Loading ...
Sorry, an error occurred while loading the content.

Re: [textualcriticism] Re: p73--Byzantine, or just original?

Expand Messages
  • Mike Holmes
    Sorry, a correction: for 2013 read 2012 : whereas Brill printed 2013 in the 2nd ed. of Ehrman and Holmes, the book actually appeared in November 2012.
    Message 1 of 23 , Mar 20, 2013
    • 0 Attachment
      Sorry, a correction: for "2013" read "2012": whereas Brill printed '2013' in the 2nd ed. of Ehrman and Holmes, the book actually appeared in November 2012.
      Mike

      On Wed, Mar 20, 2013 at 11:09 PM, Mike Holmes <holmic@...> wrote:
      Hello, Steven,


      No, the date is not 1995. The 1st edition of Der Text de Neuen Testaments was published, as you indicated, in 1981; the ET was published in 1987, and the ET of the 2nd edition was published in 1989.  The 1995 date of Googlebooks is misleading; it likely is the date of a reprint, not the year of publication. So the five-level classification created by the Alands represents their views in the 1980’s.


      The essay in which Barbara Aland and Kurt Wachtel characterize the five-level system as “obsolete” is from 2013. Here is the context:

      “All of the manuscripts have been collated in the test passages at the Institute for Textual Research. The results were summarized by assigning the manuscripts to so-called categories.19 [note 19: See K. and B. Aland, Text of the New Testament, (2nd ed., 1989) 159–163 and 332–337.] Now that all test-passage collations are published, this preliminary classification has become obsolete. The specialist is given full access to the collations of the test passages themselves.” [B. Aland and K. Wachtel, "The Greek Minuscules, etc.", in Ehrman and Holmes, The Text of the NT, 2nd ed.  (Brill, 2013],  p. 77).


      Thanks,
      Mike




      On Wed, Mar 20, 2013 at 9:10 PM, Steven Avery <stevenavery@...> wrote:
       

      Hi,

      Daniel Buck

      Category V, by definition, contains "Manuscripts with a purely or predominantly Byzantine text." A Byzantine (fka 'Antiochian') text, by definition, contains readings without any support in 'Neutral' or 'Western' manuscripts.

      Not from Aland:

      The text of the New Testament: an introduction to the critical editions and to the theory and practice of modern textual criticism, 1995
      Kurt and Barbara Aland
      http://books.google.com/books?id=2pYDsAhUOxAC&pg=PA14
      Actually there is no such thing as a "neutral" text of the New Testament. (p. 14)

      If you change it to Alexandrian, then you would be talking about that rarity:

      "distinctively Byzantine" or
      "distinctively Syrian" readings.

      However this involves some convoluted writing from Westcott and Hort than really has never been unpacked into a sensible presentation. And was particularly mangled by Kenyon. And I do not think Aland uses the phrase in any form, although he likes the weird Byzantine Imperial text type, a crafty method of poisoning the well of textual analysis.  However, since Aland does not consider these 1200+ cursives as having any "significance for textual criticism" (p. 163) one can understand his Hortian approach to phrases and constructs that support his perceived insignificance.

      In fact, I would be interested in the full section where Aland talks of "...together with omitting the witness of uncials with a purely Byzantine text " (p. 36).  It sounds like he is so adverse to Byzantine readings that he would like to omit even having to deal with Byzantine
      uncials!

      As for P73,  in 1995, it was described as "still unedited".

      The Text of the New Testament in Contemporary Research Essays on the Status Quaestionis (1995)
      The Papyrus Manuscripts of the New Testament
      Eldon Jay Epp
      http://books.google.com/books?id=TxhqBeeAqg8C&pg=PA5
       (A fifth Bodmer papyrus. P73. also of the 7th century, contains only three verses of Matthew and is still unedited.)

      Mike Holmes

      Indeed, to add to Dirk's comment, Barbara Aland and Klaus Wachtel now consider that particular classification system to be "obsolete" (Aland and Wachtel, "The Greek Minuscules, etc.", in Ehrman and Holmes, The Text of the NT, 2nd ed., p. 77).

      Could you supply the whole quote. The year is 1995, right?

      In the Text of the NT, 1995, the translation of Der Text de Neuen Testaments, Aland made a whole presentation of his classification system as a supposed improvement.  If they already considered their 1981 system as obsolete, then why keep republishing?

