Re: numeration system

On Tuesday, December 14, 2004, at 06:55 , Gary Shannon wrote:

> My impression is that the number names all sound too
> much alike. To be easily understood under a variety of
> listening conditions there needs to be greater
> contrast between individual words.
>
> Imagine giving someone a number over a bad phone
> connection. I hear "ko ko ko ko ko ko ko"
Yes - that was precisely my reaction. The system may work on paper, but it
would not work in speech. Natlangs show this.

The old British initial s- has become h- in modern Welsh; but there is one
notable exception: _saith_ = 7. The initial s- was almost certainly
preserved because *haith /hajT/ was too easily confused with _wyth_ /ujT/
in certain environments.

In the early day of the telephone, it was found that "nought" and "four"
were readily confused over bad lines, so people got into the habit of
saying "oh" for 0 ("zero" is becoming quite often used nowadays also). 5
and 9 still cause problems and people will use forms like 'fivah' or
'ninah' in order to make the final consonant clearer.

Similarly in german the standard 'zwei' is not used when giving numbers
over the phone, because it gets confused with 'drei'; the (archaic?
dialect?) 'zwo' is used instead.

I am sure other examples can be given from natlangs

>>> 0 zo
>>> 2 ko
>>> 3 no
>>> 4 mo
>>> 5 lo
>>> 6 vo
>>> 7 fo
>>> 8 bo
>>> 9 po
>>>
>>> 10 koko
>>> 11 koko ko
We still need a word for 'one'.

Apparently the philosopher Leibniz had a similar sort of scheme for
numerals:
1 = b
2 = c
3 = d
4 = f
5 = g
6 = h
7 = l
8 = m
9 = n

To these consonants are appended vowels thus:
-a = x1
-e = x10
-i = x100
-o = x1000
-u = x10000

So the numbers 1 to 9 would be: ba, ca, da, fa, ga. ha, la, ma. na (cf.
the ko, no, mo... list above). IMO it suffers from just the same problem
that Gary has pointed out with #1's proposed system.

81 374 in Leibnitz's system would be: mubodilefa. But, unlike most systems,
  the syllables can be written in any order, thus 81 374 could also be
bodifalemu, lemudibofa or any of the other possible arrangement of
syllables. I am far from convinced that having 120 different ways of
saying "eighty-one thousand, three hundred (and) seventy four" is a good
thing!

As to how Leibniz proposed initial |c| to be pronounced or what he
proposed for numbers about 99 999, I do not know.

In my notes I have a system designed by one G. de Kolovrat. Unfortunately,
  when I assembled these notes some 40 years ago, I did not record any
other details about G. de Kolovrat or the source of my information. He put
forward a system of 100 simple CV syllables for the numbers 0 to 99. I won'
t write them all out here; I'll just give 0 to 9:
ba, be, bi, bo, bu, ca (5), ce, ci, co, cu.

(As the system does contain ka, ke, ki, ko, ku - 35, 36, 37, 38, 39 - I
assume |c| cannot be /k/. But what it is I do not know.)

To express numbers greater tha 99, you just broke the number up into
groups of two digits, prefixing a leading 0, if necessary, thus 81 374 -->
08.13.74 --> codotu. Unlike Leibnitz's system, the syllables cannot be
re-arrange without changing the meaning; for example dotuco = 137 408. The
system can also, of course, express any number till infinity.

Unfortunately, it still suffers from having too mny similar sounding
syllables for related numbers. However, if you are after just ten CV
syllables for 0 to 9, then I would advise using as many different vowels
as possible and make the initial consonants added to any similar sounding
vowels as different as possible.

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com
===============================================
Anything is possible in the fabulous Celtic twilight,
which is not so much a twilight of the gods
as of the reason."      [JRRT, "English and Welsh" ]

Replies