Latin effectively lost its dual number. It left behind some remnants, most notably duo and ambo. However, all examples or relics of the dual number in Latin I have seen are in declension. I would assume the grammatical number to be present in conjugation also, so that between unus canit and quinque canunt there would be duo canX, where the X stands for a third person dual ending.

Classical or later Latin as I know it does not really have a dual; all examples of dual forms can be practically treated as irregular plurals. This leads me to phrase my confusion as this series of questions:

  • Assuming there ever was such a thing as dual conjugation, when was it lost?
  • Did Proto-Indo-European ever have it to begin with?
  • Are there remains of originally dual conjugation that survived in Latin, presumably as plural forms?

1 Answer 1


PIE appears to have had dual verb forms, as can be seen from e.g. Greek ἐστόν "you two are", Sanskrit ithás "you two go", Gothic baírats "you two carry". (Anatolian, though, lacks dual forms, which makes it uncertain whether these existed in PIE and were lost in Anatolian, or only developed after Anatolian had branched off from the rest of the IE languages; the same question arises for many other morphological features that are absent from Anatolian.)

The dual (in both verbs and nouns) disappeared en route to Italic -- neither Latin nor any of its Italic sister languages retain it. It's possible that there is at least one relic of the PIE dual in the Latin verb system, though: namely the 2pl. ending -tis, which has sometimes been explained as originating in the dual ending. (Cf. the -thás and -ts- of the Sanskrit and Gothic forms above.) But this is controversial.

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.