From: Philippe Verdy (firstname.lastname@example.org)
Date: Tue Aug 21 2007 - 23:53:59 CDT
Peter Constable wrote :
> In your message below, you say that a spectrum analysis will reveal that
> there is no such thing as a conjunct and that /ksh/ is two consonants.
> Spectra analyses and Unicode have no connection whatsoever. (...)
> Rather, Unicode pertains to the real of graphic characters – the visual
> realm, not the audio realm.
I would add to this that the Indic scripts themselves do not encode speech
directly; it is used for noting orthographically what was pertinent to some
speakers at some period of time, and the text remains by tradition, even if
the words are spoken differently today.
A written script is just a conventional notation for denoting the actual
speech, but only for mutual understanding, it does not really dictate the
effective phonetic, and sometimes not even the phonology.
This occurs with any script (even though there will exist orthographic
reforms from time to time to make the written script nearer from the vocal
speech, but there is a strong resistance because of the various spoken
regional accents, and the written script is already an accepted form for
If we just look at the written notations for the "actual" speech, you'll see
immediately that, despite of the international agreement about the IPA and
the rich set of symbols it contains, no one uses it the same way, and more
or less simplified notations are used (this results from phonologic analysis
within the same language to match various local accents for words that are
generally recognized the same), and nobody seems to agree about a "correct"
orthography for the actual phonetic (so noting the actual phonetic will be
overkill, and useless for many people, as it is not essential for the
language and will be wrong within several accents, as various people are
using the same phonologic phoneme with different phonetic realizations,
which also varies according to context including some forms of phonetic
mutation of initial and final phonemes).
Some languages are also spoken using so strong mutations that they are
written orthographically, but this is much less common than languages that
don't note these mutations, as it helps recognizing the written radicals,
i.e. its intended semantic (contextual phonetic mutations are just implied
by the reader and knowledge of the language phonology, that offers other
unwritten hints to exhibit the semantic, including tonality and rhythm but
also gesture and facial expressions).
The beauty of the written text is that it allows the reader to choose the
way it reads it, according to what he feels, but text still remains linked
to denote the semantic rather than those subjective variations left as a
freedom for the readers (or singers...). Remember the well known exercise
given to candidate actors: trying to say exactly the same words, but
saying/singing them in many different ways to express various feelings. It
certainly affects the phonetic, sometimes also the common phonology; anyway
these are the same words, theoretically meaning the same thing semantically,
written identically, but with the additional interpretation expressed by the
reader, some other meaning may appear if the actor or song interpret is
This archive was generated by hypermail 2.1.5 : Tue Aug 21 2007 - 23:57:08 CDT