This article is available in Korean.

I begin with a confession. For I must confess here at the outset that I am now utterly confident in the exegetical case for a universalist reading of the Bible as a whole, and I am also persuaded that the standard arguments against such a reading are wholly untenable. I say this because virtually every opposing argument I have encountered seems to me badly flawed in one of two ways: either it rests upon a rather elementary logical confusion, or it is easily reversed in a way I’ll try to explain in due course. But given that I am neither an expert in the languages of the Bible nor an expert in the historical background of its various documents, how can I so confidently (or even reasonably) reject so many arguments of so many distinguished scholars who read the Bible very differently than we Christian universalists do?

Behind that question lies the more basic question of just what it might mean to interpret the Bible as a whole. Some scholars (especially those of a more liberal persuasion) are understandably suspicious of any such effort; some would even dismiss it, though I do not, as an incoherent project. For as even religiously conservative scholars typically acknowledge, the Bible is not a single text with a single (human) author; neither is it, as the New Testament scholar Donald Hagner once put it, a systematic theology that has floated down from above like a balloon. It is instead a rich and diverse set of documents that appeal to the religious imagination in a variety of complex ways. Given the diversity of interests and writing styles of its various authors, the history of some of its documents, and the variety of perspectives it includes, a fertile imagination can almost always find a congenial way of putting things together. And for that reason alone, a theological interpretation of the Bible as a whole is as much an art, as much a work of the imagination, and as much a product of theological reasoning as it is of historical and linguistic study. Just as proponents of the geocentric theory of the solar system found many ways to account for the anomalous behavior of planets, so those who interpret the Bible from the perspective of a given system of theology inevitably find many ways to account for anomalous texts in the Bible.

Mind you, I would never minimize the contribution of Bible scholars to our understanding of the text. To the contrary, I have always tried never to challenge an expert on any specific point in his or her area of expertise. Nor is it necessary to do so, because large scale theological disputes, particularly those between Calvinists, Arminians, and universalists, almost never turn decisively upon scholarly minutia, and that is but one reason, perhaps, why one can find accomplished scholars in each of these theological camps. In any case, I shall now try to explain why a universalist reading of the Bible as a whole seems to me far more plausible than any competing reading of it.

‍

‍Saint Paul’s Apparent Universalism

‍That many Pauline texts at least appear, when taken in their own context, to teach an explicit universalism should be obvious to any careful reader of the New Testament. For again and again, Paul made explicit statements to the effect that God will eventually bring all things into subjection to Christ (1 Cor. 15:20-28) and reconcile all things in Christ (Col. 1:20) and bring justification and life to all persons through Christ (Rom. 5:12-21)” These statements are neither obscure nor incidental; indeed, the lengths to which some have gone to explain them away is itself a testimony to their clarity and power.

As a good illustration, consider more closely a single text, namely Romans 5:18 [1], and consider first its parallel structure:

Therefore just as one man’s trespass led to condemnation for
all [humans],
so one man’s act of righteousness leads to justification and life
for [them] all.

The whole point of such a parallel structure, so typical of Paul, is to identify a single group of individuals and to make two parallel statements about that single group of individuals, and the practical effect is therefore to eliminate any possibility of ambiguity. The very ones who came under condemnation, as a result of the first Adam’s act of disobedience, will eventually be brought to justification and life, as a result of the second Adam’s act of obedience. Or, as Paul put it in verse 19: the very ones who were constituted sinners, as a result of the first Adam’s act of disobedience, will eventually be constituted righteous, as a result of the second Adam’s act of obedience.  I do not know how Paul could have expressed himself any more clearly than that.

So does anything in the immediate context of Romans 5:18 justify the widespread supposition that Paul did not intend to say what his words in fact do say? One of the more popular arguments at this point appeals to 5:17, where Paul spoke of “those who receive the abundance of grace and the free gift of righteousness.” According to Douglas J. Moo, for example, “the deliberately worded v. 17, along with the persistent stress on faith as the means of achieving righteousness in 1:16-4:25, makes it clear that only certain people derive the benefits from Christ’s act of righteousness.” [2] Others, such as the New Testament scholar Ralph P. Martin, point to Paul’s use of the expression “the many” in verse 19, which Martin interprets as “a Semitic way of saying that ‘all’ are included with the assurance that ‘the all’ [included] are not a few in number.” [3]

