Transliteration from Greek into ASCII

From: Edward Hobbs (
Date: Wed Jun 26 1996 - 17:40:08 EDT

Al Kidd wrote:

My face is red. Well, after all my effort to get the transliteration
scheme to conform with what seem to be acceptable here, I
still got part of it wrong..... I did correct
more of the transliteration.

*Am I correct in holding that b-greek readership prefers all caps for
transliterating Greek characters (but exception is made when "i" is
used in transliterating "iota subscript")--and that it prefers a
one-to-one correspondence in the transliteration scheme?


Transliteration: The answer is iffy!

Several years ago there were various proposals for uniformity, with good
reasons advanced in favor of each of six schemes. At the time, two of those
pushed heavily used a rather elaborate scheme (such as the Beta-code) to
represent accents and breathings. Since these are in fact editorial decisions,
and not in the older MSS. (including the autographs [which I must have seen,
to make that comment!]), most of us objected, allowing every poster to use any
scheme he/she found comfortable, since we all would know what text was meant.
In some matters, however, such as textual variants, this is a bit confusing.

Interestingly enough, a kind of scheme has found general acceptance, with two
or three letters still not settled.

(1) Yes, most of us now follow a practice urged by Carl Conrad, to use CAPITALS
when transliterating Greek letters, on a one-to-one basis,,and reserving
lower-case /i/ to represent iota-subscript and lower-case /h/ to represent
rough breathing. No accents, no smooth breathings. And no distinction between
medial and final sigma, that also being a later affectation, not a spelling
variation. (George Kilpatrick refused to use final sigma when transcribing
texts, for that very reason.)

(2) A few characters without Roman single-character form are usually done with
almost-look-a-like Roman characters otherwise unneeded:

        theta = Q
        eta = H
        psi = Y (upsilon is always U)
        omega = W

(3) Digraphs (in the usual Roman transliteration) are handled in three
different ways, usually, to avoid two-letter transliterations.

        (a) THeta uses Q ("lookalike"), as above.
        (a) PSi uses Y ("lookalike"), as above.
        (b) PHi uses F (sound equivalence).
        (c) CHi uses C (only letter left!).

(4) There being no single Roman letter for XI other than X, the "lookalike"
use of X for "chi" is confusing, though some use it. And some seem to like to
use C for sigma (I suppose using the old lunate-sigma of ancient days, mimicked
by Westcott-Hort in their original edition of 1881). Since S is otherwise
unused, and poses no confusion whatever, using C for sigma makes for problems
in decoding back to Greek, especially since it is the only letter available
for "chi" (unless X is used, thus posing a problem for "xi"). And occasionally
someone uses P for "rho", making problems for how to represent "pi".

AGREEMENT, CERTAINLY NOT A REQUIREMENT! You will catch me not following it,
since I use the SBL or else the Chicago Manual system when I write for
publication, unless Greek itself can be printed (now often the case), and I use
Beta-code when communicating with those who require it -- thus leading to
lapses when I must use ASCII 0-127 on the Net.

Two of our Listers several years ago posted six different schemes available
using ASCII; but no agreement was reached, and what is described above is a
kind of evolved semi-accepted, semi-used way of doing it.

Edward Hobbs

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:37:45 EDT