Peru: Vanished Inca may have used binary code language

[Harold Schiffman]

Vanished Inca may have used binary code language
Rossella Lorenzi, Discovery News
 Thursday, 3 July 2003



The vanished Inca civilisation of the Andes, long thought to have no
writing, invented a seven-bit binary code to store information more
than 500 years before the invention of the computer, argues an
American anthropologist. Begun in the Andean highlands of Chile and
Colombia around 1200 AD, the Inca ruled the largest empire on Earth by
the time their last emperor, Atahualpa, was garroted by Spanish
conquistadors in 1533. But the voice of the Inca has never been heard;
it has long been considered the only major Bronze Age civilisation
without a written language. Professor Gary Urton, an anthropologist at
Harvard University in Boston and a specialist in Pre-Columbian
studies, is now challenging that assumption in a new book, Signs of
the Inka Khipu. He argues the Incas had a written language disguised
in the form of elaborate knotted strings known as khipu.

Derived from the word for 'knot' in the Quechuan language of the Inca
- still widely spoken in the the Andes highlands - these decorative
objects consist of one main cord to which are attached several pendant
strings. These, which can carry subsidiary or tertiary strings, bear
clusters of knots. In 1923, science historian L. Leland Locke proved
that the khipu were more than decorative; they were a sort of textile
abacus, their knots used to record calculations.  But Locke's rules
decoded only a small percentage of the existing 600 khipu that
survived the Spanish destruction, failing to take into account even
one-half of the total information encoded in them, Urton said.

"The most convincing evidence for this three-dimensional writing
system is the khipu. Their complexity would have been unnecessary if
they were just mnemonic devices understood only by their makers,"
Urton said. In the book, published by the University of Texas Press,
Urton has for the first time systematically analyses the khipu's
essential elements. It emerged that there are seven points in making a
khipu, where the maker chooses between two possibilities. The binary
choices include the type of material (cotton or wool), the spin and
ply direction of the string, the direction (forward or reversed) of
the knot, and so on. A strict seven-bit code would produce 128
permutations (or 27). But Urton calculates that there were 24 possible
colours that could be used in khipu making.

Thus the khipu code can store 1,536 units of information (26,
multiplied by 24). This is comparable to the Sumerian cuneiform
symbols used between 1000 AD and 1500 AD, and more than twice the
Egyptian and Maya hieroglyphic signs. A definitive way to crack the
intractable code would be the discovery of what Urton calls a 'Rosetta
khipu', something similar to the deciphering of Egyptian hieroglyphics
from the Rosetta stone: a basalt slab unearthed in Egypt in 1799 with
text in Greek and Egyptian hieroglyphs, allowing linguists to decode
the language.

"We have a sizeable number of khipu, and we have about a dozen
documents that are written up from the khipu. What we don't have yet
is a match between a document and a khipu," said Urton. While
searching, Urton is attacking the khipu code with 21st century
technology, creating a database packed with any possible data on each
khipu: length of the main string, number of pendants, details on the
knots, spin, ply of each string, and so on, in order to search for
common patterns. "Just 10 days ago, I discovered three khipu that
share part of the information," he said. "This is a pretty strong
evidence that they were not made by single people. On the contrary,
there was a shared code." "It is an interesting study," said Professor
Laura Laurencich-Minelli, a specilaist in Pre-Columbian studies at
Bologna University and author of several books on the Inca and the
khipu. "Certainly, khipu were much more than mnemonic devices."

http://www.abc.net.au/science/news/stories/2003/893964.htm

-- 
**************************************
N.b.: Listing on the lgpolicy-list is merely intended as a service to
its members
and implies neither approval, confirmation nor agreement by the owner
or sponsor of
the list as to the veracity of a message's contents. Members who
disagree with a
message are encouraged to post a rebuttal. (H. Schiffman, Moderator)
*******************************************