In the previous post, I described some passages in the Gospels where the rival variants may receive different treatment in future editions of the critical text of the Greek New Testament, and/or in English translations based on it – including a few passages where the editors may adopt readings with no Greek manuscript-support, as the editors of the 28th edition of the Nestle-Aland Novum Testamentum Graece recently did in Second Peter 3:10. Today, let’s look at a dozen passages in the rest of the New Testament which may be similarly vulnerable to the effects of thoroughgoing eclecticism.
● Acts 6:9 – The scholar Friedrich Blass (1843-1907), in the course of his detailed study of the Greek text of the Gospels and Acts, detected something abnormal about the mention of Libertines in Acts 6:9: why did Luke resort to Λιβερτίνων, a Latin-based term, rather than simply write Ἀπελεύθεροι? And why, followed by various geographically based terms, is this one not also geographically based?
Such questions elicited a search for answers. Blass discovered that he was not the first reader to hum upon encountering the term Λιβερτίνων in this verse. A long line of researchers, going all the way back to Beza, had sensed that something about this word was amiss.
Blass was informed by J. Rendel Harris that in the Armenian version, the reference was not to Libertines, but to Libyans. How, though, could a reference to Libyans ever be misconstrued as Libertines? And if the Armenian version’s base-text had referred to Libertines, how did the Armenian version end up with Libyans? With remarkable determination, Blass dug a little deeper into this puzzle, and discovered, among the Latin poems of Catullus, the use of a rare term that satisfied his curiosity; transliterated into Greek, it is Λιβυστίνων – inhabitants of the area west of
This conjectural emendation resembles the extant text; it fits the context, it makes sense, and it introduces nothing problematic. Future editors may reason that the rareness of the term Λιβυστίνων provoked early copyists to misread it, and also provoked the translators of the Armenian version to loosely approximate its meaning with a more familiar term.
● Acts 7:46 – Working within the extant evidence, textual critics must choose between the statement that David asked to be allowed to find a dwelling-place for the God of Jacob (a reading supported by Codices A, C, E, 1739, the vast majority of manuscripts, and broad versional evidence), or a dwelling-place for the house of Jacob (which is the reading in the Nestle-Aland compilation, and which is supported by a small cluster of early manuscripts, including Papyrus 74, Vaticanus, Sinaiticus, and Bezae).
The Alexandrian reading is certainly more difficult, because it seems to say that David asked to build a house for a house. Even when the second “house” is understood to refer to the nation descended from Jacob, the problem does not go away, since the temple was for God, not for the people, who were not looking for a new residence in the days of David.
The reading οἴκω (“house”) has been considered too difficult by some textual critics, including Hort, who wrote in 1881, “οἴκω can hardly be genuine,” but rather than accept the Byzantine reading, he proposed that probably neither reading is original. Instead, he conjectured that the original text was τω Κυριω (“the Lord”), which was contracted to ΤΩ ΚΩ, which was misread by inattentive copyists as ΤΩΟΙΚΩ. If future editors of the Nestle-Aland compilation are unwilling to adopt Hort’s conjecture, they might at least acknowledge the force of his admission of the implausibility of the Alexandrian reading, and adopt the other reading, for which the diversity of the external support is very impressive.
● Acts – In the description of the action taken by Barnabas and Saul in this verse, there is a four-horse race, so to speak:
ἀπο Ἰερουσαλὴμ (“from Jerusalem”), supported by Codex D, Ψ, 614, several Old Latin copies, a significant minority of Byzantine manuscripts, the Vulgate, et al.
εἰς Ἰερουσαλὴμ (“to
”), supported by Codices א, B, and most Byzantine manuscripts. Jerusalem
ἐξ Ἰερουσαλὴμ (“from
”), supported by Papyrus 74, Codex A, et al.
ἐξ Ἰερουσαλὴμ εἰς Ἀντιόχιαν (“from
to Jerusalem ”), supported by Codex E, 1739, the
Peshitta, the Sahidic version, et al. Antioch
The most difficult option is the variant with εἰς, because (1) when last seen in the narrative (in ), Paul and Barnabas were already going to
, and (2)
in the very next scene (at the beginning of chapter 13), Paul and Barnabas are
present at Jerusalem , not at Antioch .
