Romans 9:5

Romans 9:5 sa,rka( o` w'n evpi. pa,ntwn qeo.j euvloghto.j eivj tou.j aivw/naj

Since the earliest manuscripts of the New Testament are without systematic punctuation, editors and translators of the text must insert such marks of punctuation as seem to be appropriate to the syntax and meaning. The present passage has been the object of much discussion ¹ as to whether or not Paul intended to refer qeo,j to o` Cristo,j. The chief interpretations are the following:

(a) Placing a comma after sa,rka and referring the following words to o` Cristo,j (“… who is God over all, blessed for ever”).

(b) Placing a point (either a colon or a full stop) after sa,rka and taking the following words as a clause independent of o` Cristo,j. (Several translations are possible: “God who is over all be blessed for ever!”; or “He who is God over all be blessed for ever!”; or “He who is over all is God blessed for ever.”)

(c) Placing a comma after sa,rka and a point (a colon or a full stop) after pa,ntwn. (This, which is a modification of (b), is to be translated, “… who is over all. God be [or, is] blessed for ever!”)

In deciding which punctuation should be used, the Committee was agreed that evidence from the Church Fathers, who were almost unanimous in understanding the passage as referring to o` Cristo,j, is of relatively minor significance, as is also the opposing fact that four uncial manuscripts (A B C L) and at least twenty-six minuscule manuscripts have a point after sa,rka, either by the first hand or by subsequent correctors. ² In both cases the tradition, whether patristic or palaeographical, originated at a time subsequent to Paul’s writing (i.e. dictating; cf. Romans 16:22) the passage, and is therefore of questionable authority.

On the one hand, some members of the Committee preferred punctuation (a) for the following reasons:

(1) The interpretation that refers the passage to Christ suits the structure of the sentence, whereas the interpretation that takes the words as an asyndetic doxology to God the Father is awkward and unnatural. As Westcott observes, “The juxtaposition of o` Cristo.j kata. sa,rka and o` w'n k)t)l) seems to make a change of subject improbable.” ³

(2) If the clause o` w'n k)t)l) is an asyndetic doxology to God the Father, the word w;n is superfluous, for “he who is God over all” is most simply represented by o` evpi. pa,ntwn qeo,j. The presence of the participle suggests that the clause functions as a relative clause (not “he who is …” but “who is …”), and thus describes o` Cristo,j as being “God over all.”

(3) Pauline doxologies, as Zahn points out, ⁴ are never asyndetic but always attach themselves to that which precedes: with o[j evstin ( Romans 1:25); with o` w;n ( 2 Corinthians 11:31); with w|- ( Galatians 1:5; 2 Timothy 4:18; cf. Hebrews 13:21; 1 Peter 4:11); with auvtw|/ ( Romans 11:36; Ephesians 3:21; cf. 1 Peter 5:11; 2 Peter 3:18); with tw|/ de. qew|/ ( Philippians 4:20; 1 Timothy 1:17).

(4) Asyndetic doxologies, not only in the Bible but also in Semitic inscriptions, are differently constructed; the verb or verbal adjective (euvloghto,j, Heb. %WrB', Aram. %yrIB.) always precedes the name of God, and never follows it, as here. ⁵

(5) In the light of the context, in which Paul speaks of his sorrow over Israel’s unbelief, there seems to be no psychological explanation to account for the introduction of a doxology at this point.

On the other hand, in the opinion of others of the Committee, none of these considerations seemed to be decisive, particularly since nowhere else in his genuine epistles ⁶ does Paul ever designate o` Cristo,j as qeo,j. ⁷ In fact, on the basis of the general tenor of his theology it was considered tantamount to impossible that Paul would have expressed Christ’s greatness by calling him God blessed for ever. As between the punctuation in (b) and (c), the former was preferred.

The Committee also considered the possibility that by accident in transcription o` w;n had replaced an original w-n o` (cf. the preceding ver. Romans 9:4 w-n h` ui`oqesi,a … , ver. Romans 9:5 w-n oi` pate,rej), but was unwilling to introduce a conjectural emendation into the text. ⁸

¹ Among many earlier discussions pro and con, two may be singled out for special mention as representative of the two points of view. In favor of taking the words as an ascription to Christ, see William Sanday and A. C. Headlam, A Critical and Exegetical Commentary on the Epistle to the Romans, 2nd ed. (New York, 1896). pp. 233—238; in favor of taking the words separately from the preceding clause, see Ezra Abbot, “On the Construction of Romans ix:5, ” Journal of the Society of Biblical Literature and Exegesis, 1881, pp. 87—154, and idem, “Recent Discussions of Romans ix:5, ” ibid., 1883, pp. 90—112 (both articles are reprinted in Abbot’s posthumously published volume entitled, The Authorship of the Fourth Gospel and Other Critical Essays [Boston, 1888], pp. 332—410, and 411—438). For a more recent discussion, see the present writer’s contribution to Christ and Spirit in the New Testament; Studies in honour of C. F. D. Moule, ed. by Barnabas Lindars and Stephen S. Smalley (Cambridge University Press, 1973), pp. 95—112; reprinted in Metzger’s New Testament Studies (Leiden, 1980), pp. 56—74.

² So Abbot, op. cit., 1883, pp. 107 f. [= pp. 431 f.]. The presence of marks of punctuation in early manuscripts of the New Testament is so sporadic and haphazard that one cannot infer with confidence the construction given by the punctuator to the passage. For example, in Romans 9:2-4 codex Alexandrinus has a colon after mega,lh in ver. Romans 9:2, one between Cristou/ and u`pe,r and another after sa,rka in ver. Romans 9:3, and one after VIsrahli/tai in ver. Romans 9:4. Codex Vaticanus has a colon at the end of Romans 9:3, after both occurrences of VIsrah,l in ver. Romans 9:6, after VAbraa,m in ver. Romans 9:7, ~Rebe,kka in ver. Romans 9:10, and auvtou/ in ver. Romans 9:22!

³ B. F. Westcott in “Notes on Select Readings,” in Westcott and Hort, The New Testament in the Original Greek, [II], Introduction [and] Appendix, 2nd ed. (London, 1896), p. 110. Similarly Nigel Turner declares it to be grammatically unnatural that a participle agreeing with Cristo,j “should first be divorced from it and then given the force of a wish, receiving a different person as its subject” (Grammatical Insights into the New Testament [Edinburgh, 1965], p. 15).

⁴ Theodor Zahn, Der Brief des Paulus an die Römer (Leipzig, 1910) p. 433, Anm. 78.

⁵ The only instance that appears to be an exception is Psalms 68:19-20 [= LXX 67:19—20], where the Septuagint reads ku,rioj o` qeo.j euvloghto,j( euvloghto.j ku,rioj h`me,ran kaqV h`me,ran. Here, however, the first euvloghto,j has no corresponding word in Hebrew and seems to be a double translation.

⁶ Titus 2:13 is generally regarded as deutero-Pauline.

⁷ In reply it was argued that if Paul could refer to Cristo.j VIhsou/j as i;sa qew|/ ( Philippians 2:6), it is not inconceivable that on another occasion he could also refer to o` Cristo,j as qeo,j.

⁸ For an account of the history of the conjecture, see W. L. Lorimer in New Testament Studies, XIII (1966—67), pp. 385 f.