as you can see from the title I would like to know what was the original Germanic agent suffix before Proto-Germanic speakers borrowed -er from Latin speakers. All I know is -a in Old English and I think it's just like British pronunciation of -er (I know it's was only for masculine agents) but I don't know about feminine agent suffix, other Germanic languages and Proto-Germanic.

Thanks in advance.

  • I think in old English the feminine agent suffix was -ster
    – Anixx
    Commented Nov 1, 2019 at 10:11
  • 1
    *-ter and *-tor go back to PIE as agentive markers, and showed up in Latin and Germanic. What's one /t/ more or less?
    – jlawler
    Commented Nov 3, 2019 at 2:37
  • It is complicated, basically Germanic languages lacked a kind of uniform agent suffix but employed different methods to denote the agent dependent on conjugation class. This is one of the reasons why a foreign suffix caought on so easily. Commented Nov 3, 2019 at 18:54

2 Answers 2


As in many languages in Germanic the present participle could be used as an agent. In some Germanic languages such as Old English the participle and agent could sometimes be differentiated with the participle ending in '-ende' and the agent in '-end'. So in Beowulf line 254 we get 'būend' (= 'dweller, somebody who dwells') from the verb 'būan' (= 'to dwell'). The present participle is 'būende' (= 'dwelling, who dwells').

  • What about the the endings *-arijaz and *-jo, that Wiktionary mentions as markers of (verb-derived) agent nouns? Commented Dec 19, 2019 at 19:07
  • It seems that it is a French borrowing as in : apprendre (learn) / apprenant (learner). So I doubt this construction came from proto-germanic.
    – amegnunsen
    Commented Dec 19, 2019 at 20:01
  • 2
    @amegnunsen I doubt this is a French borrowing as Beowulf was written well before the Norman conquest. I don't doubt that French, like many tongues, uses the participle as an agent. It is a natural usage in any language where an adjective can stand on its own as a noun. Ringe notes (2014:7.2.3) that 'when pres. participles were extended as *ija-stems in PWGmc ... some nominalized examples were relexified as consonant-stem nouns'. Examples he gives are 'frēond' (friend), 'fēond' (enemy), 'hettend' (enemy), 'ēhtend' (persecutor), 'hælend' (saviour), 'sċieppend' (creator), 'wealdend' (ruler)’
    – Ned
    Commented Dec 22, 2019 at 14:22
  • @Ned You are right. I saw it is the same in Swedish, so that can date back to proto germanic.
    – amegnunsen
    Commented Dec 22, 2019 at 14:57

It's worth reading this article on academia.edu: Gąsiorowski, Piotr "Cherchez la femme: Two Germanic suffixes, one etymology" in "Folia Linguistica" vol. 51 (2017) (Berlin: de Gruyter, 2017) pp 125-147 Basically Gąsiorowski is arguing that the Germanic -er ending is not borrowed but rather that -er and -ster are Verner variants. He argues that the distribution of -er in early Germanic languages does not reflect what one would expect in a borrowed suffix and it's similarity to the Latin -arius is co-incidence. He also notes that the original meaning of '-er' was not as agent but that it was first added to nouns (as -ster still is in English in words such as 'gangster') and only gained it's current agentive meaning after being added to verbal nouns.

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.