Two Wills in God: Not Just for Calvinists

| | Comments (32)

In this discussion, one of the commenters makes the following argument against Reformed views of divine providence:

On a related topic, I still don't quite get Reformed theology. God desires all to repent, but He doesn't desire all to repent. How does one believe something one is incapable of understanding? It's like saying I "believe" that the round plate before me is also a square, as if my saying it makes it so.

What follows is an expansion of my response in the comments there.

What the commenter has hit on is a formal contradiction, at least if no fallacious equivocation is going on. If the word "desire" is being used in the same sense, then the statement that God desires all to repent and the statement that God does not desire all to repent do indeed result in a formal contradictiom.

But there's no problem if the two uses of "desire" are in fact different senses in which God desires. That is in fact what the Reformed view means by both claims, but the basic distinction required to take such a view isn't limited to Reformed theology. Any adequate response to the problem of evil needs something like that, as has been known at least since Thomas Aquinas. (At least you need something like this if you want to avoid open theism, but I've long thought open theism doesn't really have the resources to respond to the problem of evil anyway, because it can't guarantee a full victory over evil, not to mention being overkill, so that becomes a null option.)

You need to have some sense in which God wants to evil to happen if God in any sense knowingly allows it, so those with models of divine sovereignty that are more commonly associated with Wesleyan or Arminian theology will need to say the same thing this commenter is criticizing. God allows something rather than preventing it. Why? Perhaps the reason is because God thinks human freedom is more desirable than the desire to prevent that particular evil. You need not be a Calvinist to appeal to this sort of thing. But you better not say that God wants it to happen in every sense. God certainly disapproves of the evil, and wouldn't desire it if it weren't for whatever issue led God to allow the evil.

Once you have that distinction between desiring for its own sake and desiring for some other reason, when for its own sake God would want it removed, you have exactly the thing you're criticizing. God can desire something and not desire the same thing.

I would say that Arminians need to say this even about the salvation of non-believers if they want to avoid universalism. If anyone dies in their sins and goes to hell as a result, then God will be desiring that fate for them given their rejection of him, even if God desired them to repent and thus avoid that fate. So God both desires it and desires that it not happen, even with Arminianism. Only an open theist or a universalist can avoid saying something like that about these cases, and I don't think either can avoid saying it entirely. Even to allow one bit of evil or even the risk of it is a tradeoff in one sense, with God choosing one thing over another that would be good and desirable if all things were equal.

[cross-posted at Evangel]


Was it your intention simply to point out that even the non-Reformed have concepts of the will of God that allows them to speak of it in two different ways? I ask because the distinction you go on to explain, while valid, is not actually the one that the Reformed draw. In speaking of cases in which God allows evil rather than preventing it, you rightly point out that even those who hold to Arminian theology must admit some sense in which God desires this outcome. The possible reason you give for God doing this- “because God thinks human freedom is more desirable than the desire to prevent that particular evil”- is not in line with the Reformed viewpoint, which holds to a compatibilist and not a libertarian view of the will. God is perfectly capable of preventing evil without violating anyone’s freedom.

Your last paragraph talks about Arminian distinctions (and perhaps you meant to so limit it): God desires that all men repent and go to heaven, but, given the fact that some don’t, there is another sense in which he desires that this group does not go to heaven. I’ve done something similar myself. I desire that every last one my students get passing grades. But, for those who have not done the required work, I also desire that they fail; moreover, my grading reflects the latter desire. In both cases, only the last will is an effectual will. The first will is that of preference. So, God would prefer that all people repent, but, if some don’t, it is in some sense his desire to condemn them. The Reformed view is to draw a distinction between God’s prescriptive will and God’s effectual will. The first refers to his commands. It is his desire that everyone cooperate with these, but oftentimes we don’t. The second refers to his eternal decrees. This is the sense in which, if God wills something, it must inevitably come to pass.

Concerning the reconciliation of the statements, “God desires all to repent,” and, “God does not desire all to repent,” I think that the validity of your solution has to do with which verse one has in mind for the first proposition. It works with Acts 17:30, “The times of ignorance God overlooked, but now he commands all men everywhere to repent” (ESV). While it is God’s prescriptive will that everyone repent, it is not his effectual will; otherwise, everyone would repent.

This is not the first passage that comes to my mind, though. Instead, I think of I Timothy 2:3,4, “This is good, and it is pleasing in the sight of God our Savior, who desires all people to be saved and to come to the knowledge of the truth.” In this case, a Reformed reconciliation of the statements would not focus on different meanings of “desire” but on different meanings of “all.” Two things in the context point to this being an all without distinction rather than an all without exception. First, after asking that prayers be made for all people, Paul begins to classify these people- “for kings and all who are in high positions.” While the list is my no means exhaustive, it does point to the conclusion that he is urging prayer for all categories of people and not necessarily for every single person. Had Paul meant all without exception, further classification would have been redundant. The second contextual clue is, from a Reformed perspective, the stronger of the two. It is found in v. 6, which states of Jesus that “he gave himself as a ransom for all.” Given the Reformed doctrine of a definite atonement, if Christ gave himself as a ransom for all without exception, then Paul must be teaching universalism. While it is God’s effectual will that all classes of people repent, it is not his effectual will that every single individual repent.

All I'm saying in this post is that the particular formulation presented as an argument against Calvinism is a formulation that non-Calvinists will want to affirm as well, and I gave some examples. I wasn't intended to give the impression that Calvinists would handle all those examples in the same way.

That being said, I disagree with your interpretation of the passages you give. The I Tim reference is taken the way you discuss it by George Knight, but plenty of Calvinists don't assume definite atonement requires no sense in which the atonement is universally open to all. Limited or definite atonement means the atonement actually applies only to those who are saved. It doesn't mean there's no general offer to all, including the reprobate. I've seen Reformed types argue that there can be no potentiality in God and thus no offer in any sense to the unelect, but I consider such a view hyper-Calvinist and not mainstream Reformed.

For more detailed arguments, see my posts on Limited Atonement and Is There Potentiality in God?, John Piper's Are There Two Wills in God?, and John Hendryx's Is It God's Desire for All to Be Saved?

I don’t deny that there is a general offer of the gospel to everyone or that we are to proclaim such an offer without discrimination. I do deny that this is in any way related to either the intent or the extent of the atonement. For the sake of argument though, let me grant your own view that, while the atonement is limited in the more fundamental sense, there is an equally true yet less fundamental sense in which it was for everyone. Even so, I do not see how the use of the word “ransom” (antilutron) in I Timothy 2:6 can allow the passage to apply to anything other than the more fundamental sense. This is about payment actually rendered for individuals such that requiring them to pay it again in the form of eternal punishment, as could be the case under the less fundamental sense, would be unjust. I have no problem with the idea of potentiality in God, and I am granting the idea of a less fundamental, universal sense to the atonement. It’s just that, in this particular passage, that’s not what Paul’s talking about. Consequently, given my belief that there exist individuals who are eternally punished for their own sin, I must conclude that the use of the word “all” in this passage does not have universal intent.

Piper does a good job of giving a scriptural basis for distinguishing between God’s “will of command” and his “will of decree.” However, when he tries to reconcile the two, he equivocates. He writes of Arminians and Calvinists, “Both can say that God wills for all to be saved.” For neither side can this be the will of decree; otherwise, everyone would be saved. It must, according to the parameters that Piper has established, be the will of command. But what Arminian is going to accept this limitation? God’s will for the salvation of everyone is not so much based on his law, they will say, as on his love. This effectively introduces a third category- “will of passion.” Piper accepts this uncritically, and I may know why. It has to do with a criticism that I’ve seen you level against him. You have written, “John Piper reduces all of God's emotions to God's desire for promoting his own glory” (I’ve also read your other posts on the matter). In general, I think I’m more sympathetic than you to Piper’s views on God’s glory. While I affirm that God’s glory is a motivation behind everything that he has decreed, I don’t believe that it’s the only motivation (even though I think that all other divine motivations ultimately result in God being glorified). I’ve questioned your charge that Piper is reductionistic in this regard. Until now.

