Philosophical Theology

An entailment of the Reformed doctrine of limited atonement is p:

If Christ died for S, then S will be saved.

Therefore, if p is true, S’ salvation is in some sense guaranteed by Christ’s death on behalf of S.

Davenant Hypothetical Universalism (HU) rejects p by affirming that (a) Christ died for all and (b) not all will be saved. Consequently, even in spite of the cliché: “sufficient for all, efficient for the elect”, HU actually denies the logical sufficiency of the cross!

The force of the argument is, He who willed and ordained that Christ the Mediator should sustain the wrath of God due to the sins not of certain persons, but of the whole human race, He willed that this passion of Christ should be a remedy applicable to the human race, that is, to each and every man, and not only to certain individual persons; supreme power being nevertheless left to himself, and full liberty of dispensing and applying this infinite merit according to the secret good pleasure of his will.

Death of Christ

Consequently, HU is not Reformed in any sense of the word because it denies the very heart of Reformed theology with respect to the finished work of Christ:

To all those for whom Christ hath purchased redemption, he doth certainly and effectually apply and communicate the same; making intercession for them, and revealing unto them, in and by the Word, the mysteries of salvation; effectually persuading them by his Spirit to believe and obey, and governing their hearts by his Word and Spirit; overcoming all their enemies by his almighty power and wisdom, in such manner, and ways, as are most consonant to his wonderful and unsearchable dispensation.

Westminster Confession of Faith 8.8

Not only does HU deny the Reformed gospel principle that Christ died only for the elect and in doing so secured only their salvation, HU further denies confessional Reformed theology by alleging that it is truly possible that a non-elect adult freely (and savingly) believes:

The death of Christ is applicable to any man living, because the condition of faith and repentance is possible to any living person, the secret decree of predestination or preterition in no wise hindering or confining this power either on the part of God, or on the part of men. They act, therefore, with little consideration who endeavour, by the decrees of secret election and preterition, to overthrow the universality of the death of Christ, which pertains to any persons whatsoever according to the tenor of the evangelical covenant.

Davenant, Loc. Cit.

Sound philosophical theology:

As previously noted, Reformed theology entails p:

If Christ died for S, then S will be saved.

That entails an instance of the necessity of the consequence (not consequent) given that Christ’s death for S is a sufficient condition for S’ salvation. Consequently, given p, the cross (on behalf of S) is not only a sufficient condition for S’ salvation – S’ salvation is a necessary condition for Christ’s substitutionary death on behalf of S. Therefore, if S does not become saved, then Christ did not die for S! HU denies these basic and essential Reformed tenets.

Reconciling redemption accomplished with the application of redemption:

If the only freedom that can account for moral responsibility and do justice to the Reformed doctrine of total depravity is compatibilist freedom, then it is not possible for a non-elect living person to believe freely and responsibly unless it is also possible for God to incline a person’s will to Christ after he has determined not to do so. Consequently, unless God can deny himself by acting contrary to his decree, HU (perhaps unwittingly) consigns itself to an Arminian notion of incompatibilist freedom, which entails an implicit denial of the need for effectual grace to cause one to believe freely.*

Philosophically speaking, incompatibilism, which is not a Reformed position, does allow for the possibility of a non-elect person to believe by exercising libertarian free will. Consequently, HU implies libertarian freedom given HU’s axiom that “the condition of faith and repentance is possible to any living person.” It’s crucially noteworthy that because HU refers to any living person, the possibility of a non-elect person repenting and believing does not refer to uninstantiated essences but instead to actual persons, who from a Reformed perspective are subject to divinely predetermined choices, which from a distinctly Reformed perspective cannot be otherwise given the causal nature of the creative decree. In other words, we’re no longer considering possible counterfactuals that are objects of God’s natural knowledge, but rather we’re considering determinatively true counterfactuals that are freely known logically posterior to the creative decree.

An ironic twist to HU:

Only incompatibilism makes room for the possibility of saving faith for the non-elect. Or as Davenant would have it, the decree of predestination “is in no wise hindering or confining this power either on the part of God, or on the part of men.”

What must be grasped is that libertarian freedom cuts two ways. If all have libertarian freedom and it is truly possible that a non-elect living person freely believes the gospel, then it would seem equally possible that an elect adult could forever freely reject the gospel. In which case, saving faith can be uncaused and according to resistible grace. Consequently, HU cannot consistently maintain that (a) Christ’s death must be efficacious for the elect if they’re to be saved given (b) the implicit possibility of an elect person not believing according to libertarian freedom. In other words, the libertarian freedom that HU (unwittingly) requires for the possibility of the non-elect to believe and be saved ends up undermining the need for the causality of effectual grace that must be determined upon the free will of anyone who would believe.

