Julian

Argh, I wasn't going to get involved in this thread and I am not going to. However, there are a few issues that struck me as particularly annoying.

First of all, what Wilker did was a Principal Component Analysis which, contrary to Nazaroo's bizarre claim that it is non-standard (you say you are a scientist yet you do not understand PCA?), is a standard technique that can be applied to many types of statistics. It is a essentially an n-space filter designed bring to the front the most significant components and throw away noise.

Wilker also explains how he determines the textual variations and their relevance. And, while he does not list the exact variations (why should he?), he makes it clear what his techniques are and how he classifies the variations. The steps can be reproduced easily by anyone with the patience to do so.

The study is here: http://www1.uni-bremen.de/~wie/pub/Analysis-PCA.html

Nazaroo also seemed to imply that Wilker was being dodgy on the issue of B (03) umaluts which is strange, and maybe I am misreading Nazaroo here, seeing how Wilker has written many pages on the topic: http://www1.uni-bremen.de/~wie/Vaticanus/umlauts.html

He also discusses Didymus and many other things here: http://www1.uni-bremen.de/~wie/TCG/TC-John-PA.pdf (Not sure if this has been linked already but a second time can't hurt).

Nazaroo fails to see a few things that appear obvious to the rest of us. Having read a significant portion of his PdA website I noted right away that it reads like someone from Fox News wrote it. While he frequently assures us that he will be scientific and balanced, he proceeds into a lopsided presentation of a plethora of vague theories most of which seem born from a desparate need for his preferred text to be valid in the form that he desires. Typical tail wagging the dog approach and unimpressive.

No one has denied that the story is ancient, very likely from the 1st century although Papias via Eusebius is typically vague.

While one can certainly speculate on the early history of the pericope it must be held clearly in mind that musings serve little real purpose and hold no impact for the simple reason that the evidence is against it. If the evidence should change then it would be reasonable to change one's mind, but before that happens one merely appears tendentious, not to mention tedious. Enthusiasm and desire are no substitutes for solid evidence.

I guess what I am saying is that Wilker's methods are sound and his presentations adequate (or better). I would be happy to take up his defense here if there are specific attacks on his methods.

Julian

Nazaroo

First of all, let me state clearly that I am in no way challenging Mr. Willker's abilities at chemistry, or his basic grasp of university level mathematics.

Nor did I suggest that Mr. Willker has been particularly dishonest, at least any more dishonest than anyone else in this field.

My criticism re: Willker's online paper on the PA was limited to pointing out that his presentation of the 'textual evidence' was biased in favour of his predetermined position on authenticity, and he had perpetuated a few scientific errors of fact and methodology.

These errors in methodology and fact I carefully outlined in the post on Willker, and you have not addressed them at all. Just review the post, and show me where the error or inaccuracy in my critique of his work actually is.

Willker's Article <-- Click Here.

My Critique of it <-- Click Here.

So lets not exaggerate what I said about Willker's work. He was not accused of fraud, or lying, but biased presentation and faulty methodology, which I carefully documented. This is the very same thing you are now accusing me of, WITHOUT ANY DOCUMENTATION.

Facts first: Then conclusion, please.




This is an inaccurate statement at least two levels.

(1) Willker does not reproduce his calculations or exact details of his methodology. Which was my original complaint.

(2) Willker's techniques are not standard for at least two reasons:

a) there is no current standard for the application of Profile methods in this field. In fact, every textual critic has applied his own peculiar techniques to the problem. Ehrman, who has been previously mentioned here has is own technique too, which is nothing like Willker's.

b) it can't be standardized, because no one has adequately quantified exactly how to even define a 'variation unit', or the actual amount of information contained in a variant. This requires a deep sub-structural analysis of language which has never been done.

Here at the beginning of the age of computer application to code-breaking and other military techniques, it is true that there are some practical methods, but each of these must be applied with varying success and 'experience', as well as intuition.


The issue is specifically this: Willker IN HIS PAPER ON THE PA minimized the significance of these findings, and avoided the entire current discussion on their meaning for the authenticity of the PA. This cannot be denied.

