Italian and Romanian plural

Discussion in 'Etymology, History of languages, and Linguistics (EHL)' started by francisgranada, Jun 16, 2012.

  1. francisgranada Senior Member

    Hi everybody!

    I'd like to start a discussion about the origin of the Italian and Romanian plural (ending in -i and -e).

    (This argument was treated partially also in other threads, but I was not able to find a thread dedicated especially to this question [sorry, if I am mistaken ...])

    Generally, the Romance nouns and adjectives are supposed to derive from the the Latin accusative (or "casus obliquus") which is valid also for the Italian in singular. Instead, in plural we have pochi, poche, case, amori etc.... instead of the "expected" pocos, pocas, casas, amores.

    So the question is if the Italian (Italo-Romance) and Romanian (Balcano-Romance ?) plurals derive directly from the Latin nominative (even though e.g. the Latin for amori is amores) or it's rather the result of a later evolution?
    Last edited: Jun 17, 2012
  2. fdb Senior Member

    Cambridge, UK
    French (France)
    Yes, the Italian and Romanian plurals continue directly the Latin nominative plural. amori is an example of paradigmatic levelling.
  3. CapnPrep Senior Member

    There was already some discussion of Italian plural endings in this thread, in particular about 3rd declension nouns:
    Accusative declension from Latin -> Italian

    And there is also this Wikipedia article:
    Romance plurals
    And this paper, which you may have on-line access to:
    Yves D’hulst (2005) "Romance plurals". Lingua 116: 1303–1329.

    Finally, you might like to check out this project on Romance plurals at Oxford:
    The Romance noun
    Final s would not be expected to survive in Italian. The question is what it would have turned into and the possible effects on the final vowel.
    Last edited: Jun 17, 2012
  4. francisgranada Senior Member

    Of course (I haven't expressed myself precisely enough)

    Thanks for the links. I've read some opinions also before and as far I can understand, there is still no "general consensus" on this question. That's the reason why I'd like to hear (read) the opinion of others, too. The Wikipedia article says: The "nominative" theory appears more straightforward at first; however, the "accusative" theory is more common currently.
  5. Ben Jamin Senior Member

    Is there a possibility that there was an influence of invading nations (Goths and Lombards) on development of Italian from Latin?
  6. francisgranada Senior Member

    In case of the plural endings, I don't believe.
  7. XiaoRoel

    XiaoRoel Senior Member

    Vigo (Galiza)
    galego, español
    Desde el punto de vista de un latinista (mi profesión) no hay duda de que esta teoría del "acusativo" para explicar los plurales del italiano (don de hay hablas con plural en -s), ni es mayoritaria, sino todo lo contrario, es la teoría del nominativo la que está mayoritariamente aceptada.
    Sabida es la cercanía a la lengua madre, en cosas gramaticales especialmente, del rumano y del italiano.
    Estos plurales vocálicos trazan además una línea de separación, una isoglosa que separa el romance occidental, mayoritariamente de sustrato céltico, del italiano, del dálmata y del rumano, que además sufren el adstrato y superestrato germánico, y después el dálmata y el rumano la presión adstrática y superestrática de las lenguas eslavas, que no sufren las lenguas hispanicas centrooccidentales, por las que los elementos suevo o visigodo casi ni dejaron rastro.
  8. Forero Senior Member

    Houston, Texas, USA
    USA English
    I think that when speakers of Vulgar Latin quit distinguishing between nominative and accusative nouns, they still needed a way to distinguish plural from singular. Most chose to use "-s" for plural, and some chose to use a vowel change (and there were some other things too), but none chose nominative or accusative per se as their model because they saw no clear difference. Some regularly eliminated final s whatever its source, but all regularized to some degree or another.

    Rumanian is heavily influenced by Slavic and other non-Romance language habits. How do Slavic languages form plurals?

