The Textual History of the New Testament and the Bible Translator

Jude 5

ὑπομνῆσαι δὲ ὑμᾶς βούλομαι, εἰδότας [ὑμᾶς] πάντα ὅτι [ὁ] κύριος ἅπαξ λαὸν ἐκ γῆς Αἰγύπτου σώσας τὸ δεύτερον τοὺς μὴ πιστεύσαντας ἀπώλεσεν

I want to remind you, although [you] know all this, that [the] Lord at one time, after saving the people from Egypt, on the second occasion destroyed those that did not believe. ¹

There is substantial variation in the manuscript tradition regarding κύριος, the principal contenders being ὁ κύριος, Ἰησοῦς, and the singular reading θεὸς Χριστός found in the Bodmer papyrus of the Catholic Epistles (P72). ² The textual decision in the end comes down to the weighting one affords the lectio difficilior principle, that is, whether “Jesus” must be the right reading because it seems so theologically difficult (as the NET translators have in effect argued) ³ or whether it cannot be right because it is “difficult to the point of impossibility,” as the UBS4 committee argued in their rejection of it, giving their preference to the more ambiguous ὁ κύριος (Metzger 1994, ad loc.). The Editio Critica Maior of 2005 and its offspring NA28 and UBS5 have now definitively placed Ἰησους in the running text, and its lead is also followed by the SBL critical text of Michael Holmes. ⁴ Its presence in UBS5 will no doubt cause it to find its way into many translations around the world.

However one resolves the problem of reconstructing the critical text, of greater interest for our purposes are rather the readings that are attested in a variety of early witnesses, which indicate for us how the verse was being interpreted in the early church. The Bodmer papyrus, for instance, has the singular reading θεὸς Χριστός, a collocation so odd and apparently at variance with NT usage that the UBS4 committee considered it no more than a scribal blunder (Metzger 1994, ad loc.). The reason they so readily dismiss its claim to be a serious variant may be found in their reticence to grasp how anybody could really believe that it had been Christ himself who had saved the Israelites in Egypt, the same aversion to theological aberrations that led to their rejection of the much better-attested Ἰησοῦς (whose presence in both Vaticanus and Alexandrinus was presumably a major factor in its inclusion in the Editio Critica Maior). Even a theologian fully committed to the pre-existence of the Logos may very well baulk at such an unsophisticated expression of that theologoumenon as would be expressed by the clause, “Jesus saved the people out of Egypt.”

Yet, if even the well-trained and theologically astute scribes of the Alexandrinus and Vaticanus codices wrote “Jesus,” it would appear that the UBS4 committee is in reality representative of that same tendency that affected the scribes of the Byzantine tradition, who preferred the simpler reading “God.” The text of the Bodmer papyrus shows us that even a pre-Nicene scribe was quite capable of believing that the one who saved the Israelites was Jesus, and was willing to make this fact clear in his text. In fact, it was not such an unusual idea in early Christianity (cf. 1 Cor 10.4, 9). Many early Christians may well have identified the angel whom God sent to effect the exodus (Num 20.16) with Jesus himself (Hannah 1999, 139–42). By the mid-second century, the notion that Jesus had been present in Israel’s history, putting God’s plans into action in the form of an angel, had become almost a cliché. ⁵ If scribes from the third to fifth centuries not only acquiesced in readings that ascribed the miracle of the exodus to Jesus, but actively propagated it, there is really no reason to believe that it was not the original intention of the author. The theological “fear” of this meaning belongs to a later age of the church, and to some extent to our own too.

What is significant for us is not, however, the original reading. Even if Jude wrote ὁ κύριος, what he meant by that was certainly Jesus, ⁶ and that is, in effect, what the scribe of P72 is showing us by writing “God Christ.” It is not a good reading, but it is a helpful interpretation, and an early one at that. Any modern translation that seeks to make the meaning of the original as clear as possible to its readers ought to have the courage to follow these ancient interpreters. To translate “Lord,” without further qualification, may well mislead the modern reader, unaccustomed to the early church’s belief that Jesus was active in Old Testament events, into imagining that Jude simply meant the God of Israel. To translate it as “Lord” (as if the name Yahweh was hovering underneath) would surely misrepresent what Jude meant, as well as do an injustice to the very high Christology of this letter.

