and mercy and his wrath and justice in his created order (2003a, 134; 2010, 5; 2014, 111).
Universalism is compatible with these stipulations. For God’s justice could be satisfied by the atonement of all human sins through Christ’s self-sacrifice. Thus all human persons can be numbered among the elect, since Christ died for all. Moreover, there is no compromise of either divine mercy or justice: mercy is shown to all humans by their election and Christ’s death on their behalf; justice is shown in the punishment of sin through the crucifixion of Jesus (2003a, 135-7; 2010, 6; 2014, 113-6). But now this generates a problem of evil for the Augustinian: for if God could have saved all human persons while not compromising either his mercy or justice, why wouldn’t he have done so, rather than mercilessly damn some persons?
Crisp’s response to this problem constitutes his objection to universalism. He attempts to demonstrate the conceptual possibility of Augustinian particularism in the face of the universalist challenge, but his suggestions actually suggest more than that. He begins his argument by considering some further theological premises palatable to Augustinians. First, following Aquinas and Swinburne, there is no best possible world but rather an infinity of possible worlds, all compatible with God’s essential goodness (2010, 15-6; 2014, 145-6). Second, he affirms that “divine grace is necessarily not part and parcel of divine distributive justice” (2010, 16). God can act graciously if he chooses, but he is never obligated to be gracious since “it is of the nature of an act of grace that it is free and not obligatory” (2010, 17). On the other hand, “divine justice is inexorable and must be satisfied on an Augustinian way of thinking” (ibid.). Finally, Crisp proposes that God has no obligations whatsoever unless he places himself under one (ibid.).
With these in mind, Crisp considers different worlds God might create. Imagine a world in which humanity sins and God justly damns everyone. Crisp insists that such a world is compatible with divine goodness (2010, 17) and benevolence (2014, 147) because salvation is a matter of grace, and thus God is not flouting any obligations in failing to save anyone. But this world is not desirable for the Augustinian because God’s mercy and goodness are not displayed in salvation. Suppose instead that God creates a world in which some persons are saved from their sins while others are justly damned. Crisp asks: Need God do more than this? The answer is ‘No,’ because God isn’t obligated to do more than what is consistent with his nature (2010, 18; 2014, 148). Rather, “All that God has to do is act in accordance with the divine nature” (2014, 149). Thus, Augustinian particularism is preserved as a conceptual possibility alongside Augustinian universalism.
Crisp considers the objection that God might be obligated to create a world where everyone is saved if possible. His response, drawing from Robert Adams, is effectively that God does not have any obligations towards the uncreated that do not yet exist (2010, 20; 2014, 149). Suppose God elects to create a world in which some are saved and others damned, and that the following conditions are satisfied in that world:
(a) None of the individual creatures in it would exist in an Augustinian universalist world.
(b) None of the creatures in it has a life that is so miserable on the whole that it would be better for that creature if it had never existed.
(c) Every individual creature in the world is at least as happy, on the whole, as it would have been in any other possible world in which it could have existed (2014, 149-50).
Crisp insists that in actualizing such a world, God does not wrong any of the creatures within it (2014, 150), though he admits that satisfying these conditions may require a revision of the traditional doctrine of hell along annihilationist lines (2010, 21-2).
Towards the end of his 2010 paper, he suggests further that the Augustinian might justify God’s choosing to create a world in which only some are saved by appeal to divine justice. He suggests a modification of the principle that God must display both divine justice and divine mercy in his created order along these lines:
(c*) The need for the display of both God’s grace and mercy and his wrath and justice in his created order for some number of deserving humanity (2010, 22).
Crisp considers this “a clarification of the nature of distributive justice on the Augustinian way of thinking” (2010, 23 n. 16), since God’s justice is fundamentally connected to desert (2010, 22). Because Christ’s punishment is not deserved but rather suffered vicariously, it would not perfectly satisfy the revised strict justice condition here proposed. Consequently, justice must be visited upon some deserving human persons, and this means their eternal damnation.
This latter line of argument actually rules out the possibility of universalism altogether. This is because, as Crisp says, “any world that God creates that includes creatures that are sinners must have provision for the display of divine justice to some number of deserving humanity” (2010, 23). Thus Crisp proffers two powerful arguments against universalism which must be addressed.
Now whereas Oliver Crisp argues against universalism “from above,” appealing to God’s justice, Jerry Walls argues in large part “from below,” insisting on the inviolability of human freedom. His argument against dogmatic universalism is formed in critical dialog with the universalist philosopher of religion, Thomas Talbott. The latter had argued that choosing eternal damnation was an incoherent notion. For a person to choose eternal damnation freely, she would have to choose damnation in conditions of absolute clarity of vision, knowing that God is the source of all happiness and sin is the source of all misery (Walls 1992, 125). Talbott grants that we can choose life without God for a while, but eventually the inevitable misery of sin would convince us that we have made a huge mistake (ibid.). Moreover, the sufferings in hell are forcibly imposed punishments of God which lead to unbearable suffering (Walls 2015, 78). There could be no intelligible motive for remaining in hell under such conditions, and therefore no one will voluntarily remain damned forever.
