Acquiring numerals

[Richard W]

Has the origin of 'one' even been figured out yet?

It's a failed sound change, apparently successful in vowel-initial words in some dialects. The spellings of whole and whore suggest that it made good inroads into standard English before failing. Now, as to why it should be retained in this word, I don't know.

[Lambuzhao]

Regarding how languages acquire higher numbers from within (native constructions) + without (L2 influence, extra-linguistic factors), consider the case of the Oksapmin.

The Oksapmin are a tribe in Papua New Guinea. They are one of a few languages that utilize "whole body" counting system (Cf. Torrres Strait Islanders), rather than a "numerical base" language.
See here:
http://www.culturecognition.com/tit-fu
http://www.jimrodslz.org/cultural_math/australia.html

Especially interesting is how they wrap numbers greater than 27 around the second wrist.



Geoffrey Saxe and Indigo Esmonde worked with the Oksapmin for years and wrote this study of Tradestores as the locus for mathematical exchanges in Oksapmin, Melanasian Pidgin and English:

Making change in Oksapmin tradestores: A study of shifting practices of quantification under conditions of rapid shift towards a cash economy
http://spjp.massey.ac.nz/issues/2004-v15/v15_saxe.pdf


It seems here that, at least for the speakers of Oksapmin, two interesting factors seem to be molding the native tally system.
One is the adopted system of pounds to shillings. The Oksapmin have developed a preferential system of the 20 shillings to one pound ratio, and carried it over to the newer currency, electing to use bills and coins that preserve that ratio.

Also, Cassowaries, butterflies, boars, other animals or symbols appear on coins or notes. Oksapmin speakers use these as readily accessible pars pro toto quantifying markers. This further reduces mathematical quantities that would become greater or more cumbersome to compute in the native tallying system.

[TaylorS]

PIE *septm seems suspiciously similar to Proto-Semitic *sabʕatum. I wonder if PIE borrowed from a Semitic source.

[Nannalu]

In the Berber languages, or at least a majority of them, only the numerals one and two (Shenwa. iğ (f. ict) & sen (f. senat̠). Everything higher is a borrowing from Arabic.

[jal]

PIE *septm seems suspiciously similar to Proto-Semitic *sabʕatum. I wonder if PIE borrowed from a Semitic source.

You call that "suspiciously similar"? That's probably because of the Basque monks I guess...


JAL

[Nortaneous]

It's a failed sound change, apparently successful in vowel-initial words in some dialects.

No. wh > h before rounded vowels.

[hwhatting]

PIE *septm seems suspiciously similar to Proto-Semitic *sabʕatum. I wonder if PIE borrowed from a Semitic source.

You call that "suspiciously similar"? That's probably because of the Basque monks I guess...

YMMV concerning the suspiciousness of similarity, but to make that comparison is certainly not crackpottery, that idea has come up before in serious linguistics. There's at least a bit of a debate among IEanists on the direction of borrowing (PIE -> Semitic or vice versa, or from a third language), which also includes the number "6".

[jmcd]

It's a failed sound change, apparently successful in vowel-initial words in some dialects.

No. wh > h before rounded vowels.

That's because the other words Richard mentioned aren't related:
'whore' is due to a sound change wh > h before rounded vowels whereas 'one' and the likes of 'open [wʌpən]' is due to a sound change o>wʌ at the start of words.

[KathTheDragon]

'open [wʌpən]'

Who the fuck says that?

[hwhatting]

'whore' is due to a sound change wh > h before rounded vowels.

Wrong example. The "wh" in "whore" is an unetymological spelling, the word goes back to Old & Middle English hore. The process you describe took place, e,g. in the forms of "who".

[Nortaneous]

PIE *septm seems suspiciously similar to Proto-Semitic *sabʕatum. I wonder if PIE borrowed from a Semitic source.

You call that "suspiciously similar"? That's probably because of the Basque monks I guess...

YMMV concerning the suspiciousness of similarity, but to make that comparison is certainly not crackpottery, that idea has come up before in serious linguistics. There's at least a bit of a debate among IEanists on the direction of borrowing (PIE -> Semitic or vice versa, or from a third language), which also includes the number "6".

Yeah, the consonants are almost identical, and vowels do weird shit in that part of the world, so why not?