      Shalom,
      Steven Avery
      Bayside, NY

      Daniel Buck  wrote:
      > The wikipedia article for p73 states,  http://en.wikipedia.org/wiki/Papyrus_73 > "The Greek text of this codex is a representative of the Byzantine text-type, but text is too brief for certainty. Aland placed it in Category V."  The Greek text of p73 in fact consists of about ten clearly identifiable letters and another seven that appear to match what would be expected in the adjacent text of Matthew 25:43-44 and 26:2-3.  Although there are textual variants just on either side of these excerpts (there's an h.t. later in v. 3 that would show up if we had even one more line fragment of identifiable text), there are none in the running text that could be determined even by letter-count. So there is absolutely no evidence of any distinctly Byzantine reading in this seventh-century papyrus. By assigning it to Category V, Aland was virtually admitting that the Byzantine text was original.

      Jeff Cate, Just curious... why would assigning P73 to Category V be an indication that Byz was "original"? The Alands date P73 to the VII century. They date some Cat V uncials earlier than that (026, 061, 022, 023, 024, 027, 042, 043, 064, 065, 0246, 0253, 0265?)... but none before the V century.

      Daniel Buck
      Inasmuch as the text of p73 contains no such readings, it is not Byzantine. Call it Byzantine, and you are identifying Byzantine with the Ausgangstext.
       



    • yennifmit
      Hi James, I don t have that Editing the Bible article. What the group should be called is a matter of discretion. I use the group medoid though that is not
      Message 2 of 23 , Mar 21, 2013
      • 0 Attachment
        Hi James,

        I don't have that "Editing the Bible" article. What the group should be called is a matter of discretion. I use the group medoid though that is not an ideal solution as the medoid can change if the composition of the data set changes.

        I don't know much about the method used to produce the plot in the SWH article. I imagine that changing the set of variation units used to do the analysis would change the result somewhat.

        Best,

        Tim Finney

        --- In textualcriticism@yahoogroups.com, "Vox Verax" <james.snapp@...> wrote:
        >
        > Tim,
        >
        > My brief comparison of 2816 to 2412 in James leads me to suspect that 2816 is a fellow-member of the same group. But I have a full plate already, what with Easter approaching, and I don't have the time to partially collate the General Epistles in 2816. (I might go back and sift through James more thoroughly, though.) Maybe someone else here would like this task??
        >
        > I wonder where 2816 would fit in Diagram 3 in Klaus Wachtel's chapter on CBGM (on page 125 of "Editing the Bible").
        >
        > What should this group be called? Wachtel called it the Harclensis Group (HG). But in the article at http://rosetta.reltech.org/TC/v07/SWH2002/index.html the authors acknowledge (however cautiously) that "the form of text preserved in the Greek witnesses is prior to that translated into Syriac in 616 A.D." Wouldn't it be best to name a group after its earliest or most distinct member?
        >
        > Also: (I'm not sure if I quite grasped what I think is the basis for this question, but,) does anything in the Circular Maximum Parsimony diagram, or in a diagram like Diagram #3, change when all members are arranged using only the 495 passages in which the text produced via a reverse-translation from the Harklean Syriac consists of an unambiguous variant, instead of the 3,046 passages that Wachtel mentioned?
        >
        > Yours in Christ,
        >
        > James Snapp, Jr.
        >
        >
        > --- In textualcriticism@yahoogroups.com, "yennifmit" <tfinney@> wrote:
        > >
        > > Hi James,
        > >
        > > Unfortunately, I don't have any data on 2816 so can't say which group it belongs to. It may be in the same textual complex as 2138 and company. (If you are really keen, you could add a row for 2816 to one of the UBS4 data matrices for the General Letters available at my Views site. I would then redo the analysis and be able to say where 2816 fits.)
        > >
        > > Concerning what to call that complex, I prefer to use medoids to name groups. (The medoid of a group is its most central member.) Just what the medoid is depends on which other texts are there -- adding or subtracting even one can cause the medoid to change. That means that one needs to look at group members to find corresponding groups when comparing across data sets. As it happens, some partitions of the INTF-General and UBS4 data sets for the General Letters shown at the Views site have 2138 as the medoid of that complex.
        > >
        > > http://www.tfinney.net/Views/index.xhtml
        > >
        > > I don't know how early the 2138 complex is. Its roots may be in pre-Chalcedon Eastern Christianity but its link with the Harclean Syriac suggests a later date for the form found in 2138 and co.
        > >
        > > Best,
        > >
        > > Tim Finney
        > >
        > > --- In textualcriticism@yahoogroups.com, "Vox Verax" <james.snapp@> wrote:
        > > >
        > > > Tim Finney,
        > > >
        > > > I don't see 2816 listed in any of the groups. Based on a brief comparison of its text in James to some members of the 2138-group, I reckon that 2816 should be considered a member of Group-2138 � and somewhat less Byzantine than 2412 (images of which are online at the University of Chicago's Goodspeed Collection, where 2412 is called the Ira Maurice Price Praxapostolos -- look, btw, at how 5:20 is mangled in 2412, at Image #264! And no correction?!). Robert Waltz has some interesting things to say about Group-2138, including, "Family 2138 is an early text-type."
        > > >
        > > > Does anyone think HMWN in 5:8 has a shot at being original?
        > > >
        > > > Yours in Christ,
        > > >
        > > > James Snapp, Jr.
        > > >
        > >
        >
      • yennifmit
        Hi Jeff, Sorry to chime in late. (I tried yesterday but there was some kind of technical glitch.) You said there has to be some kind of way to associate MSS
        Message 3 of 23 , Mar 21, 2013
        • 0 Attachment
          Hi Jeff,

