If you read through my papers, articles and monograph, you will note that I never have claimed to do discourse analysis. Never. I have made reference to what it has to offer, but have been extremely meticulous about classifying myself as a grammarian rather than an analyst. Why? Because I have not DONE any discourse analysis. The exception might be when I give a reading of a passage or in the High Definition Commentary: Philippians, but the latter offers up little detail about how I arrived at my conclusions.

Discourse analysis, in my view, is the synthesis of all linguistic, rhetorical, literary, semantic or pragmatic information in a discourse into a coherent and cohesive whole. Synthesis is a key word here, with the objective being to demonstrate the cohesion of the various parts into a coherent whole. If you are not doing synthesis of parts into a coherent reading, you aren’t doing discourse analysis. This is why I have not claimed to do discourse analysis before this summer.

So what exactly have I been doing? Linguistic description of the pragmatic pieces and parts. In my view, this description is a necessary first step if one has any hopes of synthesis. If you miss some of the parts or fail to correctly understand their unique contributions, what are the consequences for your synthesis? In light of this inauspicious potential for derailment, I have focused–perhaps inordinately so–on really understanding the parts, and hopefully helping others to do so in the process.

So you might be wondering at this point if I have any interest in discourse analysis. Indeed I do. However I have known for some time that I was missing a piece in my framework that prevented me from effectively working in this area. What was it? Drawing a meaningful distinction between continuity/discontinuity and cohesion. How did I know this? Since roughly 2003 Stephen Levinsohn has been trying (rather unsuccessfully, I’m afraid to say) to help me understand this difference, but I have turned out to be a slow learner. He’d explain it and I’d give him a blank stare in response. You see, I have done my best to conflate the two as though they were the same thing, as though they described the same parameter.

Continuity and discontinuity have to do with how we break a discourse down into smaller chunks. Generally speaking, we chunk a discourse where there are changes in time, place, participants or kind of action. For instance the change in location from Lydda to Joppa in Acts 9:36 suggests a change in the story, a new chunk. Givon has claimed that there is a relationship between the number of changes there are in these parameters and the level of the break in the discourse. For example, if there is simultaneously a change in time, location and participants, chances are there is a pretty big transition in the discourse. Continuity of these factors leads to judgments of continuity in the discourse, discontinuity in one or more of these factors leads to judgments of discontinuity. So continuity has to do primarily with segmentation and chunking, determinations about what things go together and which ones don’t.

But wait, there’s more! There is another, completely separate parameter that deals entirely with how things hang together The parameter? Cohesion. It is only concerned with how things hang together. And here is where I stumbled. I thought that cohesion and continuity sounded a lot alike, and they do. However they are focused on different parameters. You see, even discontinuity adds to cohesion. Say what? Think about it. If there is a change from Lydda to Joppa in Acts 9:36, this can indeed be viewed as a discontinuity. But on the other hand, since there is a thematic link between the places, in terms of cohesion this change is actually cohesive link binding the two parts together.

Recall the definition from p. 194 of the Discourse Grammar for a frame of reference: it provides the primary basis for connecting what precedes to what follows. So although there may be a change of topic, time, location, etc, the frame of reference provides a cohesive link between the two parts. In other words, although a frame of reference represents a discontinuity, it also represents a cohesive link between what may be viewed as two disparate things. It is a cohesive bridge across the gap of discontinuity to make sure the reader successfully navigates the transition. Think about it. If there is a major discontinuity or transition in a discourse, you risk losing your reader. They might miss it or lose the bigger cohesive picture in the process. A writer needs to make sure that the reader recognizes the transition, hence frames and other devices to clearly signal the discontinuity. But if the discourse truly builds into a cohesive whole, there must still be something connecting the parts. This is where cohesion comes in, tracking those things the writer uses to help the reader make the proper connections.

Here is the kicker, the thing that I failed to grasp. The same devices that are used to mark discontinuities on the continuity/discontinuity plane also function to cohesively link the parts on the cohesion plane. They are related yet distinct planes, not to be conflated.

