The First Resurrection: A Reaffirmation

By Meredith G. Kline

Gordon-Conwell Theological Seminary, South Hamilton, Massachusetts

The characteristic courtesy of Ramsey Michaels in discussion, even when the discussion is a disagreement, is amply evidenced again in his Response to my earlier article. I admire the virtue and appreciate this particular instance of it, but as one notably deficient in irenic grace I could almost wish he had set a less noble example! Despairing of matching it, I tender my apologies beforehand, ere the ardor of offensive defensiveness has quite carried me away.

So far as formal exegesis is concerned, Professor Michaels’ Response is not so much a criticism of my article (the distinctive point of which he in fact endorses) as it is a novel proposal of his own setting himself against the commentators of all millennial schools on what has been a point of fundamental, if formal, agreement among them. Though he characterizes his approach as a “rather conventional premillennial” one, at this critical point it is anything but conventional.

He charges the commentators on all sides with complicating the picture in Revelation 20:4–6 by reading as a double pair of first and second death and first and second resurrection what he wants to restrict to the single pair of first resurrection and second death. In spite of the bold manner in which this proposal is introduced as being “of course” what John says, it must strike most readers of the Book of Revelation as a strained exegesis, unnatural in the extreme.

How he arrived at such a position seems to be clear enough, however. Having acknowledged the validity of my observation that in the Revelation 20 and 21 context “first” and “second” do not refer simply to mere sequence but denote a qualitative difference, he then perceives that if that fact is incorporated into the double binary pattern interpretation that has all along been accepted on all sides, it will have to be admitted that there is a difference in kind between the first and second resurrections as well as between the first and second deaths. That, he is also aware, spells the end of the premillennial exegesis—and millennial opinions go down swinging!

It is under point two that the Response takes up this decisive issue, asserting that Revelation 20:4–6 does not contain the idea of two death experiences and two resurrection experiences but only one of each. The Response says that John does not pair “first” with death or “second” with resurrection, and it then makes the quite unfounded assumption that if first death and second resurrection are not present as explicit terms in the text they are not present in the thought of the text either. But the attempt of the Response to avoid the force of the fact that the concepts of first death and second resurrection are found in this context is evasive and futile. The concept or experience of the first death is, in fact, brought into immediate juxtaposition to the term “second death” every time the latter is used (Rev 2:10, 11; 20:4–6; and 20:13, 14). This is not fortuitous. Whatever proleptic or parenthetical aspect may be involved in the use of “second death” in these passages, its use in combination with the concept or experience of the first death is still eminently appropriate in each instance, and thus rather clearly deliberate. The concept of a second resurrection is also prominent in the context where the term “first resurrection” is used (Rev 20:12ff). These data flatly contradict the contention of the Response that we are not to think (as we find ourselves naturally inclined to do) of a first death and second resurrection in Revelation 20:4–6 in order to fill out the pattern that is initiated there by the explicit terms “first resurrection” and “second death” and is further intimated by the reference to the experience of the first death in this very passage (v. 4) as well as in the surrounding context. These data are far more than an adequate basis for recognizing the double binary pattern; they are completely compelling, and it is this that accounts for what has been the general consensus on the subject.

Beyond this it should be noted that the Response hardly gives an adequate account of all the facts concerning the explicit usage of the term “first” in Revelation 20 and 21 when it says that “first” is paired only with “resurrection” and not with “death.” Revelation 21:4 says that death and the related phenomena of sorrow, crying, and pain will be no more in the new world and explains this absence of death by the statement: “for the first things are passed away.” The death in view here is identified as one of the “first things” (i.e., as belonging to the pre-consummation order). The facts are obscured when this evidence is dismissed with the ambiguous and misleading remark that “Kline can point to the use of the word ‘death’ in 20:13f and 21:4 as evidence that the concept, though not the term, ‘first death’ is present in the context.” The fact is that Revelation 21:4 does virtually use the term “first death.” And accordingly the assumption of the Response discussed above is not merely unfounded and contradicted by contextual data, but it (and the whole argument of point two) is quite pointless.

