Future Imperfect

By: | Post date: 2017-11-11 | Comments: 6 Comments
Posted in categories: Linguistics, Mediaeval Greek, Modern Greek

Latin and Greek both had an indicative tense called the Future Perfect. The tense described a event occurring in future time, but with perfective aspect—something complete in the future. The future perfect fits neatly into the matrix of possible tenses of Greek: it has the reduplication of Greek perfect tenses, but the -s- ending of future tenses.

European languages that have constructions with comparable meaning have had them called the same thing, which is no surprise: the description of verb tense and aspect in European languages was done in the shadow of the description devised for classical languages.

Now, there was a Perfect tense in Greek and Latin, denoting past completed actions, and there was an Imperfect tense in Greek and Latin, denoting past ongoing actions. Moreover, the adjectives perfect and imperfect generalised in meaning, from referring to completeness vs incompleteness, to referring to perfection vs flaw. So “Future Imperfect” was inevitable as a witticism (Get it? It’s Future, right? But it’s Not Perfect At All!!!), and it may be unearthed on Google as both the title of a Star Trek: The Next Generation episode, and a TV Tropes appellation for flawed reconstructions of the past.

It turns out that Coptic has a Future Imperfect tense after all. And as I discovered when going through Humbert’s account of the Greek dative,1 while Greek inscriptions and papyri in Egypt are normally in correct (if often vernacular) Greek, there is an instance of the Future Imperfect that has made it into Greek, in a late Nubian inscription: the fifth century Silko inscription on the Temple of Kalabsha.

The Future Imperfect is a straightforward combination in Coptic, of the imperfect prefix or particle, and the future prefix. If I can reconstruct from Wikipedia’s grammar:

  • nere pə-tʲoeis krine ən-nə-Laos “The Lord was judging the nations”
  • pə-tʲoeis na-krine ən-nə-Laos “The Lord will judge the nations”
  • nere pə-tʲoeis na-krine ən-nə-Laos “The Lord will was judging the nations”

Similarly in the Silko inscription, the future imperfect is a combination of the augment of Greek past tenses (imperfect, aorist, and perfect), and the -s- ending of future tenses: ἐ-φιλονικ-ήσ-ουσιν “will was combatting”.

There’s two copies online of the inscription, one with the Greek in a picture and one with the Greek in a Norwegian blog. This will be the third. And I’m putting the diacritics back in:

Ἐγὼ Σιλκω, βασιλίσκος Νουβάδων καὶ ὅλων τῶν Αἰθιόπων, ἦλθον εἰς Ταλμιν και Ταφιν. Ἅπαξ δύο ἐπολέμησα μετὰ τῶν Βλεμύων, καὶ ὁ θεὸς ἔδωκέν μοι τὸ νίκημα. Μετὰ τῶν τριῶν ἅπαξ ἐνίκησα πάλιν καὶ ἐκράτησα τὰς πόλεις αὐτῶν. Ἐκαθέσθην μετὰ τῶν ὄχλων μου τὸ μὲν πρῶτον ἅπαξ, ἐνίκησα αὐτῶν καὶ αὐτοὶ ἠξίωσάν με. Ἐποίησα εἰρήνην μετ’ αὐτῶν καὶ ὤμοσάν μοι τὰ εἴδωλα αὐτῶν καὶ ἐπίστευσα τὸν ὅρκον αὐτῶν, ὡς καλοί εἰσιν ἄνθρωποι. Ἀναχωρήθην εἰς τὰ ἄνω μέρη μου. Ὅτε ἐγεγονέμην βασιλίσκος, οὐκ ἀπῆλθον ὅλως ὀπίσω ἄλλων βασιλέων, ἀλλὰ ἀκμὴν ἔμπροσθεν αὐτῶν. Οἳ γὰρ φιλονικοῦσιν μετ’ ἐμοῦ, οὐκ ἀφῶ αὐτοὺς καθεζόμενοι εἰς χώραν αὐτῶν, εἰ μὴ κατηξίωσάν με καὶ παρακαλοῦσιν. Ἐγὼ γὰρ εἰς κάτω μέρη λέων εἰμι καὶ εἰς ἄνω μέρη ἄρξ εἰμι. Ἐπολέμησα μετὰ τῶν Βλεμύων ἀπὸ Πριμεως ἕως Τελμεως ἐν ἅπαξ, καὶ οἱ ἄλλοι Νουβάδων ἀνωτέρω ἐπόρθησα τὰς χώρας αὐτῶν, ἐπειδὴ ἐφιλονικήσουσιν μετ’ ἐμοῦ. Οἱ δεσπόται τῶν ἄλλων ἐθνῶν, οἳ φιλονεικοῦσιν μετ’ ἐμοῦ, οὐκ ἀφῶ αὐτοὺς καθεσθῆναι εἰς τὴν σκιάν, εἰ μὴ ὑπὸ ἡλίου ἔξω, καὶ οὐκ ἔδωκαν νηρὸν ἔσω εἰς τὴν οἰκίαν αὐτῶν. Οἱ γὰρ ἀντίδικοί μου ἁρπάζω τῶν γυναικῶν καὶ τὰ παιδία αὐτῶν.