This does not in any way take away from the fact that elsewhere Willker has indeed PROMOTED knowledge of the UMLAUT question on the internet, by posting several good articles by other researchers on this question.

In fact, Willker should be praised for this, as well as for his open-minded attitude.

This however does not excuse the fact that Willker of all people knows full well how important the Umlaut question is to the evaluation of Codex B and the meaning of the 'evidence' concerning the authenticity of the PA.



As a matter of record, I was the first person to praise Willker for popularizing hard to find information about the testimony of Didymus the 4th century Greek commentator who quotes the PA as scripture.

But thanks for confirming what I already have reported.



The 'us and them' attitude is quite inappropriate, and the lame attempt to isolate me and group everyone else with you is childish.

If I have 'failed to see a few things', by all means point them out, instead of engaging in innuendos. Its so boring.



Again, what we have here is the old rhetorical technique of 'damning the source' by slanderous characterization. In the same breath, offer an unsubstantiated 'pop-psychology' analysis of my 'motives', while you promote yourself as an 'authority' who needs to be 'impressed', or it must be crap.

If you are an expert psychologist who wants to be taken seriously for his 'expert' opinion, by all means provide your credentials please.

My question to you is, why don't you stop the high-school debating techniques and get down to the scientific analysis of the evidence?

For instance, why have you not addressed a single criticism of mine about Willker? Why don't you show what I 'failed to see'?

Or better yet, why don't you have a go at the internal evidence I just posted earlier? Let's see how with your uncanny psychological analysis skills, you are able to refute that evidence.



Actually, many have denied it. And they are wrong of course, but so are you on an important historical point of the ongoing history of the debate.

The point is we are not speculating at all, and the 'evidence' is not against the passage.

Opponents of the PA would like everyone to believe that the only 'acceptable' evidence or the most important evidence is the late manuscript evidence, as they choose to interpret it.

Any NEW evidence is then classed as 'speculation'. But this is not scientific enquiry, its just rhetoric and debate.

I have no real interest in 'debate' or 'rhetoric'. I want to actually examine the evidences, independantly of the various parties and 'experts' who already obviously have taken an unscientific ideological stand.

This is the whole point. The evidence HAS changed. It has been changing for a steady 100 years, and at this point, experts like Petersen or M. Robinson can say quite rightly that the evidence is complex and ambiguous at best.

Those who claim otherwise are simply engaging in smokescreens and rhetoric.



Nor is rhetoric any substitute for fresh scientific research, investigation and analysis.

I guess what I am saying is that Willker's 'methods' are not sound, and his presentations have been shown to be biased.

So I guess what you are really saying is that you'd like to give an 'expert' opinion, without any qualifications to do so, while ignoring the evidence already presented against your assertion.


I think he would be better off without your defense, if this entails unsubstantiated rhetoric and a 'methodology' which ignores the evidences that have been presented.

By contrast, Mr. Willker himself is far more honest and humble about his skills and his work:


Mr. Willker's self-defense is refreshing, in comparison to your rhetorical posturing.


It is interesting however, that other engineers and scientists (amateur TCs) on the topic of the suitability of MA as THE TOOL tend toward a lot of other mathematical techniques:

http://www.julian.textcrit.com/?p=5

Steven Avery

This is actually doubly-wrong. Presentations are designed specifically to give the misimpression that the Pericope Adultera was a very late addition, going into the 2nd millennium. The methodology is to have the huge number of early references (manuscripts and early writers and lectionaries) bypassed, instead offering word-parsing and component-units designed to deceive.

Yet even among those who are informed about the references we have such as from Peter Kirby -

"It appears to have been added to the text of John in the third or fourth century."

If folks on this forum can now more accurately begin with the idea that the Pericope Adultera was ancient, 1st century, that would be very good.

Then the opponents have to go through a triple-stage explanation, how it was unwritten by John, yet recognized as scripture by some early, then undone in the supposedly critical handful of socalled "earliest and most reliable" manuscripts (the same ones who could not even get the Johannine place-name Bethesda right) and then the Pericope went ahead and massively overwhelmed the Greek and Latin textlines anyway.

