ΕΝ ΕΦΕΣΩ

Some have called the claim that Modern Greek has something to add to the study of Hellenistic Greek controversial.

Perhaps.

But I think that depends on what you mean and what context you intend to use Modern Greek. For the first or second year student, I would heartily agree that Modern Greek isn’t going to be terribly helpful.

But when it comes to understanding the Greek text by the more advanced scholar, particularly for being able to teach the beginning student, Modern Greek is much more important.

Let’s look at an example.

According to Artemis Alexiadou, Liliane Haegeman & Melita Stavrou, the two demonstrative positions in Greek have distinctly different meanings (the footnotes are theirs):

Context A: Mary is at the butcher’s pointing to a pork joint that she wants to buy.

a. Mary: Thelo afto to butaki.
want-1SG this the joint
‘I want this pork joint.’

b. Mary: ??Thelo to butaki afto.[48]
want-1SG the joint this

Context B: A paragraph from a guide book about a Greek town.

c. I poli eci pola istorika ktiria pu xronologhunte apo ti vizantini epoci.
the town has many historical buildings that date back to the Byzantine period

d. Ta ktiria afta episceptonte kathe xrono ekatondadhes turistes.
the buildings these vistis-3SG every year hundreds tourists

e. ??? Aftra ta ktiria episcenpttonte kathe rono ekatondadhes turistes.
these the buildings visit-3SG every year hundreds tourists[49]

Manolessou & Panagiotidis (1999), Manolessou (2000), Panagiotidis (2000), and Grohmann & Panagiotidis (2005) observe that in Greek the pre-article position of the demonstrative afto entails greater deictic strength, in contrast with the post-nominal position in which the demonstrative is used as discourse anaphoric, namely to refer back to an entity that has been previously mentioned. Only the pre-article demonstrative can normally be used along with a pointing gesture: based on this observation it is generally agreed that only the pre-article demonstrative is genuinely deictic.[50]

Artemis Alexiadou, Liliane Haegeman & Melita Stavrou, The Noun Phrase in Generative Grammar (Studies in Generative Grammar 71; Berlin: Mouton De Gruyter, 2007), 120-121.

---

[48] For some speakers (b) is acceptable with contrastive intonation on the demonstrative and an accompanying deictic gesture.

[49] Notice that the ‘strategy’ discussed in the preceding note is not available for rescuing (83e) since the buildings in question are not physically present, and so one cannot point to them (unlike the butcher’s customer who actually sees the joint).

[50] Interestingly, a different proposal with respect to the interpretation of post-nominal and pre-article demonstratives is made by Tasmowski De Ryck (1990), who argues that the pre-article demonstrative has a thematic interpretation (i.e. it represents an entity already known/given) while the post-nominal demonstrative has a rhematic (i.e. new) interpretation.

These words, sparked my curiosity as to what we can say about the status of demonstratives in Hellenistic Greek. So I perused the texts for a bit. I didn’t do anything comprehensive, just perusing. But what I saw, in my randomness, fit – if we assume that footnote 48 was generally true for the Hellenistic Period.

Likewise, it *appears* that pre-nominal demonstratives are not allowed in situations like what we see in Context B above. Granted, this was more difficult to test. What I did was search through the non-narrative texts of the New Testament and examined the kinds of contexts and pragmatic situations where the author used pre-nominal vs. post-nominal demonstratives. The different, I though, was quite striking. For example, in Paul, pre-nominal demonstratives are only used in a couple different situations:

1. Pronominal situations: Rom 11:30, νῦν δὲ ἠλεήθητε τῇ τούτων ἀπειθείᾳ, “but now you have received mercy as a result of their disobedience.”
2. Quotes from Direct Speech: 1 Cor 11:25, τοῦτο τὸ ποτήριον ἡ καινὴ διαθήκη ἐστὶν ἐν τῷ ἐμῷ αἵματι, this cup is the new covenant in my blood.”
3. References to previous words written (i.e. the external real life referent is the text itself): 2 Cor 7:1 ταύτας οὖν ἔχοντες τὰς ἐπαγγελίας, “Therefore, since we have these promises…” (Cf. Eph 3:1
4. Abstact Concepts that cannot have an external referent: 2 Cor 11:17, ὃ λαλῶ οὐ κατὰ κύριον λαλῶ ἀλλʼ ὡς ἐν ἀφροσύνῃ ἐν ταύτῃ τῇ ὑποστάσει τῆς καυχήσεως, “ In this self-confident boasting I am not talking as the Lord would, but as a fool.”

If what Alexiadou, Haegeman, & Stavrou say is correct for Hellenistic Greek as well, then it would have been ungrammatical for Paul to have said: ἐν τούτοις τοῖς κλίμασι instead of ἐν τοῖς κλίμασι τούτοις in Romans 15:23.

Now whether I’m right or not, I’m not sure, which is why I used the word “appears” above. I don’t think we can draw definite conclusions on the issue from this very cursory & ad hoc study I did this afternoon.

But if I am correct on this, then there are two important things to remember:

1) Greek word order issues are never discussed in beginning classes in terms of the why?. New students only learn the what?. They need to learn the reasons for different word orders.

2) What I just covered here is something you could very easily teach to your students: “Students, this is why the order of the demonstrative in the NP matters and this is what it means.” See, that was simple, wasn’t it?