    The Italian peninsula is not just the part of the Empire nearest the city of Rome but the home of highly varied dialects, some with roots in non-Latinate languages. Is some sort of "missing link" still spoken somewhere that, for example, sometimes allows an -s ending and sometimes an -i (for plural nouns, second person verbs, etc.)?
  9. fdb Senior Member

    Cambridge, UK
    French (France)
    I agree with you. In linguistics, as in other sciences, the truth is not decided by a majority vote (and certainly not by what gets into the unsigned and unrefereed entries in Wikipedia), but by the strength of argument. For me, the argument that explains ALL Italian plural forms as somehow truncated accusatives seems extraordinarily weak.
  10. CapnPrep Senior Member

    Which particular version of the argument do you have in mind here? I don't know if anyone has proposed explaining ALL Italian plurals as etymological accusatives, but Martin Maiden comes pretty close, and I do not consider his work "extraordinarily weak". The nominative hypothesis has its share of problems, too.
  11. merquiades

    merquiades Senior Member

    USA Northeast
    Sí, eso es lo que siempre me han enseñado y la verdad es que es muy lógico, pero entonces como latinista ¿cómo explicarías tú la pérdida sistemática de la s final en italiano incluso donde (más) normal hubiera sido conservarla... por ejemplo, además del nominativo, también en las desinencias verbales de segunda persona singular... amas> ami? ¿Será analogía? Ya sé que a lo mejor es pura casualidad pero me llama la atención que antes donde había /s/ (y aún lo tenemos en las lenguas occidentales hispánicas) ahora hay /i/ (italiano y rumano).
  12. killerbee256 Senior Member

    American English
    How do/did plurals functions in other Italian and eastern Romance Languages? Is it the same?
  13. merquiades

    merquiades Senior Member

    USA Northeast
    This should answer your question. There's a line going through Northern Italy. North and West of this line plurals are taken from the accusative and end in s (this includes French, Spanish, Portuguese etc.), South and East of this line they end in vowel sounds and surely derive from the nominative (Italian and most dialects and Romanian).
  14. berndf Moderator

    German (Germany)
    I don't know why you say "most". Today Western Romance languages have more speakers than Eastern Romance languages. But this was not necessarily so some 1800 years ago or whenever the the divide developed.
  15. francisgranada Senior Member

    Typical Slavic plural endings today in nominative :
    Masculine: -y, -i, -e/é/ia, -ove/ové/ovia
    Feminine: -y, -i, -e
    Neuter: -a/á
    In other cases we have a plenty of different endings (-om, -am, -ach, -iach, -och, -ami, -imi, -mi, -ov ….)

    I myself can hardly imagine the influence of the relatively complex Slavic plural system on Rumanian, and absolutely not on Italo-Romance.
  16. berndf Moderator

    German (Germany)
    If you read the discussion above, you will discover that this is not at all "sure".
  17. merquiades

    merquiades Senior Member

    USA Northeast
    I did read it but I'm not really convinced. The nominative theory makes sense. That's why I said surely with a very slight doubt and not definitively, obviously, completely.

    Xiao the latinist has said it is obvious and accepted by everyone, so I've asked him to explain it.

    For me surely means probably not certainly. See #8
    Last edited: Jun 20, 2012
  18. francisgranada Senior Member

    The accusative theory makes sense, too ...:)

    (see e.g. Romance plurals , a link given by CapnPrep)
  19. merquiades

    merquiades Senior Member

    USA Northeast
    Personally I believe the nominative theory, but my only little doubt is that final /s/ has completely disappeared in Italian (even where it shouldn't have): for example, the "tu" verb endings have "i" in Italian, not "s" which cannot be easily explained. It seems wherever there is /s/ in Western Romance, Italian has /i/, which would lead to believe there was a phonetic change across the board.