I gave two papers at the International Society of Biblical Literature meeting in London this summer. Questions after my first one got me thinking, but the real credit for helping me understand the error of my conflating ways goes to Dr. Cynthia Westfall of McMaster Divinity School. Her incisive questions after my second paper led to a huge “aha” moment, a long and contemplative “ohhhh” the likes of which written language cannot do justice. Lights went on, pangs of anguish struck as I realized what I had been missing.  Levinsohn spoke up right after her, further prosecuting her point.

On the train ride out of London and for the next few days, Levinsohn and I discussed cohesion in great detail. If you are looking for a tidy introduction, I’d recommend the one in Dooley and Levinsohn’s Analyzing Discourse. Only be sure not to conflate cohesion with continuity. I know the terms sound similar at face value, but in linguistics they are describing very distinct concepts. Continuity/discontinuity concerns chunking into smaller parts, cohesion describes how the parts fit into a cohesive whole.

I have always made an intentional distinction between discourse grammar and discourse analysis. I have known I was missing the boat on something important, so I avoided making claims about discourse analysis like I avoid the plague. However a great many people who have followed my research had claimed on my behalf that I am doing discourse analysis, something which I have sought to correct.

I have always viewed the grammatical description of devices as the necessary first step to any kind of analysis, hence writing Discourse Grammar. It has been my belief that once I understood what each of the pieces does, I would be in a much better place to synthesize the parts into a whole. I have also known that I failed to grasp something very important about cohesion. Thanks to Westfall and some strategies I learned from Levinsohn, I am in a much better place to venture into cohesion and full-blown discourse analysis.

This November at ETS,((Discontinuity of time and place, cohesive link between the two parts)) I’ll be participating in a panel discussion on different approaches to discourse analysis. I think it’s pretty cool that God saw fit to help me figure things out before I had to give the paper! The focus will be on the book of James, and I’ll be analyzing chapter 2. In light of this, you’ll be seeing more posts on analysis as I practice what I’ve learned. I am very thankful for having had such a great mentor for learning about the pieces, and for his persistence in pressing me about considering the whole.

And that’s the rest of the story.

1. I like bananas and I like grapes.
2. *I like bananas but I like grapes.

Now if I had said that I loved, hated or some other kind of action in clause 2, there would be no problem using but. The same would hold true if I changed the subject to someone else but kept the action the same.

So what’s my point? The use of but constrains us to read the two things as somehow semantically discontinuous. If there is not some kind of contrast inherently present in the context, using but sounds wrong or strange. Take a look at the thing but with noun phrases this time.

1. I like bananas and grapes.
2. *I like bananas but grapes.
3. I like everything but grapes.
4. *I like everything and grapes.
5. I like everything even/including grapes.

Note what happens in 2. with the but. You feel like there should be another sentence like “…. but grapes I strongly dislike.” Using it makes you expect that there is some kind of contrast. If it ain’t there, it doesn’t sound right. Now in 3. but is needed to indicate that grapes are being excluded from the everything. Note how using and in 4. sounds wrong. Why? Because you are trying to add grapes to something that theoretically already includes it: everything. However, using an “additive,” an adverb that either intensifies or portrays the added element as though it was the least likely possibility, is completely acceptable. Another candidate additive would be especially.

So just to summarize the point so far, certain conjunctions like but can constrain you to read the linked elements as contrastive. However, these words to not create contrast that wasn’t already there, they simply amplify it. If there is no contrast present, using a contrastive conjunction is infelicitous as the linguists say. It comes across as wrong.

Now let’s hop over to Greek. In contrast to and and but which indicate semantic continuity or discontinuity respectively, και and δε are unmarked for this feature. Since they do not indicate anything about semantic continuity/discontinuity, we find them used in both contexts. These Greek conjunctions are differentiated by development, not semantic continuity. If you have the Logos version of the Discourse Grammar, begin reading here. Otherwise, you can begin reading about the main Greek conjunctions on page 17 of the sample I’ve posted online.