With the Arminians, Piper affirms that “God loves the world with a deep compassion that desires the salvation of all men” (in context, he means all without exception). I will admit a sense in which God loves everyone because they have been created in his image, but I cannot grant the same universality to God’s greater, salvific love. The latter is an efficacious love that, accomplishes the salvation of those on whom it is bestowed. To Paul’s question in Romans 8 on who can separate us from this love, Piper would have respond, “Nothing, except for the manifestation of the full range of God’s glory in wrath and mercy.”

I don't see it. If I say that I'm giving a ransom payment to a kidnapper for the return of anyone kidnapped who wants to return home, and some choose not to come home, then it doesn't make it a ransom only for the ones who return home. This is so even if the children who stay are doing so on a prearranged mission to spy on the kidnappers. The ransom that's offered is potentially for any of them to be returned. The deal made with the kidnappers is that it would cover any who accept it. If some don't, then they don't actually get covered by it. But it's potentially for them too. So the atonement is actually restricted only to those who receive it but potentially open to all. I Tim 2 is, in my mind, almost certainly speaking of this potential for all, given that its background is in God's statement in Ezekiel of not wanting any to perish but wanting all to come to him. This repentance never came. Ezekiel's message fell on dead ears, as had Jeremiah's, and God had told both of them that this would be so.

So I have to side with Carson, Grudem, Mounce, Piper, and Packer on this verse. (I believe Calvin is with you on this verse but ended up rejecting it later when he commented on a similar verse later in the Bible, perhaps in II Peter). John Owen, Charles Hodge, and Robert Dabney all have insisted on potentiality (in a compatibilist sense) in the atonement, and I believe R.C. Sproul has as well.

I'm not sure just from what you've said how Piper's reductionism on God's glory relates to the third category of will. You're right that Calvinists and Arminians have to think of the (at least) two wills in different ways. It's not the same set of two wills, for sure. But my point is merely against the claim that Calvinism is dead in the water for asserting this contradiction, because the same sentence would be affirmed by an Arminian who doesn't take one of the minority approaches I mentioned (open theism, universalism, etc.). I wasn't saying the Calvinist and Arminian mean the same thing by the expression, just that Arminians also have to say the same apparent contradiction and have to use some method of resolving it to avoid a genuine contradiction, which means someone can't just assert that Calvinists accept a contradiction without probing to see if the Calvinist resolution is any worse than the Arminian one.

You're right that God's effective-for-salvation love is not directed at the reprobate. But that doesn't mean there's no love related to salvation directed at them. It's been a long time since I've looked at Carson's Difficult Doctrine of the Love of God, but I think he deals with this somewhere in there. One of his overarching themes in that book is that there are quite a number of senses of God's love in scripture, and we too easily conflate them. I think it speaks of more than one kind of salvific love. Maybe Carson doesn't actually get into this particular point, but I think he might. At any rate, I think we can speak of salvific love of the sort that does lead to God saving people, and I think we can think of salvific love in the sense that God would, other things being equal, save everyone (but other things aren't equal, and God has reasons that are stronger for not saving everyone than the reasons motivated purely by God's mercy that (absent justice and perhaps many other things) would lead God to save everyone.


Are you saying Piper agrees with your explanation of "ransom" in 1 Tim 2:4? (I haven't read him on this so I am really asking.) If he is agreeing, how would that fit with the gymnastics involved in his explanation of whether we can tell an unbeliever that Christ died for them in gospel presentations? Working from memory, Piper says that we can't tell them Christ died for them but rather we tell them that Christ died for sinners and if their affections are moved by the fact that God would die for sinners, then they can believe that God died for them. (He also has parallel answers on assurance issues.) It doesn't seem to me that your understanding of limited atonement and 1 Tim 2:4 would require such contortions.

It's been a while since I've read it closely, but I thought he was offering a view similar to the one I've outlined here. Even from looking quickly over it this time around, I got the impression he was much closer to what I've been saying than, say, to George W. Knight.

Given what I know about the contortions he goes through in answering the issue of evangelism and assurance, I would be surprised if he handled the "ransom" issue the way you do. But, again, I have not read his actual handling of 1 Tim 2:4.

It has been so long since I have looked at this issue and I have never really worked out my own views on the issues to where I feel like I have a solid position. Limited atonement is the only part of Reformed soteriology that I am not sure about, partly because I have heard versions like you present and then the traditional Owen version and I am not sure which Reformed theology is stating.

I can think of at least two different senses in which one could acknowledge potentiality. It depends upon if there is any sense in which the atonement is effectual at the time of the cross. As I understand those like MacArthur/Sproul, part of what the cross/resurrection achieves is the gift of faith. The value of the cross is such that God could have counted it for more sinners, but God counted it for a specific group of people at the time of the cross. So there is already an element of particularity at the time of the cross. So while the value had the potential to save more, God counted it effectual for a specific group, to earn them the gift of faith, at the time of the cross.

To use your analogy, a person makes a deal with kidnappers to give 1 million dollars to get back 5 of 10 people. But let's say the person knew that the kidnappers would have given back all 10 for the same 1 million. So the ransom had the potential for all, but the person specifically limited it to 5. I think MacArthur/Sproul/Owen are saying something like this. While the atonement could have saved more, God accounted it to pay for the sins of certain people at the cross; there is specificity at the cross in a certain effectual aspect.

I tend towards your view, but I have to work through it a lot more.

Given your professed Calvinism, the kidnapper analogy is surprising. It reads more like a blueprint for the Arminian view. The kidnapper is the one to whom the ransom is paid. I’ll say, since this is what I believe and many Arminians agree, that this is God (in full wrath mode). The ransom payment is the atonement, the kidnapped are the lost, and being returned home is salvation. Choosing to be returned is faith. In themselves, there’s nothing objectionable to the elements of the analogy. There is in the way it’s been constructed. Choosing to be returned is that condition upon which the payment is applied. In the Reformed doctrine of the atonement though, the gift of faith is among those blessings that are purchased. Basically, everyone who’s been kidnapped has Stockholm syndrome. None of them wants to go home. Not only do you need to pay the kidnapper, but you need to supply him with medication that will change the minds of the kidnapped people along with a list of which of them are to receive it. The ransom is intended only for those on the list, and the list is not exhaustive.

As it stands, your version of the analogy either eliminates or severely compromises at least four of the five points of Calvinism. Since the victims are capable of choosing whether or not to be returned home after payment has been made, there is either no total depravity, or we should assume the Wesleyan doctrine of prevenient grace. Election, if present at all, is based on the choice to be returned home and is, consequently, conditional. Among non-universalists, there is no disagreement over which of the kidnap victims are actually covered by the payment. The question of limited atonement is one of divine intent. If the ransom offered is intended for the potential return of anyone, whether or not they actually return, then the atonement is not limited (or, more accurately, it is limited in a different way). If it is possible for the choice to stay or be returned home to determine whether or not the payment is actual for any one person, then there is no irresistible grace. This, too, is among those blessings purchased by the atonement and must, therefore, follow its application.

My comments on the different ways in which Calvinists and Arminians think of the will of God were aimed at Piper. I understand what you were doing. After reading up on the issue to refresh my memory, I do believe that the third category of the will is legitimate. God can will something in the sense of wishing that it were so, even if this will is neither a decree nor a command. That being said, Piper had eliminated this option. He was trying to reconcile the first two. When he writes that both Calvinists and Arminians can say that God wills for all to be saved, he is not speaking in terms of mere formal equivalency. It is essential to his point that both sides mean the same thing. But no Arminian would understand this to refer to any of the two wills to which Piper has restricted his discussion. He ends up agreeing with the Arminian view on this point. Although I have reconsidered my position and have no problem, in itself, with the common ground he establishes concerning what kind of will is at play, I still maintain that Piper has not been consistent within his own parameters. And I wondered what could might explain the inconsistency.

