If you’ve read even a little of my blog you’ll know that I’m a Thomist. I’m also Reformed, which if you don’t know is one of the two magisterial Protestant traditions, the other being Lutheranism. Those unfamiliar with the history of Reformed theology may find this combination strange, but there is in fact a long history of using Scholastic (and specifically Thomistic) methods and thought in service of Reformed theology. While many contemporary (or popular-level) Calvinist teachers seem to eschew natural or philosophical theology, this is by no means the dominant trend in Reformed theology. My aim in this post is to sketch how a more historically-informed and philosophically-nuanced understanding of the doctrines of grace can avoid many of the shortcomings of the “five points of Calvinism.”

Based on the discussions I’ve seen or been part of, you’d swear that the five points of Calvinism are the defining features of Reformed theology. In fact, prior to the Calvinist-Arminian debates, predestination itself wasn’t a major issue, let alone these particular doctrines. Indeed, these doctrines do not really capture the most important contributions that the Reformed tradition has made to the Christian theology. To my mind, more important are its sacramentology, covenant theology, and focus on the union with Christ. The Calvinist-Arminian controversy only occured in the century after the Reformation, and even then the Synod of Dort allows for more flexibility in our understanding of the doctrines of grace than you find in modern treatments.

I say all this in order to contextualize this post, which is a short summary of my current thoughts on the  Reformed doctrines of grace, commonly summarized as TULIP. Since discovering the Thomistic model of divine providence I have called omni-instrumentality, I have not returned to this topic, but only made some piecemeal and tentative conclusions. This is normal and healthy—whenever we change our ideas on something it is good to hold on to the other ideas we’ve had, even if they need revision, until such time as we can give them the time they deserve, lest we inadvertently throw the baby out with the bathwater. Two recent discussions have led me to collect my hereunto disconnected thoughts into something more coherent, which I summarize under the headings common in the Calvinist-Arminian debates: depravity, grace, election, atonement, and perseverance.

First, some general thoughts. Like most theological debates in the post-apostolic era, there is a tendency to read the terms of the Calvinist-Arminian debate back into the text of scripture. Numerous times, I have ended up studying this or that prooftext outside of the context of the debate only to find that it has nothing to do with the debate, or that if it did have something to say it was not as straightforward as I had been led to believe. It is not always appreciated how far removed some of the issues are from the primary concerns of the authors of scripture. This debate requires speculating about the hidden things of God, such as his intentions, the logical ordering of his actions before creation, and how he organizes history “behind the scenes”. Though the authors of scripture definitely do give us a window into these every now and then, if you look at how this debate proceeds you’d swear it’s almost the only thing they ever speak about! I’m by no means against systematic or speculative theology, but I do think an appreciation of this point would do a lot to temper expectations and force us to rethink methodology.

On the topic of methodology, as a Thomist I think part of the problem is that this debate often proceeds as though the only source of knowledge is scripture. But if we acknowledge that it concerns itself with “behind the scenes” questions, then we should rather think about how to integrate biblical theology with a robust natural and philosophical theology. Compare this to questions about the Trinity. Most agree that we need to synthesize the doctrine of the Trinity from what scripture teaches, because the doctrine as such is not laid out anywhere. Such synthesis requires great care, and its success also depends on our philosophical categories. Indeed, a large part of the development of Trinitarian orthodoxy was the development of philosophical categories rich enough for us to affirm all that scripture teaches about God. But when it comes to the doctrines of grace, for some reason people seem to assume that the opposite were true. The ironic result is that eschewing philosophical theology in favor of scriptural prooftexts leads to contrived interpretations forced into ill-informed philosophical constraints. We should seek to avoid this, along with the related error of pitting one passage against another when a better alternative exists. Now, I think the best philosophical framework for doing this sort of theology is Thomism. Thus, I will be discussing this in relation to the Reformed understanding of each of the doctrines of grace.

Depravity

The Reformed doctrine of total depravity affirms that sin negatively impacts every part of the human, both in our external acts as well as every part of the soul. It does not claim that any given part is as bad as can be (total in a qualitative sense) but that every part is corrupted to some degree (total in a quantitative sense). In this regard, there is not much disagreement between Reformed theology and Thomism—they would perhaps cash it out in different ways, but both get to the same point.^[1] Nature is utterly and inescapably corrupted by sin, and needs grace if it is to come to God.