Talbott’s argument is initially plausible, but Walls’s response is strong and worth considering. Walls first specifies what it could mean to choose damnation, and second provides the motive for which a person could plausibly choose damnation. For Walls, the choice of damnation does not occur in conditions of absolute clarity of vision, as Talbott conceived. Walls agrees with Talbott that a person with such utter clarity of vision about God and sin would have no motive for continuing in sin, and thus could not choose it (1992, 133; 2004, 206, 210). After all, happiness, as good in itself, is the ultimate goal of all human agency (2015, 20-1). But he insists that such absolute clarity of vision cannot be acquired apart from cooperation with the grace of God (1992, 130-1; 2004, 206-7). Because sinners in hell will not have lived such a life, consequently they will never have the sort of clarity of vision which makes a choice for damnation impossible. Rather, their choice of hell comes as a result of various decisions made throughout the course of their lives, pursuing evil rather than good. Following Kierkegaard, Walls suggests that the choice of evil becomes a principle of consistency for a person’s character over time (1992, 120). Agents can even actively resist the temptation to do what is good should it appear within them, just as Goebbels, feeling pity for the devastated Poles he saw in a German newsreel, told himself, “Be hard, my heart, be hard” (1992, 121). The choice of evil becomes a unifying principle for the sinner’s character, around which she organizes and structures her life. It becomes a part of the sinner’s identity, so that the possibility of change seems all but lost.
So the choice of eternal damnation does not occur at single point in time. In a way, it would seem that persons choose damnation over the course of a lifetime as they give in to evil temptations. They choose wrongly over and over again until it seems all but impossible that they change. And if given the opportunity in the afterlife to leave hell, such a person would plausibly reject it, just as C.S. Lewis envisions in The Great Divorce. But why would they reject it? Certainly the experience of damnation is not pleasant and enjoyable as is the experience of the redeemed in God’s kingdom. Here Walls provides an answer to the second crucial question regarding the motivation for remaining in the torments of hell.
He writes: “if the choice of hell is an intelligible one, there must be something about the subjective experience of choosing evil which can account for why some may prefer it to goodness” (1992, 126). The choice of eternal damnation “is possible because hell can somehow be judged better than heaven, just as evil can be seen as a good to be desired” (1992, 125). The damned see something preferable about the infernal life. Walls gives an example drawn from Kierkegaard:
a person who has suffered some form of earthly distress and comes to despair that he will ever be relieved of it. His anger at what he has had to undergo turns into resentment at the whole of existence. He imbibes this so deeply that he reaches a point where he is no longer willing to be relieved of his distress. Indeed, it is now a part of his very identity. He considers himself a proof against the goodness of existence and he must hold out against repentance in order to maintain his protest (Walls 1992, 127).
Even if offered assistance by God and all the angels of heaven, such a person would arguably reject it. Walls notes, “Even if this person is not in bondage to sin in such a way that he cannot repent, we can still see how he may never want to. For his resentment seems to perpetuate itself” (ibid.). Walls moreover insists that sinners in hell, even if they come to some sort of knowledge about God and about their condition, can deceive themselves (1992, 129). They trick themselves into thinking that it is “better to reign in Hell than to serve in Heaven,” like Milton’s Satan.
Now Talbott insisted that the misery of life apart from God would eventually convince the sinner that she has chosen wrongly and her illusions would be shattered. After all, God is punishing the sinners in hell for what they’ve done, and their sufferings are unbearable. Furthermore, God could provide the damned with compelling evidence that they are mistaken, just as Saul’s vision of Jesus on the road to Damascus compelled him to change. But here Walls demurs. He poses a dilemma. Repentance, for Walls, cannot be forced or compelled; it must come as a result of a genuine change of heart (1992, 78). If God is punishing sinners in hell forcibly to the point that their suffering is unbearable, presumably they would repent, but their repentance would not be genuine. Rather God would be acting quite objectionably, forcing conversions by the sword. On the other hand, if the sufferings of the damned are not unbearable, there is no reason to think they could not remain in hell in spite of them. Even if their sufferings are quite painful, and even if God intends to use them to move sinners to repentance, nevertheless these punishments might simply harden sinners even more. Moreover, “the more we harden our heart and dull our conscience, the more we will form a character that is comfortable with sin… the hardness of heart would make the misery [of hell] more tolerable” (2015, 81).
So Walls concludes: “As I see it then, hell is indeed a place of misery but not unbearable misery. This is why it can be freely chosen forever as one’s eternal destiny” (2004, 212; 2015, 84). The choice of sin hardens the sinner’s heart, stabilizing her identity as a sinner, and this hardheartedness, alongside the irrational preference for some petty evil instead of true good, makes the damned existence of the sinner in hell sufficiently tolerable that she may never freely choose to leave. Even if she knows some things about God and sin, she may nevertheless reject him because the absolute clarity of vision that makes choosing sin impossible can only come in perfection, after a life lived in cooperation with God’s grace. Rather, she sees something preferable about life in hell—perhaps her self-righteous protest at the offer of God’s grace, perhaps a prideful insistence that she live her own life without need of anyone else—and thus freely chooses to remain there, despite the offer of salvation. She pursues happiness the same as everyone else, but in the deformity of her character, she satisfies herself with the shadowy pseudo-happiness of hell (2015, 89-90).
