The Interpretation of Romans 8:28

By Carroll D. Osburn

Harding Graduate School of Religion, Memphis, Tennessee 38117

Romans 8:28 is a popular text, the exegetical and doctrinal complexities of which have perplexed its readers since the early patristic period. The basic problem arises from the fact that συνεργεῖ is capable of having three different subjects. The reading which has achieved the widest currency takes πάντα to be a neuter nominative plural and reads, “and we know that to them that love God all things work together for good.” This reading is prominent in the old English translations from Tyndale to KJV and is read by the Latin vulgate. Alternatively, relying upon the texts of A B, supported by copsa syrpal arm, Lachmann printed ὁ θεός after συνεργεῖ in the 1831 edition of his Greek text. Now further supported by P46 81, this reading is translated by Goodspeed, RSV, NIV, and Jerusalem Bible as, “We know that in everything God works for good with those who love him.” Yet a third possibility takes τὸ πνεῦμα of vv. 26–27 as the understood subject of συνεργεῖ, as in NEB, and reads, “and in everything, as we know, he (i.e., the Spirit) co-operates for good with those who love God and are called according to his purpose.” While each of these readings has possible claim to represent Paul’s true meaning, each also involves not only complex grammatical considerations, but serious doctrinal implications as well, and to those we now turn.

1. “All things work together for good”

In his Moffatt Commentary C. H. Dodd[1] has written that “the familiar translation is not an admissible rendering of the Greek. Paul did not write: ‘All things work together for good to them that love God.’ The literal translation is: ‘With those who love God He’ (or according to the other reading, ‘God’) ‘co-operates in all respects for good.’“ In explaining his reasons for rejecting the familiar view in favor of “God” as subject, Dodd observes, “No doubt many readers will regret the loss of a text which expresses the truth in a form so congenial to the ‘modern mind,’ which thinks so much of the universe as an orderly system of laws, and likes to believe that ‘it will all come right to the end.’ But we must be quite clear that this is not the attitude of Paul or of any other New Testament writer…. He does not say calamity is ‘somehow good,’ because it belongs to a general scheme which will all come right in the end. He says it cannot separate us from the love of God. In calamity, as in all things, God will co-operate with us for good.”

Although it might appear that in the final analysis not much difference exists between Dodd’s view and the traditional interpretation, in reality Dodd has charged the popular view with a kind of “universal optimism” more akin to Stoicism than to Paul. In the Stoic attempt to explain the presence of evil in the world, the doctrine of Fatalism was advanced in which there is a universal “reason according to which past events have happened, present events are happening, and future events will happen” (Stobaeus, Eclogae). God, whom the Stoics envisaged as a creative “force” within things, was thought to will good in his overall scheme of things, “but to achieve good, he has to resort to the use of means which in themselves are not without drawbacks.”[2] Stoicism, then, advocated a resignation to one’s Fate; whatever happens must be construed as “good,” the will of God (Cicero, De finibus 2.34; 3.14). Thus when Plutarch (De Stoic. repugn. 1050e) and Seneca (Ep. 74, 20) proclaim the Stoic view that “all things work unto good,” it is meant by them that everything happens through universal reason, or Fate, and one therefore should live in harmony with universal nature, resigning himself to whatever happens as “good.” As Epictetus (Discourses 2.14.7) put it, “We must put our will in harmony with events so that whatever happens will be to our liking.” Dodd’s observation that such “universal optimism” is un-Pauline is cogent,[3] for the tribulation, distress, persecution, famine, nakedness, and peril of v. 35 are not presented as “good.”

