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

[Synoptic-L] the nature of the data

Expand Messages
  • Brian E. Wilson
    Brian Wilson wrote -- ... Stephen Carlson replied -- ... Stephen, But the aim of the game is to discover the relationship between the synoptic gospels, not the
    Message 1 of 27 , Dec 2, 2001
    • 0 Attachment
      Brian Wilson wrote --
      >
      >The more I look at the HHB categorization the more I wonder whether
      >analysing the stats reveals the decisions of the editors of the
      >Concordance rather than the relationship between the synoptic gospels.
      >
      Stephen Carlson replied --
      >
      >Of course, it reveals the decisions of the editors about the
      >relationship between the synoptic gospels.
      >
      Stephen,
      But the aim of the game is to discover the relationship between
      the synoptic gospels, not the editors' decisions.

      I would suggest that a computer could be programmed to categorize every
      word of the synoptic gospels into 222, 221, 220, and so on, without
      assuming the Two Document Hypothesis (as do the editors of the HHB
      Concordance), or any other supposed documentary relationship between the
      synoptic gospels. The program might be designed first to search for
      strings of words in one synoptic gospel containing at least a stated
      number of words (or word-roots) in the same order as strings of words of
      the same size in another synoptic gospel. This could be extended to find
      similarities of wording in the same order in all three synoptic gospels
      within stated parameters. The program would establish triple parallel,
      double parallel, and Sondergut pieces of material. The size of each unit
      of triple, double, or special, could be, say, at least 8 word-roots in
      the same order, though of course a triple, double or special could be
      longer than this minimum. Whatever minimum size is laid down would be
      arbitrary, but at least the procedure using it would be independent of
      any assumed documentary hypothesis. On this basis, the categorization
      into 222, 221, 220 and so on could be carried out without making any
      assumptions of the documentary relationship between the synoptic
      gospels. 222 would represent a word-root present in each synoptic gospel
      in a computer-generated triple parallel, and so on. This would give
      "clean data", to use Dave Gentile's terminology. It would produce
      results that would not be dependent on the decisions of the editors of
      the HHB Concordance. I think that ideally this is where the analysis
      might have started.
      >
      >The main question, however, is whether and to what extent the editors'
      >judgment is biased to a particular solution to the synoptic problem.
      >
      I agree. That is one question I have been asking. The other question I
      have been raising is to what extent the editors' have applied their
      criteria inconsistently.

      What is really needed is computer-produced data that is not dependent on
      any synoptic documentary hypothesis and that applies criteria
      consistently. The HHB Concordance does not provide such data.

      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@...
    • Stephen C. Carlson
      ... If you (or anyone else) could produce such a computer program, I d be very impressed. However, one cannot even identify a parallel without supposing some
      Message 2 of 27 , Dec 2, 2001
      • 0 Attachment
        At 06:07 PM 12/2/2001 +0000, Brian E. Wilson wrote:
        >I would suggest that a computer could be programmed to categorize every
        >word of the synoptic gospels into 222, 221, 220, and so on, without
        >assuming the Two Document Hypothesis (as do the editors of the HHB
        >Concordance), or any other supposed documentary relationship between the
        >synoptic gospels.

        If you (or anyone else) could produce such a computer program,
        I'd be very impressed. However, one cannot even identify a parallel
        without supposing some documentary relationship between the synoptic
        gospels. For example, the decision to identify Mark's healing of
        the paralytic as a parallel to Matthew's healing of the paralytic --
        but not to Matthew's genealogy -- supposes that there is a documentary
        relationship with the former but not the latter. Of course, most
        decisions about parallelization will not be controversial, but
        they will be for not only for the so-called Mark-Q overlap texts and
        for doublets in one gospel where there is only one instance in the
        others (which of the two in the doublet is "the" parallel?). Even
        in noncontroversial parallels, there is discretion in deciding which
        of two KAIs in Mark correspond to the one KAI in Matthew. Frankly,
        I doubt it can be done objectively, because the theories and hypotheses
        we hold influence our perception of the data. This extends to the
        programmer of the computer as well.

