The Majority Text

"The third illusion that attends the King James-Only position involves the evaluation of the actual evidence. King James-Only advocates are extremely reluctant to allow the empirical evidence to stand on its own merits. On the one hand, they are fond of insisting that “the majority rules” in textual matters. On the other hand, they are very careful about what they allow to count as a majority. For example, if all manuscripts of the ancient translations of the New Testament are counted, then manuscripts that support the Textus Receptus form a distinct minority. Moreover, according to the actual manuscript evidence, the manuscripts that support the Textus Receptus are not in the majority even of Greek manuscripts until the fourth century or even later. If the theory that “the majority rules” is correct, then the next two questions are, Majority of what? and, Majority from when?

The King James-Only movement can survive only by deploying a highly prejudicial definition of the word majority. Its defenders insist that very late Greek manuscripts be included in this majority but that very early translations be excluded from it. They revise history to explain the paucity of manuscripts that support the Textus Receptus before the fourth century. In fact, historical revisionism is a mainstay of the King James-Only argument. Their carefully reworked history is filled with heretics who deliberately miscopied the Scriptures; churches that rejected Alexandrian manuscripts; ecumenical councils that endorsed the Byzantine tradition; secret plots of Jesuits, Masons, Nazis, and Communists; and a variety of other irresponsible speculations, none of which can be shown to have happened."

K. Bauder,
One Bibly Only? (pg. 160)

"...Also, the Greek support for the TR wasn’t really a majority of manuscripts until the 9th Century, per James White. The impression I got in my experience of King James Onlyism was that the “evidence” and the role of “the majority of manuscripts” was quite important. That is what made the whole theory appeal to me as solid. When I found out that often King James Onlyists manipulated the evidence to suit their cause, I started down the disillusionment path.

Bob Hayden,
KJV Only? (Net blog)

Paragraph 1: – starts out fairly even-handed, and correctly points out that KJVOs “are reluctant to allow the empirical evidence to stand on its own merits”.

But why should they? Christianity and Empiricism are hardly equivalent. Christianity requires the rejection of many whole lines of philosophy, including Empiricism, materialism, deism etc.

We in the 20th and early 21st century have been essentially “bowled over” by the immediate successes and rapid advances of technology, and this has given a dubious credibility to the (many) bogus philosophies of scientific practitioners. But philosophers themselves have spotted the non-sequiter in this embracing of “scientific thought” in place of Christian (or any other) beliefs.

Empiricism has some utility, and nobody should deny that. As a physicist, I have great respect for scientific methodology, including independent testing and peer review, as well as skepticism over mysticism or naivity. But I and any other physicist should be the first to tell you, that it is *NOT* the various hair-brained philosophies proposed by scientists that is responsible for the largely gratuitous (and mostly accidental) progress of “science” (read: ‘technology’). Most scientists are lousy philosophers and religious quacks. Empiricism as a belief-system or even methodology, ironically lacks any Empirically derived credibility for itself! Saying Empiricism in any form won the “science” war, would be like saying that Roman Catholic policy defeated the Nazis.

This is precisely what is wrong with quacks like Hawkins, who want to make the ridiculous claim that belief in evolution deserves the credit for all scientific advance, or that atheism deserves the credit for electricity, computer-science, TV, and gasoline engines.

My first point about Empiricism should be clear from above. My second is equally important: A lot more than KJVOs need to reject Empiricism as the ‘ultimate arbitrator’ of truth, and a lot more people than KJVOs *DO* reject Empiricism.

There is a long tradition in philosophy of critical thought and rejection of Empiricism in all its popular forms, and any good philosophy course can furnish brilliant illustrations of its limitations.

It is perfectly respectable and easily defensible to reject modernist “Empiricism” as the only or final methodology for fact-finding, or establishing either physics or historical truth.

Next (still Para.#1), Bauder goes on to specifics, and attempts to find either inconsistency or hypocrisy in the position KJVOs take in interpreting the textual evidence.

Again, the first point is that a lot more than KJVOs are quite willing, quite capable, and quite within scientific standards to prefer the “Majority text-type”. Those who are *NOT* KJVOs, and who have impeccable credentials, like Dr. Maurice Robinson, or John Paul Heil, are quite within scientific procedure and plausibility to prefer the Majority text.

Its all about interpretation, and the relative plausibility of the historical reconstructions of the textual MS copying stream(s). Those arguing for the “earliest and best MSS” have hardly demonstrated their own historical reconstruction is the best option, and their scientific case is far, far from the natural legal mode of “beyond reasonable doubt”.

Whether “Majority rules” is the ultimate Canon or guide, is quite debatable even WITHIN the KJVO camp, with many rejecting this as the highest Canon in authority, overriding all others. Obviously, those who defend minority readings like the Johannine Comma, and other strong Latin readings (found in the KJV) just don’t fit the stereotype Bauder is painting at all. Majority obviously *DOESN’T* rule KJVOs.

Does it rule Majority Text supporters? Apparently not. Hodges/Farstad rejected Majority Rule when there was excellent genealogical evidence to use instead, for instance, in reconstructing the text of John 8:1-11, or the readings of Revelation. This is the Classic Majority Text, yet the very editors responsible clearly don’t believe “Majority (Over)Rules.”

And when we turn to a text like the Robinson/Pierpont Majority Text, we find the editors aren’t even KJVOnly. Dr. M. Robinson has gone out of his way to explain that he plainly *DOESN’T* hold KJVO views and doesn’t want to be associated with them. How then can his majority text-type be evidence of the attitudes of KJVOs, when they REJECT the R/P Maj. Text where it differs from the KJV, in favour of the TR?