Now, we do sometimes see the idea creep into Reformed thought that, if left to our own devices, corrupt humans would be as bad as they possibly could be.^[2] One way we see this come up is in some attempts to explain the relationship between God and sin. Wishing to affirm that God is the source of all good and not the author of sin, they propose that there is a metaphysical “tug-of-war” within the soul of each fallen human: the human always and only draws themselves to evil, while God always and only draws them to good. Wishing to affirm God’s meticulous sovereignty, they propose that God can choose to overpower the fallen human to any degree he requires. While I agree with each of the three desiderata—God as the source of good, creatures as the source of evil, and meticulous sovereignty—I find the actual proposal problematic for two reasons.

First, it legitimizes the caricature of total depravity as the claim that we are as bad as we could be. Yes, we don’t actually always and only act for evil, but the proposal is that this is what we would do if not held back by God. But this goes beyond the corruption of nature to its complete destruction! To be a creature with a particular nature is (in part) to be ordered to the good, for it is nature which so orders us—this is why things with different natures have different goods. Even if our corruption always and only orders us to self-interest, this would still have some measure of goodness in it and therefore be excluded by these sorts of proposals—that’s how radical they are. Second, these proposals seem to make God responsible for sin. Granted, they avoid making him the author of sin because he does not cause the sin to be present in the first place. But since God acts in every action by pulling the sinner some degree away from absolute evil, he is still the cause of how much sin ends up in every action. Thus, while he may not be the author (i.e., source) of sin, he is still the cause of how much sin there is, which seems equally as problematic.

Thomistic models of divine providence avoid both issues by conceiving of evil as the privation of good rather than as an opposing force in the fallen soul. As I have explained at length elsewhere,^[3] God is the divine agent who orders the human as a natural instrument to its end, which is its good. Evil is introduced by impediments in the instrument, which limit its ability to fully realize the ends to which it is ordered. God working through human choice is analogous to a human using a stick to move a stone: if the stick is rotten or weakened then it will fail to achieve the ends to which it has been ordered. The stick does not impose some contrary influence which somehow comes up against the human agent’s influence, but simply falls short of its purpose. Furthermore, just as a human agent can work with the limitations of their artificial instruments, so too can God work with the limitations of his natural instruments to bring about his purposes.

Since the Thomistic position holds that the fallen human preserves their natural order to the good, we may wonder whether it is possible for some humans to avoid actual sin without the aid of grace. Here Aquinas makes an interesting suggestion, based on moral psychology: although the fallen human can freely avoid any particular sin, they cannot avoid all sin. The fallen state of the human soul and world make some sin inevitable—we may avoid a particular sin through error in the opposite direction, or the pressures at play in a circumstance may prove too much for us, or we may be ignorant of the right course of action, or we may respond instinctively with hate before we catch ourselves, etc. Human experience provides us with many ways in which we can sin, and without the divine aid of grace, our fallen nature cannot avoid every one of them. Furthermore, as we proceed in life we will acquire various vices (i.e. habits not correctly ordered to the good), so that the corresponding sins become easier to fall into and harder to recognize as such. The upshot of this answer (which I have only briefly summarized) is that we can uphold the inevitability of sin and guilt before God without resorting to the extreme supposition that left to ourselves we are as bad as can be. Indeed, it does justice to the insidious nature of sin by saying that despite pursuing the good we inevitably fail to do so properly and instead fall into pitfalls like selfishness, anger, pride, laziness, cowardice, and so on. This is arguably a more helpless situation than one in which our sins are the result of  the “success” of something intrinsic in us.

