Greek vowels in Romance languages

< Previous | Next >

vaprwev

New Member
Italian
I never studied Ancient Greek at school, unlike Latin, so I have a natural question to which I haven't found an answer on wikipedia yet:
how were Greek vowels affected by the sound changes from Classical Latin into Vulgar Latin? And then Romance languages? In particular Italian, as it should be the simplest to understand.
In Italian I have noticed that many words of Greek origin are peculiar in that they have an open E/O in positions where words of Latin origin could never or hardly could, to the point that I can guess they're Greek derived because of it.
Some examples: tèsi, fenòmeno, lògica, ètica (according to treccani.it).

I even wonder if there are Greek words that in Italian have a close E/O, since I haven't been able to spot any.
Could it be that the original distinction between close mid (ε , ο) and open mid (η , ω) of Ancient Greek was lost? And if so, why?
For instance, I have searched and found out that lògica comes from λογική, so 'o' should actually be close mid rather than open mid in Italian, shouldn't it? Although I guess the matter is complicated by the shifts in stress that Greek words typically undergo in Latin.
 
  • entangledbank

    Senior Member
    English - South-East England
    There is insufficient evidence as to whether Greek short /e/ /o/ were open, close, or mid-height, but when taken into Latin they followed the Latin system: short /e/ /o/ were open. Parts of the Greek vowel system had started to change by the time borrowing into Latin was widespread, but most of the familiar technical words, such as those you quote, were assimilated into Classical Latin and underwent the same changes in Late Latin as native Latin vowels. So the likes of tèsi and lògica are the expected forms.
     

    Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    Those are all learned borrowing dating from the Renaissance, actual Greek borrowings into vulgar Latin show all the usual changes: episcopum -> vescovo, ecclesia -> chiesa, Petrum -> Pietro, petram -> pietra, piper -> pepe (see poivre in French, which shows the expected diphthong from open syllable /e/ in that language).

    There's also semi-learned borrowings, where a word shows some of the expected changes but not all, which is usually attributed to the influence of the Latin word that was still in use among the literate class that'd have used that kind of vocabulary: chorum -> it. coro instead of cuoro; librum -> it. libro, fr. livre instead of lebro and loivre.
     

    Catagrapha

    Member
    Malagasy
    The less expected the change is, the less likely the word is borrowed, yes?
    e.g., organum > organo, likely borrowed, argano, less likely
    angelus > angelo, likely borrowed, angiolo, less likely
     

    vaprwev

    New Member
    Italian
    Thinking of this in Spanish...
    Petram > piedra seems consistent with the explanation that Greek borrowings dating back to the "Vulgar" period (for lack of a better word), or pre-dating it, were completely latinized.
    Other words, like the ones I mentioned in OP, or "tono", "Europa", "génesis" instead are treated oppositely than in Italian, as in Spanish they are pronounced with a close o/e.
    Sure enough, Spanish doesn't have an open-mid vs mid/close-mid opposition unlike Italian, but it usually does keep trace of the original distinction between the two pairs:
    (Vulgar) ĕ, ē, ŏ, ō > (Spanish) ie, e, ue, o.
    This is at least true with most Latinate words that I can think of.

    So, my question is why didn't this happen:
    (Vulgar Latin) *tesis, *fenomeno, ... > *tiesis, *fenuémeno, ...
    Is the written nature of the spread of these words, opposed to the oral modality of most Latinate words, the cause?
    Or may it be that by the time these words were introduced into Spanish, Romance diphthongization no longer was an active process in Spanish? And most importantly, that by that same time the open-mids had already ceased to exist?

