I have seen two versions:

a) *-ia ending actually derived from the collective number form, which also ends in *-ia. So the collective number first started to represent abstract things (compare Latin animalia, Greek zooia), and then came to mark feminine gender.

b) From Eurasiatic *haia, "mother", from Borean *haia, "mother" (see Nostratic etymology) which explains similar feminine *-ia endings in Semitic, IE and Altaic.

The account for Laryngeal theory is only compatible with version a) and cannot explain how Eurasiatic *-ia became PIE *-ih2.

  • 5
    most linguists do not accept the validity of genetic groupings higher than IE, but you might find someone familiar with the super-lumping hypotheses who will answer.
    – user483
    Jan 11, 2012 at 3:55
  • 1
    Good question! Just a side note: as far as I can tell, tetrapoda comes from Greek, not Latin. Jan 11, 2012 at 10:34
  • 1
    if the late PIE m/f/n gender system arose from an earlier two-way animacy system as is often conjectured, it seems like that presents an additional difficulty for option (b). But in support of (a), I have seen this theory of the collective ending leading to the feminine before, e.g. at the sci.lang FAQ #28 Jan 12, 2012 at 2:21
  • 3
    Looking at the link you posted, all I can see is Eurasiastic *ʔVjV ? and Borean *HVJV. Doesn't V there stand for an unspecified vowel?
    – Alex B.
    Jan 12, 2012 at 16:42
  • The last sentence looks like it's describing a difficulty for laryngeal theory, but if anything, it's describing a difficulty for the Euroasiatic theory.
    – TKR
    Dec 21, 2018 at 4:43

1 Answer 1


It is impossible to really know for sure how a feminine marker "originated" in Indo-European, since the exact nature of gender there is unknown, since Hittite conveys only two genders (or states): animate and inanimate. Furthermore, "animalia" is a neuter plural. Nevertheless, it suffices to say that the Semitic ending in Arabic ة-/ية (-ah/iyah) is not at all related to the Indo-European one. Take for example a Berber word for "house": "tamdint"; the t-t indicates a feminine. The root is "mdn" like "مدينة" (medīnah, city) in Arabic, which becomes "مدينات" (medīnāt, cities) in the plural. Therefore the Eurasiatic theory is far-fetched. I might not have fully answered your question, but at least this hopefully narrows it. I think it is important to ponder this sort of thing, but the answer will likely never be known.

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.