Even though εἰς is in the Nestle-Aland compilation, some translators of
modern versions have rejected this reading; for example, the Jerusalem NASB states, “And Barnabas and Saul returned
from ,” implying a base-text with either ἀπο or
ἐξ. The Jerusalem ESV reads identically (as of , August 4, 2017):
“And Barnabas and Saul returned from .”
The NIV also rejects εἰς, stating, “When Barnabas and Saul had finished
their mission, they returned from Jerusalem .” Jerusalem
Nobody (other than avid advocates of the Peshitta) seems to think that the longest reading is original, because it looks like just the sort of textual adjustment that a copyist might make to alleviate a difficulty. The contest, then, is between εἰς, ἐξ, and ἀπο. Theoretically, if, in a scriptorium where a group of copyists worked from dictation, their supervisor read ἐξ, a copyist could mishear it as εἰς – but the theory works as well in the opposite direction.
Thoroughgoing eclecticism turns the race into a five-horse contest. Hort suggested in 1881 that the original word-order has been garbled by copyists, and that the original text read τὴν εἰς Ἰερουσαλὴμ πληρώσαντες διακονίαν, so as to merely report that Barnabas and Saul returned, having completed their service in Jerusalem. F. F. Bruce, whose confident comments about the reliability of New Testament manuscripts have been thoroughly recycled by many apologists, did not refuse to embrace a conjectural emendation in this passage; he held that in the original text there was no prepositional phrase at all, and that marginal glosses have impacted the text in Acts 12:25 in all extant manuscripts.
Although textual critics often regard a higher degree of difficulty as a quality of the most-likely-original reading, it is possible that future editors may regard the currently printed reading here as simply too difficult, and either adopt ἐξ, or adopt ἀπο, or resort to conjectural emendation.
● Acts – Bruce Metzger dedicated a full two pages of his Textual Commentary to a consideration of this passage. The initial question is, did the original text refer to “the
,” or to “the church of the Lord,” or to
“the church of the Lord and God”? Like
John 1:18, the contest between “God” and “Lord” is a contest amounting to the
difference of a single letter, once one accounts for the contraction of sacred
names: Θεοῦ (“of God”) becomes ΘΥ
and Κυρίου (“of the Lord”) becomes ΚΥ. church of God
If that contest is decided in favor of Θεοῦ (on the grounds that this is supported by ﬡ, B, the Vulgate, the Peshitta, and a significant minority of Byzantine manuscripts, including the Textus Receptus), then a second question arises: did Luke report that Paul stated that God purchased the church with His own blood? Many an apologist has used this verse to demonstrate Paul’s advocacy of the divinity of Christ, inasmuch it was neither the Father, nor the Spirit, whose blood was shed. Hort, however, expressed a suspicion (which, it seems, was first expressed in 1797 by Georg Christian Knapp) that at the end of the verse, following the words διὰ τοῦ αἴματος τοῦ ἰδίου (“through His own blood”), there was originally the word υἱοῦ (“Son”).
It is possible that future editors, may decide that the inclusion of υἱοῦ in this verse is required by internal evidence, and that it is feasible that the word υἱοῦ was accidentally lost very early via a common parableptic error (when a copyist’s line of sight drifted from the letters ΙΟΥ at the end of ἰδίου to the same letters at the end of υἱοῦ. Already, the Contemporary English Version, advertised as “an accurate and faithful translation of the original manuscripts,” has the word “Son” in its text of Acts 20:28b: “Be like shepherds to God’s church. It is the flock that he bought with the blood of his own Son.” A footnote informs the CEV’s readers about the meaning of the extant text.
● First Corinthians 6:5 – The Greek evidence, from Papyrus 46 to the Textus Receptus, is in agreement about how this verse ends. However, the Peshitta – a Syriac version traceable to the late 300’s (followed by a period of standardization), but possibly earlier – disagrees. The reading in the Peshitta implies that its Greek base-text included the phrase καὶ τοῦ ἀδελφοῦ (“and a brother”).
