Deciphering Incan khipu

[Vin McLellan]

Boston Globe reporter Gareth Cook <cook at globe.com> was awarded the 
2005 Pulitzer Prize for Explanatory Journalism for "explaining, with 
clarity and humanity, the complex scientific and ethical dimensions 
of stem cell research."  He's an unusually talented writer.

_Vin

------------------------------------

R. Hirschfeld <ray at unipay.nl> quoted Gareth Cook, who wrote:

<snip>

>I am wondering if you know anyone who might be able to help me with this
>
>I wrote a while ago about a fascinating project focused on 
>deciphering the Incan khipu (see below). The basic idea is that they 
>are collections of knots used in the Incan empire to record 
>information. It is known that some of them contain numbers, perhaps 
>recording census data or tax information for the empire. But some 
>believe that the knots records language -- perhaps histories or 
>other narratives. Cracking this code would be hugely important, not 
>to mention interesting, because it would open up the still very 
>mysterious Incan empire the same way that ancient Egypt has been opened up.
>
>All this is a rather long-winded prelude to my question, which is 
>whether there are people out there who are working on computational 
>techniques to decipher ancient scripts, not necessarily the khipu 
>problem. I am thinking of doing a story on this. Any thoughts or 
>leads at all would be most appreciated. It would even be a help to 
>talk to someone who has done cryptography who could explain how the 
>ancient scripts problem would be similar to, and different from, the 
>problem of cracking a present-day encryption scheme.
>
>Let me know if you have any thoughts.
>
>Best,
>Gareth

----------------

SCHOLAR SEES STRANDS OF ANCIENT SECRETS

  Author: By Gareth Cook, Boston Globe
Date: 07/04/2003



CAMBRIDGE - For centuries, the mighty Incan empire has confounded researchers.

The Incas controlled territory up and down the spine of South 
America, with a sophisticated system of tributes and distribution 
that kept millions fed through the seasons. They built irrigation 
systems and stone temples in the clouds.

And yet they had no writing. For scholars, this has been like trying 
to imagine how the Romans could have administered their vast empire 
without written Latin.

Now, after more than a decade of fieldwork and research, a professor 
at Harvard University believes he has uncovered a language of binary 
code recorded in knotted strings - a writing system unlike virtually any other.

The strings are found on "khipus," ancient Incan objects that look 
something like mops. About 600 khipus (also spelled "quipu") survive 
in museums and private collections, and archeologists have long known 
that the elaborately knotted strings of some khipus recorded numbers 
like an abacus. Harvard's Gary Urton said the khipus contain a wealth 
of overlooked information hidden in their construction details, like 
the way the knots are tied - and that these could be the building 
blocks of a lost writing system which records the history, myths, and 
poetry of the Incas.

The theory has Incan scholars abuzz. The discovery of true Incan 
writing would revolutionize their field the same way that deciphering 
the Egyptian hieroglyphics or Mayan glyphs lifted a veil from those 
civilizations. But it also has broader interest because the khipus 
could constitute what is, to Western eyes, a very unorthodox writing 
system, using knots and strings in three dimensions instead of 
markings on a flat expanse of paper, clay, or stone.

"What makes this work so interesting is that what is being expressed 
is being conceptualized in such a different way than we 
conceptualize," said Sabine MacCormack, a historian of the Romans and 
the Incas who is a  professor at the University of Notre Dame. "This 
is about an expression of the human mind, the likes of which we don't 
have elsewhere."

The only way to prove Urton's theory correct would be to translate 
the khipus, which no one has yet done. In his new book, he proposes a 
new method for transcribing the knotted strings which he believes 
could lead to breakthroughs.

And his work, funded in part by a genius grant from the MacArthur 
Foundation, has helped fuel a resurgence of scholarly interest in 
khipus. Later this month, the Chilean Museum of Pre-Columbian Art in 
Santiago is opening the world's first exhibit dedicated to the khipu.

"We are on the cusp of a very hot period," said Frank Salomon, a 
professor of anthropology at the University of Wisconsin who has 
studied khipus extensively.

The khipu mystery dates to the early 16th century, when the Incas 
were conquered by Francisco Pizarro and the Spanish set about 
destroying their culture. The missionaries sent to South America 
tried to eliminate all touches of the old gods, including the strange 
stringed textiles that the Incas said held their histories.

The Spanish chroniclers often exaggerated, but they did record 
histories of tributes and other stories they said were "read" to them 
by khipukamayuq - or knot keepers - from strings of knots.

In 1923, researcher L. Leland Locke was able to show that many khipus 
recorded numbers like an abacus, with knots in positions representing 
the hundred's, ten's, or one's place. He concluded that khipus were 
an accounting tool and scholars largely lost interest.

Locke, however, missed many subtleties in the khipus, which could 
make them a richer tool for communication, said Urton, whose research 
was described in a recent issue of the journal Science, and whose new 
book is called "Signs of the Inka Khipu."

The attention to khipus has its roots in insights from Marcia and 
Robert Ascher, a husband-and-wife team who began an extensive survey 
and analysis of khipus in 1968, and on the observations of Bill 
Conklin, a textile specialist at the Textile Museum in Washington, 
D.C., who noticed that khipus were spun and tied in surprisingly 
complex and varied ways.

Urton is proposing a system for tackling the meaning of the knots. 
Each knot, Urton suggests, can be thought of as a series of 
decisions, such as whether to make it of cotton or wool, to tie the 
knot with a crossing string that begins in the upper left or the 
upper right, and to use string that is spun clockwise or counterclockwise.

Not all scholars are persuaded by Urton's ideas.

"I don't see that this proposal arises from the actuality of the 
khipus," said Marcia Ascher, an emerita professor of mathematics at 
Ithaca College. "I don't see it being shown to fit or explain any of them."

Using money from the National Science Foundation, Urton has 
undertaken a comprehensive project to record as many khipus as 
possible in great detail, including the binary information he says 
could be so important. He hopes to place it all in a single computer 
database and give access to other scholars and the public in the 
hopes that somebody will see ways to crack the code. He is being 
helped by Carrie Brezine, a weaver and database specialist who did 
her undergraduate thesis in mathematics.

Last week, Brezine brought in a printout of transcriptions taken from 
khipus found recently in a cave overlooking the Lake of the Condors 
in northern Peru.

As he sat in his office, surrounded by Andean textiles, he noticed 
long strings of numbers that were virtually identical on three of the 
khipus - an indication that information was being copied from one to 
another, the way medieval scribes copied books by hand.

"It was one of those eureka moments," he said with a boyish grin. 
"This is really cool."

----------



---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at metzdowd.com