          Sorry to chime in late. (I tried yesterday but there was some kind of technical glitch.)

          You said "there has to be some kind of way to associate MSS with important witnesses such as 03 or 05 or Byz-Maj." Well, there is -- multivariate analysis (MVA). It lets you see where witnesses in a data set lie relative to each other. We've been doing things the wrong way for a long time, only comparing with a select few (quantitative analysis). Also, Colwell's grouping criteria (70% agreement; 10% gap) have serious problems.

          Better representations of the actual group structure emerge when multivariate analysis is used. As an example, here are two representations produced by multivariate analysis of data from the UBS Greek New Testament (4th ed.) apparatus for the Gospel of Mark:

          http://www.tfinney.net/Views/cmds/Mark-UBS4.15.SMD.gif

          http://www.tfinney.net/Views/dc/Mark-UBS4.15.SMD.png

          The first item is the result of classical multidimensional scaling (CMDS) using (only) three dimensions; the second is the result of a technique called divisive clustering (DC). Other techniques can be used as well. My favourite is partitioning around medoids (PAM) because it indicates preferable numbers of groups. Some PAM results for the Gospel of Mark are available at my Views site:

          http://www.tfinney.net/Views/index.xhtml

          (An interesting aspect of the MVA results for Mark is that Streeter's Eastern text -- P45, W, Theta, Family 1, Family 13, 28, 565, Sinaitic Syriac, Armenian, Georgian, Origen -- stands out as separate from the Alexandrian, Old Latin, Byzantine, and Latin Vulgate clusters. There are many other interesting things about these results as well. One is that there is a group structure. Another is that the Alexandrian and Old Latin clusters are about as "wild" as each other if their respective diameters serve as a guide.)

          Concerning P73, it could only be located relative to the others if its text were defined at enough variation sites. I set a minimum number of 15 defined variation sites for a witness to be included in the analysis results. Even fifteen is a bare minimum: the sampling error associated with a distance estimate based on this many points of comparison is about 50%.