Jude 14

The scribe of P72 (or that of his exemplar) will help us again in similar fashion at v. 14. There is little doubt that NA28 and UBS5 accurately reflect the earliest recoverable text, and most likely the autograph, that is, ἰδοὺ ἦλθεν κύριος ἐν ἁγίαις μυριάσιν αὐτοῦ “behold the Lord [more likely Jesus than Yahweh, but that is another discussion] is coming among his holy ones in their thousands.” The question is one of interpretation—that is, to whom does ἁγίαις refer? To the people in the church, or to heavenly beings? P72 and א both supply an explanatory “angels” (ἁγίων ἀγγέλων μυριάσιν P72; μυριάσιν ἁγίων ἀγγέλων א). A third-century Latin author (omitted by the NA apparatus) also quoted the text as “ecce venit cum multis millibus nuntiorum suorum” (see he comes with many thousands of his angels). ⁷ In the light of other hints within the New Testament that the parousia would occur in the midst of a company of heavenly beings (1 Thess 3.13 and 2 Thess 1.10, with Luke 9.26 providing the religio-historical background), we could hardly do better in a meaning-based translation than to follow the interpretation offered by the Bodmer papyrus and pseudo-Cyprian (even against many modern English versions).

2 Peter 1.10

In contrast to the previous example, here our guide P72 offers us the shorter reading widely accepted by most editors, σπουδάσατε βεβαίαν ὑμῶν τὴν κλῆσιν καὶ ἐκλογὴν ποιεῖσθαι (be eager to make your calling and election firm). ⁸ This undoubtedly original reading has thoroughly exercised the minds of commentators as to what, exactly, the author might have meant. One early expansion of the text (presumed early on account of its widespread attestation) reads σπουδάσατε διὰ τῶν καλῶν ὑμῶν ἔργων βεβαίαν ὑμῶν τὴν κλῆσιν καὶ ἐκλογὴν ποιεῖσθαι (be eager to make your calling and election firm by means of your good works). In this understanding of the verse the author is exhorting his readers to live a life in keeping with their calling by the good things they do, and this indeed is how they will confirm and ratify their election. As understood by these scribes, who were carrying out a role not very different from that of a translator committed to a meaning-based approach, this is just what the author meant.

John 6.11

ἔλαβεν οὖν τοὺς ἄρτους ὁ Ἰησοῦς καὶ εὐχαριστήσας διέδωκεν [τοῖς μαθηταῖς, οἱ δὲ μαθηταὶ] τοῖς ἀνακειμένοις

So Jesus took the loaves and, after giving thanks, he gave them [to the disciples, and the disciples gave them] to those who were sitting

The bracketed words are not to be found in the papyri or best uncials, but were rather added from the parallel at Matt 14.19 by an early corrector of Sinaiticus, possibly as early as the fifth century and using an Alexandrian exemplar. They are also transmitted by the quite independent tradition from which the Old Sinaitic Syriac was translated, ⁹ and also by the “Western” text as witnessed by Codex Bezae (D) and a variety of old Latin copies. ¹⁰ This is not, therefore, even the work of a single corrector who had imitators, but of perhaps as many as four independent witnesses to a particular “correction.”

It is instructive, therefore, to note that the UBS Translator’s Handbook on the Gospel of John (Newman and Nida 1980) advises just the same alteration to the text in the name of clarity. Again it is clear that many early scribes approached their task in a fashion not dissimilar to that of contemporary translators who adopt an approach that encourages them to amplify the text where it is felt that a stage in the narrative has been missed.

2 Peter 1.2

The expression τοῦ θεοῦ καὶ Ἰησοῦ τοῦ κυρίου ἡμῶν (of God and Jesus our Lord) was also subjected to P72’s (or his exemplar’s) inquisitive pen, and lost its καὶ. The result is the singular reading, ¹¹ “of God, Jesus our Lord,” which was not, of course, an accident. Rather it is an honest attempt to lay to rest any doubts that may have arisen over the co-reference of the two noun phrases concerned. The same pair of referents, “our God” and “Savior Jesus Christ,” were already mentioned in the previous verse. What we know of the scribe of P72 may assure us that his reading of v. 2 is not a theologically motivated alteration of an originally different meaning, but rather an interpretation of how he read the text of v. 1. As a mother-tongue speaker of Greek, the scribe has no problem understanding that the words of v. 1, τοῦ θεοῦ ἡμῶν καὶ σωτῆρος Ἰησοῦ Χριστοῦ, could only be taken to mean “of our God and Savior Jesus Christ,” that is, that Christ is himself called both “God” and “Savior.” ¹² He thus simply extended that understanding also to the following verse, which on its own terms, and without the preceding words of v. 1, might be open to alternative interpretations. The scribe’s aim in omitting the καὶ was purely to clarify the meaning as he already understood it. This does not give modern translators carte blanche to follow P72’s text, but it does provide them with the confidence, should they need it, to understand v. 1 aright, and might well be permitted to influence our translations of the passage as a whole.

