Indus Valley Script: Artificial Intelligence Says It's a Language

"Artificial Intelligence Cracks 4,000-Year-Old Mystery"
Wired Science (April 23, 2009)

"An ancient script that's defied generations of archaeologists has yielded some of its secrets to artificially intelligent computers.

"Computational analysis of symbols used 4,000 years ago by a long-lost Indus Valley civilization suggests they represent a spoken language. Some frustrated linguists thought the symbols were merely pretty pictures.

" 'The underlying grammatical structure seems similar to what's found in many languages,' said University of Washington computer scientist Rajesh Rao...."


(from Wired Science, used without permission)

The Wired Science title is a little optimistic, I think. Unless the question is whether or not the Indus Valley civilization script is language, or "nothing more than political and religious symbols." That was the opinion of a linguist, in a published paper.

I can see the linguists point, in a way. When "generations of archaeologists" haven't been able to decipher a language, it must be tempting to say that the language isn't a language.

The article gives a very brief overview of how researchers gave an artificial intelligence system examples of spoken languages, programming language, and an artificial language - had it study that, and then showed it Indus Valley script.

We don't know what the script means, but it's organized to a degree that spoken languages are.

I think this is an example of how archeology can use today's information technology to analyze information - faster and (perhaps) more effectively than a person can.

Indus Valley Script - Not Much Data, and No Rosetta Stone

There isn't much Indus Valley script to study: the longest example is 27 signs long.

Star Trek's Universal Translator notwithstanding, trying to figure out what another language means is hard work - and nearly impossible if there isn't some frame of reference. Egyptian hieroglyphs were as untranslatable as Indus Valley script: until Napoleon's army found the Rosetta Stone. That slab of rock had the same message in three languages: Greek, demotic and hieroglyphs. That gave linguists something to work with. Knowing what the hieroglyphs said, they were able to figure out how the information was coded.

There are better explanations than that, but I'll let it stand.

Something I Heard in Class - - -

The Indus Valley script reminds me of a situation a professor mentioned, decades back. Somewhere, in the Middle East, I think, archeologists came upon a library of clay tablets. Generally, with finds like this, the tablets are scattered and broken: not at all in the order they were supposed to be in. Think of the contents of a filing cabinet, shaken out onto the floor, then trying to make sense of the mess.

In this case, the library had been abandoned but not disturbed. The shelves had collapsed, but they were in their original order, with the tablets stacked exactly as an archivist had left them. Some of the tablets were cracked, but otherwise complete.

It was a wealth of information from ancient times.

Just one problem.

The tablets, every last one of them, was in an unknown language.