Ockham will be having a lot of head-scratching with itchy whiskers on that one.

Those who look at the vast number of hand-copied manuscripts from a wide geographical range as significant (a la the textual analysis of Professor Robinson) and feel that we can get a good window on the early manuscripts by the comments and inclusions of the early church writers will have our decision on the Pericope rather quickly and decisively. With the compelling internal evidences as a rather fine cake-icing, not to be left out in the rain.

Granted there are basic underlying issues that go beyond the Pericope itself that will affect ones view. The likelihood of additions versus omissions is a major one where the modern textcrit paradigms are very strange, essentially accepted in order to bolster two manuscripts that have lots of scribal corruption that is basically ignored. And ones view of the text being consistent and pure. This is often a factor in lectio difficilior however in this section the most significant internal evidences are not of the lectio difficilior type. More the type (internal consistency and style and text-flow) that simply favor the Pericope.

In fact one can see a parallel between the John 5 case, recently discussed here, where the omission of one or two verses (part or all of 3 or 4 or both) makes the remaining text hard to be original as we have verse 7 in the text. Take out the Pericope and the transition is similarly difficult, perhaps not as severely. The claimants of the additions then have to hand-wave, or ignore, or struggle mightily with, the difficulties in the resulting claimed-as-original text.

Shalom,
Steven Avery

Julian

Sigh. I knew this was going to happen.
Fine, here is what you said:
1) 'all the major MSS groups?' Where does he say this? The whole point of his study is to show the groups. He starts by saying that he is considering variations between NA and Maj., both texts that are easily accessible.

2) 'fails to provide the text' Wrong again. He clearly states that he is using John 1:1 - 5:11 and going to Swanson for 361 variants and twice as many from unnamed sources which I suspect are probably NA27, von Soden and Tischendorf since Swanson's coverage is limited. But who cares since he has given you the text section he is working with and the more variants we have only adds to the accuracy.

3) 'provide an explanation of his methodology' Another miss by you. He clearly states that he is using PCA and he also tells you what implementation he is using (XLSTAT and Excel).

4) 'not a standard technique' Wrong again. PCA is quite standard and well understood. He even explains what it does in layman's terms. He could have described eigen vectors and so forth but I doubt most people would have understood him. Besides, anyone can look up the technique, it being quite standard and all that.

5) 'This is wholly unacceptable in a scientific work making a scientific claim.' Errr, no. He is using a standard technique and explains why and how. Just because you don't understand it doesn't mean it is wrong. He also provides more than enough information for anyone to replicate the results.

There, that's quite enough time to spend on this.

Except he does exactly that. He is using an Excel package. He tells you which one. He doesn't explain the exact details of PCA because he probably doesn't understand them. Understandable since eigen values/vectors are quite complex. I may not know all the inner workings of a car but I do know that it gets me to work. I do understand the inner workings of PCA and his methods are sound. Again, you just don't understand what he is doing.
There are no standards because it is only very recently that people with some mathematical/programming skill have gotten involved in this field. Besides, that is irrelevant in this case since any valid technique can be applied as one wishes. Whether the results are useful is an entirely different issue and not what you commented on here.
Wrong again. There are a number of ways to determine the degree of variation. It is not used because there are not enough technical people in this field yet. I have applied proper measurement techniques on numerous occasions, here is one way: http://www.julian.textcrit.com/?p=11

What does deep sub-structural analysis even mean? Did you just make that up?
Code breaking is not a military technnique. There are lots of practical methods and we apply them with varying success...?!? Which is the standard scientific way to do things. I am beginning to doubt your claims regarding you being a scientist since you display a dismaying lack of understanding of even basic principles.
I deny it. Willker shows you the plots resulting from the PCA. Just because you don't like the mathematical hard evidence doesn't mean that it is biased, skewed, or wrong. Perform a similar (or better) experiment and show that Willker is mistaken. Until you do that you are merely whining. Again, this is basic science stuff. You think an experiment is in error? Show how or do a better one. Simple.
I point them out in the following text. Read before you reply.
Because you wouldn't understand it? Since you have already been confronted with scientific evidence and completely failed to grasp it I see only fruitless endeavors in my future should I pursue this course.
I did address them, you failed to understand them. I have now addressed them again. I suspect that you will fail to understand them once more.
Nope, not interested. I do not understand enough about the pericope to have a valid opinion yet. I was merely interested in correcting your views on scientific techniques and skewed viewpoint.
You have shown no such thing. You have merely been whining and displaying ignorance regarding Willker's work. I have explained to you how to refute his facts. Do so with better facts.

