Asked by Luis Andrade
The rumor started by Farmer and colleagues, to a great extent, seems to have settled down. I do not think he has really convinced many Indus specialists.
I may be partial to speak about the present stand on this matter, but in my opinion none of the arguments of Farmer & co. holds good -- the Indus script is definitely logo-syllabic writing, which operates with partially phonetic partially ideographic pictograms.
We have done an extensive study of the syntactic structure of the Indus script and we are confident that the Indus script has a definite, well defined structure and a firm grammar which also allows a degree of flexibility typical of linguistic writing. Farmer is clearly off the mark when he claims that it is not structured. Harappans were certainly literate. However, several models to attempt a semantic fit to later languages (Dravidian and Sanskrit), pure numbers or some other form have been shown to be inconsistent with the syntactic structure. So we do not have a self-consistent semantic model but the syntax is clear. So one can assign meanings to sign based on the syntax and write in Harappan, but it will not be what the Harappans meant.
Above: Inscribed storage jar, Harappa. The paper referenced is http://www.safarmer.com/fsw2.pdf.