          Best,

          Tim Finney

          --- In textualcriticism@yahoogroups.com, "jjcate" <jjcate@...> wrote:
          >
          > Yes, and others have questioned the whole Cat system too. And the text type system (as defined by Colwell) has its critics. But there has to be some kind of way to associate MSS with important witnesses such as 03 or 05 or Byz-Maj. I think you're right that in this instance, labeling P73 can be misleading... so maybe it would've been best to leave it uncategorized. But even though it was labeled Cat 5, I don't think this indicates anything about Byz being initial or original. But maybe some day more of p73 can be found & identified and its affinities can be more clearly demonstrated. All the best,
          > --Jeff
          >
          > --- In textualcriticism@yahoogroups.com, Daniel Buck <bucksburg@> wrote:
          > >
          > > I would propose that p73's inconvenient existence shows that something is fatally wrong with the whole Category system.
          > >
          > >
          > > Daniel Buck
          > >
          > > ________________________________
          > > From: jjcate <jjcate@>
          > > To: textualcriticism@yahoogroups.com
          > > Sent: Wednesday, March 20, 2013 3:43 PM
          > > Subject: [textualcriticism] Re: p73--Byzantine, or just original?
          > >
          > >
          > >  
          > > Ok, but p73 is certainly not Cat 1, 2, or 4. So the only other options are Cat 3 or 5, neither of which are good fits. Looking at how the Alands described Cat 3 on pp. 335-36 (Engl. trans.), does p73 have "significant readings from other sources not yet identified" (ala W or "Caesarean" mss)?
          > >
          > > I suspect (and this is speculation on my part) that the VII date is what caused them to label p73 as Cat 5, with the caveat "text too brief for certainty." In other words, being papyrus, it could be easy to falsely assume p73 was Cat 1, when it's actually late (for papyrus) and there's no indication that it is Cat 1. But categorizing a VII century MS (albeit papyrus) as Cat 5 is no indication that Cat 5/Byz should be considered default for manuscripts prior to the fifth century since the Alands never labeled any pre-V MS as Cat 5.
          > >
          > > Hope this is helpful,
          > > --Jeff
          > >
          > > --- In textualcriticism@yahoogroups.com, Daniel Buck <bucksburg@> wrote:
          > > >
          > > > Category V, by definition, contains "Manuscripts with a purely or predominantly Byzantine text."
          > > > A Byzantine (fka 'Antiochian') text, by definition, contains readings without any support in 'Neutral' or 'Western' manuscripts. Inasmuch as the text of p73 contains no such readings, it is not Byzantine. Call it Byzantine, and you are identifying Byzantine with the Ausgangstext.
          > > >  
          > > > Daniel Buck 
          > > >
          > > >
          > > > ________________________________
          > > > From: jjcate <jjcate@>
          > > > To: textualcriticism@yahoogroups.com
          > > > Sent: Wednesday, March 20, 2013 1:44 PM
          > > > Subject: [textualcriticism] Re: p73--Byzantine, or just original?
          > > >
          > > >
          > > >  
          > > > Just curious... why would assigning P73 to Category V be an indication that Byz was "original"? The Alands date P73 to the VII century. They date some Cat V uncials earlier than that (026, 061, 022, 023, 024, 027, 042, 043, 064, 065, 0246, 0253, 0265?)... but none before the V century.
          > > > --Jeff Cate,
          > > > Riverside, CA
          > > >
          > > > --- In textualcriticism@yahoogroups.com, Daniel Buck <bucksburg@> wrote:
          > > > >
          > > > > The wikipedia article for p73 states,  http://en.wikipedia.org/wiki/Papyrus_73
          > > > > "The Greek text of this codex is a representative of the Byzantine text-type, but text is too brief for certainty. Aland placed it in Category V."
          > > > >
          > > > > The Greek text of p73 in fact consists of about ten clearly identifiable letters and another seven that appear to match what would be expected in the adjacent text of Matthew 25:43-44 and 26:2-3. 
          > > > >
          > > > > Although there are textual variants just on either side of these excerpts (there's an h.t. later in v. 3 that would show up if we had even one more line fragment of identifiable text), there are none in the running text that could be determined even by letter-count. So there is absolutely no evidence of any distinctly Byzantine reading in this seventh-century papyrus. 
          > > > >
          > > > > By assigning it to Category V, Aland was virtually admitting that the Byzantine text was original.
          > > > >
          > > > > Daniel Buck
          > > > >
          > > >
          > >
          >
        • clearbrush
          Dear Dr. Finney, you wrote : (An interesting aspect of the MVA results for Mark is that Streeter s Eastern text -- P45, W, Theta, Family 1, Family 13, 28, 565,
          Message 4 of 23 , Mar 24, 2013
          • 0 Attachment

            Dear Dr. Finney,

            you wrote :

            (An interesting aspect of the MVA results for Mark is that Streeter's Eastern text -- P45, W, Theta, Family 1, Family 13, 28, 565, Sinaitic Syriac, Armenian, Georgian, Origen -- stands out as separate from the Alexandrian, Old Latin, Byzantine, and Latin Vulgate clusters.

            There are many other interesting things about these results as well.

            One is that there is a group structure.

            Another is that the Alexandrian and Old Latin clusters are about as "wild"as each other if their respective diameters serve as a guide.)

            My question : could this data be interpreted as pointing to very ancient centers where prestigious and-or exemplary Mark texts resided in and the groups (or clusters) emanated from; e.g. Rome, Alexandria, and Antioch-Smyrna ?

            Also curious about the situation with Luke and John.. and especially - the comparison with Hebrews..

