As I carry on my dissertation work on “periphrastic” constructions (more specifically, the syntactic pattern of εἰμί + participles, as the name periphrastic in the grammatical tradition is problematic) a few instances of a different type of periphrasis have caught my eye: ἔχω + participle. This periphrasis plays essentially no role in the NT or LXX (at least, I did not find any doing a rudimentary search with Accordance; there a possibly a few in Josephus—a few promising candidates but I didn’t look any more deeply), but appears scattered throughout various other writings from the time.

Significance of the construction

This construction, known in grammars both ancient and modern as the σχῆμα Ἀττικόν or σχῆμα Σοφοκλειον due to the major use of it by Sophocles and Attic Tragedians, ends up being roughly equivalent in meaning to a perfect indicative.[i] In a recent extensive study of the periphrasis, Klaas Bentein has this brief summary of it:

“One HAVE-perfect construction that does seem to increase, albeit slightly, in frequency is ἔχω with the present/aorist participle and a temporal adjunct…. As already mentioned, this construction is used with one specific anterior subfunction–that is, as a perfect of persistence, which could also be expressed by the synthetic perfect or present/imperfect, or periphrastic εἰμί with the perfect participle. It denotes an event that has started in the past and is ongoing until the time of speaking (in the case of the present perfect), as in:

ἰδοὺ τοσούτους χρόνους ἔχω καταναλίσκων τὸ ἔθνος τῶν Χριστιανῶν καὶ οὐκ ἐπάγη ἐν ἐμοὶ βέλος (V. Sym. Styl. J. 186.16-17)

‘Behold I have been destroying the Christian people for so many years and there has not been stuck any arrow in me”

Klaas Bentein, Verbal Periphrasis in Ancient Greek: Have- and Be- Constructions (Oxford: Oxford University Press, 2016), 194.

He mentions that it is encountered most often in Christian writings.  It is doubtful that there is any context in which this construction is obligatory, though, as Bentein notes, it does often occur with a temporal adjunct. It is perhaps not surprising, given its close association with the Attic Tragedians, that this construction tends to appear in more literary Greek. That it is not in the LXX or NT, or many other early Christian or Jewish texts, is not surprising. A brief glance at a few statistics shows its distribution.

Status of construction in Hellenistic Greek

The ἔχω periphrasis is a fairly marginal one in Hellenistic Greek, as also in Classical. Further, it appears that it is fairly limited to higher(-ish) register documents. It seems to be virtually absent from the papyri. I have not found any discussion of it in Basil Mandilaras’ The Verb in the Greek Non-Literary Papyri or in Edwin Mayser’s Grammatik der griechischen Papyri aus der Ptolemäerzeit (though, I only have a pdf of Mayser, which is difficult to search through and quite a pain, so I may have simply overlooked any mention he has of the discussion). The general absence of discussion in these grammars is consistent with the recent work of Klaas Bentein. Searching through the excel data set he made available, I’ve rounded up some numbers from his corpus to illustrate the status of this construction. In the period from 300 BC to AD 300 he has the following:

  • BC 300-1
    • periphrasis with εἰμί: 1247x
    • periphrasis with ἔχω: 83x
    • periphrasis with εἰμί in the papyri: 326x
    • periphrasis with ἔχω in papyri: 1x
  • AD 1-300
    • periphrasis with εἰμί: 1468x
    • periphrasis with ἔχω: 215x
    • periphrasis with εἰμί in the papyri: 300x
    • periphrasis with ἔχω in papyri: 11x

Being only a novice papyri reader (with aims of improving), I will not embarrass myself by posting any of the texts here and attempting a translation. Suffice to say, ἔχω periphrasis is not widely used in general, and not much used in papyri. This could be a sampling bias, but the general point stands that it is no where near as widely used as εἰμί periphrasis. Also, ἔχω periphrasis occurs markedly more frequently in the non-papyri (while this doesn’t exactly equate to higher register, that is often the case).


