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

[Synoptic-L] Difficulty in interpretation

Expand Messages
  • David Gentile
    As I look at these results more, they are proving difficult to interpret. Mark Goodacre seems to have been correct, in that the impact of the editors is hard
    Message 1 of 10 , Dec 4, 2001
      As I look at these results more, they are proving difficult to interpret.
      Mark Goodacre seems to have been correct, in that the impact of the editors
      is hard to separate from the source. I think 222 may look a lot like 202
      simply because they both passed the approval of both Luke and Matthew. It
      part 122 looks like 112 because Matthew did not choose either. It's
      instructive to add some of the cases together, to see what happens,
      particularly in relation to 222. Staying only with the triple tradition:
      x22 and x21 are negative.
      x22 and x12 are negative.
      22x and 21x are negative.
      22x and 12x are negative.
      222 is just kind of floating in the middle. It's like 221 because Matthew
      choose both, and unlike it because Luke choose one and not the other. The
      effects seem to cancel, leaving it uncorrelated. It does reveal things about
      the relationships, but it might be hard to say what, in most cases, without
      looking much closer. I think the only relatively unambiguous result relates
      to proto-Matthew.

      There does not seem to be any reason on the 2SH for 202 and 200 to look
      alike at the 99.99% confidence level, when 200 does not even look much like
      211 or 210.
      200 looks as much like 202 as other things that clearly have the same
      author.

      Here are the relevant results again:
      200 looks like 201 and 202
      202 looks like 201 and 102
      002 does not look much like the other x0x types.
      But 002 looks like 112, whereas 200 does not look like 211.

      I think a proto-Matthew or Matthew&Q hypothesis has much less trouble with
      this than the 2SH or the FH. 200-202 is problem for the 2SH. 102-202 for
      the FH.

      The study also has given me some other ideas to look at.

      I'm mostly through entering the E-I data, so I'll have that to work with
      soon. Also, I found out that my econometrics class (which is really just
      data analysis), will let me work on this for the final class project, so
      I'll have some time for it.

      Dave Gentile
      Riverside, Illinois
      M.S. Physics
      PhD Management Science candidate

      -------------------------------------------------------------------------
      "When you have eliminated the impossible,
      whatever remains, however improbable, must be the truth."
      - Sherlock Holmes,
      in The Sign of Four, by Sir Arthur Conan Doyle.

      "Why sometimes I've believed as many as
      six impossible things before breakfast."
      - The Red Queen,
      in Through the Looking Glass, by Lewis Carroll



      Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
      List Owner: Synoptic-L-Owner@...
    • Brian E. Wilson
      Dave Gentile wrote -- ... Dave, I am still finding this fascinating. I am wondering whether the above was intended to be a summary of the state of play now
      Message 2 of 10 , Dec 5, 2001
        Dave Gentile wrote --
        >
        >Here are the relevant results again:
        >200 looks like 201 and 202
        >202 looks like 201 and 102
        >002 does not look much like the other x0x types.
        >But 002 looks like 112, whereas 200 does not look like 211.
        >
        >I think a proto-Matthew or Matthew&Q hypothesis has much less trouble
        >with this than the 2SH or the FH. 200-202 is problem for the 2SH.
        >102-202 for the FH.
        >
        Dave,
        I am still finding this fascinating. I am wondering whether the
        above was intended to be a summary of the state of play now that you
        have made use of the stats from both Alpha-Delta and Epsilon-Iota in the
        HHB Concordance? Is this a list of your statistical findings so far,
        please?

        I think these could be set out as eight statements, five correlations
        and five non-correlations (anti-correlations?) --

        (1) 200 <==> 201
        (2) 200 <==> 202
        (3) 202 <==> 201
        (4) 202 <==> 102
        (5) 002 <==> 112

        (6) 002 >xx< 202
        (7) 002 >xx< 201
        (8) 002 >xx< 102
        (9) 002 >xx< 200
        (10) 200 >xx< 211

        I understand both (1) ad (2) to be inconsistent with Sondergut Mt and
        the Double Tradition not being originally mostly from the same author.

