The history of the doctrine of universal salvation (or apokastastasis) is a remarkable one. Until the nineteenth century almost all Christian theologians taught the reality of eternal torment in hell. Here and there, outside the theological mainstream, were some who believed that the wicked would be finally annihilated (in its commonest form. this is the doctrine of 'conditional immortality'). Even fewer were the advocates of universal salvation, though these few included same major theologians of the early church. Eternal punishment was firmly asserted in official creeds and confessions of the churches. It must have seemed as indispensable a part of universal Christian belief as the doctrines of the Trinity and the incarnation. Since 1800 this situation has entirely changed, and no traditional Christian doctrine has been so widely abandoned as that of eternal punishment. Its advocates among theologians today must be fewer than ever before. The alternative interpretation of hell as annihilation seems to have prevailed even among many of the more conservative theologians. Among the less conservative, universal salvation, either as hope or as dogma, is now so widely accepted that many theologians assume it virtually without argument.
The history is a complex one, partly because the issue of hell and universalism is closely interconnected with other difficult and debated theological issues, such as predestination and free will, the validity of retributive punishment, the authority of the Bible, and (most centrally) the nature of God, the meaning of and the relationship between His love and His justice. The issue of universal salvation is not related to these other issues in a straightforward way. Absolute predestination, for example, has been held to be the basis either for a doctrine of hell (Augustine, Calvin) or for a doctrine of universal salvation (Schleiermacher), while, conversely, free will has been held to support a doctrine of hell (C. S. Lewis) or a form of uni-
versalism (Origen). Nineteenth-century advocates of universalism frequently emphasized the role of retributive punishment in their scheme, but more modern universalists often reject hell as a result of rejecting the idea of retributive punishment, Thus the problem of universalism cannot be reduced to a simple choice of alternatives. Only the belief :hat ultimately all men will be saved is common to all universalists. The rationale for that belief and :he total theological context in which it belongs vary considerably.
The most famous and influential advocate of universalism in the early church was Origen, whose teaching on this point was partly anticipated by his predecessor Clement of Alexandria. Origen's universalism belongs to the logic of his whole theological system, which was decisively influenced by his Platonism and depended on his hermeneutical method of discerning the allegorical sense of Scripture behind the literal sense. According to Origen all intelligent beings (men, angels, devils) are created good and equal, but with absolute free will. Some, through the misuse of free will, turned from God and fell into varying degrees of sin. Those who fell furthest became the devils, those whose fall was less disastrous became the souls of men. These are to be restored to God through a process of discipline and chastisement, for which purpose this material world has been created and the pre-existing souls incarnated in human bodies. The process of purification is not complete at death but continues after this life. Nor is it an inevitably upward path: the soul remains free to choose good or evil, and so even after this life may fall again as well as rise. Within this scheme punishment is always; in God's intention, remedial: God is wholly good and His justice serves no other purpose than His good purpose of bringing all souls back to Himself. Thus the torments of hell cannot be endless, though they may last for aeons; the soul in hell remains always free to repent and be restored.
Logically it might seem that Origen's conviction of the inalienable freedom of the soul ought to prevent him from teaching both universalism (for any soul is free to remain obstinate for ever) and the final secure happiness of the saved (who remain free to fall again at any time). In fact Origin seems to have drawn neither conclusion. Given unlimited time, God's purpose will eventually prevail and all souls will be finally, united to Him, never to sin again. The final restoration includes even Satan and the devils.
Origen's scheme conforms to a Platonic pattern of understanding the world as part of a great cycle of the emanation of all things from God and the return of all things to God. Despite the appeal to such texts as 1 Cor. 15: 28 ('God shall be all in all' this has always been a favourite universalist text) the final unity of all things with God is more Platonic than biblical in inspiration. The Platonic pattern of emanation and return was widely influential in Greek theology and provided the same kind of general world-view favouring universalism as Darwinian evolution was to provide for some nineteenth-century universalists, In both cases universalism is achieved by seeing both this earthly life and hell as only stages in the soul's long upward progress towards God, whereas mainstream Christian orthodoxy has always regarded this life as decisive for a man's fate and hell as the final destiny of the wicked.