So returning to the opening statement, many descriptions of Greek conjunctions treat them as though the contrast is inherently present in the conjunction, at times as though adding this conjunction creates contrast that another would not have created. If there is semantic contrast present in the context, then certain conjunctions can amplify what was already there. This is the case with αλλα versus και or δε. The former constrains what follows to be correcting or replacing some aspect of what precedes.

Continuity and discontinuity are not binary items, but two ends of a continuum. Here’s an older post on the matter. My kids are badgering me to take them to their youth group bbq, but head over to Rick Brannan’s blog for a great paper on αλλα that unpacks these issues in more detail. I’ll try to add the link later.

• RBL: http://www.bookreviews.org
• JETS: Contact Joel F. Williams, NT Book Review Editor, Cedarville University.

Hendrickson has released a second printing which includes indexes and a number of corrections. For those of you with the first printing, a PDF of the indexes can be downloaded here.

Levinsohn at the presentation in London

On July 4, 2011, more than two years’ of preparation came to fruition when I presented  Discourse Studies and Biblical Interpretation: A Festschrift in Honor of Stephen H. Levinsohn to the honoree at the International meeting of the Society of Biblical Literature. Levinsohn has played a significant role in shaping my approach to linguistics, a process that first began via email in 2000. Over the years he has served as an informal doctoral adviser and mentor. I’d send him questions and he’d reply with reading lists, very much in keeping with the British model of research.

For those unfamiliar with the term, a Festschrift is a collection of essays typically written by students and colleagues that is presented to a scholar upon retirement, turning 65 or 70, or some other milestone. Since Levinsohn has never formally held a post at a college or university, the contributors were all scholars with whom he’s interacted over the years: at conferences, on publication projects, or both.

When the location for the 2011 ISBL meeting was first announced in 2009 I began looking into possibility of organizing a Festschrift volume. Potential contributors were contacted, a special session was requested from the SBL, but the real glitch was a publisher. No one I contacted expressed much interest in producing the project  since these projects typically lose money. I approached Logos Bible Software about the matter, and they were willing to take on the project despite the financial outlook. They agreed that Levinsohn deserved to be honored, and committed to invest the resources needed to see the project come to fruition.

From left-to-right, Margaret Sim, Jan and Buist Fanning, Jenny Read-Heimerdinger, Stephen Levinsohn, yours truly, Bryan Harmerlink of SIL, and Ronnie Sim

When the last articles from contributors arrived this January, John Barry and his publication team took over the project, doing a fantastic job. The final version was sent off to production by early Spring, more than enough time to ship the presentation copies of the book to the UK. On the morning of July 4, Jenny Read-Heimerdinger, Margaret Sim, Buist Fanning,  and I presented papers, with Ronnie Sim presiding over the session. The papers were new, not the ones presented in the book. At the end, a copy of the Festschrift was presented to Stephen. In his usual style, Stephen questioned each of the presenters, pointing out areas that needed clarification or correction. That evening Logos sponsored a dinner for the presenters in attendance.

Few projects I have worked on have proven more difficult or more rewarding. Seeing Levinsohn’s work honored has been a bucket-list item of mine for some time, and it was really a blessing to see everything come together. Here is a list of the contributions:

22 καὶ ὑμεῖς οὖν νῦν μὲν λύπην ἔχετε· πάλιν δὲ ὄψομαι ὑμᾶς, καὶ χαρήσεται ὑμῶν ἡ καρδία, καὶ τὴν χαρὰν ὑμῶν οὐδεὶς αἴρει ἀφʼ ὑμῶν. 23 Καὶ ἐν ἐκείνῃ τῇ ἡμέρᾳ ἐμὲ οὐκ ἐρωτήσετε οὐδέν. * ἀμὴν ἀμὴν λέγω ὑμῖν, ἄν τι αἰτήσητε τὸν πατέρα ἐν τῷ ὀνόματί μου δώσει ὑμῖν. 24 ἕως ἄρτι οὐκ ᾐτήσατε οὐδὲν ἐν τῷ ὀνόματί μου· αἰτεῖτε καὶ λήμψεσθε, ἵνα ἡ χαρὰ ὑμῶν ᾖ πεπληρωμένη.¹