Bauder is right to point out that the mass of Latin MSS are rejected by just about everybody (except conservative Roman Catholics). If the 10,000 some odd Latin MSS were merely counted, they would swamp the paltry 2,500 Greek MSS.

But Bauder is playing fast and loose here, since ALL TEXTUAL CRITICS including all those pushing the Westcott/Hort Aleph/B (‘Alexandrian’) readings, utterly reject any claim of priority or equality with Greek MSS for the Latin MSS. How can this be a fault of KJVOs, or even associated with them, when EVERY Textual Critic in the field since Hort has taken the very same position?

But we have to go beyond the mere fact that Latin MSS are ignored in reconstruction of the Greek, to the very strong scientific reasons WHY they are rejected (by everyone!). Its apples and oranges (or kumquats according to Mr. Avery).

The Latin MSS ARE good, good for constructing the LATIN text, but this is admitted by all hands to be a secondary text (translation of the Greek). Even the best possible reconstruction of the whole Latin textual stream would still only give us an early translation, with all the faults that any translation has, mainly its ambiguity in key TC questions.

Finally, it is equally obvious that contrary to Bauder’s assertion, KJVOs DO use the Latin, and even give it priority in many readings where the KJV/TR differs from the Greek Majority text. How then can he claim that the Latin MSS are ignored? They are certainly counted in hundreds of such places, and it seems that “majority rules” after all, *without* excluding the Latin MSS.

Again everyone (on all sides of all readings) acknowledges that translations have to be “devalued” or given less than equal voting power with Greek MSS. Even Greek MSS which have been influenced by the Latin are for the most part discounted or devalued by all parties. Its no shame to accept this, and indeed we would be surprised if KJVOs rejected this judgment, and instead began blindly counting MSS regardless of the age or language they presented, and giving them all equal voting power. In this KJVOs are not strange or unique, just the same as everyone else.

But our second bleat against this development of Bauder’s argument is this: Since all parties acknowledge that mere ‘voting’ is not scientific, practical, or acceptable, then *some* form of MS or Witness “weighing” or “weighting” must be put in place.

KJVOs can hardly be faulted for that, since no one else in the last 200 years has offered ANY weighted voting system for MSS, which can be objectively and consistently applied with deterministic and unique results independent of personal opinion or conjecture. The Eclecticists can hardly claim to have any such scientific methodology in place. How then can the KJVOs be singled out? They have acknowledged as loudly as anyone the need for a scientific “weighting” system for textual witnesses. Just look at Burgon’s “Seven Principles”. These all operate independently of MS age or ‘majority’ voting.

Bauder then makes an unsupportable claim about the early MS copying stream that is simply impossible to determine. He says, “according to the actual MS evidence, the MSS that support the TR are not in the majority even of Greek MSS until the 4th cent. or later”.

But this is a meaningless statement. Since we only have a mere handful of MSS for the first 6 centuries, what possible claim can be made about the “majority of MSS” for this period? The number of LOST manuscripts has been estimated to be in the 10,000 range for this early period. How then can a sample-size of less than 10 MSS on average adequately represent the vast and complex transmission streams spread across the Empire? Any statistician would laugh at this. There are accepted calculations for “poll-size” that can be applied here to show that the “poll” is worthless.

Finally at the end of the paragraph Bauder makes his first accurate observation of fact: “Majority of what?”, “Majority from when?” This is precisely the problem, but the two pieces can be more clearly identified than his vague questions:

(1) How do we weigh witnesses?

(2) What mutually exclusive ideal features do we adopt in creating our Voting System?

For the fact of the matter is, NOBODY can escape having to create *some* kind of voting system. And there are plenty of “Scientific American” articles on the Intransitivity of Voting Systems and the contradictory ‘desirable qualities’ we would wish for in any adopted voting system.

Any mathematician worth a tinker’s cuss can explain to Bauder that EVERY voting-system is a compromise of competing interests, and the designer must make hard choices.

peace
Nazaroo

Nazaroo,

First off, I’m presenting Bauder’s conclusion to a book which deals with specific arguments, so in a sense he’s recapping.

Secondly, I’m presenting his arguments piecemeal. He is well aware that there are many other arguments used besides the majority rules argument.

Finally, you may not have come across some of the language that I had in my KJVO background concerning the majority. The books I read weren’t nearly as nuanced as your comment here, explaining many factors that limit the usefulness of a bare majority rules mindset. The populist books act as if majority rules applies to the TR as a whole (not bothering to explain the many minority readings). This is the problem Bauder is exposing.

And as I said in the text, I think he is bringing up a point that the extant Byzantine texts don’t become a majority of available Greek texts until the 9th Century, actually, although he stays with a safer 4th Century. We can’t argue on the basis of non-extant texts that we don’t know about.

Hi Bob: I’m sorry if my post appears “critical” in the modern vernacular sense. That was certainly not my intent.

It is my understanding that one of the big reasons for having a discussion blog like this is to encourage people, especially KJVOnlyers, to think more critically (in the scientific sense) and logically about this.

My purpose is not even to take a side per se, but to show clearly the non-scientific components of any theories or arguments on any side. This is how science progresses, and discovery is made. We clear away the chaff and dross, and see what remains.

Bauder’s efforts are far from worthless, and your post of this summary is far from pointless. But again, as you say, the purpose of a blog-format is to reason together about the arguments people offer, and see what can be accomplished. I fully intend to engage with yourself, and anyone and everyone who has a real interest in the issues.

Your points are well taken regarding Bauder:

I think your intent (you can correct me if this is misguided), is to discuss the key points raised by Bauder’s statements. I am very interested in hearing all sides on this. Perhaps if I rephrase some of my points in the form of questions, they won’t appear to have the “tone” of ‘criticalness’ they may seem to convey in print (part of the problem of internet communication, still an experimental art!)

