Companion Terms for 7 and 8 (Re: 'eight' some more)

[Koontz John E]

On Sun, 9 May 2004, R. Rankin wrote:
> As I recall, my essential point was that, if you use the Plains Sign
> Language system for the numerals, then the number 6 is (re)interpretable
> as simply the second fist (with a nearly invisible thumb actually
> counting for the numeral). Then 7 extends one VISIBLE finger, giving the
> morpheme /-wiN/ 'one'.  And so forth.  The problem was figuring out what
> kind of a counting system would incorporate 'one' in 'seven' instead of
> 'six', and the hand signs accomplished that.

Bob?  I don't remember - were you placing a particular analysis on *pe in
'six'?  In Denig's description that Linda cites, fingers are closed down in
counting, rather that opened out.  Denig:

> In counting with the hand, an Indian invariably begins with the little
> finger of the left, shutting it down forcibly with the thumb of the
> right; when the five fingers are thus shut he commences on the thumb of
> the right, shutting it with the left fist.

So, 'six' involves the closed left fist holding down the right thumb.
The closest I can get to a morpheme *pe that would describe this is the
fist syllable of *pe'thaN 'to fold' (Dakotan *pehaN, OP bethaN, etc.), or
maybe *pe'priN 'to twist' (Da pemni(N)', OP be'bdhiN, etc.), One would
have to assume some sort of truncation - or, of course, another morpheme
entirely!

But, when the right index finger is closed down on the right thumb and the
left fist placed on it you do have a locative + 'one' situation, htough
I'd have expected the a-locative rather than the o-locative.  I guess I
should think of it as into the palm, not onto the thumb.

> The analysis may be a bit strained, but I still find it attractive.

It is a bit unexpected.  I commented much earlier in this thread that it
was surprising that 'seven' involved 'one', recapituating your earlier
observation though not your insight.  I think that the form speaks for
itself, however, so that the problem is one of discovering, if possible,
how 'one' could be intended, and not of justifying to ourselves that it is
actually 'one'.  It's obvious that 'one' isn't the X in 5 + X = 7.  The
*s^aak-based higher digits don't use that approach, whereas the *hpe- and
*hpa-based higher digits do.

One thing I notice about the counting scheme is that it starts with the
little finger on the first hand, and the thumb on the second, so that,
in effect, one could use a single hand - thumb out is 'four', but little
finger ot is 'nine', and so on.  The other hand acts to hold down the
little finger for 'one', since little finger don't move well independently
for most of us.  This scheme can also be thought us as proceding
iconically across the finger of the hands held up before one.

Clark's The Indian Sign Language describes counting by extending or
unfolding fingers, starting with the closed right hand, but, again,
beginning with the little finger and continuing on the left hand with the
thumb.

By contrast, when I count with my fingers - and I assume I do this in the
canonical way for my own culture, though perhaps I aven't been paying
attention, as usual! - I start with my closed right fist and extend
successive fingers starting with the index finger, using the thumb to hold
down the little finger until it is needed for 'four', and adding the thumb
to make five.  Then I start over again in the same order on the left hand,
perhaps holding up the open right hand to remind people that I've already
reached five.  I might tap each finger as it extends using the index
finger of the other hand.  Alternatively, I can start extending fingers
with the thumb, holding the little finger in with the base of the thumb,
and almost certainly tapping the extended fingers in this case.  This
approach avoids using the other hand to hold in the little finger, but it
doesn't distinguish 1-5 from 6-10 by using different hand configurations.

My understanding is that systems of finger counting differ culturally,
though the same themes, obviously, tend to repeat themselves from the
antomical limitations of the situation.

> It's clear they were using the hand signs since 'nine' is something like
> /naNpciyuNka/ 'one in the palm' [the little finger] if memory serves.
> 'Ten', then, involves all the fingers extended on both hands.

Denig describes folding in fingers in counting, but also extending them in
showing numbers.  Denig, again, from Linda:

> When wishing to telegraph by signs a certain number less than 10 he
> holds up that number of fingers, beginning with the little finger of the
> left hand and keeping the others shut. Should the number be 7, then all
> the fingers of the left and thumb and finger of the right would be
> extended, ...

I assume understands the Dakotan 'nine' construction is na(N)p- 'hand'
plus c^iNyuNka, where the latter is something like (i)c^iyuNka <
*i-ki-yuNka meaning something like 'it (one's own)  sitting/lying with
respect to it'.  I have to assume the *i- to get c^i from *ki.
Alternatively, it might be *raNraNpyiruNka (putting it in PMV terms) or
*na(N)p-c^hiyuNka, with c^(h) < *y, but in that case I don't understand
the construction.

> 'Eight' is odd man out.  What is the meaning of -loghaN/-yoghaN of
> /$aglogaN/?? If I ever knew, I've forgotten.

I can't find an etymology in terms of something like PMV *roghaN (with the
forms Bob cites in Dakotan) either.  The *ro might be 'flesh', but that
doesn't seem to lead anywhere.