Atonement

The question of atonement is arguably the least tightly coupled to the other four points in the Calvinist-Arminian debate. This is in part evidenced by the fact that there is such a thing as Amyraldism (sometimes improperly called “four-point Calvinism”), and in part because limited atonement really only arises in a form we would recognize today with John Own in the century following the Reformation. Everyone affirms that atonement is sufficient for all and efficient for some. Everyone affirms that there is a sense in which Christ died for the whole world and another sense in which he died specially for the elect. So what is it that distinguishes limited atonement from unlimited atonement? Some have suggested that it comes down to the divine intentions for Christ’s work at the cross,^[4] but the only way we can reliably reason about these is with reference to the concrete actions of God in redemptive history and the effects of the cross. Oliver Crisp offers a helpful analogy: suppose a company is producing a vaccine to combat some widespread disease knowing that some people will refuse it. The company could either (a) produce only enough vaccines for those they know will accept it, or (b) produce enough vaccines for everyone in the world and leave it up to them to refuse if they so wish. In both cases they have the potential to create vaccines for everyone, while in (a) they only actualize this potential for some and in (b) they actualize it for everyone, leaving it up to them whether they take advantage of this. Similarly, Christ’s death on the cross has unlimited potential to save everyone, but what was actualized? Like (a), limited atonement says that God only actualized this potential for the elect, whereas like (b), unlimited atonement says that he actualized some unlimited intermediate result, which can be taken hold of by anyone. Of course, as we shall see later, God elects only some to take hold of it, but the present point concerns only what the cross primarily achieved.

The precise nature of this intermediate result will depend on which version of unlimited atonement you’re considering, but from a Reformed perspective the most natural choice would have something to do with the union with Christ. Union with Christ has played a central role in Reformed   theology since the beginning, although different theologians have cashed it out in different ways. In my own work, I have suggested that it be understood as our incorporation into Christ’s glorified life, which is a “divine-human hybrid” made possible through his resurrection from death. I have also explained how this can be understood within the context of Thomistic metaphysics, and how it provides a better way of understanding penal substitution (another common Reformed doctrine).^[5] Even if we don’t accept my particular formulation, it is easy to see how the union with Christ might inform our understanding of atonement: Christ’s death secured a way of relating to God, which we partake in by being found in him. This union with Christ has various dimensions, each of which is part of what it means to be atoned: we are freed from sin, forgiven our guilt, reconciled to God, adopted as his children and co-heirs with Christ, clothed with immortality, glorified with Christ, seated in the heavenly places, and so on. It is this new reality made accessible to humanity which Christ secured at the cross. And, importantly, this new reality is unlimited in access and universal in nature: it doesn’t matter whether one, or ten million, or infinitely many people receive it, its power is never exhausted or reduced. It is not relevant whether God knows whether or intends for only a particular set of people to take hold of this gift—that is a question of election rather than atonement.

Thus, we have a thoroughly Reformed way of understanding unlimited atonement, and my particular version of it is fleshed out in terms of Thomistic metaphysics. Biblically, this also has the advantage of a straightforward reading for passages that speak of God saving “the world” or sinners in general: such passages merely describe how the unlimited gift achieved by Christ is accessible to all without qualification. On the other hand, passages that speak of a more limited scope are understood in terms of how, through God’s grace and election, only some come to actually accept this universal gift.

Grace

This leads us to the question of grace. The term “grace” can refer to many different things, even in scripture. It can refer to the gift of faith itself (Acts 18:27), the salvation that comes by faith (Eph 2:8–10), the state into which we enter through faith (Rom 5:1–2), the personal gifts we receive to serve the church (Rom 12:6), and more. Similarly,  in the present discussion we must distinguish between different kinds of grace. Calvinists speak of common grace, which God gives to everyone, and irresistible grace, by which God turns sinners to himself. Irresistible grace fits naturally with the account of depravity mentioned earlier, wherein the fallen human always and only acts for evil unless God draws them toward the good. God’s action must be irresistible in order to secure meticulous divine providence, and we call it irresistible grace when God acts so as to bring the sinner to faith and repentance. The same problem I raised with the general account is also applicable here: though God is not the source of the sinner’s rejection of him, he is the cause of the extent to which the sinner does or doesn’t reject him. Thus, irresistible grace seems to imply double predestination, which affirms a symmetry between God’s electing some and reprobating others: he is as responsible for the faith of the former as he is for the unbelief of the latter.

