Re: Types of numerals; bases in natlangs.

Hi!

John Vertical <johnvertical@...> writes:
> Mark J. Reed wrote:
>...
> >What constitutes "unavoidability" for a new word?  It seems like you
> >could go on indefinitely with ten, ten tens, ten tens of tens, ten
> >tens of tens of tens, etc.
>
> Those can get confusingly long, and probably also ambiguous depending
> on the language (is "fmof fmof fmof fmof fmof" 10010 or 1100?) Then
> again, other, more syntactically limited languages might run into
> problems already with expressing "ten hundred myriads", so you do have
> a sound point there.
My number system first developed for Tyl Sjok system only uses base
words up to ten if you use base ten numbers and encodes long numbers
very efficiently without needing and words for large numbers.  It uses
a base-exponent notation, just as you write mathematically.  This is
then applied recursively.  Basically, number work like this:

   Number ::= Exponent Base Sequence_of_digits

Where Exponent is itself a Number.  Base is the base word, (often
enough, this is therefore '10').  Any of the three Exponent, Base and
Sequence_of_digits may be empty if trivial, but at least one of them
must be given.

To understand the system:

   1 = ling
   2 = kul
   3 = hen
  10 = kjox

  20 = kjox kul          = '10 2'
  12 = kjox ling kul     = '10 1 2'

 100 = kul kjox          = '2 10'
 300 = kul kjox hen      = '2 10 3'
 120 = kul kjox ling kul = '2 12 1 2'

And now:

 3*10^20    = kjox kul kjox hen    = '10 2 10 3'
 10^300     = kul kjox hen kjox    = '2 10 3 10'

Is this understandable?

Anyway...  I think I tried to make this system understood before on
this list.  Mainly the result was, the this system is too scientific
for ordinary language users.  I disagree, but as there are no L1
speakers, we will never know. :-)

**Henrik

--
Relay 13 is forthcoming:
http://www.conlang.info/relay/relay13.html

Reply