We have seen, then, that Crisp and Walls offer powerful arguments against universalism, but from theological traditions with seemingly opposite emphases. For Crisp the Augustinian, the inexorable demands of divine justice require that some number of human persons receive justly eternal damnation for their sins. It may even be impossible in principle that God create a world in which everyone is saved. In any case, divine justice is inexorable and divine goodness is not compromised by the sinner’s damnation. Meanwhile Walls the Wesleyan upholds the inviolable power of human persons to choose between alternatives: a life with God in righteousness, or self-damnation apart from God in sin. God in his goodness wants all to come to salvation, but he calls them to come freely, on their own, and so “God may not be able to save some people, even though he is willing to do so and is all-powerful” (Walls 2015, 70).
So Crisp the Augustinian responds to the Athanasian challenge by questioning the implicit understanding of God’s goodness, whereas Walls the Wesleyan questions God’s power to ensure the salvation of all. Now in addition to the Augustinian and Wesleyan traditions, there is also something of a classical universalist tradition. Ilaria Ramelli’s recent tome, The Christian Doctrine of Apokatastasis (Brill, 2013), follows the development of this tradition during the first millennium or so of Christian history. Her findings give excellent reason to suppose that universalism, as much as its competitors, deserves to be named a “traditional” Christian doctrine. Its patristic defenders included “Bardaisan, Clement, Origen, Didymus, St. Anthony, St. Pamphilus Martyr, Methodius, St. Macrina, St. Gregory of Nyssa (and probably the two other Cappadocians), St. Evagrius Ponticus, Diodore of Tarsus, Theodore of Mopsuestia, St. John of Jerusalem, Rufinus, St. Jerome and St. Augustine (at least initially), Cassian, St. Isaac of Nineveh, St. John of Dalyatha, Ps. Dionysius the Areopagite, probably St. Maximus the Confessor, up to John the Scot Eriugena, and many others” (Ramelli 2013, 11). Drawing from Ramelli’s research, I will expound some elements of the classical universalist tradition, supporting them with varied arguments where they contradict those tenets of Augustinianism or Wesleyanism which motivate Crisp and Walls’s objections.
Beginning with Oliver Crisp’s arguments, I first address the notion of God’s goodness. Crisp’s understanding of God’s goodness is subject to serious objection. He affirms, on the one hand, that it would not be contrary to God’s goodness or benevolence if he were to damn everyone for their sins, because salvation is by grace and grace cannot be obligated. This connects goodness or benevolence with keeping obligations. On the other hand, he says that God has no obligations at all, except perhaps the pseudo-obligation to act consistently with his nature. Divine justice, however, is inexorable, and its demands cannot be set aside. Therefore divine goodness and benevolence are either nothing at all, or else reduce to God’s retributive justice. But goodness and benevolence are one thing, whereas retributive justice is entirely another. The latter can be defined as an attention to merit or desert in our dealings with others, but goodness and benevolence are the disposition to act for the benefit of the other person, especially when she is undeserving. This is plausibly enough the way the terms are used in various scriptural passages. Ps 136 connects divine goodness with love/mercy: “Give thanks to the LORD, for he is good [טוֹב, χρηστός], for his steadfast love [חֶסֶד, ἔλεος] endures forever” (v. 1 and passim). So does Ps 23: “Surely goodness and mercy shall follow me all the days of my life” (23.6). In Ps 25, the psalmist prays: “Do not remember the sins of my youth or my transgressions; according to your steadfast love remember me, for your goodness’ sake, O LORD!” (25.7). He goes on to say: “Good and upright is the LORD; therefore he instructs sinners in the way” (25.8). Ps 68 says: “In your goodness, O God, you provided for the needy” (68.10). Ps 86 says God is “good and forgiving, abounding in steadfast love for all who call on [him]” (86.5). Ps 145.8-9 connects divine goodness with graciousness, mercy, slowness to anger, steadfast love, and universal compassion. A final obvious example is Paul’s remark in Titus 3: “when the goodness and loving kindness of God our Savior appeared, he saved us … according to his mercy” (3.4-5). Now because goodness and benevolence are connected to mercy and compassion as dispositions to act for the benefit of the other, especially if she is undeserving, therefore it is very much contrary to God’s goodness not to save even a single person! These things are what motivate salvation.