James 1.27

The foregoing examples should be enough at least to suggest that the interpretations of ancient scribes (encoded in the “variants” that are attributed to them) are often of interest to translators who are concerned to communicate the meaning of the text as clearly as possible to a particular audience. Another of the Bodmer papyri, P74, although not as ancient as P72, remains nonetheless a valuable and important witness to the text of the General Epistles. In place of the standard text of Jas 1.27, ἄσπιλον ἑαυτὸν τηρεῖν ἀπὸ τοῦ κόσμου (keep oneself spotless from the world), the papyrus in question reads ὑπερασπίζειν αὐτοὺς ἀπὸ τοῦ κόσμου (protect them [the orphans and widows mentioned in the previous clause] from the world). ¹³ Of course, as text critics we need not accept the late papyrus’s singular reading as being in any way original (though this variant has found one supporter in Roberts 1972), but this should not prevent us from taking an interest in the apparatus, recognizing the work of a dynamic “translator” and perhaps taking our cue from his lead.

Acts 24.17-19

¹⁷ δι᾽ ἐτῶν δὲ πλειόνων ἐλεημοσύνας ποιήσων εἰς τὸ ἔθνος μου παρεγενόμην καὶ προσφοράς, ¹⁸ ἐν αἷς εὗρόν με ἡγνισμένον ἐν τῷ ἱερῷ οὐ μετὰ ὄχλου οὐδὲ μετὰ θορύβου, ¹⁹ τινὲς δὲ ἀπὸ τῆς Ἀσίας Ἰουδαῖοι, οὓς ἔδει ἐπὶ σοῦ παρεῖναι καὶ κατηγορεῖν εἴ τι ἔχοιεν πρὸς ἐμέ.

Following the grammar closely, we ought to translate something like this:

After some years I arrived [intending] to bring money for my people, and sacrifices too, ¹⁸ in [doing] which things they found me after I had purified myself in the temple, without [attracting] a crowd or [causing] a disturbance. ¹⁹ But some Jews from Asia who ought to be present before you and to bring charges, if they have anything against me.

There are two grammatical problems as the text stands: first, the subject of εὗρόν με (they found me) is unspecified and not at all obvious from the context; second, in v. 19 “Jews from Asia” stands alone without an expressed predicate. The δὲ at least indicates for us that it cannot have been the Asian Jews who found Paul in the temple. Almost the only way to read the text without resorting to emendation is to take as the subject of εὗρόν με those other Jews who had actually brought Paul before the governor and who are mentioned again in v. 20. However, in that verse they are referred to as οὗτοι (these ones), which is just how we should expect Paul to refer to those present in a courtroom. Had he intended them as the subject of εὗρόν με in v. 18, he most likely would have said οὗτοι here too. Of course, a critic with a penchant for emendation might well feel obliged to insert it, and a meaning-based translator is unlikely to flinch at making just the same move, which is in fact what the New Living Translation does (“my accusers saw me in the Temple as I was completing a purification ceremony”). In any case, all the versions are then forced to make some sort of an existential clause out of v. 19 (TEV: “But some Jews from the province of Asia were there; they themselves ought to come before you . . .” and many others have similar renderings).