If we try to introduce the necessary condition of irresistible grace for any living person to believe, then the possibility of any non-elect living person freely believing is confounded along with HU. That’s because the non-elect, after having been passed over in the eternal decree, cannot possibly be the recipients of irresistible grace, which in Reformed theology is a particular bestowal determined upon the elect alone, which is, also, necessary for the causal efficacy of the cross.

Only in an arbitrary last-ditch effort might one insist that the non-elect could (but never would) be saved according to libertarian freedom and resistible grace, yet only the elect can be caused to believe freely by the efficacious work of the Holy Spirit. Obviously, that would entail two different ways of appropriating the benefits of Christ, a monstrosity indeed.

In summary: (a) the metaphysical assumptions entailed by the possibility of any living person freely believing undermine (b) the causal guarantee that any living person will certainly believe.** HU has no answers to these things.

Possible worlds:

Possible world semantics can be useful here. There is no possible world (or consistent state of affairs of how things might have been) in which a decree that entails causal divine determinism affords room for the possibility of a living non-elect free moral agent repenting and believing in his depravity. Consequently, given a Reformed understanding of the decree and the condition of man’s unregenerate heart, Davenant’s “possibility” is an impossibility for non-elect living people! Stated differently, Davenant’s “possibility” cannot be made actual by God, which makes it an impossibility by Reformed standards. Notwithstanding, HU “possibility” does comport with a non-determinist (Arminian) view of the decree, but at the high cost of undermining the impossibility of an elect living person not being saved. Yet even from a libertarian freedom perspective, the “possibility” of a decreed non-elect person freely repenting and believing, although claimed to be philosophically “possible” by sophisticated Molinists – is nonetheless an infeasibility by their own standards, which is to say, it too is not actualizable! (None of this is controversial among those who share a common understanding of modal concepts and contemporary philosophical taxonomy, whether Calvinists or not.)

It’s possible not to recognize one’s own position:

It’s worth interjecting at this time that one can hold to HU while not recognizing its logical implications, if not also vehemently objecting to them. But even affirming truth elsewhere is insufficient to overcome incoherence, especially when foundational premises undermine stated truths. This is not uncommon. For instance and similarly Molinism posits libertarian free will while rejecting the implications of such autonomous freedom, which is a contingent God who is not exhaustively omniscient.

HU is not confessional in any sense of the word:

Not only does HU explicitly deny a core principle of the Reformed doctrine of limited atonement – namely that Christ’s death on behalf of S is a sufficient condition for S’ salvation – it cannot consistently maintain its own claim upon the Reformed teaching that Christ’s death is effectual for the elect. Nor can HU do justice to the alleged possibility of the salvation of all the living without affirming Arminianism. So, not only does HU deny the Reformed teaching that all for whom Christ died will be saved – it is also inconsistent within its own watered-down version of the atonement by undermining the Reformed teaching of total depravity and how Christ’s death can be effectual for the salvation of the elect.***

HU does not warrant “honorable mention” by Reformed standards:

Kevin DeYoung appears to have little use for the doctrine of HU, and I think his instincts are spot on, which isn’t surprising. However, like so many who find the doctrine of little use (or even incoherent!), DeYoung still finds HU within the bounds of Reformed orthodoxy. (I must disagree lest we allow the system of confessionally Reformed orthodoxy to be internally incoherent. Why should we do that?)

Hypothetical universalism appears to do more for the Calvinist’s psyche than for the state of the non-elect. To be sure, hypothetical universalism—at least of the non-Amyraldian kind—has not been considered outside the bounds of Reformed orthodoxy, but this is owing to its congruence with stricter notions of particular redemption, not because of a marked departure from them.

Kevin DeYoung

Because HU must implicitly deny the necessity of irresistible grace in order to explicitly maintain the possibility of a non-elect living person believing, the (unintended) ramifications of HU entail a non-Reformed view of the extent of the fall while implicitly affirming a doctrine of resistible grace and libertarian freedom. All of that too disqualifies HU from being confessionally Reformed. But the additional (and ironic) rub is that given HU’s “possibility” of faith in the life of all persons, HU undermines the causal efficacy of the cross as it relates to the certain future-fruition of saving faith in the life of the elect. Therefore, there is no “congruence with stricter notions of particular redemption” but rather an implicit theological denial of it.

