CHAT: IE numbers (was: Icelandic Numbers (was: Re: CHAT: San Marino))

At 1:43 am +0000 30/8/00, Oskar Gudlaugsson wrote:
[....]
>number declension system. When reading about IE, I also somehow got the
>understanding that even in that language, the number 4 was undeclined. But
>couldn't be true, I mean, how could Icelandic/Old Norse just make up such a
>declension? It usually goes the other way, right?
Exactly.

'four' was certainly declined in PIE.

The numbers 'one', 'two', 'three' and 'four' were fully declined in ancient
Greek and IIRC in Sanskrit.  Only 'dyo' (two), which had the moribund dual
endings, became invariable before the end of the ancient Greek period.
From there on, things have remained much the same till the present day.
Modern Greek still declines the numbers 'one', 'three' and 'four'.  The
declension of 'four' in the modern language is:
       Masc. & fem.    neuter
Nom.    téssereis      téssera
Acc.    téssereis      téssera
Gen.    tessérwn       tessérwn  (w = omega)

Not much variation, I admit, but still some  :)

The Brittonic languages have long given up case distinctions, but they
retain gender distinctions in the numbers 1 through 4, thus:
              WELSH    BRETON         CORNISH
                               'Unified' Kemmyn 'Modern'
  two (masc.  dau       daou       deu    dew    deaw
      (fem.   dwy       diou       dyw    diw    dew

three (masc.  tri       tri        try    tri    try
      (fem.   tair      teir       tyr    teyr   tayr

 four (masc.  pedwar    pevar     peswar  peswar   pager
      (fem.   pedair    peder     peder   peder    peder

'one' also originally declined, but is now invariable in all the languages
(un; unan; un, unn, edn); however, in all the feminine causes soft mutation
of a following noun, whereas the masculine does not.

Ray.


=========================================
A mind which thinks at its own expense
will always interfere with language.
                   [J.G. Hamann 1760]
=========================================