I’m totally ripping this off directly from the B-Greek list, but its definitely worth it if I can get a few more people to see this little discussion.

Several days ago, Elizabeth Kline wrote this:

In James 3:3 we see an example of a preposed genitive noun TWN hIPPWN.
One way to read it, TWN hIPPWN modifies TA STOMATA (J.Huther) which is
within a prepositional phrase EIS TA STOMATA. Another reading, TWN
hIPPWN modifies TOUS CALINOUS. J.B. Mayor suggests TWN hIPPWN modifies
both TOUS CALINOUS and TA STOMATA. Alford suggests that TWN hIPPWN
might mean something like “in the case of horses” thus linking ideas
CALINAGWGHSAI … SWMA from the immediately preceding context.
Alford’s suggestion is interesting in light of proposals in recent
text-linguistics about fronted constituents. S.Levinsohn talks about
TWN hIPPWN in Js. 3:3 (Discourse Features SIL, 2000, p63). Levinsohn
suggests that TWN hIPPWN is preposed to indicate a shift from talking
about people to talking about horses. What I find interesting here, is
that Alford in the second half of the nineteenth century had a very
similar reading without the trappings of text-linguistics. The
terminology is quite different and Levinsohn certainly frames the
question differently, nevertheless … draw your on conclusions about
this.

(My Emphasis)

To which Steve Runge replied,

You seem surprised about the agreement you find in their analysis. IMO, it is foolish to begin anywhere else than with what the “Dead Grammarians” have said. I include Carl and Wallace here out of respect. Most had internalized the language more than I could ever hope. The main problem I have with them is the fuzziness of explanations at times, ones that are dependent on English or Latin. I mean to say that they had a poor medium to communicate their thoughts, not that their thoughts were poorly formed. If you look at the commentaries and grammarians that Levinsohn cites for support, most of them are the same vintage as Alford.

Another related issue with the dead grammarians is their ability to explain WHY they claim what they do. They know a duck when they see one, but at times fail to provide the meaningful characteristics that make a duck a duck and not a seagull or goose. This is the area that I think text-linguistics has the most to offer. This kind of description should, if done properly, allow those following behind to understand the thought process step by step, compared to choosing from a long list of potential syntactic forces. This is particularly the case where there are several complicating factors at work in a single context, as in James 1:22-23. Text-linguistics can help one understand the role played by each component that makes up the whole. Does Alford offer any basis for his conclusion for you to weigh it against the competing alternatives that you cite? You called it interesting, making me think it was the outlier compared to other more modern explanations. Levinsohn reaches the same conclusion, but leaves a trail of bread crumbs.

Simply having an answer leaves the exegete in a poor position to evaluate the alternatives. Too often, statistical factors are called into play, even in traditional grammar. “The vast majority of the time X functions as Y, therefore in this context the evidence leans toward reading it as Y.” The exegesis of hOUTWS in John 3:16 is a good example of such argumentation.

If someone claims a brand new reading that has hitherto been overlooked by scholarship, watch out. Unfortunately, linguistics is being used by some as a hip new way of tickling the ears, being little more than smoke and mirrors with nice poly-syllabic words. Linguistics, if it is used properly, should help you understand what is going on under the hood; it does/should not change the car. In my experience, I have found that any claim I might make on the basis of discourse grammar and linguistics can be supported by the previous claim of a dead grammarian. They knew their stuff, knew whom they were stuffing, and stuffed them with eloquence. It behooves us to linger over this same stuff, as you so often demonstrate in your research. Thanks for passing it along on the list.

And finally, Elizabeth Kline again replied,

Actually the surprise is more rhetorical than real. I have been reading “Dead Grammarians” (sans Wallace who is a special case) along with linguists for 25 years. Just reminding some of the “young turks” that the old NT scholars are not to be despised, the linguistic framework you (not Steve R.) are all excited about now will be dead and gone and people will continue to read the old NT scholars.

…

One significant difference between Alford and Levinsohn, Alford doesn’t attempt to draw any broad generalizations about fronted constituents and fit them into some sort of large theoretical framework, Levinsohn does. Alford is commenting on the text and not writing a grammar of discourse. That is one justification for reading linguistics.

Levinsohn (DFNTG 2000 p63) marks TWN hIPPWN TOUS CALINOUS as a “point of departure” and states that TWN hIPPWN is fronted with the “point of departure” and marks a “switch of attention” from humans to horses.
Alford tells us in language almost anyone can understand that TWN hIPPWN is new information whereas we already know about CALINOUS and STOMATA since they are mentioned in the preceding context. Alford claims that the position of TWN hIPPWN is emphatic and is connected with the (head) noun STOMATA not CALINOUS. J.Huther agrees with him on both points. The notion that fronted constituents are emphatic is a commonplace generalization in the older exegetical works. However, this observation isn’t set into a large theoretical framework like functional grammar (S.Dik, T.Givon). Perhaps this is one reason we still read Alford and Meyer, we aren’t required to demystify the dynamics of a very complex hybrid (eclectic) theoretical framework to read Alford.

The notion of “points of departure” in Levinsohn isn’t present in Alford (stating the obvious) but Alford’s treatment does have implications for discourse cohesion and thematic development. The difference is that Levinsohn pulls out, highlights and labels aspects of the surface structure that contribute to discourse cohesion, thematic development and so forth. This is a useful project, not to be despised by advocates of the traditional grammar.

So.

Read dead grammarians.