The doctrine of the final restoration of all souls seems to have been not uncommon in the East during the fourth and fifth centuries. It was clearly taught by Gregory of Nyssa and is attributed to Diodore of Tarsus, Theodore of Mopsuestia, and some Nestorian theologians. Others, such as Gregory of Nazianzus, regarded it as an open question. Augustine took the trouble to refute several current versions of universalism, as well as views on the extent of salvation which stopped short of universalism but were more generous than his own.
Origen's universalism was involved in the group of doctrines known as 'Origenism', about which there were long controversies in the East. A Council at Constantinople in 543 condemned a list of Origenist errors including Apokatastasis, but whether this condemnation was endorsed by the
Fifth Ecumenical Council (553) seems in doubt. At any rate the condemnation of Origenism discredited universalism in the theological tradition of the East. In the West, not only Origen's heretical reputation but also Augustine's enormous influence ensured that the Augustinian version of the doctrine of hell prevailed almost without question for many centuries. During the Middle Ages universalism is found only in the strongly Platonic system of John Scotus Erigena (dc 877) and in a few of the more pantheistic thinkers in the mystical tradition, for whore the divine spark in every man must return to its source in God.
The intellectual and religious upheaval of the sixteenth and seventeenth centuries produced some examples of almost every possible religious opinion, and so it is not surprising to find some universalists. A few sixteenth-century Anabaptists and Spiritualists, notably John Denck, and a few of the most radical religious thinkers of the English Interregnum, notably Gerrard Winstanley and Richard Coppin, were universalists.
Universalism in the seventeenth century should be seen partly as reaction to the particularism of high Calvinism, which with its doctrine of limited atonement excluded any kind of divine will for the salvation of all men. Revulsion against the apparent cruelty of the God who created the reprobate for no other purpose than to damn him, led firstly to Arminianism, in which the Gospel genuinely offers salvation to all men; a further step leads to the Quaker doctrine that saving grace is given to all men, but may be resisted; the extreme position is that all men will actually be saved. A further factor promoting universalism was the Platonic tradition, revived during the Renaissance, along with an interest in Origen and the early Greek Fathers, who could plausibly be thought to represent a form of Christian doctrine earlier, and therefore purer, than Augustine, to whom the Calvinists appealed.
So it is no surprise to find that some of the Cambridge Platonists in seventeenth-century England were universalists. Peter Sterry and Jeremiah White held the Platonic scheme of emanation and return, the preexistence of souls, and the remedial character of all punishment. The love of God is His supreme attribute. His wrath is an aspect His love, not against the sinner bat against the sin. So the sinner's torment in hell will be the agony of enduring God's holy burning love until his sins are burned up and he himself is pure.
Universalism also appears at the end of the seventeenth century among some of the German Pietists, and was again popularized in eighteenth-century England especially by the devotional writer William Law.
One very strong objection to universalism in these centuries was the deep-rooted belief that the threat of eternal torment was a necessary deterrent from immorality during this life. So weighty was this objection felt to be, that some who believed in universal salvation (or even in annihilation) held that this belief must retrain an esoteric, secret doctrine for the few, while hell must continue to be preached as a deterrent for the masses. Even in the nineteenth century, where such esotericism was seen to be indefensible, universalists found it necessary to meet the objection by emphasizing as much as possible the severity and length of the torments which the wicked must endure before their eventual salvation.
F. D. E. Schleiermacher was the first great theologian of modern times to teach universalism. He taught a predestination as absolute as that of Augustine and Calvin, but he rejected any form of double predestination. All men are elected to salvation in Christ, and this purpose of divine omnipotence cannot fail. In this respect Schleiermacher represents a 'Reformed' universalism, founded on the all-determining will of God. Only a Pelagian, on this view, could argue from human
free will to the possibility of hell. But Schleiermacher rejects what he sees as the capriciousness of the Calvinist God who arbitrarily elects only some men to salvation. For Schleiermacher a sense of the unity of the human race is a high virtue in then and cannot be thus disregarded by God.
