From ob ("towards") + ferō ("bear, carry")


From ab ("from") + ferō ("bear, carry")

Both prefixes of them end with "-b-", but why do their compounds differ from each other, namely "-b-" vs "-u-"? Is this a regular sound change?

Another question is L "ostendo",


From obs- for ob ("before") + tendō ("I stretch").


From ob + teneō ("hold; restrain").

Finally, L "obs-" turned into L "os-" in "ostendo", while "ob-" survived in "obtineo", is that also a regular sound change? Can anyone show me some other examples?

  • 2
    Obstineo is also possible (think obstinacy), so I think different transformations happened depending on what period the prefigation happened in. As to aufero, that is probably mainly to avoid confusion with affero, from ad-fero, contrastive dissimulation or whatever it's called. I can't think of any other example where ab- becomes au-. Nor obs- => os-: I think that one is just irregular, possibly dialectical, I don't know. – Cerberus Mar 20 '13 at 3:18
  • Oh, there is aufugio. And affugio doesn't exist, so it's probably not because of contrast. Perhaps by analogy, from aufero? Either way, au- for a(b(s))- is not regular. // I can't find any other example of a verb with os-. So os- for obs- is not regular, but then not everything is regular in Latin. Think of compounds with manus, like mancipium... – Cerberus Mar 20 '13 at 3:25
  • It's best to not tack on additional questions to your main question. You should ask one thing. Your main question is also not entirely clear. You wonder if "this" is a regular sound change - but what does "this" refer to? – Sverre Mar 21 '13 at 14:45
  • @Sverre, Thanks for your advice, in the main question, "this" refers to the change from "ab-" > "au-" – archenoo Mar 21 '13 at 16:28

In Latin, there was total regressive assimilation in a combination of an occlusive (Verschlusslaut) followed by f:

OCC+f> -ff-


affero < *at-fero < *ad-fero (recomposition also possible - adfero)

offero < *op-fero (recomposition also possible - obfero)

effero < ec-fero (recomposition also possible - ecfero)

cf. Weiss 2009/2011: 172, "Latin does not normally have f in non-initial positions. The few examples are in compounds and the family of inferus."

Also note that the group "bt" in Latin was most likely pronounced [pt] (Tronskii 1960: 114). Prefixes ab-, ob-, and sub- are usually reconstructed as PIt. *ap-, *op-, and *sup- (Tronskii 1960: 113, de Vaan 2008).

As for aufero (and aufugio), here's what you can find in de Vaan 2008:


In a combination of a labial/velar followed by s+C, the first consonant (labial or velar) was lost:

ostendo < *ops-tendo

suscipio < *sups-capio

suspicio < *sub-specio etc.

cf. Leumann 1977 "Neben ab ob sub stehen die mit s erweiterten Formen abs- obs- subs- bzw. aps- ops- sups-, fast nur als Praeverbien [...] Die s-Formen stehen nur vor Tenues, besonders in altertuemlichen Komposita; oefters sind sie vereinfacht zu os- sus-, vor p auch as- [...]" (p. 157).

Ernout & Meillet 1985/2001, "ce dernier [obs-; Alex B.] usité seulement en composition; [..] il est souvent réduit à os-" (p. 454).

So, as you can see (Ernout & Meillet 1985/2001, Leumann 1997), the form obs- was usually used in compounds.


Ernout, Alfred, Antoine Meillet, and Jacques André. 2001. Dictionnaire étymologique de la langue latine: Histoire des mots. Paris: Klincksieck.

Leumann 1977, § 203. - Leumann, Manu, J. B. Hofmann, Anton Szantyr, and Friedrich Stolz. 1977. Lateinische Grammatik: Lateinische Laut- und Formenlehre. München: Beck.

Pfister 1977, §137.2. - Pfister, Raimund. 1977. Handbuch der lateinischen Laut- und Formenlehre. Band 1, Einleitung und Lautlehre. Heidelberg: Winter.

Tronskii 1960, §232, 268. - Tronskiĭ, I. M. 1960. Istoricheskai︠a︡ grammatika latinskogo i︠a︡zyka. Moskva: Izd-vo lit-ry na inostrannykh i︠a︡zykakh.

Vaan, Michiel Arnoud Cor de. 2008. Etymological dictionary of Latin and the other Italic languages. Leiden: Brill.

Weiss, Michael L. 2011. Outline of the historical and comparative grammar of Latin. Ann Arbor: Beech Stave Press.

  • 1
    Epic answer! And thank you so much!!! May I ask you the origin of L "ecfero"? Maybe it derives from ex- + fero, but why does the "x" change to "c"? – archenoo Mar 21 '13 at 16:34
  • Alex, fantastic answer! @archenoo: In many words, ec- is from Greek: ecloga, ecclesia, ecdicus, etc. Because pherô means the same thing in Greek, it is possible that Latin somehow borrowed this from Greek (also possible for ecfatus), but I doubt it. There is probably some Proto-Latin sound change at work. – Cerberus May 9 '13 at 5:54
  • @ewawe re: suspicio < *sups-capio. Sorry, I meant to write suscipio - it was a typo, I fixed that. – Alex B. Apr 3 at 15:02
  • @ewawe re: a velar being lost before sC: this is a well-known sound change, see e.g. Weiss 2020, page 197, rule D5. *TsT > -sT- (T stands for tenues). There are two exceptions listed in Tronskii 1960 (paragraph 268): cst is preserved or restored in formal (literary) Latin after a short vowel, cf. dexter vs. lustrum or Sextus vs. Sestius. – Alex B. Apr 3 at 15:19
  • 1
    @AlexB.: Thanks, I didn't think about the possible role of vowel length in affecting this change – brass tacks Apr 3 at 19:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.