Re: IE right & 10

From: Richard Wordingham
Message: 34215
Date: 2004-09-19

--- In cybalist@yahoogroups.com, enlil@... wrote:
> Harald:
> > Quite often the have fingercounting which ranges up to 5, 10 or
20.
> > Extremely seldom, if at all, are there monomorphemic numerals up
to
> > exactly 6.
>
> I think this is an incorrect way of thinking about language. If you
> really wrap your head around what it would be like to draw out a
> linguistic map of languages during the neolithic, you'd realize
that
> you'd end up with a pattern like a continuum where there aren't any
> hard edges. What does that have to do with anything? Well, if you
take
> one dialect of a particular language, they may indeed have numbers
up
> to "ten". You take another dialect and they may also have numbers
up
> to "ten".
>
> But what if the two dialects of that language have different number
> sets? What if the higher numbers differ in the two dialects? This
is
> what we'd expect since higher numbers would tend to be replaced
more
> often than lower ones, particularly in a neolithic environment
where
> we have nothing but hunter-gatherer bands roaming the wilderness
for
> good hunting grounds. All the speakers of a particular language
> aren't going to band together, have a meeting and decide how to say
> 'eight' but that doesn't mean that they didn't have their own local
> word.

Do you mean palaeolithic? Also, I would have expected members of
the same tribe (= many bands) to understand one another's number
words. Another issue is that you could very easily see a mixture of
hard bondaries and far ranging dialect continua (rather like today,
in fact.)

> So, now let's expand this thought experiment.
>
> What if the two dialects both contribute to the development of a
> proto-language that us modernday people are trying to reconstruct?
> We all should know by now that proto-languages are abstract
> representations of the real language. The _real_ protolanguage,
such
> as the 'real IE', would not have been a single, homogeneous
language.
> Rather, the true protolanguage would be a collection of
interrelated
> dialects all contributing to form what we call things like 'Proto-
IE'
> or 'Proto-Uralic'.

Speculation, again.

> Now, back to Uralic itself. Understanding the above fully means
that
> we understand that Uralic is not a single language and never was.
It's
> a conglomeration of dialects, ever merging, fracturing and
remerging.
> One can still have number systems from one to ten in these
neolithic
> languages without them ever being reconstructable in the artificial
> construct we call 'protolanguage' simply because there may not have
> been consensus amongst the neighbouring dialects on what 'seven',
> 'eight', 'nine' or 'ten' should be.

Who needs dialects? An individual's speech can have multiple
unrelated forms - e.g. 'dozen' v. 'twelve', 'score'
v. 'twenty', 'brace' v. 'pair' in English. (I know the starts
of 'dozen' and 'twelve' are cognate, but I think it reasonable to
take them as unrelated.)

Richard.