The momentum for this reading is drawn from a grammatical oddity in the usual Greek text. The first part of Paul’s statement in this verse is something to the effect of, “Is there not even one person among you – just one! – who shall be able to judge between” – and that’s where the difficulty appears, because the Greek text just mentions one brother, whereas the idea of judgment between two parties seems to demand that more than one brother should be mentioned.
The KJV’s translators, although the Textus Receptus reads ἀνα μέσον τοῦ ἀδελφοῦ αὐτοῦ (“between his brother” – which is clearly singular), concludes the verse with “between his brethren” (which is clearly plural). The NET is similar (“between fellow Christians”); the difference is due to the
NET’s enlightened gender-neutral treatment of
the term ἀδελφοῦ, not to any new feature in the Greek base-text. The CSB,
and even the NASB
likewise render the text as if the verse ends with a plural word rather than a
singular one. All such treatments of the
text make the problem all the obvious:
the first part of the sentence, in Greek, anticipates two brothers, while the second part of
the sentence mentions only one.
In light of such strong internal evidence, Michael Holmes, the compiler of the SBLGNT, recommended the adoption of a conjectural emendation at this point, so that καὶ τοῦ ἀδελφοῦ (“and the brother”) appears at the very end of the verse. It is possible that future compilers of the Nestle-Aland text will concur. If that happens, it will have hardly any effect on English translations, most of which already translate the passage as if the wording proposed by Dr. Holmes is extant in the manuscripts.
● Galatians – For almost 300 years, a scholarly debate has orbited part of this verse. The phrase “Now this Hagar is
Sinai in Arabia” is presently in the Nestle-Aland
compilation (as τὸ δὲ Ἁγὰρ Σινᾶ ὅρος ἐστιν ἐν τῇ Ἀραβίᾳ, though this is
contested by four slightly different rival forms). However, it has been proposed that the entire
phrase originated as a marginal note, and does not belong in the text. This
conjecture goes back at least as far as Richard Bentley
(a gifted British cleric, 1662-1742, who advanced the field of New Testament
textual criticism more than anyone else in his generation). Recently Stephen
Carlson, who has conducted a
stemmatics-based analysis of the text of Galatians, has argued in favor of
the same idea. (Robert Waltz, however,
retained the phrase in his
compilation of the text of Galatians.)
If future editors of NTG concur with Carlson, the phrase might be exiled
to the footnotes.
|A Syriac manuscript at Saint|
Catherine's Monastery displays
an adjustment of the text of
(Charley Ellis pointed out this
feature of the MS to me.)
● Hebrews 2:9 – In Bart Ehrman’s 2005 book Misquoting Jesus, the author noted that instead of reading χάριτι θεοῦ (“by the grace of God”), a smattering of witnesses supports χωρὶς θεοῦ (“without God”). Origen, in the 200’s, was aware of both variants. Opposing an array of witnesses that includes Papyrus 46, Vaticanus, Sinaiticus, Claromontanus, the Byzantine Text, and broad versional support, Ehrman proposed that χωρὶς θεοῦ was the original reading, and that an early copyist altered the text so that it said something less provocative – and this unknown copyist was so influential that his alteration has affected almost all extant Greek copies of Hebrews.
As Ehrman noted in his book, the existence of the variant χωρὶς θεοῦ has been accounted for by some textual critics via the idea that it was written in the margin by someone who intended for the phrase to be a qualification of the sentiment of the preceding verse – the idea being that all things except God were subject to the authority of Christ – an exception mentioned by Paul in First Corinthians 15:27. Resisting this proposal, Ehrman objected that if this had been an annotator’s intent, “Would he not have written “except for God” (EKTOS THEOU – the phrase that actually occurs in the I Corinthians passage) rather than “apart from God (CHŌRIS THEOU – a phrase not found in I Corinthians)?”