Julian

P.S. I have no time to proof this so I apologize for any errors.

Nazaroo

Fine. I'll just pick out a few of the real zingers, so that others aren't misled by your hand-waving:


(1) Your complete incomprehension of the actual subject at hand:

Incredible. By group here in the context of the PA everyone familiar with the problem understands that we are talking here about von Soden's groups, M1, M2, ...M7. These are groups of MSS which contain the PA, and not only this, but contain a specific version of the PA (i.e., 1 of 7 main versions), a variant Profile that matches the Profile for each group.

This as nothing to do with John 1:1-5:11, since the very nature of the problem is that the text for many of these MSS has been inserted from another source. That is, the text a manuscript has in 7:53-8:11 often has nothing at all to do with the text a manuscript displays elsewhere.

My complaint was Willker's lack of provision of the actual seven variant texts identified by von Soden, but which SHOULD have been refined and associated with known manuscripts. von Soden was not specific in naming many of the MSS he had classified as belonging to each group, so finding representative MSS for each group has been a headache for textual critics ever since.

The only other person to have collated the manuscripts himself is Dr. Maurice Robinson, and he has not published his detailed collations yet.

We have every right to ask what groups (and the MSS assigned to them) that Willker used for his 'PCA' experiment.


Again, your inability to grasp the problem floors us. Of course Principle Component Analysis (PCA) has been refined and discussed in the mathematical literature over the last two decades. It has even been applied with varying success in a variety of fields, from economics to linguistics.

No one is questioning the validity of the mathematical component of the technique(s) and their variations.

But there is no 'standard' way to use PCA in textual criticism or even with texts generally. This is all a new area of application, and many different researchers are trying dozens of ideas out.

There is no concensus among textual critics as to how PCA should be applied to the NT mss, nor is there any concensus on how relevant or valid the results are.

Because of this, it is quite appropriate to request that any researcher in this field with a claim explain his particular application of the mathematical technique and the philosophical and methodological basis for it, as well as present quantified 'error' estimates in the technique. As a chemist Willker knows this.

No chemical paper would be accepted without a proper explanation of the application of any set of methods, or a calculation of error ranges in measurement and calculation.

Astounding. After insisting that Willker HAS explained everything, you admit he hasn't because he 'probably doesn't understand' the details of the way the PCA algorithms are being applied by his Excel software.

Your idea of Willker explaining his 'technique' is that he has announced 'he is using an Excel package', and that this is adequate.


(2) The self-contradiction in your position:

Thanks for conceding my original point, making me right, not wrong.

This was after you just finished 'proving me wrong' because PCA according to you IS A STANDARD TECHNIQUE.

Ouch.


May I quote you on this? In case the CIA need to re-assess a third of their budget.


Do you even know what the topic is? My statement is regarding the UMLAUTS in Codex B, and Willker's minimization of their importance. That is, he fails to discuss them in his PA article.

You apparently deny this criticism because...? "Willker shows you the plots resulting from the PCA".

What?!? For someone who repeatedly calls into question my abilities to 'understand' things, your thinking processes seem pretty ...nebulous.

Wow. For someone giving an 'expert' opinion, you don't seem very well read. Any linguistics analysis text will explain to you that language has surface features of meaning, which fall under the categories of morphology (word formation) and syntax (sentence/paragraph structure).

Language also has 'deep structures' which sometimes overrule surface features, due to extra-contextual conventions (e.g., idiomatic usages of phrases and metaphorical expressions), inter-contextual connections (e.g. prior pronomial referents etc.) and even reader/writer assumptions.

