Loading...In this post I’m going to try to do something I’ve never done before: actually explain by way of example the extent and kind of variations you find in our surviving Greek manuscripts. In doing so I hope to show: (a) there are lots of variations and (b) most of them involve nuances of meaning but rarely anything of huge significance (and lots of them don’t affect the meaning at all).
By way of introduction: I have previously indicated that virtually all translators use the Greek text established by an international committee of scholars for the United Bible Societies. I have also mentioned that this form of the text comes in two published versions. One is for translators around the world who are translating the NT into various languages into which it has not yet appeared. This is the kind of “student” edition that many first year Greek students use. That one is called the United Bible Societies Greek New Testament, and it is now in its fifth edition. That’s the one I’ve been describing.
The other is for scholars. It is called the Nestle-Aland Novum Testamentum Graece (Nestle and Aland are the names of two textual scholars who were largely responsible for producing the edition; the title is given in Latin to show that it is serious business for scholars, not for the faint of heart). This edition gives far more textual variants than the other, because it wants scholars to see all the relatively significant variations in our surviving witnesses. But I need to stress, it is nowhere near being exhaustive in its presentation of the textual evidence. There are hundreds of thousands of variations it does not note, but virtually all of those hundreds of thousands are completely insignificant and immaterial and really don’t affect a solitary thing.
To give you a sense of the difference between these editions, and to show what kinds of variation we’re talking about, I have chosen a random passage….
THE REST OF THIS POST IS FOR MEMBERS ONLY. If you don’t belong, JOIN, or you will NEVER KNOW! It costs little, gives lots, and every penny goes to help the needy. So JOIN!!!
Completely random, in fact. I opened my Greek New Testament and it fell to this page.
The passage is Mark 14:27-31, just five verses in length. Here is what it says in the NRSV translation:
27 And Jesus said to them, “You will all fall away; for it is written, ‘I will strike the shepherd, and the sheep will be scattered.’ 28 But after I am raised up, I will go before you to Galilee.” 29 Peter said to him, “Even though they all fall away, I will not.” 30 And Jesus said to him, “Truly, I say to you, this very night, before the cock crows twice, you will deny me three times.” 31 But he said vehemently, “If I must die with you, I will not deny you.” And they all said the same thing.
The UBS text (that is, the fifth edition of the United Bible Societies version) indicates one textual variant for these five verses, in v. 30. In several of our oldest manuscripts and in a number of Latin manuscripts and in the Armenian and Ethiopic versions, instead of saying “before the cock crows twice” it says “before the cock crows.” In other words for those witnesses, Peter will deny Jesus before the cock crows at all, not before the cock crows the second time.
Who cares? Well, maybe not a lot of people. But what is interesting is that this variant reading (“crows” instead of “crows twice”) is exactly what is found in the Gospels of Matthew, Luke, and John). And why does that matter? Because it could be conceived of as a contradiction: if in Mark Peter denies Jesus once or more after the cock crows the first time, then that would be at odds with what the other Gospels say.
It is precisely the fact that “crows twice” seems to stand at odds with what the other Gospels say that makes textual scholars pretty sure that this is what Mark originally wrote. The logic is that scribes who knew the other Gospels would be inclined to make them consistent with one another, and so would make a simple change of Mark (eliminating the word “twice”) to harmonize the passages. It would be much harder to explain why a scribe committed to the truth of Scripture would create a problem where there wasn’t one to begin with. And so “twice” is probably the original reading (there are other reasons for thinking so, but I’m just giving a brief analysis).
Now, just to illustrate the difference between the two most widely used editions of the Greek text I can compare this one variation with what is found in the Nestle-Aland (scholarly) edition. There you will find eight variations noted in these five verses. Eight times as many. But, at the same time, most are not as interesting or significant:
- v. 27: After “fall away” some mss (= manuscripts) add the words “because of me”; others add the words “because of me on this night.” That latter reading is what is found in the other Gospels (note: I’m counting this as one variant but you could see it as two variants since there are three different options for what the text originally said).