Read linguists.

It’s not an either or. It’s both and.

This is a continuation of my series examining Devine and Stephen’s book Discontinuous Syntax: Hyperbaton in Greek. Today’s post consists of much rewriting of about 15 pages I lost back in February when my hard drive died.

See also:

Part I

Part II

Part III

Part IV

Weak Focus Y₂ Hyperbaton[1]

Most commonly, Y₂ Hyperbaton will denote weak focus on the initial head noun with the following adjective marking an additional restriction on the head noun. This additional restriction also tends to be weak focus. The first three instances of Y_­2 Hyperbaton in the New Testament provide an excellent example of this combination of syntactic structure and pragmatic meaning seen below.[2]

(1) τὸ γὰρ ἐν αὐτῇ γεννηθὲν ἐκ πνεύματός ἐστιν ἁγίου
For what is conceived in her is from the Holy Spirit. Matt 1:20.

(2) ὁ λαὸς ὁ καθήμενος ἐν σκότει φῶς εἶδεν μέγα
The people who live in darkness have seen a great light. Matt 4:16

(3) καὶ ἰδοὺ ἄνθρωπος χεῖρα ἔχων ξηράν.
And behold, a man who had a withered hand. Matt 12:10

In these three examples, the author/speaker provides brand new information to the reader/listener. Example (50) is from the monologue of the angel to Joseph regarding the source of Mary’s pregnancy. The clause consists of the known information shared by both the Angel and Joseph: Mary is pregnant. There is a child conceived (τὸ γὰρ ἐν αὐτῇ γεννηθὲν). The predicate of this clause is the completely new, non-contrastive information: the child is from the Holy Spirit (ἐκ πνεύματός ἐστιν ἁγίου).

Example (51) is found within a quotation of Isaiah 9:1-2 in the Gospel of Matthew.[3] The discontinuous phrase appears late enough in the quotation from Isaiah to provide a large enough context for understanding the phrase’s information structure without recourse to the text of Isaiah.[4] The phrase appears following a description of the land of Israel, “Land of Zebulun and land of Naphtali, / the Way of the Sea, beyond the Jordan, / Galilee of the Gentiles— / the people living in darkness have seen a great light / on those living in the land of the shadow of death a light has dawned” (TNIV). In the greater context of the Gospel, the writer has just described Jesus as traveling through these regions, fulfilling Isaiah’s prophecy. Verse 15 is entirely accessible, known information; as does the Subject of the clause in question here (the people). The brand new, prominent information is “light” –a great one.

In the last of these examples (52), Matthew introduces a new scene within the periscope that describes Jesus’ relationship to the Sabbath. With the new scene, come new participants. Since the clause itself introduces brand new information, there is little doubt about the information status of the constituents: καὶ ἰδοὺ (and behold) marks the shift to a new Topic, ἄνθρωπος (a man), which also is the Subject of the clause. The rest of the clause is a participial clause (χεῖρα ἔχων ξηράν, “having a withered hand”) in the predicate position with ἄνθρωπος (a man).

In light of DS’s claims about the discourse function of Y_2­ ­Hyperbaton, these first three examples are rather striking because of them, only the last one even partially parallels their predicted discourse context which Y₂ Hyperbaton should occur. They write, “Two generalizations suggest themselves: first, the verbs are all intransitive, and second … they are all simple verbs of occurrence or existence.”[5] In our examples above, only example (50) is an existential verb and none of the three examples are intransitive. [6]_­

How do we deal with this? On the one hand, it is possible that DS limits their generalizations only to Weak Focus ­Y₂­ Hyperbaton examples where the discontinuous phrase bears the grammatical relation of Subject. This is not explicit in their discussion, but could easily be inferred from the text itself. Earlier they write, “In the simplest type of Y_{­2­ ­}Hyperbaton, [Y₁X] is an ordinary weak focus and the Y₂ Adjective specifies an additional restriction on the noun, and so represents a second weak focus. Some of the clearest instances involve split subject phrases.”[7]

1 Intransitive, Existential, or Occurrence Verbs

If we accept this perspective on the intention of DS, as well as assume that their two generalizations function as separate possibilities rather than two requirements, a clearer picture emerges. Of those instances of Subject ­Y₂ Hyperbaton, all of them fit their generalizations and many of the cases can be paraphrased in English with either, “an expletive there or an extraposed relative clause.”[8] Consider examples (53-55) with an existential verb from the New Testament and Josephus.

(4) καὶ ἐγένοντο ἀστραπαὶ καὶ φωναὶ καὶ βρονταὶ καὶ σεισμὸς ἐγένετο μέγας
And there was lightning and sounds and thunder and there was a great earthquake. Rev 16:18

(5) λίθος δὲ ἦν μέγας ἐπὶ τῷ στόματι τοῦ φρέατος
But there was a large stone on the mouth of the well. Gen 29:2

(6) κατὰ νυκτὸς ἐνάτην ὥραν τοσοῦτο φῶς περιέλαμψε τὸν βωμὸν καὶ τὸν ναόν, ὡς δοκεῖν ἡμέραν εἶναι λαμπράν
By the ninth hour of the night, such a light shone around the altar and the Temple, as to appear that it appeared to be bright day time. Josephus, Wars 6.290

In all three of these examples, the author is not seeking to contrast two items or events (i.e. Strong Focus), rather he is merely reporting that X exists or happened.[9]

There are also Hellenistic examples of Y₂ Hyperbaton with intransitive verbs shown in the following examples.

