"Une femme" and "Un homme", but not "Deuce femmes" and "Deux hommes"?

Is there a reason that the number one receives special treatment?

  • 5
    Note that premier/première, second/seconde, tiers/tierce, quart/quarte, quint/quinte are gendered too. – jlliagre Dec 21 '18 at 16:18
  • 3
    @jlliagre But there are adjectives or nouns, not articles. – Matthieu Brucher Dec 21 '18 at 17:15
  • 1
    @MatthieuBrucher They are indeed but if you look closely to the question, you should notice nowhere does appear a restriction or even a mention about articles. The question tells un is the only number having a special treatment, my comment was just about a few ordinal numbers that are also gendered. – jlliagre Dec 21 '18 at 18:56
  • 2
    @jlliagre fully agree with both your comments ;) – Matthieu Brucher Dec 21 '18 at 18:57
  • 1
    if you take "number" litterally, you could also say vingt-et-un(e), trente-et-un(e)... so you have an infnity of gendered numbers. But of course they all end with un(e), so this doesn't answer the question. – radouxju Dec 23 '18 at 20:10

It's not really unusual at all for a French word to not have distinct forms for masculine and feminine. It's not obvious whether this is best analyzed as "lacking gender inflection" or "having the same form for both genders" (or different ways in different contexts).

But in any case, many adjectives such as jaune, rouge and orange don't change between masculine and feminine (and in many accents of French, the distinction between bleu and bleue is purely graphical), while vert(e), gris(e), blanc(he) do change. We see the same form for masculine and feminine with the plural articles les, des and the plural "possessive adjectives" mes, tes, ses, nos, vos, leurs.

In general, it's fairly common for languages with grammatical gender to only show gender agreement for a subset of cardinal numbers (usually only lower ones), and this is the case for Latin, the ancestor of French, even though Latin had gender agreement on a few more numbers than French. In Latin, the words for "one", "two" and "three" agreed in gender with the noun, but higher numbers did not (except for, oddly, multiples of 100).

So if French were to show gender agreement on four, five, six, seven, etc. this would represent an innovation from the situation in Latin. While there are some places where French has added inflection relative to Latin, the general trend has been a reduction rather than an increase in the amount of inflection of this kind.

The Latin word for "three" had the same form for masculine and feminine (tres); the distinct form tria was used with neuter-gender nouns, but as French lost the neuter gender, it may not be too surprising that it also lost gender inflection for the number 3.

In Latin, 2 in the accusative case (the main source of Romance forms) had the masculine form duōs and the feminine form duās. The modern French spelling with "x" is just from a spelling change of "us" to "ux"; this isn't a word that historically had a "c" sound in either the masculine or feminine. The "x" just represents the plural suffix that is spelled as "-s" in other contexts, and which has become silent (outside of liaison contexts) in modern French for both masculine and feminine words.

I'm not entirely sure about the phonetic development of duōs to deux /dø/, and whether duōs and the Latin feminine form duās would both regularly correspond to modern French /dø/. Maybe in theory there could have been a feminine form "deues" or something like that, but I don't see how it could have been anything as different from deux as your hypothetical form "deuce".

  • Duos to deux would have been something like /'du.o:s/ > */'do.os/ > */'do:.os/ > */'doʊ̯.os/ > /deʊ̯s/ > /døs/ > /dø/ while duas should have been /'du.a:s/ > */'do.as/ > */'do:.as/ > */'doʊ̯.əs/ > /'deʊ̯.əs/ > /'dø.əs/ > /dø:/ (> /dø/), as you surmise. The issue here is that all the other /ø:/ rhymes are either later innovation (bleue, which was an innovation, the old french form didn't inflect for gender), or the result of a VCa sequence like leuca (lieue) or cauda (queue) so it's possible that /oa/ might have had a different outcome from /oda/ – Eau qui dort Dec 22 '18 at 14:14
  • On the other hand, via and fica show identical outcomes in -oie, so diphthong+schwa was clearly not prohibited – Eau qui dort Dec 22 '18 at 14:17

"un"/"une" is also the equivalent for "a" in English, so the undetermined article, whereas there is no other term for two, three...

  • 1
    And in English, there is a/an, and they feel the same as un/une when used. – Émile Jetzer Dec 22 '18 at 14:11

Your Answer

By clicking "Post Your Answer", you agree to our terms of service, privacy policy and cookie policy

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