So far, we have gotten five separate queries about a list of counting words supposedly from an American Indian language. Though the words
do not bear any resemblance to any Native American language we know of, they do bear an interesting resemblance to each
other. Here they are in the five versions that were sent to us (collected in Pennsylvania, Indiana, Alabama, Arkansas, and Kentucky--
all from white Americans who were "taught how to count in Indian" as children in the early 1900's):
This almost certainly is not the counting system from an actual Native American language. For one thing, there really was not any single Native American language
that was natively spoken in all five of those far-flung locations. For another, the numeral systems of Native American languages have been very extensively
recorded, and this number set does not match any of the known language families of Native North America. It is possible that the numbers might
have come from an African language, since speakers of the same African language were often separated and sent to different states as slaves. A Native American
connection is also possible in that case, since African-Americans and Native Americans frequently intermarried. It's also possible that the words came from
a fictional source, like the made-up lyrics "hana mana ganda" in Peter Pan or the faux-African lyrics "oo-ee oo-ah-ah ting-tang walla-walla bing-bang"
in the song "Witch Doctor."
In my opinion, though, these numbers are probably part of a counting-out rhyme or counting game. If you look at the most complete wordset (from
Pennsylvania,) there's a very strong pattern to these words: after 3, every even number ends in "many" (making the words more rhythmic to
recite,) and every odd number is a combination of the numbers immediately before and after it, minus the "many."
(cutlong = cut + long, longco = long + co, etc.) The odds of that ever happening in the number system of a natural language seem
close to zero to me, but it makes total sense as wordplay. Furthermore, the fact that the same word elements are used in all three versions but sometimes
refer to different numbers (a variant of cutlong means "three" in one list but "five" in another; the word longco means "five" in
two lists but "seven" in two others; the element ben appears at "five" in one list but "fifteen" in another, etc.) It would be very, very
unusual for the meaning of a word to change from "three" to "five" over time in a real language!
All of this only leads to an even more mystifying question, though: if this is a counting rhyme, where did it come from?
It must have been fairly well known since people in five different states sent copies of it to us; yet we we've been unable to find hide nor hair
of it online. And except for the rhythm, it doesn't strongly resemble other well-known counting-out rhymes like
"eeny, meeny, miney, moe," "intery, mintery, cuttery, corn," "on-ery, or-ery, ick-ery, ann," "hickory, dickory, dock,"
or (one from my own childhood) "eeny, meeny, pepsameeny, oopah, popsameeny." Neither does it resemble any of the old European counting
rhymes of the "yan, tyan, tethera, methera, pimp" or "een, teen, tether, fether, fitz" varieties
(which are also often mistaken for American Indian numbers) -- you can see a good list
of counting rhymes of Great Britain here. And the only counting-out
rhyme we've ever heard of being used specifically by Native Americans is the Passamaquoddy and Penobscot children's rhyme "oni, kibi, lawes, akles, untip."
So in the end, we have some good ideas about where this way of counting didn't come from, but where it did come from, we're still
scratching our heads over. If anyone out there knows another version of this count or has some knowledge about its origins,
please drop us a line and we'd be glad to add it to this page.