(7) θύρα γάρ μοι ἀνέῳγεν μεγάλη καὶ ἐνεργής
For a door has opened up to me, a wide and effective one. 1 Cor 16:7

(8) πηγαὶ συνεσχέθησαν αἰώνιοι ἐξ ἀβύσσων ἀπὸ ὀρέων ὑψηλῶν, ὅτι οὐκ ἦν ἐν αὐτοῖς ποιῶν δικαιοσύνην καὶ κρίμα.
Eternal Springs were shut up from the depths from the high mountains, because there was no one among them performing righteous and justice. Psalms of Solomon 17:19

(9) Εἰπὸν αὐτῷ ἐπὶ τῷ ἐμῷ ὀνόματι Κρύψον σεαυτόν, καὶ δήλωσον αὐτῷ τέλος ἐπερχόμενον, ὅτι ἡ γῆ ἀπόλλυται πᾶσα, καὶ κατακλυσμὸς μέλλει γίνεσθαι πάσης τῆς γῆς καὶ ἀπολέσει πάντα ὅσα ἔστʼ ἐν αὐτῇ.
Say to him [Noah] in my name, “Hide yourself!” And show him that the coming end: that the entire earth will be destroyed and a severe flood is about to come upon all the earth and will destroy everything on it. Book of Enoch 10.2

Of these three examples, (56) could easily be paraphrased as, “There is a door that has opened to me, wide and effective.” Also, example (58) is particularly striking because it states explicitly the reportive nature of the construction with the angel Uriel commanding the son of Lamech to bring this new information to Noah.

Finally, there is one clear example of Y_{2 ­}Hyperbaton with an occurrence verb in Josephus.[10]

(10) ἀσπασάμενος δὲ τὴν φωνὴν ὁ τοῦ Σαούλου παῖς ὡς νίκην αὐτῷ σημαίνουσαν παραυτίκα μὲν ἀνεχώρησαν ἐξ οὗπερ ὤφθησαν τόπου τοῖς πολεμίοις, παραμειψάμενοι δὲ τοῦτον, ἐπὶ τὴν πέτραν ἧκον ἔρημον οὖσαν τῶν φυλαττόντων διὰ τὴν ὀχυρότητα.
So Saul’s son accepted the voice, as it indicated to him victory. So he immediately came out of the place where they had been seen by their enemies; so he changed his place, and came upon a deserted rock guarded by its own strength. Josephus, Antiquities 6.112

The discontinuous phrase is “ἐπὶ τὴν πέτραν ἧκον ἔρημον” (he came upon a deserted rock). And once again, it is clear that Josephus is merely reporting what he views as the necessary new information in preparation for the more salient events which follow where Jonathan and his armor bearer route the Philistine army.

2 Challenges for DS’ Generalizations?

Now that we have covered the examples which can easily be explained by the generalizations of DS, what shall be done with the rest? Indeed, the majority of instances of Weak Focus Y­₂ Hyperbaton do not , at face value, align themselves to their description. These must be discussed and explained below, either by showing how they fit within the already stated generalization or by adapting and expanding the generalization so that all the data can be explained. This will be the focus of Part VI, which will appear in much better time than Part V did.

[2] There is one construction that could technically appear here, but will be discussed elsewhere, the construction: ἐγὼ χρείαν ἔχω X (cf. Matt 3:14). This structure is technically a Y_2­­ Hyperbaton, but the modifier constituent is so varied it cannot be discussed consistently in this section with its focus specifically on Adjectives. Secondly, the nature of the construction as a stereotyped phrase deserves it own independent discussion later on.

[3] The quotation and its word order differ from both the LXX (Rahlfs) and the Masoretic Text. The LXX reads πορευόμενος instead of καθήμενος and neither the LXX nor the MT has the discontinuous phrase we find in Matt 4:16. For this reason, I do not view this instance of Hyperbaton influenced by its Hebrew source text.

[4] This assumes, of course, that Matthew did not adapt the text himself, but drew from an Isaiah manuscript that included the discontinuous phrase.

[5] DS, 92; they provide no definition for “verbs of occurrence.” But according to Angela Downing and Philip Locke (English Grammar: A University Course [London: Taylor & Francis, 2006], 85), verbs of occurrence are those such as, “appear, disappear, go, come, arrive, depart, vanish, fade, happen…”

[6] Though according to our reading of (52), the discontinuous phrase, with its participle, functions as the semantic predicate of a nominal clause and, by definition, nominal clauses are existential.

[7] DS, 91.

[8] Ibid, 93.

[9] Of note in the Josephus example is that the existential infinitive within the discontinuous phrase is governed by an occurrence verb: to appear.

[10] Since this is the only clear example, the context is quoted more fully.

Michael Whitenton over at Ecce Homo has linked to a very interesting PDF/power point on word order in Greek and Indo-European by Dag Haug. The PDF is pretty good, though I wouldn’t subscribe to all the conclusions. Do check it out: For the Über-Nerds Among Us.

As for Dag Haug, himself, I didn’t recognize the name, so I did a google search for him and the word Greek. I got some very interesting hits. The Google Books result (third below) in particular looks interesting and I think I’ll be reading through that chapter. The last one, the BNET article on the Greek Perfect might be worth reading through as well. Finally, his project site looks incredibly interesting: HERE.