I think it fair to say, however, that neither of these arguments is even remotely plausible. As for Moo’s appeal to 5:17 in an effort to limit the number of those who eventually receive justification and life, there are, I believe, two decisive objections. First, Moo never even considers those contexts in which Paul obviously used the verb “to receive” (lambanō) in a passive sense, and this has nothing to do, by the way, with the grammatical idea of the passive voice. When Paul declared, “Five times I have received [active voice] . . . the forty lashes minus one” (2 Cor. 11:24), we understand that he received these 39 lashes in the same passive sense that a boxer might receive severe blows to the head; and when he spoke of those who “have received [active voice] grace and apostleship to bring about the obedience of faith” (Rom 1:5), we again understand that such persons are the recipients of some divine action in the same passive way that a newborn baby receives life. Similarly, in Romans 5:18 and 19 Paul was comparing the effect of Christ’s one act of righteousness on the whole mass of humanity with the effect of Adam’s disobedience, pointing out in verses 15 and 17 that the latter is far greater, and far more extensive, than the former. So even though the Reformed New Testament scholar John Murray rejected altogether the universalist interpretation of our text, he nonetheless pointed out that the “word ‘receiving [in 5:17] . . . does not refer to our believing acceptance of the free gift but to our being made the recipients, and we are regarded as the passive beneficiaries of both the grace and the free gift in their overflowing fullness.” [4] According to Paul, in other words, we no more choose to experience the beneficial effects of Christ’s one act of righteousness than we chose to experience the destructive effects of Adam’s disobedience.

Second, Moo has attributed to Paul a fallacious argument of the following form:

(1) Only those sinners receiving the abundance of grace will “derive the benefits of Christ’s act of righteousness” and thus be saved.
Therefore,
(2) Not all sinners will “derive the benefits of Christ’s act of righteousness” and thus be saved.

The premise sets forth a necessary condition of salvation, namely that a sinner must receive “the abundance of grace” in order to be saved, and the conclusion draws the inference that, therefore, some sinners will never meet that necessary condition. But that inference is obviously fallacious—as is the following inference of exactly the same form: only those believers who remain faithful to the end will be sanctified; therefore, not all believers will be sanctified. So even if Paul were not using lambanō in a passive sense, as he surely was, Moo’s appeal to 5:17 in an effort to explain away 5:18 would merely attribute to Paul the same fallacious inference that Moo brings to the text. For unless Paul himself had drawn a similar fallacious inference, neither “the deliberately worded v. 17” nor the “persistent stress on faith as the means of achieving righteousness” carries any implication that Paul intended the second “all” in 5:18 to be more restrictive than the first. Much less would it justify Moo’s conclusion that, according to Paul, “only certain people [that is, only some sinners and not all of them] derive the benefits from Christ’s act of righteousness.” Quite the contrary. Paul’s explicit affirmation in 5:18 that Christ brings “justification and life” to all humans already entails that all of the necessary conditions of such justification and life will eventually be met. So you can hardly challenge the universal scope of the second “all” in 5:18 merely by pointed out, as Moo does correctly, that the right kind of faith is one of these necessary conditions.

Accordingly, if you want to understand Christian universalism accurately before criticizing it, as any competent critic would want to do, the first lesson to learn is this: proponents of such universalism not only do not deny, but even insist, that the salvation of any sinner requires that certain conditions be met. But they also believe that Christ’s ultimate victory over sin and death is what guarantees that all of the relevant conditions will indeed be met in the end.

Consider next Martin’s suggestion that Paul’s use of “the many” in verse 19 reduces his “all” in verse 18 to something like more than a few in number. Unfortunately, that ignores Paul’s own clarification in verse 15—where Paul distinguished within the group or class of all human sinners between “the one” and “the many”—“the one” being Adam himself, who first sinned, and “the many” being all of those who died as a result of Adam’s sin. So once again it is John Murray who, despite his own vigorous opposition to universalism, has nonetheless pointed out the fatal flaw in Martin’s kind of argument:

When Paul uses the expression “the many”, he is not intending to delimit the denotation. The scope of “the many” must be the same as the “all men” of verses 12 and 18. He uses “the many” here, as in verse 19, for the purpose of contrasting more effectively “the one” and “the many”, singularity and plurality—it was the trespass of “the one” . . . but “the many” died as a result. [5]