Finally, DeYoung states:

After all, as Lee Gatiss has pointed out, Calvinistic hypothetical universalism is, in the end, still a variant of limited atonement: Christ died effectually for the elect and only conditionally for the non-elect. The conditional intent for the non-elect is not in place of particular redemption for the elect (as in Arminianism), but in addition to or prior to this effectual atonement for those who will believe (For Us and For Our Salvation, 99).

Given all that HU denies and affirms, I fail to see how it can be a “variant” form of particular redemption or an “addition” to particular redemption. Instead, HU would “replace” particular redemption with a work of the cross that is not a sufficient condition for the salvation of all those for whom Christ died. Moreover, HU introduces, by implication, several non-Reformed soteriological notions, and we haven’t even begun to consider HU’s implicit impact on orthodox theology proper, for instance the inseparable operations of divine persons and exhaustive omniscience.

“Reformed” but not confessional or ordainable:

It has become fashionable to identify Reformed theology by the theology of individuals as opposed to the official confessional documents that were produced after much debate. By this measure of Reformed theology, must we consider historical premillennialism “Reformed” or a doctrine that purports an infusion of grace for the elect at the font? Or did certain doctrines win the day, as it were?

It should be apparent, if we were to allow the unfiltered theology of individual Westminster Divines to define Reformed theology for us, our confession would not be a fair representation of Reformed theology! Our confession could become contra-Reformed depending upon the particular theologian to which one might appeal for doctrinal precedent. Consequently, true Reformed theology cannot be defined by particular Divines but instead must be elucidated by the doctrinal standards they produced.

At the time of the formulation of the confessions, we would expect that vastly different views on baptismal regeneration, faith, atonement etc. would have been unavoidable. In other words, conflicting doctrine would have needed to co-exist among the delegates because they hadn’t yet debated, voted and formulated a Confession to rule out certain doctrines! But now that we have a defined Confession, Reformed denominations operate in a vastly different context than the seventeenth century with respect to defined confessional boundaries for church officers, especially on things like baptismal regeneration and HU. 

Closing:

It’s indubitable that HU stands in stark contrast to the Westminster standards. Consequently, if HU is Reformed, then so are all non-confessional doctrines that were represented at the time of the formulation of the Westminster standards. Historical madness is what it is!

Lest we forget, the Westminster standards teach:

To all those for whom Christ hath purchased redemption, he doth certainly and effectually apply and communicate the same; making intercession for them, and revealing unto them, in and by the Word, the mysteries of salvation; effectually persuading them by his Spirit to believe and obey, and governing their hearts by his Word and Spirit…

WCF 8.8

Consequently, HU explicitly denies the Reformed tenet that the work of Christ will be applied to all for whom Christ died. That alone makes HU un-confessional.

Moreover, because HU posits the possibility of the salvation of the non-elect, HU unwittingly undermines the need for an effectual appropriation of the salvific benefits of the cross. And lastly, by maintaining the possibility of the non-elect appropriating Christ’s benefits by repenting and believing, HU implicitly denies the depths of man’s depravity and the Reformed doctrine of reprobation as it relates to the efficacious nature of the divine decree:

The rest of mankind God was pleased, according to the un-searchable counsel of his own will, whereby he extendeth or with-holdeth mercy, as he pleaseth, for the glory of his sovereign power over his creatures, to pass by; and to ordain them to dishonor and wrath for their sin, to the praise of his glorious justice.”

WCF 3.7

Plain and simple, HU implicitly denies not just a few but several core features of Reformed soteriology, and cannot consistently make good on its own claim upon the efficacy of Christ’s death for the elect. But why should that be surprising given the intricate nature and interdependence of Reformed Christian doctrines? Given a system of doctrine, shouldn’t we expect that to be off in one area of doctrine will ripple through to other areas of doctrine, especially as we try to reconcile redemption accomplished with redemption applied?

Now of course and as alluded to above, when one holds to any faulty position, (whatever it is – it does not need to be theological), the necessary implications of the position are often not recognized as being part an parcel to the original position. In fact, the logical trajectory of one’s axioms can entail consequences that are even abhorred by the one holding to their sufficient antecedents! But I suppose that’s just a hallmark of not recognizing the implicit incoherence of one’s own foundational premises. Consequently, the initial reaction can be to scream foul, or think that the original position is being unfairly misrepresented or simply misunderstood. The impasse is all the more pronounced when opposing views operate with a vastly different grasp of modal concepts and the tagging of contemporary philosophical terms.