But before turning to other sections of the Response it should be noted that its own positive exegetical proposal will not bear close scrutiny. As previously mentioned, Michaels agrees that a qualitative contrast is denoted by the first-second pair in Revelation 20 and 21. But he ignores the particular nature of that difference as a contrast between old, pre-consummative and new, consummative. On his view of a single pair, namely, first resurrection and second death, “first” and “second” denote the bare notion of difference, supposedly underscoring the polar contrast between death and resurrection. In the attested usage in the context, however, the first-second pair is used primarily and essentially to contrast the old and the new varieties of one entity (like apples), i.e., death or resurrection, not, as in Michaels’ view, to denote in some abstract way the contrast between totally different entities (like apples and rocks), i.e., death and resurrection. Michaels has not faced up to the real nature of the contextual evidence, which offers no justification for interpreting “first” and “second” as he does. Moreover, stripped of the specific antonymic significance (i.e., old and new) which they have in the context and even of every trace of sequential force, as they are in Michaels’ view, it is not evident how the terms “first” and “second” would convey the notion of contrast at all. The idiom would be quite opaque and would not, therefore, serve to reinforce the inherent contrast between death and resurrection but only, bv perplexing the reader, to blur it. In fact, as Michaels expounds “first” and “second,” they do not express difference but similarity. The “first resurrection” is said to be so named because it is the only one to deserve the name. Apparently, then, “first” means the real thing. And the “second death” is said to be so designated because it has finality. So, “second” seems to mean final. But “real” and “final” are, if comparable at all, synonymous, not antonymic. It is a question, therefore, whether the exegetical proposal of the Response is intelligible, let alone credible.

Revelation 20:4–6 does, then, contain the idea of two kinds of death experiences and two kinds of resurrection experiences, so that we all have to come up with identifications for these—and there are no options in sight that would salvage the premillennial exegesis. The Response contends under points one and three that the particular experiences with which my article identified the two kinds of death and resurrection are not different kinds but just one kind of each.

Under point one, the Response argues that the second death is not different in kind from physical death, which introduces man into the intermediate state, but only finalizes it. This is a strange contention which obliterates the significance of the bodily resurrection and the consummation. For if the second death simply finalizes physical death, man’s disembodied condition would be perpetuated; there would be no resurrection of the wicked. Michaels does not wish to deny the resurrection of the wicked and, in fact, he acknowledges elsewhere that it is the resurrection of the wicked that is paradoxically called “the second death,” since it is “only formally a resurrection.”

Certainly a death that consists in disembodiment differs in kind from death as re-embodiment to suffer eternal perdition. The Book of Revelation introduces a new image, the lake of fire, to portray the new, eternal kind of death. Revelation 20:14 depicts death—death in the sense of the grave and intermediate state—as being terminated by means of this very lake of fire, the new kind of death to which it gives way. This is, of course, one of the places where the context compels us to think of a first death as the intended contrast to what is called “the second death.”

Whether we do well to speak of this second death as “metaphorical” is no doubt debatable. If that term suggests the mistaken idea that the second death is not a real future event or that it is spiritual death in distinction from an experience of the total person raised from the grave, that is a serious disadvantage.

On the other hand, since it is the resurrection (of the wicked) that is called “the second death,” the appropriate adjective for this usage does not seem to be “literal.” But whatever adjectival term we use to distinguish them, we are dealing here with two very distinct kinds of death.

In the discussion of the two kinds of resurrection under point three, the Response again betrays a lack of appreciation for the decisive significance of the resurrection-consummation event. My article took the position that just as “the second death” is the paradoxical designation for the resurrection of unbelievers, so the death of believers is paradoxically called “the first resurrection.” The two support and really demand each other. Michaels accepts the first but balks at the second (an inconsistency that is necessary if one is not to give up premillennialism). He maintains that “first resurrection” is a contradiction in terms, if “first” is understood as pre-consummative.

Professor Michael’s argument here succeeds in showing only how seriously he has been influenced by the Kantian dialectic, to which as a matter of fact he refers with approval. He speaks of a “first resurrection as resurrection” and says that we cannot speak of a particular “resurrection” as consummative and at the same time as belonging to the present order (in an intrusive-anticipatory sense). The resurrection (and consummation) concept of the Response is radically Kantian, not biblical; it is an ideal abstraction of the noumenal realm that cannot enter the phenomenal realm and participate in history. Hence, time with its chronological distinctions of earlier and later has no significance for it. It floats immune above history’s eschatological movement from this world to the consummation world. The familiar biblical eschatological structure of already not yet ceases to be the biblical structure when it is reinterpreted within the Kantian system. Set in that framework, the already and the not yet can only stand as abstract entities in dialectical tension; they cannot come together as coordinate dimensions of individual historical experiences. It is only in terms of such a profoundly unbiblical approach that it could be denied that there is a difference in kind between the “resurrection” which the Christian experiences when he passes into the intermediate state at death, absent from the body though at home with the Lord, and the resurrection he experiences at the day of the redemption of his body and glorification. It is precisely because my colleague and (I hope still) friend has let himself come under this alien Kantian-Barthian influence that we repeatedly find that the biblical reality of the consummation does not come into its own in his thinking.

