Wiktionary gives us the following etymology of amare.

Probably from Proto-Indo-European *am-a-, *am- (“mother, aunt”), a lost nursery-word of the papa-type. Compare amita (“aunt”), Old High German amma (“nurse”). Alternatively, O. Hackstein suggests Proto-Indo-European *h₂emh₃- (“seize”).

There's a fascinating conjecture in this etymology that amare comes from a nursery word meaning "mother" or "aunt", which I take to be a babble word such as "ma-ma". But unfortunately, I do not find this theory in de Vaan's entry.

amo, -are 'to love' [v. I] (Naev.+) Derivatives: amascere 'to begin to love' (Naev.), amasius 'lover' (PL+); amicus friend' (Naev.+), 'friendly' [o/a] (PL+), arnica 'female friend' (Naev.+), inimlcus [adj. / m.] 'unfriendly; an enemy (P1.+), inimicitia 'enmity, ill will' (P1.+); amor 'sexual passion, love' (Naev.+).
Pit. *ama- 'to take, hold\ It. cognates: Marr. amatens 'they have received'.
PIE *h3mh3- to take hold of. IE cognates: Olr. namae 'enemy' (< *n-AjmAj-^«/-); Skt pr. amfsi, amanti 'to take hold of; swear', ama- [m.], OAv. §ma-9 YAv. ama- [m.] 'attacking power'; Gr. όμνυμι 'to swear' [aor. όμόσαι, fut. όμοΰμαι] , άνώμοτος 'not under oath'.
The Latin meaning has developed from 'to take the hand of > 'regard as a friend'. Pit. *ama- could reflect PIE *h3mh3- in front of a vowel, cf. Schrijver 1991: 318. Schrijver 1991: 398ff. argues that the stative meaning of amare and the presence of a derivative amor point to a stative verb *ama-e-. Its formation must then post-date Pit., since Marr. amatens is not likely to have a stative suffix *-e-3 and has the transitive meaning * receive' of the IE cognates. The form amasius, because of its s, seems to be dialectal; maybe it is the same suffix as CLat. -arius. The adjectival function of amicus is original; the suffix -icus might be decasuative from an ins. in *-/A; plus *-Ao- (parallel cases in Latin are pudicus 'chaste' and mendicus 'needy'). The PIE root is reconstructed with initial *h2- in LIV, but Gr. points to *h3-,
Bibl.: WH 1:40, EM 29, IEW 778, Rix 1999: 523-24, LIV *h2emh3-. -> amplus

De Vaan gives the PIE root *h3mh3- ("to take hold of") for its etymology. This could be consistent with a babble word, but there is no explicit mention.

Is there any plausibility behind the idea that amare comes from a nursery word meaning "mother"? If so, how probable is this conjecture? Is there a consensus?

  • 4
    You should change the question to get rid of amare and instead talk about the root of amare, ama-.
    – cmw
    Oct 31, 2017 at 20:26
  • @cmw That's an interesting suggestion. I'll think about it and perhaps revise when I have time.
    – ktm5124
    Oct 31, 2017 at 20:46

1 Answer 1


Regarding the infinitive specifically, as opposed to conjugated forms of amare "to love", which is a generally accurate definition of mother in relation to her infant:

The “R” sound is hard for some children because it is difficult to see the tongue when you say it and it is hard to explain to a child how to make it... Many children can say a correct “R” sound by the time they are five and a half years old, but some do not produce it until they are seven years old.
SOURCE: Why is “R” So Hard to Say? K.F. Bedsole, M.S., CCC-SLP and C.M. Johnson, M.A., CCC-SLP

"t" sounds would also be quite difficult, but the "o" sound is more accessible because the infant can see the shape of the mouth.

  • "ama" is quite common as baby babble, in many ways "ma" is the easiest word because it only requires opening the mouth in a fully relaxed state. As one of the first utterances humans will make, I think it's a reasonable assumption that amare might partly derive from "ma" or "ama".

What I mean by reasonable assumption is that adults, in my experience, generally try to interpret baby babble. In this case, the adult might assume the infant is trying to say "you love" (amas), which, again, is a generally accurate description of a mother in regards to her infant.

  • 5
    Thanks for your answer. I think I might disagree with your grounds for ruling out the conjecture. I believe the word would have evolved to take on the "r" sound in Latin. After all, -are is an infinitival suffix, not a root, and the root would have existed long before Latin came about, i.e. in Proto-Indo-European.
    – ktm5124
    Oct 31, 2017 at 20:08
  • @ktm5124 It's an interesting idea in that an infant's first love might be thought of as for the one who suckles the child, but as you note, it would have to originate from "ama" only.
    – DukeZhou
    Oct 31, 2017 at 20:11
  • Indeed! But I find it plausible that this happened. Because de Vaan gives a root very similar to the am- sound... he gives a laryngeal followed by the consonant "m", which seems consistent with am- or ama-.
    – ktm5124
    Oct 31, 2017 at 20:13
  • @ktm5124 See: en.wikipedia.org/wiki/Amah_(term)
    – DukeZhou
    Oct 31, 2017 at 20:13
  • 3
    If I have understood correctly, an older form of the infinitive is amase which became the classical one we know by rhotacism. And more importantly, the infinitive is quite irrelevant for simple use, and a child using forms like amo, amas, and amat would have no need to fight with the R.
    – Joonas Ilmavirta
    Oct 31, 2017 at 20:22

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.