Re: The English/French counting system (WAS: number systems from conlangs)
| From: | Ray Brown <ray.brown@...> |
|---|
| Date: | Tuesday, September 16, 2003, 19:14 |
|---|
I know this thread has drifted into quaint ways of telling the time
and now onto Calendar reform. But I'll stick with the original
points (and try and resist Calendar reform; there seems to be as many
proposals for this as there are for IALs).
=====================================================================
On Monday, September 15, 2003, at 07:58 , Pavel Iosad wrote:
> Hello,
>
>> Some days ago, I wondered about
>> why English/German/French etc.
>> (I guess all European languages)
>
> Umm, no. T ex Welsh 12 is dau ar ddeg, i. e. twoteen.
No - 12 is either 'deuddeg' in the traditional vigesimal
system or 'un deg dau' in the modern decimal system.
Traditional Modern
11. undeg un deg un
12. dau ar ddeg un deg dau
13. tri a ddeg un deg tri
14. pedwar ar ddeg un deg pedwar
15. pymtheg un deg pump
16. un ar bymtheg un deg chwech
17. dau ar bymtheg un deg saith
18. deunaw un deg wyth
19. pedwar ar bymtheg un deg naw
20. ugain dau ddeg
'pymtheg' is derived from 'pump' (5) + 'deg' (10). It
can thus been seen that altho the tradition forms
works in groups of 20s, the numeral, apart from 18
(deunaw <-- dau + naw, i.e. 2x9) & 20 (ugain), all the
other traditional forms are derived from a morpheme
denoting 10 and another denoting the units numeral. The
word 'ar' BTW means "on" and is followed by the soft
mutation.
The modern system is clearly 100% decimal. The traditional
forms are more less confined now to telling the time &
giving a person's age, especially up to 30. In all other
situations the modern forms prevail.
>> have separate names for 11 and 12: eleven, twelve;
>> elf, zwölf; onze, douze, instead of oneteen, twoteen;
>> einzehn, zweizehn; dix et un, dix et deux. We count
>> in tens, but have numbers which you can count in
>> twelves with.
>
> The Germanic ones come from one and two + *lib, meaning "rest, excess,
> something which is left". So it's "one extra [from ten]" and "two
> extra". The Gothic ones are ainlif and twalif, cf. bileiban "to be left"
Spot on - and the 'twe-' bit of twelve is still clearly cognate with 'two'
,
'twain' & 'twice' (and Scots 'twae'). Admittedly the watered down 'one' at
the beginning of 'eleven; is not so obvious.
And of course another clear exception to the generalization (and IME
generalizations always have exceptions) that all European langs have
special
words for 11 & 12 is Romanian:
11. unsprezece (i.e. un spre zece 'one on ten')
12. doisprezece
13. treisprezece
........
19. nou@sprezece (@ is actually written as a-breve)
=====================================================================
On Monday, September 15, 2003, at 08:34 , John Cowan wrote:
> Mark J. Reed scripsit:
>
>> That I've also wondered about. My impression was that the
>> Swiss system is a logical extension of the French, rather than
>> that the French is a reduction of the Swiss, but since
>> Latin had the full set of numbers up to 100 the French
>> system must have involved a reduction somewhere.
>
> Probably a calque on a Celtic substrate; Celtic numbers do the base-20
> bit in full.
That indeed is the general opinion. Only modern seems to developed new
wholly
decimal forms. But in Breton, Cornish & in the Irish, Scots & Manx
Gaelics where
vigesimal system still operates for 20, 30, 40, 50, 60, 70, 80 & 90, the
numbers
from 11 to 19 inclusive (apart from 18 in Breton & Cornish) are clearly
derived
from a decimal basis.
Similar vestiges of vegesimal counting are occasionally encountered in
English, e.g.
according the the King James version the span of a man's years are "three
score
years and ten" where a literal translation would merely say "seventy" and
the mark
of the beast in Revelation is "six hundred three score and six" where the
Greek
original merely has the prosaic 666 (Yes, it uses numerals). And, of
course,
Lincoln's famous Gettysburg address with its "four score years and
seven". It's
simply that these are noe archaism in English whereas in standard French
in the
Hexagon relics of the older system have endured.
=========================================================================
==========
On Monday, September 15, 2003, at 08:39 , Jean-François Colson wrote:
[snip]
>
> Yes, that's strange. And in French, we could count in sixteens too: 13,
> 14,
> 15 and 16 are treize, quatorze, quinze and seize, not dix-trois,
> dix-quatre,
> dix-cinq and dix-six. ;-)
>
> In French I think that comes from a distortion of the Latin words:
> 11 onze undecim
> 12 douze duodecim
> 13 treize tredecim
> 14 quatorze quattordecim / quatuordecim
> 15 quinze quindecim
> 16 seize sedecim
You do your native language a disservice to call these 'distortions' :-)
They are, in fact, quite normal derivatives from the colloquial Latin
forms of the above. 'Quattuordecim' has 'tt' in good, classical Latin
Yep, but the Latin forms are quite clearly formed by one+ten, two+ten
etc. All the Romance langs, except Romanian, have forms derived from
Latin for 11 to 15; Italian (sedici) & French retain words derived from
'sedecim'. Latin also had 'septendecim' [sic] but now Romance lang has
continued that.
But the Latin for 18 and 19 are intersting: duodeviginti, undeviginti -
2 from 20, 1 from 20. Similarly we have:
28. duodetriginta (2 from 30)
29. undetriginta (1 from 30)
38. duodequagraginta
39. undequadraginta
etc.
This is sort of the inverse of the Germanic 'one left' and 'two left' idea
for 11 & 12.
=========================================================================
On Monday, September 15, 2003, at 08:52 , Christophe Grandsire wrote:
> En réponse à Carsten Becker :
>
>
>> Some days ago, I wondered about why English/German/French etc. (I guess
>> all European languages) have separate names for 11 and 12: eleven,
>> twelve; elf, zwölf; onze, douze, instead of oneteen, twoteen; einzehn,
>> zweizehn; dix et un, dix et deux. We count in tens, but have numbers
>> which you can count in twelves with.
>
> Using French here is incorrect, since the system is: onze, douze, treize,
> quatorze, quinze, seize, dix-sept, i.e. it's until 16 that you have the
> "strange-looking" tens.
I agree entirely. To lump all Europeans langs together like this is quite
misleading. The Romance
and Germanic languages do not behave the same way in this respect. And so
far no one's mentioned the
Slav languages of Europe, which have a very considerable number of
speakers, nor others such as
Finnish or Hungarian.
> And actually, they are not that strange, since they are directly coming
> from the Latin forms undecim, duodecim, tredecim, quattuordecim, etc...,
> which actually correspond exactly to "oneteen, twoteen, thirteen,
> fourteen, etc...".
Quite! In fact, the odd ball is Romanian which has abandoned its ancestral
Latin forms entirely for these numbers.
[The rest snipped but agreed with 100%]
>> Another question: Why are the French counting so odd? Quatre-vingt (4
>> times 20) for 80, soixant-dix (60 and 10) for 70 etc. (instead of Swiss
>> "huitante" and "septante" (and "nonante")) is really difficult when you'
>> re not used to it. How did this develop?
>
> Some say that it's a substrate influence from Gaulish which, as a Celtic
> language, probably counted in twenties. I personally don't know if it
> really holds water.
I have grave doubts also.
> I know that Old French had both numeration modes (i.e. decimal and
> vicesimal),
Just like middle and early modern English :-)
Ray
===============================================
ray.brown@freeuk.com (home)
raymond.brown@kingston-college.ac.uk (work)
http://home.freeuk.com/ray.brown
===============================================