An easier course does, however, lie to hand, a course already followed by some ancient scribe(s) of the Byzantine recension, namely to delete the δὲ of v. 19 and so make the Asian Jews themselves the subjects of εὗρόν με. Without the δὲ, and without the unnecessary comma at the end of v. 18, our text may be easily rendered with good sense as “in doing which act [i.e., bringing the offering] some Jews from Asia found me, after I had purified myself in the temple without [attracting] a crowd or [causing] a disturbance, who [i.e., those Jews from Asia] ought also to be present . . .” This not only solves the problem of the subjectless εὗρόν με but also has the advantage that it comports with precisely what Luke tells us had actually happened in the temple (Acts 21.27). In fact, after the author specified rather precisely that those who caused Paul all this trouble in the temple were Jews from Asia, it stretches coincidence to the breaking point that here in Paul’s account of the same event in ch. 24 he should mention both his being found in the temple and the Asian Jews, but without explicitly linking the two.

In a sense, therefore, our scribes have offered an excellent critical emendation by their removal of the problematic δέ. ¹⁴ I do not mean by this that they have preserved for us either the original or the earliest recoverable text. It is not inconceivable that they have in fact done so, but the probability is much greater that the emendation was made upon a text that presented itself to (at least one) scribe as a grammatical conundrum, and that he was not prepared to leave a confusing, perhaps meaningless, verse in the text he was offering his readers. Many translators feel the same way about awkward or unclear texts. This ancient scribe viewed his own function in a similar light. He has only made explicit what he believed (probably rightly, though the original text seems to be lost to us) the meaning to have been, and produced what we would today call a “dynamic” version of the text.

Acts 26.20

τοῖς ἐν Δαμασκῷ πρῶτόν τε καὶ Ἱεροσολύμοις πᾶσάν τε τὴν χώραν τῆς Ἰουδαίας καὶ τοῖς ἔθνεσιν ἀπήγγελλον

I preached first to those in Damascus, and then to those in Jerusalem, and the whole land of Judea, and then to the Gentiles

Although the earliest recoverable text is not in doubt, the content remains problematic since Paul clearly never preached in the province of Judea (cf. Gal 2.9). At least two critical emendations were made, probably independently, by early medieval scribes, who believed that the original must have meant “to Jews and to Gentiles.” Their lead was followed by one modern editor. ¹⁵ The emendation thus originated with these medieval scribes and testifies to their efforts to present their readers with a clear text.

As with the scribes of the Byzantine tradition, whose work we mentioned earlier, these more recent transmitters of the book of Acts, two of them anonymous medieval scribes, another an early modern textual critic, have been concerned to present a more meaningful text. Their efforts are in line with the modern movement towards clear, unambiguous translations, whether or not this specific “emendation” be accepted by translators.

It appears that when a contemporary translator “alters” the base text in some way so as to clarify (in his or her understanding) its meaning, in doing so he or she is in a sense accepting a critical emendation—not in the strict sense of reconstituting the earliest recoverable text by a process of emendatio, but insofar as the translator is the heir of former tradents, whose task was not substantially different.

John 8.8-10

The Pericope Adulterae appears to have been fertile ground for variants and scribal elaborations of the text. In some of these we are able to discern just the sorts of clarifications and explanatory renderings that many modern translators judiciously insert into their renderings. As “interpretations” of what the underlying text actually says, some of these elaborations appear more acceptable than others.

In v. 8, to the words κατακύψας ἔγραφεν εἰς τὴν γῆν (bending down, he wrote on the ground), one scribe added ενος εκαστου αυτων τας αμαρτιαις (the sins of each one of them). Answering the curiosity of readers wanting to know just what Jesus had written in the sand, this particular “addition” to the text most likely would not commend itself to many translators or consultants today. ¹⁶

In v. 9, one rather unexpected and theologically questionable addition (καὶ ὑπὸ τῆς συνειδήσεως ἐλεγχόμενοι, represented in the AV as “they which heard it, being convicted by their own conscience”) was followed by three others much more likely to be repeated by modern translators: one adds to the expression ἀπὸ τῶν πρεσβυτέρων + εως των εσχατων (beginning with the elders + until the last of them), another specifies the subject of κατελείφθη μόνος as Jesus, and the last has the woman ἐν μέσῳ ἑστῶσα (standing in the middle) rather than merely οὖσα (being in the middle). All these four variants are well attested in the tradition and, although assuredly not original, they are ancient, with three of them going back at least to the very oldest Latin manuscripts. The obvious facility of the addition of the name “Jesus” here has been recognized by almost all modern English versions, which make precisely the same move.

