Matt 1:20 word order: ἐκ πνεύματός ἐστιν ἁγίου

Greek Koine and Biblical and Medieval Greek
21

Hi all, thanks for your replies!

For word order questions in (Attic) Greek, I find the combination of Dover 1960 plus the newer approach (most recently summarised in the Cambridge grammar of classical Greek) most helpful. Neither is sufficient on its own (for me).

Bill, I agree with everything in your post too—yes, it was just the clitic form ἐστι in second position that I was referring to in those examples with a genitive modifier in first position; agreed also that the other examples in the first five chapters don’t split noun + adjective, and so further investigation would be required. I haven’t checked elsewhere yet in Matthew (I’ll need to read him first! I’m way behind). Since other clitics in second position can split syntactically-related groups, I don’t see any intrinsic reason to doubt this, but only comprehensive reading of Matthew will tell.

I note in passing that one would also need to check Aeschines before determining whether the Aeschines 1.122–23 quote that Joel helpfully gives above, containing two copulas, follows the same pattern that I noted in Matthew: the first ἐστι fronted under the influence of the demonstrative pronoun, the second ἐστι in second position (with οὐκ ἀνδρός in first position in that word group):

αὕτη μέν ἐστιν, ὦ Τίμαρχε, ἀνδρὸς ἀγαθοῦ καὶ σώφρονος ἀπολογία, καὶ πεπιστευκότος τῷ βίῳ καὶ καταφρονοῦντος εἰκότως ἁπάσης βλασφημίας· ἃ δὲ πείθει σε Δημοσθένης, οὐκ ἀνδρός ἐστιν ἐλευθέρου, ἀλλὰ πόρνου περὶ τῶν τόπων διαφερομένου.

Each author can have their own individual approach to word order: the resources only give us a few general tools and patterns, to allow us to begin further investigation…

I wonder whether anyone has already studied the position of postpositives in Matthew? Any suggestions for the leading (and, if different, the latest) reference works for NT Greek would be highly appreciated, many thanks all!

Cheers, Chad

22

May I suggest that this is an example where the author split an ‘entity’ defined by a noun and its attributes (ἀνδρός + ἐλευθέρου, πόρνου, etc.) in order to emphasize the contrast between the first attribute (ἐλευθέρου) and the following one: not only does πόρνου come closer (by one word) to ἐλευθέρου, sharpening thus the contrast, but the ἀλλὰ between the two amplifies the contrast all the more. (The noun remains ‘constant’, whatever the attributes, but the attributes are not in harmony with one another). The author does not need to do this with the first ἐστιν, because all attributes used in that case are of similar positive quality. Just first sight impression…

23

Responding to Vasile: My initial reaction was the same as yours. But on further reflection, I think that if instead of πορνου, the word were another adjective agreeing with ανδρος, say for example πονηρου, there would be a stronger case that the placement of ελευθηρου after εστι was intended to highlight the contrast between the two terms. However, the contrast is between ανδρος ελευθηρου and the noun πορνου. To my mind, this clouds the issue of whether the placement of enclitic εστι as the second element of the word group is intended to make ελευθηρου more salient or, as Chad suggests, merely “normal” Greek word order consistent with Wackernagel’s Law..

24

The predicate position (explicit, with articles) is often a matter of emphasis, as Smyth and others mention.

Getting back to ἁγίου, my suggestion in the other thread, was that a conception εκ πνευματος is hardly enough to eliminate fears of cuckoldry, but the special word order here (not normal at all! not English prejudice! despite the claim above, just try to find parallels to it if you don’t believe me) tells us that it is specifically the ἅγιος element, “moral purity”, that makes it all right. “…from spirit that is morally pure”. Or “…from a…”

25

After refreshing my memory of Wackernagel’s Law, I don’t think it explains the placement of the clitic εστιν in either the Matthew or the Aeschines sentences. W."s maintains that in Proto-Indo-European clitics occupied the second element in the sentence or clause. In both Matthew and Aeschines, εστιν doesn’t occupy the second element in the sentence: in fact, the issue is precisely that εστιν anomalously does not occupy the second element.

26

So I’m back to where I started.

Denniston, Greek Prose Style, pp. 51-2 (emphasis added; examples in next post):

The milder forms of hyperbaton, where only one or two words intervene, are common in all authors: > particularly the interposition of a verb between substantive and adjective in agreement> , or between a substantive and a partitive genitive dependent on it.

