The ‘necessity of faith for salvation’ argument

The first argument in favour of Augustinian particularism was suggested to me upon reading through book 3 of John Owen's The Death of Death in the Death of Christ, although I am not primarily concerned to expound Owen here, and the argument has merit independently of whether or not Owen actually espoused it.

The central requirement of this argument for particularism depends on the notion that faith is a necessary condition for salvation. If salvation cannot be had without faith and many human beings never have faith in Christ, then many human beings are not saved by Christ's work. (Owen uses this reasoning against the Arminian view of the atonement, but I think we may also usefully appropriate it here.) For Augustinians, it would be useless to say that God could effectually procure the salvation of all human beings through the work of Christ, but then not ensure that they all had faith in order to make that effectual salvation effective in the life of the individual human being.Footnote ¹⁶ So, on this way of thinking, there is a requirement of faith for salvation. God cannot be said to ‘save’ anyone without ensuring that such a person has appropriated that salvation through the exercise of faith.

Let us call this the Requirement of faith condition for salvation, or RF condition for short. The idea here is that faith is a necessary, though not sufficient, condition for salvation that is supplied by God acting upon, or concurrently with, the human agent in choosing salvation. According to most Augustinians, saving faith is regarded as a gift of God's grace, not an action that originates with human choice (see e.g. Ephesians 2:8–9). Even where the gift of saving faith is committed to human beings in a way that involves human action in concurrence with divine action (rather than the divine agent acting upon a passive, or receptive human agent), the point is that human choice alone is insufficient for saving faith. At the very least, on an Augustinian model of salvation, God and the human being in question are both necessary but only jointly sufficient conditions for saving faith. Some Roman Catholic Augustinians seem to take a view similar to this one (some Thomists, for instance). Other Augustinians advocate a stronger position than this. In fact, Owen's view, in common with other Calvinists, is considerably stronger than this, denying that the human will is able to choose divine grace without the interposition of prevenient divine grace to enable that choice. Be that as it may, my point here is simply to show that, even if an Augustinian were to take the weaker view and allow a concurrence between human choice and divine choice (however that might come about), the human choice is not sufficient for saving faith without the divine will acting upon it.

So, let us assume for the sake of argument that faith in Christ is necessary, but not sufficient for salvation that is supplied by God acting upon, or concurrently with, the human agent in choosing salvation as per the RF condition. Then either God gives faith to all, and all are saved, or he gives faith to some and they are saved, or to none. Owen, and Augustinians who hold the RF condition, say God does not give faith to all, because all are not saved. Many human agents are born and die without any knowledge of Christ and without faith in him. So, according to the logic of his position, these individuals cannot be saved. Perhaps this is true. But how does this rebut Augustinian universalism? Recall that Augustinian universalism says merely that God could save all people and that this is consistent with the tenets of Augustinian thinking. The problem Augustinian universalism raises is that God could have brought about a world both Augustinian and universalist, not that he has done so. For the idea here is that if God could have done this, he should have done this.

In fact, aspects of Owen's argument (or an argument inspired by Owen's thinking) can be incorporated into a version of Augustinian universalism – although, I am sure Owen himself would be appalled at this prospect. We could say God creates an Augustinian universalistic world, and ensures that all human beings attain salvation by faith in Christ. Naturally, not all will do so in this life, because not all people will hear about Christ or have the chance to trust in him. But God could surely ensure that all have such a chance post-mortem. Presumably, God could ensure that all those who have not believed in Christ during this life do so in some post-mortem existence, and are acted upon by divine grace in such an irresistible way that they all (eventually) choose Christ and are saved. There is no element of contingency here as there is with contingent universalism, because God ensures all have a saving faith, as per Ephesians 2:8–9. If a follower of Owen were to respond by saying that, whether God could have done this or not, God has not done this, so such reasoning is idle, then they have missed the point of the Augustinian universalistic argument. The point is this: if God could have done this, and there is no moral impediment to him doing this, why has he not done so? Hence, the addition of the RF condition to traditional Augustinianism does not rebut Augustinian universalism.