Most interesting of Schleiermacher's arguments against hell is his deeply felt conviction that the blessedness of the redeemed would be severely marred by their sympathy for the damned. This is precisely the opposite of the conviction of many earlier theologians that the blessedness of the redeemed would be actually enhanced by their contemplation of the torments of the damned. The latter view has a kind of reason on its side: Those who are wholly at one with God's will should rejoice to see His justice done. But it has largely disappeared from the doctrine of hell since the seventeenth century, and the modern Christian's instinctive sympathy with Schleiermacher's contrary view places him on Schleiermacher's side of a great transition in the history of attitudes to suffering. With Schleiermacher we now feel that even the justly inflicted suffering of other men roust be pitied, not enjoyed. Schleiermacher's argument is typically modern in its appeal and is one element in the increasing popularity of universalism since his day.
Schleiermacher's universalism had surprisingly few successors in nineteenth-century Germany, but in nineteenth-century England the problem of hell and universal salvation (with other aspects of the future life) became a matter of widespread concern. This can be gauged from the attention given to three cases, all notorious in their day. In 1853 F. D. Maurice was dismissed from his professional chair at King's College, London, for what was little more than a cautious modification of the traditional doctrine of hell: a storm of controversy brake over this 'proto-martyr of the wider hope'. There in 1862, for his very tentative assertion of universalism in Essays and Reviews (1860), H. B. Wilson was condemned in the Court of Arches, guilty of contradicting the Athanasian Creed, though the judgment was subsequently reversed on appeal by the Lord Chancellor. Thirdly, F. W. Farrar denied eternal punishment in a famous series of sermons in Westminster Abbey in 1877 (published as Eternal Hope, 1878), though he remained agnostic as to the alternatives. But he was commonly understood to be teaching universalism, and his sermons provoked a learned defence of the traditional hell from E. B. Pusey.
Dogmatic universalism was in fact much less common in nineteenth-century England than a general uneasiness with the traditional doctrine of hell. This led to arguments for conditional immortality; to undogmatic hopes for universal salvation; and to the idea that a man's fate is not sealed at death, but that the intermediate state offers fresh opportunities for attaining salvation. The two leading dogmatic universalists were Andrew Jukes (The Second Death and the Restitution of all things, 1867) and Samuel Cox (Salvator Mundi, 1877).
Common to almost all versions of the 'wider hope' was the belief that death was not the decisive break which traditional orthodoxy had taught. Repentance, conversion, moral progress are still possible after death. This widespread belief was certainly influenced by the common nineteenthcentury faith in evolutionary progress. Hell - or a modified version of purgatory - could be understood in this context as the pain and suffering necessary to moral growth. In this way evolutionary progress provides the new context for nineteenth century universalism, replacing the Platonic cycle of emanation and return which influenced the universalists of earlier centuries.
The transition from Victorian to more modern forms of universalism is marked by some changes of which the most important concerns exegesis. Almost all universalists before this century thought it necessary to argue for a universalist interpretation of those texts of the NT which seem to teach eternal punishment or final condemnation, and the standard approach to such texts was to deny the everlasting or final character of the punishment. Texts such as Matthew 25: 46 or even Revelation 14: 10f. were held to refer to a very long but limited period of torment in hell, from which the sinner will eventually emerge to salvation. The nineteenth-century debates always included extensive exegetical discussions, especially over the meaning of aionios. In this century, however, exegesis has turned decisively against the universalist case. Few would now doubt that many NT texts clearly teach a final division of mankind into saved and lost, and the most that universalists now commonly claim is that alongside these texts there are others which hold out a universal hope (e.g. Eph. 1: 10; Col. 1: 20).
There are two ways of dealing with this situation. One is a new form of exegesis of the texts about final condemnation, which acknowledges the note of finality but sees these texts as threats rather than predictions. A threat need not be carried out. This, as we shall see, is the approach adopted by the most persuasive of modern universalists.
The second approach to the exegetical problem is simply to disagree with the NIT writers' teaching about a final division of mankind, which can be said to be merely taken over from their contemporary Jewish environment, while the texts which could be held to support universalism represent a deeper insight into the meaning of God's revelation in Christ. Here the doctrinal authority of the Bible is understood much more flexibly than by most nineteenth-century universalists. C. W. Emmet's essay, 'The Bible and Hell' (1917), is something of a landmark. After a survey of the NT material, showing that final division and judgment are clearly taught and hesitating to find full universalism even in Ephesians and Colossians, Emmet declares: It is best in fact to admit quite frankly that any view of the future destiny of [unbelievers] which is to be tolerable to us today must go beyond the explicit teaching of the New Testament.... [This] does not really give us what we want, and it only leads to insincerity if we try to satisfy ourselves by artificial explanations of its language. And we are in the end on surer ground when as Christians we claim the right to go beyond the letter, since we do so under the irresistible leading of the moral principles of the New Testament and of Christ Himself.'