I, Silko, king of the Noubades and all the Ethiopians, came twice to Talmis and Taphis. I warred with the Blemmys and God gave me victory. For a third time I defeated them again, and I took possession of their towns, I occupied them with my troops. At the first time I defeated them, they subjugated themselves to me, and I made peace with them. They swore for me to their idols. And I trusted their oaths and that they were decent people, I withdrew to my upper part of the country. When I became king, I was not ranking behind the other kings, by any means, but rather in front of them. Those who contend with me—I do not let them dwell in their country unless they give me esteem and devotion. For my bottom half is a lion and my top half is a bear. I have warred with the Blemmys from Primis to Telmis at the same time. Also the others, above the Noubades—I have laid waste to their areas because they “will were contending” with me. The rulers of the other peoples who contend with me—I do not allow them to sit in the shade, but keep them rather in the sun outside, and they did not drink water inside their houses. But my adversaries: I drag away their women and children.

All very Ozymandias. But that still doesn’t quite tell us what the future imperfect means. How do Coptic grammars gloss it?

When they get around the awkwardness of accounts like “casting the past into the future”, they turn out to give a straightforward rendering in English:

“I would contend.”

Well shit. By that notion, English has a future imperfect. And Greek definitely has a future imperfect.

Would, after all, is merely the past tense of will. If I will contend is the way of expressing the future in Modern English, then the past tense of will is used to cast the future event into the past. And presenting a future event as being in the past is a strategy for expressing a hypothetical. A future event from the perspective of the present, after all, is not guaranteed to take place: it is a hypothetical that “I will contend”. If we situate that hypothetical in the past, and don’t add the information that it did in fact take place, then we are asserting that the hypothetical remained unrealised. The other tribes would contend with Silko; but he laid waste to them before they got very far.

(The other English form for this makes it even more explict: I was going to contend.)

The same occurs in Modern Greek. The particle θα, derived like will from the verb for “I want to”, combines with present tenses to express futurity. In fact, Modern Greek has three present tenses in the subjunctive, which θα can combine with: a progressive, leading to a future progressive; a punctual (perfective), leading to a future punctual (which is in fact the unmarked tense); and a perfect tense (formed itself with an auxiliary), which leads to a future perfect. θα φιλονικώ “I will be contending”, θα φιλονικήσω “I will contend”, θα έχω φιλονικήσει “I will have contended.”

Already, there’s a mismatch with the traditional, Classical Greek-derived account of tenses. I called the punctual subjunctive φιλονικήσω a present tense. That dates from Hesse’s account of Greek verbs from 1980 at least2 and I wouldn’t be surprised if it dates from structuralists like André Mirambel. It does not match the history of the Greek verb: that form is historically the aorist subjunctive. But the ancient patterning is not the most economical way to make sense of Modern Greek verb tenses; a 2×2 matrix of present vs past, perfective vs imperfective is far more sensible.

