It looks like you're using an Ad Blocker.

Please white-list or disable in your ad-blocking tool.

Thank you.


Some features of ATS will be disabled while you continue to use an ad-blocker.


Artificial Intelligence Cracks 4,000-Year-Old Mystery

page: 1

log in


posted on Apr, 24 2009 @ 12:06 AM

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.

The Indus script, used between 2,600 and 1,900 B.C. in what is now eastern Pakistan and northwest India, belonged to a civilization as sophisticated as its Mesopotamian and Egyptian contemporaries. However, it left fewer linguistic remains. Archaeologists have uncovered about 1,500 unique inscriptions from fragments of pottery, tablets and seals. The longest inscription is just 27 signs long.

In 1877, British archaeologist Alexander Cunningham hypothesized that the Indus script was a forerunner of modern-day Brahmic scripts, used from Central to Southeast Asia. Other researchers disagreed. Fueled by scores of competing and ultimately unsuccessful attempts to decipher the script, that contentious state of affairs has persisted to the present.

Among the languages linked to the mysterious script are Chinese Lolo, Sumerian, Egyptian, Dravidian, Indo-Aryan, Old Slavic, even Easter Island — and, finally, no language at all. In 2004, linguist Steve Farmer published a paper asserting that the Indus script was nothing more than political and religious symbols. It was a controversial notion, but not an unpopular one.

Rao, a machine learning specialist who read about the Indus script in high school and decided to apply his expertise to the script while on sabbatical in Inda, may have solved the language-versus-symbol question, if not the script itself.

"One of the main questions in machine learning is how to generalize rules from a limited amount of data," said Rao. "Even though we can't read it, we can look at the patterns and get the underlying grammatical structure."

Rao's team used pattern-analyzing software running what's known as a Markov model, a computational tool used to map system dynamics.

They fed the program sequences of four spoken languages: ancient Sumerian, Sanskrit and Old Tamil, as well as modern English. Then they gave it samples of four non-spoken communication systems: human DNA, Fortran, bacterial protein sequences and an artificial language.

The program calculated the level of order present in each language. Non-spoken languages were either highly ordered, with symbols and structures following each other in unvarying ways, or utterly chaotic. Spoken languages fell in the middle.

When they seeded the program with fragments of Indus script, it returned with grammatical rules based on patterns of symbol arrangement. These proved to be moderately ordered, just like spoken languages.

This is a great leap forward to understand the past

posted on Apr, 24 2009 @ 12:18 AM
This is amazing no?

I mean i'm very interested in what they thought was so important that they must write it down.

I hope to hear alot more out of this. India has a lot of strange things in it.

posted on Apr, 24 2009 @ 12:22 AM
Very interesting read; I will be following the story (or at least trying to).

We keep pushing the circa on civilization further and further back. I wonder what our picture of the past will be in 50 years?

In 1000 years?

posted on Apr, 24 2009 @ 12:33 AM
reply to post by spines

While I agree with you here, the real breakthrough in this article is not the timeframe, since that was already known, but that a previously undecipherable language was cracked using new computer software. I'd love to see them apply this to other as-yet unreadable languages. I know there's at least one South American language, and then Linear B. (or maybe A; one we can read, one we can't, I mix it up all the time) I'm sure theres tons of other languages we can't read yet, but those are just what I can remember.

posted on Apr, 24 2009 @ 01:11 AM
reply to post by DragonsDemesne

Unless I am mistaken, they haven't 'read' this language yet. It has been ran through a program which shows it has a similar variance of order, but does it say they have figured any of it out yet?

I wonder if a computer program could be relied on to accuratley translate a previously non-translated language (that is believed by some to not be a language at all) into something understandable/eye-opening.

If it already is working off the assumption of variance of order similar to that of known language...could it not find something that isn't there?

posted on Apr, 24 2009 @ 12:42 PM
reply to post by imd12c4funn

I want my universal translator... NOW!.. J/k

This is an awesome find. I have always believed our history is much older than we are led to believe.

posted on Apr, 24 2009 @ 11:59 PM

Originally posted by spines
reply to post by DragonsDemesne

Unless I am mistaken, they haven't 'read' this language yet. It has been ran through a program which shows it has a similar variance of order, but does it say they have figured any of it out yet?

Yeah, that's what I read from this as well.

It seems as though the machine has looked at the symbols which archeologists have concluded are simply insignia, or political symbols, and has concluded that it is in fact a language, not just symbols.

... but it doesn't look like the machine has actually made sense of the language... all it's done here is said "Yep, it's a language."

Still though, from a standpoint of an echo-generation roboticist, I still find it interesting.

posted on Apr, 25 2009 @ 02:31 AM
When you take a step out into thin air how can you call that progress. When you merely reaffirm a possibity of a pattern that might or might not be that of an unknown language. It seemed to be more a critique of the test rather than any useful gain.. I don't see the next step. How can alineing with different languages open the code? How do the invisible codes interact?

posted on Apr, 26 2009 @ 12:53 AM
I agree that this is an interesting find but I don't see it as ground breaking. Someone said just before me here, the computer essentially gave a yes out of a yes or no. It returned a 1 instead of a 0. Or whatever. I don't speak binary.

Now, if this "AI" ends up deciphering this language into something we can understand, i'll say that's amazing. I'll post a youtube video with me in my underwear wearing a monkey mask on my head dancing around to "Head" by Prince if that happens.

posted on Apr, 30 2009 @ 03:02 AM
Seems I misunderstood then! Hmm...
In that case, all they actually proved was that this is a language, rather than artwork or something. I suppose that's somewhat useful, as you don't want to try and decipher artwork, but I thought there was more to the story than that.

posted on Apr, 30 2009 @ 06:13 AM
maybe they can translate the book of death and life?

which shows some coooooooool lil... pics ;-)

posted on Apr, 30 2009 @ 06:46 AM
reply to post by imd12c4funn

DAMN... I tried to post this story with the same source 3 or 4 times the other day and every time I hit post the whole thing disappeared... I complained to management and never got a response back.

new topics

top topics


log in