No, this post isn’t going to provide a comparison of how word order in Classical Greek differs from that of Koine Greek. Rather, I want to compare how such studies have been done in Classical Greek and Koine, specifically with regard to their use of statistics:

Ivan Shing Chung Kwong, in his study of word order in Luke writes the following:

The statistical result of the study contributes to the figure of relative positions between the three main constituents in two aspects: unmarked word order patterns and tendencies of certain word order patterns–a very high percentage of occurrences of a certain word order pattern indicates an unmarked order (a very regular/typical order patter, e.g., subject precedes the complement in a clause [as stated below]); as relatively high percentage of occurrences of a certain word order pattern indicates a certain degree of tendency of having such a word order (the percentage of a certain pattern is not as high as the unmarked ones, but it is still relatively high to demonstrate a certain degree of tendency of having such a word order pattern, e.g., subject tends to precede its  predicate in independent clauses [as stated below]).

The Word Order of the Gospel of Luke: It’s Forgrounded Messages, Page 45.

Now compare that with what Helma Dik says in her first book on Greek word order in Herodotus:

We have seen 48 instances in this section, many of which discussed in the previous sections of this chapter. Throughout, I have refrained from giving any statistics of the ordering patterns found, mainly because I do not think that statistics on the order of, for instance, the three core constituents would be very illuminating. In any case, the number of instances examined here is so small, that we cannot even expect ‘significant’ results in the technical sense. Be that as it may, Table 4.1 presents the distribution over the various ordering patterns as introduced in section 4.1.

Table 4.1 Ordering patterns with two arguments expressed

Pattern	Instances	Total
A1-A2-P	1.66.3; 1.188.1; 4.160.2; 6.7	4
A2-A1-P	3.1.1	1
A1-P-A2	1.161; 3.25.2; 4.173; 5.14.2; 6.28.1; 6.108.4 [3.19.3; 3.52.7; 3.151.1; 4.80.2]	6 [4]
A2-P-A1	3.44.1	1
P-A1-A2	1.73.1; 3.47.1	2
P-A2-A1	1.166.1	1

What conclusions can we draw from this distribution? At first sight A1-P-A2 [e.g. Subject-V-Object] appears to be the preferred pattern. If we include 3.19.3, 3.52.7, 3.151.1, and 4.80.2, this pattern covers more than half of the total number of instances. Does this make A1-P-A2 the unmarked order and should we proceed to consider only the marked instances? Clearly this would be a complicated procedure; with five different ‘marked’ patterns, there is a lot left to explain. In the preceding chapter I have tired to show an alternative way of handling the data. From this it appears that we need not describe one pattern as the preferred or unmarked pattern, but that differences in pragmatic function assignment can account for the different orderings. In such a description, we would not regard A1-P-A2 as unmarked, but explain its frequency from the fact that it is often the first argument that has Topic function and, therefore, ends up in initial position, and that in continuous narrative the predicate is the primary candidate for Focus function. Ironically, if we wanted to describe one pattern as unmarked in this description, it would be P-A1-A2, for it is very simple to describe the five other patterns in terms of ‘deviations’ (resulting from pragmatic function assignment) from this pattern.

In the clauses in which only one argument is present, we can see a similar phenomenon. A1-P is the ‘statistically unmarked’ option; I have argued that in A1-P the first argument is pragmatically marked.

Word Order in Ancient Greek: A Pragmatic Account of Word Order Variation in Herodotus, Page 93-4.

And now for my commentary on these two discussions.

First of all, Kwong and Dik’s perspectives are polar opposites.

Secondly, the quote from Kwong above is a single long and cumbrous sentence that really gets under my skin.

Third (and to the point), Dik is highly critical of the sort of statistics that Kwong has based his entire study upon. In light of her statements on statistics, Kwong’s literature review comments about Dik are rather striking:

The method and approach of Dik’s work is strange and largely disappointing. . . . Dik’s attempt to study the area pragmatically on the theory of Topic and Focus is a good try, but she fails to have a reasonable methodology and thus her work cannot give a satisfactory figure on Herodotus’ pragmatic word order (Kwong, 20; you can look up this page on Google Books if you’d like).

A couple thoughts on these words. I truly find them to be incredible. The Bryn Mawr review of Dik’s first book was incredibly positive – the exact opposite of Kwong. The only explanation I can think of consists of a combination of Kwong simply not understanding Dik with the extremely high value that he places on statistics in his own study (the few books that he speaks positively of in his lit review tend to be statistical). The vast expanse between the two approaches, one statistical and the other wholly pragmatic seems to have also introduced a gap in comprehension.

Personally, and this is my opinion (of course, I consider my opinion is correct, but am willing to be convinced otherwise), I think that Dik’s approach does far more justice to the data itself and is significantly closer to reality for Greek word order. In the end, Kwong’s work on word order will be viewed as idiosyncratic, which is unfortunate, because there is some good stuff in the book, particularly on semantic chaining (there’s also some good discussion of Topic change in chapter 8). He is simply too dependent upon statistics to be the last word, especially considering that word order statistics often vary from author to author and even from book to book within an author. I’d be curious as to whether Kwong’s word order proposals in Luke’s Gospel also work for Acts. I haven’t checked, but I’m a bit cynical.

With all of that said, there is also some strength in his exegetical claims, but I think that has more to do with how he relates his statistical findings to other linguistic/grammatical factors within the text than it does with any sort of true discovery about word order.