The classical universalists understood God’s goodness in this way, too. Theophilus of Antioch wrote that God’s goodness was manifested in his preventing the sin of human beings from being eternal through the imposition of physical death and removal from Paradise and the tree of life (Ramelli 2013, 66). Bardaisan and Origen rejected the Marcionites’ hard distinction between God’s justice and goodness, insisting that both alike work for the good of the human being (111-2, 175). Clement of Alexandria held that it is always open to human beings to repent unto salvation, since God’s goodness is at work both now and in the next age (126), and that God “almost compels people to salvation, out of a superabundance of goodness” (129). For Origen, God’s goodness will be shown in the eventual and mysterious salvation of even the worst of sinners (177). Diodore of Tarsus writes that the punishments of the damned in Gehenna are diminished out of God’s goodness (523). Gregory of Nazianzus, speaking about the incarnation of God and the subsequent deification of humanity, writes: “What a richness of goodness!” (459). Gregory of Nyssa connects goodness with salvation when he writes, “God’s purest goodness will embrace every rational creature, and none of the beings that have come to existence thanks to God will fall out of the Kingdom of God” (412). Evagrius describes Christ as “the leaven of the divinity who, in its goodness, has hidden himself in the unleavened lump of humanity” so as to “raise the whole lump to all that God is” (479). These authors see goodness as a disposition to act for the benefit of the other, especially when undeserving, following biblical usage.
Now if God is essentially good, if it is a part of his existence as such to be good, then this will inform everything that he does. Crisp apparently denies this, maintaining instead that God’s essential goodness or benevolence entails merely that he does nothing morally bad (2014, 109-10). But if God’s goodness is his disposition to act for the other’s benefit, then badness would be to act for the other’s harm. Consider the analogous case of an agent who is essentially malevolent: either she would always act malevolently, or else perform a proximate good for the sake of an ultimate evil. So also mutatis mutandis in the case of God: if he is essentially benevolent, then he always acts for ultimate good, even if for the moment he harms.
On this point, the classical universalists differ with Crisp’s Augustinianism about divine providence. Clearly on Crisp’s Augustinian picture, God does not always act for the ultimate benefit of everyone. Sometimes he makes use of human persons in order to realize some other end at their extreme cost; for example, the damnation of reprobate persons serves for demonstrating God’s justice. On the contrary, the ancient universalists affirmed that God always acts for the good of every person in what he does, even in punishment. Clement writes that the God’s work is to save, and his punishments “save and educate” (126-7). Origen affirmed that punishment must ultimately be given for the sake of reforming the sinner and bringing her to salvation in order to be “worthy of the goodness of the God of the universe” (195). Following Origen and others, Gregory of Nazianzus later would interpret the imposition of death upon humans as a hidden grace, an act of divine φιλανθρωπία, because in this way human sin would not go on undyingly, saying, “I am persuaded this is the way in which God punishes” (457). Elsewhere he writes that God makes use of threats and beatings and the effects of his wrath as remedies for sin in his pedagogic work (452). Athanasius insisted that it would be unworthy of God’s goodness for human beings created in his image to be destroyed, whether deservedly or by the deception of demons (247); this implies that it would be unworthy of God’s goodness ever to intend the irreversible destruction or damnation of any human person. Eusebius called God a φιλάνθρωπος, the savior of the whole universe [ὁ τῶν ὅλων σωτήρ] who acted out of love for humanity (320). Isaac the Syrian wrote, “Among all his deeds, there is none that is not entirely dictated by mercy, love, and compassion. This is the beginning and the end of God’s attitude towards us” (762). Because God’s character is like this, he is not satisfied creating a person who will ultimately be lost forever.
In this matter, it seems to me the universalists have the exegetical advantage. I want to offer an argument for this conclusion by comparing the creation mythology of Genesis to the Babylonian Atrahasis myth. In the Atrahasis myth, the creation of humanity was a practical solution to an inter-pantheon struggle. There were two classes of gods: the Igigi, laborers who worked the earth; and the Annunaki, who mostly slept and partied. One day the Igigi tired of their toils and began rioting, disturbing the sleep of the Annunaki. Of course this problem had to be resolved as quickly as possible, so the solution was to create humanity to work the earth instead. The cry at the creation of man was: “Let man bear the load of the gods!” (Dalley 2000, 14). On this scheme of things, the gods have no fundamental commitment to humanity’s well being as such. Rather, the relationship is economic and pragmatic: human persons were created for the benefit of the gods, though the gods may act for humanity’s benefit if adequately incentivized. Genesis paints a very different picture. God creates the world freely, by the power of his word, without needing anything or expecting anything from his creatures. He makes each animal and life form in its proper place, commanding all only to go forth, multiply, and flourish in the world he has created for them. Human beings are made to reflect the divine goodness and benevolence in the creation, caring for the garden and for one another. And when God sees that it is not good for the human to be alone, he creates him a proper helper (Gen 2.18). In contrast to the Babylonian picture, God seems in everything to act for the sake of his creation out of his own goodness; he has no ulterior motives or goals. He is specifically not modeled on human leaders or benefactors who act quid pro quo; rather God creates out of the abundance of his generosity, for every creature’s benefit.
