Is there a way to ascertain whether a third declension noun ending in -r in the nominative and -ris in the genitive is r-stem or s-stem?


I understand that the s in some s-stem words has entirely given way to r.

Sometimes it is possible to find out that this has happened.

For example, for honor Wiktionary lists honos as an alternative form; so you can guess that it was s-stem.

For victor, Wiktionary gives the etymology of vic- + -tor; so you think this was r-stem.

But for luror, I don't find in Wiktionary anything I would want to go on (just that the word is related to χλωρός and helvus‎).

Is there some reliable way to ascertain whether an apparently r-stem noun is in truth (or originally) s-stem?

Should I for example assume that an apparently r-stem word is in fact r-stem unless I find some contrary indication?

  • For what reason are you distinguishing them? I'd consider honor an R-stem for pretty much all intents and purposes, even though it started out S-stem, because its declension has been entirely regularized.
    – Draconis
    Commented Jan 12, 2017 at 4:03
  • @Draconis I am a beginner in Latin and don't want to lose sight of some distinction that may later become important. For example, I found that it was sometimes difficult to make sense of Greek noun declensions without knowing their earlier forms (e.g. the accent on the antepenult of πόλεως, from πόλῐος). I suppose you are telling me that the r/s will not have that sort of importance?
    – Catomic
    Commented Jan 12, 2017 at 4:19
  • 1
    Welcome to the site! I agree that the distinction will not be important in Latin. Elaborating on that would make a good answer. I would still like to know the answer to the question, not out of practical need but curiosity.
    – Joonas Ilmavirta
    Commented Jan 12, 2017 at 6:49
  • 2
    Yeah, I think it was smart to do Greek first and then Latin. I did it the other way around and going from a deeply regular language to a deeply irregular one really did a number on me. There are a VERY few words in Latin whose original forms it's helpful to keep in mind, but if I count them off the top of my head I still have fingers left over. Commented Jan 12, 2017 at 14:34

2 Answers 2


Look at older forms.

(See the end of this post for a tl;dr.)

For example, consider the third-declension noun lār, laris ("home spirit"). In Classical Latin this seems to be a standard R-stem noun: lār, laris, larī, larem, lare...

But in some very archaic Latin (such as the Carmen Arvale), the nominative plural is given as lases. This could mean that it used to be an S-stem lās, lasis instead.

How does that make sense? There are a set of standard sound changes which happened between the Old Latin of Plautus and Cato, and the Classical Latin of Cicero and Vergil.

  • S became R in between vowels
    • Compare flōris (formerly *flōsis) against flōs
    • (More precisely, S became a Z sound between vowels. Later on, the Z sound became something like the American R; even later, that turned into the trilled Italian R.)
  • VO usually became VE
    • Compare vester against vōs
  • DV became B
    • Hence duellum (> "duel") became bellum, and duis (from duo) became bis
    • Duo was used commonly enough to escape this change, just like it preserved the dual endings which otherwise died out.
  • OS and OM at the end of a word became US and UM
    • Compare tempus (earlier *tempos) with temporis (earlier *temposis)
    • Or compare the Latin second-declension endings with the Greek ones
  • And many others...

As you can see, following these backward is not a trivial process. (Though it is very very interesting.)


This is almost never necessary.

Latin regularized forms much more than Greek did. In Attic Greek especially, inflections often don't make sense unless you know the underlying forms. But in Latin, these irregularities were smoothed out over time: lār and honor became normal R-stem nouns; flōs and tempus are normal R-stems except for the nominative singular.

So if you know the nominative singular, which is sometimes unusual, and the genitive singular, which never is (because the ending starts with a vowel), you know everything you need to decline the noun properly.


The general rule is that in Latin etymological s becomes r between two vowels. Thus nom. honos, but gen. honoris (but also nom. honor by analogy to the other cases). Words like honestus confirm that this is indeed an s-stem.

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.