In the same context, moreover, Paul insisted that “the one,” namely Adam, was “a type” of Jesus Christ (vs. 14), presumably because Jesus Christ, the second Adam, stands in the same relationship to “the many” as the first Adam did. But with this difference: “if the many died by the trespass of the one man, how much more did God’s grace and the gift that came by the grace of the one man, Jesus Christ, overflow to the many!” (vs. 15—NIV). It seems to me indisputable, therefore, that Paul had in mind one group of individuals—“the many,” which included all human beings except for the first and the second Adam—and he envisioned that each of the two Adams stands in the same relationship to that one group of individuals. The first Adam’s act of disobedience brought doom upon them all, but the second Adam’s act of obedience, whose effects are even greater and more extensive than the effects of Adam’s disobedience (thus the expression “how much more”), undid the doom and will eventually bring justification and life to them all. In the words of M. C. de Boer, “Unless the universalism of vv. 18-19 is taken seriously . . . ‘how much more’ is turned into ‘how much less,’ for death is then given the last word over the vast majority of human beings and God’s regrasping of the world for his sovereignty becomes a limited affair.” [6] Or, as Arland J. Hultgren has put it, “As Adam was the head of humanity in the old eon, leading all to destruction, so Christ is the head of humanity in the new age which has dawned, leading all to justification and life. The grace of God in Christ amounts to ‘much more’ than the trespass of Adam and its effects (5:17). All of humanity is in view here without exception.” [7]

‍

‍The Wider Context of Pauline Thought

‍Even opponents of universalism sometimes admit that, taken in its own context, Romans 5:18 at least appears to teach an explicit universalism. Indeed, no less a conservative authority than John Murray appears to have made such a concession in the above quotations; he therefore appealed to the broader context of Pauline thought in an effort to escape the clear universalistic import of our text. He thus wrote:

When we ask the question: Is it Pauline to posit universal salvation? the answer must be decisively negative (cf. II Thess. 1:8, 9). Hence we cannot interpret the apodosis in verse 18 [of Rom 5] in the sense of inclusive universalism, and it is consistent with sound canons of interpretation to assume a restrictive implication. In I Cor. 15:22 Paul says, “As in Adam all die, even so in Christ shall all be made alive”. As the context will demonstrate the apostle is here dealing with the resurrection to life, with those who are Christ’s and will be raised at his coming. The “all” of the second clause is therefore restrictive in a way that the “all” in the first clause is not. In like manner in Rom 5:18 we may and must recognize a restriction in the “all men” of the apodosis that is not present in the “all men” of the protasis. [8]

The first sentence in the above quotation is especially relevant to our main topic of how to read the Bible from a universalist perspective, and I’ll return to it shortly. But first I want to point out that Murray too has fallen prey to an obviously fallacious inference, indeed the same kind of reasoning that we encountered above in Moo. From Murray’s premise that the second “all” in 1 Corinthians 15:22 is restricted to “those who are Christ’s and will be raised at his coming,” it simply does not follow that the second “all” is therefore more restrictive than the first. To get that conclusion, you must adopt the additional assumption that the first “all” includes some who will never belong to Christ—a question begging assumption, if ever there was one. As a good Calvinist, Murray held that the elect have belonged to Christ from the very foundation of the world—long before any of them knew they belonged to him and long before they consciously identified with him. So even if one should accept Murray’s understanding of the context—which is itself debatable—why not draw the inference from 1 Corinthians 15:22 and its immediate context that the entire human race has belonged to Christ from the very foundation of the world? That inference, unlike Murray’s, would at least have the virtue of being a valid inference.

In fact, Murray appears to have gotten the matter exactly backwards because the first “all” in 15:22 is, if anything, more restrictive than the second. For right after making his two parallel statements about the entire human race, Paul immediately expanded his second “all” to include not only the descendants of Adam, but every competing will as well. Christ must continue to reign, he insisted, until he finally brings all things, including every will and opposing power, into subjection to him (15:24-27), and there is but one exception to this “all things,” namely the Father himself (15:28). The last enemy to be destroyed is death (15:27), which in the larger context of Paul’s thought includes all separation from God. When Christ finally overcomes all separation from God, all persons will then be in subjection to Christ in exactly the same sense that Christ places himself in subjection to the Father (15:28)—a sense that, as I have argued in various places, implies spontaneous and glad obedience. Then and only then will the Father truly be “all in all,” because then and only then will all persons belong to him, or at least know that they belong to him, through his Son.