How could it have gone from PIE to Semitic, given the ʕ?

[hwhatting]

How could it have gone from PIE to Semitic, given the ʕ?

Probably not from late PIE, but perhaps from some Pre-PIE... in fact, I don't know, and it's a long time since I read about this. Personally, I'm not of the opinion that Semitic and PIE were neighbours, so there anyway needs to have been an intermediary language in any borrowing, whatever the direction.

[CatDoom]

Probably not from late PIE, but perhaps from some Pre-PIE... in fact, I don't know, and it's a long time since I read about this. Personally, I'm not of the opinion that Semitic and PIE were neighbours, so there anyway needs to have been an intermediary language in any borrowing, whatever the direction.

Ooh, that might be an interesting bit of speculative pseudo-historical conlanging; a "Septumian" language family, spoken somewhere between the Indo-European and Semitic homelands (probably somewhere in upper Mesopotamia, north of Sumeria), from whence both families borrowed the word for "seven."

[KathTheDragon]

Do we know what the Etruscan word for 'seven' was?

[Richard W]

Do we know what the Etruscan word for 'seven' was?

Etruscan: 7 = semph (uncertain, though), and either 4= śa and 6 = huth or 6 = śa and 4 = huth
Basque: 7 = zazpi and 6 = sei
Egyptian: 7 = *sáfḫaw and 6 = *yVssáw (-w marks masculine plural, V = uncertain vowel)
Georgian: 7= švidi and 6 = ekvsi (not the best of matches)

There are hints of '6' and '7' words like these through much of Afroasiatic. Finno-Ugrian does not evidence these words, and I don't see anything in NE or NW Caucasian.

[CatDoom]

Based on a quick internet search, the best candidate for a cognate I've been able to find in the region outside of the Indo-European and Semitic families is *ʃwidi, which is reconstructed for Proto-Kartvelian. It's got a sibilant, a labial, and a coronal stop in the right order, though that's not terribly strong evidence.

[Richard W]

'open [wʌpən]'

Who says that?

Onions' dictionary implies it's to be found in a 'westerly' dialect, and records /wʌk/ and /wʌts/ for <oak> and <oats>. <Whore> and <whole> (cf. Scots hale and German heil) indicate that the change wasn't necessarily just word-initial. The <w> of whole has been pronounced, and the spelling with <w> appeared in the 16th century.

[TaylorS]

Probably not from late PIE, but perhaps from some Pre-PIE... in fact, I don't know, and it's a long time since I read about this. Personally, I'm not of the opinion that Semitic and PIE were neighbours, so there anyway needs to have been an intermediary language in any borrowing, whatever the direction.

Ooh, that might be an interesting bit of speculative pseudo-historical conlanging; a "Septumian" language family, spoken somewhere between the Indo-European and Semitic homelands (probably somewhere in upper Mesopotamia, north of Sumeria), from whence both families borrowed the word for "seven."

What is 7 in Proto-NE-Caucasian and in Hurrian?

[vokzhen]

Probably not from late PIE, but perhaps from some Pre-PIE... in fact, I don't know, and it's a long time since I read about this. Personally, I'm not of the opinion that Semitic and PIE were neighbours, so there anyway needs to have been an intermediary language in any borrowing, whatever the direction.

Ooh, that might be an interesting bit of speculative pseudo-historical conlanging; a "Septumian" language family, spoken somewhere between the Indo-European and Semitic homelands (probably somewhere in upper Mesopotamia, north of Sumeria), from whence both families borrowed the word for "seven."

What is 7 in Proto-NE-Caucasian and in Hurrian?

According to Wikipedia, PNEC is *u̯ərδ, Hurrian is šinti. Six is *renɬə- and šeše. Though if I'm not mistaken, while NEC isn't as bad, you take reconstructed NEC/NWC words with a mountain of salt. Hurrian, though, seems to hint that it may have gained 6 and 7 in similar fashion to others.
EDIT: Woops, should point out. Hurrian š is probably just /s/.

[Richard W]

One reason for borrowing largish number words is a change of base. Khmer has changed from a mix of bases 5 and 20 to decimal, and therefore the words for the multiples of 10 changed. In this case, the new words were borrowed rather than being formed as calques. English vigesimal numbers are formed using a single borrowed element, score.