Consider also the following analogy from film. In Ridley Scott’s Blade Runner, the replicants rebel against the humans and the Tyrell Corporation because their situation is miserable: they are given false memories and identities; their life span is very short, only four years; they are not permitted to pursue what is good for them, but rather are created for the purpose of effective off-planet labor. Though they can and want to experience life in all its goodness, yet this is denied them by their creators. They are free agents, yet created for an end from which they cannot benefit. Blade Runner expects us to sympathize with the replicants’ cause and their injustice. To the extent that we do, we must also object to the goodness of God on Crisp’s Augustinian picture, in which the divinity more closely resembles Dr. Eldon Tyrell or the Babylonian Igigi than the LORD in Genesis’s opening chapters.
Now if this is the first picture we get of God when reading his scriptures, it isn’t unreasonable to interpret what follows in light of this beginning: as Isaac the Syrian says, even in places where at first glance it would seem God is acting otherwise, we should understand a hidden providence aimed at the good of all (cf. Second Part 39, 19). I will give some examples from scripture: Hannah’s infertility, which initially seemed God’s curse, was actually the providential means by which God would secure the birth of Samuel and his dedication to the LORD (1 Sam 1). Likewise, the crucifixion of Christ, who seemed stricken and afflicted by God (Isa 53.4), was actually the propitiation for the whole world’s sins (1 John 2.2). Though he was not himself a universalist, T.F. Torrance remarked that Christ’s crucifixion reveals something about God to us, namely that “God loves us more than he loves himself” (Torrance, Torrance, and Torrance 2010, 14). To my mind, the universalist conviction is ultimately grounded in the experience of the crucified God, accepting death for the sake of sinful humanity. If God “loves us more than he loves himself,” then he always acts for our good.
The Augustinian may question: where does God’s justice come in? What of the infinite punishment which sin deserves? Among the classical universalists, Isaac the Syrian denied that God has any concern for retribution at all. He objects strongly to the notion, calling it “abominable.” Rather, God makes use of apparent punishments as means of bringing salvation and repentance. Thus Isaac says: “This is how everything works with [God], even though things may seem otherwise to us: with Him it is not a matter of retribution, but He is always looking beyond to the advantage that will come from His dealings with humanity” (Second Part 39, 5). His principle evidence is the self-sacrifice of Christ on behalf of sinners, which is infinitely disproportionate to the deserts of wicked humans (39, 16). Other universalists affirmed the role of justice in the divine economy, but they did not consider that sin merited an infinite punishment, nor did they separate justice from a concern for the good of the sinner. Clement of Alexandria held that “everything, both in general and in the single cases, is ordered by the Lord of the universe for the purpose of universal salvation... the task of salvific justice is to lead each being to what is better” (Ramelli 2013, 124). Origen likewise did not conceive that God could punish for merely retributive purposes (153, n. 410), and that there was no irreconcilable difference of intention between God’s justice and goodness (175). This was essential to their anti-Marcionite rhetoric. Theodore of Mopsuestia and Diodore of Tarsus affirmed that sinners would be punished in Gehenna, but as they deserve, sometimes less, and not infinitely (523). Theodore appealed to Jesus’ words in the gospel of Luke that one disobedient servant who knew much would be beat with many stripes while another ignorant servant would be beaten with few, concluding that the punishments are limited and measured according to desert. And Diodore affirmed that God “chastises the wicked sparingly, … far less than they deserve, just as He [rewards] the good beyond the measure and period (of their deserts).” These figures would agree with Isaac’s sentiment that God’s concern is not justice strictly upheld: “God wishes for our salvation, and not for reasons to torment us” (Second Part 40, 12). They did not consider that sin against God deserved unending or infinite punishment, and God is not obliged to punish us as we deserve anyway. I think this notion is supported by Ps 103: “[God] will not always accuse, nor will he keep his anger forever. He does not deal with us according to our sins, nor repay us according to our iniquities… For he knows how we were made; he remembers that we are dust” (vv. 9-10, 14). This psalm suggests that God does not repay us strictly according to desert because he knows our weaknesses. Thus, contrary to the Augustinian argument, not only the dignity or value of the offended party but also the condition of the offender must also be considered in judging guilt.
There is one more thing to be said on this topic. I’ve defined retributive justice as an attention or respect to merit in one’s dealings with another. Gregory of Nyssa writes in his Catechical Oration 35 that persons who have not been purified of their sins by repentance or baptism or invocation of the divine power have to get what is proper to them, which means they must be purified of their sins “just as the furnace is the proper thing for gold alloyed with dross,” so they will be restored to God afterwards. The suggestion seems to be that the proper course of action with regards to a sinful human person is not infinite punishment or annihilation, akin to throwing away gold because of impurities. Rather, what is valuable must be restored by the necessary means so that so great a good is not lost. So also Athanasius argued that God’s goodness would be compromised if human beings, created rational in his image, were destroyed—whether deservedly or undeservedly (On the Incarnation 6). This is because human beings, alone bearing the divine likeness, are eminently valuable, so much so that God gave them dominion over his entire creation (cf. Ps 8.5-8). If God will restore the creation from bondage (Rom 8.20-1), how much more valuable is the human race for whom it was created! Jesus said himself: “Of how much more value are you than the birds!” (Luke 12.24) Maybe the great value of the human person doesn’t permit God in his justice to destroy or damn him forever.