But we might take this Owen-esque argument in a different way. Perhaps saving faith is not a condition of salvation because knowledge of this salvation is not a prerequisite of salvation. In which case, the RF condition is false. We might put it like this: a particular salvific event could be ontologically necessary for salvation but not epistemically necessary for salvation. Then, Christ's saving work is ontologically necessary for my salvation (his death has to be efficacious for me). But it does not follow from this that I have to know that it is efficacious for me. Famously, Karl Rahner maintained something like this in his ‘Anonymous Christianity’ version of inclusivism:

Let us call this second line of reasoning the Inclusivist-Universalist Augustinian argument, or IUA argument. Perhaps an Augustinian universalist could say that those who do not know and trust Christ may yet be saved via Christ's work, even if they are ignorant of that work.Footnote ¹⁸ It may be thought that if I do not know that some event is salvific for me, it cannot be salvific for me, since I would not know I was saved. But this is false. Let us say that my wife will treat me to a special birthday meal at a local restaurant this evening. I, however, am blissfully unaware that she has arranged this. Clearly, my present ignorance of a future birthday treat has no bearing on whether or not my wife will actually treat me to this meal. She will take me out for a meal although I do not know this now, and will not know it until we walk into the restaurant foyer. In a similar fashion, a person's current ignorance of some future salvation has no bearing upon whether he or she is amongst those who will be saved or not.

Transpose this thinking onto the Christian scheme of salvation: A cannibal from Papua New Guinea does not know the great things of the gospel. He eats every missionary sent to him before the poor creatures manage to utter a word about the gospel. And, lacking any other means by which he might come to believe in Christ (he is illiterate, so unable to read the Bibles that the missionaries bring with them), the cannibal dies ignorant of the gospel. Is he damned? According to proponents of the RF condition he is, because he cannot fulfil the RF condition. But why should we concede this? Assume with Augustinian universalism that the cannibal is, like the whole of humanity, amongst the elect, saved via Christ's work. Although he does not know this now, and will not know it in this life, he is one of the elect. To be sure, his salvation will be a great surprise to him when he discovers it. But it could be that he is saved via the salvific event of Christ's work, even if he is ignorant of that work. The point of the proponent of the RF condition is that the only effective way in which a person can be in a position to benefit from the work of Christ is if one is given the faith to participate in the benefits of that work. And, it would seem, he or she does not think a purely dispositional faith will do (that is, a faith that is somehow given to an individual who may be ignorant of the object of faith, and who may never be in a position to exercise that faith – a view that can be found in the writings of several doctrinally orthodox CalvinistsFootnote ¹⁹). But that need not apply if God actualises a world where the work of Christ just is automatically applied to each individual person whether they are aware of it or not (that is, where conscious faith in Christ is not a requirement of salvation).

This IUA argument may be stretching the ‘Augustinian’ component of Augustinian universalism somewhat. It might be thought that someone who takes such a view ceases to be an Augustinian because Augustinians maintain that faith is a necessary condition of salvation. Well, perhaps something like that is right. But then again, perhaps not – at the very least, there are minority voices in the Augustinian tradition, like Jonathan Edwards and William Shedd, who seem to think faith that is exercised (as opposed to dispositional) might not be necessary for salvation. Maybe a weaker version of the IUA view can be made compatible with the biblical claims that faith is a prerequisite of salvation for human beings, adding that this is the normal or conventional, but not necessarily the only way in which human beings are saved. Maybe some human beings, like the Old Testament saints, have no saving knowledge of Christ (no conscious faith in Christ), and yet are saved via Christ's work.Footnote ²⁰ It seems to me that many Christian theologians would accept this. But this does not denigrate faith. Faith, or rather, a faith that is exercised not merely dispositional, is still the normal means to salvation if one accepts this particular version of IUA universalism. Nor does it mean that, on the basis of such an argument as this, Christians have no reason to proselytise those who are ignorant of salvation since they will be saved whether or not they hear of the gospel. Surely it is better to know you are saved than to be in ignorance of the fact? Surely it is better to know where one is bound rather than to go there blindly? If I was worried about an uncertain future, although, in point of fact and unbeknown to me, my future was secure – and you knew this but did not tell me, though you could have – would we not consider your inaction to be immoral? Yet this is the state of those ignorant of their salvation, according to the IUA argument. And you and I are those able to relieve them of their ignorance.

The point I am making here is that a person could be saved irrespective of whether they knew they were saved or not, not that there is a moral equivalence between being in a state of ignorance about one's salvation and being in a state where one is aware of it.

