Sunday, April 26, 2009

Seal Wars: The Unsolved Indus Valley Mystery

Since the accidental discovery of the lost world of Harappa and Mohenjo-Daro, we have found ourselves beset by countless unanswered questions: Who were the people? Where did they come from? What language did they speak? Where did they disappear? Did they have anything to do with the Aryans? Some of these answers, if not all, can be hoped to be found by deciphering the numerous seals with engraved hieroglyphics excavated from the Indus Valley ruins. About 6 years back, a team of linguists/historians came up with the somewhat disappointing conclusion that the symbols did not represent any language at all and were merely a part of a "nonlinguistic sign system." Today, however, the media is abuzz with the work of Rao et al, which recently appeared in Science, where the team has sought mathematical justification based on an information theoretic approach to support their hypothesis that the Indus Valley hieroglyphics indeed represent a natural language!

The team looked at conditional entropies of 5 natural linguistic systems (including English, old Tamil, and Sumerian) and 4 nonlinguistic ones (including DNA and protein sequences!) They concluded that the conditional entropy of Indus Valley inscriptions is akin to those of the tested linguistic systems. The publication has sparked off a flurry of news reports with catchy headings (some of which seem to give off the impression that the script will be deciphered by early tomorrow morning!) Well, the results do kindle hope! Indeed I was quite excited about it until I came across this speedy rebuttal by Farmer et al, the persistent propounders of the earlier theory suggesting the nonlinguistic nature of the Indus Valley script! Farmer and his team cast doubt on the reliability of conditional entropy as a statistical measure in the world of linguistics. (Linguists, what say you all on that?) It appears that they will come up with a more detailed, full-length rebuttal some time in the near future. It does seem like this battle will wage on, and I'll stay tuned for further updates.

There was a time, many years ago, when my dream was to become an archaeologist who would one day crack the Indus Valley code. (To be honest, back then I was terribly addicted to mystery novels. Archaeology then seemed like the closest real-life approximation to a career as a detective! Well, Sherlock Holmes is fiction, Harappa ain't!) As you can see, that childhood dream was forgotten as I eventually chose a career in engineering. But this current series of events has reignited that spark. Well, I don't mean to say I'm going out for digs now! But I sure will keep looking out for what others have dugg instead. And somewhere inside I'll keep hoping that someday we too will find our Rosetta stone! :-)


  1. AMEN for your Rosetta stone !!! :)

    "Seal Wars: The Unsolved Indus Valley Mystery" ... for a moment (few moments, actually) i thought: wow, that's a whole new kind of mystery !!

    Though i wish all the people working to understand our history better all the luck in the world, i don't think it'll change anything significant.
    Any self-evolving system should ideally forget its initial condition after a while. And nature/life is as ideal as it can get. So, even if we actually crack the mystery, i guess it'll just be a happy feeling, the satisfaction of knowing.
    But then, it's better to know than not to know. So... fingers crossed !! :)

  2. hmm... interesting!

  3. Conditional entropy :-) .... it sure sounds like something you could get into .. even now..
    although, unless someone could conclusively prove that linguistic systems necessarily have to have a certain type of conditional entropy, it's really not going to be a convincing argument.

  4. Are these scratches on seals forerunners to letters?
    With new tools our understanding now betters...
    The seals' closeness to language...
    With smart math we can now gauge!
    Though the old-schooler his misgivings still mutters!

  5. Praveen3:07 PM

    I wish the truth comes out at the end of the day. Right now, it seems more like the fight between big egos, which may lead to complete deviation from the main objective. The "detective" task should not be just to fight "only" over the question of whether Indus people were "non-literate" or "literate", but to really decipher what those symbols meant - after all, an urban civilization cannot last for over 2000 years based on some random symbols.

  6. very interesting and informative!! Nice to know what other career you would have landed in!! ahaa! detective Joyita.. a.k.a JD ;)

  7. nice read ...looks like you are going to give Lara Croft a run for the 'archaeological' treasure

  8. @adarsh, PK: As PK said, it's more like a search for the truth. And it makes a lot of sense because it might change history (rather our knowledge of it) by leaps and bounds.
    @AP: hmmmmmmmm
    @Addie: Haha.. one more reason to take a class on Information theory now. :-)
    @Arvind: That was one hell of a limerick!!
    @Neha: Walking down memory lane, guess who was the detective in my debut play "Defective Detective" ??
    @Subhankar: "Archeology" as the Americans would say it. :-)

  9. How many careers are you planning to juggle simultaneously?