Ehrman’s objection loses much force when one observes that the phrase EKTOS THEOU does not, in fact, occur anywhere in First Corinthians. Ehrman also overstates the evidence when he claims that “Origen tells us that this [χωρὶς θεοῦ] was the reading of the majority of manuscripts in his own day,” for Origen cites this reading and then says that some copies have the other reading, χάριτι θεοῦ; nowhere does Origen say that his collection of manuscripts at Caesarea was typical of the manuscripts of Hebrews that existed throughout the world. Most readers of Ehrman’s book, however, will probably not double-check his confidently worded assertions.
Centuries ago, the devout scholar John Bengel (1687-1752) cautiously favored the reading χωρὶς θεοῦ and argued that its meaning is not scandalous, but theologically profound. Bengel proposed that it was intended to mean that the Son of God, and not God the Father, tasted death for everyone – an idea that is consistent with the text of Hebrews 1:3, where, in Papyrus 46 and the Byzantine Text, Jesus is said to have made atonement by Himself.
Another theory, wounded but not killed by the grammatical quirk that it involves, proposes that χωρὶς θεοῦ is original, and that the author intended to thus qualify the words ὑπὲρ παντὸς (“for all”), so as to convey the idea that Christ tasted death for everyone except God. This is how Origen interpreted this variant, without dogmatically deciding in favor of either variant.
Meanwhile, F. F. Bruce proposed an alternative solution: he conjectured that χωρὶς θεοῦ originated as a note in the margin, and that subsequently a copyist replaced that note with one that read χάριτι θεοῦ, and that both readings have slid into the text – in other words, Bruce suspected that neither phrase is original!
● Hebrews –As the list of variants in the textual apparatus of the STEP-Bible shows, the textual racetrack in Hebrews is crowded with rival variants. This sort of contest is difficult for the Coherence-Based Genealogical Method – the newly developed mapping-program used by some of the Nestle-Aland editors – to handle. Researcher J. Harold Greenlee (a scholar in the same league as Bruce Metzger), proposed that the original text of this verse did not contain σπεῖρα or ἡ σπεῖρα (that is, it did not specifically say that Sarah was barren). Michael Holmes’ SBLGNT likewise does not have σπεῖρα or ἡ σπεῖρα in its text.
This constitutes preference for a Byzantine reading (supported by ﬡ, A, and minuscules 33 and 1175, et al). The effect of this textual decision (and some nuanced syntax-related translational decisions) can be seen via a comparison of the text of the NIV 2011 (which adopts it) and the text of the 1973 NIV (which does not):
NIV 1973: “By faith Abraham, even though he was past age – and Sarah herself was barren – was enabled to become a father because he considered him faithful who had made the promise.”
NIV 2011: “And by faith even Sarah, who was past childbearing age, was enabled to bear children because she considered him faithful who had made the promise.”
Whether future editors will continue this trend remains to be seen, particularly because the loss of σπεῖρα can be attributed to parableptic error, when a copyist’s like of sight drifted from the last two letters of Σάρρα (“Sarah”) to the same two letters in σπεῖρα (“barren”).
● Hebrews – In the list of the sufferings of spiritual heroes, one of those things is not like the others: they are all somewhat unusual experiences, except for ἐπειράσθησαν, “they were tempted.” Some textual critics have suspected that this word originated when a copyist committed dittography – writing twice what should be written once; in this case, the preceding word ἐπρίσθησαν (“they were sawn in two”), and that subsequent copyists, not realizing the mistake in their exemplar, changed it into something meaningful. Others have thought that this relatively common term replaced one that was less common – perhaps ἐπάθησαν (“they were pierced”) or ἐπράσθησαν (“they were sold”).