Looking back at my original post, It could have been much shorter and clearer. My points were essentially these:

(1) Empiricism or some (likely inadequate) form of “scientific methodology” cannot be the final arbitrator in resolving the issues of the NT text. For instance, one’s world-view and faith must be a contributing factor in how the problem is handled. There is no shame in introducing this element for Christian scientists. Any world-view affects how one formulates theories and interprets “facts”, including materialism, naturalism, and atheism.

(2) The more important, serious, and credible proponents of the Byzantine Text, (Sturz, Hodges, Farstad, Pierpont, Robinson etc., even Burgon) all want to treat the evidence comprehensively and fairly. They are not selling any quick and easy solutions, like a blind MS count.

(3) The 10,000 Latin MSS (and other versional evidence) should indeed be included in any comprehensive theory about the manuscript transmission and copying.

(4) No acceptable or universally recognised scientific method for weighing or counting MSS has been proposed or tried. Instead critics have drifted into an arbitrary “reasoned eclecticism”. This is a huge and glaring problem with Textual Criticism as a “science”. Until such a convincing systematic method can be created and applied, the field can have no scientific credibility whatever. Observers are left with trying to form some kind of tentative or interim opinion about various claims, theories and arguments by individual practitioners about minor variation units.

(5) Bauder raises the problem of Voting Systems, but does not solve it. This is an essential part of the task to form a plausible and credible scientific method for this field, and there is a vast literature already developed on this, which no textual critic has ever even referenced, let alone applied.

(6) Bauder points out, and Bob Hayden makes more clear, the central problem of the MISSING MANUSCRIPTS. This is a problem for everyone however, not just KJVOs. As Bob has succinctly said:

We have however, found a partial solution to what happened to these manuscripts, and expounded the problem more fully here:

http://adultera.awardspace.com/TEXT/missingMSS.html

This probable explanation for at least some of the MSS however does not bode well for the Alexandrian camp. If the MSS were recycled in the manner we propose, then so were their texts. This would also explain the large number of early independent text-types found in the later MSS. Many of them are copies of much earlier documents. If so, then the Majority Text position is far stronger than anyone has realised.

The main question hanging was:

(Q1) What was the “majority text” in the first 5 centuries, if there was one?

The answer, if the recycling theory is substantially correct, would be:

(A1) Essentially the same as that found in the majority of later MSS.

peace
Nazaroo

Hi Folks,

Nazaroo, there is a tension between your (2) and (3) when it comes to the Greek Byzantine proponents like Hodges-Farsted and Robinson-Pierpont. (Sturz and Burgon have their own complications.)

They simply do not consider the Latin (or early writers or internal evidences) as significant, except in tie-break mode for the small amount of Byzantine close-calls .. eg. 79-30%.

Afaik, there is only one major textual theory that considers highly the mass of Byzantine manuscripts yet also looks closely at the mass of Latin manuscripts (especially in inclusion-omission verses) and that is the Reformation Bible understandings brought forth by Erasmus and Stephanus and Bezae, unto the Genevan and AV and the historic Bibles throughout the world.

Shalom,
Steven Avery
Queens, Ny

Bob,

If this sort of argument is the best that Bauder’s got, I don’t think he’s going to get very far. He claims that KJV-Onlyists “are extremely reluctant to allow the empirical evidence to stand on its own merits.”
What does that mean? If it means that they interpret the empirical evidence instead of just listing witnesses and saying “So there’s your proof,” then such a statement is equally true of non-KJV-Onlyists. The canons of NTTC virtually demand that interpretation be an integral part of the task.

Bauder stated, “They are fond of insisting that “the majority rules” in textual matters.”
Yes; there is definitely a tendency among KJV-Onlyists to treat the Textus Receptus as if it is the same as the Majority Text.

Then Bauder claimed, “They are very careful about what they allow to count as a majority.”
The KJV-Onlyists count a majority of Greek MSS as a majority. That is normal. Bauder seems to be strangely criticizing the KJV-Onlyists for using the same definition of a “majority reading” that everyone else uses. (Now, if he were saying that the KJV-Onlyists falsely claim that the Textus Receptus has a majority reading where it doesn’t, that would be different.)

Bauder stated, “If all manuscripts of the ancient translations of the New Testament are counted, then manuscripts that support the Textus Receptus form a distinct minority.”
I am not sure what Bauder is suggesting; perhaps you can help me out here: maybe he meant that where the TR contains the same reading that is in the majority of Greek MSS, if we were to consult all the non-Greek copies, a different reading – one found in a minority of Greek MSS but in a majority of non-Greek witnesses – would have more supporters than the reading in the TR. To which the KJV-Onlyists might appropriately respond that this simply shows the folly of counting versional copies individually; since all such copies descend from the creation of their first non-Greek ancestor their weight should be boiled down; to allow them individual votes would be comparable to allowing a person in a political election to vote as many times as he has limbs, or digits, or hairs. Is Bauder seriously criticizing the KJV-Onlyists for failing to engage in such a flawed approach? Don’t you agree that such a criticism would be foolish?

Next Bauder wrote, “According to the actual manuscript evidence, the manuscripts that support the Textus Receptus are not in the majority even of Greek manuscripts until the fourth century or even later.”

If Bauder had just focused on TR readings which are supported by a tiny minority of MSS, he might have had a point. Instead, he ignored two key aspects of the MS-evidence:

All that Bauder is observing about the MSS prior to the 300’s is that the climate in Egypt treated papyrus more gently than the climate elsewhere.

Next, Bauder asked, “If the theory that “the majority rules” is correct, then the next two questions are, Majority of what? and, Majority from when?”
But the KJV-Onlyists have already answered that question quite clearly: What = Greek MSS (primarily continuous-text Greek copies of NT books), and When = from antiquity onward. (Generally, Greek MSS made after 1500 are considered unimportant, though.) It’s a standard definition.

