Theiling Online    Sitemap    Conlang Mailing List HQ   

Re: Historical Sound Change & Numbers Puzzle

From:Joseph Fatula <joefatula@...>
Date:Sunday, March 11, 2007, 8:16
Alex Fink wrote:
> On Sun, 11 Mar 2007 12:47:22 +0900, Sanghyeon Seo <sanxiyn@...> wrote: > 2007/3/10, Joseph Fatula <joefatula@...>: > >> For those of you interested in language change over time, here's a >> descendant version of the interlinear I posted a few days ago: >> http://www.geocities.com/altyaltynalma/northwindandsun_nfolk.html >> >> And the original for comparison: >> http://www.geocities.com/altyaltynalma/northwindandsun_gn.html >> > > Hmm, no morphosyntactic change at all? >
There is some, but it just didn't happen to show up in this sample. I've got a further descendant that messes it up even more, and then it starts to really change how the system works.
>>> So if you've figured out how the one descends from the other (not too >>> complicated), here's a puzzle for you. The following numbers are in the >>> descendant language. They make a fairly standard base-10 number >>> system. Can you figure out how the ancestral number system worked? >>> >> I tried, but I can't! >> >> I guess 5=si is from ancestral se, and 10=neru is from ancestral nelru. >> Apart from that, I'm completely lost. Especially, >> >> 3=wec, 8=nel, 30=bornel, 50=borwec? >> >> Others, any idea? >> > > Alright, some thoughts (that I suspect give a fair amount away, so take this > as a spoiler warning).
For anyone else reading this, Alex isn't kidding about it being a spoiler... (spoiler below)
> > > > > > > > > > > > > > > > > > > > > > > Up through 10 we've got a nice tidy Fibonacci system with 1, 2, 3, 5, 8 > fundamental, and some changes in combination (the ones you mentioned plus > |p| < |cb| in 4). Trying the tens in this vein: >
Spot on so far...
> 20 |ceseru| = 13 + 5 + 2, beautifully. |ce| seems to be 13. >
"Ceseru" is from an original that's exactly as you explain it: 13 + 5 + 2.
> 30 |bornel| = 21 + 8 + 1: |bor| could be 21, and |nel| is there, but no > element for 1? Probably just dropped when saying 29 this way fell out of > currency. >
"Bornel" would be "bornelba", but you're right - once you've started saying "ceseru nelba" for 29, there's no reason to say the whole "bornelba" when the shorter "bornel" works just as well.
> 40 |mospa| = 34 + 5 + 1. |mos| seems to be 34. >
Your second message got it right - "mo" is what remains of the old word for 34. "Mospa" is 34 + 5 + 1.
> 50 |borwec| = 34 + 13 + 3. So yeah, it's bogglesome that there should be > another |bor| here; there's nothing in common between this expansion and the > one for 30. >
You're right that this is a bogglesome one. It comes from "moborwec", which (given the skill you've shown so far) you can clearly see is 34 + 21 + 3 = 58. It's not fifty, but it's one of the "fifty" numbers, and got connected to the number 50 when the old number system broke down. This is the only real speed bump in the list.
> 60 |consi| = 55 + 5, and we've found 55 |con|. >
Right on.
> 70 |contef| = 55 + 13 + 2. This and the next one suggest a _progressive_ > assimilation in |nC| clusters, so this |te| < |ce|. |ru| must get mogrified > into |f| somehow. >
"Contef" by itself would be just 68. We've got another "bornel" situation. The original would be "contefru" - the "f" is from "cef", the original word for 13.
> 80 |condor| = 55 + 21 + 3 + 1. |dor| < |bor| with assimilation, 3 + 1 dropped. >
Right again. "Condor" is short for "condorwepa", but again, there's not much reason to use the long form anymore.
> 90 |yuca| = 89 + 1. Difficult to segment out what 89 is... |yuc| maybe, but > the behaviour of the |cb| cluster isn't consistent with 4. >
It is indeed "yuc", and the "cb" cluster is reducing differently. One of the two numbers became treated as a single word before the "b" > "p" change, the other one happened after it.
> 100 |yushnel| = 89 + 8 + 3, perhaps involving |c| > |sh| before |n|. 3 drops. >
Absolutely right.
> 1000 |shnew| = 987 + 13, but a definite absence of |ce|, especially since > this word is in the Northwind story and has protoform |cnew|. The story > suggests it might have originally had a nonnumeric meaning like > 'everybody'...? > > Alex >
There sure is an absence of a 13 word here - "shnew" is from "cnew", as you mention, which is 987. Once you have a word that close in meaning to 1000, there isn't much need to add other numbers to it to be exact. And you're right about it not being a precise number to begin with. "Cnew" meant 987, but metaphorically meant something like "zillions" or "tons". You've solved the puzzle. I for one am impressed, and hope you enjoyed it.