Middle Knowledge, Gratuitous Evil and the Presence Model of Hell

That Damned Problem of the Creation of the Damned.

Two books that have been incredibly formative for me are “Luis De Molina: The Life and Theology of the Founder of Middle Knowledge” by Kirk MacGregor and “Sinners in the Presence of a Loving God: An Essay on the Problem of Hell” by Zachary R. Manis. MacGregors work should be the very first place anyone seeking to understand the concept of middle knowledge goes. While MacGregors work has informed my thinking on God’s omniscience, providence, concurrence, and most profoundly the problem of evil (more on that later), Manis has definitively shaped my thinking when it comes to hell. This however leads to a problem, as Manis writes:

“We have seen, first, that an adequate view of hell… must reject any account of meticulous providence, according to which God’s control extends to everything that occurs in the world, including creaturely free choices, for all such views founder at least on the question of why God creates the damned (with some views, such as Calvinism, facing intractable difficulties along a number of other lines, as well). Although classically considered a theological ideal, meticulous providence turns out to be a serious liability when it comes to the problem of hell, and the benefit–cost ratio is unfavorable to any non-universalist theology that includes it…” (Manis 247).

            So, as a middle knowledge affirming Calvinist (look forward to a separate post on that) this would seem to be a problem for me, that an adequate view of hell, specifically the presence model being put forward by Manis (which I endorse!) seems to run counter to my acceptance of God’s meticulous providence. But to get the full force of Manis’ argument we need to understand the context in which this problem is raised, how the question of God’s creation of the damned figures into it, and why it’s a problem for middle knowledge. So, let’s motivate the argument.

The Context

            First, we need to understand Manis’ project, he is putting forward a model of hell termed “the presence model” which understands hell as the experience of the divine presence by unrepentant sinners. His model is within the eternal conscious torment family of views but as he demonstrates, avoids or provides solutions to serious problems facing the traditional ECT view. Beyond weighing this model against traditional ECT he also puts it up against the Lewisian “choice model”, annihilationism, and universalism (and towards the end of the book offers hybrid accounts of his model and each of these views). It is in the context of his discussion of universalism that this problem for middle knowledge arises. Manis notes a powerful argument from free will against universalism put forward by Kvanvig saying:

“The conclusion to which we are led by Kvanvig’s argument is this: even if everyone is in fact eventually saved, it must be at least possible that some are lost, from which it follows that creaturely damnation must not be incompatible with divine goodness and love, after all.” (Manis 117-9)

            Faced with the free will argument, the universalist may avail herself of the resources of middle knowledge to respond. This is a straightforward move, as middle knowledge allows God to providentially order the world without violating the libertarian (incompatibilist) free will of creatures. Thus, reasons the universalist, God could actualize a world where he knows that every person will freely be saved ensuring universal salvation without violating the libertarian free will of his creatures. Thus, molinism becomes a great aid to the universalist and a problem for non-universalist models of hell. But after elaborating how “molinism and universalism are a match made in heaven” (Manis 133) he goes on to survey molinist anti-universalism and this is the context in which he raises the problem of God’s creation of the damned.

The Problem

            The problem is that if God is perfectly loving (loves all people perfectly) and could actualize a universalist world where he knows everyone freely comes to be saved then he would actualize such a world. If he were to actualize a world where he knew people would be damned when the universalist world was feasible for him then this would mean that he isn’t perfectly loving. We have a quartet of inconsistent propositions, God’s moral perfection, middle-knowledge, anti-universalism, and feasible universalist worlds. According to Manis the proposition that makes the most sense for the Christian affirming middle knowledge to reject is the anti-universalism and this is due to the untenability of the molinist anti-universalist responses to the problem of God’s creation of the damned. Manis notes that Jerry Walls and William Lane Craig both reject the proposition from our quartet that there are feasible universalist worlds (NFUW) though Craig argues there are no feasible universalist worlds where heaven is sufficiently populated and includes transworld damnation in his account (NFUW_sp). Manis quotes Walls as acknowledging the problem saying:

“Perhaps God cannot create a world of free creatures in which all respond positively to His grace. Perhaps the best creatable world with free creatures includes some who will not respond to even an optimal measure of grace. It may be that all God can do in this regard is create as few such persons as possible. Of course, it would require further work to suggest why God could not altogether avoid creating persons who would decisively reject His grace.” (Manis 129).

            Manis goes on to say:

“In this passage, Walls tentatively endorses NFUW, but he does not offer any suggestions as to why it is true...He suggests that it is unnecessary for his project to provide such reasons, but it is not clear why he thinks this; he admits, after all, that the problem of God’s creation of the damned is a “weighty objection.” In my judgment, it is the weightiest of all objections to M-AU, and it remains unanswered until such reasons are produced.” (Manis 129).

The Solution

            Now that we have properly laid out the problem, we will look at a possible solution, as points of clarification I will not at this time be addressing Manis’ first point concerning the favorability of molinism to universalism (I’ll share my thoughts on universalism in a separate post). Further I agree with Manis’ assessment of Dr. Craigs (NFUW_sp) argument and as such it will not feature in my solution, though transworld damnation will. Lastly, I am aware of Kirk MacGregors excellent response to Manis in his latest book “Molinist Philosophical and Theological Ventures” which utilizes traducianism to support (NFUW) and while I’m sympathetic to this response I will be utilizing a separate bit of Kirk’s work in formulating mine.