Piper goes beyond agreeing that God desires the salvation of everyone to agreeing on the proximate cause of that desire; namely, God’s love. Moreover, he does not disagree with the Arminian as to the type or extent of this love. If he did, his discussion of how the Calvinist accounts for the fact that some are unsaved despite God’s will on the matter would have taken a different direction. As he explains it, both sides account for the unsaved by positing something that overrides the will/love of God. For the Arminian, this is God’s refusal to violate free will. For the Calvinist, it is God’s desire to manifest the full extent of his glory. With Piper, I will affirm that God’s primary motivation for everything he does is to bring glory to himself. But I wouldn’t make it the proximate cause for explaining the existence of the non-elect. Instead, I would put this in the fact that God has an effective love for the elect that he does not have for the reprobate (even though he may love them in other ways). God’s desire for his own glory would explain why some are so loved while others are not; it would not cancel the effects of a love that would otherwise save.

I would say that the effective sense and the potential sense are both present at the time of the cross. It's hard to make it time-specific unless you think God changes his mind, and that doesn't fit well with Reformed theology.

I was just trying to find an analogy where there's something both effectual and potential and covers a broader group insofar as it is potential than it does insofar as it's effectual. You can mess around with the details to make it more similar to Calvinist soteriology. Compatibilists say this kind of thing all the time in philosophy, though, so I would think compatibilist Calvinists should be willing to do so also. If we're all predetermined by physical causes, then a compatibilist will say that we're still free. Some of them will say we're free despite not having the ability to do otherwise, but I think the right compatibilist thing to say (and this is David Lewis' view, and he's as close as you get to orthodoxy among contemporary metaphysicians) is that we do have the ability to do otherwise. The Stoics insisted that we do as well, and they were the first systematically-defended compatibilists in Western philosophy.

The way I've always seen Phil 2:12-13 is that we have as full a role as can be in our own salvation even if it's also only due to God's grace that we can do so. It's monergistic in the Reformation sense, because we could do none of it alone. It's not as if God supplies some of the work while we fill out the rest. But it's compatibilist in the philosopher's sense because we actually do the things God requires for salvation, i.e. to repent and believe. It's not something God does in a way that bypasses our own moral choices and character. God works through our own decision-making and our own understanding of our situation, our moral thought-processes, our recognition of our sin, our choice to repent and follow him, and our continued choices to live in his will.

It's a true statement that we would do none of it if it weren't in God's plan for us to do it. It's also a true statement that if we didn't do it, it must not have been in God's plan. Explanatorily, God's plan is prior. So to say that there's potentiality is not to say that God's plan has gaps for us to fill in. But one issue that matters here, I think is which approach is correct about the logical order of God's decrees. Supralapsarians have the choice to save some and allow others to be reprobate before God's reasoning about what will justify saving some while rejecting others. Infralapsarians have the logical basis for rejecting some and saving others before the actual conclusion that that's going to happen. So in the logical order of God's choices, there's a point at which the plan is open to all who may respond, and then you eventually reach a point at which that's not the case in terms of some being chosen to respond and others not. So one level of potentiality is that God's choice that there will be some saved and some not comes before God's choice to save some particular people and reject others. That's a metaphysical difference.

Another level of potentiality is epistemological. It applies to the command to seek the evangelization of all. The fact that this command is grounded in God's desire to save all shows that it's not merely epistemological (and I don't think the I Tim 2 trick discussed above will apply to all such passages, even if, contrary to my view, it's the right way to take I Tim 2). In other words, I don't think we can just say that for all we know people are elect, so for all we know the atonement applies to them. Potential application is more than that. It's actually true that, if they were to respond positively to the gospel in repentance and belief, they would be saved.

This is where the compatibilist analysis is crucial. For them to respond positively, something would have to be different in God's plan. But the sense in which it's open to them is that if they did respond they'd be saved. The atonement thus covers them potentially but not actually, if it's someone who never will respond. What makes it true in English to say that it's open to them and that it's possible for them to be covered is the counterfactual claim that if they did believe they'd be saved. That is a true counterfactual, so the atonement potentially covers them.

I am not sure that I see time specification with the cross any more difficult for a timeless/immutable God then there being time indexing in his single timeless decree.

So God timelessly decreed that at time T (time of the cross) the atonement of the elect's sin is accomplished, securing the gift of faith and all other gifts of salvation for them.

At time T1 the atonement of my sin was either accomplished or not accomlished already. If it wasn't then how can I change that? It isn't just that to believe something would have to change God's plan. It is that who Christ effectively died for is already a past event and thus there is nothing that can be done to change that at T1. So if I believe at T1 then I won't be saved since my believing can't change the past event of Christ's death on the cross that accomplished atonement for the elect. (I get the sense there is something similar to Ockhamism going on here, but I can't put my finger on it.)

At time T1 the atonement of my sin was either accomplished or not accomlished already

But isn't this what the already/not-yet tension throughout scripture insists is not true? In one sense it's accomplished at the cross, and in another it's not accomplished yet. What I'm saying here is one aspect of the already/not-yet.

I never said anything about changing anything. In the possible world where Hitler had a last-minute genuine conversion due to a work of God's grace regenerating him, he doesn't change anything about the future. In that world that's the only future. He just does what brings that future about. In this world we all do what brings the actual future about. To say that it's possible that I could have worn flip-flops instead of sneakers tonight doesn't mean that if I'd worn the flip-flops I would have changed anything. In that world where I wear the flip-flops, it was never true that I wouldn't wear them. This isn't about time travel with past-changing occurrences (not that I think even those are coherent). It's about how things would have gone if the world had been different in some important way. In the world where Hitler was saved from a deathbed conversion, it was always in God's providential plan for Hitler to be saved. But that doesn't mean it would have been impossible in that world for Hitler to die in his sins, since the actual world would have been a possible world with an entirely different but possible causal history for Hitler to lead to his different decision.

I'm not sure about Ockhamism, but you're right that evaluating these counterfactuals requires getting into the metaphysics that underlies discussions of truth about the future.

If something is monergistic in the Reformation sense, it means that we have nothing to do with it at all. God’s grace enabling us to do something that, otherwise, we could not do alone, defines synergism. When Paul tells the Philippians to work out their own salvation, he is speaking of their sanctification, which, properly speaking, is synergistic. He is not speaking of the grounds of their justification, which is monergistic. The principle of sola fide is violated not only when works are added to faith, but also when faith itself is considered to be a work. This is done, albeit unwittingly, by Arminians when they make faith a condition for receiving the benefits of the atonement rather than making it one of those benefits.

The consequent of the counterfactual, “If they did believe they’d be saved,” is too broad. Being “saved” encompasses everything from foreknowledge to glorification. The validity of any one of these elements as a consequence for the antecedent depends on its position in the ordo salutis. One may rightly say, “If they did believe they’d be justified,” or, “they’d be sanctified,” or “glorified.” Each of these follows faith. It is not right, however, to say, “If they did believe they’d be foreknown,” or “they’d be predestinated,” or “called.” Each of these precedes faith. Whether or not your implied counterfactual, “If they did believe they’d be covered by the atonement,” is, in fact, true depends on the relation of faith to the application of the atonement. In Arminian sotereology, the atonement is applied if and only if faith has been exercised. In Reformed sotereology, faith is given if and only if the atonement has been applied. Inasmuch as you have professed an alignment with Reformed (or, at least, Calvinistic) theology, your counterfactual is false.

