Alphabet, Abjad and so forth

James E. Clapp jeclapp at WANS.NET
Wed Mar 20 19:51:31 UTC 2002


Benjamin Fortson writes:

> > [JC:] It was not until around the fourth century BC that stunningly modern
> > phonetic analysis made possible a reordering of the consonants into neat
> > rows and columns according where the sound was made in the mouth (the rows)
> > and whether it was voiced or unvoiced, aspirated or unaspirated, continuant,
> > and the like (the columns).  Unfortunately for us, that intellectual
> > achievement occurred in India, too far away and too late to influence the
> > order of the Greek alphabet.  So we are stuck with an alphabet that is
> > pretty much in random order (except in a historical sense), but at least it
> > includes letters for the vowels; South and Southeast Asian alphabets derived
> > from the Indic offshoot of the original Semitic writing systems have neatly
> > organized alphabets based upon scientific phonetic principles, but generally
> > lack full-fledged letters for the vowels sounds.
>
> [BF:]All true except for the end--these alphabets *do* have full-fledged signs
> for the vowels (and diphthongs). But because the consonant signs "include"
> in their pronunciation a following vowel "a", which can be modified by the
> addition of superscript, subscript, preceding, or following marks, the
> vowel letters are only needed word-initially, or word-internally following
> another vowel sound.


You're being generous.  My effort to summarize the 3500-year history of
alphabetic writing in three or four paragraphs necessarily involved such
oversimplification that it is a stretch to call any of it "true."  I'd hoped
that by including enough weasel words (". . . *generally* lack *full-fledged*
letters for the vowel sounds") I could allude to a fundamental structural
difference between western alphabets derived from Greek and eastern writing
systems derived from the Brahmi script that took form in India circa 4th century
B.C.E. without going into detail about the latter--which is pretty far afield
for this list.

But of course you are right: the Brahmi-based scripts do have marks for most
of the vowel sounds.  As stated by Richard G. Salomon in the introduction to
Part VI ("South Asian Writing Systems") in _The World's Writing Systems_ (cited
in full in my first posting in this thread), p. 372:

          ". . . in systemic terms, the Indic scripts typically
          share the same basic principles of the aksara system,
          i.e. a modified consonantal syllabary representing most
          vowels by diacritic signs attached to the consonants."

William Bright, in the book's article on "The Devanagari Script," characterizes
such scripts as follows (p. 384):

          "Typologically, it is what I call an alphasyllabary:
          that is, it writes each consonant-vowel sequence as a
          unit, called an aksara, in which the vowel functions
          as an obligatory diacritic to the consonant; in the
          terminology of Daniels . . . , it is an abugida."

[See the definitions of "alphasyllabary" and "abugida" quoted from that book in
my original post.]

Of course, as you point out, there is no mark for the vowel sound regarded as
"inherent" in the consonant, and in most such scripts there is, in addition to
the set of diacritics, a separate set of entirely different symbols for the
vowels that occur at the beginning of a line (or, in scripts that separate the
words, at the beginning of a word).

In Thai--the only such language I've actually studied--the unwritten (or "zero"
or "inherent") vowel has different phonetic values in syllable-medial and
syllable-final position, and the written vowel signs take different written
forms in those two contexts.  As to the written symbols, some are placed after
the consonant with which they are associated (just as a vowel letter in a Greek-
derived script would be), others over or under or before the consonant, others
in two parts before and after the consonant, and still others in three parts
before and over and after the consonant.  There are no special symbols for
initial vowels as in Devanagari; instead, a dummy consonant is put at the
beginning of words whose pronunciation begins with a vowel sound as a
placeholder to write the vowel symbol(s) around.  When the first sound in the
word is one of the unwritten vowels, you have the surreal phenomenon of an
unpronounced consonant inserted to indicate the presence of an unwritten vowel.
(It's like the pantomime in the final scene of Antonioni's "Blow-Up": in an
imaginary tennis game, it shows where the viewer is to imagine the ball would be
if the game were played with a ball--which it's not, because the pantomime *is*
the game.)  And just to torture us foreign students further, that placeholder is
a symbol that, depending upon the context (which is extremely hard to determine
since the Thais view spaces between words as a crutch appropriate only for
children's primers), can itself function as a vowel symbol--or occasionally
neither as a vowel nor as a dummy consonant, but as a tone indicator.  Because
the vowels--even the unwritten ones--exist conceptually and physically only in
relation to consonants, the words on a vertical Thai sign (on a shop or theater)
are stacked syllable by syllable; there isn't any way to sever the vowels from
the consonants to which they belong so as to stack them as we might the letters
R-O-X-Y or B-O-O-K-S.

So that's what I was trying to get at with my one admittedly inadequate
sentence: Thanks to the Indian pandits the Thais have an elegantly ordered array
of consonants, but their system for indicating (or not indicating) vowels is a
mess; thanks to the Greeks we have an alphabetical order that only a Phoenician
of the year 1000 B.C. could love, but it includes vowels as letters.  As one who
has worked with dictionaries in both languages, I'm grateful to the Greeks!

James E. Clapp
Author of Random House Webster's Dictionary of the Law



More information about the Ads-l mailing list