            So why are there no feasible universalist worlds? I believe Kirk MacGregor’s response to the probabilistic problem of gratuitous evil can help us here, he writes:

“My proposed demonstration of the coexistence of God and gratuitous natural evil begins by defining perfection as completeness or wholeness in each and every respect. It then argues that perfection is an essential attribute of God that may only reside intrinsically in God as Necessary Being. In each and every respect, contingent being depends on Necessary Being for its existence and therefore lacks intrinsic completeness and wholeness. Thus in each and every respect, contingent being is finite and limited. Hence it is logically impossible for God to create anything that is intrinsically perfect, for the simple reason that it constitutes contingent being that is in every respect ontologically dependent on Necessary Being. Since Augustine, Christian theologians have overwhelmingly held that evil is not an independently existing thing but rather a lack, limitation, or incompleteness in something that is good, namely, an absence of the complete limitless fulfillment that equals perfection. It then follows deductively that evil is necessary to the creation, both to the natural order and to the human constitution, as God obviously cannot create anything with an intrinsic attribute exclusively proper to the divine nature. So, it is logically impossible for God to create a world without natural evil; if God chose to create anything at all, natural evil would necessarily come into existence, not because God created or caused it, but because whatever God created would not be God. Notice that all such evils are, in and of themselves, gratuitous; their only reason for existence is the logically unavoidable lack of ontological necessity exhibited by created entities.” (MacGregor 261).

So, because perfection is an incommunicable property of God anything he decides to create will be imperfect. This imperfection manifests as natural evil in worlds without agents, and as moral evil in worlds with agents. Thus, in any world with agents with libertarian free will (whether of the sourcehood or PAP variety) this imperfection will manifest in choosing wrong, including the choice to reject God, to remain in unrepentant sin. Thus, there are no feasible universalist worlds because “Gratuitous evil is logically necessary to the universe, and God simply has to put up with it if he chooses to create a universe at all” (MacGregor 262). Let us call this position no feasible universalist worlds: incommunicable perfection (NFUW_IP) But Manis raises a further problem:

“Even granting there is no feasible universalist world...a troublesome follow-up question arises for the defender of M-AU: Are the particular individuals who are lost in one feasible world saved in another? More specifically: among those feasible worlds containing the same persons, does the roster of those who are damned change from one to the next? If so, then it appears that, in choosing which world to create...God has to sacrifice some individuals… there are some persons whom God could have saved by arranging the world differently, but He chooses to sacrifice these individuals in order to save others.” (Manis 130-2).

            Here we can adopt Plantinga’s Transworld depravity and argue that God through his middle knowledge knew that any world he created would necessarily be imperfect, that there were no feasible universalist worlds due to this fact and thus actualized the world where the only people who are damned are those who would persist in unrepentant sin in any possible world. Knowing that damnation was a manifestation of the imperfection necessary to contingent being, God ensured that anyone who would respond to his free offer of salvation in any possible world would not be damned in the world he actualized. Rather damnation would only accrue to those who persist in unrepentant sin in all possible worlds. Manis concedes that “This allows one to avoid the view that God sacrifices anyone. Those who are lost could not be saved by God: no matter how He had arranged circumstances for them, their final destiny would be the same.” (Manis 133) Considering the combination of (NFUW_IP) and Transworld depravity it follows that the only way God could have avoided creaturely damnation altogether would be to refrain from creating a universe at all.

            Now Manis rejects the combination of (NFUW_sp) and Transworld damnation as “simply incredible” (Manis 133) saying:

“One must be prepared to affirm that it is actually true that (i) there is no feasible world in which all are saved and heaven is sufficiently populated and (ii) that it is worse for only a few to be saved than for many to be saved and some lost and (iii) that all those who are lost suffer from transworld damnation…If this is the price Molinism charges to shore up anti-universalism, the anti-universalist is best advised to keep shopping.” (Manis 133).

But for the person who accepts (NFUW_IP) the conjunctive proposition is (i) there is no feasible world where all are saved because imperfection is necessary to contingent being and (ii) that it is worse for none to be saved than for many to be saved and some lost and (iii) that all those who are lost suffer from Transworld depravity/damnation. This conjunctive proposition is more plausible in that the first conjunct entails the next two as the only way God could avoid creaturely damnation was by not creating at all, and Transworld depravity/damnation is a species of gratuitous evil which is the manifestation of the imperfection necessary to contingent being. Further conjunct two is rendered plausible in light of our being here in the world God actualized, that God chose to create when he knew that his creation would necessarily manifest such imperfections supports that it was worse for none to exist at all.

            This brings me to my last point which is that MacGregors response to the probabilistic problem of gratuitous evil and Manis’ presence model of hell are mutually reinforcing in that the presence model and MacGregor’s incommunicable perfection response are both doctrines of double effect. They show that God forsees/foreknows but does not intend or desire these imperfections. He foreknows through his middle knowledge that there will be imperfections in any world he creates, and that they will manifest in human agency as unrepentant sin, and that individuals who continue in unrepentant sin will experience his presence as hell, but he does not desire or intend any of these things, they are merely the necessary consequences of the imperfection of contingent being. As such with (NFUW_IP) we can reconcile middle knowledge with the presence model of hell and affirm with Mannis

“But, crucially, insofar as this suffering is not an artificial or arbitrary punishment imposed by God for the sake of retributive justice, insofar as God does not intend this suffering for the damned at all, and insofar as God does everything in His power, in the time leading up to the Day of the Lord, to spare each person from it, there would seem to be no basis for objecting that God’s treatment of the damned is unfair or unjust.” (Manis 290).

And the same goes for his creation of a world in which he middle-knew they would exist.

Works Cited

MacGregor, Kirk R. Luis de Molina. Zondervan Academic, 2015.

---. Molinist Philosophical and Theological Ventures. Wipf and Stock Publishers, 2022.

R Zachary Manis. Sinners in the Presence of a Loving God : An Essay on the Problem of Hell. Oxford University Press, 2019.

“Transworld Depravity.” Www.youtube.com, www.youtube.com/watch?v=keBw5Z-QWKs. Accessed 10 May 2024.

(Art by Beksinski)