Kühner-Gerth, Ausführliche Grammatik der griechischen Sprache, sec. 606a3:

Überhaupt ist sowohl die erste als die letzte Stelle als eine Kraftstelle anzusehen, wenn sie von Satzteilen, die nach der gewöhnlichen Ordnung nicht hierher gehören, eingenommen werden. Pl. Ap. 18, e ἔπειτά εἰσιν οὗτοι οἱ κατήγοροι . . ἀτεχνῶς ἐρήμην κατηγοροῦντες ἀπολογουμένου οὐδενός. *h, 29 νῦν μὲν παύσωμεν πόλεμον καὶ δηιοτῆτα | σήμερον: ὕστερον αὖτε μαχήσοντ̓



Generally, the first as well as the last position is to be seen as an emphatic position, when they are occupied by parts of the clause that don’t belong there according to the usual order. . . .

http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0021%3Asmythp%3D606

The unusual word order in both the Matthew and Aeschines sentences, placing the important word at the end of the clause, is normal in the sense that it is a normal way of emphasizing an important word in Greek. The hyperbaton is foreign to English, but, as the examples in Denniston demonstrate, not uncommon in Greek.

27

Examples from Denniston p. 52 of “mild” hyperbaton:

https://www.flickr.com/photos/127666006@N08/51355699469/in/datetaken/

These examples show how hyperbaton of the sort exemplified in the sentences from Matthew and Aeschines is really not very unusual in Greek.

(If you follow the link, you can access my entire photostream on Flickr.)

28

I gave several examples of this with a tightly related verb/participle from 1Pe in the other thread. (Where the article was present, they happened to be predicate.) Michael gives ἐλαίου ῥεῦμα ἀψοφητὶ ῥέοντος in his new thread, with the common genitive separation.

However, here we have a verb between a substantive and adjective of a prepositional phrase (the substantive being neither the subject nor the object of the verb). It seems closer to the “intrusion of external elements into the articular structure” that Denniston discusses on pg. 56., giving two examples in the second paragraph of phrases without the article. (Though the εστιν here to me seems rather more intrusive then either of those.)

Regardless, I recommend reading a few pages of Matthew to see how common this sort of thing is. How many other examples could we find in the Gospels if we looked? 2-3? It is not normal.

Also, the added emphasis should be a clue that we’re closer to prepositional territory than to attributive with this. But the only way to tell for sure would be to find examples with the article, as I did in 1Pe. There has been an assumption made so far that attributive is somehow the normal Greek and predicative is the exception. That’s not really the case.

29

The Flickr photographs of DC (and elsewhere?) are amazing, by the way.

30

Hi all, I’m definitely inclined to follow the NT experts in this thread on this point! I hope I made clear that I wasn’t proposing this (just asking the question as a total NT newbie) and that one needs to understand the author in question, rather than take the dangerous step of applying general rules to specific authors.

Bill, just to pick up one thing however (zooming out from NT and the position of the copula in this quote): regardless of how it worked in Indo-European (and I’m not familiar with that), the iteration of Wackernagel’s law in Greek does not necessarily operate just at the level of the sentence or clause. It also operates at the level of the word group (hence those vertical bars in my examples above).

Dover 1960 p. 17 (where q stands for a postpositive):

The example suggests that > q > are not necessarily placed after the leading > p > or > M > of what would traditionally be defined as a ‘clause’, > but may occupy a similar position within one of the word-groups which constitute the clause> …

Cambridge grammar of classical Greek sec. 60.7:

Postpositive > words tend to occur after the first word of the sentence, clause > or word group > they belong to, that is, in > second position> : this rule is known as > Wackernagel’s Law> . Depending on several factors ‘second position’ may mean slightly different things in different contexts.

The treatment of the copula as a postpositive or not (or something special by itself) is always tricky (same issues arises in Latin: the position of the copula is not quite the same as even the auxiliary…). e.g. Dover 1960 p. 13:

The verb εἶναι cannot be classed as > q > without many qualifications and reservations, not all of which admit of a satisfactory classification. > εἶναι as a copula tends, in most authors, to be treated as > q> …

Coming back to the NT quote, the only reason I asked was that there did seem to be a pattern in the first five chapters at least of Matthew; as to whether a copula splitting a noun + adjective was ‘tolerated’ in Matthew, I simply didn’t know: different authors ‘tolerate’ different types of postpositives splitting syntactically-related elements (Dover talks on p. 16 about differences between Herodotus and Attic prose on this point).

Since I don’t know anything about Matthew’s practice, and saw a pattern, I thought it was worth asking. The examples I gave set out what motivated my question, not a proof of any particular position. As noted above, only a deep reading of an author will answer the question, and I’ll go with the NT experts on the thread as the best available information. Thanks all!

Cheers, Chad

31

Chad, the verb is not a constituent of the prepositional phrase in which it’s embedded. I think that‘s a significant distinction. So a fortiori εστι doesn’t belong in the position of the second element by W’s Law, even taking for granted that W.’s Law applies locally to prepositional phrases in Greek. This, it seems to me, is a pure hyperbaton — but one that’s not foreign to Greek in general.

I should mention that I’m probably even less familiar with New Testament Greek than you, and certainly not specifically with Matthew, and I’m not inclined to do a thorough philological analysis, but fools rush in where angels fear to tread. When I originally read the sentence, the placement of εστι and the hyperbaton didn’t strike me as terribly unusual.