Regrettably, a personal note must be intruded here. For in this connection the Response links my hermeneutics with the Kantian world view, and I may not let my Christian witness be thus compromised. To dispel any false impression that might exist, let me say there is no later Kline, only an older version of the Van Tilian Reformed, Covenantal, garden variety of Christian he was in the 1948–65 Westminster period. I still reprobate the Kantian dialectic that comes to expression in Barthian hermeneutics, exemplified in a commentary like that of Caird on Revelation, and I can only deplore its insidious influence within evangelical circles.

One’s judgment on the exegetical question under discussion will have been determined on the basis of the decisive issues already reviewed, but the remainder of the Response calls for comment. In point four the problem once again is the logical one of unfounded assumptions. The Response isolates Revelation 20 from Revelation 21, identifying chapter 20 (quite erroneously) as the “climactic vision of the Christian hope,” and arbitrarily insists that the topic of the bodily resurrection of believers must be dealt with within these literary confines, suggesting—the last desperate swinging of the premillennial cause now becomes very wild indeed!—that otherwise the scent of incipient gnosticism will contaminate Revelation 20.

Involved here, of course, are larger questions about the structure of the Book of Revelation and the theme of chapter 20 and how it fits into the total structure. I would just briefly observe that, as I understand it, Revelation 20 is one part of an extensive section of the book in which the final disposition of all the major figures who have appeared is dealt with, each in turn. Revelation 20 is Satan’s turn, and it is no wonder then that this chapter does not feature the final resurrection of the redeemed. Their turn comes in Revelation 21:1–8, the last part of the long section of final judgments. If one is going to speak of a “climactic vision of Christian hope,” a more appropriate selection would be Revelation 21:9ff, which does present the resurrection-consummation event under the imagery of the revelation of the bride of the Lamb in the glory of the eternal city. This theme is also present in Revelation 21:1–8. Whether or not one accepts precisely this analysis of the structure, it should be apparent that one cannot just blithely proceed on the basis of the unvalidated assumptions made in point four of the Response.

As for point five, the Response slights the relevance of Revelation 2:10f and 14:13, especially the “from henceforth” in 14:13. In the comparison of Revelation 6:9–11 with Revelation 20:4–6, faulty logic is again in evidence. Noting that these two pictures of the church differ in that the number of the Christians is incomplete in the vision of the intermediate state in Revelation 6, the Response argues that if the relation between the two is not that of simple identity but rather of continuity, Revelation 20 “must refer” to a subsequent stage which “can only be” the bodily resurrection beyond the intermediate state.

But the logical options are obviously not so limited. There is indeed progression in the perspective from incompleteness to completeness as we move from Revelation 6 to Revelation 20, but both passages portray the intermediate state. Revelation 20:4–6 views the entire period of the church in the intermediate state as a whole, while Revelation 6:9–11 sees it at a particular (early) point. And, contrary to the Response, the eschatological position and posture of the Christians is precisely the same in both passages. A judgment has already been rendered in favor of the martyrs in Revelation 6, just as for the martyrs in Revelation 20. It is attested by the bestowal of white priestly robes, indicative of the verdict of righteousness. And like the saints in Revelation 20, who share with Christ in the reign of his sabbatical session on the heavenly throne, so too the overcorners of Revelation 6 (cf. Rev 3:5) are granted royal sabbatical repose. The priestly aspect of the service of the church before the throne is also present in both passages. Moreover, in each case, while favorable judgment has already been rendered for the martyrs, the punitive vengeance of God against their wicked slayers is still future. In Revelation 6 the martyrs pray for that ultimate vindication and in Revelation 20 that consummation of divine judgment is yet to come after “the thousand years.” This full and striking parallelism between Revelation 20:4–6 and the vision of Revelation 6:9–11, which admittedly refers to the intermediate state, is a powerful confirmation that “the first resurrection” of Revelation 20 refers to the experience of death through which the Christian enters that blessed and holy state.

As an appendix to the foregoing review of the Response, I shall venture a few comments on certain broader aspects of the millennial question.

The policy adopted in some of our Reformed churches allowing considerable confessional latitude on the subject of the millennium would suggest that this is an isolable exegetical question, not affected by the general body of Reformed doctrine and not necessarily affecting the latter in any confessionally significant way. But that stance might well stand some rethinking. Of special interest is the way the doctrine of common grace fares in different millennial reconstructions, for that doctrine is a cornerstone of the Reformed view of history.

