Whenever I read about the Indus Valley civilization, I am reminded of how their script remains undecipherable [1]. Do we need a lucky discovery like the Rosetta Stone was for Egyptian hieroglyphs? If we find a long enough passage, can we do some kind of frequency analysis to pick out common terms (food, water etc.)?
One thing that's often missing with mentions of the rosetta stone (and I'm not assuming you don't know this, just didn't mention it) is how ridiculously, insanely unlikely that find was.
Each of these things is incredibly unlikely, each more unlikely than the last:
- it was made at all! these were never common. multilingual cultures have always been common or even the norm but for various reasons decrees and monuments tend not to be.
- it's in languages that remained in wide use after, and of interest to a broad swathe of historians and linguists today.
- it was preserved in a usable form; it was moved and reused as building materials, how many times must it have been almost destroyed or buried? that's the normal end for similar artifacts.
Without a similarly astoundingly lucky combination of events yeah we're stuck with just grinding away with frequency analysis and similar techniques. I'm not up to date on the indus valley script but I think a stumbling block there is you have to have at least a good guess what you're looking at before frequency analysis starts to work.
We don't know what the "atoms" of that script are or what its purpose was. Are the marks similar to words, or letters, or like strokes in chinese, or like dots in braille? Is it even a writing system for a language? it could be musical notation, court shorthand, stage directions for religious ceremonies, complex tally marks for trade inventory, etc. It's probably not but that indicates the problem space a little.
Moreover the first attempts to decipher hieroglyphs were in the 14th century, there was an enormous body of examples, and by 1798 they had been connected to Coptic (a language partially descended for the ancient Egyptian language) and there had been great strides in deciphering them.
A rosetta stone for the Indus script might not be enough.
To answer your last question, we would need to have a good idea of what the language sounded like. Linear B was deciphered, we think, by doing frequency analysis but using the Greek language. so it was a script for the Greek language in Crete.
It seems impossible to decipher the script without a sound of the language.
So even if you could translate it perfectly, there would be nothing to read.
They might have had a massive literature, but written on something that is all dust now. There was an Asian tradition of scratching letters in palm leaves stitched together.
[1]: https://www.worldhistory.org/Indus_Script/