There are two substantial additions found in the Majority Text of v. 10:

ἀνακύψας δὲ ὁ Ἰησοῦς (+ καὶ μηδένα θεασάμενος πλὴν τῆς γυναικός) εἶπεν αὐτῇ, ποῦ εἰσιν ἐκεῖνοι (+ οἱ κατήγοροί σου)

The difference may be most readily observed by a simple comparison of AV (“When Jesus had lifted up himself, and saw none but the woman, he said unto her, Woman, where are those thine accusers?”) with the shorter NIV (“Jesus straightened up and asked her, ‘Woman, where are they?’”). At least the second of these additions consists in nothing more than making a pronoun explicit, a move that none of the translations baulked at in the previous verse. Yet the modern versions stick steadfastly to their base text here.

2 Thess 2.13

ὅτι εἵλατο ὑμᾶς ὁ θεὸς ἀπαρχὴν εἰς σωτηρίαν

ἀπαρχὴς א and many other mss

because God chose you as the firstfruits of salvation [or, chose you to salvation from the beginning]

There is little agreement as to whether the earliest recoverable text was απαρχην or απαρχης. The UBS4 text opted for the former (“firstfruits,” rather than “from the beginning”), although both readings are feasible and both well attested.

There are two arguments which most likely tilt the issue in favour of the UBS decision for ἀπαρχὴν as the oldest recoverable reading.

NET and Wanamaker 1990 both argue for what they call the “easier” reading, ἀπ’ ἀρχῆς, though not for the same reasons. The commentator seemingly only refuses ἀπαρχῆ because he cannot conceive of what its meaning would be in the present context—which of course is just the same reason that would have tempted a scribe to have written ἀπ’ ἀρχῆς in the first place. The notes to NET prefer ἀπ’ ἀρχῆς on the almost contrary grounds that it is the lectio difficilior (being an uncharacteristic Pauline expression). It is very unclear, however, what the author(s) of this note mean when they say that ἀπ’ ἀρχῆς > ἀπαρχῆν was a more likely alteration because “scribes . . . thought of the early churches in general in this way.” This type of lectio difficilior argument, such as the NET note offers, can only ever be a rather general guide. The lectio difficilior must still be plausible and defensible, and a good knowledge of authorial style, as Housman frequently used to insist, remains an invaluable touchstone of editorial judgement. Thus in the current case, the argument for the harder reading most likely has the opposite force, for although ἀπ’ ἀρχῆς is not a characteristic Pauline expression, nevertheless it is rather common in other NT authors (esp. Matthew and 1 John), hence it is much more likely to have been a scribal correction applied to the original ἀπαρχῆν than vice versa (scribes, after all, tended not to distinguish the grammatical or lexical usages of different authors). Thus in this case we have good reason to follow Paul’s idiom, which has here been assimilated to NT usage.

Translations have been split as to which reading to follow, and one may instructively trace the genealogy of readings among translations and the “pulls” that they experience. The AV reads “from the beginning,” naturally following the Received Text, whereas many recent revisions (including NIV2011, NRSV, TEV, NLT, ESV) have moved to reading “firstfruits.” ¹⁷ However, if we turn to the history of Swahili versions, different “pulls” are at work. In contexts such as this, tradition holds a stronger sway. For instance, the most commonly used Swahili version, known as the Union Version, imitated the AV here as in so many places. The more recent Habari Njema (Good News), however, although in many ways constituting a direct translation of TEV, evidently felt the tug of tradition more keenly than that of the UBS editors or of its guide, TEV, and so still reads, “from the beginning.”

Thus despite the fact that the NRSV committee recognized that the most likely reading was, in fact, “firstfruits” and altered the text of their RSV base to read “God chose you as the firstfruits for salvation,” in which they have been followed by many other English versions, yet the reading “firstfruits” remains wholly unknown to readers of Swahili Bibles.

The issue of “acceptability,” sometimes touted as the fourth great principle of Bible translation after accuracy, clarity, and naturalness, significantly impacts translators around the globe. The perception that “modern” translators are changing the words of Scripture is an ever-present concern among church leaders. The criterion of “acceptability” may well motivate translators to retain the AV reading “from the beginning” in this verse. To do this would mean also rejecting the reading of NA28/UBS5, and the translator is faced with the possibility that he or she may have to submit a translation which includes a reading that is almost certainly not original in the interests of acceptability. Thus even if in principle a particular translation project has chosen to make the UBS text its base-text, this principle will inevitably be compromised from time to time.