Note also that Murray seems to have thought it sufficient merely to cite 2 Thessalonians 1:8-9 as proof that we should reject the apparent universalism of Romans 5 and 1 Corinthians 15. In a similar vein, Charles Hodge once wrote: “As, however, not only the Scriptures generally, but Paul himself, distinctly teach that all men are not to be saved, as in 2 Thess. I.9, this [universalistic] interpretation [of Romans 5:18] cannot be admitted by any who acknowledge the inspiration of the Bible.” [9] On its face, that is a remarkable claim for two reasons: first, because many Christian universalists have believed as strongly as Hodge did in the inspiration of the Bible, and second, because one could just as easily, if one wanted to be uncharitable, use the same kind of argument against Hodge. For surely, the following argument, which is just the reverse of Hodge’s argument, is equally cogent at this point: “Because not only the Scriptures generally, but Paul himself, distinctly teach universal reconciliation, as in Romans 5, Romans 11, and 1 Corinthians 15:20-28, Hodge’s interpretation of 2 Thessalonians 1:9 cannot be admitted by any who acknowledge the inspiration of the Bible.” As this counterargument illustrates, the issue of inspiration is a distracting irrelevancy in the present context; it is the correct interpretation of a text, not the inspiration of the Bible, that is here at issue. And concerning that issue—the correct interpretation of Romans 5:18—the appeal of Murray and Hodge to 2 Thessalonians 1:9 suffers from a serious weakness. For without any trouble at all, one could simply reverse their argument and argue just as plausibly in the opposite direction.

Herein lies an important key, I believe, to reading the Bible with confidence from a universalist perspective: to do so, you need only become aware of how easy it is to reverse the most common arguments against such a reading. There are, after all, two prominent New Testament themes that may initially seem difficult to harmonize: that of Christ’s total victory and triumph over sin and death, on the one hand, and that of God’s wrath, judgment, and punishment of sin, on the other. Such texts as Romans 5 and 11, 1 Corinthians 15, and the old creedal hymn reproduced in Colossians 1:15-20, among others, illustrate the first theme, whereas such texts as 2 Thessalonians 1:9, the parable of the sheep and the goats (Matt. 25:31-46), and the parable of the rich man and Lazarus (Luke 16), among other texts, illustrate the second. Now one possible response is to claim that these two themes are simply incompatible and hence irreconcilable. But those who reject such an inconsistency, as do most Arminians, Calvinists, and George MacDonald-type universalists, can reason in one of two directions, each of which is just the reverse of the other. Those, such as Murray and Hodge, who interpret various judgment texts as teaching that some sinners are destined for unending retribution in hell, will adjust their understanding of Christ’s ultimate victory and triumph in light of that presumed teaching; and we universalists, who interpret certain texts as teaching that in Christ God will eventually reconcile the entire human race to himself will in a similar manner adjust our understanding of God’s wrath and judgment in light of that presumed teaching. The one pattern of argument is just the reverse of the other.

So do we here encounter a mere impasse or stalemate? Not in my opinion. For Paul himself, I believe, teaches us exactly how to understand God’s wrath, judgment, and severity as an expression of his boundless love and inexhaustible mercy.

‍

‍God’s Severe Mercy

‍According to Paul in the eleventh chapter of Romans, God’s severity towards the disobedient, his judgment of sin, even his willingness to blind the eyes and harden the hearts of the disobedient, are expressions of a more fundamental quality, namely that of mercy, which is itself an expression of his purifying love. In 11:7 he thus wrote: “What then? Israel failed to obtain what it was seeking. The elect obtained it, but the rest were hardened” (or blinded). He then immediately asked, “Have they [the non-remnant who were hardened or blinded] stumbled so as to fall?” And his answer was most emphatic: “By no means!” (11:11). By the end of the following verse, he was already speaking of their full inclusion: “Now if their stumbling means riches for the world, and if their defeat means riches for the Gentiles, how much more will their full inclusion mean!” (11:12). [10] And three verses later he was hinting that their acceptance would mean “life from the dead” (9:15). He then generalized the whole thing: God blinded the eyes and hardened the hearts of the unbelieving Jews, we discover, as a means by which all of Israel might be saved (Romans 11:25-26)—all of Israel including those who were blinded and hardened. There is simply no way, so far as I can tell, to escape the universalistic implication here. The specific point that Paul made in Romans 11 was this: though the unbelieving Jews were in some sense “enemies of God” (11:28), they nonetheless became “disobedient in order that they too may now receive mercy” (11:31-NIV). But the general principle (of which the specific point was but an instance) is even more glorious: “For God has imprisoned all in disobedience so that he may be merciful to all” (11:32—my emphasis). [11]

According to Paul, therefore, God is always and everywhere merciful, but we sometimes experience his mercy (or purifying love) as severity, judgment, punishment. When we live a life of obedience, we experience it as kindness; when we live a life of disobedience, we experience it as severity (see 11:22). Paul himself called this a mystery (11:25) and admitted that God’s ways are, in just this respect, “inscrutable” and “unsearchable” (11:33), but nothing could be clearer than his own glorious summation of the whole thing in 11:32. If the first “all” of 11:32 refers distributively to all the merely human descendants of Adam, if all are “imprisoned” in disobedience, then so also does the second; they are all objects of divine mercy as well. And if one should insist, as some have in a seemingly desperate effort to escape universalism, that neither “all” literally means “all without exception,” the obvious rejoinder is that here, no less than in Romans 5:18 and 1 Corinthians 15:22, the parallelism is even more important than the scope of “all.” According to Paul, the very ones whom God “shuts up” to disobedience—whom he blinds, or hardens, or cuts off for a season—are those to whom he is merciful; his former act is but the first expression of the latter, and the latter is the goal of the former. God hardens a heart in order to produce, in the end, a contrite spirit, blinds those who are unready for the truth in order to bring them ultimately to the truth, “imprisons all in disobedience so that he may be merciful to [them] all.”

But if God is truly merciful to all, according to Paul, how are we to reconcile that with the claim in 2 Thessalonians 1:9 that some “will suffer the punishment of eternal destruction,” or the claim in Matthew 25:46 that some will “go away into eternal punishment,” or the claim in Luke 16:26 that an unbridgeable chasm exists between the place where the rich man was being tormented and the place where Lazarus was being comforted in Abraham’s bosom? One common strategy among Christian universalists is to point out that the word “aiōnios,” which many of our English Bibles translate as “eternal” or “everlasting,” literally means something like age enduring or perhaps that which pertains to an age. But neither annihilationists nor proponents of an unending hell typically find such a move persuasive. According to Arthur W. Pink, for example, “The Greek word . . . ‘aionios’ and its meaning and scope has been definitely defined for us by the Holy Spirit in” 2 Corinthians 4:18, where Paul noted that “the things which are seen are temporal; but the things which are not seen are eternal” (KJV). According to Pink, then, “it is obvious that . . . if the things ‘eternal’ are merely ‘age-long,’ then they cannot be properly contrasted with things that are temporal.” [12] But of course, if that which is eternal in some quasi-Platonic sense is to be contrasted with that which is temporal, then that which is eternal, being timeless, would carry no implication of temporal duration at all and hence no implication of unending temporal duration either.

In any case, I typically adopt a very different strategy against those, such as Pink, who insist that “aiōnios” implies unending temporal duration. For we can always make them a present of this Greek adjective and turn our attention, in specific cases, to the noun it qualifies. As an illustration, consider the expression often translated as “eternal destruction” in 2 Thessalonians 1:9. Just how should we understand the relevant destruction here? Should we understand it as the annihilation of some individual person who is a sinner? Or, should we understand it as the complete destruction of what Paul called the old person, the very thing whose destruction is required for salvation? Which understanding fits better with Paul’s own insistence that God’s severity is always an expression of his mercy? In fact, Paul used exactly the same term for destruction (“olethros”) when he wrote this in 1 Corinthians 5:5: “you are to hand this man over to Satan for the destruction of the flesh, so that his spirit may be saved in the day of the Lord.” Here the relevant destruction is explicitly a means of correction, albeit a very severe means, and nothing in the context of 2 Thessalonians 1:9 excludes a similar meaning there. So once we come to understand that the idea of destruction is itself a redemptive concept, the issue of the correct translation of “aiōnios” pretty much fades into the background.

