The verb εὔχομαι means "to pray", and it shows up before the Lord's Prayer (Matthew 6:9-13) as προσεύχεσθε. I was curious to learn more about this word, so of course I looked it up in Wiktionary, and found this:

From Proto-Indo-European *h₁wegʷʰ-. Cognates include Sanskrit ओहते (óhate), वाघत् (vāghat), Avestan 𐬀𐬊𐬘𐬀𐬌𐬙𐬈 (aoǰaite), Old Armenian գոգեմ (gogem) and Latin voveō.

εὔχομαι • (eúkhomai)

to pray, offer prayers

I was surprised to find the Latin word voveo listed as a cognate. Could someone explain how these two words share similar descent? Unfortunately, I do not know how to read IE asterisk notation (is there a more appropriate name for this?) so I request that you go easy on me :)

  • 1
    It's not that important, but the Avestan script is not visible to me (unless they used empty boxes). If you want exotic scripts to be visible, it's best to include them as images, but it's your call to decide if it's worth the trouble.
    – Joonas Ilmavirta
    Commented Mar 30, 2017 at 7:11
  • @JoonasIlmavirta Good to know! I think I can read the Avestan script owing to some fonts that Cerberus recommended in the chat room.
    – ktm5124
    Commented Mar 30, 2017 at 7:21
  • 3
    @JoonasIlmavirta I am reading the question from my iPhone and the Avestan script is correctly visualized. Perhaps you should upgrade your system or simply install a good Unicode-savvy font (you can try Google Noto that has special fonts for many ancient languages). Commented Mar 30, 2017 at 15:17
  • @JoonasIlmavirta Not to rub it in... But the Avestan script is the most elegant. You are missing out!
    – ktm5124
    Commented Mar 30, 2017 at 18:12

1 Answer 1


The reconstructed Proto-Indo-European form listed is *h₁wegʷʰ- "to promise, vow, praise".

  • h₁ is a "laryngeal consonant", so-called because we don't know what else to call it. The most common theory is that there were three consonants that acted similarly, and since nobody knew what they were, they were written as h with a subscript. h₁ was most likely either like Latin h, or a glottal stop.
  • w is a "consonantal u", like Latin v.
  • e is a short vowel probably like Latin e.
  • gʷʰ is the weirdest one. The standard reconstruction of PIE involves voiced aspirated consonants, but these are not particularly common, and this analysis leads to several other very strange features (such as the lack of a b sound). Other theories have been put forth but they have their own flaws. So in the standard theory, gʷʰ is like Latin qu, but voiced and aspirated.
  • And the hyphen at the end means this is just a stem.

Latin derivation

One particular form of this verb was the causative, *h₁wogʷʰ-éye-.

  • The e became an o, which is a common though not entirely understood occurrence in PIE: consider Greek leg-ō and log-os. It's generally called Ablaut because it was first noted by German linguists.

This led to a Proto-Italic form *wogʷ-e-ō.

  • The laryngeal disappeared entirely, which they tend to do.
  • The gʷʰ lost its aspiration, which regularly happened in Italic.
  • was a standard verb ending, as seen in Greek and Latin.

And this became Latin vov-eō.

  • became plain w, written as v in Latin. This was again a regular process.

Greek derivation

Another form of this verb was *h₁ewgʰ-.

  • The e and w switched places, a not-unheard-of process called metathesis.
  • Then the gʷʰ lost its ʷ when right next to the w.
    • This is called the Boukolos Rule, because the first example discovered was Greek βουκόλος = boukólos < *gʷou-kolos < *gʷou-kʷolos.

This became eukʰ- in Hellenic.

  • The laryngeal vanished as usual.
  • Greek kept the aspiration distinctions that Latin lost, but removed voicing from aspirated consonants.
  • The consonant w became the vowel u.

In Greek, this was written eúkhomai (εὔχομαι).

  • -o-mai is simply a thematic vowel + personal ending.
  • 1
    Also: do you know of a good resource that explains the conventions of written PIE forms? It's something I've wanted to learn more about for a long time.
    – brianpck
    Commented Mar 30, 2017 at 16:49
  • 1
    @ktm5124 The voiced aspirates are quite diverse like that! In Latin a whole bunch of them became f, while in Greek they mostly turned into phi, theta, chi. Also if there weren't metathesis we might see a tau in the Greek: labialized consonants (kw, gw) tended to become taus, as in tis ~ Latin quis and friends.
    – Draconis
    Commented Mar 30, 2017 at 17:00
  • 3
    Beekes argues that the Greek word is from the reduplicated stem *h₁e-h₁ugʷʰ-e-. This accounts for the initial diphthong.
    – fdb
    Commented Mar 30, 2017 at 17:30
  • 2
    @fdb, that does seem potentially a better theory than unexplained metathesis / Schwebeablaut. The problem, though, is that reduplicated presents in Greek have Ci-, not Ce-.
    – TKR
    Commented Mar 30, 2017 at 18:01
  • 1
    I don't know that there's a reason to think that gʷʰ merged into gʷ in Italic before becoming Latin w -- the PIE voiced aspirates didn't deaspirate in Italic, they spirantized (and later re-hardened intervocalically into stops -- at least this seems to be the most common theory). (Btw, yes, the Boukolos rule is what explains the loss of labialization in Greek.)
    – TKR
    Commented Mar 30, 2017 at 19:04

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.