middle voice in Latin

Question

Does the sequence esse plus past participle (of a non-deponent Verb) occure in middle function in latin? Is the middle function restricted to the mediopassive r-form in the imperfective tenses (infectum) or does the middle function also occur in the perfective tenses (perfectum)?

Answer 1

The "middle function" can also be claimed to occur in perfective tenses (of both non-deponent and deponent verbs). For example, it's one of the three readings an ambiguous sentence like Porta clausa est can have: please see the so-called anticausative reading (i.e., 'The door closed by itself') in this previous question .

Although it is not morphologically distinct from them (in the sense that the three of them involve 'be+participle'), note that the anticausative/"middle" reading (i.e., 'The door closed by itself') is different from the verbal eventive passive reading (i.e., 'The door {was/has been} closed (by someone/by something')) and from the adjectival resultative/stative passive reading (i.e., 'The door is closed').

For example, note that an agentive by-phrase is only possible with eventive passives but it is excluded from anticausatives/"middles" ('*The door closed by someone') and from statives ('*The door is closed-Adj by someone'). NB: the use of asterisk indicates ungrammaticality. Same in Latin: if you add an ablative phrase like a Gaio to the sentence Porta clausa est, the anticausative/"middle" and the stative readings become impossible. Similarly, if you add a subordinate final clause (e.g., introduced by the conjunction ut) or an adverb oriented to the agent (e.g., libenter), same happens again: the anticausative/"middle" and the stative readings become impossible. So an example like Porta libenter clausa est is not ambiguous (only the verbal passive reading is possible). Please see the link above for more discussion.

Here is another example where the "middle"/anticausative reading is possible:

Postquam omnis res mea Ianum ad medium fracta est, aliena negotia curo, excussus propriis. (Hor. Sat. II 3.18-20) ‘After all my business collapsed at the central arcade of Janus…’

---

Lisa's examples in her comment above, which are taken from the great grammar by Kühner & Stegmann (1955: 104f.), are indeed very relevant. Note that some of them become even more interesting if more context is provided. For example, a nice contrast ("middle" vs. passive) is provided by the following example from Livy, where circumactus est is interpreted as a "middle"/anticausative, whereas reducerentur (sc. 'by someone') is a verbal eventive passive.

prius circumactus est annus quam a Velitris reducerentur legiones (Liv. 6, 38, 1). 'A year rolled round before the legions could be brought back from Velitrae' (transl. Benjamin Oliver Foster, Ph.D., 1924)

Similarly, note that the next example from Livy contains a VERY nice pair between the "middle" (i.e., anticausative) reading of a non-deponent verb (suffusus (sc. est)) and that of a deponent verb (obortae (sc. sunt)):

Masinissae haec audienti non rubor solum suffusus, sed lacrimae etiam obortae. (Livy 30, 15, 1). 'On hearing these words not only did Masinissa blush, but tears also came' (Loeb transl.).

NOTA OPTIME: these two examples from Livy are very interesting since circumactus est and suffusus are NOT ambiguous: i.e., these forms (of non-deponent verbs) can ONLY be interpreted here as having a "middle"/anticausative reading.

Answer 2

In this paper (a book actually) it is claimed that the analytic perfect reflects an inactive structure (p. 85). You might find it interesting; the author, Laura Migliori, also explores the syntax of Latin –r forms.