        I understand (3) to be inconsistent with Lk and the Double Tradition not
        being originally mostly from the same author.

        I think (4) is inconsistent with Mt and the Double Tradition not being
        originally mostly from the same author.

        I would suggest that (5) is inconsistent with Sondergut Lk and triple
        tradition material not being originally mostly from the same author.

        I see (6), (7) and (8) as consistent with Lk having added wording of his
        own at various points in his gospel, the sort of words being favoured by
        Lk in his additions being generally unlike the sort of words found in
        the Double Tradition.

        I think (9) is consistent with Mt and Lk independently having added
        wording of their own in various parts of their gospels, the sort of
        wording favoured by one in his additions being generally different from
        the sort of wording favoured by the other in his additions.

        I understand (10) to be consistent with Mt having added wording of his
        own at various points in his gospel, including within the triple
        tradition, the words favoured by Mt in his additions being mostly unlike
        those found in the triple tradition.

        Would you agree with the above interpretations?

        I would suggest that, if the above interpretations are correct, then
        (1), (2), (3), (4) and (5) show that the Two Document Hypothesis is in
        difficulties, since the 2DH does not fit well these phenomena.

        I would also suggest that, if the above interpretations are correct, my
        Logia Translation Hypothesis fits well (1), (2), (3), (4), (5), (6),
        (7), (8), (9) and (10).

        Best wishes,
        BRIAN WILSON

        >HOMEPAGE http://www.twonh.demon.co.uk/

        Rev B.E.Wilson,10 York Close,Godmanchester,Huntingdon,Cambs,PE29 2EB,UK
        > "What can be said at all can be said clearly; and whereof one cannot
        > speak thereof one must be silent." Ludwig Wittgenstein, "Tractatus".
        _

        Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
        List Owner: Synoptic-L-Owner@...
      • David Gentile
        Hello Brian, ... That s not a complete list, I ll put that together at some point. Everything else was pretty much consistent with the 2SH, so I just reported
        Message 3 of 10 , Dec 5, 2001
          Hello Brian,

          > >
          > Dave,
          > I am still finding this fascinating. I am wondering whether the
          > above was intended to be a summary of the state of play now that you
          > have made use of the stats from both Alpha-Delta and Epsilon-Iota in the
          > HHB Concordance? Is this a list of your statistical findings so far,
          > please?

          That's not a complete list, I'll put that together at some point. Everything
          else was pretty much consistent with the 2SH, so I just reported the
          interesting results there.

          >
          > I think these could be set out as eight statements, five correlations
          > and five non-correlations (anti-correlations?) --

          > (1) 200 <==> 201
          > (2) 200 <==> 202
          > (3) 202 <==> 201
          > (4) 202 <==> 102
          > (5) 002 <==> 112
          >
          > (6) 002 >xx< 202
          > (7) 002 >xx< 201
          > (8) 002 >xx< 102
          > (9) 002 >xx< 200
          > (10) 200 >xx< 211

          6 , 7 ,and 9 are anti-correlation. 8 and 10 were non-correlated

          >
          > I understand both (1) ad (2) to be inconsistent with Sondergut Mt and
          > the Double Tradition not being originally mostly from the same author.

          I believe #1 is inconsistent with that. #2 could just be the result of
          Matthew having the same style editing Q and producing sondergut. Actually
          with the Alpha-Delta data, I can make a stronger statement. (I'm not sure of
          confidence levels in E-I) 102+202 (Luke's Q) correlates with 200. (Matthew's
          sondergut). That seems very difficult for the 2SH.
          Yes GNH works here.

          >
          > I understand (3) to be inconsistent with Lk and the Double Tradition not
          > being originally mostly from the same author.

          (202-201) It means Matthew's Q is reasonably consistent. One could use it to
          argue Matthew did not copy Luke, because in that case 201 would all be
          changes to the original.
          I think 3 would be expected on the 2SH or FH. On the FH they are both
          Matthew. On the 2SH both often represent Q, and also reflect words Matthew
          liked enough to keep. It's also generally supposed on the 2SH that Luke
          followed Q closely, so 202 and 201 would be expected to be similar.
          Yes. GNH works here.

