[Basic] Learning languages

[vicky_molokh]

Originally Posted by Ketsuban

Hexadecimal and binary are abstract concepts; they're not used by any languages I know of. The languages (substituting "Latin" for "Roman") you mentioned all inherited their words, and in all three cases have no known analysis. (I did read a tentative attempt to derive the Indo-European numerals from other IE roots which were related to counting, but it was pretty speculative and I can't find it again now.)

I meant Roman numbers as opposed to the (modernised) Arabic numbers that we are currently using.

Originally Posted by Ketsuban

As for the specific examples:

• 万/萬 is borrowed from Chinese, and hence of Sino-Tibetan origin. Numbers higher than twenty are more likely to be derived either from numbers below twenty or from words describing enormity - an example is Greek myrios "infinite, countless, abundant" > "ten thousand".
• "Pint" is from Latin picta "painted" - you painted a mark on a vessel to indicate "this much". (Measurements don't indicate numbers, they indicate a quantity of something. A furlong was the length of a furrow; "acre" originally meant "field"; a poronkusema is how far a reindeer can go before it needs to ****.)

If measurement of quoantity something in units is not indicating numbers, then I don't know what is. I have 5 apples and 2½ soldiers and 0.3km/s of delta-V.

[malloyd]

Originally Posted by Ketsuban

Hexadecimal and binary are abstract concepts; they're not used by any languages I know of.

Systems for writing numerals aren't numbers in a languages sense - and some of them do use non-decimal bases. The Roman numbers are "unus duo tres quattor quinque..." not "I II III IV V..."

It is sort of odd there aren't languages that group words for larger quantities by dozens or scores rather than tens.

[vicky_molokh]

Oh, and there are languages that use binary numbers.
https://en.wikipedia.org/wiki/Kumbainggar_language (I'm pretty sure there was a scientific journal listed as a source when I first saw this; maybe it was another language)
http://www.ancient-origins.net/news-...-mathematician

[Ketsuban]

Gumbaynggir is a language which has been brought back from the dead, so it's not a great example. Mark Rosenfelder has produced an exhaustive list of numbers from 1-10 in languages from all over the world, and cites two sources - Dixon & Blake's Handbook of Australian Languages, published 1979, for the Gumbaynggir words for "one" and "two", and "W.E. Smuth 1948" for the principle behind the numbers three through six.

I can't find any W.E. Smuth (or similar), and the citation of Dixon & Blake is questioned by the Swadesh list collected by the Rosetta project and put into the public domain on the Internet Archive which suggests there's more to this story which was lost when the language all but died off.

one: arugurum
one: galugun
other: galugun
...
two: bulari
two: yulwari
...
you two: bulaː
you two: bulaːɲa
you two: yulwari

At any rate, it's a single data point versus many examples along the lines of Choctaw 5 = tahlapi ‘the first (hand) finished’; Klamath 8 ndan-ksahpta ‘three I have bent over’; Unalit 11 atkahakhtok ‘it goes down (to the feet)’; Shasta 20 tsec ‘man’ (considered as having 20 countable appendages).

e: The Mangareva example only applies to numbers between 10 and 80, which are more malleable in general. (Once you get above 20 you're already working with more than there are usable appendages on the human body.)

ee: I just realised how far off track we've gotten with this. I'm not going to discuss this anymore, in the interests of serving the actual topic.