While some verbs in french with intransitive sense use "être" for complex tenses (Je suis entré), other verbs with transitive sense use "avoir" to show subject-object relations: Je les ai vus. ("Je" - subject, "les" - object).

Pronominal verbs in french are always directed at oneself, which means they are some kind of transitive and should use "avoir" as well.

So why do they always use "être" as their auxiliary? Is the situation with "etre (to be)" and pronominal verbs the same in other roman languages?

  • 1
    Unrelated, but in German être/sein also is used for verbs expressing movement in perfect tense (wir sind gekommen) - i.o. the common avoir/haben (wir haben geschlafen) – Joop Eggen Aug 25 '17 at 12:18
  • @JoopEggen The same was true in English through EModE. – StoneyB Aug 25 '17 at 17:01

In An Outline of Middle Voice in Syriac by M. Farina, she connects this with unaccusativity (some layout changed in the examples):

Unaccusativity can surface in different ways in languages. In Italian, for example, unaccusative intransitive verbs receive a different auxiliary (essere 'to be') from unergative intransitives (which take avere 'to have'). Moreover, the unergative verbs have the same auxiliary received by active transitive verbs (such as mangiare, 'to eat'), while unaccusatives take the same auxiliary which is used to express the passive. This proximity to the passive shows how the subject of the unaccusative verb is considered as undergoing an action or a process, just like the subject of passive verbs. Both subjects are objects at a deeper level of analysis. The situation of Italian can be exemplified as follows:

Italian partire 'to leave' vs. camminare 'to walk'

  • Stefania ieri sera è partita — "Yesterday evening Stefania left."
  • Stefania ieri sera ha camminato — "Yesterday evening Stefania walked." (with a Terminative Aoristic interpretation)

Italian mangiare 'to eat'

  • La mela è mangiata da Stefania — "The apple is eaten by Stefania."
  • Stefania ha mangiato la mela — "Stefania has eaten the apple."

So, her claim (for which she sadly does not give any references) is that "to be" is used when the syntactical subject is affected / a semantical agent, whereas "to have" is used when the syntactical subject is also the semantical agent.

Pronominal verbs are syntactically "directed at oneself", as you put it, but are semantically often middle voice. They are not true active transitive verbs, since it concerns an affected agent. This affectedness is reflected by using the same auxiliary as the one used for passive constructions.

In An Outline of Middle Voice in Syriac this is really a remotely related issue; Farina uses it just to explain the concept of unaccusativity and how it can surface in a language. So, surely there are better references for this available (which also compare other languages), but this is not my field so I wouldn't know.

  • This seems to make sense. The possessive past comes from the expressions like "have something done" etc. while such a construction would probably emerge withe less likelihood in place where the possessed object is the same as the subject (you do not have yourself, you simply are). – Eleshar Aug 28 '17 at 21:27

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.