Next, Bauder asserted, “The King James-Only movement can survive only by deploying a highly prejudicial definition of the word majority.”
Wrong. And by wrong, I mean, absolutely false! It is not “highly prejudicial” to simply count how many MSS support one reading, and how many MSS support another reading. And it is not “highly prejudicial” to realize that each version (or, to put it more precisely, each stage of each version), regardless of how many copies it contains, boils down to one witness, and that nine or ten non-Greek witnesses – even if they all oppose the majority of the Greek MSS – are not going to have much impact on elections in which 85-99% of the Greek MSS agree. (What is wrong is to adopt a variant merely because it’s supported by over 85% of the MSS, but that’s not the question at hand.)

Bauder continues: “Its defenders insist that very late Greek manuscripts be included in this majority but that very early translations be excluded from it.”

No; KJV-Onlyists are quite content to include early translations in the counting, provided that there’s a one-version-one-vote principle at work, instead of letting each non-Greek copy have its own vote. *
[this idea was picked up and commented on by Nazaroo further on below]

Now, regarding Bauder’s statements about KJV-Onlyists’ history-revision: it is certainly true that KJV-Onlyist writers have unjustifiably called the Alexandrian Text a recension made by Origen. This sort of mistake is not theirs alone; several major textual critics have suspected that the Alexandrian Text was a recension made by Hesychius. When the KJV-Onlyists catch up to the implications of the early papyri, logic will force them to admit that Origen did not originate the Alexandrian Text. But the main question about the origin of the Alexandrian Text is thus moved (i.e., moved into the 100’s instead of the early 200’s) rather than settled.

Also, Bauder overstated his case when, after a list that includes “heretics who deliberately miscopied the Scriptures” and “churches that rejected Alexandrian manuscripts,” he said that none of these things can be shown to have happened. Is he calling Tertullian a liar, or just a bad guesser? Is he unaware of how Severus of Antioch explicitly identified copies from Alexandria as the ones which exhibited signs of tampering in Matthew 27:49? Does he detect no scent of docetism whatsoever in Codex Bobbiensis?

Then, Bob, you claimed, “The Greek support for the TR wasn’t really a majority of manuscripts until the 9th Century, per James White.” That sounds like a line straight out of Dan Wallace’s “Second Thoughts” essay.
It would take more time than I’m willing to spend today to explain why his claim is specious, so I’ll just summarize here: it’s sort of like claiming that if Island A is populated by 10 five-year old penguins, and Island B has 1,000 one-year-old-penguins, then Island A, five years ago, must have been occupied by more penguins than Island B – as if a chart about lifespans necessarily says something about past population-sizes.

So, I encourage you to reassess almost everything that Bauder has said in the statements that you presented. He’s right about the wrongness of KJV-Onlyism, but his reasoning here is flawed. It would be much, much better to concede that the TR usually agrees with the majority-reading, and go from there.

I wonder what will happen when you fathom how some NA-27 advocates manipulate the evidence.

Yours in Christ,

James Snapp, Jr.

I’d like to address the looming spectre of parties on both sides allegedly “manipulating the evidence”, then I’d like to discuss an issue raised by the interaction of Bauder, Bob, and James.

(1) Many individual writers on both sides may be guilty of some minor fudging regarding data, or minimization of evidence detrimental to their own position. This is a natural human trait, and also a standard debating practice. Its hard to see how continuously raising this spectre accomplishes anything. It should not be minimized or maximized. No one is being totally fraudulent, (one hopes), but they must for the most part believe in their own position, and can justify in their own minds what “evidence” (though inconvenient) is not seriously significant, granting their own theory or view. All humans manipulate data to some extent, and its inevitable in the process of investigation and construction of theories.

Let us admit now that in formal books, pamphlets, debates and internet sites, each proponent will maximize his favourable evidence and minimize the problems. Lets move beyond that to the evidence itself and also the question of methodology, which IS germaine, and CAN be reasoned about.

(2) I’d now like to discuss a methodological fallacy which I call “The Collapsing Fan Fallacy”. I bring it up, because it is clearly an integral component of arguments on both sides, as James’ post above illustrates all too well.

Introduction: Manuscripts(MSS) which are closely genealogically related can have their relationships (copying dependance) displayed by a tree-chart which has the shape of an upside-down fan, beginning at an archetype, and fanning outward and downward (with time proceeding down) as more copies are made from key master-copies. It matters not that some masters are copied more than others, or skipped. The basic fan-shape remains a feature of the Display-Method (DM).

This shape should not be thought of as having a Platonic ‘life’ of its own. Its just a phenomenal accident of the display-method and purpose of the chart. Such charting is not an effective DM for dealing with loosely related MSS or MSS with a lot of ‘mixture’ or cross-bleeding of readings, borrowed from MSS not directly descended from the copies. This means the DM cannot show ‘everything’ or especially think for us.

What Textual Critics (TCs) of a certain school of thought intend by grouping MSS is far more than merely organizing data and reducing the relations to symbols (such as “A-text”, “OL”, “Byz” etc. ). They do this with a very conscious motive and wilful purpose, namely to attempt to solve the “Voting/Weighting Problem” (VWP). The most popular solution is to simply count “text-types” (groups of MSS closely related by text) instead of MSS. That is really what text-typing is all about.

Thus the actual numbers of MSS supporting a text-type or reading is reduced to a single vote: The Byz. “Text-type” would get one vote, the Alexandrian (B-type) one vote, etc. One can immediately see the attractiveness of this simplification, which seems to deal with the problem of counting in a single swoop.

