In Part 3 of this series on periphrastic constructions I left off with a brief summary of a model of constituent order for periphrastic constructions. Such a model is necessary to try to make sense of the various constituent orders and try to disambiguate what is or is not periphrastic. While necessary, such a model is not sufficient.

On periphrastic and non-periphrastic

It turns out that non-periphrastic constructions of the sort copula + predicate participial phrase (that is, where the participle phrase is in predicate/conjunct/”adverbial” relationship to the copula) also pretty much follow this same syntactic ordering.[i] In other words, there is no hard syntactic line between periphrastic and non-periphrastic constructions. Thus, simply noting the order of constituents is not going to solve the problem of what is or is not periphrastic in cases where that isn’t already obvious. But that does not mean all hope is lost.

Noting these constituent order patterns gives us an intelligent and methodical way to think through borderline cases. We have tools at our disposal to make arguments about why a certain passage should or should not be considered periphrastic. That is more useful than fiat statements about what is or isn’t periphrastic based on nothing more than grammatical intuition (which tends to be strongly shaped by the regular use of periphrastic verbs in English, for those of us who are English primary). We can make these arguments by paying attention to the syntax of εἰμί and the order of constituents and considering how the different elements fit together with the surroundings.

An (easier) test case

Back in Post I (of what I then foolishly believed would be a 2-post series), I pointed to this text from Joseph and Aseneth as an example:

τὸ ἐμὸν ὄνομα ἐν τοῖς οὐρανοῖς ἐστιν ἐν τῇ βίβλῳ τοῦ ὑψίστου γεγραμμένον τῷ δακτύλῳ τοῦ θεοῦ ἐν ἀρχῇ τῆς βίβλου πρὸ πάντων

My name is in the heavenlies, having been written in the book of the Highest by the finger of God in the beginning of the book before all others

This is a non-periphrastic understanding of the text. Any avid footnote readers would find that I provided a possible way to translate this passage as a periphrastic construction. It is possible to understand it that way, but very improbable. Let’s use the constituent order model and other tools developed in the last three posts to talk through the example and see some of the reasons why this passage is unlikely to be periphrastic.

First, on any reading of it we note that the subject and a spatial prepositional phrase both precede the copula:

  • [τὸ ἐμὸν ὄνομα] [ἐν τοῖς οὐρανοῖς] ἐστιν…

The subject in front of the copula (often) signals a contrastive topic. That is, a subject in this position alerts the reader that this topic is different than the topic of the sentence/clause/section prior. When the subject is the same it is most often omitted, but can also appear following the verb. The prepositional phrase immediately preceding the copula is in the slot of the Greek sentence that is used to put focus, or emphasis, on an element of the predicate (predicate being everything that is not the subject in the sentence). In simple terms, the position of the prepositional phrase indicates the author wants us to process it as the most important element.[ii] This is similar again to the way we use emphatic stress on certain portions of sentences when we speak in English to single them out as important (or italics in writing).

Not only is this prepositional phrase singled out as important, but it also is spatial. As pointed out in Post II, one of the ways εἰμί can make a full predication is when it stands with a spatial adjunct. It is reasonable to assert that τὸ ἐμὸν ὄνομα ἐν τοῖς οὐρανοῖς ἐστιν is a complete predication: “My name is in the heavenlies.” Supporting this reading is an interesting syntactic pattern. When a participle follows the copula it is unusual for part of the participial phrase to precede the copula. It can happen, like in Ephesians 2.8:

Τῇ γὰρ χάριτί ἐστε σεσῳσμένοι διὰ πίστεως

For you are being saved by grace through faith

“By grace” syntactically is in the orbit of the participle, but here precedes the copula even though the participle follows the copula. Suffice to say that this sort of syntax is not common. Generally when the participle follows the copula, its arguments and/or adjuncts stay with it unless there is a compelling reason for them to occur pre-copula.[iii] So, in Joseph and Aseneth “in the heavenlies” is not likely to be in the syntactic orbit of the participle. Given that it makes great sense of the passage to take “in the heavenlies” with “my name is…”, we have a solid argument here that this passage is not periphrastic.