          >
          > I think (4) is inconsistent with Mt and the Double Tradition not being
          > originally mostly from the same author.

          On the 2SH they (102 and 202) are both Q, and both words Luke liked enough
          to keep. So the 2SH has little problem here. On the FH one is Matthew, the
          other is Luke's changes. This seems a problem. GNH seems fine.

          >
          > I would suggest that (5) is inconsistent with Sondergut Lk and triple
          > tradition material not being originally mostly from the same author.

          Here they both reflect Luke's choice of words.
          Yep, GNH works.

          >
          > I see (6), (7) and (8) as consistent with Lk having added wording of his
          > own at various points in his gospel, the sort of words being favored by
          > Lk in his additions being generally unlike the sort of words found in
          > the Double Tradition.

          Something like that, yes. GNH works, yes.

          >
          > I think (9) is consistent with Mt and Lk independently having added
          > wording of their own in various parts of their gospels, the sort of
          > wording favoured by one in his additions being generally different from
          > the sort of wording favoured by the other in his additions.

          Yes. And, yes, GNH works.

          >
          > I understand (10) to be consistent with Mt having added wording of his
          > own at various points in his gospel, including within the triple
          > tradition, the words favoured by Mt in his additions being mostly unlike
          > those found in the triple tradition.

          Hmm. Well if 211 represents the triple tradition, that would work. But
          generally one would suppose Mark&Luke are the original and Matthew is the
          change. Then one would expect 211 to look like 200.


          >
          > Would you agree with the above interpretations?

          Mostly.

          >
          > I would suggest that, if the above interpretations are correct, then
          > (1), (2), (3), (4) and (5) show that the Two Document Hypothesis is in
          > difficulties, since the 2DH does not fit well these phenomena.
          >
          > I would also suggest that, if the above interpretations are correct, my
          > Logia Translation Hypothesis fits well (1), (2), (3), (4), (5), (6),
          > (7), (8), (9) and (10).

          Well, since I was using those to argue that Luke and Matthew and a common
          source that looked like a proto-Matthew, if we suppose the GN looked that
          way, they should support both. We'd have to look through the full set (the
          rest of which are consistent with the 2SH), to see what implications they
          had for the GNH. I think it might do reasonably well. But, I think we'd end
          up trying to claim that both Mark and Matthew look closest to the original.
          (Mark in triple tradition, Matthew elsewhere).
          Mark seems to have a stronger claim on the "boarder categories" (220, 221,
          122, 022).
          But that would run into a problem with what we assumed about 211 and 200.

          Just my thoughts.

          Dave Gentile
          Riverside, Illinois
          M.S. Physics
          PhD management Science candidate




          Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
          List Owner: Synoptic-L-Owner@...
        • Brian E. Wilson
          David Gentile wrote -- ... Dave, Thanks for the clarifications and comments. On the LTH, Mk is closest to the Greek Logia in the triple tradition, and Mt is
          Message 4 of 10 , Dec 5, 2001
            David Gentile wrote --
            >
            > I think we'd end up trying to claim that both Mark and Matthew look
            >closest to the original. (Mark in triple tradition, Matthew elsewhere).
            >
            Dave,
            Thanks for the clarifications and comments. On the LTH, Mk is
            closest to the Greek Logia in the triple tradition, and Mt is closer
            than Lk in wording (though not in order of material) to the original in
            the double tradition.

            I realize that Sondergut Mk is small and therefore sufficient data may
            be lacking, but have you investigated whether 200 and 020 are
            correlated, please?

            Best wishes,
            BRIAN WILSON

            >HOMEPAGE http://www.twonh.demon.co.uk/

            Rev B.E.Wilson,10 York Close,Godmanchester,Huntingdon,Cambs,PE29 2EB,UK
            > "What can be said at all can be said clearly; and whereof one cannot
            > speak thereof one must be silent." Ludwig Wittgenstein, "Tractatus".
            _

            Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
            List Owner: Synoptic-L-Owner@...
          • Mark Goodacre
            Thanks, Dave for your further reflections. I would like to make comments on some of the assumptions underlying some of these inferences from the data.
            Message 5 of 10 , Dec 5, 2001
              Thanks, Dave for your further reflections. I would like to make
              comments on some of the assumptions underlying some of these
              inferences from the data. Perhaps I might add another word of
              caution here. Not only do we need to make sure that the detailed
              data is all fully available and analysed before inferences are made,
              but also, perhaps most importantly, we need to be quite sure about
              what kinds of inferences are legitimate and on what grounds they are
              being made. The current general conclusions about x correlating with
              y and therefore both being from the same author (etc.) do not strike
              me as particularly useful.