Everyone concedes that just “counting MSS” would always result in the Majority Text, regardless of whether this text-type was always right, or even ‘inerrant’ or not. In fact all Textual Criticism would be reduced to simply collating all MS readings and counting the MSS. In theory, it could be accomplished in our lifetime with modern technology, beyond any reasonable doubt, using scanners and sophisticated recognition software. Of course textual critics would be out of a job, because TC would be reduced to mechanical housekeeping, and could be performed by workers in China or machines.

The convenience of grouping MSS, and treating them as singular entities is seductive and compelling, and has all the appearance of a scientific procedure. But this is the fallacy.

And please note before we continue that all parties are engaging in this fallacy. The Hortians would collapse the whole Byzantine group into one entity, and give it one weak vote. They would take the reconstructed archetype and give it one vote too, and so on with the Western, Caesarean etc. And this is only an initial step in the procedure. With all the mass of data reduced to such easily defined and manageable entities, the next step is to assign each a weight also, to fine-tune as it were the voting. Thus the Aleph/B text would get say 10 votes, while the Byz only 3, and the Western say 5. The Old Latin Version might be assigned a 1. Now the voting-count is easy. There is some play in readings, with various combinations of these entities playing off one another. Aleph/B might be overridden by a combination of Byz, OL, Syr, and West., but would take precedence against Byz/West alone.

But anyone can see the ridiculous artificiality of this: the weighting has been deliberately rigged, “balanced” to give the appearance of choice, and provide jobs for Textual critics, and accomodate new discoveries and theories of weighting.

Nor can other parties claim they do not engage in the very same fallacy, for all its seductiveness.

The TR-Onlyists and Majority-Texters would do the very same thing, as James so ably illustrates:

“KJV-Onlyists are quite content to include early translations in the counting, provided that there’s a one-version-one-vote principle at work, instead of letting each non-Greek copy have its own vote.”

This is James’ remarkable answer to Bauder! “one version one vote”!

And TR-onlyists would do the same, only weight the Latin vote more heavily than the Maj-Textists, so that the Latin readings they favour are saved.

There is no denying the arbitrariness in this procedure of “collapsing” a group of related MSS (a “fan”) into a single weighted “vote” (a “stick”!).

But a moment’s reflection will undeniably reveal that the “Voting/Weighting Problem” (VWM) has not been solved at all, only deferred and simplified and made more easy to manipulate to achieve desirable results. A scientist can have no hesitation in dismissing the whole enterprise as an absurd dance of subjectivity, something like the mating ritual of Vulcans on Star Trek.

For all its seductive beauty, the method is a joke.

There is no scientific reason, let alone explanation why a group of MSS should be reduced to a single “vote”, weighted or not. And no explanation is offered or any scientific procedure for assigning “weights”.

All the participants are guilty of the most shocking naivety imaginable.

Its “The Collapsing Fan” Fallacy.

peace
Nazaroo

Nazaroo,

Since the focus throughout my comments was Bauder’s statements, it is rather out-of-place to accuse anyone of “the most shocking naivety imaginable” inasmuch as you have not seen here either my own approach or my reasons for it.

I certainly take issue with the notion that “There is no scientific reason, let alone explanation why a group of MSS should be reduced to a single vote.”

There are several valid, non-arbitrary, scientific reasons for boiling down groups of MSS, and they are not an “absurd dance”. But a detailed presentation about that would require several pages, and would not pertain to Bob’s initial entry.

You seem to be objecting to a formulaic Sturzian approach (in which variants are elected and each text-type gets a vote), but that is not my approach. I wasn’t presenting my own text-critical approach in my previous comment — at least, not in any sort of detail — so I don’t see how anyone could reject it on the basis of that comment.)

Yours in Christ,
James Snapp, Jr.

Dear James:

The “naivety” comment was a bit of fun. But the point remains. We concede that you were only illustrating the approach of conservative textual critics with your comment about “one-version-one-vote”. We accept that this isn’t necessarily your own approach.

Secondly, let the approach I illustrated in my post be classified as “formulaic” (perhaps meaning the use of a fixed formula), and let it be labelled “Sturzian” in honour of our friend Dr. Sturz. I’m good with all that.

Now lets get down to business:

On the positive side, we are asserting only that:

The prize is still available:

By all means let the games begin. Let all TCs bring forth their theories of METHOD (not theories of the NT text itself) to be examined in the light of day, and judged the worthy winner. But meanwhile, I will continue to expose the logical and methodological flaws in current attempts to reconstruct the NT text, *without* a convincing scientific methodology in place.

yours in Christ,
Nazaroo

James, You bring up some good points. One response: Bauder stated,

I am not sure what Bauder is suggesting; perhaps you can help me out here: maybe he meant that where the TR contains the same reading that is in the majority of Greek MSS, if we were to consult all the non-Greek copies, a different reading – one found in a minority of Greek MSS but in a majority of non-Greek witnesses – would have more supporters than the reading in the TR.

To which the KJV-Onlyists might appropriately respond that this simply shows the folly of counting versional copies individually; since all such copies descend from the creation of their first non-Greek ancestor their weight should be boiled down; to allow them individual votes would be comparable to allowing a person in a political election to vote as many times as he has limbs, or digits, or hairs.

Is Bauder seriously criticizing the KJV-Onlyists for failing to engage in such a flawed approach? Don’t you agree that such a criticism would be foolish?

To this I would say most KJV Onlyists I have read don’t think this nuanced. They argue against forcing all the Greek MSS which form the majority to be reduced to a single vote. If we give non-Greek mss a single vote for each development period of their version (thus excluding the thousands of Latin mss from consideration as a majority), wouldn’t we be duplicitous not to exclude the thousands of late Greek mss from having each a single vote (in the world of majority vote wins)?