Presently the Nestle-Aland compilation, deviating from the 25th edition, simply does not include ἐπειράσθησαν in the text, adopting instead the reading of Papyrus 46, which is very ancient. Papyrus 13, however, is also very ancient, and appears to support the inclusion of ἐπειράσθησαν, in which case it has a very impressive array of allies. I would advise readers to not get used to the
NTG’s current form of this verse, for it
seems to be merely a place-holder that might be blown away by the appearance of
even the slightest new evidence, and even by an intrinsically appealing
● First Peter 3:19 – What may be the most popular conjectural emendation of all time was favored by the erudite textual expert J. Rendel Harris (1852-1941), who encountered a form of it in William Bowyer’s 1782 book Critical Conjectures and Observations on the New Testament. The extant text of First Peter 3:19 says, “in which he went and proclaimed to the spirits in prison.” Verse 18 refers to Christ, and nobody else is introduced into the text, so verse 19 has been interpreted to mean that during the time between Jesus’ death and resurrection, He visited the realm of the dead – specifically visiting the spirits of those who had been disobedient in the days of Noah, prior to the great flood – and delivered a message (ἐκήρυξεν) to them.
Harris, however, proposed that the original text was different. He thought that Peter had in mind a scene that is related in the pseudepigraphical amalgamation known as the Book of Enoch (the first section of which is quoted by Jude in verses 14-15 of his epistle), in which Enoch is depicted delivering a message of condemnation to the fallen spirits who corrupted human beings so thoroughly that the great flood was introduced as the means of amputating the moral infection they had induced.
Specifically, what Harris proposed was that the opening words of the original text of were ἐν ᾧ καὶ Ἐνώχ (“in which also Enoch”), thus assigning the subsequent action not to Christ, but to Enoch. (A variation on this idea is that the original text read Ἐνώχ instead of ἐν ᾧ καὶ.)
(It should, perhaps, be noted that Irenaeus, in Against Heresies Book 4,16:2, took for granted the veracity of the tradition that Enoch had brought God’s message to fallen angels; these fallen angels being the “sons of God” mentioned in Genesis 6:2-4.)
1. If the original text were simply Ἐνώχ (without ἐν ᾧ καὶ), then, in uncial letters, the χ was susceptible to being misread as a και-compendium (that is, a common abbreviation for the word και (“and”)). A copyist could easily decide to write the whole word instead of the abbreviation, and thus Enoch’s name would become ἐν ᾧ καὶ.
2. If the original text were ἐν ᾧ καὶ Ἐνώχ, a copyist, reading the χ as a και-compendium, could assume that the scribe who made his exemplar had inadvertently repeated three words, and, attempting a correction, remove “Ἐνώχ.”
Against the charge that the introduction of Enoch’s name “disturbs the otherwise smooth context” (as Metzger claimed in 1963) the answer may be given that a reference to Enoch is not out of place, inasmuch as Enoch’s story sets the stage for the story of Noah and his family, whose deliverance through water Peter frames as a sort of pattern of the salvation of the church.
Future compilers willing to engage in conjectural emendation might consider the internal arguments in favor of the inclusion of Enoch’s name in First Peter 3:19 to be too attractive to resist. If that turns out to be the case, then it would certainly have some doctrinal impact, significantly diminishing the Biblical basis for the phrase “He descended into hell” found in the “Apostles’ Creed.”
● Jude verses 22-23 – Even though Tommy Wasserman has collated every known Greek manuscript of the book of Jude, plenty of questions remain about how that data should be interpreted. Like Hebrews 11:11, verses 22-23 of the Epistle of Jude verse have multiple rival variants. Here I shall spare readers the fine details of the case, and simply note that it is possible that future editors may discern here a threefold command, and that the first command from Jude is to refute (ἐλέγχετε) those who cause divisions. Some copyists, with the earlier mention of mercy in verse 21 (ἔλεος) fresh in their minds, may have allowed their memory of it to complete the similarly started word in verse 22.
The reading with ἐλέγχετε is supported not only by Codices A and C* but also by members of manuscript-clusters represented by minuscule 1739 and minuscule 1611; both of these clusters have special weight (as the echoes of distinct and ancient transmission-lines) in the Catholic (General) Epistles. Even though the Nestle-Aland editors have already sifted through the text of Jude, they might take another look at this variation-unit.
