It is generally accepted that the nominal forms in the Romance languages represent reflexes of the Latin accusative rather than the nominative. (This is even true for those languages that have masculine plurals in -i, which is actually a regular reflex of -os.)

However, I'm aware of exactly one word in Romanian which is only explicable as a survival of the Latin nominative:

om: "man (sg)" < Latin homo

oameni: "men (pl)" < Latin homines

(The hypothetical, unattested Romanian reflex of Latin acc. hominem would be *oamene.)

Is there any other similar reflex in other Romance languages that reflects the Latin nominative? Is there evidence for the survival of the Latin nominative into Proto-Romance in any systematic way (outside of the pronouns)?

  • 1
    French loi and roi?
    – Timwi
    Sep 15, 2011 at 13:59
  • 4
    @Timwi, I'm pretty sure that those are regular reflexes of *lege(m) *rege(m). Sep 15, 2011 at 16:25
  • 1
    @JSBngs: That is correct.
    – Cerberus
    Sep 16, 2011 at 5:38
  • 1
    Are you sure that the Romanian example comes from the nominative? Seems to me that it could be a later "correction" to avoid confusion with neuter and feminine plurals ending in -e...
    – user493
    Nov 7, 2011 at 1:48
  • 1
    @E.C., but there are other Romanian masculine nouns ending in -e: soare, frate, etc. I doubt that disambiguation from the neuter/feminine plural is the reason. Nov 7, 2011 at 13:19

5 Answers 5


Apparently, many subjective and objective (cas régime) forms of words were still used in Middle French, at least into the late Middles Ages, if you go through the etymologies of French words. Here are some examples of words in modern French that are reflexes of the Latin nominative, from the Centre National de Ressources Textuelles et Lexicales:

French on:

... Du lat. homo, cas suj. du subst. signifiant «homme», développé en position proclitique, et qui, à basse époque, est relevé dans qq. ex. comme suj. indéterm., emploi aboutissant à sa fonction de pron. indéf.: Peregr. Aether. 13, 1: "ubi homo desiderium suum compleri videt"

French fils, "son":

... Du lat. class. filius « fils, enfant »; « descendants » en b. lat. La forme actuelle représente l'anc. cas sujet conservé en raison de son emploi fréq. comme vocatif et prob. aussi pour éviter la confusion avec fil.

A similar cross-pollination from the vocative occurred with French soeur, "sister" (the Old-French oblique case seror falling out of use):

Du lat. soror « sœur ». L'a. fr. avait les formes suer, cas suj., et seror, avec différentes var., cas régime issu de l'acc. sororem, très tôt le cas suj. devint le plus usuel à cause de son utilisation fréq. au vocatif.

Another good candidate is corps (OFr. cors); the nominative is not mentioned as such, but, the word being neuter in Latin, nominative and accusative were identical in form, so it would seem likely for that form to be used rather than some declined form.

... Du lat. class. corpus, attesté aux sens de base A 1, 2, 3, B 1; sens C en lat. class., repris en fr. surtout au xvie s.; le sens D est également latin; ...

There's also croix and champs, but those probably come from plurals, in which nominative and accusative are formally identical.

Italian uomo (pl. uomini) appears to come from the nominative homo too. The Vocabolario Etimologico della Lingua Italiana by Pianigiani is not entirely clear; lacking further knowledge about the phonology of that time and a reliable source, we cannot be absolutely certain that the word wasn't shortened from huomino, or something like that. The same applies to Provençal hom.

This Spanish etymological-dictionary website says hombre comes from hominem but luz, paz, voz, cruz from lux, pax, vox, crux; however, I would not trust this dictionary: it sounds rather amateurish all over. I couldn't find any good, relatively modern etymological dictionary for Spanish.

Update: This quotation from Wikipedia lists more examples for French:

La forme unique du français moderne dérive le plus souvent du cas régime. Il y a cependant un certain nombre d'exceptions où c'est le cas sujet qui a survécu, concernant les noms de personnes : ex. prestre / provoire (« prêtre »), ancestre / ancessor (« ancêtre »), traïtre / traïtor (« traître »), suer / seror (« sœur ») et de nombreux prénoms. Dans quelques cas, le cas sujet et le cas régime se sont tous deux maintenus dans la langue moderne, parfois avec des sens différents : c'est le cas pour gars / garçon, copain / compagnon, sire / seigneur, pâtre / pasteur, nonne / nonnain et pute / putain. [The first word of each pair is the Old-French nominatif.]

