Computer's artificial intelligence managed to uncover an ancient script secrets. 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. But few linguistic remains passed the test of time. More than 1,500 unique inscriptions from fragments of pottery, tablets and seals were found by archaeologists. The longest inscription is just 27 signs long. Computer analysis of symbols used 4,000 years ago by a lost Indus Valley civilization suggests they represent a spoken language. At first linguists thought the symbols were just nice pictures.
"The underlying grammatical structure seems similar to what’s found in many languages," said University of Washington computer scientist Rajesh Rao.
In 1877, British archaeologist Alexander Cunningham hinted that the Indus script preceded modern-day Brahmic scripts, used from Central to Southeast Asia. Other researchers disagreed.
Rao, who is a machine learning specialist, read about the Indus script in high school. He decided to analyse the script while in Inda and try to solve it.
"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.
The input sequeces for the compuer analysis belong to four spoken languages: ancient Sumerian, Sanskrit, Old Tamil and modern English. The theam also used for input data 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.
The output that computer delivered consisted in grammatical rules based on patterns of symbol arrangement. These proved to be moderately ordered, just like spoken languages. Unfortunately the program couldn't decipher the meaning of the script.
"It’s a useful paper," said University of Helsinki archaeologist Asko Parpola, an authority on Indus scripts, "but it doesn’t really further our understanding of the script."
The analysis already performed is a step ahead in understanding the Indus script grammar and finally its meaning.
"It’s only recently that archaeologists have started to apply computational approaches in a rigid manner," said Rao.