              On 5 Dec 2001 at 10:00, David Gentile wrote:

              > (202-201) It means Matthew's Q is reasonably consistent. One could use
              > it to argue Matthew did not copy Luke, because in that case 201 would
              > all be changes to the original. I think 3 would be expected on the 2SH
              > or FH. On the FH they are both Matthew.

              These are QC (double tradition words common to Matthew and Luke) and
              Matthew's QD (double tradition words differing in Matthew). It's not
              quite right on the Farrer Theory that "they are both Matthew" -- QC
              words are (so to speak) Luke-pleasing selections from Matthew.

              > On the 2SH both often
              > represent Q, and also reflect words Matthew liked enough to keep. It's
              > also generally supposed on the 2SH that Luke followed Q closely, so
              > 202 and 201 would be expected to be similar.

              It is generally supposed on the Two-Source Theory that Luke follows
              Q's order more closely than does Matthew but it is not generally
              supposed that Luke follows Q's wording more closely and it is the
              latter that is relevant here. On the question of wording, it is
              sometimes Matthew and sometimes Luke who are held to have the more
              original wording.

              > On the 2SH they (102 and 202) are both Q, and both words Luke liked
              > enough to keep. So the 2SH has little problem here.

              Again, this needs some nuancing. 102 and 202 are Luke's QD and QC
              respectively. On the Two-Source Theory, 102 (Luke's QD) sometimes
              reflects the wording of Q and sometimes reflect Luke's own additions.

              > On the FH one is
              > Matthew, the other is Luke's changes. This seems a problem.

              QC on the Farrer Theory are the words retained by Luke from Matthew
              in double tradition, viz. the Luke-pleasing elements in the material
              taken over by Luke; they are not just "Matthew".

              Brian writes:

              > > I would also suggest that, if the above interpretations are correct,
              > > my Logia Translation Hypothesis fits well (1), (2), (3), (4), (5),
              > > (6), (7), (8), (9) and (10).

              To what extent do you think that this is simply a result of the way
              in which the Logia Translation Hypothesis is set up, viz. as the
              hypothetical source of all three Synoptic Gospels? In other words,
              what correlations / anti-correlations can you imagine that would
              falsify the hypothesis?

              Mark


              -----------------------------
              Dr Mark Goodacre mailto:M.S.Goodacre@...
              Dept of Theology tel: +44 121 414 7512
              University of Birmingham fax: +44 121 414 4381
              Birmingham B15 2TT UK

              http://www.bham.ac.uk/theology/goodacre
              http://NTGateway.com


              Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
              List Owner: Synoptic-L-Owner@...
            • David Gentile
              In A-D 020 and 200 show a non-significant negative (75% level) In E-I I would estimate a moderately significant negative. Dave Gentile Riverside, Illinois M.S.
              Message 6 of 10 , Dec 5, 2001
                In A-D 020 and 200 show a non-significant negative (75% level)
                In E-I I would estimate a moderately significant negative.

                Dave Gentile
                Riverside, Illinois
                M.S. Physics
                PhD Management Science candidate

                -------------------------------------------------------------------------
                "When you have eliminated the impossible,
                whatever remains, however improbable, must be the truth."
                - Sherlock Holmes,
                in The Sign of Four, by Sir Arthur Conan Doyle.