This would leave the (quite long) participial clause  to take account of:

τὸ ἐμὸν ὄνομα ἐν τοῖς οὐρανοῖς ἐστιν ἐν τῇ βίβλῳ τοῦ ὑψίστου γεγραμμένον τῷ δακτύλῳ τοῦ θεοῦ ἐν ἀρχῇ τῆς βίβλου πρὸ πάντων

My name is in the heavenlies, having been written in the book of the Highest by the finger of God in the beginning of the book before all others

The big question remaining is how to take the prepositional phrase ἐν τῇ βίβλῳ τοῦ ὑψίστου. It could be read in one of two ways: 1) as a complement to “in the heavenlies,” yielding “it is in the heavenlies, in the book of the Highest,” or 2) as a focal element in the participial phrase, “being written in the book of the Highest…” While I favor number 2—mainly on the grounds that it gives two clauses (or one clause and one quasi-clause) with an emphatic spatial prepositional phrase, further pursuing this argument leads us astray from the main topic of periphrasis. In either case, it does not affect the overall conclusion that this passage is not periphrastic.

Returning Full-circle to Barnabas

Now we can return to the question posed at the end of Post 1: how should we understand the Greek in the following two example from The Epistle of Barnabas?


  • Καὶ ἦν Μωϋσῆς νηστεύων ἐν ὄρει Σινᾶ
  • And Moses was fasting on Mount Sinai

and 4.7

  • Καὶ ἦν Μωϋσῆς ἐν τῷ ὄρει νηστεύων ἡμέρας τεσσαράκοντα καὶ νύκτας τεσσαράκοντα
  • And Moses was on the mountain, fasting 40 days and 40 nights

Based on the ground we have covered, I will unequivocally claim that 14.2 is periphrastic (which is how the translators I have checked all take it). It follows the default word order of copula + subject + participle + adjunct and there is no reason to not take it as periphrastic.

4.7 remains difficult. The single largest reason this is the case is that 4.7 is an embedded quotation from the Septuagint (probably a loose citation from memory as it does not correspond to anything in any extant LXX manuscript). This is difficult in that it is hard to assess, in the context given, whether the point of the sentence is to predicate “Moses was on the mountain, fasting for 40 days and nights,” or to make an emphatic statement of the order, “Moses was fasting on the mountain for 40 days and nights.”

The argument in the passage seems to put less weight on his location on the mountain than on the broader complex of receiving the covenant on the stone tablets, which Moses subsequently breaks. In other words, there seems slight weight in favor of taking this prepositional phrases as non-emphatic, which leaves two options: 2) either the text is non-periphrastic, or, to introduce a new option in the middle of an argument, 3) the prepositional phrase serves as an anchor to the context. In option 3) the prepositional phrase occurs between the copula and participle as an anchor to the context, which is not uncommon in periphrastic constructions. Such a construction is not focal, but rather topical and similar to clause-initial setting elements in that they belong mainly to the structure of the text. The possible trigger for such a reading would be the common knowledge that Moses, when he received the covenant, was on Mt. Sinai. As the receiving of the covenant is mentioned at the end of the preceding clause, “on the mountain” could be considered accessible and topical. This reading is quite possible in the context from which this quotation was “lifted,” wherever that happens to be. I lean towards this reading: periphrastic with a spatial anchor to the context.

Finishing thoughts

Of course, at this level of discussion we are really splitting hairs. Part of what I appreciate with Levinsohn’s model is pointing out that the syntax between these two readings is actually the same. By that I mean on the surface it is not possible to draw a clear distinction between periphrastic and non-periphrastic syntax for copula + participle constructions that always holds true. Obviously what elements go with what other elements differs between two such readings, but there is no way to see for sure what goes with what based on what is written in all cases. It is probable that such ambiguity would be removed in speaking, perhaps via a slight pause or intonation change. But, for us today as readers (and for Ancient Greek readers as well), there remains an ambiguity in certain instances which cannot be cleared up. Rather than yielding certainty in all cases (though they do cover a great deal of them), these tools discussed provide ways of making arguments about what is or is not periphrastic which are sensitive to both the various constituent orders attested in copula + participle constructions and to the how these constituents fit within the broader context in terms of the way the text structures information for the reader to understand it.

[i] I have not, as of yet, detected any systematic differences in constituent order between the two, though there are a few possibilities.

[ii] This follows the assumption that topic and focus positions are similar in clauses with εἰμί as in ones with “normal” verbs. This is a reasonable, though not unproblematic assumption.

[iii] Incidentally, the pre-copula position of Τῇ γὰρ χάριτί in this verse is not because it is focal, as the concept of “by grace” is already part of the immediate context. Here it serves a function of linking this verse, and the following, to the immediate context, showing the continuity point “by grace” which these verses add upon. It is more like a pre-verbal participial clause giving necessary setting for understanding what comes after it than a focal element.