        >>The main question, however, is whether and to what extent the editors'
        >>judgment is biased to a particular solution to the synoptic problem.
        >>
        >I agree. That is one question I have been asking. The other question I
        >have been raising is to what extent the editors' have applied their
        >criteria inconsistently.

        The editors admit their bias on the Mark-Q overlaps, but does that
        bias extend to other portions? You've pointed out some inconsistencies,
        but unless one can show they somehow favor the 2ST, it may be better
        to attribute the inconsistencies to noise rather than bias.

        >What is really needed is computer-produced data that is not dependent on
        >any synoptic documentary hypothesis and that applies criteria
        >consistently.

        I'm not holding my breath. Perhaps a better approach is to ask
        the proponents of the various solutions to produce their own
        "partisan" synopses and concordances that sets forth the data
        in such a manner that renders their solution in the most favorable
        light. Then we can compare them to see how well they handle the
        data. Maybe Dungan was right after all that there is no neutral
        synopsis.

        Stephen Carlson
        --
        Stephen C. Carlson mailto:scarlson@...
        Synoptic Problem Home Page http://www.mindspring.com/~scarlson/synopt/
        "Poetry speaks of aspirations, and songs chant the words." Shujing 2.35


        Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
        List Owner: Synoptic-L-Owner@...
      • Brian E. Wilson
        Brian Wilson wrote -- ... Stephen Carlson replied -- ... Stephen, I do not understand your argument here. The occurrence of parallels between Mark s healing of
        Message 3 of 27 , Dec 3, 2001
        • 0 Attachment
          Brian Wilson wrote --
          >
          >I would suggest that a computer could be programmed to categorize every
          >word of the synoptic gospels into 222, 221, 220, and so on, without
          >assuming the Two Document Hypothesis (as do the editors of the HHB
          >Concordance), or any other supposed documentary relationship between
          >the synoptic gospels. The program might be designed first to search for
          >strings of words in one synoptic gospel containing at least a stated
          >number of words (or word-roots) in the same order as strings of words
          >of the same size in another synoptic gospel. This could be extended to
          >find similarities of wording in the same order in all three synoptic
          >gospels within stated parameters. The program would establish triple
          >parallel, double parallel, and Sondergut pieces of material. The size
          >of each unit of triple, double, or special, could be, say, at least 8
          >word-roots in the same order, though of course a triple, double or
          >special could be longer than this minimum. Whatever minimum size is
          >laid down would be arbitrary, but at least the procedure using it would
          >be independent of any assumed documentary hypothesis. On this basis,
          >the categorization into 222, 221, 220 and so on could be carried out
          >without making any assumptions of the documentary relationship between
          >the synoptic gospels. 222 would represent a word-root present in each
          >synoptic gospel in a computer-generated triple parallel, and so on.
          >
          Stephen Carlson replied --
          >
          >If you (or anyone else) could produce such a computer program,
          >I'd be very impressed. However, one cannot even identify a parallel
          >without supposing some documentary relationship between the synoptic
          >gospels. For example, the decision to identify Mark's healing of
          >the paralytic as a parallel to Matthew's healing of the paralytic --
          >but not to Matthew's genealogy -- supposes that there is a documentary
          >relationship with the former but not the latter.
          >
          Stephen,
          I do not understand your argument here. The occurrence of
          parallels between Mark's healing of the paralytic and Matthew's healing
          of the paralytic but not between Matthew's healing of the paralytic and
          his Genealogy, is precisely the sort of result such a computer program
          would produce without assuming a documentary hypothesis such as the 2DH,
          FH or GH. There are many instances of strings of, say, no more than 20
          words in the Markan healing that contain at least 8 word-roots the same,
          and in the same order, in a string of no more than 20 words in the
          Matthean healing. On the other hand, there is no string of no more than
          20 words in the Matthaean healing that contains at least 8 word-roots
          the same and in the same order in a corresponding string of no more than
          20 words in the Genealogy in Matthew. The criterion would establish that
          the two healing stories include parallels in the sense of having many
          word roots the same and in the same order within strings of specified
          lengths, but that the Matthaean healing and the Genealogy in Matthew do
          not include parallels in this sense. It would be easy to set up a
          computer program to show this.

          Indeed, we could carry out such a check by hand, without using a
          computer program to spot parallels. I have just done this. There are no
          pairs of strings that meet the criteria above between the healing of the
          paralytic in Mt and the Genealogy in Mt. However, the following string
          of 20 words is found in the Matthaean healing of the paralytic --

          KAI IDWN O IHSOUJ THN PISTIN AUTWN EIPEN TW PARALUTIKW QARSEI TEKNON
          AFIENTAI SOU AI AMARTIAI KAI IDOU TINEJ TWN

          And it has 17 word roots the same and in the same order as the string of
          words in the Markan healing of the paralytic --

          KAI IDWN O IHSOUJ THN PISTIN AUTWN LEGEI TW PARALUTIKW TEKNON AFI/ENTAI
          SOU AI AMARTIAI HSAN DE TINEJ TWN

          Similarly, in the same two passages, the following are pairs of strings,
          one string from each passage, of no more than 20 words each such that in
          each pair there are at least 8 word roots the same and in the same order
          as in a string of no more than 20 words the same and in the same order
          in the other member of the pair --

          EN TAIJ KARDIAIJ UMWN TI GAR ESTIN EUKOPWTERON EIPEIN AFIENTAI SOU AI
          AMARTIAI H EIPEIN EGEIRE KAI

          EN TAIJ KARDIAIJ UMWN TI ESTIN EUKOPWTERON EIPEI=N TW PARALUTIKW
          AFI/ENTAI SOU AI AMARTIAI H EIPEIN EGEIRE KAI

          ===

          PERIPATEI INA DE EIDHTE OTI ECOUSIAN EXEI O UIOJ TOU ANQRWPOU EPI THJ
          GHJ AFIENAI AMARTIAJ TOTE LEGEI TW PARALUTIKW

          PERIPATEI INA DE EIDHTE OTI ECOUSIAN EXEI O UIOJ TOU ANQRWPOU AFIENAI
          AMARTIAJ EPI THJ GHJ LEGEI TW PARALUTIKW

          ===

          EGERQEIJ ARON SOU THN KLINHN KAI UPAGE EIJ TON OIKON SOU KAI EGERQEIJ

          EGEIRE ARON TON KRABATTON SOU KAI UPAGE EIJ TON OIKON SOU KAI HGERQH

          ===

          Each of these strings is a pair of parallels between Mt and Mk, and many
          words within these strings (after doing the same sort of comparison with
          Lk) could be categorized 220, 221, and so on. These are by no means all
          the parallels of this type that can be listed. These parallels, and
          others, between Mt and Mk could have been spotted by a computer
          systematically comparing every string of 20 words in Mt with every
          string of 20 words in Mk, and looking for at least 8 word-roots the same
          and in the same order in each pair of strings compared.

          The point is that it is totally unnecessary to assume a synoptic
          documentary hypothesis such as the 2DH, FH or GH, to establish that two
          pieces of material, one in Mt and the other in Mk, are parallels, or
          that two pieces of material in Mt are not parallels.

          You also wrote --
          >
          >Maybe Dungan was right after all that there is no neutral synopsis.
          >
          I am sure he was right. The HHB Concordance, however, is not a synopsis,
          and the question is whether such a Concordance can be produced without
          assuming a documentary hypothesis such as the 2DH, FH or GH. The answer
          is that it can.

          In the same way, a colour-coded text showing similarities of wording
          between the synoptic gospels, like W. R. Farmer's SYNOPTICON, is not a
          synopsis. Such a colour-coded text of the synoptic gospels can be
          constructed without assuming any particular solution to the synoptic
          problem, and without using any synopsis.