The doctrine of limited atonement has to do with divine intent. In order for the atonement to potentially cover more than it actually covers, you need something that could have been otherwise and that actually is antecedent to it. For instance, God could have intended to save someone else. This intent is a part of his decree and, barring outright contradiction, he could have decreed anything. We could say, “If God had decreed it, then x,” where x stands for every outcome in all possible worlds. Surely, however, this is far too expansive a category to say with any meaning beyond a philosophical technicality that x is, in some sense, true. Moreover, it is highly unlikely that the human authors of scripture were thinking in such abstract terms. Paul’s concept of how many are included among the ransomed would be informed, not by his concept of God’s pre-decree options, but by his concept of God’s purposeful election.

Let me try an analogy for what I think Sproul (etc.) are saying. Think of the atonement as like purchasing tickets for people to go on the journey of salvation. The tickets are purchased at the time of the cross, but people go on the journey on the basis of these tickets at different times and it only comes to completion at the eschaton. Faith is acceptance of the ticket to go on the journey. (On the Reformed story the ticket ensures its acceptance, but this isn't relevant for my point; the issue is not compatibilism.) At the time of the cross the tickets are purchased for specific people (and thus not purchased for other people). Although, the cross was of such value that it could have purchased tickets for all people (hence the sense in which they could affirm potentiality), it was only purchased for specific people.

So at time T a ticket was purchased for me or it wasn't. Let's say I am not one of the elect. This means that at time T, on the cross, no ticket was purchased for me. At time T1 there is nothing for me to accept since there is no ticket for me to accept since it wasn't purchased for me. Thus, it is false at T1 that if I believe then I will go on the journey, because my faith only takes me on the journey if I have a ticket. But there is no ticket for me.

This is why I was talking about the issue of changing the past even though you were not. The conditional "If I believe at T1 then I would go on the journey" is only true if a ticket was purchased for me at T. But if no ticket was purchased for me at the cross at time T then the conditional is false. The only way the conditional could be true would be if I could change the purchasing of the tickets at T, but I can't do that at T1.

The already/not-yet tension concerns that which is actual in the present and that which will be actualized in the future. It does not deal with mere potential, but with that which will definitely come to pass. So, in the case of the atonement, it would exist for someone who, though elect, had not yet been regenerated. Neither side of the tension would ever be true for the non-elect.

Kevin, Warfield would disagree with the claim that monergism requires hard determinism rather than compatibilism. Compatibilism is still a monergistic view. To get synergism you need libertarianism. Because compatibilism is still as deterministic as hard determinism, it's still considered monergism. You don't have to be a hyper-Calvinist to be a monergist. In the comments here, Phil Johnson agrees.

Here is Jonathan Edwards:

As to that objection against the doctrine, which I have endeavoured to prove, that it makes men no more than mere machines; I would say, that notwithstanding this doctrine, man is entirely, perfectly, and unspeakably different from a mere machine, in that he has reason and understanding, and has a faculty of will, and is so capable of volition and choice; and in that his will is guided by the dictates or views of his understanding; and in that his external actions and behavior, and in many respects also his thoughts, and the exercises of his mind, are subject to his will; so that he has liberty to act according to his choice, and do what he pleases; and, by means of these things, is capable of moral habits and moral acts, such inclinations and actions, as, according to the common sense of mankind, are worthy of praise, esteem, love, and reward; or, on the contrary, of disesteem, detestation, indignation, and punishment.

In these things is all the difference from mere machines, as to liberty and agency, that would be any perfection, dignity, or privilege, in any respect; all the difference that can be desired, and all that can be conceived of; and indeed all that the pretensions of the Arminians themselves come to, as they are forced often to explain themselves. (Though their explications overthrow and abolish the things asserted, and pretended to be explained,) For they are forced to explain a self-determining power of will, by a power in the soul to determine as it chooses or wills; which comes to no more than this, that a man has a power of choosing, and in many instances, can do as he chooses,-- which is quite a different thing from that contradiction, his having power of choosing his first act of choice in the case.

Now I don't think you want to claim that Edwards is a synergist, but he quite clearly says that we do the things we do of our own will and choice, that we do them and aren't just passive recipients of God's actions. Compatibilism like that of Edward and what I'm defending has been called monergism by those who prefer to speak in such terms (I don't, because it comes across to philosophers as hard determinism even though that was never what was meant).

In that case, Warfield and I would be on the same side. It was never my intent to come across as a hard determinist. The difference between monergism and synergism is that between a single worker and co-workers. It is not about which theory of the will accounts for the actions of the human co-worker. Although often connected, it is not the case that you need libertarianism to get synergism. They are distinct concepts, as are compatibilism and monergism. By explaining the “work out your own salvation” passage in Philippians as monergistic “because we could do none of it alone,” you have contradicted yourself. The implication of not doing it alone is that we cooperate; however, the etymological origins of “cooperation” (Latin) and “synergy” (Greek) translate one another. Even though each is favored in specific contexts, they are essentially the same word. Moreover, the contrary of your explanation, which would be “because we can do it alone”, would not describe the contrary of monergism but would describe another kind of monergism. In Pelagianism (as opposed to semi-Pelagianism) man is the only necessary agent in salvation. Nothing prevents God from cooperating, but his assistance is not required.

Philippians 2:12-13 is a synergistic passage. It does not contradict Reformed monergism because it does not address those areas in which we cannot cooperate. Paul writes his letter to “the saints in Christ Jesus who are at Philippi.” Other than their future resurrection and glorification, all divine monergistic work for their salvation had already been accomplished and applied. Paul is talking about progressive sanctification; nothing else. There is no real quarrel between monergists and synergists concerning the synergistic nature of our sanctification. For that matter, a synergist may consistently hold that there are many aspects of our salvation that God must accomplish all by himself. The dispute is not one of absolute categories: God does everything vs. We help out in everything. Rather, it comes down to the issue of our justification. For the Reformed monergist, our sanctification is not a factor for God to declare us righteous. For the synergist, it is. This is what it means to be justified by faith alone.

I think we need to distinguish between "monergism" as a general slogan/synonym for Calvinism and the more specific usage of monergism in Reformed theology. For instance, in Reformed theology, regeneration is monergistic, but sanctification has a cooperative aspect.

Keith, think about what would be true in the closest possible world where the antecedent is true. We're talking about you as if you're reprobate. Then in the actual world you have no ticket. But in the closest possible world where you believe, it's a world with the same rules from God, and you have an explanation for your belief, namely that you're elect in that world. So in the closest possible world where you believe, you do have a ticket. It doesn't matter that you don't have one in the actual world. In the closest possible world where you do believe, you do have a ticket.

You're right that the already/not-yet tension only applies to believers when it comes to salvation. We're already saved in one sense, and we're not yet saved in another. So too it applies to the unelect in the opposite way. They're already damned in one sense but are not yet damned in another sense. The only reason I introduced this is to show that there is language that I know you to accept that is like what you described as a contradiction. You said it's either happened or hasn't, and the already/not-yet shows that that's not the case. One way that it can both have happened and not have happened is the already/not-yet.

What I'm saying here is another way something can have both happened and not have happened. What has happened is the actual atonement for the actual elect. What hasn't happened (and in this case won't) is the actual atonement for the non-elect. But does that prohibit us from describing the potential atonement as able to cover all if they turned out to be elect? Of course not.

It sounds like you're saying the monergism Calvinists hold to is hard determinism for justification but compatibilism for sanctification but then insist that you're not being a hard determinist.

What hard determinism says is that determinism makes us such that we don't make choices. We're not involved. Things just happen to us, and we're passively moved along. What you're describing monergism as is exactly that. Compatiblists, on the other hand, insist that our choices are our own choices and that we cause what we do, even though that causing is ultimately explainable ultimately only in terms of God's causing of us to do it.

Synergism as I've always understood it is that God does some of the work, and we need to complete it. Synergism is not compatibilism, which has God doing all the work but with it also true that we do the work. Synergism is more like holes in the work of God that we need to fill in. Compatibilism is a two-level approach that allows us to speak of ourselves as causes on a level of description that fits best with our experiences. Hard determinism removes the human level of description entirely. These are separate issues, but your resistance to potentiality-talk strikes me as resistance to compatibilist talk of the ability to do somethim

To clarify according to Steve’s request, I am not using monergism as a general synonym for Calvinism. Instead, I am taking it in the specific sense that it applies to regeneration and whatever may immediately follow from that. I concede that, as it is applied to regeneration, monergism is a matter of hard determinism. It just happens to us, and we are passive. This is, nevertheless, entirely irrelevant as far as theories of the will are concerned. It’s not that if either libertarianism or compatibilism obtained, then we would have a choice in the matter. It’s that there is absolutely no choice to be made. You might as well question the compatibilism of someone who disagrees that we had choice in being conceived, or born, or in who our ancestors are. Even libertarians will admit that there are some things that simply happen to us and that have nothing to do with our free will.

A compatibilistic concept of the will is essential in order for orthodox monergism to be coherent. Under compatibilism, it is not the will itself that is free. The will is a natural faculty of the person who possesses it and is subject to the inclinations of that individual. We are free agents in that we can do whatever we want to (within the same natural constraints that libertarians concede). In our fallen state, we are totally depraved and dead in our sins. None of us wants to repent; consequently, none of can will to do so. Even so, our free agency is intact. In order for anyone to choose God, the basic underlying nature of that individual must be radically altered. This is where regeneration comes in. Regeneration is that which moves a free agent to cooperate in a compatibilistic sense with the preceptive will of God. It, therefore, begs the question to make regeneration subject to our will in any sense. It is, of necessity, something that happens to us as passive recipients.

Regeneration precedes faith. Insofar as faith is understood as our choice to believe God, this can be explained by the relationship between regeneration and the will. Perhaps this is the case; yet, if so, faith, as something that we do of our own volition, is a work. Personally, I don’t think that this is the proper understanding of faith. Believing anything, such as my belief while writing this that I’m sitting at the dining room table, is not a matter of choice but of perception. As long as I remain here, and my senses are functioning properly, I cannot believe otherwise. In the same way, regeneration brings to life certain senses that are capable of perceiving the objective realities of the heavenly realm. We neither choose to have faith nor to exercise faith. It is God’s gift to us, and all who posses it are declared righteous, i.e., justified.

You distinguished between, “God does some of the work, and we need to complete it,” and, “God doing all the work but with it also true that we do the work.” The distinction is valid; nevertheless, both are contrary to monergism. Prior to being declared righteous, any work that we may try to contribute would be deemed sinful. God would never accept it towards the goal of our salvation. Afterwards, the situation changes. Because of our regeneration, we are capable of willing to do what God wants from us. Because of our justification, God accepts what we do. This is our sanctification, which is synergistic (in the sense that you have labeled “compatibilism”).

Back to the already/not-yet tension. While you concede that it applies only to believers when it comes to salvation, my point was that it is not language that would explain away what Keith described as a contradiction. The biblical concept of already/not-yet is always a tension between what actually is and what actually will be. It does not cover vacuous potentiality. If the sins of a certain number of people were atoned for on the cross, then they either were or they weren’t. There are no two ways about it. Having said that, it still possible to speak of the atonement as potential. For anyone x, where x is a person whose sins have not actually been atoned for, but we don’t know that, the following is true: if x turns out to be elect, then the atonement covers the sins of x. In this case, it is true that the atonement actually does not apply, and it is also true that the atonement potentially does apply. But this only works because of our ignorance. If God were to reveal to us the actual status of x’s election, then the antecedent would not be true and the consequent wouldn’t necessarily follow. [Actually, the syllogism should be modified. Election is both sufficient and necessary to inclusion in the atonement, so the antecedent should begin with “if and only if.” In this way, a denial of the antecedent is a valid negation of the consequent.]

Since we do not know who is elect the counterfactual applies to anyone, and so the atonement potentially covers anyone. But this potentiality is still limited. We cannot move to saying, “The atonement potentially covers everyone,” without a different syllogism. If and only if everyone turns out to be elect, then the atonement covers everyone. The form is valid, but the antecedent is not true. Even if we cannot know who is excluded, we know for a fact that God has not elected everyone. Consequently, the atonement does not potentially cover everyone. A potential universal atonement cannot be the case.

But you have to distinguish between regeneration and faith in one sense. Regeneration does just happen to you. But faith, while not volitional, isn't merely passive. A tree can't have faith. A robot can't have faith. It takes desires being transformed to have faith, and even if the regeneration that causes that is a passively-received act of God it doesn't follow that the faith itself is like hard determinism. Once I'm regenerated, I don't just find myself believing and trusting. My desires become reordered so that I'm wanting things in a way much more in line with their actual worth. My beliefs begin to be transformed to fit more closely with reality on spiritual matters. Those are things that run through my own intellectual processes in the same way as the things you're saying are synergistic in sanctification. Compatibilism applies to anything done by an agent that's caused by God, and these are things done by agents. Agents don't regenerate themselves, but agents believe and trust.

So when I ask if it was genuinely possible for someone to repent and believe, which are actions an agent does, I have to say that it's not possible given the persons unregenerate state but possible not given that state, and the compatibilist analysis applies when the second sense is what's relevant. In evangelistic contexts, it's the second sense, because no one involved, including the person, knows whether regeneration will occur. I'm not denying monergism in the sense you have in mind about regeneration. I'm just saying the potentiality that matters for the atonement depends on compatibilist possibility for the things that would follow regeneration if it were to occur.

The biblical concept of already/not-yet is always a tension between what actually is and what actually will be. It does not cover vacuous potentiality.

Isn't that special pleading? I'm taking a whole bunch of passages to say exactly that, and the reason you say they can't mean that is that the scriptures don't use language that way. Well they do if my view is correct, and they don't if your view is correct. Pointing out that the passages that don't speak the way I'm talking don't speak the way I'm talking doesn't show very much.

As with all cases of possibility, the key is the context of discussion. I've never denied that, when considering all the facts, the potentiality is removed. This is just like time travel cases, where if you consider all the facts you should know that you can't kill your paternal grandfather before your father was conceived. But if you just consider the facts in front of you about the time you
are at, there's nothing that has to stop you. So in one sense you can kill him, and in another sense you can't. The same goes for the atonement. I've been saying all along that in one sense it's limited, the sense that matters when you consider all the facts. We can tell someone what God did for them, and we mean that God did it for humanity as a whole, with them as a potential beneficiary given that the offer itself doesn't specify which individual people are elect to respond positively.

If we knew who was elect, and the people we were talking to knew we were elect, it's sort of like the person who time-travels and tries to change the past deliberately. But we don't, so it's more like the case where we time-travel to somewhere when we don't know what already happened. We could do something or not, and we don't know which we will do until we do it. So we'll describe all manner of possibilities, and it's right to do so.

Once we know which will happen, we can still describe them as possibilities, though. We'll just know that none of them are actual futures. Similarly, if we know that someone was not elect (which we won't until we know they died in their sins) we might lament that they refused God's genuine offer, because if they'd responded (which would have taken God's regenerative act first) they would have been saved. We'd have to trust that in the end there are good reasons for God leading to that result. But we could still see it as something worth lamenting, and God in scripture seems to be shown doing exactly that. If you remove the potentiality-talk once complete knowledge is available, it strikes me as contrary to how scripture actually presents these scenarios.

My definition of “compatibilism” is narrower than the way you’re using it here. Since I limit it to a theory of the will, then I wouldn’t apply it to something that I admit to being non-volitional. There is a sense in which faith grows in much the way that you have described it. Since we have some control over this growth by willing those things which may promote or hinder it, this is very much a part of sanctification. Yet, the ability to grow in faith does not imply that there is insufficient faith for justification at the moment of regeneration. And while faith may involve something that we do, insofar as believing is a verb, it does not follow that the possession of faith is something that could have been otherwise. Seeing, hearing, breathing, heart-beating, aging, etc. are all things that we do. As long as the relevant systems are functioning properly, these things necessarily occur. Justifying faith is the immediate and necessary consequence of regeneration and so does not exist as a compatibilist potentiality thereafter.

It may be possible to derive a tension between actuality and mere potential from several scriptural passages, but, unless the concept is poorly named, this would be something distinct from the already/not-yet. If something is not actually actual, such as the potential application of the atonement for anyone for whom Christ did not die, then there is no sense in which it can be already. And “not-yet” implies that, while something is not actual now, it will be actual in the future. You first mentioned this in response to Keith’s statement, “At time T1 the atonement of my sin was either accomplished or not accomplished already.” A contradictory statement is one which presupposes that x can be both x and non-x at the same time and in the same relationship. The specification of time T1 pretty much locks down the specific time, so now the question is one of relationship. Concerning the atonement, this was specified as “accomplished.” You claim that at T1 there is a sense in which it is accomplished and sense in which it is not accomplished yet. Unless there is some meaningful sense in which it is not accomplished yet after time T, this is still a contradiction. Now it possible for the atonement to be accomplished at T1 but not applied until T1+n, and at T-1 there is a sense in which the atonement is potentially accomplished for all (universally) but actually accomplished for none. However, at time T, for any given person x, Christ either did die for x or he did not.

Consider the possible interpretations for I Timothy 2:7, “[Jesus] gave himself as a ransom for all.” This verse is about what Jesus did to accomplish the atonement (the application thereof is the work of the Holy Spirit). The question of extent is this: for whom did Christ intend to die? If universal intent is assumed, then I believe we’re both agreed that all inferences other than potential universal /actual limited application are untrue. The question is whether this inference is true. The relation between universal intent and potential universal application is valid and unobjectionable. It’s the actual limited application that constitutes the problem. The purpose of the atonement was the full satisfaction of the justice of God, which demands death in payment for sin. When Christ finished his atoning work, this was sufficient unto its purpose: he had done everything necessary to pay for the sins of those for whom he intended to die, and God had accepted the payment. This is what it means for the atonement to be accomplished. Hell exists as an alternate method to satisfy divine justice. Since the atonement accomplished its purpose of satisfying God’s justice regarding the sins of all those for whom it was intended, since it was intended for every human being without exception, then God would be a most unjust tyrant if he were to allow anyone into hell. However, God is not unjust, and people do go to hell. Barring the existence of an unknown and unconsidered consequent of sufficient difference, and having eliminated all others, I conclude that the antecedent of universal intent is unsubstantiated. Jesus did not give himself as a ransom for all without exception.

Justifying faith is the immediate and necessary consequence of regeneration and so does not exist as a compatibilist potentiality thereafter.

You say faith can grow, but one of the ways it grows seems to me to count against this. There are people whose process of coming to faith strikes me as taking time, with parts of the process occurring before faith is fully present, but it takes a gracious act of God for the early parts, and thus regeneration is already taking place. It's not as if there's one complete act of regeneration and an immediate consequence of full faith.

As for compatibilist potentiality after regeneration, I'm not sure what you're referring to. That would be a separate issue, i.e. whether we should speak of the compatibilist possibility of falling away for believers, whereas here we're speaking of the compatibilist possibility of belief for those who aren't regenerate. We might answer them the same way as each other, but they're separate issues.

As I said before, I wasn't claiming this to be an example of the already/not-yet. I was saying that the already/not-yet is an example of the same category that this belongs to, where something that seems to be a formal contradiction is nonetheless true because the two things are meant in different senses, and some of the language is similar between the two cases.

I would insist that the question "For whom did Christ intend to die?" has the same ambiguity. His intention, on one fundamental level, was to save some, and the particular ones he intended to save were already selected. But the will of God contains logical priority in the order of the decrees, and the decree that some will be saved (de dicto) precedes the decree to save some particular people (de re). Because of that ordering, there's a sense in which God deemed the sacrifice of Christ potentially before narrowing who would receive it, and in that sense at one logical point in the decrees of God there's an openness about its recipients. For God as manifested in time in Christ, you can say that all this was already decided before the foundation of the world, but God outside time is simultaneous with every event in time, and thus the reasoning about this is going on during every moment in time, including at the cross. So the openness and the fixedness are both present at every time, or else God isn't omnipresent throughout time.

The sense in which the atonement is limited is exactly what justifies most of what you say at the end. Since the sense in which it is open or universal is not in the way that you're discussing there, it doesn't have the consequence that some person's sin is actually covered while they go to hell. So it doesn't bring the injustice charge against God.

Inasmuch as it is always possible to grow further in faith, I doubt that anyone in this life has full faith. The relevant question to justification is whether or not they have sufficient faith. I will agree that the process of coming to faith can appear to take time, but there are plausible ways to account for this other than saying that regeneration and justifying faith are themselves a process. Either the elements of noticia and assensus are accounted for by common grace (for these can be present even in those who never come to saving faith), and then regeneration results in fiducia; or, regeneration first results in fiducia, and then the subject is gradually made aware that this has already occurred by going through the process of noticia and assensus. Regeneration itself is not a process. It does not describe the presence of life, but the passing from death to life. This is a one-time, complete act of God.

Even though the distinction between de dicto and de re is valid regarding possible interpretations of the question concerning for whom Christ intended to die, they don’t necessarily describe distinct decrees. If they do, then the order you place them in would follow. It is just as plausible, however, that the decree to save was personalized from the beginning. There is a de re decree with no need for a de dicto follow-up. For instance, when I order a turkey sandwich at Subway, I want it with some bread, cheese, and vegetables. Now, my desire for bread and cheese is de dicto: I’m not going to choose what kind until I’m actually ordering. Vegetables are another matter. I want lettuce, spinach, green peppers, and pickles. This desire is de re. Since they offer more vegetables than that, it turns out that I only want some of the vegetables. If they only offered what I wanted in the first place, then I would want all of the vegetables. That I want some rather than all of the vegetables is entirely accidental. If there is but one decree of salvation, namely, the de re decree, then there is no sense in which God deems Christ’s sacrifice potentially.

Let’s say you’re right about the decrees, though, that the first is a de dicto decree for an unspecified some. “Some” is logically compatible with “all,” which you may mean. Essentially, this would be Amyraldism. Also known as hypothetical universalism, this system places the decree of atonement before the decree of election. After decreeing an atonement sufficient for all, to be applied on condition of faith, God notes that no one in their natural state will actually exercise faith. And so he decrees to elect particular persons whom he will call with irresistible grace leading to their belief. Arminianism differs from this in that the decree of election follows that of a, now resistible, call. The potentiality of the atonement in Amyraldism is found only in its application, not in its accomplishment, and this subjects it to the same criticism I outlined in my last response. If God fails to apply what has been accomplished for someone and, instead, demands satisfaction from them, then he is unjust.

Or, one might understand the de dicto decree to mean that God decrees to save some but not all. In this case, both “Some S are P” and “Some S are not P” would be true. Since this precedes the de re decree, then any S could potentially be P. It would be true of anyone we met that God potentially died for them. However, because we are also acknowledging that some S are not P, there can be no warrant for moving from the particular, “God has potentially died for anyone,” to the universal, “God has potentially died for everyone.” It would, in fact, be impossible, for the contradiction of “some S are not P” is not “no S is P” but “all S are P.”

I wasn't intending this in an Amyraldian way. I'm an infralapsarian on those issues.

What I meant by the order of decrees issue is that "Jesus' death potentially covers everyone" is read in a de dicto way from the de dicto decree, such that God has willed, logically prior to anyone's being selected for salvation, that each person will receive an offer and either be saved or not, but that nothing in the de dicto decree that some will be saved interferes with anyone's salvation. It takes a later decree for that.