As for Dik, she doesn’t have everything right either. Her basic paradigm is great, though I’m suspicious about her claim that Verbs can be Pragmatic Topics, which is a unique distinction from typical Functional Grammar (and in every other framework I’ve read on this subject). As for the issue of simplicity, her explanations of constituents that follow the Verb leaves something to be desired. Post-verbal constitutents are  relegated to the category, “everything else” (in her 1994 book). I think there’s more structure following the verb than that. At times, she seems to want everything to be explained via pragmatics rather than other components of Grammar, which makes me nervous — though I’m not even sure if that’s an accurate description of her view either, so we’ll put that question aside for now.

But at least Dik is aware of the problems and dangers of overdependence upon statistics, which Kwong almost seems to embrace with open arms. Also her focus is on looking at word order as a discourse level phenomenon, while Kwong is entirely focused upon the sentence. I suspect that this is the cause of Kwong’s misreading of Dik. At least, that’s my working hypothesis.

So there are some reflections on two very different books on the same topic. I hope I haven’t bored you all to death yet.

Typical/Default word order in English requires Core Arguments to appear closer to the verb than Oblique Arguments:

We do not tend to say things like:

The doctor removed with a knife her kidney.

Rather, we would tend to say:

The doctor removed her kidney with a knife.

In both sentences “her kidney” is a core argument and “with a knife” is an oblique argument.

Now, I know that both these sentences and the ones in the translations below are comprehensible. That’s not what’s at stake here. They mean basically the same thing, so when you comment on this post, please don’t give that whole argument about how this stuff about naturalness doesn’t really matter as long as the translation is “literal.” For one I’m really tired of that argument and I think its nonsense anyway, so you won’t convince me with it. But secondly, and more significantly, such an argument shows little awareness of how the languages work. When we translation Greek word order that is both natural and default in such a way that makes the English word order neither natural nor default, we also subtly change the meaning of the text. Specifically, we change the pragmatics of the text. When we place a prepositional phrase between the verb and the object, we draw greater attention to it. Were someone to actually speak the first sentence above, they would likely stress the phrase “with a knife” in their intonation. They may even pause after saying, “with a knife.” Third, its debatable whether this could be consider a “literal” translation anyway.

Now, on to the text at hand: Acts 15:37.

Last night I came across this in a KJV tradition translation and it caught my eye. So I looked at a variety of other translations. Here are the ones with a rather awkward sounding “with them” placed directly between the verb and the direct object – so much for the translations that are supposedly “good for public reading…”

Barnabas wanted to take with them also John, who was called Mark, NAB

And Barnabas was minded to take with them John also, who was called Mark. ASV

Now Barnabas wanted to take with them John called Mark. ESV

Barnabas wanted to take with them John called Mark. NRSV

And Barnabas wanted to take with them John called Mark. RSV

And Barnabas determined to take with them John, whose surname was Mark. KJV

There’s another problem from an English perspective with these translations. Its the combination of the strange position of “with them” and the “also” of the NAB & ASV. In English, “also” marks a focused constituent in a given clause. So what we have here, in the NAB & ASV, is essentially focus overkill and its often ungrammatical. We have two emphatic constituents in the same clause. This sort of thing can only confuse the readers. Now the translators of the KJV seem to have recognized this fact, which is why they do not have an “also” (though it is in Greek). The ASV, in following the the KJV, while also wanting to be the most literal translation it could, reinserted the “also” and it seems that the NAB, independently of the ASV (but perhaps unconsciously influenced by the KJV?), did the same. In contrast, the RSV (and thus also the ESV & NRSV) followed the KJV more closely.

Here are the translations that maintain the “with them” phrase, but place it in a significantly more natural spot.

Barnabas wanted to take John, also called Mark, with them, T/NIV

Barnabas wanted to take John, called Mark, along with them also. NASB95

Barnabas wanted to bring John called Mark along with them too, NET

Barnabas wanted to take John Mark with them, NCV

Barnabas wanted to take John Mark with them. NIrV

Barnabas wanted to take John Mark with them, GNB

These translations sound infinitely better in English (I’m willing to give the KJV the benefit of the doubt that its odd word order could have been felicitous in 1611). But not all of them have draw attention to John, called Mark, with an “also” or “too,” loosing the Focused nature of the Direct Object here (the NIrV, GNB, and NCV).

But this is where I get nitpicky (which is a silly word). Let’s talk about accuracy/literalness now. And this relates to examining why the first group of translations put the prepositional phrase, “with them,” where they did. Here’s the Greek text:

Βαρναβᾶς δὲ ἐβούλετο συμπαραλαβεῖν καὶ τὸν Ἰωάννην τὸν καλούμενον Μᾶρκον·

It seems that the first group of translations, the KJV tradition, were trying to keep “with them” as close to the verb as possible because in the Greek text, “with them” is part of the verb – the derivational prefix συν-. That makes sense, I suppose, but its about as silly as saying that you shouldn’t split an infinitive in English because Latin can’t. So, is it more literal to keep this prepositional phrase as close to the verb as possible?

Well, perhaps, but it introduces problems. A truly literal translation would require the translation of the adverbial καί, which modifies the direct object. So if you do not include it, you do not have a literal translation (ESV, NRSV, RSV, KJV). But if you do include it, you still don’t have a literal translation because then we have some sort of double Focus construction that does not exist in the Greek and is barely grammatical (if at all) in English.