The only question left is, what is unconfessional Reformed doctrine? It is in that light we ask that God might be pleased to grant confessional Reformed churches humble discernment in holding fast to their rich theological tradition, especially with respect to the glory of God in the person and work of Christ.****

*By distinguishing determinism from necessitarianism, causal divine determinism allows for contingency that is rooted in the will of God. However, given the causality that the decree contemplates, contingent truths become causally secure and, therefore, impossible to be otherwise, though they’re not absolutely necessary. (Richard Muller misses modal distinctions as James Anderson and Paul Manata have shown.)

**From a Reformed perspective God is free to determine the salvation of anyone. However, that sovereign latitude pertains to the decree. After the decree God is not free to deny himself and effect saving faith in the life of the non-elect. Consequently, given the creative decree it's not possible for the non-elect to believe unless they possess libertarian freedom, in which case salvific efficacy (along with foreknowledge from a Reformed and open theist perspective!) becomes impossible. Yet if a Reformed understanding of efficacy is in play, then there is no libertarian freedom under consideration, in which case God's prior determination not to cause all to believe makes it functionally impossible for all living persons to believe.

As I’ve written elsewhere, yet with relevance to the present discussion due to the non-Reformed assumptions of HU, certain modal concepts are useful:

“Although from a Molinist perspective foreknowledge presupposes decree (just like it does with theological determinism), the decree does not limit possibilities like it does with theological determinism; that’s because within Molinism the decree does not include free choices that are determined to be causally secure. Hence, the contingency of any free choice carries through to the actual world, as there remains the possibility of freely choosing otherwise within Molinism (even after the decree and consequent foreknowledge). In short, from a Molinist perspective God’s foreknowledge exerts no causal influence upon free choices.”

“Yet from a Reformed perspective, the contingency does not carry through in the same sense as it does with Molinism. That’s because from a consistently Reformed perspective contingencies relate to possibilities that God can determine, which are not dependent upon man’s libertarian freedom. Consequently, given a Reformed view of the decree, contingent truths become, according to God’s creative decree, causally secure, having been freely determined by God. (These truths do not become necessarily true, precisely because God can determine and actualize another contingent outcome 
from identical states of affairs.)”

As I recently wrote:

We must attend to logical moments and modal concepts. The possibilities that are an object of natural knowledge aren't applicable in the way Muller and Davenant think. Given the creative decree, we're no longer dealing strictly with divine pre-volitional possibilities, which are known according to God's natural knowledge, but rather we are considering counterfactual truth, which from a theological determinist perspective are objects of divine free knowledge of the creative decree. God is the truth maker of all truth bearing propositional counterfactuals of creaturely freedom (CCFs). They're grounded in God's free determination. Therefore, they're causally secure and cannot be otherwise in the decreed world because it's impossible for God to act contrary to his decree. 

Whereas with incompatibilism, CCFs can be otherwise after the decree without God contradicting himself; that's because CCFs are not causally secure given libertarian freedom. They're objects of middle knowledge. In Arminianism and Calvinism free choices are contingently true, but from a Reformed perspective their truth values are determined and, therefore, causally secure and can no longer be otherwise. Within Molinism (for instance), the truth values are pre-volitional to the decree and aren't casually established and can be otherwise.

***Clichés that obfuscate:

It is inescapable, the atonement is a matter of divine intent, which is equivocally obscure within Davenant’s hypothetical universalism.

Little clichés like Christ’s death is “sufficient for all, efficient for the elect” have their place but not in rigorous systematic theology. A sufficient condition entails a state of affairs that if met ensures another state of affairs. In that sense, the cliché literally implies actual universalism. Sufficient and efficient become functionally indistinguishable and the cliché, tautological. Yet if “sufficient for all” is merely intended to convey that Christ’s death has infinite value and would save you if you believe, then redemption becomes necessary (and not sufficient!) for saving faith, which isn’t very interesting in this particular context. That one cannot have saving faith without the work (or future promise) of the cross, although true, doesn’t advance the atonement discussion. Accordingly, how are we not back to election and irresistible grace, which are anything but sufficient for all! The historia salutis and ordo salutis must coincide.

****Fence posts on Reformed doctrine:

Regarding confessional status, any (a) direct contradiction of the confession or (b) extra-confessional teaching that leads to intra-confessional doctrinal contradiction may be confidently rejected for being un-confessional even if not explicitly refuted by the church's standards (regardless if a delegate to the assembly held the view in question). Otherwise, we unnecessarily introduce incoherence and confusion into our system of doctrine. Also, any doctrine that is theologically derivable from other confessional doctrines must be considered no less confessional than the doctrines from which they come.

Otherwise, we would not be able to refute on confessional grounds doctrinal claims that oppose the necessary implications of our own theology!