Now, when it comes to grace we want to avoid the Pelagian error of making the human praiseworthy for their acceptance of God, but we also want to avoid the other extreme of making God responsible for the unbelief of the sinner. Augustine (and Aquinas after him) sought to navigate the precarious path between both of these, and I think we should follow them. They begin with the distinction between operative and cooperative grace. Common to both is that God moves the human will to desire him, but  they differ in the state of the human so moved. In operative grace, God acts on (and in some sense against) the will that was previously rejecting him, making it turn to him and desire him. In cooperative grace, God works with a will that already desires him (as an effect of an earlier operative grace), strengthening it in this desire.  Naturally, we may ask: does operative grace contradict free will? In some sense yes, but in a more fundamental sense it actually establishes the freedom of the will. Consider the example of someone who wants to be free of their alcohol addiction, and prays to God for help. In order to do this, God would need to override their will so as to weaken their desire for alcohol. But in doing so, there is another sense in which he is actually establishing the person’s will, by freeing them from a desire they didn’t want.^[6] In order to apply this to operative grace, recall what we said about depravity. I suggested that we think of depravity as the inevitable failure to achieve the good to which we are ordered by nature. That is, God directs us as his instruments toward the good, but because of sin we fail to fully realize these ends. Indeed, some self-reflection shows us that everything we do, we do for the sake of some good we perceive in it, in virtue is which we think it is worth pursuing—indeed, pursuit-worthiness is all we mean by good in the minimal sense we’re considering, rather than something more specific like moral goodness.^[7] However, God is goodness itself, and the most worthwhile of any possible pursuit! Thus, when God acts with operative grace to turn someone to himself, he is turning them to the thing they have always desired, goodness itself. As Augustine said, “our heart is restless until it rests in you.” Similar to the alcoholism example, then, God overrides the sinner’s rejection of him in order to establish their more fundamental desire for goodness itself, and therefore frees their will rather than constraining it.

Now, just as we can impede God ordering us to the good, which is sin, so too we can impede God turning us to himself in operative grace or helping us in cooperative grace. In other words, both operative and cooperative grace are resistable. This is not because we can somehow overpower God (we have already rejected the idea of a tug-of-war in the human soul) but because God works his grace in us as his instruments, which we may fail to fully receive through weakness, ignorance, or some other impediment. The result of this is that God is the source of our faith, in which we play only a passive role, while we are the source of our continued rejection to operative grace. God, who desires all to be saved (1 Tim 2:3–4), is always ready to give operative grace to each person in their own way if they would only not impede its effect on them. Aquinas notes that this is similar to how the sun helps everyone to see except those who choose to close their eyes (SCG III.159).

He goes on to make the following important point. Just as God is capable of giving sight to the blind and raising the dead to life, so too is he capable of overcoming the impediments to grace in the human soul. He could therefore exercise an additional overcoming grace in order to prepare the soul to receive operative grace, and then exercise operative grace to turn the person to him. Thus, while operative grace may be resistable, its outcome may nevertheless be guaranteed if God also chooses to exercise overcoming grace.

Election

This brings us to the question of election, which concerns God’s choice from eternity of who will be saved (the elect) and who won’t be (the reprobate). The primary point of debate is the basis for God’s election: Calvinists hold that it is not conditioned on any good thing seen in the human, while Arminians hold that it is conditioned on foreseen faith. That is, the Arminian thinks that God chooses only the condition under which he will save anyone, whereas the Calvinist thinks that God also chooses which people will satisfy this condition. Thus, for the Arminian election is conditional while for the Calvinist it is unconditional. The kind of election made possible by the view of grace I have articulated above lies somewhere between these two, but is closer to the Calvinist. Similar to the Calvinist view, a person is saved because God works a specific grace in them which brings them to faith; but similar to the Arminian view, a person remains unsaved because they rejected God’s grace. So, for the elect, God is the reason they have faith and the reason their faith results in salvation, but for the reprobate, God is only the reason for why their rejection of him results in judgement. The Arminian holds that both election and reprobation are conditional, the Calvinist holds that both are unconditional, but what I’m suggesting is that election is unconditional and reprobation is conditional—there is no double predestination here.