- v. 27: the words “the sheep will be scattered” are changed in terms of their word order (literally: scattered will be the sheep; it means the same thing) in some mss; and in some of these mss a different form of the Greek word “scattered” is used, which means almost precisely the same thing. So again you could count these as two different variants.
- v. 30: When Jesus says “I say to you, this very night…” some mss say “I say to you today, on this night…” (the word “very” – which in the Greek is actually “today” – is missing)
- v. 30: The same words “on this night” are changed in various ways in terms of their word order in the Greek; all of the variant ways of arranging the words, though, mean almost exactly the same thing. (The mss present three different ways of arranging the words)
- v. 30: “Before the cock crows twice”: as noted for the UBS text, “twice” is omitted in some mss; and the words get rearranged in various ways, all of which basically mean the same thing.
- v. 31: The words “if I must die” are rearranged (“if die I must”) in a way that doesn’t affect their meaning
- v. 31: the verb deny is changed from a future indicative (I will not die) to an aorist subjunctive (I shall not die). It means the same thing either way.
- v. 31: “And” is missing in some mss, including our single oldest and best manuscript for the passage.
So, that is the kind of thing you will find on every page of the Nestle-Aland text. Lots and lots of variations (even though many more variants found in the mss are not given) and most of them not terrifically important for the meaning of the passage (the ones not given are even less so). But some are kind of interesting, and one of them actually matters because it may involve a contradiction between Mark and the other Gospels.[/mepr-show]
Fascinating. You give good logical (and convincing) reasons why a scribe might delete “twice” from Mark. But can textual scholars point to other “examples” (not just in the Bible but in, say, other ancient writings) where it’s absolutely certain (eg, scribes confessed to doing it or multiple, contemporaneous, trustworthy writers had noted it had been done) that scholars dropped a seemingly unimportant word to make it conform to other sources? So that this example from Mark is another specific instance of a general rule about what happens in situations like this?
This question probably isn’t clear. What I’m trying to get at is a more “empirical” or “law-like” justification for decisions about, say, the original wording as compared to a rational, logical justification.
Yup, it is very common for scribes to harmonize their texts.
Are there differences in the text of the early church fathers like there are in the new testament ? For example, did scribes change some of what Justin Martyr or Irenaeus said ?
Yes indeed. But we don’t have nearly as many manuscripts so it is harder to detect scribal habits in those cases
Kind of along the same lines, do you believe there are interpolations in the writing of Josephus is regard to mentioning Jesus, or the Christ?
Yes, though I think *most* of the passage in Antiquities 18 goes back to Josephus himself.
I don’t understand how some people can hold to a strict biblical inerrancy view in light of these kinds of issues.
They tend to insist that the *originals* were inspired in their very words, but that human scribes messed up the words in places
Which sounds great to other fundamentalists, till you realize then we don’t have originals then…
I prefer Mark’s version, which isn’t to say I think it’s more likely to have happened that way. Possible that Peter told a story along these lines this afterwards (it’s not as if anybody else could have possibly born witness to his cowardice). Equally possible that it’s a completely made-up story, meant to represent the general failure of the disciples to intervene on their master’s behalf, or to go to the cross with him. Extremely likely that whatever account Peter might have given of his conduct that night would have been greatly embellished in the decades that passed between that night and the writing of the gospels.
But Mark, as so often elsewhere, has the more finely tuned ear for a line. “Before the cock crows twice” sounds better. And given his distressed mental state, it makes sense that Peter wouldn’t connect the commonplace sound of a cock crowing at dawn with Jesus’ prediction until he’d heard it twice.
And of course it doesn’t take supernatural abilities to know that frightened men will deny knowing someone about to be crucified by the Roman state, anymore than it does to know that roosters will crow in the morning–and Jesus might well have planted that notion in Peter’s head on purpose, because his purpose was to motivate his people to carry on his teachings after his death. And there’s no better motivator than guilt, is there.