            All the best, George Eller


            --- In textualcriticism@yahoogroups.com, "yennifmit" <tfinney@...> wrote:
            >
            > Hi Jeff,
            >
            > Sorry to chime in late. (I tried yesterday but there was some kind of technical glitch.)
            >
            > You said "there has to be some kind of way to associate MSS with important witnesses such as 03 or 05 or Byz-Maj." Well, there is -- multivariate analysis (MVA). It lets you see where witnesses in a data set lie relative to each other. We've been doing things the wrong way for a long time, only comparing with a select few (quantitative analysis). Also, Colwell's grouping criteria (70% agreement; 10% gap) have serious problems.
            >
            > Better representations of the actual group structure emerge when multivariate analysis is used. As an example, here are two representations produced by multivariate analysis of data from the UBS Greek New Testament (4th ed.) apparatus for the Gospel of Mark:
            >
            > http://www.tfinney.net/Views/cmds/Mark-UBS4.15.SMD.gif
            >
            > http://www.tfinney.net/Views/dc/Mark-UBS4.15.SMD.png
            >
            > The first item is the result of classical multidimensional scaling (CMDS) using (only) three dimensions; the second is the result of a technique called divisive clustering (DC). Other techniques can be used as well. My favourite is partitioning around medoids (PAM) because it indicates preferable numbers of groups. Some PAM results for the Gospel of Mark are available at my Views site:
            >
            > http://www.tfinney.net/Views/index.xhtml
            >
            > (An interesting aspect of the MVA results for Mark is that Streeter's Eastern text -- P45, W, Theta, Family 1, Family 13, 28, 565, Sinaitic Syriac, Armenian, Georgian, Origen -- stands out as separate from the Alexandrian, Old Latin, Byzantine, and Latin Vulgate clusters. There are many other interesting things about these results as well. One is that there is a group structure. Another is that the Alexandrian and Old Latin clusters are about as "wild" as each other if their respective diameters serve as a guide.)
            >
            > Concerning P73, it could only be located relative to the others if its text were defined at enough variation sites. I set a minimum number of 15 defined variation sites for a witness to be included in the analysis results. Even fifteen is a bare minimum: the sampling error associated with a distance estimate based on this many points of comparison is about 50%.
            >
            > Best,
            >
            > Tim Finney
            >
            > --- In textualcriticism@yahoogroups.com, "jjcate" jjcate@ wrote:
            > >
            > > Yes, and others have questioned the whole Cat system too. And the text type system (as defined by Colwell) has its critics. But there has to be some kind of way to associate MSS with important witnesses such as 03 or 05 or Byz-Maj. I think you're right that in this instance, labeling P73 can be misleading... so maybe it would've been best to leave it uncategorized. But even though it was labeled Cat 5, I don't think this indicates anything about Byz being initial or original. But maybe some day more of p73 can be found & identified and its affinities can be more clearly demonstrated. All the best,
            > > --Jeff
            > >
            > > --- In textualcriticism@yahoogroups.com, Daniel Buck <bucksburg@> wrote:
            > > >
            > > > I would propose that p73's inconvenient existence shows that something is fatally wrong with the whole Category system.
            > > >
            > > >
            > > > Daniel Buck
            > > >
            > > > ________________________________
            > > > From: jjcate <jjcate@>
            > > > To: textualcriticism@yahoogroups.com
            > > > Sent: Wednesday, March 20, 2013 3:43 PM
            > > > Subject: [textualcriticism] Re: p73--Byzantine, or just original?
            > > >
            > > >
            > > >  
            > > > Ok, but p73 is certainly not Cat 1, 2, or 4. So the only other options are Cat 3 or 5, neither of which are good fits. Looking at how the Alands described Cat 3 on pp. 335-36 (Engl. trans.), does p73 have "significant readings from other sources not yet identified" (ala W or "Caesarean" mss)?
            > > >
            > > > I suspect (and this is speculation on my part) that the VII date is what caused them to label p73 as Cat 5, with the caveat "text too brief for certainty." In other words, being papyrus, it could be easy to falsely assume p73 was Cat 1, when it's actually late (for papyrus) and there's no indication that it is Cat 1. But categorizing a VII century MS (albeit papyrus) as Cat 5 is no indication that Cat 5/Byz should be considered default for manuscripts prior to the fifth century since the Alands never labeled any pre-V MS as Cat 5.
            > > >
            > > > Hope this is helpful,
            > > > --Jeff
            > > >
            > > > --- In textualcriticism@yahoogroups.com, Daniel Buck <bucksburg@> wrote:
            > > > >
            > > > > Category V, by definition, contains "Manuscripts with a purely or predominantly Byzantine text."
            > > > > A Byzantine (fka 'Antiochian') text, by definition, contains readings without any support in 'Neutral' or 'Western' manuscripts. Inasmuch as the text of p73 contains no such readings, it is not Byzantine. Call it Byzantine, and you are identifying Byzantine with the Ausgangstext.
            > > > >  
            > > > > Daniel Buck 
            > > > >
            > > > >
            > > > > ________________________________
            > > > > From: jjcate <jjcate@>
            > > > > To: textualcriticism@yahoogroups.com
            > > > > Sent: Wednesday, March 20, 2013 1:44 PM
            > > > > Subject: [textualcriticism] Re: p73--Byzantine, or just original?
            > > > >
            > > > >
            > > > >  
            > > > > Just curious... why would assigning P73 to Category V be an indication that Byz was "original"? The Alands date P73 to the VII century. They date some Cat V uncials earlier than that (026, 061, 022, 023, 024, 027, 042, 043, 064, 065, 0246, 0253, 0265?)... but none before the V century.
            > > > > --Jeff Cate,
            > > > > Riverside, CA
            > > > >
            > > > > --- In textualcriticism@yahoogroups.com, Daniel Buck <bucksburg@> wrote:
            > > > > >
            > > > > > The wikipedia article for p73 states,  http://en.wikipedia.org/wiki/Papyrus_73
            > > > > > "The Greek text of this codex is a representative of the Byzantine text-type, but text is too brief for certainty. Aland placed it in Category V."
            > > > > >
            > > > > > The Greek text of p73 in fact consists of about ten clearly identifiable letters and another seven that appear to match what would be expected in the adjacent text of Matthew 25:43-44 and 26:2-3. 
            > > > > >
            > > > > > Although there are textual variants just on either side of these excerpts (there's an h.t. later in v. 3 that would show up if we had even one more line fragment of identifiable text), there are none in the running text that could be determined even by letter-count. So there is absolutely no evidence of any distinctly Byzantine reading in this seventh-century papyrus. 
            > > > > >
            > > > > > By assigning it to Category V, Aland was virtually admitting that the Byzantine text was original.
            > > > > >
            > > > > > Daniel Buck
            > > > > >
            > > > >
            > > >
            > >
            >
          • yennifmit
            Hi George, It s good to hear from you! You wrote, could this data be interpreted as pointing to very ancient centers where prestigious and-or exemplary Mark
            Message 5 of 23 , Mar 25, 2013
            • 0 Attachment
              Hi George,