I have thus responded at length to Oliver Crisp’s objection. But now Wesleyans like Jerry Walls will rightly ask: how can God ensure that all will be saved through the punishments of Gehenna? What of human free will? At this juncture, I do well to obey Prov 30.6: “Do not add to [God’s] words, or else he will rebuke you, and you will be found a liar.” Isaac the Syrian held that Gehenna is a “hidden mystery” which God uses “as a way of bringing to perfection” sinners being punished (Second Part 39, 20). Likewise Origen said that the worst sinners will be restored “in some way I do not know” (Ramelli 2013, 177). We cannot know how it will happen that sinners are restored. But against Plato, who held that the vicious suffering endless torment in Tartarus become incurable, Origen insisted: “nothing is impossible for the Omnipotent; no being is incurable for the One who created it” (153). Christ too, speaking of salvation, said: “For mortals it is impossible, but for God all things are possible” (Mt 19.26; Mark 10.27; Luke 18.27). When Jesus visited the synagogue leader’s dead daughter, he said: “The child is not dead but sleeping” (Mark 5.39; Luke 8.52). Before Christ, even the irreversible, unchangeable state of death becomes transient and curable, like sleep. God can give life to the spiritually dead, too, as Paul writes (Eph 2.1, 4-5). So I cannot say how sinners in hell will be saved, even while remaining confident that they will be. Still, I must address Walls’s powerful argument that sinners may freely choose to remain in hell.
I start by noting that the classical universalists had a particular understanding of the nature of freedom. Clement, Origen, Gregory of Nyssa, and others affirmed ethical intellectualism. On this view, “How one behaves depends on what one knows and how one thinks and regards reality; will depends on intellect and is not an autonomous force” (Ramelli 2013, 178). The will has an intrinsic orientation towards what the intellect perceives as good. For this reason, “evil is never chosen qua evil, but because it is mistaken for a good, out of an error of judgment, due to insufficient knowledge and/or obnubilation” (ibid.). This teleological conception of will implies that freedom is evaluated according to an agent’s ability to act on the basis of knowledge rather than ignorance. Conditions of hazy ambiguity may enable an agent to make a choice between competing alternatives, but this is not what freedom consists in. On the contrary, ignorant uncertainty impedes freedom.
Now some classical universalists described what happens in Gehenna as persuasion, which does not violate but sustains freedom. Bardaisan writes: “And there will come a time when even this capacity for harm that remains in [sinners] will be brought to an end by the instruction that will obtain in a different arrangement of things; and, once that new world will be constituted, all evil movements will cease, all rebellions will come to an end, and the fools will be persuaded, and the lacks will be filled, and there will be safety and peace, as a gift of the Lord of all natures” (113). Clement, Origen, and others speak of the illumination of the intellect which will follow upon purification by the next world’s punishments (125, 129, 178, 332). This illumination enables sinners freely to cling to the true Good, which is God. Furthermore, Origen suggests that there will be a sort of destruction of identity in the next world. He appealed to Ps 78.34: “When he killed them, they sought him; they repented and sought God earnestly.” He gives Paul’s example: the informer and persecutor is killed, and the apostle of Jesus Christ comes to life afterward. Even Paul himself uses this sort of language: “I died to the law… I have been crucified with Christ” (Gal 2.19).
Now I will address Wall’s argument. It seems to me that Walls implicitly endorses ethical intellectualism when he writes that sinners choose hell because “hell can somehow be judged better than heaven” and “evil can be seen as a good to be desired” (Walls 1992, 125). Likewise, he writes that the damned “see some advantage to be gained in the choice of evil” (129). He also grants that “anyone who has a fully formed awareness that God is the source of happiness and sin is the cause of misery” could not choose evil (135; emphasis original), implying that will is oriented towards the good as understood by the intellect. My response is to pose a dilemma. I take it that where disagreement between persons is grounded in matters of intellect, persuasion is possible so long as adequate evidence is available. If the differences of opinion between Crisp, Walls, and myself are grounded in the intellect, then we should be able to convince each other if we have adequate evidence. Only if other non-intellectual factors get in the way—e.g., emotions, habits, fears, impulses, etc.—can persuasion prove impossible. This is because the intellect in itself is open to reality, and takes in new data and forms judgments accordingly. Now, by Walls’s own admission, if the choice to remain in hell is to be intelligible, it must be grounded in some conviction of the agent’s intellect. And if the choice is made on intellectual grounds, as I’ve said, the agent can be convinced to act otherwise. God could reveal to this person various things which would persuade her to leave hell. But if her choice is not grounded solely in the intellect, then its freedom and intelligibility is compromised. Consequently Walls’s defense fails.
