1. Cilquiestsuens Senior Member

    French
    Good day to all, I recently read an article claiming that the etymology of 'chien' as coming from latin can- was doubtful. Mostly because, it was claimed of the impossiblity to have this (i) sound in 'chien' derive from the (a) in 'can-'. I am therefore asking the expert etymologists from this forum: what is your opinion?
     
  2. CapnPrep Senior Member

    France
    AmE
    :confused: This i sound is a textbook example of Bartsch's Law. It is true that the phonetic mechanism behind this sound change is poorly understood, but the facts themselves are not in question. The development of chien from cane(m) is totally regular.

    Could you please provide a reference for the article you read?
     
  3. Cilquiestsuens Senior Member

    French
    Thanks CapnPrep for your answer and for the very interesting link. I can't find the article right now. It is in French the abstract of a dissertation from some Polish university, about the Gaulish substratum in French, asserting - rightfully in my opinion, that in a number of etymologies, the possibility of an influence of the Gaulish substratum wasn't even considered... (one example: ''char'' given from 'Latin' carrus). He quoted a few more convincing examples and.... chien... claiming a possible connection with the Celtic root 'ki' (plural: koun) (breton) / cu (Irish) that would better explain the (i) in French... This part, I thought was the weakest of his dissertation, hence my question. I'll find the abstract tomorrow and give you the exact reference. That part of his thesis isn't very well documented, it seems he is just throwing ideas around, food for thought...
     
  4. CapnPrep Senior Member

    France
    AmE
    Again, :confused:… The Gaulish origin of Latin carrus or carrum is widely accepted. See e.g. Ernout & Meillet:
    The passage from Latin carru(m) to French char is again completely regular, and I don't see any need for, or evidence in favor of, supposing additional influence from Gaulish, although it could very well be the case that Gaulish speakers in the proto-French period used a word like karro with the same meaning.
     
  5. Cilquiestsuens Senior Member

    French
    Moderator note: Off-topic question about Gaulish substratum theory moved to new thread.

    In the case of 'chien', my query was motivated by the fact that in Modern Breton, 'chien' is -ki (plural koun) as well as in all Celtic languages and the word is attested in Old Irish too.

    I was struck by the similarity with a few northern 'dialects' of French (picard, chti), in which 'chien' is pronounced as 'kien' (sometimes spelt 'quien').
     
    Last edited: Sep 13, 2013
  6. berndf Moderator

    Geneva
    German (Germany)
    The non-palatalization of Latin /k/ in can- in Northern dialects (today only survived in Piccard today but originally also typical for Norman dialect; cf. modern French chasser (> English chase) and AF cacher (> English catch)) is completely regular and does not require any special explanation. The spelling with <k> or <qu> just means that the word should not be pronounced with (originally [ts]) front of the letter <i>.
     
    Last edited: Sep 13, 2013
  7. CapnPrep Senior Member

    France
    AmE
    Sure, there are cognates in Celtic that also have the sounds k and n (and several different vowels). How does this help explain the phonetic form of French chien (or Picard kien) any better than the regular development from Latin cane(m)?
     
  8. Cilquiestsuens Senior Member

    French
    Who said it does?

    It is just that I thought that if I were a linguist, I'd think it would be an interesting track to follow, and wouldn't dismiss it wholesale at first sight.

    But you might have good reasons to think it doesn't make sense. I am no expert to contradict you.
     
  9. JeanDeSponde

    JeanDeSponde Senior Member

    France, Lyon area
    France, Français
    Dans la Chanson de Roland on trouve
    Ou encore
    Ce n'est que vers le XIIe que "chien" a remplacé "chen".
     
  10. CapnPrep Senior Member

    France
    AmE
    Quentel did, in the article that lead you to start this thread:
    So he thinks that the Latin etymology could be wrong and that a Celtic source could be better. But he provides no solid arguments here, either against Latin or in favor of Celtic. He just points out that chien looks pretty different from canis while c'hi looks kind of like chien. (Does he know that Breton c'h is not the same sound as French ch?)
     
  11. Cilquiestsuens Senior Member

    French
    Merci JeanDeSponde, this is the kind of hard facts I was looking for. Interesting. Thanks for the research!
     
  12. berndf Moderator

    Geneva
    German (Germany)
    It is usually a good idea to check the easy facts first, like looking up the development of forms here. I didn't mention the early form chen because I assumed you were aware of it.
     
