Dr. Ehrman, Textual Instability, and the Number of Variants
I think I'll chime in as well. First, Dr. Ehrman's name has
one "n," not two. Second, when Christian apologists refer to the
wealth of manuscripts attesting to the NT text, that should be
understood as relative to the much smaller number of MSS attesting
to other ancient texts. Now, on one hand, the existence of so many
MSS is not an absolute guarantee that they preserve the text
accurately. On the other hand, one could illustrate the situation:
if only one person sat down to hand-copy a text, and made mistakes,
it would be difficult for future examiners of the text, without the
master-copy, to locate and identify the differences between the
master-copy and the copy. But if four people were to sit down to
hand-copy a text, one of them might make a mistake where the other
four did not. As the number of independent copyists increases, the
probability that the original text will be completely lost at any
given point decreases. The NT MSS can be categorized into different
groups, with the result that each group, or text-type, occupies a
role like those four copyists. Even though it still cannot be
proven that the original text has been preserved in one or more text-
types, at any given point where the MSS disagree, a scientific case
can almost always be made for the primacy of one reading.
What exactly is "almost always"? Well, in 1881 Hort listed 60
passages in which no text-type appears to preserve the original
text. That's 60 variation-units, out of thousands (and hundreds of
thousands of variants). Even if one were to double (to 120) or
triple (to 180) the number of passages where one feels that no MS
preserves the original text, the amount of non-extant original text
would be extremely tiny. In the most popular hand-held Greek New
Testament currently used by NT textual critics, the number of
passages where the adopted text is not extant in any Greek MS = one
(in Acts 16:12).
But possessing the original text, somewhere in the MSS, and
discerning where it is, are two different things! And that's where
textual criticism enters the picture.
Yusuf: "How accurate is this line of argument?"
If one meant that a plurality of local texts safeguards against the
loss of original text, it would be accurate to say that over 99.5%
of the text is preserved in the manuscripts more like 99.99%. Dr.
Ehrman is correct that we don't *know* the history of the text prior
to our earliest witnesses, but that does not preclude the
construction of an archetype-text which is 99.99% stable (as long as
the evidence remains the same). It's a tentative sort of stability,
though, because there's always a possibility that new evidence may
result in some drastic adjustment. Nevertheless, the claim that a
higher percentage of the text has been lost is just as faith-based
as the claim that none of the original text whatsoever has been
lost. Probably, though, those making the "99.5%" claim mean
(rightly or wrongly) that the current critically revised Greek NT
text is assuredly the same as the original text 99.5% of the time
and the remaining .5% is presumed to be inconsequential.
Yusuf: "When variations and omissions are pointed out to them, they
appeal to the fact that this does not amt. to or have any effect on
The claim that no variant affects a Christian doctrine is simply not
true. But no basic Christian doctrine hinges on any one passage.
There are some secondary teachings regarding which a lot stands or
falls depending on a textual variant: for instance, did Jesus
declare all food to be clean in Mark 7:19? It depends on which
variant one accepts! Accept one particular variant, and the
question is settled. But even if one considers the text unclear or
questionable in Mark 7:19, the teaching that all foods are clean is
stated and/or conveyed elsewhere in Peter's vision in Acts 10, in
the apostolic decree in Acts 15, and in the letters of Paul. That
may be a very important consideration for some groups within
Christendom who want to maintain that they should eat only kosher
foods, but it is not exactly a tenet of Christianity. Perhaps a
good way to put it would be to say that the adoption of any one
variant does not have a pivotal effect on basic Christian teachings,
either because the variant is inherently untenable (i.e., no case
for its authenticity can be made), or because it does not affect the
meaning, or because the previously-conveyed meaning is communicated
elsewhere in Scripture, or because the doctrine being affected,
however important it may be to a particular sub-group of Christians,
is not basic to Christianity.
Yusuf: . . . "The bizarre claim that all of the NT can be
reproduced by means of the patristic citations if the manuscript are
It would be a large project to demonstrate that this is true, but
when one considers how many patristic writings exist, and also
considers that some of them are commentaries in which large portions
of Scripture are cited, it would not be surprising to find that the
entire NT text, as represented by one text-type or another, is
embedded in patristic writings up to John of Damascus.
Yusuf: . . . "The allegation that there is not much difference bet.
The Byzantine text type and the Alexandrian text-type(?)."
Of course there are significant differences between the Byzantine
Text and the Alexandrian Text. To illustrate this very point I
recently added to the Files at the TC-Alternate discussion-group an
English translation of the Gospel of Mark, based on the traditional
Byzantine Text, displaying sense-affecting variants adopted in
critical texts almost all Alexandrian in superscripted bracketed
text wherever they appear. There are something like 550 sense-
affecting variants between the Alexandrian Text and the Byzantine
Text in Mark. But the number of startling, hard-impact differences
between the Alexandrian text of Mark and the Byzantine text of Mark
is much smaller, as you will see if you access the file and read it
It is no overestimate to state that there are probably 400,000
variants in the Greek NT manuscripts. By my calculations, the
number of variants is probably higher, over 500,000. One variant
shared by 1,111 MSS does *not* count as 1,111 variants. That was a
claim made by Norman Geisler, and carelessly echoed by his readers.