Considering whether God could give the damned compelling evidence that they are wrong, Walls writes, “I am very dubious, however, that evidence is ever compelling, strictly speaking. This is especially so when we are dealing with matters as controversial as religious beliefs. The reason this is so is because belief is far more than a matter of the intellect. Our emotions, will, and desires are also involved. And if we are unwilling to repent, we cannot be compelled to do so by evidence” (2015, 81). But in response, I simply reiterate the same dilemma argument at another level. These emotions, volitions, and desires which impede a person’s repentance—are they grounded in the perception of the intellect as well or not? If they are not, it is difficult to see how hell is thus freely chosen. How free is a choice made on irrational, blind emotions, impulses, and desires with no connection whatsoever to the agent’s understanding? But if they are grounded in her understanding, then God could simply address these intellectual stumbling blocks so as to enable repentance.
The evidence offered does not need to be compelling but only adequate, by which I mean sufficient to convince the damned agent. Consider Kieslowski’s film Dekalog I. An atheist university professor’s son dies after the frozen town lake inexplicably melts and his son, who had been ice skating on it, goes under and drowns. They performed measured calculations numerous times, all of which suggested the ice would certainly have held under three times the boy’s weight. The professor himself had tested the resistance of the lake’s frozen surface the night before. After this death, he goes into the unfinished Catholic cathedral of the city, first knocking over the unfinished altar, and then crossing himself in repentance with frozen holy water. In this case, God addressed the professor’s cause of unbelief—his conviction that the world operated strictly according to certain, quantifiable natural laws—and offered adequate evidence to convince him he was wrong. Moreover the treatment is painful, which also motivates repentance. God may act the same way with sinners in hell.
There are still more objections to make. Essential to Walls’s argument is that human psychology will function roughly the same post-resurrection as it does pre-resurrection. Thus sinners will be able to deceive themselves about what is good for them, choosing to remain in hell rather than to repent, they will not have a clear vision that God is the ultimate Good, and so on. He supposes that clarity of vision about God and sin can only be won gradually in cooperation with God’s grace. But perhaps this isn’t true. Basil suggested that at the resurrection, “The veil will be removed from each one’s spiritual sight, which will return to being like that of angels” (Ramelli 2013, 348). Ramelli comments: “In this condition it is difficult to imagine how some people will still stick to evil” (ibid.). Perhaps at the resurrection, human psychology will function very differently than it presently does. We know through modern neuroscience that the connection between the brain and the mind is very tight. It could be that the brain of the resurrected body will provide human beings with a different perception of things. Gregory of Nyssa, following Origen and others, stressed that the resurrection affects human psychology as much as physiology because Christ assumed the whole human nature, not merely the body (422). Certainly in this world, people gain clearer vision about God, etc. by cooperating with the Holy Spirit and participating in the life of the next world. In that world, however, with another body, it could be natural. Maybe the damned will have a clear vision of God as the Good, sin as the cause of misery, and so on, and their misery consists at least partly in their realization that they have chosen wrongly. Isaac the Syrian wrote, “Those who are to be scourged in Gehenna will be tortured with the stripes of love; they who feel that they have sinned against love will suffer harder and more severe pangs from love than the pain that springs from fear” (523). Andrew Louth describes that popular tradition of Eastern Orthodoxy, according to which “The ultimate state of human beings, after the final judgment, is to behold the glory of God’s love” (Louth 2007, 242). This experience will be bliss for the redeemed but torment for the damned. Walls seems to endorse something like this (2015, 86). But if things should turn out like this, what would keep the damned person from accepting God’s offer of salvation, since she sees clearly that God is love, and she is tormented by her rejection of that love?
Here too we have to address the quality of the experience of hell. Walls writes suggestively that “hell is indeed a place of misery but not unbearable misery. This is why it can be freely chosen forever as one’s eternal destiny” (2004a, 212). I suggest that Walls’s picture of hell is insufficiently biblical. The scriptures consistently use images which suggest the experience of Gehenna will be unbearable. Jesus often uses the metaphor of unquenchable fire (Mt 5.22, 7.19, 13.40) tied with “weeping and gnashing of teeth” (13.42). He says it would be better to mutilate your body than to go into Gehenna’s fire (Mt 18.8-9). Fire suggests an unbearable and consuming experience, to be avoided at all costs. Indeed, that is the rhetorical function of these threats: the experience of hell has to be as utterly tormented as possible, so that it will be avoided at all costs. But it’s clear that Walls’s defense of the possibility of freely chosen eternal damnation is only plausible to the extent that the biblical description of hell is minimized and tamed.