    Edit: @Francisgranada.
    I guess both theories could go hand and hand then. Both together could explain everything. Italian might derive from the Proto-romance nominative and not the Classical Latin nominative (which is not the same) and then out of analogy all unstressed /as/ become /e/, all unstressed /es/ become /i/. Complicated..... :confused:
    Last edited: Jun 20, 2012
  20. francisgranada Senior Member

    This is not the only "problem" that cannot be easily explained. According to the nominative theory we should expect amice, lunge etc... instead of amiche, lunghe etc ... Another question is the appearence of -i also in cases when in Latin there is -s: amori, spiriti, voci ... (of course, this can be a later generalization of this -i, but ...)

    P.S. Sorry, I haven't noticed your last edit before ...
  21. merquiades

    merquiades Senior Member

    USA Northeast
    Yes those words with /k/ that have palatized masculine plurals but not feminine plurals is an enigma: amico amici amica amiche. It is also strange that /k/ nouns are palatized but not /g/ nouns: albergo alberghi. Maybe this is another issue. All romance languages have palatized in different ways before different vowel sounds

    Incidently, Andalusian Spanish shows vowels opening after final s is eliminated. Las amigas > Lah amigah > Lae amigae. Los hombres > lohombreh > lɔ hombrei
    So I can believe that final /s/ could eventually become /i/ or /e/.

    Edit: At any rate there must be nominative. It could have been that there was hesitation between nominative and accusative at a certain time with a general switch to nominative then in the mind of people since -s was disappearing in favor of -i they generalized the rule to all nouns then to final -s in general. *this is just my pure mental speculation
    Last edited: Jun 20, 2012
  22. francisgranada Senior Member

    Not so complicated ... But the Proto-romance nominative mostly coincides with the Latin accusative. This should suggest that rather the accusative theory is "valid" (or at least nearer to the "reality").

    But yes, the evolution of the Proto-Romance plural endings could be more complicaded and not necessarily a straightforward accusative-to-nominitave development. However, the main question is, if the "Western Romance" and the "Eastern Romance" plural systems are of the same common origin or not?

    (The conservation of the -a plurals as uova, braccia, dita ... etc. in some Italian nouns is another question, but it does not contradict neither to the accusative nor to the nominative theory)
    Last edited: Jun 20, 2012
  23. merquiades

    merquiades Senior Member

    USA Northeast
    Isn't this based entirely on Old French which has the advantage of keeping the case system long enough for it to show up in writing? If that also occurred in the Italian pelninsula would it have to be the same? We know that French eventually moved toward accusative anyway.
    It depends on how early this system developed. It's known that Romans tended to ignore the Iberian peninsula in later years of the empire. The lack of contact caused those languages to retain some archaic characteristics. Yet, it is obvious (no doubt here) for example that Spanish derives completely from the Latin accusative. The question is did western vulgar latin develop this characteristic on its own?

    Weren't these nouns neuter in Latin which would make -a the natural nominative plural?
  24. CapnPrep Senior Member

    This particular part of the Wikipedia article could use some editing… It is true that some scholars believe that Romance inherited a vulgar nominative -as ending, but it is very strange to call that "the accusative theory"… :rolleyes:

    The accusative theory derives Italian (masculine and feminine) plurals from VL accusative forms, not only in the a-declension, but also for o-declension nouns and adjectives.
  25. francisgranada Senior Member

    It's a perfect example ! After 1500 years some scholars will say that the Andalusian plural comes from the Latin nominative (illae amicae) :D (I am just joking ...)

    The evolution like -os > -oi > -i is very improbable (according to your opinion)?
    Last edited: Jun 20, 2012
  26. CapnPrep Senior Member

    Actually, no, if anything, Old French and Old Occitan/Provençal show that Proto-Romance had a nominative derived from the VL nominative and an accusative derived from the VL accusative. I suppose that francisgranada was referring to the general movement, throughout Romance, towards a form that looks more like the Latin accusative than the nominative. But (as discussed for example in the other thread) this is an impression based primarily on the stem of the noun in the singular, so I don't think it can be used to argue anything about the ending of the noun in the plural.
    It's not unreasonable. The problem is that, as far as I know, it would be pretty much limited to this single grammatical ending. Noi and voi show the first part of the change (if they derive from nos and vos), and according to D'hulst in the article I cited above, voi > vi shows the second part (but personally I think vi < ibi is a better explanation).