In this case, grammarians and linguists refer to the meaning-units as 'deep-structures' or 'sub-structural' because they do not correspond to the surface morphology or rules of syntax on a one-to-one basis.

Read a few books on linguistics.


Well, since you are not an expert on mathematics, or linguistics, or textual criticism, exactly what kind of 'expert' opinion are you offering, if any?


Good thing you corrected those.


Thanks dad. But do you have to embarrass me in front of everyone?




...

Julian

Wow, just wow. Well, that's the end of my patience. Truly, that last one was one of the more astoundingly bad replies I have ever seen around here, and I've seen a few.
You are doing that just fine without my help.

However, in fairness to the lurkers and other spectators, just so no one will think that Nazaroo actually knows what he is talking about, I will present a few more facts.

Even though I supposedly need to read a few books on linguistics I will still offer these observations. 'Sub-structural,' outside of construction, is a term applied to a particular form of logic. Since logic is a mathematical field, the same as computational linguistics, it follows that there has been research into applying the former onto the latter in the field of formal grammar studies. What Nazaroo was talking about was semantics, something entirely different from substructural logic.

He also said "inter-contextual connections (e.g. prior pronomial referents etc.)" which is normally referred as anaphora (and cataphora if the reference goes forward, collective term is endophora), if I am reading his bizarre terminology correctly.

PCA is a standard technique, how you apply it is not. I trust everybody else understood that rather obvious point.

I also assume that everybody else understood that MSS groupings were irrelevant to Willker's study since they were the whole point of the result. My reference to von Soden was obviously just to clarify another source besides Swanson. Since groups never form part of Willker's input data we must conclude that Nazaroo have not read Willker's study or he would realize this.

I trust it was also obvious that one does not need to know how a car works internally in order to drive yourself to work, for example. The same holds true for PCA: You do not need to know how the internals are working in order to apply it to data.

Despite me even bolding it in a prior post, Nazaroo completely missed it. No surprise there. Just so that others won't make the same mistake, it should be Principal Component Analysis. I bolded it again. Not that it makes a difference.

There is no concensus on how to apply PCA among text crits. Most text crits have never even heard of PCA. Nor does it matter how you apply it as long as it is done in a mathematically defensible manner. A concensus would defeat the purpose. BTW, document-term matrices/vectors are fairly standard and take a PCA (or SVD or whatever) quite readily.

Code breaking is not a military technique although it is used very heavily by both the military and intelligence agencies. One sees code breaking in many other fields like banking, research and, of course, hacking. It has its basis in a field of mathematics called cryptography. I said it was not a military technique, I never said that the military couldn't (or didn't) use it extensively, which they obviously do. My mistake is that I expect my readers to apply some common sense to what I write. Sometimes that backfires.

As for reading some textbooks on linguistics, I guess that the nine (9) filled to bursting 3" binders and various computational linguistic textbooks within easy reach are insufficient.

Just wanted to clear up these things. Someone else can take my place in this discussion if there is one with the stomach and stamina.

Julian

Doug Shaver

No atheist or agnostic of my acquaintance thinks anything of the sort.

If it were proved beyond doubt that every last story in the KJV was in the author's original manuscript of the book it's found in, my opinion of the Bible's authority would not change in the least.

Nazaroo

Thank you for your clinical survey. If you put some numbers and dates to your experience, we could collect it with others and do a statistical survey.

Then you should have no psychological issues, or conflict of interest concerning the authorship of John 7:53-8:11.

We can proceed on a scientific basis to examine all the facts and arguments.

Diogenes the Cynic

It was your own assertion that atheists/agnostics care particularly about discrediting the provenance of the PA in John. It's your burden to prove that's the case, not the burden of others to prove the negative. It's also an assertion you've already backed off of earlier in this thread.
I have no such issues at all with the PA. I don't particularly care if it's authentic to GJohn or not. I am quite capable of observing, though, that all the best scholarship (including most CHRISTIAN scholarship) and manuscript evidence points to the peripcope being an interpolation. This is merely an objective observation. It's not an expression of any a priori assumptions or desires. It's not a pericope whose authenticity would be troublesome to atheists. For that matter, atheists have no reason to care whether ANYTHING in the NT is original to the author or interpolated. I assume that all of the most outlandish claims in the Gospels are original to the authors. So what?