Then if you follow the path of the of the second group of translations, particularly the first three, you’ve distanced the “with them” from the verb, so that we have a natural and default ordering of the Arguments as well as the marking of Focus on the Direct Object – though unfortunately, by placing the “also” right before “called,” the TNIV fails by placing Focus in the wrong place. It should be on the entire direct object “also John, called Mark,” not “John, also called Mark.”

So under normal circumstances, I would definitely say that the approach of the NASB95, and NET is by far the best and most literal approach to dealing with this verb.

But that was before I read what the NLT said at this verse. That’s right. I said the NLT. Here’s its translation:

Barnabas agreed and wanted to take along John Mark. NLT

For the verb itself, the NLT has got it right. They’ve successfully kept the meaning of the “with them” phrase directly with the verb by recognizing that the Greek prefix συν- doesn’t mean “with.” Rather it denotes the abstract concept of accompaniment, which is often expressed by “with” in English, but not necessarily so. In terms of literal translation, “along” does a far better job of maintaining cohesion with the verb, but also accurately and naturally conveying the meaning in English. After I read it, I check and found two others that roughly take this approach:

Barnabas persisted in wanting to take along John, who was called Mark, ISV

Barnabas wanted to take along John Mark. HCSB

But of course, there is still an issue with all four of these. None of them translation the καί, which, I would argue, is necessary for accurately conveying the emphatic nature of the Greek’s καὶ τὸν Ἰωάννην τὸν καλούμενον Μᾶρκον.

And that’s where things got amusing for me in this little search. I decided to check The Message:

Barnabas wanted to take John along, the John nicknamed Mark. The Message

Peterson essentially gets it all right where 15 other committee translations got it wrong.

“Wait,” you say, “He didn’t translate the καί either!” Ah, but there you’d be wrong, though I doubt Peterson did this intentionally. Let me show you why:

When we deal with phrasal verbs such as “take along.” We are invariably going to be dealing with two potential word orders:

Take along X

Or

Take X along

Now, think about how you would say both of these out loud. Which one of them would more likely lead you to place intonational emphasis upon X? I’m willing to bet that you’re more likely to say the second one (that and I’ve already surveyed people in the library where I wrote this). The fronted direct object makes it more appealing.

And that’s exactly how The Message translates the καί, with its English word order.

Who would have expect that The Message would get it right where everyone else failed?

I was reading Dik’s book Word Order in Greek Tragic Dialogue in Amazon’s preview, curious about what she says about Devine and Stephen’s work (Discontinuous Syntax: Hyperbaton in Greek, 1999), which came out after her first book on word order in Herodotus (Word Order in Ancient Greek: A Pragmatic Account of Word Order Variation in Herodotus,1995).

Looking through the discussion, it seems like her second book is likely going to be the more important work simply because she has a larger number of other recent books to argue and dialogue with. What I’ve read has been so-so. There’s some good and some bad – I think she’s wrong to reject D&S’s combination of pragmatics and semantics for their explanation of adjective orderings (she prefers pragmatics only with at least one caveat), but I think (only intuitively, I haven’t checked) that her criticism of D&S’s claim (that adjective ordering differences between Herodotus and Thucydides is best explained by dialect difference) is spot on. With that said, D&S tend to make many more typological comparisons than Dik ever does, which is a clear strength to their book. And then there’s the issue of price. You can get D&S for less than $20 used while either of Dik’s books go for at least $75. Who knows, that might end up being the factor in who wins the debate – though at the same time, Dik is infinitely easier to read than D&S.

What is interesting, and I’m curious if Stephanie Bakker will continue the trend (see below), is that Dik in her 2007 book consistently follows D&S’s terminology for the different types of discontinuous phrases in Greek syntax – e.g. Y1 Hyperbaton (AXN) & Y2 Hyperbaton (NXA).

There are more books coming out on the subject though. Stephanie Bakker (any relation to Egbert J. Bakker?) is writing a book on the Noun Phrase in Greek (The Noun Phrase in Ancient Greek: A Functional Analysis of the Order and Articulation of Np Constituents in Herodotus). And according to Dik (Word order, 2007; page 88), Rijksbaron is currently researching definiteness in Greek and S. R. Slings is currently working on a functional analysis (as opposed to D&S’s generative analysis) of Hyperbaton in Greek. I don’t know how I feel about this. I think that D&S are generally correct in their explanation of hyperbaton. Why can’t we simply work from different theoretical perspectives and build on each other? Framework rivalry is always such nonsense.

Finally, what is up with Herodotus? Why is he so popular for word order studies? I don’t know…

So all in all, it looks like the next ten years are going to be quite productive for Classical Greek word order studies.

I hope we actually get somewhere with all of this writing (yes, that’s cynicism you hear).

And I hope some of it spills over into the Hellenistic Period (more cynicism – though I do hope to make my own contribution to this subject at some point in the relatively near future – 3 to 5 years).

I retract all of my criticisms of the Opentext.org Syntactically Analyzed Greek New Testament. From this point on, I view Opentext.org as an excellent database.

Why the change of heart?

Well, today it hit me what Opentext.org is and what it is not.

As for what it is not, well, simply put, it is not a Syntactic Analysis; at least generally speaking. And regarding what it is, Opentext.org is a semantic database.