So, although this IUA argument does require the Augustinian to give up the necessity of the RF condition for human salvation beloved of Augustinians like John Owen, this need not be so great a concession, provided some argument can be mounted to show that those who never hear the gospel and die in ignorance of it might be sufficiently like the Old Testament saints, or those incapable of conscious faith (e.g. the severely mentally handicapped) who are in just this sort of position with respect to salvation. Thus, for example, William Shedd:

Owen is not unaware of the sorts of responses that might be made to his position (although he does not respond to exactly the sorts of arguments I have mounted here – and, of course, Shedd postdates Owen in this matter). On the question of the requirement of faith as a necessary condition for salvation, Owen has this to say:

This passage is rather compressed, so I shall unpack it a little in order to make its structure clear. First, Owen's reasoning depends on the claim that unbelief is a sin and is a cause of damnation. It is a sin because it involves not giving due honour and regard to God. Presumably, unbelief is not the only cause of damnation. Possession of a corrupt moral nature supplied by original sin would be another cause, in fact the very thing that gives rise to unbelief in the first place.

Second, either Christ died for the sinful unbelief of all humanity, or he did not. If, according to Owen, Christ did not die for the sinful unbelief of all humanity, then he did not die for all the sins of all humanity and all are not saved. The reason being that, if Christ had died for the sinful unbelief of all humanity, he would have atoned for the unbelief of all humanity as one of the sins of all human beings, and all humanity would have been saved. But, says Owen, the evidence of scripture and experience show that not all are saved. So Christ did not atone for the sinful unbelief of all humanity.

Alternatively, if Christ did die for the sinful unbelief of all humanity, then all humanity would be saved because he would have atoned for all the sin of humanity, unbelief included. But then, reasons Owen, unbelief would not be an obstacle to salvation, and (presumably) all would be saved (irrespective of their unbelief and whether they continued in their unbelief during this life). But of course, Owen has already said that all are not saved according to scripture and experience. So it cannot be that Christ died for the sinful unbelief of all humanity, irrespective of whether or not all human beings are aware of this fact.

Owen thinks that this counts against the sort of reasoning mounted by his Arminian opponents.Footnote ²³ But it might also be applied to universalism of an Augustinian kind. Let us do just that.

First, we will presume, in common with Owen (and the majority voice in the Christian tradition) that all of humanity are not saved, so Christ cannot have died to save all of humanity from their sinfulness, including their sinful unbelief, otherwise they would be saved from their sinful condition, including, as one of the constituents of this condition, their unbelief. Then, presumably, Christ dies for their unbelief conditionally, that is conditional upon their accepting Christ, or something of that nature. However it is that Christ does atone for the unbelief of those that do believe in him, and only those (not the whole of humanity), our question is whether or not Owen is correct about this Augustinian form of the problem. And I think he is not correct in this matter. For on Augustinian universalism Christ does atone for unbelief because he atones for all sin. How is it that some human beings go to the grave without knowledge of Christ and are saved? Perhaps because, like the Patriarchs, they go to some post-mortem existence where Christ is revealed to them in a saving fashion, consistent with theological compatibilism (I am assuming Augustinianism is consistent with theological compatibilism). Or, perhaps conscious faith is not a requirement for salvation and the work of Christ can be applied to all human beings irrespective of whether or not they have conscious faith in Christ (as per Shedd, and, perhaps, EdwardsFootnote ²⁴). So, it seems, Owen's case does not offer a convincing argument against Augustinian universalism and in favour of an Augustinian account of particularism.

A new argument for particularism

In the last section of this article, I want to offer a new argument in defence of an Augustinian account of particularism that succeeds where Owen's account does not. The question of whether or not God must create the best possible world is a hoary old chestnut in the philosophy of religion. Philosophers have been divided over the question of whether or not the notion of a best possible world even makes sense, and if it does, whether God has to create the best anyway. I side with Richard Swinburne on this question. He maintains that there can be no best possible world because, for any optimal world we care to imagine, it can always be made better by the addition of one more person (or one less person in some cases), or one more planet, such that the overall objective moral worth of that particular world is greater than the previous world, and so on ad infinitum.Footnote ²⁵ Swinburne is not alone is his denial of a best possible world thesis. Thomas Aquinas famously denied that there was a single optimal sort of world that God had to create. There are, Thomas argued, an infinite number of possible worlds all compatible with divine goodness, and God could have chosen to bring about any one of them:

This sort of argument seems plausible to me. In setting out this new argument for an Augustinian particularism, let us assume that it is plausible. And let us also suppose, with classical theism, that God is essentially good, that is, he is essentially morally and metaphysically good. Thomas thinks God is not merely good but goodness itself, as is well known.Footnote ²⁷ But we need not make that assumption in addition to the idea that God is supremely morally and metaphysically good. (It may be that being morally and metaphysically good in the relevant sense are compatible with being goodness itself, but I do not propose to explore these matters here.) I take it that God is supremely or maximally metaphysically good, provided he has, as Thomas Morris puts it, ‘ontological completeness’. That is, he has all those compossible great making properties requisite for being the one ontologically ‘complete’ being, as per Anselmian perfect being theology. And God is supremely morally good provided he is the most exalted and morally pristine being, who acts in accordance with all moral principles relevant to his actions.Footnote ²⁸ To this rough-and-ready conception of divine goodness we may join Thomas’ reasoning that there is no single best possible world God could and should have created. In fact, there are an infinite number of worlds he could have created consistent with his goodness.

To these three assumptions let us add several others that fit with the Augustinian way of thinking. The fourth is that the actual world, and worlds very similar to this one, is among the set of worlds that God could have created because they are worlds consistent with divine goodness. This seems plausible. Fifth, a theologically commonplace principle that Augustinians share with a number of other Christian theologians: divine grace is necessarily not part and parcel of divine distributive justice.Footnote ²⁹ The two notions are distinct. God may act in a gracious manner towards his creatures, but he need not act in this way, and there can be no requirement for him to do so, because it is of the nature of an act of grace that it is free and not obligatory. By contrast, divine justice is inexorable and must be satisfied on an Augustinian way of thinking. It cannot be set aside, passed over, or abrogated. So there is an important distinction between the nature of a gracious act and the nature of a just act that bears upon the present concern.

Our sixth and final assumption is this. It may be that God is under no moral obligations until and unless he places himself under such obligations, e.g. he ordains that he will create a particular world where sin occurs and that, as a result of this, he will freely provide a means of salvation in the person of Christ. If this were the case, then ‘prior’ to this decision to create the world (i.e. the actual world) God is under no obligation to create a world, or the particular world he does create. Nor is he under any obligation to provide a means of salvation for human sinners once he has ordained to create a world where human beings freely rebel and sin. (I shall return to this matter presently.)

With these provisos in mind, let us consider an argument in defence of particularism: if salvation is a work of divine grace, then God could create a world where no one is saved and this would be a world consistent with his divine goodness. Call this world, W1. In such a world, all human beings would be damned because of their sinfulness, and God would be acting according to his just nature if he were to damn all of fallen humanity. Not only would he be acting according to divine justice, he would also be acting in a way consistent with his goodness. For we are working on the assumption that it is good for God to create a world, and the particular world he does create (in this hypothetical case, W1), and that God is not obligated to save any human sinner who exists in the world he does create, provided the sinner has freely sinned against God – for salvation is a matter of divine grace, which is gratuitous. Hence, in W1 God saves no one, which is consistent with his goodness. However, W1 does have a significant disadvantage for the Augustinian. It is this: in such a world divine benevolence and divine grace are not both displayed, because divine grace is not offered to any of the human agents God has made in W1. But, as we have already seen, it is an important constituent of traditional Augustinianism that both divine justice and divine mercy are displayed in the creation in order that God's name is vindicated (namely, the third constituent of the restricted elect view). Thus, although such a world would be consistent with exercise of divine goodness and justice taken in abstraction, as it were, from the particular concerns of Augustinianism, it would not be consistent with the way in which God exercises his goodness and justice – indeed, must exercise these attributes – according to traditional Augustinianism.

Now, consider a different world. This second sort of world is a particularist world, where some but not all of humanity are saved. Call this, W2. If God were to create W2 then we might think he would have created a better world than the previous one, a world that is consistent with his moral nature, as seen through the lens of Augustinianism. For W2 would be consistent with the requirement of the restricted elect view that both divine justice and divine mercy are displayed to the creation. Such a world would better comport with divine metaphysical goodness, which is the aspect of goodness that has to do with ontological completeness. For, according to the Augustinian way of thinking, God's nature is such that he must display his justice and mercy in his creation and it is a good thing that both of these aspects of the divine nature are displayed in the created order. The reason given for thinking this on the restricted elect view was that this arrangement somehow vindicates the divine character by displaying both his justice and mercy to his creatures, provided that (at least some of) those upon whom this distributive justice is visited are sinners, in the sense outlined previously. Then, W2 might be thought to be an objectively morally better sort of world than W1 because it is a world where some human beings are saved, rather than no human beings. Of course, a world where some people are saved rather than no people will only seem to be an objectively morally better world if we think that, for any world God could create where n number of people are saved, a world where n+1 people are saved is an objectively morally better world.Footnote ³⁰ Whilst such a principle may not be unassailable, it seems plausible.

