From: Rob
Message: 34085
Date: 2004-09-08
--- In cybalist@yahoogroups.com, enlil@... wrote:
"Rob:
'At least for the purposes of playing Devil's Advocate, how is there
conclusively "no **dek- meaning 'ten'"? More on this below/later.'
Well, have you seen it? I haven't."
No, I have not. So (thus far) I conclude that there is no root
for "ten," only a root *dek- meaning "right (side)." Your hypothesis
that *dekm should be analyzed *de-km is interesting, but there again
seems to be no evidence of a prefixal *de- anywhere else in IE.
"Rob:
'However, what's the evidence for the supposed prefixal **de- "one"?'
"Altaic seems to show a similar word. I think it's related to *dekm
which would mean that the 'prefix' is an IE-Altaic isogloss. It
produces Manchu /zhuwan/, Old Japanese /töwö/, Classical
Mongolian /arban/ and Turkish /on/. The Proto-Steppe word would have
been *t?u-kum (or *t?u-k&m if there is a fourth vowel *&). In Altaic,
the *k would soften out of existence with only a glide *w in its
place and ejectives like *t? regularly deglottalize in Altaic so
there shouldn't be a problem in the connection. This would support *t?
u = "one"... but interestingly we find something similar on the
Tyrrhenian side of IndoTyr in Etruscan as /tHu/."
I'm not familiar with Altaic sound changes... I'm guessing that the
initial */d/ was lost in Mongolian and Turkish, while it softened
to /Z/ in Manchu and hardened to /t/ in Old Japanese? Where did
the /r/ and /b/ in Classical Mongolian /arban/ come from?
"[Rob:]
'Perhaps (Pre-)IE's speakers counted routinely from the left hand to
the right hand.'
We went over the reason for this connection before but while that's
possible, there still is no **dek- which by itself means "ten" nor
does anyone on this Forum no [sic] anything about an instance of
**dek- meaning "right side", only *deks-. A connection can't be
logically established between the two roots at all."
I know I'm rehashing old arguments here, but is it possible that
*deks- is really *dek-s-, that is, with the genitive **-s?
However, this makes less sense when considering forms like *deksi-.
Assuming the *-i is the common adjectival formant *-i, there is no
reason why it would be put after a frozen genitive ending. So the
root probably was *deks-.
"Rob:
'Ultimately, all numerals come from body-counting terms. But the
origins for many of these are buried in prehistory.'
That's an unprovable assertion. We don't know that for certain at
all."
As Harold also pointed out, the words for larger numbers are likely
to not be from body-counting terms, but from words like "group."
However, the words for digits (itself coming from the Latin word
for "finger") likely came from body-counting terms, I think.
- Rob