What has caught my eye recently in reading are three examples, two from early Christian sources and one from a Greek literary writer. These examples clearly display how the periphrasis functions. First, one from the NT textual tradition and one from the Martyrdom of Polycarp, both of which perfectly fit Bentein’s model of ἔχω + participle + temporal adjunct.

Martyrdom of Polycarp 9:3 has the following text, as Polycarp responds to the magistrate:

… ἔφη ὁ Πολύκαρπος· Ὀγδοήκοντα καὶ ἓξ ἔτη  ⸂ἔχω δουλεύων  ⸂αὐτῷ, καὶ  ⸂οὐδέν με ἠδίκησεν⸃· καὶ πῶς δύναμαι βλασφημῆσαι τὸν βασιλέα μου τὸν σώσαντά με;

Polycarp said, “I have been serving him (Jesus) for 86 years and he has done me no wrong. How could I blaspheme my king who saved me?”

One could just as well render this text, “I have been his servant.” Here the periphrasis highlights the ongoing persistence of the action expressed in the periphrasis, with a special focus on the duration of time, provided by an accusative of duration temporal adjunct.

A second example comes from the NT textual tradition. Intriguingly, it is with the same verb. I was alerted to this instance by Jean Putmans on an email list. He noted that a variant reading of Luke 15.29 appears in at least three manuscripts.[ii] This variant reading is attested in a homily of John Chrysostom on the Parable of the Prodigal Son, as seen here (text from Migne’s Patrologia Graece):

Ὁ δὲ ἀποκριθεὶς εἶπε τῷ πατρί· Ἰδοὺ ἐγὼ τοσαῦτα ἔτη ἔχω δουλεύων σοι, καὶ οὐδέποτε ἐντολήν σου παρῆλθον· (In parabola de filio prodigo, 59.520.65)

Now he [the elder son] answered and said to his father: “behold, I have been serving you for all these years, and not once have I transgressed your command.

Again, this is a textbook example of the construction with the periphrasis communicating the ongoing nature of the action, from past into present, with a temporal adjunct provided by the accusative of duration.

This construction was also used by  Greek literary writers. Here is an example from Lucian’s The Ass.  It differs from the former two examples in that there is no temporal adjunct.

“Ἀλλὰ πάντα,” εἶπεν ἡ γραῦς, “εὐτρεπῆ ὑμῖν, ἄρτοι πολλοί, οἴνου παλαιοῦ πίθοι, καὶ τὰ κρέα δὲ ὑμῖν τὰ ἄγρια σκευάσασα ἔχω.” [sec. 20]

“Everything is ready for you,” replied the old woman. “There’s plenty of bread along with jars of old wine and I’ve also cooked you venison.”

Lucian, Soloecista. Lucius or The Ass. Amores. Halcyon. Demosthenes. Podagra. Ocypus. Cyniscus. Philopatris. Charidemus. Nero., trans. M. D. MacLeod, Loeb Classical Library 432 (Cambridge (MA): Harvard University Press, 1967).

Here the emphasis is something like this, “I have prepared (and it is still ready) venison for you.” Perhaps the usage of this form is meant to bring out a subtle difference in the work necessary for the different aspects of meal preparation. The wine and bread were already prepared, requiring no work in the recent past. The venison, by contrast, was prepared by the woman “just now” and the usage of this form possibly draws attention to that fact.

The ἔχω periphrasis is an interest periphrasis form, not encountered too often, but certainly worth knowing about for any point in time you stumble across one while reading outside the NT.

[i] Evert van Emde Boas et al., The Cambridge Grammar of Classical Greek (Cambridge: Cambridge University Press, 2019), 635

[ii] His comment, in full, is:

“Zu Lukas 15:29 findet sich in (wenigstens) drei Hss.. (251, 716 unbd 1093) statt  des Präs.Ind.Sg.1 δουλευω den Ausdruck εχω δουλευων.”