Now, here is the question: need God do more than this? That is, must he create a world better than any of this infinite number of particularist worlds like W2, consistent with his goodness, where his justice and mercy are shown forth to all creation in the work of Christ? In other words, for our purposes, must God create an Augustinian universalist world? The answer is clear: given the logic of the Augustinian view, he need not do so. The fact that God could do so does not mean he must do so. All that God has to do is act in accordance with his divine nature. Creation of a particularist sort of world rather than a universalist world is consistent with his divine nature (where particularism is the notion that God saves some number of humanity less than the total number of humanity). But is God obliged by his own nature to create a universalistic world because it is objectively morally better than all other, particularist worlds (given our caveat about what, in matters of salvation, makes a world of greater objective moral worth than another world)? No he is not obliged to do more than act in a way that is consistent with his divine nature. (In fact, as I have already indicated, I do not think language of obligation applies to God at all, apart from the extended sense in which God may place himself under a prima facie obligation through promising or ordaining to do a particular thing.Footnote ³¹) In any case, as Robert Adams has shown, God has no obligations to creatures he has not created. One can only have obligations to things that exist. If God eternally decrees to create a particularist sort of world consistent with his goodness, he cannot be said to have an obligation to bring about better worlds because he cannot have an obligation to worlds and creatures that do not exist. The reasoning behind this claim is as follows:

At this point it would be helpful to distinguish between God having obligations to himself, that is, his own nature, and his having obligations to some non-existing but possible persons. I am claiming that God may have obligations of the first sort, which are really nothing more than God's acting in a way consistent with his own metaphysically good character, but that he does not have obligations of the second sort. It might be thought that on the Augustinian way of thinking God could have an obligation to his own nature to redeem (at least some) human beings ruined by sin. For consider this analogy: I might not be obligated to let you stay the night in my house – yet I consent to do so. Once I have agreed to have you stay overnight, I place myself under an obligation to provide you with a place to stay for the night. Having promised this to you, I must see it through. I am suggesting that it is this sort of reasoning that applies to God. Once he has decreed that he will create a world that will be populated by sinful creatures he appears to have placed himself under certain moral constraints. If that is right, then it looks like God does have obligations to possible but uncreated creatures after all, in this limited sense: for any world that God creates, if that world is populated by sinful creatures whose sin is the result of morally free rebellion, God is under a moral constraint generated by the internal configuration of his own divine nature to redeem (at least some of) these sinful creatures, in order that his justice and mercy be displayed in the created order. But clearly this generates a problem for the Augustinian particularist.

There are several possible responses to this difficulty consistent with the Augustinian way of thinking outlined thus far. First, the Augustinian could simply deny that God has any such obligations. This would mean tinkering with the notion that God must display his justice and mercy in the created order, which, even if this can be made to work on a revised version of Augustinianism, seems like an unnecessary complication to the argument, if there is some simpler solution. Second, one could concede that God does have such an obligation to his own nature once he has decreed to create the world he does create, because it is a requirement of the Augustinian view that God displays both his justice and mercy in the creation. This need not mean God is required to create an Augustinian universalist world, only that he is required to create a world where these two attributes or predicates can be displayed. This is consistent with either Augustinian universalism or the Augustinian particularism developed thus far. But it might be thought to tell against Augustinian particularism if this requirement means God is not free in his act of creation. However, there is no reason to think this restricts divine freedom in a way that is damaging to the case for Augustinian particularism. On this way of thinking, God is still free to create or not, and to create the particular world he does create, or not, as he pleases. What he is not free to do is to create a world where his justice and mercy are not displayed, because this would be inconsistent with his divine nature. But this is true on Augustinian universalism and Augustinian particularism.

With this made tolerably clear, we can return to Robert Adams and the notion that God has no obligations to possible but uncreated creatures. Suppose, with Adams, that the following conditions apply to the world God does create (call it W3 for the sake of clarity):

1. (1) None of the individual creatures in W3 would exist in an Augustinian universalistic world.