Acts 20.28

The early history of the text is of course closely bound up with the early history of the church. The one sheds its light upon the other, and the interpretations of the early tradents are, once again, of value to the translator.

ποιμαίνειν τὴν ἐκκλησίαν τοῦ θεοῦ, ἣν περιεποιήσατο διὰ τοῦ αἵματος τοῦ ἰδίου אB etc.

τοῦ κυρίου P41 P74 A.D. 33 etc.

τοῦ κυρίου καὶ θεου Maj.

Be shepherds of the church of God [or, of the Lord, or, of the Lord God], which he purchased through his own blood

The conflate reading of the Byzantine tradition merely reinforces the critic’s uncertainty over the original. “Church of the Lord” and “church of God” are both well attested in the manuscripts. As Metzger concedes, the decision must be made on grounds of internal probability, and the critic must decide which reading the tradents of the early church are more likely to have found “easier” to transmit to posterity (1994, ad loc.).

This is precisely where different criteria clash, and the question of “what changes scribes of the early era were likely to make” becomes subject rather to the expectations of modern commentators than to the hard evidence of the case.

First, different criteria will not always agree: “church of God” is a much more common NT expression than “church of the Lord,” and hence the latter is clearly the harder reading if we award preference to the criterion of what is stylistically more common. On the other hand, “of God” becomes the harder reading from a theological perspective since it involves Paul (or Luke) saying that God shed his own blood, which would indeed be a very unusual, perhaps unexpected, expression, a difficulty from which the reading “church of the Lord” extracts us. Thus the issue of deciding upon the harder reading turns upon whether a scribe might be expected to opt more readily for a normative biblical expression, or for the theologically comfortable one. ¹⁸

In fact, even Metzger’s assurance that the external evidence is equally divided is based upon the weight placed by Westcott and Hort (1882, appendix: 98–99) upon the combination of אB, which, when placed together with the testimony of Cyril of Alexandria, was enough to clinch the issue within their system of criticism. Yet really this only tells us that the reading “of God” was well known in fifth-century Egypt, where it naturally tended to support a miaphysite reading of the relationship between the divine and the human in Christ, in which whatever is said of the human nature may be predicated of the divine with equal facility, the so-called communicatio idiomatum.

Two general facts cast doubt on Westcott and Hort’s certainty. The first is that, Alexandrian witnesses aside, there is widespread and excellent attestation for κυρίου; ¹⁹ second, for the theologians of the first centuries of the church, this verse was never wielded in defence of any dogmatic position. Perhaps the idea that God purchased his church with his own blood would not, in fact, have been quite such a “hard reading” as modern critics suppose. If the theological difficulty of the reading can be disposed of, then there can no longer be any significant argument against a scribal adaptation κυρίου > θεοῦ, conforming the text to biblical usage, and the way to a solution becomes clearer.

In the last resort, much hangs upon how we view the thorny problem of just how “difficult” the idea of God shedding his own blood was. In his aforementioned discussion, Hort presumes its great theological difficulty, that the church’s fear of monarchian tendencies in those very centuries (fourth to seventh) that were crucial for the transmission and dissemination of the NT text, and their equal fear of the eastern doctrines of miaphysitism from the fifth century onwards, make the idea of an alteration in the direction κυρίου > θεοῦ close to inconceivable. The UBS committee has in essence followed this lead (Metzger 1994, ad loc.).

In fact, Hort felt particularly challenged by this reading, such that he was inclined to suggest that διὰ τοῦ αἵματος τοῦ ἰδίου must really mean “through the blood of his own [son]” (Westcott and Hort 1882, appendix: 99). He thereby in effect removed the difficulty from the text altogether, unwittingly causing a rather circular argument in his reasoning, since once τοῦ ἰδίου is granted such a meaning as this, the originally problematic ἐκκλησίαν τοῦ θεοῦ ceases to be a difficult reading at all, and no longer demands the critic’s attention as much as it did at first! ²⁰

But as we have already suggested, the theological difficulty may have been Hort’s rather than anyone else’s. After all, an argument regarding theologically motivated textual alterations can cut both ways. Shortly after the publication of the Westcott–Hort New Testament, Page defended ἡ ἐκκλησία τοῦ κυρίου on the basis that κυρίου > θεοῦ was just as likely at the hand of an anti-Arian scribe as was θεοῦ > κυρίου at the hand of an anti-monarchian or anti-miaphysite one (1886, 216–17).

