LINGUIST List 6.1199

Fri Sep 1 1995

Sum: English Numerals

Editor for this issue: T. Daniel Seely <dseelyemunix.emich.edu>

Directory

1. Picus Sizhi Ding, Sum: English Numerals

I'd like to thank the following people who replied my query on the
English 'eleven' and 'twelve': Anton Sherwood, Bob Hoberman, Ann
Lindvall, Keith McCormick, Mark Mitton, Marc Picard, Paul Purdom,
Steven Schaufele, Richard Sproat, Larry Trask, Henk Wolf, Paul
Woods, Deborah Yeager, and halaszkewszeg.norden1.com.

As several respondents point out, English 'eleven' and 'twelve' are
etymologically related to 'one' and 'two' meaning 'one left', 'two
left' (over ten) respectively. This can also be found in other
Germanic languages such as Gothic, Old Norse, German, Dutch, West
Frisian and perhaps in some Scandinavian tongues like Swedish. Now
it is clear to me that my original impression on these two numerals
is wrong. There is another mistake made in my original posting.
This time it is with the French 'sixteen', which I carelessly put
it in the dix- group. It should be _seize_, as noticed by Mark
Miltton.

Below are extracts from the responses with additional information.


>From Bob Hoberman (RDHOBERMANccmail.sunysb.edu):

<Biblical Hebrew has two words for 'eleven', one of them formed
with a form of the ordinary word for 'one' plus the same form for
'teen' (derived from 'ten') that appears in the numbers 12-19, and
the second of them formed with a totally different first element,
and the same form for 'teen'. I don't have etymological reference
books handy, but I know that the former one is cognate with the
words for 'eleven' in the related languages Arabic and Aramaic,
while the strange element for 'one' in the latter form for 'eleven'
is cognate with aword for 'one' in another related language,
Akkadian. 'Twelve' works just like 13-19, with the usual
morphological properties of the word for 'two' in the language. If
this example is useful to you, I can look up some more of the
etymological information.

I realize after writing the above that it might be confusing.
Hebrew numbers such as 13, 14, etc. have the same order of
morphemes as English: three-teen, four-teen, five-teen (both the
morphemes are slightly different phonologically than their isolated
forms). 'Eleven' is one-teen, 'twelve' is two-teen. However, there
is an alternative form for 'eleven', which we could model asbip-
teen, where bip is also the word for 'one' in a related language.
If you want the actual forms of the morphemes, let me know.>


>From Ann Lindvall (ann.lindvallling.lu.se):
<Still, there IS something mysterious with 11 and 12. Here is one
example of reversed order:

(modern) Greek
1 ena
11 endeka (one+ten)
2 dio
12 dodeka (two+ten)
3-9 tria etc
13-19 dekatria etc (ten+three)

Singalese
1 ek
11 ekolos
2 de
12 dolos
3-9 tun
13-19 exc 15 dahatun (ten+three)

I have more examples of languages with a regular derivation from 1-
9 to 11-19, either ten+one or one+ten. I list them here:

ten+one one+ten
Turkish Persian
Inuktitut Polish
Mapudungu Arabic
Chinese
Kammu
Tamil
Mongolian
Georgian
Adygean

Still other examples with regularity 1-9 and 11-19:
Sami (11 = one+second of ten)
Finnish (11 = one+second)
Estonian (11 = one+second)>

I must add that only the data from German, Swedish, Greek, Turkish,
Persian and Polish come from my own knowledge in these languages.
All the other examples are taken from second-hand sources.>


>From Anton Sherwood (dashernetcom.com):
<In the Romance languages:
 Latin Italian Spanish French
10 decem dieci diez dix
11 un-decim undici once onze
12 duo-decim dodici doce douze
13 tre-decim tredici trece treize
14 quattuor-decim quatordici catorce quatorze
15 quin-decim quindici quince quinze
16 se-decim sedici diez y seis seize
17 septen-decim diciassette diez y siete dix-sept
18 duo-de-viginti diciotto diez y ocho dix-huit
19 un-de-viginti diciannove diez y nueve dix-neuf
20 viginti venti veinte vingt

The -ze or -ce is simply a remnant of the word for 10.

In Welsh, the higher teens are named as 1+15, 2+15, 2x9, 4+15.>


>From Larry Trask (larrytcogs.susx.ac.uk):
<The only language I can mention here is Basque. In Basque, the
numerals from 12 to 19 are of transparent formation. Thus `12' is
hamabi, from hamar `10' and bi `2', and so on. But `11' is
different: it's hamaika, in which the first element is obviously
hamar `10' but the second element is mysterious: it looks nothing
like bat `1' (< *bade or *bada). Many have tried to interpret this
-ika (or better *-eka, on the basis of internal evidence) as a lost
ancient numeral for `1', but there is no supporting evidence for
this and good evidence against it.

Since hamaika is everywhere used in Basque to represent an
indefinitely large number ("I told him eleven times" = English "I
told him a thousand times"), I am inclined to wonder whether
hamaika might not represent a fossized relic of a time when the Basque
counting system stopped at ten, so that hamaika might originally
have meant merely `ten-something', in other words `lots'. This,
however, is pure speculation.>


>From Paul Woods (ncxrs.ucc.okstate.edu):
<Slavic languages start at 11 for the teens.>


>From Henk Wolf:
West Frisian:

1 ien 11 alve
2 twa 12 tolve
3 trije 13 tretjin
4 fjouwer 14 fjirtjin
5 fiif 15 fyftjin
6 seis 16 sechstjin
7 s^an 17 santjin
8 acht 18 achttjin
9 njoggen 19 njoggentjin
10 tsien

Dutch:

1 een 11 elf
2 twee 12 twaalf
3 drie 13 dertien
4 vier 14 veertien
5 vijf 15 vijftien
6 zes 16 zestien
7 zeven 17 zeventien
8 acht 18 achttien
9 negen 19 negentien
10 tien

>From Deborah Yeager (kyeagerprism.nmt.edu):

<In Gothic:
 1 = a'ins
 2 = twa'i
 11= a'inlif is unattested, but the dative a'inlibim exists
 12= twalif (dat twalibim)
Other '-teens' just add tai'hun 10.
>From Wright's Grammar of the Gothic Language, p. 115>