              It's good to hear from you!

              You wrote, "could this data be interpreted as pointing to very ancient
              centers where prestigious and-or exemplary Mark texts resided in and the groups (or clusters) emanated from; e.g. Rome, Alexandria, and
              Antioch-Smyrna?"

              I would say yes to ancient centres but no to always using prestigious exemplars. I imagine that every Christian community would try to get its own copies of the text (comprised of various combinations of the Gospels, Paul's Letters, Acts + General Letters, Revelation). Thinking about how the text propagated, it is not unreasonable to expect that most copies would be made where there were the most Christian communities. If a significant proportion of the Christian population resided in and around the provincial capitals then one would expect much manuscript copying to happen in and around Rome, Ephesus, Syrian Antioch, Caesarea, and Alexandria. (BTW, Asia Minor was the number one Christian population centre in the second century according to Harnack.)

              I don't know whether there was much effort to reference prestigious exemplars. I'm more inclined to think that copyists would get the nearest convenient exemplar -- the principle of least effort applies.

              I do think that the scenario of population centres being centres of manuscript production combined with least effort in securing exemplars would lead to development of local texts. My PhD research found that multivariate analysis of (1) textual and (2) spelling variations produced similar results: both had clusters of texts, with the same manuscripts usually being in the same clusters for both kinds of variation. A theory of local texts can account for this result: scribes would spell according to local custom (and accent!); they would also tend to use the textual variants that were most popular where they lived. (There were viral variants too -- ones that did "Levy flights" to leap from region to region.)

              So, which texts should we associate with which regions? (I have an opinion on this.) And the $64,000 question: what happened to the ancient text of Asia Minor?

              As for books besides Mark, there are now multivariate analysis results for many of them at my Views site:

              http://www.tfinney.net/Views/index.xhtml

              Considering the classical multidimensional scaling (CMDS) results, there is often a similar structure. One has to be careful to compare the same kinds of data. The results based on INTF data sets cover a lot of Greek manuscripts but do not include versions of patristic evidence. The UBS data sets include more classes of evidence (Greek MSS, versional, patristic) but do not cover as many variation sites.

              As examples of similarity, compare the INTF map results for a few of the General Letters:

              http://www.tfinney.net/Views/cmds/Peter-A-INTF-General.15.SMD.gif

              http://www.tfinney.net/Views/cmds/Peter-B-INTF-General.15.SMD.gif

              http://www.tfinney.net/Views/cmds/John-A-INTF-General.15.SMD.gif

              CMDS maps for the Gospels have some broad similarities to each other but significant differences as well -- see e.g. the maps derived from UBS2 data originally encoded by Maurice Robinson.