    In short, I'm wondering if Greek borrowings past a certain "Vulgar" period were imported into Italian and Spanish with this approach:
    (Greek) ĕ, ē, ŏ, ō > (Italian) è, è, ò, ò, (Spanish) e, e, o, o.
    as this latinization of Greek words doesn't seem to be the same as the earlier, the one responsible of pĕtra > It. piètra, Sp. piedra.
    And what the explanation could be.
     

    vaprwev

    New Member
    Italian
    The less expected the change is, the less likely the word is borrowed, yes?
    e.g., organum > organo, likely borrowed, argano, less likely
    angelus > angelo, likely borrowed, angiolo, less likely
    I'm not an expect as it should be clear by this point, but I'll try to answer.
    Assuming that by borrowed word you mean a scholarly/learned borrowing (organo), as opposed to a more "popular" borrowing (argano), for what Italian is concerned the answer may be yes.
    But I think that, very much in general, words actively used in the speech are just more subject to unexpected changes than unused ones (or later introduced ones).
    For example, It. "chiedere" comes from Latin "quaero", definitely not a borrowing, with an unexpected change of r > d.
    Likewise, some words have open-mids in place of close-mids or viceversa in all Romance languages for no good reason.
    For example, the Latin verb "mitto" has an expected form in "méttere" in Italian. But if you check this old vocabulary, you can see that the form "mèttere" was also present in Italian: Etimologia : mettere, mettere;.

    In fact, unexpected changes are perhaps the biggest reason for the existence of Latinate doublets in Italian.
     

    Penyafort

    Senior Member
    Catalan (Catalonia), Spanish (Spain)
    Thinking of this in Spanish...
    Petram > piedra seems consistent with the explanation that Greek borrowings dating back to the "Vulgar" period (for lack of a better word), or pre-dating it, were completely latinized.
    Other words, like the ones I mentioned in OP, or "tono", "Europa", "génesis" instead are treated oppositely than in Italian, as in Spanish they are pronounced with a close o/e.
    Sure enough, Spanish doesn't have an open-mid vs mid/close-mid opposition unlike Italian, but it usually does keep trace of the original distinction between the two pairs:
    (Vulgar) ĕ, ē, ŏ, ō > (Spanish) ie, e, ue, o.
    This is at least true with most Latinate words that I can think of.

    So, my question is why didn't this happen:
    (Vulgar Latin) *tesis, *fenomeno, ... > *tiesis, *fenuémeno, ...
    Is the written nature of the spread of these words, opposed to the oral modality of most Latinate words, the cause?
    Or may it be that by the time these words were introduced into Spanish, Romance diphthongization no longer was an active process in Spanish?
    Well, I don't know if I really get your point, but certainly you cannot compare words that are part of the core vocabulary of the language, such as stone (used by everybody, and which were there in the gestational process during the Middle Ages, undergoing the popular changes associated with each Romance language) with words such as thesis and phenomenon, which are more recent in time and haven't really been used in the spoken language until even more recently.

    And yes, petra in the Romance languages clearly comes from Vulgar Latin.

    Greek πέτρα ['petra] (short e)
    > Latin pĕtra ['petra] (short e)​
    > Vulgar Latin ['pέtra] (open e)​
    Portuguese pedra ['pɛdra]​
    Spanish piedra ['pjedra]​
    Catalan pedra ['pɛdrə]​
    French pierre [pR]​
    Italian pietra ['ptra]​
    Romanian piatră ['pjatrə]​
    etc.
     

    Catagrapha

    Member
    Malagasy
    I saw an interesting example: estratagema, stratagemma, from strategema, most likely borrowed - aren't estrategema and strategemma more expected?
     

    pollohispanizado

    Senior Member
    Inglés canadiense
    It should also be noted that the short Latin O and E weren't diptongized in every single instance, and some other vowels that "shouldn't" have been, like short I and U, were. Furthermore, any word ending in -sis is obviously borrowed quite directly; otherwise, we would expect *la tese/tiese -- *las teses/tieses instead of "la tesis/las tesis". Spanish tends to naturally dislike invariable words that end in S, so pretty much any that still has it was borrowed directly from somewhere. (cf. "cuerpo" comes from "corpus", the accusative of "corpus" which left an S in the singular -- *el cuerpos-- that was rather quickly discarded by the Vulgar Latin speakers in Spain as it clashed with the then newly minted fixed plural system.)
     
    Last edited:

    berndf

    Moderator
    German (Germany)
    There is insufficient evidence as to whether Greek short /e/ /o/ were open, close, or mid-height
    I don't think that is entirely accurate. ε sometimes transcribed Latin ĭ as in κομέτιον for comitium. This suggest that ε was relatively high, closer to [e] then to [ε].
     
    < Previous | Next >
    Top