But Greek does the same as Coptic and English does, in combining θα with past tenses to express counterfactuals:

  • θα φιλονικούσα (Future + imperfect = past imperfective): I would contend
  • θα είχα φιλονικήσει (Future + pluperfect): I would have contended
  • θα φιλονίκησα (Future + aorist = past perfective): I must have contended (I don’t know that to be the case, but I conclude it is so from indirect evidence)

But traditional grammars of Modern Greek don’t call these tenses Future Imperfect, Future Pluperfect, and Future Aorist. In fact, they don’t list them as tenses at all: you will never see them mentioned in grammars, showing verb conjugation. You will only see them in accounts of syntax.

And the reason for that is that those tenses did not exist in Ancient Greek in the same mood (which is why Ancient Greek never bothered to come up with a Future Imperfect, outside of the L2 Greek of Silko). Ancient Greek certainly had ways of expressing those concepts; but it did not use the indicative to do so, like Modern Greek (and English) do. Those concepts were the territory of the Optative in Ancient Greek: the modern Future Imperfect and Future Pluperfect correspond to the Ancient Present and Future Optative with ἄν, in both main clause and conditional clause use. The Future Aorist (“must have contended”) appears to have been expressed with the optative too, at least in Herodotus:

1829. The present and aorist are rarely used of the past: (a) in Hom. of past possibility: καί νύ κεν ἔνθ᾽ ἀπόλοιτο and now he might have perished E 311 (Attic ἀπώλετο ἄν, 1784), ἀλλὰ τί κεν ῥέξαιμι; but what could I do? T 90. (b) in Hdt. of a mild assertion: ταῦτα μὲν καὶ φθόνῳ ἂν εἴποιεν they may have said this out of envy 9. 71, εἴησαν δ᾽ ἂν οὖτοι Κρῆτες these would prove to be (might be, must have been) Cretans 1. 2. Both uses are doubtful in Attic prose.


  • “But traditional grammars of Modern Greek don’t call these tenses Future Imperfect, Future Pluperfect, and Future Aorist. In fact, they don’t list them as tenses at all: you will never see them mentioned in grammars, showing verb conjugation.”

    I think you’re right about traditional grammars, but I’m sure I’ve seen these formations with the particle “θα” and future tenses being called “δυνητική έγκλιση” by analogy to “να” forming the subjunctive mood.

    By the way, I can’t recall any use of the “θα + past tense” form other than in conditionals.

    • I haven’t seen a δυνητική έγκλιση (potential mood), but I’m glad it’s happened. Surely θα πήγε “he must have gone” is not conditional?

      • It still seems conditional (υποθετικός λόγος) to me, but one where the condition is implied. For example:
        – “Πού είναι ο Νίκος;”
        – “(Αφού/αν δεν είναι εδώ,) θα πήγε κάπου”

        • “Since he’s not here, he must have gone somewhere”.

          You’ve given me a twinkle in the eye, Vangeli; thank you!

          The causation here is illocutionary: me not being here does not cause me to have gone somewhere; it causes you to infer (and say) that I’ve gone somewhere. So while it’s a conditional, it’s not conditional in the same way: it’s about evidence and inference, rather than just cause and effect. That’s why it’s clearer to refer to it as an inferential, or a dubitative. (As indeed is the case for the να counterpart of that tense: πιστεύω να πήγε “I believe he might have gone”, δεν πιστεύω να πήγε “I don’t believe he went = surely he didn’t go”.)

  • David Marjanović says:

    The problem with the Latin and perhaps the Georgian evidence is that to use it we first need to date the change of [w] to [v] in those languages. Modern Japanese routinely renders foreign [v] as [b]…

    • David Marjanović says:

      Bizarrely, the HTML entity for the “greater than” sign gives me an error message upon submission: “Invalid security token”.

Leave a Reply

  • Subscribe to Blog via Email

  • June 2024
    M T W T F S S
    1. Humbert, Jean. 1930. La disparition du datif en grec du Ier au Xe siècle. Collection de la Société de linguistique de Paris, t. XXXIII. Paris, Champion. p. 17

    2. Hesse, R. 1980. Syntax of the Modern Greek Verbal System: The Use of the Forms, Particularly in Combination with θα and να. Copenhagen: Museum Tusculanum.

%d bloggers like this: