Latin has some suffixes that turn nouns into adjectives. But there doesn't seem to be much rhyme or reason to which suffixes get applied to which nouns. For example:

  • felis->felinus
  • canis->caninus


  • apis->apianus
  • avis->avianus

Does anyone know where this distinction comes from? Was there perhaps an ending in Old Latin that influenced the choice of suffix and then disappeared?


I don't have a source for this at hand, but I would assume that the -anus suffix originated in a-stems: that is, the original suffix was -nus, but added to an a-stem it was -a-nus. This -anus would then have spread by analogy to nouns that aren't a-stems, too.

This is a common type of process; for a close parallel cf. the Greek feminine suffix -ssa, which originally arose by the addition of *-ya to t- and k-stems (e.g. *melit-ya > melissa "bee", *Phoinik-ya > Phoinissa "Phoenician woman"), but then became productive with all kinds of nouns (e.g. basilissa "queen").

  • Sure. The -a stem verbs were all regular, for instance, and any borrowed or derived verbs (repetitives in -to, for instance -- cano/canto) all were regular first-conjugation verbs. For derived nouns, the -a/-us thematic declensions were also the model. That's the norm; in English reified irregular nouns or verbs become regular (Toronto Maple Leafs/*Leaves) – jlawler Aug 14 '14 at 2:17
  • Can you be a bit more specific in this case? What would the old stems have been for apis and avis? – Anschel Schaffer-Cohen Aug 14 '14 at 4:16
  • @AnschelSchaffer-Cohen I'm not sure if your comment is addressed to me or jlawler, but if the former: it's not that apis and avis once had different stems, but that the suffix originated in a different stem class and then spread out from there. – TKR Aug 14 '14 at 16:41
  • OK. But is there any good reason why it spread to some words and not others? There doesn't seem (to me) to be any phonological or semantic distinction at play. – Anschel Schaffer-Cohen Aug 15 '14 at 11:33
  • @AnschelSchaffer-Cohen I don't know; I haven't look at the data. These kind of analogical extensions tend not to be too predictable. – TKR Aug 15 '14 at 18:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.