Re: A Challenge
- On Mon, 11 Aug 1997, "Vinton A. Dearing" <dearing@...>
wrote, in part:
>Thanks to Bob Waltz for his interest and to Jimmy Adair forLet me also thank the other two for their clarifications. All
>explaining some things in my original message. Some additional
is not yet clear to me -- but then, that is what the books are
> My method of textual criticism is not "mine": it is a genealogicalIf it helps anyone in reading my Encyclopedia articles, the former
>method that expands Walter Greg's rules in The Calculus of Variants
>(1927). Greg divided variations into simple and complex types.
>Simple variations have only two readings, e.g., "yes" vs. "yea";
>complex variations have more than two, e.g. "no" vs. "nay" vs.
>"nix" vs. "nada" vs. "not on your life" etc.
are what I call "binary readings"; the latter are "ternary" or
higher order readings. It seems to me I've seen other terminology,
too. Has anyone attempted to set a standard for this?
[ ... ]
>Silva and Kirsopp used toAn interesting and noteworthy observation. Thanks for bringing it
>collate manuscripts by reading aloud to one another, and she reported
>that often a monk studying nearby would "correct" what she was
>saying. Such correction by memory must have been possible as soon as
>people set a high value on the sacred texts.
[ ... ]
> Let me try again to explain what I think might be possible inThat does help a bit. I'm not sure I could construct this, but at
>ringbreaking. Suppose we have seven texts, and variations as follows:
>tis] A; pou B,C,D,E,F,G; dh] A,B; de C,D,.E,F,G; pallakh] A,B,C;
>gunh D,E,F,G; uioi] A,B,C,D; uiwn E,F,G; eikosth] A,B,C,E; eikadi
>D,F,G; and osa] A,B,C,D,E,F; a G. Now, according to Greg's rules
>(which I accept), the relationship of the seven texts is
>A--B--C,D,E,F--G and there is what I call a ring in the group of four
>texts. Rings only occur when at least two groups of texts "exchange"
>readings, as Greg put it, so the first and last variations are not
>part of the ring. The second variation is not part of a ring because
>A and B don't exchange readings with any other pair or more of other
>texts. The ring comes when C and D have uioi where E and F have uiwn,
>and C and E have eikosth where D and F eikadi (never mind that other
>texts have these readings, C and D and E and F are the only ones to
>exchange readings. (Another way to put it is to say that the pair C,D
>overlaps both the pair C,E and the pair D,F, and so on around the
>ring; it is also true that a pair which overlaps only one other pair
>will not be part of a ring.) And what is true of pairs is true of
>larger groups of texts. Isn't there some way to start with A, let
>us say, and feel along the sequence A (which has tis), A--B (which
>have dh), recognize that we have come to a ring when we come to C,
>break it, say between C and E, and feel along once more through C
>(A,B and C have pallakh), to D (all four have uioi; we disregard the
>fact E has eikosth like A, B and C), to F (D, F and G have eikadi),
>and then separately from F to E (E, F and G have uiwn and E has the
>now-sourceless eikosth) and from F to G (all but G have osa). Maybe
>not, but it's a challenge.
least I understand it a bit better.
Unfortunately, I can't see any way to help with the code. I also
showed the problem to another programmer type (a higher order sort
of person than I am, BTW; I slog code to get work done, but he
programs for fun -- shudder!). I'm afraid we were in agreement
that the problem isn't well enough defined for the algorithm
to be evident. No doubt if the problem were better specified,
we would have better luck -- but if it were better specified,
no doubt Vinton Deering the challenge would not have been
Wish I could help more.
Robert B. Waltz
Want more loudmouthed opinions about textual criticism?
Try my web page: http://www.skypoint.com/~waltzmn
(A site inspired by the Encyclopedia of NT Textual Criticism)