Thus the modern universalist is no longer bound to the letter of the NT; he can base his doctrine on the spirit of NT teaching about the love of God. The same principle can even be extended to the teaching of the historical Jesus, though some have been able to persuade themselves that the Gospel texts about final judgment are not in any case authentic words of Jesus. This more liberal approach to Scripture has probably played quite a large part in the general spread of universalism in this century.
Neither Karl Barth nor Emil Brunner was strictly a universalist, but both regarded the final salvation of all mankind as a possibility which cannot be denied (though it cannot be dogmatically asserted either). This is a significant step beyond traditional theology, which always asserted not only that final condemnation is a real possibility but also that some men will actually be lost. It is also a position which has probably had more appeal to conservative Christians (including Roman Catholic theologians) than dogmatic universalism; it allows us to hope for the salvation of all men without presuming to know something which God has not revealed.
Barth refashioned the Reformed doctrine of predestination by making it fully Christological. It is Jesus Christ who is both rejected and elected. The rejection which sinful man deserves, God has taken upon Himself in Jesus Christ, and in Jesus Christ all men are elected to salvation. He is therefore in the true sense the only rejected one. Predestination thus becomes not an equivocal doctrine of God's Yes and No, but a fully evangelical doctrine of mood's unqualified Yes to man. The reality of man - of all men - is that in Jesus Christ the reconciliation of all men has taken place. The Gospel brings to men the knowledge of what is already true of them:
that in Jesus Christ they are already elect, justified, reconciled.
It might be thought that this line of thought logically entails universalism, much as Schleiermacher's doctrine of universal election did, but Barth refuses to follow this logic. There remains an irresolvable tension between the election of all men in Jesus Christ and the phenomenon of unbelief. The unbeliever's true reality is that he is elect, but he denies that reality and attempts to change it, to be instead the rejected man. In this perverse attempt (it is no more than an attempt) he lives under the threat of final condemnation, which would be God's acquiescence in its refusal to be the reconciled man he really is.
Will this threat be carried out? Barth does not here appeal to man's freedom to continue in unbelief: he is committed to the sovereignty of God's grace. The reason why universal salvation cannot be dogmatically expected lies in God's Freedom: 'To the man who persistently tries to change the truth into untruth, God does not owe eternal patience and therefore deliverance.... We should be denying or disarming that evil attempt and our own participation in it if, in relation to ourselves or others or all men, we were to permit ourselves to postulate a withdrawal of that threat end in this sense expect or maintain an apokalastasis or universal reconciliation as the goal and end of all things.... Even though theological consistency might seem to lead our thoughts and utterances most clearly in this direction, we must not arrogate to ourselves that which can be given and received only as a free gift.' But universal salvation remains an open possibility for which we may hope.
That universal salvation must remain an open question is also the conclusion that Brunner reaches by a different route. He stresses that we must take quite seriously the two categories of NT texts: those which speak of a final decisive division of men at the Last Judgment, and those which speak of God's single unqualified will for the salvation of all men. The two are logically incompatible and are not to be artificially reconciled by attributing to God a dual will (double predestination) or by eliminating the finality of judgment. The texts are logically incompatible because they are not intended to give theoretical information. To the question 'Is there such a thing as final loss or is there a universal salvation?' there is no answer, because the Word of God 'is a Word of challenge, not of doctrine'. It addresses us and involves us. Its truth is not the objective truth available to the neutral observer, but the subjective truth of existential encounter. The message of judgment, then, is not a prediction that some will be lost; it is a challenge to me to come out of perdition to salvation. The message of universal salvation is not a prediction that all men will be saved; it is an invitation to me to make the decision of faith which accepts mood's will to save me. The Gospel holds the two together in proclamation. Theology may not objectify either.