Accordingly, I would here stress one simple and utterly non-controversial point about the word “aiōnios”: it is an adjective and must therefore function like an adjective. Because adjectives often vary in meaning, sometimes greatly, when the nouns they qualify signify different categories of things, one of the most popular arguments concerning Matthew 25:46 is deeply flawed. Here I have in mind the argument that, if eternal life is life without end, then eternal punishment must likewise be punishment without end. We can illustrate the flaw in this argument by switching from Greek to English. Consider how the precise meaning of the English word “everlasting” can vary in different contexts. An everlasting struggle, if there should be such a thing, would no doubt be a struggle without end, an unending temporal process that never comes to a point of resolution and never gets completed. But an everlasting change, or an everlasting correction, or an everlasting transformation need not be an unending temporal process at all and certainly not one that never gets completed; it might instead be a temporal process of limited duration, or perhaps even an instantaneous event, that terminates in an irreversible state that endures forever.

Nor is there any doubt that the life and the punishment of which Jesus spoke in Matthew 25:46 belong to different categories of things. For whereas the life (zōē), being rightly related to God, is clearly an end in itself—that is, valuable or worth having for its own sake—the punishment (kolasis) is just as clearly a means to an end. And throughout the Greek world, furthermore, the word “kolasis” was widely understood to signify a means of correction. But even if the Gospel writer had chosen the word “timōria,” a common word for vengeful punishment, you cannot infer the absence of a corrective purpose from harsh language alone. Given Paul’s explicit teaching in Romans 11 that even God’s harshest judgment serves a merciful purpose, it follows that, whatever additional purpose it might serve, divine punishment is administered, at least in part, for the good of the one being punished. And once we come to understand that, it matters little whether we translate “eis kolasin aiōnion” as “into eternal correction,” “into everlasting correction,” or “into the kind of correction that pertains to the coming age.”

As for the unbridgeable chasm of which Jesus spoke in the parable of the rich man and Lazarus, not one word in this parable, even if taken as literal history, as some do take it, implies that the chasm between Hades and Abraham’s bosom will remain unbridgeable forever. Do not Christians believe that the cross has already guaranteed the ultimate destruction of sin and death, where the “last enemy to be destroyed,” as we have already noted, “is death” itself? When 1 Peter 3:19 depicts Jesus as preaching to the spirits in prison (or those who were disobedient in the days of Noah) and 1Peter 4:6 also depicts him as preaching the gospel to the dead, do these texts not illustrate perfectly the view of Elhanan Winchester [13], who wrote: “I believe, that Jesus Christ was not only able to pass, but that he actually did pass that gulph, which was impassable to all men but not to him.” [14] Even if one should take the details of this parable more literally than one should, in other words, one can still view the Cross as the means whereby Jesus Christ has bridged this hitherto unbridgeable gulf. By flinging himself into the chasm between the dead and the living and by building a bridge over it, Jesus thus brought his message of repentance and forgiveness to all of those in Hades, which is the abode of the dead.

In conclusion, our time constraints here are incompatible, unfortunately, with a thorough discussion of these matters. But however sketchy this discussion of a few specific texts has been, I hope it illustrates two points: first that, according to Paul’s teaching in Romans 11, God’s severity towards the disobedient, no less than his kindness towards the obedient, is an expression of his boundless mercy to the entire human race, and second, that anyone who takes this teaching seriously can plausibly interpret the biblical theme of divine judgment accordingly. In particular, one can interpret the so-called eternal destruction of the wicked as the final destruction of what Paul called the old person, which is our sinful nature; one can also interpret the so-called everlasting punishment of the wicked as an everlasting correction or transformation of the wicked; and one can interpret the hitherto unbridgeable chasm between Hades and Abraham’s bosom as precisely the chasm over which Jesus alone was in fact able to build a bridge.