It’s a powerful story because we all know it could have happened precisely that way–whether it did or not. Many of the most telling stories in the gospel require no belief in the supernatural at all. One of the reasons why it survived, when many a taller tale fell by the wayside. It’s a very human narrative, about human weakness, and the potential to overcome that weakness.
Is there currently a process for notating variant readings in sort of a descending order of authority? Like for instance something showing up in scores of early 4th century CE variants vs. an alternative reading only existing in a 6th century CE readings etc …
To a limited extent, yes. The apparatus of the Nestle Aland 28th edition is very dense and takes some expertise to unpack, but those who know how to do it can tell some main differences among the various variations.
In #6 do the Greek construction and conjugation really match the English version as you have it here with precision? I like that the alternative makes Peter sound like Yoda.
No, the problem is that Yoda-English depends on word order, and Greek does not depend for the most part on word order. Various sequences mean the same.
Whew! The contradiction is interesting.
How much help are computers in the study of these varying manuscripts? I was surprised when you said that nobody knows how many differences exist or even how many manuscripts exist?
Yes, computers have revolutionized the field
Clarification request: you wrote: “7.v. 31: the verb deny is changed from a future indicative (I will not die) to an aorist subjunctive (I shall not die). It means the same thing either way.”
Did you mean to write: “I will not deny” and “I shall not deny” instead of “die”?
Right! Scribal corruption of the text.
Prof Ehrman
I see the logic in thinking “crows twice” is the original reading. But how would that harmonization have worked? Are we to accept that both Matthew and Luke made the change independently of each other and then someone who had Mk, Mt & L (& J?) in front of them went back and made the change to Mk?
I realize we have to follow the textual evidence but wouldn’t it make more sense to think that the variant pre-dated Mt & L and that the variant was well known enough for both to have had access to that variant in their copies of Mk? And well known enough for John to have a variant tradition independent of the synoptics?
Thanks!
Yes, this is the kind of minor agreement of Matthew and Luke that is pretty easy to account for. They both simply thought there was no reason for the crow to crow *twice*.
Prof Ehrman,
do we have any textual variants on Matt 28:19?
We have some a church father (Eusebius) who leaves out the final bit (“in the name of the father, the son, and the holy spirit) ,but it’s not clear if he is quoting manuscripts that lack it or if he is just giving part of the verse. The words are found in all surviving manuscripts.
Hi Bart
Which English Bible version would you recommend for reading as close as possible to the original Greek? Thanks!
I prefer the New Revised Standard Version, which I especially like in an annotated edition, such as the HarperCollins Study Bible
Wassup with Mark 9:44 and 9:46? 🤔
With the notable exception of KJV/NKJV, all the popular translations I checked either omit these verses entirely or leave only the numbers behind — floating in mid-passage like the Cheshire cat’s smile with no body copy attached!
Young’s Literal Translation at Bible Gateway and all three Greek Interlinear websites I checked agree with King James, inserting in both places the lines, “where ’Their worm does not die and the fire is not quenched,’” that appear only in verse 9:48 in all other English translations.
Neither do these verses appear at the websites I found for the UBS or Nestle-Aland texts you mention as scholarly de rigueur. (In the search function that is, though I assume there is some explanation of their conspicuous absence for you who can actually read the content — but it’s all Greek to me. 😉)
Since the thrust of Mark’s passage is unaffected by this omission, the other 2/3 of the trinity can safely be returned to the Dept. of Redundancy Dept. But why did the NIV, NASB, RSV, etc. translators drop the CONTENT of verses 9:44 and 9:46, whether retaining or omitting the empty verse numbers, and reduce them to a footnote?
It is a question of the manuscripts; the most reliable ones don’t have the lines, and it is usually thought they have been imported by later scribes. (Most translations have a footnote explaining that.) The verse numbers are kept because new translations never renumber verses even if any are left out. Otherwise it would be mass confusion refereing to, say, Mark 9:47 when it would be a completely different verse in different translation.