Certainly Rom 8:28 is not to be understood that in a magical sense all will somehow turn out well in the end for the Christian. While many have used Ps 91:7 in this way, “a thousand may fall at your side, and ten thousand at your right hand, but it (death) will not come near you,” many a minister has found it difficult to explain to parents that while the others in the foxhole were spared, their Christian son was killed by enemy fire. Gen 50:20, “you meant it for evil, but God meant it for good,” is suitable when there is a happy ending, but how does one explain Rom 8:28 to parents whose missionary son has been shot by leftist guerrillas in some third world country, or to parents whose daughter has been killed in an automobile accident, while the drunk who occasioned the collision still lives? As Archibald MacLeish put it so poignantly in his J.B., “If God is God He is not good, If God is good He is not God.” Whatever else Paul may have intended to connote in this verse, it is certain that he did not intend to infer that Christians would be spared innocent sufferings. Indeed, from v 18 Paul’s point has been that the whole creation groans with distresses and injustices and the concluding section (vv 31–39) emphatically underscores that Christians will not be spared the vicissitudes of life, but on the other hand, such evils need not preclude Christian hope. In v. 28 Paul’s meaning differs radically from Stoic resignation to Fate in that the unfortunate occurrences of life are not considered “good.” His point, as we shall see, is that these unfortunate experiences need not separate us from the love of God.

Now grammatically there is no problem in taking a neuter plural substantive as the subject of the singular συνεργεῖ. Theodor Zahn,[4] Paul Althaus,[5] Otto Michel,[6] Ernst Käsemann,[7] C. E. B. Cranfield,[8] and C. K. Barrett[9] have presented the more popular rendering in their commentaries. Cranfield[10] has given the most thorough treatment of the arguments that may be made in defense of the reading, arguing that it is possible to take πάντα as the subject, yet not imply “universal optimism.” The reason why all things assist Christians, says Cranfield, is “that God is in control of all things.” But is not this actually an argument favoring “God” as subject? It is certainly not unimportant that in his argument against the NEB reading, Cranfield (p. 207) is forced to argue that ὁ θεός must be supplied with the verb in v. 28 due to the proximity of τὸν θεόν, thus destroying his own argument. The traditional rendering of the verse is incapable of withstanding rigorous analysis.

2. “God works all things together for good”

Sanday and Headlam[11] advocated nearly a century ago that those manuscripts which read ὁ θεός after συνεργεῖ have preserved the original text, and this reading was given in free translation by Moffatt, leading Dodd to observe, “In verse 28, Dr. Moffatt has corrected a serious mistranslation in the Authorized Version.” However, although this reading has the distinct advantage of not implying an unhealthy “universal optimism,” there are certain objections which have been lodged against it. For one thing, from a text-critical point of view the longer text was a very early Egyptian reading, but since συνεργεῖ may be taken to imply a personal subject ὁ θεός appears to have been only a natural explanatory addition made by a later scribe.[12] Kenneth Clark’s[13] contention that the support of the Chester Beatty papyri tips the scales in favor of an original ὁ θεός has not found acceptance among textual critics. Too, Matthew Black[14] has argued the longer reading to be stylistically difficult in view of the preceding “to those who love God.” In addition, πάντα is read with συνεργεῖ in a transitive sense and questions have been raised as to whether this is possible.

3. “He (God) works in all things for good”

Now there were a number of early patristic writers who used texts without ὁ θεός who nevertheless understood “God” to be the subject of the verb. Origen, for instance, in his commentary on John 20:23 clearly takes God as subject.[15] Sanday (p. 215) cites Chrysostom as arguing at some length as if he were viewing God as subject. Further examples of this understanding are cited from Cramer’s Catena in Gennadius and Theodoret. While these examples hardly support ὁ θεός in the text, they do point to an early interpretation which later gave rise to the reading in the Alexandrian texts.

But what about the intransitive verb? To what does πάντα refer? As we have noted, Sanday, preferring ὁ θεός as original and taking συνεργεῖ as transitive, rendered, “causes all things to work together.” However, as far back as Theodore Beza’s remark in the apparatus criticus of his 1589 Greek testament, it has been held that συνεργῶ ισ an intransitive verb, and Cranfield[16] adduces this as the strongest evidence against “God” as the understood subject. Moulton[17] and Milligan[18] advanced the notion that in Rom 8:28 συνεργῶ is indeed transitive, but against this Griffiths[19] has presented compelling evidence that in each of the examples cited by these scholars the accusative may not be the direct object after the verb, but rather an Inner Accusative after an intransitive verb. He lists: Dittenberger, Orientis Graeci Inscriptiones Selectae 45.10.11, πολλὰ συνήργησε τοῖ“ ᾿Ιτανίοις, “He provided the Itanians with many things,” Pap. Lond. 358.6, τῶν δεῖνα συνεργούντων ἀλλήλοις, “who help one another in distress,” as examples of an Inner Accusative after συνεργῶ. Similar usages are Polybius 11.9.1, ἔφη…πολλὰ συνεργεῖν τὴν ἐκ τῆ“ ἐπισκενῆ“ σ̔ρμογὴν τῶν ὅπλων εἰς τὴν χρείαν, “He said…that the fitting up of arms from the prepared store contributed a great deal to their advantage,” and Philodemus, Περὶ Κακίων 12 [ὥστε] μικρὸν ἕρ[γ]ον [συν]εργῶν, “so that contributing a small deed.” Accordingly, Black is led to suggest that if God is the understood subject in v. 28, one is not bound to take the verb as transitive (as did Sanday) and that πάντα could well be taken here as an Inner Accusative (“in all things”), thus rendering, “works for good in all things for those who love God,” as Dodd proposes. One problem Black finds with this reading is Zahn’s[20] critique that if this were Paul’s meaning he would surely have written ἐν πᾶσιν rather than πάντα. To this it might be objected that ἐν πᾶσιν could then be taken confusedly with τοῖ“ ἀγαπῶσιν rather than with the verb. One might further object that Bauer[21] cites a close parallel in Alexander Aphrodisiensis, De fato 31, εἰς ἀγαθὸν οὐδὲν ὁ Πύθιος τῷ Λαίῳ συνεργεῖ, “in no respect does Apollo work with Laius for good.” Another problem Black detects with Dodd’s proposal is that God is never said in the New Testament to “co-operate” with man, but Pack[22] rightly observes that this does not seem an important objection in view of Phil 2:13, “it is God who worketh in you.” At any rate, the possibility of reading the text with God as the understood subject and πάντα as an Inner Accusative certainly has much more to commend it than does the traditional interpretation or the reading advocated by Sanday.[23]

One might ask, then, if this view is less fraught with difficulty than the previous alternatives, why did the view that πάντα is the subject, as in KJV, become the traditional view? The answer is that in the Latin a neuter plural subject requires a plural verb and the Vulgate translation scimus autem quoniam diligentibus Deum omnia cooperantur in bonum automatically ruled out any possibility of an unexpressed “God” being taken as subject. During the millennium prior to the beginning of the English Bible tradition when the Vulgate reigned as the textus dominos, “all things work together” was the reading of the Latin Bible. It is to this tradition that the reading of the English tradition is indebted.

4. “He (the Spirit) works in all things for good”

A further interpretation of this verse occurs in the NEB, where τὸ πνεῦμα of vv 26–27 is the understood subject of συνεργεῖ. This view is an ancient one, occurring in Diodorus.[24] In more recent times it has been accepted into the commentaries of Black,[25] E. Best,[26] F. F. Bruce,[27] and J. A. T. Robinson.[28]

This interpretation has been given fuller explanation by Wilson,[29] but finds classic expression in Black’s contribution to the Cullmann Festschrift.[30] According to this view, it is much less difficult to understand “the Spirit” as the subject since this is in fact the subject of the sentences in the preceding context. The passage runs thus: The Spirit assists our weakness. The Spirit pleads for us with inarticulate groanings, and He that searcheth the hearts knows what the Spirit means. And we know that with all those who love God he (the Spirit) co-operates for good, namely, with those who are called according to His purpose. Just as the Spirit is to be read with ἐντυγχάνει in v 27, so the Spirit is to be supplied with συνεργεῖ in v 28. Thus the Spirit assists our weakness in two ways, viz., pleading and working together.

As with the other interpretations, this view also has met with certain objections. Cranfield[31] has produced the only thorough critique of this interpretation and has isolated three areas of difficulty. 1) A serious objection to the Spirit as subject is the difficulty of adducing examples of συνεργεῖν used transitively in a sense which would be appropriate here. Griffiths,[32] however, has cited numerous examples of an Inner Accusative after the intransitive συνεργεῖν, so while Cranfield’s objection has merit against the view of Sanday (who takes the verb as transitive), it has no real force against τὸ πνεῦμα here. 2) Cranfield further objects that it is difficult to envisage Paul leaving the subject of the ὅτι clause unexpressed. Wilson[33] anticipated this objection and posited a textual corruption in which πάντα is a corruption of τὸ πνεῦμα. Now Black (p. 172) is correct in observing that palaeographically this is entirely possible if an original ΠΝΑ led to the primitive error ΠΑΝ (as in P46) out of which came the πάντα of the extant manuscripts. ΠΝΑ does occur in John 7:39; 20:22; and Jude 19 absolutely for “the Spirit,” although not in Paul. However, no manuscript support exists for this conjecture and emendation at this point is unwarranted. It is best to view πᾶν in P46 as reflecting nothing more than the carelessness of the scribe. Difficulty with an unexpressed subject, however, should not be pressed too far in view of the unexpressed subject in vv 29–30 which follow.

It seems that Cranfield’s strongest objection to Black’s understanding is his contention that τὸ πνεῦμα as understood subject in v 28 introduces a harsh change of subject between v 28 and vv 29–30. No subject is expressed in vv 29–30, but it is certain that all finite verbs in those two verses must have the same subject and, in view of the reference to “his Son” in v 29, the subject must be “God.” The proximity of τὸν θεόν, as Cranfield observes, makes it easier to supply ὁ θεός as the subject of v 28, which preserves the continuation of “God” as subject in vv 29–30. The NEB, taking “Spirit” as subject, has to insert “God” at the beginning of v 29, which Cranfield holds to be an indication of the harshness of the unannounced change of subject. However, the harshness detected by Cranfield is not so severe as one might be led to suppose when one observes that all the verbs in vv 27–30 have unexpressed subjects which must be ascertained by the context.

While the apparent strength of this interpretation is in its attempt to read the verse in context rather than in isolation, a more telling criticism against “Spirit” as understood subject lies in whether discourse analysis of the context actually supports “Spirit.” A question of literary analysis exists as to whether vv 26–30 constitute the section, as in NEB and others, or whether vv 26–27 and 28–30 constitute separate units as in RSV and NIV. It may be suggested that the division of RSV/NIV is unfortunate for whereas οὖν in v 31 undoubtedly initiates a new unit of text, no such discourse marker occurs at v 28. The section in which v 28 occurs actually begins with ὡσαύτως δὲ καί in v 26, as in the UBS Greek Testament. But to what does ὡσαύτως refer? And how are we to understand the contribution of v 28 to the developing point under consideration in Paul’s discourse?

The major topic of 8:1–11 is the contrast between “life in the Spirit” and “life in the flesh.” The literary thrust of Rom 8 is then focused in vv 12–13 in the categorical imperative to live after the Spirit, not after the flesh.[34] This call to Christian commitment is underscored in vv 14–15 with the observations that “as many as are led by the Spirit are sons of God,” “have received the Spirit,” and “cry ‘Abba Father.’“ The true value of this sonship in facing life in all of its complexity is thematic through v 39. The role of the Spirit in this sonship is specified as 1) bearing witness with our spirits that we are sons of God and heirs (vv 10–17), and 2) ὡσαύτως[35] interceding in our weakness (v 26). The former task involves the confirmation of sonship in the midst of a groaning world and sustains the Christian hope that the present perplexities are nothing in view of the salvation which accrues in true sonship (vv 18–25). The latter task involves the Spirit’s awareness of the human dilemma and the ability to express accurately feelings which are not easily or carefully verbalized.[36] Within vv 26–30, the subject shifts from “Spirit” to “God,” but precisely where? Cranfield has noted some difficulty in locating that switch at the beginning of v 29. It is more likely that the change of subjects has occurred as early as v 27, however, where “He who searches the hearts” (i.e., God) “knows what is the mind of the Spirit.” This reference to the Father who listens to the Spirit quite naturally would be followed by the readers’ question as to whether God will then act on behalf of Christians. Anticipating this query, v 28 underscores Paul’s confidence that He who hears does in fact work in all things with those who believe. “For good,” then, may be taken specifically to refer to the matters mentioned in vv 29–30, [37] i.e., God understands the pliglit of His children as they must cope with a problematic world and provides a way in which stresses can be overcome which otherwise would overcome them.

Having made this point, Paul asks in v 31, “If God be for us, who can be against us?” and in v 35, “Who can separate us from the love of Christ (or God, in certain manuscripts)?” His answer is that not “tribulation, or peril, or famine, or nakedness, or sword.” The citation of Ps 44:22, “We are being killed all day long as sheep led to the slaughter,” can only mean that Paul knew that Christians, like others, must face troubles of various sorts. His abiding confidence was not that believers would be spared the unfortunate experiences of life, but that absolutely nothing in this life need separate the believer from God’s love in Jesus Christ (vv 37–39).

To conclude, the long-standing view that Rom 8:28 implies reconciliation to “Fate” is a misunderstanding of the text best relegated to the history of exegesis. The insertion of ὁ θεός in certain manuscripts has no clear claim to authenticity and the rendering of Sanday which infers “God” as subject, but treats συνεργεῖ as transitive, is indefensible linguistically. The view that “Spirit” is the understood subject is interesting, but difficult to accept in view of the harsh transition to “God” as subject in v 29 and in view of “God” being the subject already in v 27. The most acceptable view is Dodd’s suggestion that “God” from the preceding phrase is the understood subject and that πάντα is an Inner Accusative. “God works in all things for good” is an excellent rendering of the text which coheres fully with the context and with Pauline thought and which presents no serious grammatical difficulties.