What you bring up about Tertullian and Severus is interesting. My assessment as I think through this may be to become more critical of the NA27 and more interested in the Majority Text position, but I’m still studying.

Have you read populist level KJV Only materials? If so, do you chop them to pieces like this too? Just wondering. Do you give them the benefit of the doubt, where you don’t for Bauder. Have you read the entire book either?

Bob, Some answers for you:

Whether that’s the case or not (I suspect that most KJV-Onlyists, like most people, are grossly underinformed about the versional evidence as a whole), the point still stands that Bauder was essentially criticizing the KJV-Onlyists for defining a majority of Greek MSS as a majority of Greek MSS.

That’s a good question, and it’s close to being *the* question. Let me re-present the problem.

Non-Greek MSS can be collected into groups, and we can frequently identify specific points of origin for those groups.

Some versions are stratified; in the Vulgate-stream there’s the form that resulted from Alcuin’s revision-work, and much later the “Paris Bible” had a distinct form, so that what is distinct in those forms, in however many copies they occur, can be classified collectively.

What about Greek MSS? They, too, can be collected into groups. But unless the members of a group display some special feature that pins them down historically (such as a colophon or a unique format), it is not so easy to identify when and where their special readings originated. If we could identify a specific point of origin (other than the point of composition) for a large group of Greek MSS, analysis would be greatly simplified. (That is essentially what Hort did when he proposed that all the Byzantine MSS have a specific ancestor: a MS that was the product of a recension that was carried out either by Lucian of Antioch or by one of his contemporaries.)

But there is a difference between a scenario in which a group of MSS must share an earlier ancestor later than the archetype (as there must be in the case of the versions) and a scenario in which it is merely possible to posit an earlier ancestor later than the archetype.

If we don’t want to build transmission-models arbitrarily, then we need to work deductively, by tracing the history of incorrect readings (since, unless an erroneous reading independently recurred, community of reading implies community of origin) and sketching out a transmission-history on that basis. (Of course if one starts from the premise that the TR has no incorrect readings, or from the premise that the majority reading will always be correct, the shape of such a transmission-history will be a foregone conclusion!)

Reconstructing and Working with Text-types

If, in any given text-type, a group of late MSS can be shown to share numerous erroneous readings that lack support from earlier MSS in the text-type, then it would be fair to consider those late MSS collectively rather than individually. But sooner or later, as one works from the late MSS backwards, one reaches the earliest stratum of each text-type. And when that point is reached, the reconstructed sub-archetypes for each text-type can be compared.

Each sub-archetype should be given more weight than is given to any single later stratum of any text-type. But the later strata should still receive some weight, since they may reflect the effects of mixture (involving intrusions by a reading which may be early) rather than scribal inventiveness (involving novel variants).

The Versions [early translations] should be treated in basically the same way, with at least two important differences:

(Example: the early Sahidic version is very closely aligned to the same text displayed in B, so the representatives of the earliest strata of the Sahidic version, whether they number two or two dozen, collectively should be given weight equivalent to a strongly Alexandrian early MS.)

The ratio of weight given to a version should be converse to its degree of dependence upon a Greek sub-archetype.

So the answer to your question is a qualified yes; some groups of Greek MSS should be weighed collectively rather than individually, the way versions ought to be weighed.

The sub-archetypes should each be similarly weighed. Unlike what is done in the NA-27 apparatus, in which the Byzantine Text gets a single siglum ['Byz"] and Alexandrian witnesses are listed individually (giving many readers a skewed impression of the evidence), when a consensus of the major witnesses of a text-type collectively agree, it would be better to cite them collectively than individually.

(One problem with this approach is that the Alexandrian MSS frequently have no real consensus. But there’s a way to deal with that, involving the classification of MSS into text-types according to percentages of agreement at some specific points rather than according to percentages of agreement in general. But that’s another subject.)

The Prize-Fight Analogy

So let’s imagine how the evidence might look after everything has been being boiled down and assigned to a weight-class. A reading supported by all the sub-archetypes could be considered a super-heavyweight. A reading supported by most of the sub-archetypes could be considered a heavyweight; a reading in important MSS diverging from their sub-archetypes with good patristic support could also be considered a heavyweight. A reading shared by one sub-archetype and by two early versions that usually disagree with its supporters could be considered a welterweight. At the other end of the scales, a reading supported exclusively by a smattering of late minuscules could be considered a featherweight.

Now, in a Sturzian approach, after weighing the various heavyweight contenders, there would be no actual boxing-match. Each heavyweight boxer steps onto the scales, cumulative totals are calculated for competing readings, and the reading supported by the most weight wins. Instead of a Majority Text democracy, the Sturzian approach is a peaceful Ogkocracy – rule by the heaviest [textual weight].

   Accommodating Internal Evidence:

That is neat and formulaic, but I prefer an approach in which competing variants compete in the ring, not just in the weigh-in — i.e., an approach in which after the external support is taken into consideration, variants compete to see which one explains its rivals, which one is more difficult, which one is least harmonious, and so forth.

But to explore all that would get away from what you asked about. I hope this has answered your main question.

You asked some other things:

Bob H: “Have you read populist level KJV Only materials?”

Yes. Not each and every thing, but quite a bit.

Bob H: “If so, do you chop them to pieces like this too?”

If someone asks me, and if no one has adequately done so already, and if it looks like offering an analytical critique would be more fruitful than other things I could be doing. But I don’t try to steer sunken ships. If I write about something spontaneously, I probably consider it capable of being seaworthy, so to speak, even if almost everything I say about it pertains to how it should be repaired.

Bob H: “Do you give them the benefit of the doubt, where you don’t for Bauder.”

Generally not, although I like to assume that they are sincere, unless some reason arises to think otherwise.