                "Why sometimes I've believed as many as
                six impossible things before breakfast."
                - The Red Queen,
                in Through the Looking Glass, by Lewis Carroll

                ----- Original Message -----
                From: "Brian E. Wilson" <brian@...>
                To: <Synoptic-L@...>
                Sent: Wednesday, December 05, 2001 11:08 AM
                Subject: [Synoptic-L] Difficulty in interpretation
                >
                > I realize that Sondergut Mk is small and therefore sufficient data may
                > be lacking, but have you investigated whether 200 and 020 are
                > correlated, please?
                >
                > Best wishes,
                > BRIAN WILSON



                Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
                List Owner: Synoptic-L-Owner@...
              • David Gentile
                ... Mark, I understand your concerns here. All the data is not in, and more work is needed to interpret the results. I also understand your point about 202
                Message 7 of 10 , Dec 5, 2001
                  Mark Goodacre wrote:


                  > Thanks, Dave for your further reflections. I would like to make
                  > comments on some of the assumptions underlying some of these
                  > inferences from the data. Perhaps I might add another word of
                  > caution here. Not only do we need to make sure that the detailed
                  > data is all fully available and analysed before inferences are made,
                  > but also, perhaps most importantly, we need to be quite sure about
                  > what kinds of inferences are legitimate and on what grounds they are
                  > being made. The current general conclusions about x correlating with
                  > y and therefore both being from the same author (etc.) do not strike
                  > me as particularly useful.
                  >

                  Mark,

                  I understand your concerns here. All the data is not in, and more work is
                  needed to interpret the results. I also understand your point about 202
                  being Luke pleasing and 102 being words chosen by Luke.

                  One problem with this idea is that in other cases where we suspect one
                  author editing the other on the FH (x22 - x21) (22x - 12x) there are
                  distinct negatives.

                  Another problem is that, at least in the A-D data, 102 is positively
                  correlated with (202+201)
                  On the FH this would say that Luke's word choice in editing Q is very
                  similar to Matthew's Q. the document he is editing.

                  But, as you point out, more needs to be done. Hopefully looking at the work
                  in progress is interesting.

                  Thanks,

                  Dave Gentile
                  Riverside, Illinois
                  M.S. Physics
                  PhD Management Science candidate




                  Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
                  List Owner: Synoptic-L-Owner@...
                • Jeffrey Glen Jackson
                  It seems to me that these statistical tools might be tested by experimentation. My idea for an experiment would be to construct some stories that mimic the
                  Message 8 of 10 , Dec 5, 2001
                    It seems to me that these statistical tools might be tested by
                    experimentation. My idea for an experiment would be to
                    construct some stories that mimic the nature of the (hypothetical)
                    sources (as well as some that don't), then give them to a
                    large number of test subjects to combine together into
                    single stories, giving different groups different instructions
                    on how to go about doing so, then applying your statistical
                    methods to the results to see if your metrics actually
                    correlate to the known behaviors of test subjects. Since
                    you'll have all the original source documents, you'll be
                    able to reconstruct exactly what they really did vs. what
                    your analytical method supposes that they did.


                    ><> Jeffrey Glen Jackson, son of Albert, son of George, son of <><
                    ><> Henry, son of Miles, son of Randolph, son of Ephraim, son of <><
                    ><> Thomas, son of John, son of Thomas, .... sonne of Jack. <><
                    mailto:jeff@... http://www.jeff-jackson.com
                    "The blithe 'reconstruction' not only of Q, not only of its different
                    stages of composition, but even of complete communities whose
                    beliefs are accurately reflected in these different stages, betokens
                    a naive willingness to believe in anything as long as it is nothing
                    like Mark (let alone Paul)." N. T. Wright




                    Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
                    List Owner: Synoptic-L-Owner@...
                  • David Gentile
                    ... Something like that had occurred to me. You could include people with different levels of familiarity with the language too. It would make in interesting
                    Message 9 of 10 , Dec 5, 2001
                      Jeffrey Glen Jackson wrote:

                      > It seems to me that these statistical tools might be tested by
                      > experimentation. My idea for an experiment would be to
                      > construct some stories that mimic the nature of the (hypothetical)
                      > sources (as well as some that don't), then give them to a
                      > large number of test subjects to combine together into
                      > single stories, giving different groups different instructions
                      > on how to go about doing so, then applying your statistical
                      > methods to the results to see if your metrics actually
                      > correlate to the known behaviors of test subjects. Since
                      > you'll have all the original source documents, you'll be
                      > able to reconstruct exactly what they really did vs. what
                      > your analytical method supposes that they did.
                      >
                      >


                      Something like that had occurred to me. You could include people with
                      different levels of familiarity with the language too. It would make in
                      interesting project.

