I'm reading Ανάβασις by Ξενοφών. I just found this out when I came across this sentence:

"κακῶς γὰρ τῶν ἡμετέρων ἐχόντων πάντες οὗτοι οὓς ὁρᾶτε βάρβαροι πολεμιώτεροι ἡμῖν ἔσονται τῶν παρὰ βασιλεῖ ὄντων."

"ἔσονται" is 3pl fut. midd. ind. There is no active or passive voice (at least in Wiktionary). But there is also a fut. midd. optative. From the above sentence it could be tempting to think that the middle voice here is somehow used to convey "conjecture". But I believe that's the job of the subjunctive or optative, i.e. mood, not voice. The present and imperfect of εἰμί both use the active voice.

At my learning level I think of the optative a bit like a souped-up version of the subjunctive: in English terms "might" rather than "may". French imperfect subjunctive also seems to correspond quite closely to the optative.

But the choice of voice, and particularly the distinction between middle and passive (and indeed mediopassive), is still something I often find quite tricky to understand.

Is there any reason why the middle voice should be "naturally" the best fit for the future of εἰμί? Or is this just one of those "that's just the way it is" things?

  • Important for the answer: are you familiar with deponent verbs?
    – Draconis
    Dec 29, 2023 at 21:56
  • Yes indeed! ... Dec 29, 2023 at 22:20
  • 1
    Short answer: many verbs become deponent only in the future.
    – Cerberus
    Dec 30, 2023 at 5:15
  • 1
    @Cerberus That's not an explanation so much as just a differently worded statement of the phenomenon, though. The likely reason for it is that the Greek sigmatic future (probably) continues a PIE desiderative, which semantically goes well with a middle voice. A proper answer is beyond me at this time, though.
    – Cairnarvon
    Dec 30, 2023 at 6:26
  • @Cairnarvon: Well, the question asks about the translation and function of the middle voice in the future synchronically, not about its history. I agree that a good, full answer would include history, which I should also like to read.
    – Cerberus
    Dec 30, 2023 at 17:07

1 Answer 1


Basically, yes, it's just one of those "that's just the way it is" things.

The descriptive synchronic answer, as Cerberus says, is simply that many verbs have middle deponent futures. You just have to memorize this as one of the principal parts -- there's no way to predict it. There is a rough generalization that many such verbs are intransitive (like εἰμί - ἔσομαι) or at least often used intransitively (like ἀκούω - ἀκούσομαι), but some are not (e.g. λαμβάνω - λήψομαι).

But "everything is what it is because it got that way", so there are theories about the historical origin of this anomaly. The traditional one, as Cairnarvon says, is that these futures go back to a Proto-Indo-European desiderative form, expressing the idea "want to (verb)". The semantic development from "wants to do" to "will do" is a common one, as the English future with will itself illustrates. The middle endings are then supposed to be explained by the fact that a desiderative is inherently subject-focused (one typically wants to do things for one's own interest) and that this is one of the core meanings of the middle voice.

However, Andreas Willi, in his magisterial Origins of the Greek Verb (2018: pp. 443ff.), rejects this theory and doubts the existence of a PIE desiderative in -s-. For Willi the deponent futures continue subjunctives of sigmatic aorists. His explanation of why they use the middle voice is rather involved, but basically the idea is that the sigmatic aorist inherently expressed high transitivity, so when such subjunctives/futures were formed to intransitive or low-transitivity verbs, the middle endings were used to "cancel out" the high transitivity, and this became a default future formation which then spread to transitive verbs like λαμβάνω.

(The existence of a future optative ἐσοίμην isn't relevant here, as all verbs at least in theory form future optatives, not just verbs with middle futures.)

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.