(Daniel Wallace recently corrected Geisler's approach in an online
essay.) There is something to the idea that one difference can be
counted multiple times, but it goes like this: if copyist A spells
a word one way, and copyist B spells the word a different way, and
copyist C spells the word a different way, then if that word appears
in the text 20 times, each use is considered a separate variant so
if three copies, made by copyists A, B, and C, were to spell just
one word differently, and that was the only difference between their
manuscripts, this one difference would be expressed in *60*
variants. So a few differences in the spelling of often-repeated
words can produce huge numbers of variants.
Yours in Christ,
James Snapp, Jr.
- Bart Ehrman wrote:
Well, I don't find this kind of polemical shotgun back and forth (he said / she said, on multiple fronts) to be all that useful as a mode of discourse.<<
Nor do I and I wish you'd cease from engaging in it. I made 2 points. That's all. In both, you've been quite defensive and even petulant.
Nor is this present a mere "he said/she said" disagreement: its a "Bart said vs. Bart said" disagreement that you keep attempting to slide past. The best thing it seems to me to do would be either clarify the apparent contradiction, or just simply admit that your response to my comment responding to Tommy Wasserman was a bit overstated and in need of qualification. Easily done. In such a format, everyone makes statements that need qualification.
>> But if you have a specific issue you'd like me to address, I'd be happy to.<<
I believe I've already raised them. And let's recall that it was *you* who responded to my response to Tommy Wasserman regarding ancient and medieval scribal practices.
>> I will, though, deal with the final question you raise. Just to recount the conversation up to this point, to make sure we're on the same page. You indicated that I claimed that there had been a "wholesale, seamless rewriting of the original" somewhere in the manuscript tradition; I asked you what gave you the idea that that was something I claimed, and your reply (below) is that you got the idea from me. And you would like me to clarify.<<Good so far.
>> So to clarify: I don't know where you are getting the idea from that I think there was a wholesale seamless rewriting of the original;<<In hind sight, perhaps I overstated it. Are you not introducing the possibility that such happened when you claim that our ignorance is such that we have no inkling what happened after the autograph was made. Further you stated: "If someone can explain to me the logic of appealing to an author's style when you don't think you can get back to his words (hence his style), I'll eat my Westcott and Hort!"
This suggests that it the author's text is unrecoverable in any real sense: i. e. that it was changed, that we can not recover it, and that such change is undetectable since we can not make appeals to the author's style etc. Now you state that "seamless rewriting" is not what you had in mind, ok, fair enough. But it doesn't let you off the hook. Your statements that you can't get back to the author's words and so therefore can't appeal to an author's style suggest that a) the process of reconstructing a text can not get back to the author and b) if we can not get back to the author, its suggests a stage or stages of more than mere interpolation and a few changes. If we take seriously the notion that in the period between author's copy and initial text considerable changes may have taken place that left no trace in the textual tradition, then we have to apply other tools, such as authorial style. If that isn't applicable, however, then we have to posit that those changes between authorial copy and initial copy were so thorough and complete as to be indistinguishable on close study of style, syntax, vocabulary etc. Possible, but given again what we know about ancient scribal practices (and even those making personal copies), is this likely on anything except a theoretical level?
>> I don't recall ever saying such a thing, and don't recall ever thinking such a thing, and so I don't know what you're referring to. If I *did* say this, I guess I'll have to defend it!<<That's illogical. There are many other responses possible besides defense.
>> I just don't remember ever being certain that we know what happened in the early stages of the tradition. So does that clarify it?<<It does somewhat, but not entirely. Certitude, however, was not what I was suggesting or after. Certitude isn't possible. Probabilities however are.
>> I don't *know* what happened in the earliest decades when the text was being copied (I certainly don't know that their was a wholesale seamless rewriting of the autographs; and I equally don't know that there wasn't). And I don't believe you know either! *Do* you think you know? If so, how do you know?<<
Do I know? Of course not. But I don't know that the planet on which I dwell will continue to rotate on its axis in the same motion and the same speed so that the sun will give the appearance of "rising" in the morning over what is to me my eastern horizon. But based on the current evidence available to me, I can make a probable prediction that no appreciable change will occur between now and then as to change the nature of that rotation. Likewise, given what we know, we can speak in probabilities about textual reconstructions too. With Dr. Wasserman, I think the best and simplest explanation is that there is no significant difference between authorial copy and initial copy, and that it is more likely that redactional activity between authorial copy and initial copy are detectable...and hence appeal to an author's style remains a valid tool.
Surf the Web in a faster, safer and easier way:
Download Opera 9 at http://www.opera.com
Powered by Outblaze