Language name and location: Tomo Kan Dogon, Mopti, Mali [Refer to Ethnologue]

言名称和分布地区托莫坎-多贡语, 马里中部莫普提区, 塞古区及布基纳法索境内

 

1. cíː ~ hinná *

21.  pyɛ̀-lɔ́ː túré sàʔá

2. lwɛ́ː ~ lɔ́ː

22.  pyɛ̀-lɔ́ː lɔ̀ː sàʔá 

3. tàːndí ~ táːndɛ̀ː

23.  pyɛ̀-lɔ́ː tàːndí sàʔá 

4. náː

24.  pyɛ̀-lɔ́ː náː sàʔá 

5. nùnà

25.  pyɛ̀-lɔ́ː núná sàʔá 

6. kúlé

26.  pyɛ̀-lɔ́ː kúlé sàʔá   

7. sɔ́ː

27.  pyɛ̀-lɔ́ː sɔ́ː sàʔá 

8. sìlà

28.  pyɛ̀-lɔ́ː sílá sàʔá 

9. twáː

29.  pyɛ̀-lɔ́ː twáː sàʔá   

10. pyɛ́w ~ pyɛ̀lɛ̀

30.  pyɛ̀-tàːndí

11. pyɛ́lɛ̀ː túré sàʔá

40.  pyɛ̀-náː

12. pyɛ́lɛ̀ː lɔ́ː sàʔá

50.  pyɛ̀-núná

13. pyɛ́lɛ̀ː tàːndí sàʔá  

60.  pyɛ̀-kúlé

14. pyɛ́lɛ̀ː náː sàʔá 

70.  pyɛ̀-sɔ́ː

15. pyɛ́lɛ̀ː núná sàʔá  

80.  sùŋɡú (archaic form)

16. pyɛ́lɛ̀ː kúlé sàʔá  

90.  sùŋɡɛ̀ː-pyɛ́w (80+10)

17. pyɛ́lɛ̀ː sɔ́ː sàʔá 

100. sùŋɡɛ̀ː-pyɛ́-lwɛ́ [80+(10 x 2)]

18. pyɛ́lɛ̀ː sílá sàʔá 

200. sùŋɡù-lɔ́ː pyɛ̀-náː [(80 x 2)+(10 x 4)]

19. pyɛ́lɛ̀ː twáː sàʔá  

800. mùjú **

20. pyɛ̀-lɔ́ː  

1000. 

 

Linguist providing data and dateː Dr. Vadim Dyachkov, Department of African Languages, Institute of Linguistics RAS (The Russian Academy of Sciences), Moscow,
Russia, December 9
, 2019, December 16, 2020.
Referenceː  Vadim Dyachkov, A Grammar of Tomo Kan Dogon
供资料的语言学家: Dr. Vadim Dyachkov, 2019 年 12 月 9 日, 2020 年 12 月 16 日.

 

Other comments: Tomo Kan Dogon has an archaic forms for 80 'sùŋú' and 90 'sùŋú pɛ́rú'. Note that some numerals differentiate the forms used for counting and contextual forms that modify a noun in context, there are two forms for numbers 1, 2, 3 and 10ː counting form for 1, 2, 3 and 10 are tʃíː, lwɛ́ː, tàndí and pjɛ́w, and contextual form areː hìnná, lɔ́ː, táːndɛ̀ː and pjɛ̀lɛ̀ː, The tonal changes depend on the number of syllables in a numeral stem. Numerals ‘11’ - ‘19’ are formed according to the scheme: numeral ‘10’ + single-digit numeral + sàʔá ‘plus’. The numeral ‘10’ is presented in these compounds in its L-toned contextual form pyɛ̀lɛ̀ː. The single-digit numerals preserve their tones. The stem túré ‘1’ is not attested elsewhere in Tomo Kan but has evident cognates in related languages (compare Jamsay túrú ‘1’). The stem is also related to the Tomo Kan distributive quantifier tútúrú which is a frozen reduplication (Jamsay tù-tùrú ).
Decimal multiple numerals are formed according to the scheme: numeral ‘10’ + single digit numeral. The numeral ‘10’ is presented in these compounds in form pyɛ̀ which must be L-toned form of the count form pyɛ́w where the stem-final /w/ is erased. The single-digit numerals preserve their tones. Like simple ‘2 – 10’ numerals, decimal multiple compounds are represented by count forms and contextual forms – for example, the numeral ‘30’ is pyɛ-tàːndí when pronounced isolatedly and pyɛ̀-tá:ndɛ́ in context. The numeral ‘80’ is non-standard and is elicited as súŋgù (with the context form súŋgɛ̀:). Single-digit numerals can be added to decimal multiple compounds. They are added according to the same rules that derive ’11 – 19’ numerals. The numeral ‘1’ is presented in such compounds as túré. The tones of single-digit numerals are not affected by the preceding cont
ext. The original Tomo Kan system is octagonal. All compounds denoting numerals ‘> 80’ are formed by adding a numeral denoting the required number to the numeral ‘80’: sùŋgɛ̀:-pyɛ́w 90’, sùŋgɛ̀:-pyɛ̀-lwɛ́ ‘100’. The numeral súŋgù in these compounds undergoes tone-dropping. Numerals denoting huge numbers can be hardly elicited and are usually loaned form French. However, some compounds could be elicited by my consultants and exemplifies the model: 200=sùŋgù-lɔ́ː: pyɛ̀-ná:
[(80 x 2) +(10 x 4)]. Other huge numerals are rare and considered archaic. The only example known to me is mùjú '800. I've checked out the data and it seems that there is an interesting thing. mùjú was translated as 'thousand' in a previous Tomo Kan dictionary, but as '800' by my consultant, so it turned out that I don't know the truth. Probably it's archaic and that's why people misuse it. However, in the previous Tomo
Kan grammar mùjú is also translated as '800'.
Note that transcriptions between traditional phonetic symbols and IPAː c stands for IPA [tʃ], j for [dʒ], and y for [j]; and nasalized vowels areː uⁿ=ũ, oⁿ=õ, ɔⁿ=ɔ̃, aⁿ=ã,

iⁿ=ĩ.


 

Back >> [ Home ] >> [ Niger-Congo ] >> [ Adamawa-Ubanguian ] >> [ Atlantic ] >> [ Benue-Congo ]
>>
[ Grassfields
] >> [ Gur ] >>  [ Kwa ] >> [ Mande ] >> [ Narrow Bantu ]