Miguel:
>>Which got their length from...?
Me:
>The dual.
Okay, I admit it was a smokescreen so that I could get to the library quick
and refamiliarize myself with the usual reconstruction of the numerals.
Clever, huh? I finally realize why, to this day, I remain somewhat confused
on how exactly to write the IE numerals. Taking a book off a high, dusty
shelf authored by Szemerenyi dated to 1960 as are all books in my university
library, I discovered that he reconstructs *wi:knti: (with a long *i in the
first syllable!) so sumimasen if I still seem confused about length.
Now, because my brain operates on the "efficiency of thought" principle
which has been marvelous in lowering my oxygen requirements :) I sometimes
forget what I've discovered in the past and I have to have a long sit in a
cafe to re-remember what it is that's important to this particular subject.
Well, I've managed to unzip a file deep in the recesses of my brain
concerning Early and Mid IE. So let's see if we can't solve these numeric
puzzles once and for all, shall we?
Two for tea
-----------
Alright. Where shall I begin? What continues to confuse me is my
arch-nemesis known as *H1. I despise this phoneme because often times it
isn't a phoneme, just a frightening mathematical symbol signifying nothing
else but vowel length.
Let's write *-i: rather than *-iH1 for the dual, 'kay? There, that's a
little better. Why? Because, I remember now, the dual *-i: derives
ultimately from analogy with the word for "two". The Mid IE form c.6000 BCE,
before the loss of final vowels, would have been *-eixe. When the final
vowels are lost, we are left with *-eix (ending in consonantal *-x). Well,
consonantal *-x likes to make a disappearing act and so we end up with
compensatory lengthening *-e:i which later becomes *-i: because long vowels
like to rise (*-e:i > **-i:i or rather *-i:). Get it so far? Good.
So now, it must be understood that the IE dual is a relatively recent
phenomenon. Earliest stages of IE probably didn't have a "dual", only a
singular and plural. This is based on my nifty internal reconstruction
combined with the simple fact that the dual in the noun and verb was really
never worked out in full even by Late IE. As I say, the ending *-eixe (*-i:)
was derived from the word for "two", being at the time *t:Wei-xe or *t:Wa-xe
with a final plural suffix *-xe (later *-x, or rather *-H2) tacked on
needlessly as is the typical fashion of Mid IE. The word *t:Wei-xe was later
misinterpreted as *t:Wa-(e)ixe (*dwoi:) where this new termination *-eixe
(*-i:) was presumed to have a specifically dual meaning by Mid IE speakers.
Finally, some six thousand years later, IEists would reconstruct the dual
ending with an *-H1 to my chagrin and assume that the dual had always
existed.
So, how do you explain the decads, smarty pants?
------------------------------------------------
Let's look at my reconstruction of numerals for Early IE (c.7000-6000 BCE)
to get an understanding of my bizarre thinking processes:
1 t:eu (enclitic *t:e)
2 t:Wei (enclitic *t:We)
3 kWelei (enclitic *kWel)
4 kWetWan
5 penkWe
6 *reu
7 *rara
8 kWetWaxe
9 neure
10 t:e kam
20 t:We kames (declined stem: kamet-)
30 kWel kames
40 kWetWan kames
50 penkWe kames
Looking at a previous reconstructed stage is worth a thousand pictures. As
you can see, the decads, albeit spoken of rarely in Early IE no doubt, show
an opposition between singular *kam and plural *kam-es. However, early on,
*kam was found only with *t:e "one", producing the compound *t:ekam. At this
point, one ends up with an opposition between *t:ekam "ten" and *kames
"tens".
In Mid IE, *kames, like so many of the other numerals, was given useless
plural endings (remember *t:wei-xe "two"?). So *kames was given the plural
ending *-xe producing *kamet-xe since the stem was *kamet-. By 5000 BCE, we
had *kant-x (Common IE *-k^ontx). After the dual ending was formed as noted
above, the word for "20" was given the new *-i: ending for the dual. Thus
(*wi-)k^nt-i:
So in Late IE, one ended up with a yucky situation that IEists to this day
cannot accept because it destroys their pretty notions of a perfectly
regular IE. The singular stem for "ten" was *dek^m- while the plural stem
was *k^ont-/k^nt-. Granted, some regularisation appears to have been
attempted by introducing a *-t into formations with *dek^m (cf. *dek^mtos
for "tenth" instead of the original *dek^mmos). The word *dek^mt should
normally have meant "decad" and *dek^mtos should normally have meant "of a
decad, consisting of a decad" but the pressure to regularize the two
divergeant stems was great enough to override exact semantics in favour of
some consistency between a new *dek^mt-/*k^ont- opposition where *-t-
existed everywhere.
Now, what about the lengthening? Lengthening can be caused by many, many
factors from analogy to loss of laryngeal and so it drives me up the wall
that this phenomenon is used as some kind of proof that *-d- must have
existed in an entirely hypothetical decad stem **-dk^ont-. The *d just isn't
there and we will have to eventually swallow this reality when centuries go
by without ever finding this devilish leprachaun attested.
As was mentioned by Piotr, formations like *trix-k^ontx, with plural *-x,
combined with analogy, is a possible alternative to explaining the
lengthening from 30 to 90. Another factor, ironically, could be that even
speakers of IE, like IEists, may have desired that this *d existed in the
plural. This "feeling of loss" by IE speakers could in itself spur the
development of length to compensate for a perceived *d that was never there.
As you can see, there are many valid and more satisfying possibilities to be
explored than clinging to unproven IE stems.
I hope followers of the **dk^ont- cult will understand my atheism and grant
me forgiveness for losing the path to salvation. :)
- gLeN
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at
http://www.hotmail.com