                      Dave Gentile
                      Riverside, Illinois
                      M.S. Physics
                      PhD Management Science candidate


                      Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
                      List Owner: Synoptic-L-Owner@...
                    • Brian E. Wilson
                      ... Mark Goodacre replied -- ... Mark, My previous attempt at answering this point was flawed because I was mis-understanding the nature of the correlations
                      Message 10 of 10 , Dec 15, 2001
                        Brian Wilson wrote:
                        >
                        >a strong anti-correlation between 222 and 202 would place the LTH in
                        >difficulties, since this would appear to be inconsistent with the
                        >triple tradition and the double tradition being originally mostly from
                        >the same author (since the LTH posits that the triple tradition and the
                        >double tradition were originally both from the Greek Logia written by
                        >one author).
                        >
                        Mark Goodacre replied --
                        >
                        >I don't know about that. Wouldn't 202 on the Logia Translation
                        >Hypothesis be characterised by Mark-displeasing material in contrast
                        >with 222 which features Mark-pleasing material? The all important
                        >Markan element here could, on your hypothesis, explain a strong anti-
                        >correlation.
                        >
                        Mark,
                        My previous attempt at answering this point was flawed because I
                        was mis-understanding the nature of the correlations indicated by Dave
                        Gentile's tables. In the hope that I have gained some insight since
                        then, I would like to reply again.

                        On the LTH, 202 words must be virtually all Greek Logia words. Even if
                        Mk omitted the double tradition passages that contained these words, so,
                        by definition making the words "Mark-displeasing", they were
                        nonetheless Greek-Logia-pleasing words. They were in the style of the
                        writer of the Greek Logia. On the LTH also, 222 words must also be
                        virtually all Greek Logia words. They also were in the style of the
                        writer of the Greek Logia. It would therefore be very unlikely that 202
                        would strongly differ from 222 words. That is, it would be very unlikely
                        that there would be a significant negative correlation between 202 and
                        222. If, theoretically, such a significant negative correlation were to
                        be observed, then this would be a difficulty for the LTH.

                        The same argument applies to 220 and 222. And also to 022 and 222. So
                        there are at least three theoretically possible significant negative
                        correlations that would, if observed, each place the LTH in difficulties
                        --
                        (i) 202 >xx< 222
                        (ii) 220 >xx< 222
                        (iii) 022 >xx< 222
                        >
                        >Thus I'm still not convinced that there is any data that could, in the
                        >nature of the case, falsify the Logia Translation Hypothesis.
                        >
                        Would you like to reply to the above argument on this?
                        >
                        >Are you concerned that this might be one of the dangers with the
                        >hypothesis, that it is potentially so elastic that it can accommodate a
                        >great deal?
                        >
                        I really do not think it is any more elastic than the 2DH, FH or GH. It
                        is no more unfalsifiable than they are. As I think I have shown, the LTH
                        would be in trouble if certain significantly negative correlations were
                        observed. So the LTH is testable.

                        On the question of whether the LTH "can accommodate a great deal", you
                        seem to be suggesting that this would be a black mark against it. If a
                        hypothesis is true, however, surely it should be able to account easily
                        for all the observed data. It therefore should be able to accommodate a
                        great deal. That would be a mark in its favour.

                        Best wishes,
                        BRIAN WILSON

                        >HOMEPAGE http://www.twonh.demon.co.uk/

                        Rev B.E.Wilson,10 York Close,Godmanchester,Huntingdon,Cambs,PE29 2EB,UK
                        > "What can be said at all can be said clearly; and whereof one cannot
                        > speak thereof one must be silent." Ludwig Wittgenstein, "Tractatus".
                        _

                        Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
                        List Owner: Synoptic-L-Owner@...
                      Your message has been successfully submitted and would be delivered to recipients shortly.