Notes

1. Dodd, The Epistle of Paul to the Romans (MNTC; London: Hodder and Stoughton, 1932) 137–138.
2. Émile Bréhier, The Hellenistic and Roman Age (trans. W. Baskin; Chicago: Univ. of Chicago Press, 1965) 54.
3. There is no reason to suppose with E. C. Blackman, “A Further Note on Romans viii.28 ,” ExpTim 50 (1939) 378–379, that Paul has been influenced by the “words of some Stoic preacher.”
4. Zahn, Der Brief des Paulus an die Römer (Leipzig: A. Deichert, 1910) 414, n. 38.
5. Althaus, Der Brief an die Römer (Göttingen: Vandenhoeck und Ruprecht, 1932) 76.
6. Michel, Der Brief an die Römer (MeyerK; Göttingen: Vandenhoeck und Ruprecht, 1963) 209–210.
7. Käsemann, An die Römer (HNT; Tübingen: J. C. B. Mohr, 1974) 232.
8. Cranfield, The Epistle to the Romans (ICC; Edinburgh: T.&T. Clark, 1975) 1.427f.
9. Barrett, A Commentary on the Epistle to the Romans (HNTC; New York: Harper and Bros., 1957) 169, apparently following W. L. Knox, St. Paul and the Church of the Gentiles (Cambridge: Cambridge Univ, Press, 1939) 105.
10. “Romans 8.28,” SJT 19 (1966) 204–215.
11. W. Sanday and A. C. Headlam, The Epistle to the Romans (ICC; Edinburgh: T.&T. Clark, 1896) 215.
12. B. M. Metzger, A Textual Commentary on the Greek New Testament (corr. ed.; New York: U.B.S., 1975) 518.
13. “Textual Criticism and Doctrine,” Studia Paulina: In Honorem Johannis de Zwaan (eds. J. N. Sevenster and W. C. van Unnik; Haarlem: De Ervan F. Bohn, 1953) 57. Also Frank Pack, “A Study of Romans viii.28 ,” Restoration Quarterly 22 (1979) 52–53. Knox’s suggestion in St. Paul and the Church of the Gentiles, 105, that P46 might represent an original τὸν θεὸν τὸ πᾶν συνεργεῖ, “the Universe co-operates with those who love God,” is unlikely as πᾶν there is likely only an accidental change to the singular.
14. “The Interpretation of Romans viii 28 ,” Neotestamentica et Patristica: Eine Freundesgabe, Herrn Professor Dr. Oscar Cullmann (ed. W. C. van Unnik; NovT Sup 6; Leiden: Brill, 1962) 168.
15. On Rom 7:7 Origen has the traditional interpretation, but Knox, St. Paul and the Church of the Gentiles, 105, questions whether this may merely represent a revision by Rufinus in the Latin.
16. “Romans 8.28,” 208–209.
17. J. H. Moulton, A Grammar of New Testament Greek (3rd. ed.; Edinburgh: T.&T. Clark, 1967) 65.
18. J. H. Moulton and G. Milligan, The Vocabulary of the Greek Testament (London: Hodder and Stoughton, 1930) 605. F. Blass and A. Debrunner, A Greek Grammar of the New Testament (trans. R. Funk; Chicago: Univ. of Chicago Press, 1961) 82, cite ἐνεργεῖν as an example of the transitive use of an original intransitive, “to work at something,” and mention Rom 8:28 to be the same usage with συνεργεῖν.
19. J. Gwyn Griffiths, “Romans viii.28 ,” ExpTim 49 (1938) 474–476.
20. Zahn, Der Brief des Paulus an die Römer, 414, n. 38.
21. W. Bauer, A Greek-English Lexicon of the New Testament (Chicago: Univ. of Chicago Press, 1957) 795.
22. Pack, “A Study of Romans viii.28 ,” 51, n. 26.
23. See among others, Heinrich Schlier, Der Römerbrief (HTKNT; Freiburg: Herder, 1979) 270–271; Georg Bertram, TWNT 7.873; A. T. Robertson, Word Pictures in the New Testament (Nashville: Broadman Press, 1931) 4.377; and Hans Lietzmann, An die Römer (HNT 8; 5th ed.; Tübingen: J. C. B. Mohr, 1971) 87.
24. K. Staab, Pauluskatenen aus der griechischen Kirche (NTAbh 15, Münster: Aschendorff, 1933) xxv, 95, 141.
25. Black, Romans (New Century Bible; Greenwood, S.C.: Attic Press, 1973) 124.
26. Best, The Letter of Paul to the Romans (Cambridge Bible Comm., Cambridge: Cambridge Univ. Press, 1967) 100–101.
27. Bruce, The Epistle of Paul to the Romans (Tyndale NT Comm.; Grand Rapids: Eerdmans, 1963) 176.
28. Robinson, Wrestling with Romans (London: SCM, 1979) 104–105.
29. J. P. Wilson, “Romans viii.28 ,” ExpTim 60 (1948) 110–111.
30. Black, “The Interpretation of Romans viii 28 ,” 166–172.
31. Cranfield, “Romans 8:28, ” 206ff.
32. Griffiths, “Romans viii.28 ,” 475.
33. Wilson, “Romans viii.28, 111.
34. Introduced by ἄρα οὖν, this text sums up the argument of the entire section. See M. Thrall, Greek Particles in the New Testament (Leiden: E.J. Brill, 1962) 11.
35. With Sanday and Headlam, The Epistle to the Romans, 213, against H. C. G. Moule, The Romans (Cambridge: Cambridge Univ. Press, 1952) 153.
36. Note Käsemann’s suggestion (An die Römer, 230-231) that glossolalia is meant. Cf. Sanday, The Epistle to the Romans, 213, and A. Schlatter, Der Brief an die Römer (Stuttgart: Calwer Verlag, 1974) 162ff.
37. M. Zerwick, Biblical Greek (Rome: Pontifical Biblical Institute, 1963) 143, observes that ὅτι often functions to provide the basis of the preceding statements, as here.