2. (2) None of the creatures in W3 has a life that is so miserable on the whole that it would be better for that creature if it had never existed.

3. (3) Every individual creature in W3 is at least as happy on the whole for the period it exists, as it would have been in any other possible world in which it could have existed.Footnote ³³

To which we need to add:

1. (4) Each member of the elect is transworld elect and each member of the reprobate is transworld reprobate. That is, there are no possible worlds at which a W3 human creature exists and is among the elect that exists and is among the reprobate in W3.Footnote ³⁴

This is a strong modal claim, but I think it is consistent with the Augustinian notion that God elects certain individuals and reprobates others. Augustinians commonly claim that what grounds election and reprobation is the divine will (arbitrium) and nothing else (e.g. foreknowledge of creaturely actions). In which case, it would appear that what God wills for the eternal destiny of a particular individual in one world would apply, other things being equal, to the destiny of that individual in any other possible world in which they might have existed. This acts as a safeguard against the objection that some W3 human creatures might be reprobate in W3 but elect in another Augustinian particularist world, which would seem to vitiate Adams’ third claim ((3) above). The addition of this fourth claim blocks this sort of objection.

There is one other objection to this modified Adamsian argument for Augustinian particularism. It concerns the second claim ((2) above) that no creature in W3 has a life that is so miserable on the whole that it would have been better had that creature never existed. The plausibility of this claim seems considerably diminished on a particularist way of construing it, given that some such creatures end up in hell, where they are punished everlastingly. This, I concede, is a difficult matter. One potential solution would be to argue that in W3 the reprobate are not punished everlastingly in hell, but annihilated. This need not be inconsistent with an infinite punishment thesis, if, with Peter Geach, we allow that an infinite punishment could take place in a finite time (where time itself is a densely ordered continuum).Footnote ³⁵ But even if this turns out not to be compatible with an infinite punishment thesis, the addition of an annihilation doctrine might be worth the cost to Augustinianism if it makes the second Adamsian claim more plausible. For a life that ends in annihilation rather than everlasting punishment may not be so miserable on the whole that it would be better for that creature to have never existed.

With these caveats, I suggest that creating a W3-type world does not wrong any of the creatures contained in it. Nor can God be said to be obliged to create a better world (i.e. a universalistic world), if, in addition to these conditions (for the purposes of our argument), the world created is consistent with divine goodness and is a world where some number of human beings less than the total number of human beings are the object of special divine grace and are saved (which is not specified in (4) above). In fact, I have argued that God could create a world where no one is saved and all are damned because of their sinful natures and that this is also consistent with divine goodness. (Though such a world would not be consistent with traditional Augustinianism.) It may well be that this world (that is, the actual world) is a particularist world. And it seems to me that such a world is consistent with divine goodness.

Of course, both this argument for particularism and the argument that depends upon the RF condition concede to the Augustinian universalist all three conditions of the restricted elect view. But this need not be the case. The Augustinian could adopt another approach, involving revision of one (or more) of these conditions. The obvious candidate is condition (c), which had to do with ‘The need for the display of both God's grace and mercy and his wrath and justice in his created order’. But, the Augustinian might argue, it is important that the display of divine justice has some connection to desert. Were Christ to be the only human person upon whom divine justice was visited, as a vicarious substitute for sinners (as per Augustinian universalism), this would not have the right connection to desert because Christ does not deserve to be punished – he acts vicariously (and sinlessly) on behalf of sinful human beings deserving of punishment. There has to be some connection between the display of divine justice and the idea that (at least some of) those upon whom divine justice is visited are deserving of punishment. In order to reflect this, we could rephrase condition (c) as follows:

1. (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.

Let us call this the strict justice condition. There seems to be a good independent theological reason for thinking the strict justice condition is true. Unlike the Augustinian universalist argument outlined earlier, the strict justice condition requires that at least some of those upon whom divine justice is visited deserve to be recipients of that justice, which Christ is not. So the strict justice condition promises to deliver another reason for the Augustinian to resist universalism.Footnote ³⁶

Applied to the foregoing, this would mean that 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. Taken together with the modified Adamsian argument, either as a supplement to the Adamsian argument, or an additional, independent reason for particularism, this constitutes a strong case in favour of non-universalistic Augustinianism. In which case, God has good reason to create a particularist world, and no obligation to create an Augustinian universalist world instead. And this rebuts the necessity of an Augustinian version of universalism.Footnote ³⁷