Bob H: “Have you read the entire book either?”

No, just the excerpts you’ve presented and some of what is available online.

Yours in Christ,
James Snapp, Jr.

Greetings to James and Bob:

James’ description here is quite articulate and detailed. This enables us to organize it to make clear some of its foundations:

Here James lays bare a foundational difficulty in a making a truly unbiased scientific approach. In this case “premise” is probably too weak a word. Foundational Axiom would be closer to the mark:

If textual critics (TCs) accepted the assumption (as per TR-onlyists) that the Textus Receptus (TR) is perfect, then no textual criticism can even begin. Similarly, if TCs accepted the assumption (as per Maj-defenders) that the Majority Text (MT) no serious textual criticism can be performed, other than to collate majority readings, and perhaps adjudicate split readings.

Therefore, most TCs feel compelled to reject BOTH of these premises as Foundational Axioms, and indeed they feel compelled to formulate exclusionary substitute-Axioms, vis.,

These two premises, presumptions, are adopted as Axioms by virtually all academic TCs, and have been since Hort. But they are unacceptable and unscientific, for the very reason that they are (a) unproven postulates, and (b) unnecessary axioms for the engagement of TC.

That is, scientific Textual Criticism (STC) does NOT require either of these axioms, and not adopting them does not hinder scientific investigation in any way.

To continue analyzing James’ description:

What does James mean here? I think he is saying this:

This is what James means I think, when he says [early translations] *must* have an ancestor later than the archetype (the autographs). But turning to a contemporary Greek manuscript (MS), James might say this cannot be established with the same confidence:

Against this however, one must make the following caveat:

From this, the idea of the established “secondary-ness” of early translations becomes an ever-retreating mirage.

Finally, lets consider the 3rd point in James’ paragraph (although in order 2nd):

Now the rubber hits the road. James has begun articulating what textual critics must actually do to reconstruct transmission lines, and a history of the text.

He is suggesting:

Lets pause a moment to show this: James says, “by tracing the history of incorrect readings”. To make this even clearer, he gives the WHY too:

What does this mean? He is saying that in some (less probable) cases, the textual variants could give a “false-positive”. How? If a particular passage, because of a slew of similar phrases, line-endings, or clause-starts, was *prone* to a certain type of common error (say homoioteleuton), then it is possible that two manuscripts could have the same *error* by mere coincidence: Each scribe unluckily made the same blunder independently, even though they are not part of the same line of copying. In that case, the shared error would NOT demonstrate a genealogical relationship between the two MSS.

Though not impossible, this would be rarer than cases where an error of one scribe was simply copied by another.

Also limiting concern for this is the rapidly increasing improbability of coincidence as we increase the number of shared blunders: If there were a 20% chance of coincidence with one shared booboo, (1/5) then two such coincidences would be 1/5 * 1/5 = 1/25! = only a 4% chance, three (1/125) 0.8% and so on. Very quickly, it becomes clear that no matter how probable we make coincidences, whole collections of coincidences become essentially impossible.

When then, for instance, Aleph/B share some 70 such homoioteleuton errors, we know that coincidence is effectively ruled out, and these readings must come from either direct dependence or a common ancestor or conscious editing policy.

All this deductive reasoning however, is based upon two key assumptions:

All this constrains the kind of errors that can be used for tracing genealogical relationships to limited sets that are well-understood.

Now we can understand why genealogy-style reconstruction depends upon errors being created, copied, identified and understood, ...and then finally….REMOVED FROM THE FINAL reconstructed original.

This is precisely where all textual-critica practices have fallen on their face for the last whole century.

Nobody has bothered to remove the errors from the reconstructed intermediate texts. Instead they have moronically introduced them into the working Christian Bible.

peace
Nazaroo

Nazaroo, I’m glad to see that you are feeling well enough to write; may God speed your recovery.

Let’s revisit some of your recent questions and concerns.

(1) You stated that virtually all academic textual critics adopt the premises that neither the TR nor the Majority Text exactly represents the autographs. These are not arbitrarily posited premises; they are based on observation. You claimed that “they are (a) unproven postulates, and (b) unnecessary axioms for the engagement of TC.”

At the very outset of text-critical investigation, yes; these statements would be unproven. But we are not just beginning the task of textual criticism and it is clear (at least to me) from the work of earlier researchers that neither of these claims is true. And, yes, neither statement is necessary to conduct textual criticism, but such hypotheses must be tested, and if they fail, as I am convinced that they do, then this should be acknowledged.

(2) You asked what I meant when I wrote, “There is a difference between a scenario in which a group of MSS must share an earlier ancestor later than the archetype (as there must be in the case of the versions) and a scenario in which it is merely possible to posit an earlier ancestor later than the archetype.”

Simply that logic demands that a version is later than its base-text, whereas logic does not demand that the point of origin of a text-type is later than the autographs. That is something that may be logically deduced after a consideration of evidence, but it’s not a dictate of sheer logic.

That’s why it is not sound to give the Byzantine MSS only the weight of a 3rd-century MS unless one has amassed evidence showing that they all descend from such a MS, rather than from the archetype without such an intervening ancestor-MS in the equation.

(3) When I wrote that "the first step in developing a transmission-model is to trace the history of incorrect readings, since, unless an erroneous reading independently recurred, community of reading implies community of origin)” you asked, “What does this mean? He is saying that in some (less probable) cases, the textual variants could give a “false-positive”. How?”

Via the independent recurrence of the same erroneous reading, of course, as you surmised. This is not something for which the probability can be reliably statistically calculated; different erroneous variants will have different levels of probability of independent recurrence.

You mentioned that “When then, for instance, Aleph/B share some 70 such homoioteleuton errors, we know that coincidence is effectively ruled out.”