● Revelation 7:6 – More than one commentator on the book of Revelation has admitted to being puzzled by a feature of the list of the twelve tribes of the children of
inasmuch as the tribe of Joseph is included, why is the tribe of
Manasseh (Joseph’s son) also listed, but not Ephraim? Another question: why, in the extant Greek manuscripts of
Revelation 7:6, is Manasseh’s name spelled in so many different ways? Israel
A conjectural emendation answers both of those questions: in the original text, the last portion of verse 6 did not refer to the tribe of Manasseh, but to the otherwise unmentioned tribe of Dan, and an early copyist misread ΔΑΝ as ΜΑΝ, and understood it to be an abbreviation of Manasseh’s name. Thus ΜΑΝ originated, and different copyists with different orthography proceeded to spell out the name. The Bohairic version supports this idea; in this verse the Bohairic text does not refer to Manasseh, but to Dan. Perhaps future advocates of thoroughgoing eclecticism, on the strength of the Bohairic reading’s intrinsic appeal, will bring ΔΑΝ into the text.
In Conclusion . . .
Some readers, looking over these passages, and the passages from the Gospels described in the preceding post, may feel a measure of consternation, particularly because seven of them – in Matthew 1:16, Matthew 28:19, Mark 1:1, John 1:18, Acts 20:28, Hebrews 2:9, and First Peter 3:19 – have been used as a basis for establishing doctrine. However, only in the case of First Peter 3:19, and the teaching that Christ visited imprisoned spirits, could it be argued that a doctrine stands or falls on the acceptance or rejection of a particular reading or conjecture (and even then, a case could be made that Paul teaches essentially the same doctrine in Ephesians 3:9-10, minus the specificity in First Peter).
(Apologists for Islam may sense a different sort of consternation, inasmuch as even with the allowance of conjectural emendation in the picture, the application of thoroughgoing eclecticism elicits nothing remotely close to the level of textual alteration that would bring the doctrinal teachings of the New Testament into harmony with the teachings of the Quran. The charge, often made by Muslim apologists, that the New Testament agreed with the Quran until Christian copyists altered the text of the New Testament, simply lacks a historical foundation.)
It is sometimes said (because Bruce Metzger said it) that New Testament textual criticism is both an art and a science. But it should be all science, and not art, because it is an enterprise of reconstruction, not construction. Its methods may validly be creative and inventive – even intuitive – but not its product. Conjectural emendation is the only aspect of textual criticism that involves the researcher’s artistic or creative skill.
No conjectural emendation should ever be placed in a compilation of the text of the Greek New Testament. At the same time, the task of proposing possible readings which account for their rivals, or which otherwise resolve perceived oddities in the extant text, serves a valuable purpose: to demonstrate the enormous weight of the intrinsic evidence in favor of such readings in the event that they are ever discovered in a Greek manuscript.
Quotations from the
ESV have been taken from The Holy
Bible, English Standard Version. ESV® Permanent Text Edition® (2016).
Copyright © 2001 by Crossway Bibles, a publishing ministry of Good News
Quotations from the NIV (2011 edition) have been taken from the Holy Bible, New International Version®, NIV® Copyright ©1973, 1978, 1984, 2011 by Biblica, Inc.® Used by permission. All rights reserved worldwide.
Scripture quotations marked
CSB have been taken from the Christian
Standard Bible®, Copyright © 2017 by Holman Bible Publishers. Used
by permission. Christian Standard Bible® and CSB® are federally registered trademarks of
Holman Bible Publishers.
Scripture quotations marked
NASB taken from the New American Standard
Bible®, Copyright © 1960, 1962, 1963, 1968, 1971, 1972, 1973, 1975, 1977,
1995 by The Lockman Foundation. Used by permission.
Scripture quotations marked CEV are taken from the Contemporary English Version (CEV), Copyright© 1995 by American Bible Society.
NET are from the NET Bible® copyright
©1996-2006 by Biblical Studies Press, L.L.C. http://netbible.com All rights
The NIV (1973 edition, no longer in print) is Copyright © 1973 by New York Bible Society International, and published by The Zondervan Corporation,
. Grand Rapids, Michigan 49506, USA
Misquoting Jesus: The Story Behind Who Changed the Bible and Why, is Copyright © 2005 by Bart D. Ehrman, and published by HarperCollins Publishers, New York, New York. All rights reserved.