              Group structure is apparent in all of the data sets. There is often a small number of major clusters which may well be associated with early Christian population centres.

              All of these results are based on the efforts of people who have worked to collate and encode the data. I'm especially grateful to Richard Mallett who entered a number of the data matrices.

              Best,

              Tim Finney

              --- In textualcriticism@yahoogroups.com, "clearbrush" <clearbrush@...> wrote:
              >
              >
              > Dear Dr. Finney,
              >
              > you wrote :
              >
              > (An interesting aspect of the MVA results for Mark is that Streeter's
              > Eastern text -- P45, W, Theta, Family 1, Family 13, 28, 565, Sinaitic
              > Syriac, Armenian, Georgian, Origen -- stands out as separate from the
              > Alexandrian, Old Latin, Byzantine, and Latin Vulgate clusters.
              >
              > There are many other interesting things about these results as well.
              >
              > One is that there is a group structure.
              >
              > Another is that the Alexandrian and Old Latin clusters are about as
              > "wild"as each other if their respective diameters serve as a guide.)
              >
              > My question : could this data be interpreted as pointing to very ancient
              > centers where prestigious and-or exemplary Mark texts resided in and the
              > groups (or clusters) emanated from; e.g. Rome, Alexandria, and
              > Antioch-Smyrna ?
              >
              > Also curious about the situation with Luke and John.. and especially -
              > the comparison with Hebrews..
              >
              >
              > All the best, George Eller
              >
              > --- In textualcriticism@yahoogroups.com, "yennifmit" <tfinney@>
              > wrote:
              > >
              > > Hi Jeff,
              > >
              > > Sorry to chime in late. (I tried yesterday but there was some kind of
              > technical glitch.)
              > >
              > > You said "there has to be some kind of way to associate MSS with
              > important witnesses such as 03 or 05 or Byz-Maj." Well, there is --
              > multivariate analysis (MVA). It lets you see where witnesses in a data
              > set lie relative to each other. We've been doing things the wrong way
              > for a long time, only comparing with a select few (quantitative
              > analysis). Also, Colwell's grouping criteria (70% agreement; 10% gap)
              > have serious problems.
              > >
              > > Better representations of the actual group structure emerge when
              > multivariate analysis is used. As an example, here are two
              > representations produced by multivariate analysis of data from the UBS
              > Greek New Testament (4th ed.) apparatus for the Gospel of Mark:
              > >
              > > http://www.tfinney.net/Views/cmds/Mark-UBS4.15.SMD.gif
              > >
              > > http://www.tfinney.net/Views/dc/Mark-UBS4.15.SMD.png
              > >
              > > The first item is the result of classical multidimensional scaling
              > (CMDS) using (only) three dimensions; the second is the result of a
              > technique called divisive clustering (DC). Other techniques can be used
              > as well. My favourite is partitioning around medoids (PAM) because it
              > indicates preferable numbers of groups. Some PAM results for the Gospel
              > of Mark are available at my Views site:
              > >
              > > http://www.tfinney.net/Views/index.xhtml
              > >
              > > (An interesting aspect of the MVA results for Mark is that Streeter's
              > Eastern text -- P45, W, Theta, Family 1, Family 13, 28, 565, Sinaitic
              > Syriac, Armenian, Georgian, Origen -- stands out as separate from the
              > Alexandrian, Old Latin, Byzantine, and Latin Vulgate clusters. There are
              > many other interesting things about these results as well. One is that
              > there is a group structure. Another is that the Alexandrian and Old
              > Latin clusters are about as "wild" as each other if their respective
              > diameters serve as a guide.)
              > >
              > > Concerning P73, it could only be located relative to the others if its
              > text were defined at enough variation sites. I set a minimum number of
              > 15 defined variation sites for a witness to be included in the analysis
              > results. Even fifteen is a bare minimum: the sampling error associated
              > with a distance estimate based on this many points of comparison is
              > about 50%.
              > >
              > > Best,
              > >
              > > Tim Finney
              > >
              > > --- In textualcriticism@yahoogroups.com, "jjcate" jjcate@ wrote:
              > > >
              > > > Yes, and others have questioned the whole Cat system too. And the
              > text type system (as defined by Colwell) has its critics. But there has
              > to be some kind of way to associate MSS with important witnesses such as
              > 03 or 05 or Byz-Maj. I think you're right that in this instance,
              > labeling P73 can be misleading... so maybe it would've been best to