          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@...
        • Emmanuel Fritsch
          ... I see some different biases in an automatic process : - Brian pointed out that a gospel may be selected, and the priority of this gospel (typicallly :
          Message 4 of 27 , Dec 4, 2001
          • 0 Attachment
            "Stephen C. Carlson" wrote :
            >
            > Brian E. Wilson wrote:
            > >I would suggest that a computer could be programmed to categorize every
            > >word of the synoptic gospels into 222, 221, 220, and so on, without
            > >assuming the Two Document Hypothesis (as do the editors of the HHB
            > >Concordance), or any other supposed documentary relationship between the
            > >synoptic gospels.
            >
            > If you (or anyone else) could produce such a computer program,
            > I'd be very impressed. However, one cannot even identify a parallel
            > without supposing some documentary relationship between the synoptic
            > gospels. [...] Even
            > in noncontroversial parallels, there is discretion in deciding which
            > of two KAIs in Mark correspond to the one KAI in Matthew. Frankly,
            > I doubt it can be done objectively, because the theories and hypotheses
            > we hold influence our perception of the data. This extends to the
            > programmer of the computer as well.

            I see some different biases in an automatic process :
            - Brian pointed out that a gospel may be selected, and the priority
            of this gospel (typicallly : Mark) is assumed, biasing the result.
            In fact, in the treatment, the input gospels should be processed
            with equity.
            - Even if synoptic gospels are processed in equity, some a priori
            may influence the result. I mean particularly to the pattern
            of redaction process that will be seen as most probable : a
            first small document that is step by step increased, or a large
            first document that as been cut toward our gospel. the balance
            between deletion and completion in synoptic process is a global
            a priori in synoptic study.
            - Then, an automatic process would hardly integrate all the other
            information, particularly the links between Luke and John.


            > >What is really needed is computer-produced data that is not dependent on
            > >any synoptic documentary hypothesis and that applies criteria
            > >consistently.
            >
            > I'm not holding my breath. Perhaps a better approach is to ask
            > the proponents of the various solutions to produce their own
            > "partisan" synopses and concordances that sets forth the data
            > in such a manner that renders their solution in the most favorable
            > light. Then we can compare them to see how well they handle the
            > data. Maybe Dungan was right after all that there is no neutral
            > synopsis.

            Are there any referenced biases in Boismard's Synopse ?
            Do you know about a comparison of different existing synopses ?

            a+
            manu

            Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
            List Owner: Synoptic-L-Owner@...
          • David Inglis
            ... Dave, did you actually mean: why does 221 look like both 121 and 211, but 122 look only like 121 and not 112 , or have I actually got the wrong data? In
            Message 5 of 27 , Dec 19, 2001
            • 0 Attachment
              Way back on Dec 1 Dave Gentile wrote:

              > But perhaps the real question is, why does
              > 122 look like both 121 and 112, but 221 look only like 121 and not 211?

              Dave, did you actually mean: "why does 221 look like both 121 and 211, but
              122 look only like 121 and not 112", or have I actually got the wrong data?
              In any case, have you decided why?

              Dave Inglis
              david@...
              3538 O'Connor Drive
              Lafayette, CA, USA



              Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
              List Owner: Synoptic-L-Owner@...
            • David Gentile
              Hi Dave, The case reversed itself with more data. It was 121 - 122 now its 121 - 221 . These appear to be redactor effects / chance , they vanish in the macro
              Message 6 of 27 , Dec 20, 2001
              • 0 Attachment
                Hi Dave,

                The case reversed itself with more data. It was 121 - 122 now its 121 - 221
                . These appear to be redactor effects / chance , they vanish in the macro
                categories.

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

                ----- Original Message -----
                From: "David Inglis" <david@...>
                To: <Synoptic-L@...>
                Sent: Thursday, December 20, 2001 12:54 AM
                Subject: Re: [Synoptic-L] Some numerical results


                > Way back on Dec 1 Dave Gentile wrote:
                >
                > > But perhaps the real question is, why does
                > > 122 look like both 121 and 112, but 221 look only like 121 and not 211?
                >
                > Dave, did you actually mean: "why does 221 look like both 121 and 211, but
                > 122 look only like 121 and not 112", or have I actually got the wrong
                data?
                > In any case, have you decided why?
                >
                > Dave Inglis



                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.