Joel, the example from Timaeus has the verb ιζει sandwiched in a prepositional phrase.

Thanks for looking at my photos and commenting!

32

Thanks Bill! I could say more on that last point, but will leave it there as it’s digressing again from NT Greek.

When I come to reading Matthew (still on John), I’ll look out for more of these prep. + noun + adj. combinations in this position in the sentence, and see how they interact with postpositives — I’m now curious!

Cheers, Chad

33

Hylander, you mean from Timaeus? Looking up the context, I think that’s a great example. It’s predicate position, like ουτος or αυτος (self), meaning “different” here. If it had gone into attributive position and been sandwiched after the εις, I would have taken it as “into a second chair”, which wouldn’t make much sense. At least that’s how it seems to me. The complicated word order in the Timaeus passage as a whole could never appear in the NT, of course.

34

I’m loving this thread, and really appreciate all the comments to date (although I understand less than half of what I probably should!).

It might be of interest at this point to offer the results of a quick search I just did on my Accordance bible software. I looked for [PREPOSITION] [NOUN] [VERB] [ADJECTIVE], and one other similar candidate turned up. Lo and behold, it also refers to holy spirit:
Acts 1:5 ὅτι Ἰωάννης μὲν ἐβάπτισεν ὕδατι, ὑμεῖς δὲ ἐν πνεύματι βαπτισθήσεσθε ἁγίῳ οὐ μετὰ πολλὰς ταύτας ἡμέρας.
“but by/in/with spirit you will be immersed (‘baptised’) - that is, a holy one” (!!??)
For those not aware, the author of Acts is not Matthew, but considered to be Luke (a sequel to the gospel of his name).

Then I searched for πνεῦμα [VERB] ἅγιος, and found three instances: the two already mentioned, plus Luke 2:25, introducing a just and upright Jewish man, Simeon, καὶ πνεῦμα ἦν ἅγιον ἐπ’ αὐτόν.
Now that’s usually translated, ‘and the holy spirit was upon him’, whereas I would have read, ‘and (a) spirit was holy upon him’.

I understand that this last is a different case as it’s a straightforward predicate adjective (do correct me if I’m wrong), but the three seem to indicate a tendency to split πνεῦμα and ἄγιος in ways perhaps a bit uncomfortable for modern translators.

35

Further to my comment above, I note that there is a variability in the manuscripts at Acts 1:5. Metzger’s Textual commentary on the 3rd Ed. Greek New Testament (USB, NA26), which is the same here as the text I quoted (NA28), says that the great majority of witnesses have ἐν πνεύματι ἁγίῳ. However, the committee considered that the greater weight of evidence favoured the alternate construction ἐν πνεύματι βαπτισθήσεσθε ἁγίῳ (which he considers to lend the sentence a chiastic form). “The envelope construction of the Alexandrian text (placing the verb between the noun and the adjective) may be an editorial refinement, or it may reproduce an emphasis intended by the author.”

The other “less elegant”(!) construction was deemed to have arisen from a desire to harmonise with the parallel in the synoptic (Matt, Mark, Luke) gospels.

I’m intrigued that he considers the more straightforward construction “less elegant”. Apparently, for him and the committee, ἐν πνεύματι βαπτισθήσεσθε ἁγίῳ is fairly unremarkable, grammatically speaking; just more elegant and perhaps emphatic.

36

καὶ πνεῦμα ἦν ἅγιον ἐπ’ αὐτόν – ἅγιον is not clearly predicative. Could be translated “a holy pneuma was upon him” (non-predicative, attributive) or, less likely, “a pneuma was holy upon him” (predicative)..

ὑμεῖς δὲ ἐν πνεύματι βαπτισθήσεσθε ἁγίῳ – this is exactly the same hyperbaton as in the sentence from Matthew that began this discussion, with the verb intruded into the prepositional phrase. “Your were baptized in a pneuma that is holy.”

37

Yes, the first is restated in the next sentence with articles, and is definitely attributive. The second looks closely parallel to Matt 1:20 structure, but it hit me differently. At first the εν πνευματι seemed to be getting emphasized after the δε as a contrast to υδατι, but then the αγιωι comes along in a stronger position and emphasizes a different aspect of the contrast.

38

Dammit! just when I thought I was getting the hang of predicative vs attributive :wink: . Back to the books for me.

39

It’s most likely attributive. The predicative translation is strained. The predicate is ην επ’αυτον. And assuming it’s attributive, it illustrates the type of hyperbaton that where a verb splits a noun and its attributive adjective.

40

The next sentence gives it with the article as definitely attributive.

καὶ ἦν αὐτῷ κεχρηματισμένον ὑπὸ τοῦ Πνεύματος τοῦ Ἁγίου μὴ ἰδεῖν θάνατον πρὶν ἢ ἂν ἴδῃ τὸν Χριστὸν Κυρίου