    Apart from these, very few Latin words ending in -os seem to have survived in Italian, so unfortunately we won't able to find strong evidence either for or against os > i as a regular sound change.
  27. francisgranada Senior Member

    Thank you, CapnPrep, for the intersting and exhaustive answer.

    For those who are interested in, here is the Sequence of Saint Eulalia, an Old French (Walloon) text that contains both nominative and accusative plurals.
  28. LiseR Member

    First of all, I have to warn you that my Romanian is far not that good, but I'll try to bring a few examples.

    Romanian plural : scaun (neutral) = chair / scaune
    dulap(masc) = glass / dulapuri
    brad(masc) = fir / brazi
    copil(masc) = child / copii

    and so on
  29. francisgranada Senior Member

    I understand you, there are similar plural endings (i,e) in Romanian like in the Slavic languages. However, I think that this is not the consequence of Slavic influence, but rather a common "Eastern Romance" heredity.
  30. CapnPrep Senior Member

    Plural formation in Romanian is much more complicated than in Italian, but as far as I know, the various suffixes are all believed to derive from Latin. Looking at Romanian nominal declension in general, a Slavic origin has been proposed for some of the vocative endings, especially feminine singular -o, but this is not universally accepted (see e.g. Tucker 1944).
  31. robbie_SWE

    robbie_SWE Senior Member

    Trilingual: Swedish, Romanian & English
    I don't want to sound obnoxious, but your knowledge of the Romanian language does not support your statement or your hypothesis. Just because languages share a trait at first glance doesn't mean they have a common source.

    As Francisgranada stated earlier, complex Slavic plural systems have hardly influenced Romanian and I haven't yet met a linguist who states otherwise. The articles mentioned by CapnPrep illustrate the situation pretty well.

    Best Regards,

    Last edited: Jun 26, 2012
  32. LiseR Member

    You are right, comrade.

  33. Nino83 Senior Member

    Hello everyone.

    All Italian languages had or have vocalic plurals, also Gallo-Italian languages. For example, in Old Piedmontese (until today in Valsesian, grand/grend for grande/grandi), Lombard (until recent times, quest/quist, mes/mis for questo/questi, mese/mesi), Venetian (toso/tusi), Bolognese (gat/ghet) there are traces of metafonetic plural for masculine (of the second declension and for masculine/feminine of the third declension) plurals ending in -i (this feature is well mantained in Bolognese and Romagnolo).

    About Germanic influences, well, it's difficult to say that Tuscan, Roman, Neapolitan, Sicilian were influenced by those languages.

    So, all Italian languages had or have vocalic plurals.

    About amico/amici [ko/ʧi], these are exception to the rule. The rule is: for masculine paroxitones, the plural is chi/ghi [ki, gi] while for proparoxitones the plural is ci/gi [ʧi, ʤi]. Some example: bùco/bùchi, luògo/luòghi but mèdico/mèdici. So the change in the plural [ko/ʧi] is not general in Italian, so amica/amiche, buca/buche i.e the general [ka/ke] for feminine nouns could be explained in a different way.

    By the way, there is another explanation.
    Singular nouns derive from Latin accusative: rosa, muro, ragione (rationem, nominative ratio). This, because it was the most frequent case.
    Plural nouns: first declension ae/as (nominative/accusative plural), second declension i/os, thirs declension es/es.
    Some say that first declension had a nominative plural as, like in Old French la fame/les fames, so we can rewrite in this manner:
    I as/as II i/os III es/es

    As we can see, for I and III there is no difference between nominative and accusative/oblique, while for II we have two endings.

