Harappan 'non-texts'?

[Guillaume JACQUES]

Sorry to continue discussion on this tread that is only indirectly relevant
to the Indus Script problem, but I would make a few comments :

>One final point.   With so-called mono-syllabic languages like
>Chinese, there is only a limited number of word shapes that can be
>generated.

I should say that homonymy in chinese characters is the rather recent.
of phonetic erotion and increased word-compounding.
Much progress has been made in the last years in Archaic chinese
reconstruction, which reveals chinese was basically a bisyllabic
language, although it has been traditionnally assumed that one
character = one syllable (those who want more information on that
can write a private email to me, since this issue is not relevant
in an Indological list). It is not necessary to make such an equation
for the indus script either, even if it worked the same way the chinese
script did.
However, if the Indus Script had been comparable in complexity with
Chinese, we should have expected a much higher number of signs. Even
if you do not take regional and diachronic variants of the Shang oracle
bone script (that are, I concede, longer than IS inscriptions), you
still find more than 1500 deciphered characters, had many more that are
impossible to decipher. With only 400 signs, IS script writing cannot
have been a purely logographic system.
Finally, on the homogeneity of the IS : the signs of the IS are quite
simpler than that of egyptian, chinese or akkadian. There was thus
a more limited room for evolution - especially for a script that
barely lasted for a millenium.
Besides, one may suppose (please correct me if there are archeological
evidence against my claim) that these were always written on the same
medium, unlike chinese that was written on scapula, turtle shells, bamboo,
silk, clay (in the moulds used in casting bronze inscriptions) which
in my opinion contributed in some way to the great variety of pre-imperial
chinese scripts.

Guillaume