The world order of common grace is covenantally formalized in the post-diluvian revelation to Noah (Gen 8:15–9:17). This covenant of God with all the earth granted to the just and the unjust in common a measure of the blessings of earthly life, including the provision of the judicial structure of the state. And God defined this arrangement as one that would continue valid and operative until the termination of the present cosmos. The consummation will introduce through cosmic re-creation a world order in which no blessings of any kind are apportioned as a common gift in which the unjust participate but all blessings are for the beneficiaries of special redemptive grace alone. Until that re-creation, however, the order of common grace is firmly established by divine stipulation and decree. It is open to penetration at points by the world order to come, but is not subject to eclipse until the consummation.

The millennium envisaged by classical premillennialism is in conflict with the terms of the covenant of common grace. It features a theocracy on earth before the consummation, a universal kingdom of Christ in which those blessings hitherto received in common by all men and often in greater measure by the unjust than the just are no longer apportioned according to the principle of comnion grace but according to a policy of special favor to the people of God. The redeemed are in fact already in possession of glorified natures and experience their public vindication over against the world of unjust men, The premillenarian’s millennium would thus abrogate the common grace order prior to the consummation, in contradiction of God’s covenantal guarantees.

The cogency of this criticism might be questioned on the grounds that history has already witnessed an earthly theocracy established bv divine appointment. But theocratic Israel under the old covenant was a limited, local kingdom. Other nations coexisted with Israel and the common grace order thus continued uninterrupted, true to the terms of God’s covenant. In the case of the millennial theocracy of classical premillennialism, however, there would be a universal suppression of the common grace principle with its judicial order of the state, because this theocracy would itself be a universal world order.

There are, of course, non-classical forms of premillennialism that would, or at least could, avoid this difficulty since they foresee a millennial theocracy that is not universal but a restoration, with Messianic overtones, of the limited Israelite institution of Old Testament days. Such varieties of premillennialism, however, contravene Reformed theology even more blatantly in other vital respects. (See my remarks in The Westminster Theological Journal 36 [1974], 245–7.)

In so far as postmillennialism joins premillennialism in locating the realization of messianic kingdom prophecies in an earthly millennial kingdom, it too is incompatible with the covenant of common grace. Moreover, because of the particular nature of the millennial theocracy of postmillennialism, the postmillennialist is lured into a misinterpretation of the theocratic covenant mediated through Moses.

On the premillennial view, the millennial kingdom is a church-kingdom ruled over by Christ, who, on this view, has returned before the millennium. On the postmillennial view, Christ does not return with the glorified church until after the millennium and meanwhile the millennial kingdom is a state-kingdom. Now if a theocratic state fulfills the prophetic blessing ideal of the old covenant, it must also fulfill the legal stipulation ideal of that covenant. Covenantal prophecy and law must agree with each other. Hence, the covenant which God gave Israel as the constitution for the Old Testament theocratic kingdom will be interpreted by a consistent postmillennialist as though it were the constitution for an ordinary state kingdom. What was meant to apply to the special redemptive institution of the theocracy—the demand to confess God, the guarantee that obedience to the covenant stipulations will be rewarded with earthly prosperity and power, etc.—must all be regarded by the postmillennialist as normative for the state, any state. And he looks to the millennium for the universal realization of this supposed ideal of old covenant law. In this postmillennial compounding of errors the biblical concepts of the common and the holy are confounded. What the covenant of common grace makes secular is sacralized and what the covenant of redemptive theocracy makes sacred is secularized.

It appears then that certain varieties at least of premillennialism and postmillennialism are not compatible with the biblical doctrine of common grace, so important in Reformed theology.

The amillennial position, on the other hand, is altogether consistent with it.

To propose that ecclesiastical policy should be changed so as to curtail or even reduce confessional liberty as to millennial position is no doubt to propound an idea whose time has not yet come. Nevertheless, that policy is a symptom of the neglect that the area of eschatology has suffered in the enterprise of systematic theology, which is mandated by the unity of biblical truth. Perhaps, ready or not, we are being drawn indirectly into a more thoroughgoing integration of eschatology into the Reformed system of theology. For within the Reformed community movements have emerged, admirable for their concern and, in many ways, achievement in formulating a Christian world view and analyzing biblical law, but radically flawed by their failure to do justice to precisely those principles we have found to be critical for a systematic treatment of eschatology, the biblical principles of the common and the holy.