While the answer should be obvious because of the genitive in -us is for the fourth declension, my question is why they are classified as the fourth declension, seeing that they are all Greek nouns and in Greek there are only three declensions.

In my textbook from school time (a very basic one), in the chapter about the fourth declension, no nouns with nominative in -o are mentioned. They are instead mentioned in the chapter about the Latinised Greek declensions, specifically in the third declension. Only Sappho, -us and Dido, -us, though, explaining that Dido can also be Dido(n), Didonis.

Being a basic textbook, it could be that not too much detail was provided to avoid confusion. However, looking on the Internet, I couldn’t find a satisfying explanation for that.

In the fourth declension feminine noun lists on Wiktionary there are 15 nouns ending in -o, all proper nouns and all derived from Greek. By checking them, all have the -o ending for all cases with the exception of the genitive in -us. Two of them have a Latinised third declension alternative (Dido, Didonis and Io, Ionis. Sappho, Sapphonis is mentioned as well but this has been addressed in another question), and one can have an alternative accusative in -on (Echo, Echus, so Echon) that corresponds to a second Greek declension I think.

Checking Wikipedia here, it lists as example for fourth declension Greek nouns the word echo (not as proper name) that has the normal Latin fourth declension endings with the exception of the nominative in -o. Wikipedia also lists down in the page nouns belonging to a mixed declension where Dido, -us is mentioned.

Checking an online Latin dictionary for Echo, it specifies that is a third declension noun and that only the nominative Echo and the accusative Echon are attested. I checked also Sappho and it says again that is a third declension noun.

I understand that the genitive -us does not leave space to doubts, but this class of nouns seems odd to me. Is it that they were just forced into the fourth declension because of the Greek genitive sound assimilated to -us (I don't know Greek)? Is there any source that is not Wikipedia I could check for that?

1 Answer 1


They don't really belong to any Latin declension; they're just the Greek forms taken over entirely unchanged. Looking at perhaps the most broadly attested name:

Greek Latin
sg . nom. Διδώ Dīdō
gen. Διδοῦς Dīdūs
dat. Διδοῖ (unattested)
acc. Διδώ (Attic), Διδοῦν (Ionic), Διδών (Doric) Dīdō, Dīdūn, Dīdōn
abl. (none) (unattested)

Some dictionaries do, as you've observed, put these words in the fourth declension faute de mieux based entirely on the fact that they look like they have a genitive in -ūs, same as the real fourth declension nouns, but they're not meaningfully Latin words. They're in the third declension in Greek, but that's neither here nor there.

Some of the more common ones do get adapted to the Latin third declension, as Dīdō, gen. Dīdōnis did, but it doesn't then make sense to view those new third-declension forms as part of the same paradigm as the unadapted Greek forms, and I think that's one of the places where that Wikipedia article goes badly wrong.
(The other is that it tries to construct complete regular paradigms out of singular forms that certainly don't all fully show up in the corpus.)

  • 1
    Are the dative and ablative forms unattested for just Dido or all Greek names of this type?
    – Joonas Ilmavirta
    Jan 13 at 13:29
  • 2
    @JoonasIlmavirta According to L&S Pliny has an ablative Sappho and Dido herself has a disputed dative Dido, but not, AFAIK, anything else. The dative is an interesting question because -oi should be phonotactically disallowed in Latin; you probably would expect actually (or for old words), and that might be the source of Pliny's ablative.
    – Cairnarvon
    Jan 13 at 14:00
  • 1
    Why not a theoretical -oe for the dative, which Latin did use for some words they borrowed with -οι- in it (e.g. oeconomia)?
    – cmw
    Jan 14 at 21:36
  • @cmw I guess this is hard to prove, but Latin is more constrained in word-final sounds than medial ones, and original medial and final oi developed differently (-oi- became -ū- but was sometimes preserved, -oi became and never was, though obviously the morphology interferes) so it feels wrong to me. But I went and checked and nearly all the -oe words in L&S have a long and reflect Greek -οη or -ωη (Noe reflects Νῶε). One does reflect Greek -οι, and that's the interjection euhoe (εύοί) ...
    – Cairnarvon
    Jan 15 at 8:06
  • ... (Terence also has an interjection oi.) The Salian fragment pilumnoe poploe presumably also reflects Old Latin -oi. None of these are especially integrated into the language even when compared to these forms of Dido, of course. It's all interesting but not especially conclusive.
    – Cairnarvon
    Jan 15 at 8:10

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.