              > leave it uncategorized. But even though it was labeled Cat 5, I don't
              > think this indicates anything about Byz being initial or original. But
              > maybe some day more of p73 can be found & identified and its affinities
              > can be more clearly demonstrated. All the best,
              > > > --Jeff
              > > >
              > > > --- In textualcriticism@yahoogroups.com, Daniel Buck <bucksburg@>
              > wrote:
              > > > >
              > > > > I would propose that p73's inconvenient existence shows that
              > something is fatally wrong with the whole Category system.
              > > > >
              > > > >
              > > > > Daniel Buck
              > > > >
              > > > > ________________________________
              > > > > From: jjcate <jjcate@>
              > > > > To: textualcriticism@yahoogroups.com
              > > > > Sent: Wednesday, March 20, 2013 3:43 PM
              > > > > Subject: [textualcriticism] Re: p73--Byzantine, or just original?
              > > > >
              > > > >
              > > > > �
              > > > > Ok, but p73 is certainly not Cat 1, 2, or 4. So the only other
              > options are Cat 3 or 5, neither of which are good fits. Looking at how
              > the Alands described Cat 3 on pp. 335-36 (Engl. trans.), does p73 have
              > "significant readings from other sources not yet identified" (ala W or
              > "Caesarean" mss)?
              > > > >
              > > > > I suspect (and this is speculation on my part) that the VII date
              > is what caused them to label p73 as Cat 5, with the caveat "text too
              > brief for certainty." In other words, being papyrus, it could be easy to
              > falsely assume p73 was Cat 1, when it's actually late (for papyrus) and
              > there's no indication that it is Cat 1. But categorizing a VII century
              > MS (albeit papyrus) as Cat 5 is no indication that Cat 5/Byz should be
              > considered default for manuscripts prior to the fifth century since the
              > Alands never labeled any pre-V MS as Cat 5.
              > > > >
              > > > > Hope this is helpful,
              > > > > --Jeff
              > > > >
              > > > > --- In textualcriticism@yahoogroups.com, Daniel Buck <bucksburg@>
              > wrote:
              > > > > >
              > > > > > Category V, by definition, contains� "Manuscripts with a
              > purely or predominantly Byzantine text."
              > > > > > A Byzantine (fka 'Antiochian') text, by definition, contains
              > readings without any support in 'Neutral' or 'Western' manuscripts.
              > Inasmuch as the text of p73 contains no such readings, it is not
              > Byzantine. Call it Byzantine, and you are identifying Byzantine with the
              > Ausgangstext.
              > > > > > �
              > > > > > Daniel Buck�
              > > > > >
              > > > > >
              > > > > > ________________________________
              > > > > > From: jjcate <jjcate@>
              > > > > > To: textualcriticism@yahoogroups.com
              > > > > > Sent: Wednesday, March 20, 2013 1:44 PM
              > > > > > Subject: [textualcriticism] Re: p73--Byzantine, or just
              > original?
              > > > > >
              > > > > >
              > > > > > �
              > > > > > Just curious... why would assigning P73 to Category V be an
              > indication that Byz was "original"? The Alands date P73 to the VII
              > century. They date some Cat V uncials earlier than that (026, 061, 022,
              > 023, 024, 027, 042, 043, 064, 065, 0246, 0253, 0265?)... but none before
              > the V century.
              > > > > > --Jeff Cate,
              > > > > > Riverside, CA
              > > > > >
              > > > > > --- In textualcriticism@yahoogroups.com, Daniel Buck
              > <bucksburg@> wrote:
              > > > > > >
              > > > > > > The wikipedia article for p73 states, �
              > http://en.wikipedia.org/wiki/Papyrus_73
              > > > > > > "The Greek text of this codex is a representative of
              > the� Byzantine text-type, but text is too brief for
              > certainty.� Aland� placed it in� Category V."
              > > > > > >
              > > > > > > The Greek text of p73 in fact consists of about ten clearly
              > identifiable letters and another seven that appear to match what would
              > be expected in the adjacent text of Matthew� 25:43-44 and�
              > 26:2-3.�
              > > > > > >
              > > > > > > Although there are textual variants just on either side of
              > these� excerpts (there's an h.t. later in v. 3 that would show up
              > if we had even one more line fragment of identifiable text), there are
              > none in the running text that could be� determined� even by
              > letter-count. So there is absolutely no evidence of any distinctly
              > Byzantine reading in this seventh-century papyrus.�
              > > > > > >
              > > > > > > By assigning it to Category V, Aland was virtually admitting
              > that the Byzantine text was original.
              > > > > > >
              > > > > > > Daniel Buck
              > > > > > >
              > > > > >
              > > > >
              > > >
              > >
              >
            Your message has been successfully submitted and would be delivered to recipients shortly.