    For I we have:

    For III we have:

    source: G. Patota, Lineamenti di grammatica storica dell'Italiano, Il Mulino

    In Italian, plurals were formed from the nominative, for II and from the nominative (equal to the accusative) for I and III, where the last vowel was palatalized by the "s", then the "s" was lost.

    I: casa(m) > casa; casas > cases > case
    II: muru(m) > muro; muri = muri
    III: ratione(m) > ragione; rationes > rationis > ragioni

    This theory could explain the plural forms amica/amiche, buca/buche and medieval forms capres, operes, tables.

    Anyway, Italian nouns derive from nominative plurals, i.e I as, II i, III es.

    This theory says that in stressed syllables, s > i, like in nos > noi, vos > voi, vas > vai or was absorbed by the following consonant, like in raddoppiamento fonosintattico, tres casas > tre case /trèk'ka:se/, while in unstressed syllables, "s" palatalized the final vowel, as > es > e, es > is > i.
    The second person singular -i in verb conjugation is explained as it follows:
    amas > ames > ame (which are attested in Old Italian) > ami
    temes > temis > temi
    sentis > senti
    The third person singular:
    amat > ama
    temet > teme
    sentit > senti > sente (analogy with the second)

    The other theory says that they derive from I ae, II i, III i (analogy) and that plurals like amica/amiche are a particular case.
    Last edited: Jun 29, 2015
  34. Ben Jamin Senior Member

    The variety of plural endings in different Slavic languages today does not mean there was the same variety in the Slavic language at the time it influenced the Vallachian dialects.
  35. danielstan Senior Member

    Romanian - Romania
    The Romanian plurals can be reasonably explained as derived from Latin plurals (with many similarities with southern Italian dialects), so the Slavic influence cannot be taken into account.

    I explain briefly the process (I don't give an academical explanation):

    Since Vulgar Latin the final -s and final -m was lost in Balkan Peninsula. In fact there are no Romanian words inherited from Latin with their final -s or -m.

    This situation eliminated the Latin plurals with -s ending and let Romanian "choose" the -i ending for masculine plural and the -e (< lat. ae) for feminine plural.

    The Latin ending -ora was retained in Romanian as an ending for neuter plural case (along with the feminine -e ending for neuter plural). E.g. tren/trenuri vs. fapt/fapte

    By Romanian internal evolution (/a/ in unstressed position became /ǎ/ [ə], /o/ in unstressed position became /u/) this neuter termination has been transformed in -urǎ, resulting in confusions with the singular feminine ending -ǎ.
    gurǎ ("mouth", feminine singular) vs. cerurǎ ("skies", neuter plural)
    To avoid such confusions the neuter plural ending in -urǎ has evolved in -ure (as it is attested in Old Romanian texts of 16th century).
    In these religious texts the neuter plural is (with very few exceptions) ending in -ure. Examples: cerĭure ("skies"), lucrure ("things").
    Since 17th century Romanian texts contains mixed neuters plurals with -ure and -uri endigs, while after 18th century the -uri ending is generalized.

    Under the influence of the Romanian masculine plural ending in -i some other Latin words (which originally ended in -s) have gained a -i ending which never existed in Latin:
    lat. nos > rom. noi (see also it. noi)
    lat. vos > rom. voi (cf. it. voi)
    lat. duos > rom. doi
    lat. tres > rom. trei
    The influence of Slavic on Romanian plural ending in -i:

    Most of the Slavic verbs in infinitive end in (short i). E.g. pisatj ("to write") [pi-'satĭ]
    Romanian plurals use the same short -i (despite the Romanian orthography, which has no special character for distinguishing the short -ĭ- from normal -i-).
    The effect is the reduction of number of syllables:
    rom. oameni ['wa-menĭ] compare with it. uomini ['wo-mi-ni]
    rom. lupi [lupĭ] cf. it. lupi ['lu-pi]
    This is applicable also on Romanian neuter plural ending in -uri:
    rom. lucruri ['lu-krurĭ]

Share This Page