But how is this possible? Surely if one is conditional then its opposite cannot be unconditional? The answer comes down to the fact that there are two kinds of grace at play, one resistable and the other overcoming. If  there was only resistable grace, then election would have to be conditional on whether each person impedes it or not. But because there is also overcoming grace, God’s election is not bound to the impediments (or lack thereof) in the human soul. Election is unconditional, then, because of overcoming grace, while reprobation is conditional because of the resistibility of operative and cooperative grace. Desiring all to be saved, God offers (or is ready to offer) to each person grace sufficient for them to come to faith, so long as they do not impede it. Those who do not impede it come to faith, in accordance with God’s universal will of salvation. Among those who impede it, he may choose to overcome their impediments if he so desires. But if he does not, then the blame remains on the human who rejected God’s sufficient grace for them—if someone freely rejects the help offered to them, no-one is to blame for leaving it at that rather than pursuing them further.

Why, if God desires all men to be saved, does he not give overcoming grace to everyone? Aquinas makes an informative comparison here: God giving overcoming grace is similar to him healing the blind or raising the dead (SCG III.161), in that it is a special act of God in the context of a more general act of his. In the context of nature, God’s general act upholds and orders everything to its natural end, the realization of which constitutes its flourishing, while his special acts overcome the impediments to this, such as blindness and death. His general act demonstrates his desire for the well-being of all creation, even while it is impeded by corruption and death, and it is with respect to this general act that we could say that God desires all people to see and live. But, there are other factors that God considers when he acts specially to heal or raise this or that person—reasons which we do not have access to, but which will ultimately be aligned with his general desire. We get a glimpse into one such reason, namely glorifying Jesus, who is of course God’s ultimate solution to sin and death (John 9 and 14). With regards to these special acts, God’s other reasons seem to result in him more often than not permitting the evil within nature. In the context of grace, God’s general act offers resistable but sufficient grace to all, while his special acts overcome the impediments to this. Just as in the context of nature, his general act demonstrates his general desire for all to be saved, but whether he chooses to act specially will depend on other factors even as it is ultimately aligned with his general desire. Once again, we get the occasional glimpse into what some of these other factors might be (e.g. Rom 9:22–24), but we do not have the whole picture, because we cannot read other people’s minds, let alone Gods! This, I take it, is really the core of faith, that we trust that God is working all things for the best, even when our limited perspective cannot understand it—this is certainly what the Psalmists grapple with over and over again. Behind the difference between God’s general and special acts, is what John of Damascus referred to as the distinction between God’s antecedent and consequent wills. In terms of what I have said, we could say that God’s will for universal salvation (enacted through his general act) is antecedent to these additional factors I’ve mentioned, whereas his more restricted election (enacted in part through his special acts) is consequent upon these additional factors.

In my estimation, this approach to the question takes the universalistic passages more seriously than Augustine (and the typical Calvinist exegesis) was able to. He proposed that we take these passages as referring to all kinds of people rather than all individual people, which may fit with some passages but is a stretch for others. The Reformed theologian should not settle for such exegesis, but search for a model which allows for a more natural reading of these passages. I think the approach from John of Damascus and Aquinas provides just this alternative.

Perseverance

The final doctrine of grace has to do with perseverance, which has to do with whether elect individuals are guaranteed to preserve to the end and take hold of salvation. The key error I see regarding this topic, in both Arminian and contemporary Reformed thought, is to conflate the election of individuals with their faith (or regeneration). The Arminian rejects the perseverance of the saints because they think individuals move in and out of election as they move in and out of satisfying the conditions of election (i.e. having faith). The contemporary Reformed rejects the possibility of someone having true and saving faith and then falling away, because they think this would imply that that individual had lost their election. The Arminian’s position results in a fluid notion of election, wherein God is not in control of which individuals are elect; the contemporary Reformed position results in an inability to take seriously those passages which speak about believers falling away, and undermining the biblical basis for assurance. Since my focus in this post is Reformed theology, let me expand upon the latter.