I don’t deny that Aleph & B show signs of a historical relationship and common descent.

The thing to watch out for, when an analytical approach is used which assigns special favor to the reading distributed among the consensus of text-types (i.e., Sturz’s approach) is the occurrence of erroneous readings in witnesses which usually disagree. Sturz’s approach seems very vulnerable to this phenomenon.

Yours in Christ,
James Snapp, Jr.

Dear James:

Thank you for your prayers re: my health. On top of other serious matters, I also have contracted a pinched sciatic nerve, from attempting to remove the wheel of the family vehicle, as well as from sitting down to work at my desk for 10 hrs / day. I would be grateful if you were to add this to your prayers, as this makes my work much more difficult.

It may be true that accurate numerical estimates of probability for individual events cannot be given with absolute certainty: however this does not impede scientists from giving accurate estimates of accumulative results: Its done all the time in physics for instance.

Example: We cannot predict the velocity (speed and direction) of any single atom in a cloud of gas. But we can nonetheless make very accurate measurements and predictions regarding temperature and pressure and behavior of that same gas, in a known environment. This branch of physics is called, naturally enough, “statistical physics”. The same is true of quantum mechanical observations.

How does that apply here? Let me give you some example of the principles involved.

(1) It is very often the case, that although we cannot give a numerical value to the probability of an event, because we cannot give any scientific procedure for measuring or calculating it, that we can nonetheless make reasonable general statements about that probability.

(2) For instance, all probabilities must fall between .000…0001 and 1.0, since a probability of “one” = certainty. This is just the scale created by converting percentages into fractions.

(3) It is agreed that the general procedure for combining probabilities is to multiply them together. The probability that event A /and/ event B both occurring is simply the probability of A *times* (i.e., multiplied by) the probability of B. For three events, its A x B x C. This is a very accurate universal statistical law continually tested by gambling casinos (and physicists) daily.

(4) We can compensate for not having a procedure and not knowing the actual probability of individual events, by erring a very large amount on the “conservative” side, to account adequately for any error in a guesstimate.

Lets see how that would work here: We know that the probability of two scribes making the same blunder independently by coincidence is much less than 100%, or even 90%, (in fractional terms, 9/10). We know this, because for every such known singular blunder, we can find some 1,250 other MSS WITHOUT the blunder: the ratio then is more like 1/1250, much smaller than 9/10.

Almost every known error by homoioteleuton for instance presents in a very small fraction of available MSS. These other MSS represent the times a scribe did NOT blunder independently in the same way by coincidence.

By inspection, we can also observe that most other blunders, and also deliberate edits, *do not present features that would allow us to mistake them for homoioteleuton errors.* This is very important, for it adds to the confidence and accuracy of our being able to identify the majority of homoioteleuton errors.

Put another way, other changes to the text don’t look like homoioteleuton, and most of the time cannot be mistaken for it, which would throw off our ability to know how often homoioteleuton errors are made (*independently* or not). If homoioteleuton errors could “hide” in the text in significant numbers, then we could not get accurate estimates of how often the same error could occur independently.

But we *can* estimate that over 90% of the time, homoioteleuton blunders are NOT independently made, because most shared homoioteleuton blunders are only found in a small number of extant MSS. Thus, almost every homoioteleuton blunder is only found in a handful of MSS (less than 10) in a sample of over 1000. (1/100 or less).

Even here, we can quite naturally suspect that this handful of MSS may actually be related by genealogy (interdependence of text) or by mixture (borrowing of readings). How can we guesstimate the probability that a given homoioteleuton instance is copied rather than independently generated? By looking at the number of MSS that *don’t* independently generate the same error. These counts give us reasonable OUTSIDE limits to how often an error could independently occur, based on what actually *did* occur.

The same is true of flipping a coin, a truly random and uncontrolled, unpredictable event with a 50/50 (.5) probability. Although its not *impossible* to flip a coin 1,000 times and get “heads” every time, it simply doesn’t happen in real life even once in a thousand years. And we go with what is probable, not what is improbable. Its a safe bet that flipping a coin even 10 times in a row is highly unlikely: (.5)^10 = 1/1024 (once in over a thousand runs of ten) = a 0.1% probability.

Turning back to the MSS, even with the most generous estimates for individual cases of independent homoioteleuton errors (say 1/10, making the probability 10 times as likely as the data suggests!), the chance of the same two copyists making the same 10 homoioteleuton errors in a row would be (.1)^10 = 1/(10^10) = 1 in TEN BILLION. [The more realistic calculation, using 1/1000, (.01)^10 = 1/(100^10) = 1 in 100 billion. .000 000 000 000 000 000 01]

Armed with this knowledge, what is the probability of two scribes independently making 70 homoioteleuton errors in exactly the same places? (1/100)^70 = 1e-140 (in scientific notation). This number is so large, that they have a new name for it, using the “google”. A “google” is 10^100 (1 followed by 100 zeroes). Our number is a “google” to the 40th power! = (google)^40.

As with large numbers of random atoms giving statistical certainty of the behavior of a gas, so also large numbers of homoioteleuton give statistical certainty of a common genealogical ancestor, direct copying, or deliberate mixture (an editing policy during copying).

Thus it is true (as in many other branches of science) that we can through the application of large numbers of instances of an event, calculate very reliable probabilities for a set of events as a group, even without any procedure for calculating individual probabilities.

This is done empirically by application of observed data: We know the probabilities for the events that happened, by observing what actually did happen in a statistically healthy sample of cases.

P.S., to prevent anyone from being misled, its the *denominator* that contains (google)^40. This means that the chance of Aleph/B being independent is only 1 in (google)^40, and this was calculated using a very generous margin of error for the probability of the individual homoioteleuton events.

peace
Nazaroo