Now Walls will strenuously object that if hell is unbearable, the repentance that takes place therein would be compelled and not genuine. He holds that repentance is a matter of a genuine change of heart, and “cannot simply be a matter of knuckling under because the pain is so great that one cannot stand it” (2015, 78). But a person can undergo a genuine change of heart when she sees the extreme, intolerable consequences of her choices. Consider what David says: “While I kept silence, my body wasted away through my groaning all day long. For day and night your hand was heavy upon me; my strength was dried up as by the heat of summer. Then I acknowledged my sin to you, and I did not hide my iniquity; I said, ‘I will confess my transgressions to the Lord,’ and you forgave the guilt of my sin” (Ps 32.3-5). David confesses because God’s hand was pressing upon him and it became unbearable: his body was wasting away and his strength was gone. The experience seems to have taken him to the limit, but he is forgiven when he confesses. What is essential to repentance is the change of heart, but that does not mean that it cannot be compelled in some sense. Imagine a person who falls into a tiger’s den: his regret and immediate desire to escape are no less genuine simply because he could not bear to remain there indefinitely. Or consider the case of Job, whose experiences are analogous to damnation. Like Judas, he considers it better for him never to have been born (Job 3.11-3; cf. Mt 26.24); like numerous figures in Revelation, he wants to die but is unable (3.20-1; cf. Rev 6.15-6); his state is one of continual distress (3.26; cf. Rom 2.9, Rev 14.11); he foresees no escape (7.7-8); he insists on justifying himself and maligns God’s character (9.14-8, 20, 22-3); and the ubiquitous, troubling presence of God is repugnant to him (7.19). Yet in the end he repents in sackcloth and ashes (42.6), is restored to God, and learns that no purpose of God can be thwarted (42.2). Furthermore, his friends are spared by his prayer on their behalf (42.7-8). Job provides an excellent parable for the unbearable but ultimately salvific sufferings of the damned.
Finally, I note that there are tensions in Walls’s argument. Throughout, he makes extensive appeal to the strictures of libertarian freedom of the will: a free agent must be able to choose and to reject, to have multiple courses of action before him. This is possible in the case of the damned because, though they have perhaps some evidence in favor of God’s call to repent, yet there is enough uncertainty and ambiguity for them to choose to remain in hell. Walls suggests this is necessary to preserve their freedom, but this directly contradicts the ethical intellectualism which is implied in his defense. On this latter scheme, ignorance is always an impediment to freedom and not what enables it. Multiple courses of action are open to an agent only insofar as her knowledge is incomplete; far from being the hallmark of freedom, it is evidence of unfreedom. Walls perhaps might deny ethical intellectualism in favor of libertarianism as he understands it, but to my mind this would be a mistake. For the classical universalists, ethical intellectualism was a way of understanding the truth that humankind had been made for God. Ramelli writes that “human orientation toward God is part and parcel of human creatural nature” (2013, 820). The will is naturally oriented towards the good as perceived by the intellect, and God is the true Good, for whom all persons seek, knowingly or not, as Paul says: God created humans “so they would search for God and perhaps grope for him and find him” (Acts 17.26-7). The alternative—the denial that the will has any such intrinsic teleology—seems to me more existentialist than Christian. Walls could not say, for example, that the redeemed perfect their moral freedom and the damned pervert it (1992, 131) if there is not an intrinsic teleology of the will by which to make these evaluations. There is no perverting something that has no intrinsic function or goal or purpose. So I submit that denying ethical intellectualism undermines his defense against Talbott, in addition to potentially compromising an important element of Christian theological anthropology.
Objections to universalism are only as good as the metaphysical, theological, anthropological, and axiological premises which motivate them. I’ve described the objections of Oliver Crisp and Jerry Walls as informed by the Augustinian and Wesleyan traditions, respectively, and I’ve attempted to uphold the classical universalist tradition with its fundamental commitments by use of varied arguments. In conclusion I want to suggest that this latter tradition contains its competitors’ desirable elements while rejecting what is more undesirable: alongside the Augustinian, the classical universalist affirms God’s providential control over all history and human persons; alongside the Wesleyan, the universalist upholds God’s love and goodness and respect for freedom as ultimate principles which guide his providence and interactions with the world. Returning to the Athanasian challenge with which I opened, above all the classical universalist is concerned to defend God’s goodness and majesty: the former in his desire to save all human persons, and the latter in his power to accomplish this end. As Origen said, “nothing is impossible for the Omnipotent; no being is incurable for the One who created it.”
Crisp, Oliver. "Augustinian universalism." International Journal for Philosophy of Religion 53, no. 3 (2003): 127-145.
—. Deviant Calvinism: Broadening Reformed Theology. Minneapolis, MN: Fortress Press, 2014.
—. "Is universalism a problem for particularists?" Scottish Journal of Theology 63, no. 1 (2010): 1-23.
Dalley, Stephanie, trans. Myths from Mesopotamia: Creation, The Flood, Gilgamesh, and Others. New York: Oxford University Press, 1989.
Louth, Andrew. "Eastern Orthodox Eschatology." In The Oxford Handbook of Eschatology, edited by Jerry L. Walls, 233-246. New York: Oxford University Press, 2007.
Ramelli, Ilaria. The Christian Doctrine of Apokatastasis: A Critical Assessment from the New Testament to Eriugena. Boston: Brill, 2013.
Walls, Jerry L. "A Hell of a Choice: Reply to Talbott." Religious Studies 40, no. 2 (2004): 203-216.
—. Heaven, Hell, and Purgatory: A Protestant View of the Cosmic Drama. Grand Rapids: Brazos Press, 2014.
—. Hell: The Logic of Damnation. Notre Dame: University of Notre Dame Press, 1992.