There are various passages which speak seriously about a person falling away and reaping the consequences. Consider the warning passages in Hebrews, or the warning to the Gentiles in Romans 11, or even just the parable of the sower. What one finds among contemporary Reformed thought when these sorts of passages come up is less a somber reflection on the fear of God, and more a bunch of mental gymnastics to avoid the possibility of falling away. At a certain point, one wonders what a biblical author could possibly have said to convince someone of the plain fact that people can and do walk away from the faith. More egregious, to my mind, is that these mental gymnastics undermine the biblical basis for assurance in one’s salvation. The biblical authors tell us that our faith is the basis for our assurance, because God will justify us by that faith on the final day when Jesus returns. Thus, assurance should be a simple matter of reflecting on our own hearts to know that we have faith in Christ, and clinging to the truth that if we continue in this (by his help) he will save us. If someone apostatizes (falls away), then they lose the assurance they once had, as they were warned. Simple! Now, part of those mental gymnastics I mentioned includes an attempt to explain away any real apostasy as illusory: no-one ever really apostatizes, we are told, they just show that they never had true faith in the first place. Regardless of how convinced they were of their own heart and demonstrated this by the fruit in their life, the contemporary Reformed position is that they did not have real faith, but only something closely resembling it. Now, I have no problem with the suggestion that people can deceive themselves or fail to have faith truly penetrate their heart despite years in the church (human psychology is complicated, after all) but to claim that this is always the case is a claim that undercuts our very ability to cling to faith for our assurance. If we are so blind to our own heart that we cannot tell whether we have true faith or not, then how can faith ever be the basis for assurance? It can’t, contrary to what scripture teaches.

The Augustinian (and traditional Reformed) position avoids the errors of both Arminians and contemporary Reformed by distinguishing between election and regeneration (which is by faith). A person is elect before they ever have faith and remains elect even if they (really) apostatize temporarily. What election guarantees is that God will bring it about that that person will have sufficient faith at the end of their life. Although the norm is to preserve them in one stretch of faith from beginning to end, this need not always happen. On the other hand, a person who happens to have faith for some stretch of their life may well not be among the elect, in which case they will apostatize prior to and until the end. For the duration they had faith they had the assurance of future salvation, but once they apostatized they lost this assurance—not because God  failed to fulfill his promises, but because the person  stopped pursuing them.

At the end of the day, election and perseverance are things which happen “behind the scenes” whereas faith and regeneration are things we have more direct access to. We can know directly whether we have faith but only indirectly that we are elect. The doctrine of the perseverance of the saints assures us that God’s election of individuals is certain—that is, if we are among the elect then we will persevere. The promises of God assure us that our salvation is sure so long as we abide in Christ by faith—that is, if we persevere in faith then we will be saved (and were among the elect the whole time). Neither of these assure us that we are in fact among the elect—that is, we will persevere in the faith we currently have. We can at best be assured of our election indirectly and probabilistically, based on the likelihood of us persevering. This third kind of assurance just doesn’t seem to be of much interest to the biblical authors.

Conclusion

My suggestion in this post has been that the “five points of Calvinism” are an imperfect realization of the Reformed convictions about predestination. The desires for a high view of divine sovereignty and agency and the recognition of the inevitability and depth of human sin are both well-placed, but we need to analyze them within a robust philosophical theology and in the context of those church fathers who have come before us. As a Thomist I of course think that Aquinas is exemplary on both counts, but I’m open to alternatives. The key takeaway for Reformed theology is that we should aim to be more flexible in our approach, and more focused on our fundamental convictions.

---

[1] David Haines, “Thomas Aquinas on Total Depravity and the Noetic Effects of Sin,” Themelios 48.2.

[2] Loraine Boettner, The Reformed Doctrine of Predestination (1991), pp. 238–43; Guillaume Bignon, Excusing Sinners and Blaming God: A Calvinist Assessment of Determinism, Moral Responsibility, and Divine Involvement in Evil (2017), pp. 214–22.

[3] See especially my third post on omni-instrumentality and my post “God causes evil actions without causing the evil in actions.”

[4] See Terrance Thiessen’s post, “For whom did Christ die?”

[5] Roland Elliott, “The Anastatic Theory of Atonement,” Theologica 9.1 (2025).

[6] For a more detailed discussion of this point, see Eleonore Stump, “Sanctification, Hardening of the Heart, and Frankfurt’s Concept of Free Will”, in John Martin Fischer & Mark Ravizza, Perspectives on moral responsibility (1993) pp. 211-234.

[7] I’ve discussed this more general notion of goodness numerous times on this blog. Take a look specifically at my posts, “How Aristotle starts the Nicomachean Ethics”, “Aristotle and the egoist worry” (part 1, part 2), and “Aristotle’s function argument”.