‍

Thomas Talbott
Emeritus Professor of Philosophy Willamette University

‍

1. Unless otherwise indicated, all quotations from the Bible here are from the New Revised Standard Version copyrighted in 1989 by the National Council of Churches of Christ in the USA.  
2. Douglas J. Moo, Epistle to the Romans (Grand Rapids: Eerdmans, 1996), p. 344.
3. Ralph P. Martin, “Romans” in D. Guthrie, J. A. Motyer (eds.), The New Bible Commentary: Revised (Grand Rapids: Eerdmans, 1970), p. 1026.
4. John Murray, Epistle to the Romans, Vol. I (Grand Rapids: Eerdmans, 1968), p. 198. Richard Bell, following M. E. Boring, makes a similar point in “Rom. 5:18-19 and Universal Salvation”, New Testament Studies 48.3 (2002), p. 429. See also Boring, “The Language of Universal Salvation in Paul,” Journal of Biblical Literature 105 (1986), pp. 269-92.
5. Murray, op. cit., pp. 192-193.
6. M. C de Boer, Defeat of Death: Apocalyptic Eschatology in 1 Corinthians 15 an Romans 5 (Sheffield, UK: Sheffield Academic Press, 1988), p. 175.
7. Arland J. Hultgren, Christ and His Benefits: Christology and Redemption in the New Testament (Philadelphia: Fortress Press, 1987), pp. 54-55.
8. Murray, op. cit., p. 203.
9. Charles Hodge, Commentary on the Epistle to the Romans (New York: Armstrong, 1896), p. 270.
10. In order to avoid the implication that God hardens the heart as an expression of mercy, some commentators have insisted that Paul was here speaking of Israel as a corporate whole. As John Piper has put it: “Notice that this [i.e., the “they” in 11:11] is not a reference to all Jews but to Israel as a corporate whole conceived of as an entity that endures from generation to generation made up of different individuals from time to time” (“Universalism in Romans 9-11?” 12). But that will never do. For in 11:7 Paul mentioned three groups of people: Israel or the nation as a corporate whole, “the elect” or the faithful remnant, and “the rest,” that is, the non-remnant Jews who were hardened. Now the antecedent of “they” in 11:11 cannot be the faithful remnant; they are not the ones who stumbled and were hardened. Neither can it be the nation as a corporate whole, for Paul had just distinguished between two groups within that corporate whole: the faithful remnant who did not stumble and were not hardened, and “the rest” who did stumble and were hardened. Accordingly, the antecedent of “they” in 11:11 must be “the rest,” the non-remnant Jews, the very ones whom God had hardened. Even John Murray effectively admitted this when he asked: “Is not the denotation of those in view [in verse 11] the same as those mentioned in verse 7: ‘the rest were hardened’? And is not Paul thinking here of those in verse 22: ‘toward them that fell, severity’?” (Epistle to the Romans, Vol I, 75, n. 18). The answers are, “Yes” and “Yes.” But since Murray could not believe that God’s severity, or his hardening of a heart, might be an expression of mercy, he continued to insist that “those who stumbled did fall with ultimate consequences.” The “denotation of those in view” in verse 11, however, is not only “the same as those mentioned in verse 7,” as even Murray recognized; it is also the same as those mentioned in verse 12: those whose “full inclusion” will mean so much more than the stumble that made their full inclusion possible.
11. As the New Testament Scholar James Dunn correctly summarized the point, “God hardens some in order to save all; he confines all to disobedience in order to show mercy to all” (Romans 9-16, 696).
12. See A. W. Pink, “Eternal Punishment, ” Sec. 6, available at  http://www.reformed.org/eschaton/index.html?mainframe=/eschaton/pink_eternal_punishment.html.
13. Although I here adopt the interpretation of these texts that prevailed in the early church, especially in the East, until Augustine tried to circumvent it, many scholars do reject it and do so for reasons that I find utterly unpersuasive. Some insist that in 3:19 the “spirits in prison” were fallen angels rather than the spirits of disobedient humans; others insist that, according to 4:6, the gospel was preached (passive voice) by someone or another to those who were previously alive long enough to hear it and then subsequently died before 1 Peter was composed. Fortunately, however, we need not try to resolve such exegetical disputes here, because the idea that the Cross built a bridge over the hitherto unbridgeable gulf between Hades and Abraham’s bosom in no way requires some particular interpretation of these texts.
14. Elhanan Winchester, The Universal Restoration Exhibited in Four Dialogues between a Minister and His Friend (Philadelphia: Gibon, Fairchild & Co., 1845), p. 25.