I'm not following your last sentence. It sounds to me that the English sentence "X potentially covers anyone" is not logically equivalent to "X potentially covers everyone". For example, there might be a lottery, of which there will be exactly three winners, but anyone can play and potentially win. So the potentiality for anyone doesn't entail the possibility that all will be saved. OK, I'm with you up to that. But how does that lead to a problem with what I'm saying? The potential offer that matters is that each person is presented with a message -- repent and be saved, or fail to do so and remain lost. For that person, the atonement potentially covers them for salvation if they do believe. If it turns out they don't, then it didn't actually cover them. One sense in which it's potential is that, logically between the two decrees as I've presented it, the decree of the salvation message in general doesn't yet specify whether this person will be saved. Another sense in which it's potential is that the person, for all we know, is elect and will be saved. Both of them seem to me to apply in such a case. The fact that it might not be possible for all to be saved, for whatever reason, doesn't interfere with that, not for a compatibilist.

Unrelated to these last remarks, I had a chance to look through some commentaries on I John 2:2 and related verses in I John, and I found four commentators whom I believe are all 5-point Calvinists who take the sort of view I'm defending here (John Stott, Colin Kruse, Daniel Akin, Robert Yarbrough). I may be wrong on whether they're all Calvinists, but I'm pretty sure they all are. Stott describes it as "a universal pardon offered for (the sins of) the whole world and ... enjoyed by those who embrace it". Kruse uses the common phrase "sufficient to deal with the sins of the whole world, but ... not ... effective until people believe in him." Akin, I thought, was particularly helpful in giving more in the way of an actual argument. (The first two argued mostly against universalism on one side but said little against the hard-determinist interpretation.) He argues that "not only for ours but also for" indicates a contrast between the elect and the non-elect rather than between some locality of believers vs. another, which doesn't really fit the context of inside/outside. Yarbrough has a particularly detailed discussion, in which he insists that the view he thinks is implied by the text does not amount to Amyraldianism but is open to that but also open to other issues among the views that he thinks tend to be more systematic than John was being. He insists that the clear implication of the text is a wideness in the atonement in one sense and a particularity in another sense, and he gives a reference to someone who has argued that you find the same in Calvin (where his arguments in I Tim 2 and I John 2 are pretty much targeting universalism and aren't meant to deny the sort of view I'm defending here, which certain passages in Calvin clearly do support, in my view).

Did you mean to say, “Jesus' death potentially covers anyone” for the de dicto decree? If the decree is that it potentially covers everyone, then the de dicto/de re distinction is invalid. What is said (de dicto) just is the thing (de re). Since God's decree concerns whatever comes to pass, then, unless all those who haven't heard get a second chance after death, there was no decree that each person will receive an offer. I will acknowledge a decree that, due to their creation in the image of God, each person is eligible to receive an offer. Yet, even if God's decree had been such that each person actually did receive an offer, this would not be sufficient to imply that the atonement is, in some sense, universal. There are, to be sure, hyper-Calvinists who believe that the absence, in any sense, of a universal atonement implies that there is no obligation to offer the gospel indiscriminately, if at all. This is, nonetheless, a case of drawing faulty conclusions and not of having false premises.

If anyone is presented with the gospel, then the atonement potentially covers that person. This potentiality is not conditioned upon acceptance. To say, “the atonement potentially covers them for salvation if they do believe,” implies that belief is the condition for potentiality and nothing more. Those who actually did believe would find themselves in a situation in which the atonement may or may not cover them in reality. Instead, whether or not someone believes determines whether the atonement, which is potentially theirs, turns out to actually have been accomplished for them and is, therefore, actually applied to them. Anyway, if this is close enough to what you meant, I agree with this kind of a potential atonement. I also agree in the sense that, for all we know, anyone to whom the gospel is presented is elect.

So far, I haven't been convinced that there are distinct de dicto and de re decrees of salvation. Nevertheless, your conclusion that “the decree of the salvation message in general doesn't yet specify whether this person will be saved” may still be true, so I will consider it. In this formulation, the decree of a general salvation message precedes the decree of election. Since the ordering of the decrees is logical, then any given decree cannot logically presuppose a decree that follows it. You have presented this relation between the decrees of a general salvation message and election as another way in which it [the atonement] is potential. Ordinarily, I would see nothing inconsistent in the idea that a salvation message entails the potential atonement of anyone who hears it. However, since you are an infralapsarian, you also believe that the decree of election precedes the decree of atonement (only the Amyraldians and Arminians reverse the order). And if a precedes b, and b precedes c, then c follows a. The decree of a general salvation message would simply be that people should receive a salvation message, whatever that might be. The message, at the point of the decree, would be devoid of any content related to the atonement. Even if you were to adopt an ordering that allowed the salvation message to presuppose the atonement, it would only be in the sense of a potential atonement for anyone. This is the case for each of the ways in which you have demonstrated that the atonement is potential: the atonement is made potential for anyone.

The problem this presents with what you're saying is that, while you have successfully demonstrated how the atonement is potentially for anyone, this does nothing whatever to support your claim concerning I Timothy 2:6 that it should be interpreted according to a less fundamental sense in which the atonement is potentially for everyone. As you have stated, “'X potentially covers anyone' is not logically equivalent to 'X potentially covers everyone'”. I would argue, though, that you have gone further than simply failing to demonstrate the validity of your interpretation. God has decreed that the atonement is actually for some, but not all. [I believe that you would identify this limited atonement as the more fundamental sense.] To the extent that we do not know who these are, or to the extent that a de dicto decree for some precedes a de re decree specifying which some, then the atonement is potentially for anyone; i.e., for some. By the same reasoning from the intent of the atonement, that it is actually for some, but not all, it follows that the atonement is potentially not for some. Once this has been granted, then the idea that the atonement is potentially for everyone is not possible. Particular negation contradicts universal affirmation.

What I meant by the de dicto decree is a decree that says that it potentially covers all but does not specify who will actually be covered. As far as this decree goes, there's no de re decree of any individual yet whether the person will be saved. That comes in a later decree, according to the model I'm working with. That would then have de re decrees for each individual that they will be saved or not. What makes the first one de dicto is that it's a decree that some be saved that will be offered to all without it being a decree about which individuals those some will be. There's no de re component.

It's potential in two ways. One way that it's potential is that it's offered to all. The other way that it's potential is that this particular decree hasn't yet gotten to any specification of who will be saved. As far as the decree that God wants some saved goes, it's open to everyone. You need the later decree predestining individuals de re to get the particular/definite/limited atonement, and that decree lacks the potentiality of the first decree. But when we speak of God's desire for people to be saved, what we're referring to is that earlier decree. So it can be true to say that God desires all to be saved. That's how I was thinking the de dicto/de re distinction was going to make a difference here.

As I saw it, infralapsarianism is silent on when the de re decree for which individuals to be elected takes place in the ordering. There's a de dicto decree that some will be elected before there's a decree of how the atonement will be achieved. But all that could precede the de re decree of who will be elected.

I'm still not following you on the anyone/everyone distinction. I'm agreeing that being potentially for everyone doesn't follow. But potential for everyone means simply that there's a possible world in which everyone believes. That may well not be so. What may be so, though, is that for every individual there's a possible world in which that person is saved. If so, then it's potentially for all. That's all I mean when I say it potentially covers everyone. Everyone falls under its potentiality.

Your second paragraph seems to me to capture my main argument. If this de dicto/de re thing with the order of decrees won't ultimately work out, I'd still insist on that. That's my main reason for thinking there's potentiality. I was just hoping I could also capture it in the de re/de dicto way with the order of decrees in a way consistent with infralapsarianism, which I've got a much stronger commitment to than I do to any of this stuff about de re and de dicto decrees. I'm not convinced yet that I can't do both. But I'm happy to fall back to what you say in that paragraph if it doesn't work out.