Steven Avery

Hi Julian, Nazaroo and all,

I found this question of the PCA analysis interesting, since I immediately flashed over to the wonderful
"Pronoun Challenged Association"
and the difficulty in getting applicants, PTT's, Pronoun Taste Testers.

And have a couple of comments on the discussion.

First, if anyone wants to see the Wieland Willker material it is at p.21-23 of -
http://www-user.uni-bremen.de/~wie/TCG/TC-John-PA.pdf
At least that is what I know is available, I could not find anything else.

Julian, let me say that you made a good point on the over-techie usage of 'sub-structural analysis' while tripping a bit between the basic issues of standard and non-standard. It looks like you have the better mathematical logistical framework background, while Nazaroo has far more direct familiarity with the manuscript material being discussed. And maybe you have finished the code-breaking discussion, since you are both largely right.

Quote:

Originally Posted by Julian I also assume that everybody else understood that MSS groupings were irrelevant to Willker's study since they were the whole point of the result.

However they appear to be a major part of the input as well. Note that the points on the graphs appear to be individual groups, and Wieland says :

"Unfortunately I have no reliable information as to how many MSS support each group."

And the output is clearly in textual groups, with little distinction
between a group of 3 manuscripts and a group of 250. (This is
a common methodological failing of modern textcrit.)

Now to the basics:
Notice that Wieland makes a totally unwarranted presumption.

"The group that comes nearest to the reconstructed autograph (NA)"

Wow, of course if that is true, that we have a reconstructed autograph, then what is the point to the exercise ? The Pericope would not be able to be original, since it is not the -
'reconstructed autograph'.

How Wieland can use such biased language and have such a hardened a priori view is truly amazing, especially while
maintaining a scholarly and logistical pretension. The target will have to be moved to wherever the arrow landed. The arrows that land properly are the good ones, the others must be deficient. (Shades of the methodology of the 'earliest and most reliable manuscripts'.) And if there aren't any good arrows simply put the target in the garage.

The groups (eg. f1) are complimented if they are nearer the target ! While other groups must be less reliable because they are -

"roughly equally far remote from the original text"

This is not mathematical science, this is propaganda pablum.

Beyond Wieland noting which arrows might be closer to the bullseye
the graphs tell us nothing substantive that I can see. Perhaps you
can help with some sort of explanation of the exercise and its usage
to a discussion of the authenticity of the Pericope Adultera ?

Quote:

Originally Posted by Julian My reference to von Soden was obviously just to clarify another source besides Swanson. Since groups never form part of Willker's input data we must conclude that Nazaroo have not read Willker's study or he would realize this.

Since this does not seem to match up to what he has published online in the PA article, perhaps there is another paper somewhere ? It certainly looks like the whole exercise is one to see which groups can be complimented as closer to the a priori target, the 'original text' and 'reconstructed autograph'. If so, that is not much help for a discussion based on the evidences for an against the Pericope belonging in the Gospel of John.

Quote:

Originally Posted by Julian I trust it was also obvious that one does not need to know how a car works internally in order to drive yourself to work, for example. The same holds true for PCA: You do not need to know how the internals are working in order to apply it to data.... There is no concensus on how to apply PCA among text crits. Most text crits have never even heard of PCA. Nor does it matter how you apply it as long as it is done in a mathematically defensible manner.

However, since Wieland makes no claims to understanding the math who would know if his applications are 'mathematically defensible' ?
(Which might become relevant after we figure out what are his actual claims of significance.)

Here is Wieland offering the analysis to the textcrit forum:

http://groups.yahoo.com/group/textua...sm/message/262
"The group that comes nearest to the reconstructed autograph (NA) is the f1-text. This is a remarkable fact for f1"

The same methodological mishegas.

Shalom,
Steven Avery
http://groups.yahoo.com/group/Messianic_Apologetic