Hort’s argument that any normal Christian writer would avoid the connotations of God’s blood is amply disproved by Ignatius, Ephesians 1.1. Similarly, Cyril of Alexandria read θεός and understood it as meaning Jesus and as proving his divinity. Neither the second-century church of Ignatius’s day nor the fifth-century church of Cyril’s day had a problem with the reading θεός, albeit for rather different reasons.

Papyri discoveries made since Hort’s time have tended to strengthen his argument as to the contextual meaning of ὁ ἰδίος. Moulton found enough examples in the papyri of the unqualified ὁ ἰδίος referring to a close relative to suggest that this must also be the intention in our passage, which can thus be so readily rendered as “the blood of his own [one]” as to make Hort’s conjecture no longer required to meet the purpose for which it was devised. Indeed, a Roman delegate at the Council of Ephesus wrote, “pro ecclesia Dei, quam Dominus noster Iesus Christus sanguine suo acquisivit” (for the church of God, which our Lord Jesus Christ purchased with his own blood), and so appears to support the same interpretation of the difficult expression, that the church is God’s, the blood that of Jesus. ²¹

Ehrman has disposed of the problem in not dissimilar fashion. For him, the text του αιματος του ιδιου can only come to mean “God . . . with his own blood” if one already presupposes the possibility of the communicatio idiomatum (2006, 164). In any case (following Ehrman) the blood must be that of Jesus and hence the textual question may be decided on the stylistic criterion alone.

Ehrman could have quoted Apostolic Constitutions VII.56: “run together to the Church of the Lord, which He has purchased with the blood of Christ”; and VIII.12.18: “we pray for the church . . . which you purchased with the precious blood of your Christ” (seemingly a conflation of our verse with 1 Pet 1.19, ἀλλὰ τιμίῳ αἵματι ὡς ἀμνοῦ ἀμώμου καὶ ἀσπίλου Χριστοῦ, and Rom 8.32, ὅς γε τοῦ ἰδίου υἱοῦ οὐκ ἐφείσατο), as evidence that even reading “church of the Lord” does not alleviate the problem—which is in fact only really resolved by clarifying the owner of the blood in some other way. To Elliott (2010, 281), the very ambiguity of κύριος is proof of its originality.

Yet other writers clearly read the expression διὰ τοῦ αἵματος τοῦ ἰδίου as if it must naturally mean “through his own blood” (i.e., as equivalent to Heb 13.12, διὰ τοῦ ἰδίου αἵματος). Tertullian read it that way and in fact his theology was capable not only of digesting it but of glorying in it: “Quod sciam, non sumus nostri, sed pretio empti, et quali pretio? Sanguine Dei!” (so far as I know, we are not our own, but bought with a price. What price? The blood of God). ²² Both Jerome and the translator of the Vulgate Acts assumed it meant this, though for them in combination with κυρίου. Jerome could easily hold both interpretations without apparent contradiction—to Furia he says, “you are not his to whom you have been born, but His to whom you have been born again, and who has purchased you at a great price with His own blood,” but to Evangelus he quotes the verse in its “difficult” form. ²³ The Syriac poet Ephrem writes that “you are bought with the blood of God; you are redeemed by the passion of Christ.” ²⁴

It would be well to consider for a moment the interpretations of this idea by the early church, both because Greek-speaking Christians of the first few centuries are often a guide to the natural sense of the Greek of the New Testament, and also because theological reflection upon the text of the New Testament has constituted the doctrinal framework through which we too read the Bible.

What emerges is not merely that the reading “Lord” may well be preferable in terms of reconstructing the original text of the New Testament, though we can surely not be certain, but that in any case a broader issue is raised which impinges upon the translator’s task. The activities of early Christian scribes in transmitting the text of the New Testament were not in truth so very different from the task of the translator. To attend to their struggles is to look into our own, and often to become more aware of the cognitive processes that guide, or sometimes control, our activity as translators. Contemporary translators are transmitters of Scripture, just as much as were the scribes of the early church. We are one further link in a long chain, and our task has changed less than we might realize.