A few more French examples:

Moindre, and many other nominals on -dre:

Anc. cas suj., issu du nomin. lat. minor, compar. de parvus «petit en taille, nombre, quantité, valeur, âge, rang, condition, importance», le cas régime étant meneur issu de l'acc. minorem, v. mineur. Dès la fin du xiie s. mendre fut employé au cas régime sing., puis aux cas suj. et régime plur., évinçant meneur.


Anc. cas suj., issu du lat. junior nomin., compar. de l'adj. juvenis « jeune ».

Probably queux:

Étymol. et Hist. Ca 1100 cous cas régime plur. (Roland, éd. J. Bédier, 1817); ca 1165 keu cas suj. plur. (Guillaume d'Angleterre, éd. M. Wilmotte, 1818); 1174-76 coeu cas régime sing.; keus cas suj. sing. (Guernes de Pont-Ste-Maxence, St Thomas, éd. E. Walberg, 1291; 1292); fin xiies. princes queurs de la coisine (Prophéties de David, 69 ds T.-L.); ca 1280 Li maistre ques (Merveilles de Rigomer, 7684, ibid.). Du lat. coquus « cuisinier », devenu cocus.

Maire, from a substantivised nominative adjective (and probably many pure nouns on -air(e) that come from Latin -ator):

A substantivation de l'anc. adj. maire «plus grand» ca 1165 (Benoît de Sainte-Maure, Troie, éd. L. Constans, 6031), issu du nominatif lat. major, comparatif de magnus «grand», v. majeur.

If you Google for site:cnrtl.fr/etymologie/ "cas suj."|nominatif|nomin, you will probably find a whole lot more (though most of those Google hits are not it).


Re 'survival of the Latin nominative into Proto-Romance' - It survives into Old French, which had a two-case system until about the twelfth century.

So for a standard second-declension noun vicinus "neighbor":

Nom. Sg. _li voisins_ < VICINUS  Nom. Pl. _li voisin_   < VICINI
Obl. Sg. _le voisin_  < VICINUM  Obl. Pl. _les voisins_ < VICINOS

A sample remnant of the third declension, from soror "sister":

Nom. Sg. _la suer_  < SOROR      Nom. Pl. _les serors_  < SORORES
Obl. Sg. _la seror_ < SOROREM    Obl. Pl. _les serors_  < SORORES

Of course there are several innovations and Old French doesn't preserve the Latin declensions perfectly, but the original forms are retained in several declension types.

The Wikipedia article linked has a lot of example declensions, many of which appear to be sourced from Kibler's An Introduction to Old French.


In Spanish there's Dios, which I'm pretty sure is a continuation of the nominative. Also, I want to say that Carlos is from the nominative form Carolus. And of course there's Jesus.

  • 1
    Not sure about Carlos, but Dios and Jesús are both biblical, and thus likely to have been copied wholesale from Latin without intervening sound changes. Oct 15, 2011 at 4:31
  • 1
    I get what you're saying about Jesus (Hell, German used some of the Latin inflections of Jesus for a while), but I disagree about dios, as it's the general term for a deity; i.e. it's the word for god as well as God, so to speak. Also, the forms definitely based on deus are only found in Spanish & Portuguese, with Catalan deu, Friulian diu, and French dieu clearly coming from the accusative deum. My guess is that god, generally being "active" would most often be subject of a sentence, and might have contributed to the survival of the nominative form. Just my $0.02 of speculation. Oct 15, 2011 at 19:05
  • 1
    I'd put it down to the fact that the word was said frequently in prayers in Latin, rather than everyday speech. Oct 16, 2011 at 2:03

On a recent trip to Italy I noticed that the men's restrooms say "Uomini" rather than "Uomi" as I would have naively expected. I don't know much Italian so there could be more.

I'm not aware of any examples of this phenomenon in Spanish.

  • 1
    The Italian example suggests that homo/homines is common to Eastern Romance, at least. But are there any other examples? Sep 15, 2011 at 17:43
  • This is actually pretty general - plurals in Italian come from nominative in all cases as far as I am aware.
    – Eleshar
    Jan 2, 2017 at 17:09
  • 2
    @Eleshar: Actually, as far as I know that is disputed. While the feminine plural "e" could come from Latin "ae", it also could come from "as" with vocalization of word-final "s" to [i] and subsequent monophthongization of "ai" to "e". See Wikipedia en.wikipedia.org/wiki/Romance_plurals#Origin_of_vocalic_plurals Sep 8, 2017 at 1:54

In French:

  • the word meaning field champs comes from campus nominative rather than campum accusative

  • I suppose the word republique comes from res publica nominative rather than rem publicam accusative.

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.