Two of the most persuasive of recent arguments for dogmatic universalism are those of J. A. T. Robinson and John Hick. We shall conclude this survey with a brief account of their positions.
Robinson approaches the texts in a way rather similar to Brunner's. The NT contains two eschatological 'myths': universal restoration and final division into saved and lost. But whereas Brunner gives both the same status, Robinson maintains that they represent 'the two sides of the truth which is in Jesus.... Though both are the truth, one [universal restoration] is the truth as it is for God and as it is for faith the further side of decision; the other [heaven and hell] is the truth as it must be to the subject facing decision.' Hell is a reality in the existential situation of the man facing the challenge of the Gospel: the seriousness of his decision must not be weakened by universalism. But universal salvation is the reality which God wills and which therefore must come about. For all that Robinson tries to give proper weight to the myth of heaven and hell, it is clear that universalism has the last word. As God's viewpoint it has a final validity denied to the viewpoint of man in decision.
This is because, for Robinson, only universal salvation is consistent with God's nature as omnipotent love. Final judgment would be a frustration of his purpose. But what of mans
freedom to resist God's love? Omnipotent love must in the end force every man, to yield to it - not as an infringement of freedom, but as free choice elicited by love. Man's freedom is compatible with the victory of omnipotent love.
Robinson's is an eloquent expression of the case for universalism as a necessity of God's nature as omnipotent love. Hick's argument is parallel at some points. He too regards the two categories of NT texts as different kinds of statement. The warnings of hell are existential preaching, warning men that they will be damned if they permanently refuse to repent. Paul's statements about universal salvation, on the other hand, are detached theological conclusions. The two types of text are compatible because no-one will in fact permanently refuse to repent.
Hick feels the strength of the objection that universalism is incompatible with human freedom. His answer essentially is that human nature has a created bias towards God, which means that we naturally tend towards Him of our own free will. Therefore, given time, His love must in the end evoke a response from all men.
Hick's distinctive approach to universalism, however, lies in his concern for theodicy, which colours a great deal of his theology. The suffering and evil of this world can only be justified if God is going to bring to a good end every individual personal life He has created. If there is either eternal punishment or annihilation for some, then either God is not perfectly good - since He does not desire the salvation of all His creatures - or He is not omnipotent - since His purpose has finally failed in the case of some. Only universal salvation can vindicate the omnipotent good God in whom Christians believe.
More than most other modern forms of universalism, Hick's bears a striking resemblance to both the Origenist and Victorian types, in that he envisages this life as merely the first stage in a long - in many cases, unimaginably long - post-mortem progress towards final, salvation. Within this process hellish or (more properly) purgatorial experiences take their place. In his most recent work, Hick (drawing eclectically on Eastern, rather more than Christian, ideas of the future life) sketches a highly speculative account of the many subsequent lives through which men will pass in their gradual approximation to the divine purpose. It is typical of this variety of universalism. that our ultimate salvation becomes a prospect so distant as to be hardly capable of concerning us at all in this first of our many lives. This is a far cry from Jesus' message of present salvation to be apprehended or lost in immediate response to His preaching.
 For details see L. E. Froom, The Condionalist Faith of our Fathers (Washington, DC: Review and Herald, 1965-1966).
 Athanasian Creed; Fourth Lateran Council, Canon 1; Augsburg Confession, ch. 17; Second Helvetic Confession, ch. 26; Westminster Confession, ch. 33; Dordrecht Confession., art. 18.
 Already in 1914 H. R. Mackintosh could write: 'If at this moment a frank and confidential plebiscite of the English-speaking ministry were taken, the likelihood is that a considerable majority would adhere to Universalism. They may no doubt shrink from it as a dogma, but they would cherish it privately as at least a hope': 'Studies in Christian Eschatology, VII, Universal Restoration', The Expositor 8th series 8 (1914), pp. 130f.