In your first explanation of the de dicto decree, you write, “it potentially covers all but does not specify who will actually be covered.” As it stands, this is a mere assertion. I have yet to see any sense in which the mere lack of specificity implies potentiality for all. Your second explanation, which is “a decree that some be saved that will be offered to all without it being a decree about which individuals those some will be,” is better. Both a de dicto and a a de re decree that some be saved would allow for potentiality for any given subset of humanity. The de dicto decree would even provide another basis for that potentiality. However, unless you can demonstrate just how it is that the de dicto decree does potentially cover all, then the distinction is superfluous. There is no logical necessity for God to go through a de dicto step.

Your introduction of the de dicto/de re distinction was in response to the question, “For whom did Christ intend to die?” It may yet be possible for you to demonstrate that there is a distinct and prior de dicto decree; however, whenever Calvinists (whether supra- or infralapsarian), Amyraldians, or Arminians refer to the placement of the decree of election, they, for substance, are referring to the de re decree. Calvinism distinguishes itself from Amyraldism precisely on the point that the issue of election has been settled such that, when the provision of salvation is decreed, it is only intended for those specified. A mere de dicto decree of election prior to the decree of atonement is nothing more than a place holder and, as such, is insufficient for maintaining an infralapsarian viewpoint. It is the positioning of the de re decree that matters.

Although you have agreed that being potentially for everyone does not follow, it appears that you're trying to get it in through the back door. I will not dispute the idea that, for each individual, there is a possible world in which that person is saved. Even so, a possible world cannot be considered piecemeal but must be taken in its entirety. For any two possible worlds, their distinguishing points contradict one another; consequently, combining them does not constitute a valid move. The only way, then, for the atonement to potentially cover everyone is if there is a possible world in which everyone is saved. So far, those points that you have identified as capturing your main argument for potentiality (points with which I concur) do not support such a conclusion.

In your first explanation of the de dicto decree, you write, “it potentially covers all but does not specify who will actually be covered.” As it stands, this is a mere assertion

No, it's not even that. It's a hypothesis to consider. I'm not asserting or endorsing this view, and I've given no significant argument for it.

I have yet to see any sense in which the mere lack of specificity implies potentiality for all.

It implies exactly the kind of potentiality that consists of there being no fact yet determined about which people go where. As far as what's been so far determined in the order of decrees:

[for all fallen human beings x, it's possible that x will be saved].

In other words, there is a decree of salvation that in God's mind is logically prior to who gets selected, and thus at that stage in the logic of it it's still not been narrowed to any subset of the entire group and thus is still open to all of them as far as God has so far reasoned.

As I said, I'm not sure this will work, and if it works I'm not sure it's compatible with infralapsarianism but might turn out to be yet another variation not captured by the existing views. I'm convinced that what I have in mind is not Amyraldian.

Now I think the other issue is separate. I do think that what it's sufficient for what I'm saying that there's a possible world for each person in which that person is saved. I don't think there has to be a possible world in which everyone is saved. I think we need to get clear on two things we might mean when we say that the atonement is potentially for all.

1. For all human beings x: it's possible that the atonement could cover x.
2. It's possible that: [for all human beings x, the atonement covers x].

The first statement says that all human beings could be covered, and thus it potentially covers all of them (but not collectively -- just each individual is potentially covered). This is what you mean by "the atonement potentially covers any". This is equivalent to saying that for each person, there's some possible world in which that person is saved.

The second statement says that there's some possible world in which every existing fallen human being in the actual world is saved. This is what you mean by "the atonement potentially covers all".

When I say that the atonement is potentially for all, I mean exactly the first one. I don't mean the second one. I don't see why you think the expression in English has to mean the second one, because it seems to me to be ambiguous between the two readings, and I mean the first one by it, and I think it's what the biblical authors mean when they say such things. So I'm not sure we're really disagreeing on the substance. I think our disagreement (in this part of the discussion, at least) is really over whether we can say that the atonement is potentially available to all and mean 1 rather than 2.

If I understand, then, we're both saying that the atonement potentially covers all distributively but not collectively. Fair enough, but I don't think this is what the biblical authors had in mind. What I mean by this is that, for any person x, it is possible that the atonement could be applied to x. The potentiality as to its specific application lies only in our ignorance of who the elect are. The accomplishment of the atonement, however, is already settled: Jesus either died for x, or he did not. I Timothy 2:6, “[Jesus] gave himself as a ransom for all,” is moving the question back from application to accomplishment. In order for this verse to fit into the 'potentially covers all' mold, the basis for the potentiality needs to be moved to a point logically prior to the accomplishment of the atonement. This is what you have been attempting with your exploration of the decrees, but I don't believe that it will work.

The extent of the atonement's accomplishment is found in the ordering of the decrees. If the decree of election comes before the decree of salvation, the atonement is limited. If the order is reversed, the atonement is universal. In neither case does potentiality enter the equation. If you try to switch the order from the standard Calvinist understanding, then an adequate defense of an actual limited atonement is not possible. I'm not denying that, for each person, there is a possible world in which the atonement was accomplished for that person. I just don't think that any of these merely possible worlds can be a function of what God has actually decreed. They are, instead, a function of what God could have decreed. Moreover, nothing apparent prevents one of these worlds from being one in which every existing person is saved. The idea that biblical truth could ever be expressed in terms of pre-decree possible worlds is unacceptable. Not all that is possible is true. If it were, I wouldn't be lying if I claimed that the two of us had never disagreed on anything; there is, after all, a possible world in which this is the case.

In a way, though, this is beside the point. Paul's statement, “who gave himself a ransom for all,” simply does not lend itself to potentiality language. He is talking about what Jesus actually accomplished in this world. Whoever the 'all' refers to, whether to every single individual or to all classes of individuals, it is to be understood collectively. It is the Reformed view that the accomplishment of the atonement for x is both sufficient and necessary to the application of the atonement to x. Since there are people to whom the atonement will never be applied, it follows that there are people for whom the atonement was never accomplished. In other words, Jesus did not give himself as a ransom for these people. I think that you would agree with this since you have professed belief in an actual limited atonement. Our disagreement is over the exegesis of this specific passage. In your view, the verse is about a potential atonement for all and is silent concerning any limitations on the actual atonement. In my view, this verse cannot be about a potential atonement but must be about the real thing. The word 'all' should then be understood in terms of whom the atonement actually covers and, consequently, cannot refer to every single individual.

It is not the case that the most natural reading of 'all' requires maximal inclusiveness. Sometimes it does, but this should be determined contextually rather than automatically. A pervasive theme throughout the New Testament is that Jesus came to save, not just Israel, but all nations. On more than one occasion, Paul either defends or explains his mission to the Gentiles. He does so in I Timothy 2:7, “For this I was appointed a preacher and an apostle (I am telling the truth, I am not lying), a teacher of the Gentiles in faith and truth” (ESV). A common first century assumption of the Jewish church was that the Messiah was exclusively theirs. He had come to atone for their sins and those of no one else. Paul's words expand what Christ accomplished on the cross to include the Gentiles. Although, this expansion does allow us to infer the potential application of the atonement to all individuals, something which would not be true if salvation were limited to the Jews, it does not change what Paul says. He is talking about the actual accomplishment of the atonement for all nations, Jew and Gentile alike. There is no contextual warrant to press the 'all' to include every individual within these nations.

Leave a comment


    The Parablemen are: , , and .



Books I'm Reading

Fiction I've Finished Recently

Non-Fiction I've Finished Recently

Books I've Been Referring To

I've Been Listening To

Games I've Been Playing

Other Stuff


    thinking blogger
    thinking blogger

    Dr. Seuss Pro

    Search or read the Bible

    Example: John 1 or love one another (ESV)

  • Link Policy
Powered by Movable Type 5.04