That is not to say that there isn’t syntax in the database. There is. Specifically, following Opentext.org’s definitions, the labels Clause,  Conjunction, Word Group, Head Term, and Connective are the syntactically conceived labels. But they are the only such labels in the database. Every other label used in Opentext.org is a semantic label, not syntactic. This includes terms like, “Subject” and “Complement,” which rightly according to Opentext are “functional unit[s]” (a good number of linguistic frameworks call them “grammatical relations” or grammatical functions”). Subjects, Objects, Complements, and Predicates are all semantic labels.

Now all of this doesn’t mean that I think Opentext is a satisfactory database. Its not. What it does, it does quite well and that is why I call it excellent above.

But.

By placing so much focus on semantics and so little on syntax, pragmatics, or other components of grammar, they’ve missed too much information for Opentext to be a satisfactory database. This is also why breakthroughs in the study of Greek word order from Opentext.org have essentially be nonexistent. The database asks the wrong questions for dealing with word order.

Now, with that said, its possible to use Opentext for studying word order. The fact that phrases/word groups are divided out is incredibly helpful, but we’re then left with a whole lot of sorting through the information and reanalyzing it using different criteria than what Opentext provides.

Okay, so maybe I really don’t see much value in using interlinears and just wanted a title that would catch your attention.*

But now that I have it, please stay because I would like to talk about the benefit of something closely related to interlinears.

Interlinearization.

That is, of course, the process of creating an interlinear. Many of you know that I’m working on an interlinear project right now for Logos Bible Software for the LXX. And while I would not recommend using an interlinear for learning the language, I’ve been pleasantly surprised by what I’ve learned about Greek by looking at word order phenomena – specifically, what words and phrases are placed in front of the verb ad when, what words and phrases follow the verb and so forth.

I’m going to guess that at some point you were told in a first year grammar class that word order doesn’t matter – or something like that.

Well, I’m here to tell you, whoever told you that is completely wrong.

* That is to say, I don’t seem value in learning the language when it comes to interlinears. I do think there is value in them. I’ve seen first hand the busyness of the pastor’s life and I know that its hard enough for them to learn the vocabulary enough for the New Testament, much less beyond and for that reason, I think an LXX interlinear or an Apocrypha interlinear or an Apostolic Fathers interlinear would be beneficial for the busy pastor who doesn’t have the time or energy for going beyond learning New Testament vocabulary. That is exactly why I am whole heartedly behind the LXX interlinear project.

I spent a good amount of time today looking through references where Robertson talks about postpositive conjunctions appearing in the wrong place:

(i) Postpositives. A number of words are always postpositive in Greek. In the N. T. ἄν, γάρ, γε, δέ, μέν, μέντοι, οὖν, τε never begin a sentence, in harmony with ancient Greek usage. These words commonly in the N. T. come in the second place, always so with μέντοι (Jo. 4:27, etc.). In the case of μέν the third place is occasionally found as 1 Pet. 2:4, the fourth as 2 Cor. 10:1, the fifth in Eph. 4:11; Jo. 16:22, or even the sixth in Jas. 3:17. It occupies the seventh place in Herm. Sim. viii, 5:1 (Mr. H. Scott has noted). In general these words vary in position according to the point to be made in relation to other words. So also οὖν is more commonly in the second, but varies to the third (Jo. 16:22) and fourth (1 Cor. 8:4). The same remark applies to γάρ, for which see Mk. 1:38; 2 Cor. 1:19. As to δέ, it may not only go to the fourth place (Jo. 8:16), but even appears in the fifth (1 Jo. 2:2), οὐ περὶ τῶν ἡμετέρων δέ. It stands in the sixth place in Test. XII. Patr. Judah, 9:1 (Mr. H. Scott reports).
A.T. Robertson, A Grammar of the Greek New Testament in the Light of Historical Research (Logos, 1919; 2006), 424.

Looking through the references, there were some interesting patterns:

Mark 1:38 (UBS4)
[FOC: εἰς τοῦτο] γὰρ ἐξῆλθον.

John 16:22 (UBS4)

22 [καὶ [TOP: ὑμεῖς] οὖν νῦν μὲν λύπην ἔχετε·]

1 Corinthians 8:4 (UBS4)

4 [Περὶ [FOC: τῆς βρώσεως]] οὖν [τῶν εἰδωλοθύτων], οἴδαμεν ὅτι οὐδὲν εἴδωλον ἐν κόσμῳ καὶ ὅτι οὐδεὶς θεὸς εἰ μὴ εἷς.

2 Corinthians 10:1 (UBS4)

[[TOP:ὃς] [FOC: κατὰ πρόσωπον] μὲν ταπεινὸς ἐν ὑμῖν]

1 Peter 2:4 (UBS4)

4 [[FOC: ὑπὸ ἀνθρώπων] μὲν ἀποδεδοκιμασμένον]

1 John 2:2 (UBS4)

2 καὶ αὐτὸς ἱλασμός ἐστιν περὶ τῶν ἁμαρτιῶν ἡμῶν, [οὐ [FOC: περὶ τῶν ἡμετέρων] δὲ [μόνον]]

I looked through some others too that I searched for. Not all of them are focused. Some are topics, but the majority are contrastive in either case. It seems that postpositives that are not in the right place are significant, which Robertson himself notes in passing in the quote above when he write, “In general these words vary in position according to the point to be made in relation to other words.”