Are there languages for which the word "first" is built with the word "one" ? In many languages I know there exists a suppletion :

  • English : first-one
  • French : premier-un
  • Latin : primus-unus
  • Hebrew : ראשון-אחד

In Hebrew particularly, it is interesting that in the Bible one finds אחד with also a meaning of first : יום אחד (Gen 1), באחד לחודש (Deut 1).

So do you know languages for which there is no suppletion for 1-1st ?

And in particular do you know a theory for this particular suppletion, or any article about it ?


  • 1
    Note that the French example descends directly from the Latin one.
    – Draconis
    Commented Oct 11, 2020 at 23:41
  • 3
    Would you count count (a) languages that have both a suppletive and a derived form (e.g., Irish aonú ‘oneth’, céad ‘first’), and (b) languages that don’t have morphological ordinals at all (e.g., Mandarin 一 ‘one’, 第一 dì yī ‘number one, first’)? Commented Oct 12, 2020 at 8:20
  • See also linguistics.stackexchange.com/questions/30186/… Commented Oct 12, 2020 at 21:15

1 Answer 1


According to Barbiers (2007: 'Indefinite numerals one and many and the cause of ordinal suppletion', Lingua 117), suppletion between the cardinal and ordinal numerals of 1 is frequent, and he gives an explanation why one is special and incompatible with the standard derivation of ordinals (which I cannot explain in detail). He also gives references that may be interesting to you:

  • Hurford, J., 1987. Language and Number. The Emergence of a Cognitive System. Blackwell, Oxford.
  • Veselinova, L., 1998. Suppletion in the derivation of ordinal numerals. A case study. MITWPL 31.

But according to WALS, feature 53A, this is not a universal: 3 (out of 321) languages have ordinals 1/2/3 identical to cardinal numbers; 41 have regular derivation for these ordinals (among these Mandarin, on which see Janus Bahs Jacquet's comment). I would add that Esperanto has a regular derivation as well, but that's cheating.

With regards to your Biblical Hebrew examples:

  • The first (yōm 'eḥād "day one") also exists in English: day(/bus/candidate/room) one. I think we should analyse that as N-to-D raising of day and assume that one and first are allomorphs.

  • The second (bə'eḥād laḥodeš "on one to-the-month") exists in Dutch: op één november ("on the first of November"). Marijke de Belder (2007: 'Silence and the Construct State in Dutch Date Expressions', Linguistics in the Netherlands 2007) analyses this with a silent noun dag "day" and a construct state: één dag november, with the silent noun raised to D (explaining this ungrammaticality: (*de) één november). In this analysis it is the same as English day one, the only difference being that the noun is silent.

  • 1
    Notice that L. Veselinova is also co-author of the WALS article. She is a specialist on suppletion.
    – jlawler
    Commented Oct 12, 2020 at 18:33
  • 1
    Yes, her publications list contains several items that may be of interest.
    – Keelan
    Commented Oct 12, 2020 at 19:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.