Can the name of the Greek goddess Demeter come from PIE word for tamer, dōma̯tēr (especially given the Aeolic form of the goddess' name, Δωμάτηρ)? I am interested in both whether it is possible from sound change perspective and semantically.

3 Answers 3


This etymology seems unlikely to me for several reasons:

  1. A long o-grade in a Greek agent noun in -ter is at best unusual. I can't think of another such example, at any rate.
  2. The -a- of the second syllable is long, as shown by Attic-Ionic Δημήτηρ. Of course, one could invoke folk etymology here, so this isn't conclusive.
  3. The vowel in the first syllable in the more common forms of the name, Δημήτηρ / Δαμάτηρ, cannot come from PIE *ō.
  4. It's pretty clear that Δαμάτηρ is a compound of Δα-μάτηρ, since the first element seems to occur in the name of Poseidon: Potei-dāōn 'husband of Da' (which he isn't in the traditional mythology, but seems to have formerly been).
  5. It's not clear why an earth goddess should be named 'Tamer'.
  • Regarding the point 1 and 3, I actually do not know whether there was long o in PIE. Definitely the word for house had it, but in the word for tamer I speculated long o only based on the Aeolic form.
    – Anixx
    Commented Jan 27, 2015 at 9:17
  • I have doubts Poseidon is related to Demeter, as Poseidon has a digamma in the middle as evident from Mycenean Greek, that is the original form is *Poteidawōn ("a" isn't long!) I'm not sure there are examples where -aw- collapses into -ā-, that is: *Dawmāter > *Dāmāter
    – carsten
    Commented Jan 30, 2015 at 22:49
  • @carsten The Mycenaean form, AFAIK, is Poseidaon, without a w; there is a Corinthian form of the name which has -w-, but this has to be innovative. The length of the -a- is shown by Homeric meter.
    – TKR
    Commented Jan 31, 2015 at 1:04
  • @TKR Hiatuses like "ao" usually appeared due to a loss of intervocalic -w-, -j- or -s-; Corinthian as you mention has -w-, how can this be a coincidence? As for length of -a-, I'm not sure
    – carsten
    Commented Jan 31, 2015 at 19:38
  • @carsten But -ao- would appear naturally if the derivation is Dā-ōn. Whatever the origin of the Corinthian -w- it can't be old, as the Mycenaean spellings show. Not sure what you mean about the length of -a-, since it's clearly long in Homer.
    – TKR
    Commented Jan 31, 2015 at 21:28

Most probably her name derives from PIE *dhǵhōm-. Kretschmer suggests that δᾶ (Δαματήρ) is a pre-Greek version of γη/γᾶ, but I tend to believe Heubeck's (1961) and others suggestion of an Indo-European root for 'earth'. Phrygian has

Γδαν μα

so that speaks too for a IE-root. Δᾶ seems to be the Doric version of γη/γᾶ.

Now in order to be cautious, we should remember that Phrygian could (or should) have shared substratum with Greek, hence that wouldn't eliminate Kretschmer's theory. However, I don't see much support for it by others.

  • Do u mean dhǵhem-?
    – Anixx
    Commented Jan 26, 2015 at 13:19
  • What can u say about the etymology suggested in the question?
    – Anixx
    Commented Jan 26, 2015 at 13:21
  • @Anixx: Yes, sorry! I will correct it. dhǵhem-, dhǵhom- is should be. It is written like that (with an "n") in Beekes etymological dictionary of Greek.
    – Midas
    Commented Jan 26, 2015 at 13:24
  • @Anixx: Let me have a closer look and I will get back to you later today. The meaning 'earth mother' feels natural in terms of Greek etymology, but I would like to examine what you suggest in depth.
    – Midas
    Commented Jan 26, 2015 at 13:28
  • This root gave Greek khamai, khthon. Isee no retention of d- in Greek.
    – Anixx
    Commented Jan 26, 2015 at 13:31

Both the name and the cult suggest a connection with the Sumerian god Dumuzi. Another tradition that may be at the root of Demeter is that of Tiamat. This may also explain her link to Poseidon.

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.