 The following are almost random examples of 20th century English theologians who favour annihilation: C. Gore, The Religion of the Church (Oxford: Mowbray, 1916), pp. 91f.; W. Temple, Christus Veritas (London: Macmillan, 1924), p. 209; O. C. Quick, Doctrines of the Creed (London: Nisbet, 1933), pp. 257f., U. Simon, The End is Not Yet (Welwyn: Nisbet, 1964), pp. 206f.; G. B. Caird, The Revelation of St John the Divine (London: A. and C. Black., 1966), pp. 186f., 260; cf. J. W. Wenham, The Goodness of God (London: IVP, 1974, ch. 2 (leaves the issue open between everlasting punishment and annihilation). The Lausanne Covenant (ch. 3) speaks of 'eternal separation from God'. Annihilation is strictly not a possible option for Roman Catholic theologians, since it was formally condemned by the Fifth Lateran Council (1513).
 Clement's universalism is less clear than Origen's: see C. Bigg, The Christian Platonists of Alexandria (Oxford: Clarendon Press, 2nd ed., 1913), pp. 147f; W. E. G. Ford, Clement of Alexandria's Treatment of the Problem of Evil (Oxford: OUP, 1971), p. 40, n. 9, pp. 72f.
 See Bigg, op. cit., pp. 273-280, 343-351. According to H. Chadwick, Early Christian Thought and the Classical Tradition (Oxford: Clarendon Press, 1966), p. 119, universal salvation, was 'more his hope than his assured certitude'.
 Later critics of Origen accused him of denying the final security of the blessed: e.g. Augustine, De Civ. Dei 21: 17.
 Despite the general resemblance between Origen's understanding of hell and the medieval and Roman Catholic doctrine of purgatory (both concern purgatorial suffering after death); they differ very significantly in that the latter regards a man's fate as decided at death. Purgatory does not offer fresh opportunities of repentance and faith after death; it purifies those who repented and believed during their earthly life.
 E. H. Plumptre, The Spirits in Prison (London: Isbister, 2nd ed.,1893), pp. 138-140; Bigg, op. cit., p. 344.
 Bigg, op. cit., p. 343; Plumptre, op. cit., pp. 140f.
 Plumptre, op. cit., p. 141.
 Bigg, op. cit., p. 344, n. 1.
 De Civ. Dei 21: 11.7-27
 J. H. Leckie, The World to Conte and Final Destiny (Edinburgh; T. and T. Clark, 2nd ed., 1922), pp. 260-263.
 G. H. Williams, The Radical Reformation (London: Weidenfeld and Nicholson, 1962), pp. 202, 246, 252, 843; on Denck, pp. 155, 157. This Anabaptist umversalism was condemned in ch. 17 of the Augsburg Confession and in the 42nd of the English Articles of 1553.
 C. Hill, The World Turned Upside Down (London: Temple Srnith, 1972)), pp. 140-143; on Coppin, pp. 177f.
 D. P. Walker, The Decline of Hell (London: Routledge and Kegan Paul, 1964), pp. 11-18.
 On Sterry and White, see Walker, op. cit., ch. 7; for other 17th-century universalists in the Platonic tradition, chs. 8 and 10.
 Notably J. W. Petersen and F. C. Oetinger. Petersen learned his universalism from the English visionary Mrs Jane Lead. On Petersen, see Walker, op. cit., ch. 14; on Mrs Lead and the Philadelphian Society, ch. 13.
 Leckie, op. cit., pp. 267f. Another 18th-century English universalist was bishop Thomas Newton; Plumptre op. cit., pp. 203-205. Many forgotten universalist writers of the 17th and 18th centuries are listed in Ezra Abbot's bibliography appended to W. R. Alger, A Critical History of the Doctrine of a Future Life (New York: W. J. Widdleton, 3rd ed., 1878). For Joseph Priestley and Unitarian universalism, see G. Rowell, Hell and the Victorians (Oxford: Clarendon Press, 1974), ch. 3.
 Walker, op. cit., pp. 1-7. The idea goes back to Origen, C. Cels. 6: 26.
 The Christian Faith (ET, Edinburgh: T. and T. Clark, 1928), paras. 117-120, 163.
 This was taught, e.g., by Tertullian, Cyprian, Augustine, Peter Lombard, Thomas Aquinas, Robert Bellarmine: references in Walker, op. cit., p. 29.
 See Walker, op. cit., pp. 29-32.