  13. CapnPrep Senior Member

    France
    AmE
    We also find one instance of the form with i in Roland:
    See for yourselves in the Bodleian manuscript, f. 34r.
     
  14. Cilquiestsuens Senior Member

    French
    Moderator note: Part of reply moved here (about the merits of the Gaulish substratum theory).

    <...>

    Well, Old French spelling has a few lapses (less than Modern French, maybe), I guess chen and chien must be pronounced the same...

    Is it so?
     
    Last edited: Sep 13, 2013
  15. JeanDeSponde

    JeanDeSponde Senior Member

    France, Lyon area
    France, Français
    C'est le cheminement inverse de chèvre, qui n'a cessé d'être chièvre / kièvre que vers le XVe...
     
  16. Cilquiestsuens Senior Member

    French
    How were chen and chien in time of the Chanson de Roland pronounced?

    [tsjẽn] or [tʃjẽn] or [tʃẽn] ?
     
  17. sotos Senior Member

    Greek
    Notice also the similarity with the Greek κυν (pronounced kin by the Greeks).
     
  18. Cilquiestsuens Senior Member

    French
    Thanks, sotos. The word seems to be widespread and well attested throughout the Indo-European languages.

    I could get a hold of Xavier Delamarre's 'Dictionnaire de la Langue Gauloise', the article for Dog says the following (pg. 132):

    By the way, at the end of the same article, it says, after mentioning IE equivalents in other languages :

    Can someone explain me the last part here?
     
  19. CapnPrep Senior Member

    France
    AmE
    Les deux mots ont divergé en moyen français, mais leur évolution antérieure était en principe parallèle. C'est à dire que les graphies avec ‹i› sont majoritaires pendant toute la période de l'ancien français (les effets de la loi de Bartsch étant intervenus vers le VIe siècle). Les chen, chef, cher, etc. dans Roland sont un trait anglo-normand ; si le mot chèvre avait été présent dans ce texte, je suppose qu'il se serait écrit sans ‹i› aussi.
    Keep in mind that there were many dialectal pronunciations at the time, and that Roland does not represent central French. This is how Zink retraces the development of cane(m) (1999, p. 236):
    KÁNE(M)
    Ve s. k'
    tšé
    tšíene
    VIIe s. tšíen
    Xe s. tšíẽn
    XII2–XIII1 šyẽn
    XIII2 šyɛ̃n
    XVII1 šyɛ̃

    (I had to modify some of his phonetic characters.)
    We would expect Latin to have a rounded vowel u or a labial w, given the comparison with other languages (which suggest a PIE root *kwon). Canis may have replaced an earlier Latin nominative *cō (see Ernout & Meillet).

    Since Gaulish had the expected vowel u in this word, I wonder if anything is left of the suggestion that the vocalism in French chien is a Celtic substratum effect…
     
  20. Cilquiestsuens Senior Member

    French
    Thank you very much for your research and your detailed and very informative answer.

    Moreover, the vowel change from u > i in Celtic must have been taking place at the same time or even later than the a>e change in Proto-French.
     
  21. aefrizzo

    aefrizzo Senior Member

    Palermo, Italia
    italiano
    Hello.
    Thanks to this fascinating thread and with the help of TLF, I am trying to trace some french words from the classic latin items.
    "Chier" (verb) from "cacare"?
    "Cher" (adj.) from "carus"?
    "Chair" (subs.) from "caro"?

    I do not understand though the different changes from the same first syllable "ca-".
    Thank you.
     
  22. CapnPrep Senior Member

    France
    AmE
  23. aefrizzo

    aefrizzo Senior Member

    Palermo, Italia
    italiano
    Thank you, CP, very useful links.
    I have not fully grasped the evolution from "ca-ca-re" to "chier".
    According to the rule about open syllables, and assuming it holds for inside syllables too, I would expect some intermediate terms like, e.g., "checher" or anything else. TLF reports just and abruptly "chier" on 1202.
     
    Last edited: Sep 21, 2013
  24. CapnPrep Senior Member

    France
    AmE
    It does, unless the ca follows a vowel. In the case of cácat/cacáre, the k between two a sounds becomes a yod. As in e.g. pacare > payer, baca > baie. So we have:
    • cácat > chiéie > chíe
    • cacáre > cheiér → chiér
     

Share This Page