 Note that C. S. Lewis, a competent modern apologist for hell, answers this argument of Schleiermacher's by denying that heaven and hell co-exist 'in unilinear time': The Problem of Pain (London: Bles, 1940), pp. 114f. He does not argue, as earlier theologians would have done, that pity for the justly punished would be misplaced.
 Mackintosh, art. cit., p. 134, n. 1, can name only Schleiermacher's disciple Alexander Schweizer.
 A perceptive and informative study is G. Rowell, op. cit.
 Plumptre, op. cit., p. viii. For Maurice's views and the controversy, see Rowell, op. cit., pp. 76-89. In his Lectures on the Apocalypse (1861) Maurice appears to reach a more definite universalism (cf. pp. 400-405).
 Rowell, op. cit., pp. 116-123.
 Ibid., pp. 138-147.
 Ibid., ch. 9.
 E.g. S. T. Coleridge believed universal salvation to be a possibility which, in view of 'the exceeding sinfulness of sin', might not be presumed on: ibid., pp. 67f.
 The classic statement of this idea of 'extended probation' was E. H. Plumptre, The Spirits in Prison; the sermon from which this work grew was preached in 1871. Evidence was found in 1 Peter 3 : 18-20 (hence Plumptre's title) and the traditional doctrine of Christ's descent into hell; but 1 Peter 3: 18-20 cannot really be interpreted in this way: see the extensive study (including history of exegesis) in W. J. Dalton, Christ's Proclamation to the Spirits: A Study of 1 Peter 3: 18-4: 6 (Analecta Biblica 23; Rome: Pontifical Biblical Institute, 1965). For the popularity of the idea of 'extended probation' in the 19th century, see J. Orr, The Christian View of God and the World (Edinburgh: A. Elliot, 2nd ed., 1893), p. 394.
 Rowell, op. cit., pp. 129-133.
 Attempts to show that the NT texts refer to a temporary hell to be followed by ultimate salvation are still sometimes found: cf. W. Michaelis, Versöhnang des Alls (Berlin: Siloah-Verlag; 1950); M. Rissi, The Future of the World (London: SCM Press, 1972). but they no longer carry conviction.
 In B. H. Streeter ed., Immortality (London: Macmillan, 1917), ch. 5.
 Ibid., p. 212.
 Cf. J. Hick, Death, and Eternal Life (London: Collins, 1976), pp. 243-247. Hick's case is unconvincing because it does not take full account of all Synoptic sayings about final judgment. While it can plausibly be argued that much of the imagery of hell belongs to Matthew's redaction, the warning of final judgment cannot be eliminated from Jesus' authentic words even by stringent use of the generally accepted criteria of authenticity.
 Church Dogmatics IV/3 (ET, Edinburgh: T. and T. Clark, 1961), p. 477.
 Ibid., p. 478; II/12 (ET, Edinburgh: T. and T. Clark 1957), p. 418. For criticism of earth's position on election end universalism, from two different standpoints, see J. Hick, op. cit., pp. 259-261; G. C, Berkouwer, The Triumph of Grace in the Theology of Karl Barth (London: Paternoster 1956), ch. 10. But note that Berkouwer wrote before the publication of the important passage CDIV/3, pp. 461-478.
 Eternal Hope (ET, Edinburgh: Lutterworth, 1954), ch. 17; Dogmatics III (ET, London: Lutterworth, 1962), ch. 10. For a similar approach to the problem, see G. C. Berkouwer, The Return of Christ (Grand Rapids: Eerdmans, 1972, pp. 390-419.
 Eternal Hope, p. 183.
 In the End God (London: Collins, 2nd ed., 1968), chs. 10 and 11. Robinson's case for universalism as a necessity of divine love was earlier argued in 'Universalism - Is it Heretical?, SJT 2 (1949), pp, 139-155; to which T. F. Torrance replied in 'Universalism or Election?', SJT 2 (1949), pp. 310-318.
 In the End God, p. 130.
 Death and Eternal Life, pp. 247-250.
 Ibid., pp. 250-259.
 Evil and the God of Love (London: Macmillan, 2nd ed., 1977), pp. 341-345; God and the Universe of Faiths (London: Collins, 2nd ed., 197), pp. 70-74.
 Death and Eternal Life, ch. 20.
 Hick admits this: ibid., p. 420.