The issue at hand is where the phrase ἐν τῷ ὀνόματί μου belongs. Or more specifically, to which clause does it belong. The traditional reading of the text would be to understand it as “whatever you ask the Father in my name,” which assumes that the clause uses default ordering of the constituents. Take a look at the SBLGNT reading:

22 καὶ ὑμεῖς οὖν ⸂νῦν μὲν λύπην⸃ ἔχετε· πάλιν δὲ ὄψομαι ὑμᾶς, καὶ χαρήσεται ὑμῶν ἡ καρδία, καὶ τὴν χαρὰν ὑμῶν οὐδεὶς ⸀αἴρει ἀφʼ ὑμῶν. 23 καὶ ἐν ἐκείνῃ τῇ ἡμέρᾳ ἐμὲ οὐκ ἐρωτήσετε οὐδέν· ἀμὴν ἀμὴν λέγω ὑμῖν, ⸂ἄν τι⸃ αἰτήσητε τὸν πατέρα ⸂δώσει ὑμῖν ἐν τῷ ὀνόματί μου⸃. 24 ἕως ἄρτι οὐκ ᾐτήσατε οὐδὲν ἐν τῷ ὀνόματί μου· αἰτεῖτε καὶ λήμψεσθε, ἵνα ἡ χαρὰ ὑμῶν ᾖ πεπληρωμένη.², Jn 16:22–24.))

Note that the phrase is placed at the end of the main clause rather than at the end of the topical frame ἄν τι αἰτήσητε τὸν πατέρα. So in this reading, Jesus’ name is not the manner in which we ask, but the manner in which the Father gives or answers. Here is the manuscript evidence regarding the issue from the NA27 apparatus:

†5 6 1-4 P^5vid א B C* L Δ l 844 pc sa ac² (SBLGNT reading)
❙ 5 6 118 pc
❙ txt P^22vid A C‎³ D W Θ Ψ f ¹³ I. (33) M lat(t) sy pbo bo (NA27 reading)³

Regarding the † symbol on the first line, the apparatus has the following note: “A cross marks a change in the text from the 25th edition, where the reading so marked stood in the text (cf. Mt 7,18 and; 20,18). These passages always represent very difficult textual decisions.”⁴ This means that the SBLGNT reading is not so much a novel one as “going retro” so to speak. Food for thought.

The syntactical analysis of both Cascadia and OpenText analyze the NA27 reading as the prepositional phrase ἐν τῷ ὀνόματί μου modifying the topical frame, not the main clause. Interestingly, the Cascadia analysis of the SBLGNT links the prepositional phrase in question with the topical frame by crossing the main clause. Here is a snapshot.

I am not really sure what motivated the crossing, as it doesn’t really seem to represent what is going on in the syntax. This does not seem like a likely instance of discontinuous syntax, something normally associated with fronting for marked focus/emphasis.

The NA27 reading is somewhat ambiguous regarding which clause ἐν τῷ ὀνόματί μου should be affiliated with. The traditional reading has been that it goes with “whatever you ask,” which is consistent with v. 24a, that up to this point they have not asked for anything in his name. The alternative understanding, one which reconciles the manuscript evidence regardless of its placement, is to understand ἐν τῷ ὀνόματί μου as part of the main clause rather than part of the topical frame. If this is correct, then the change in the SBLGNT is a reduction in prominence on the phrase from either marked focus/emphasis or a frame of reference to default focus. Thus, our receiving things from the Father would not be based on the manner in which we ask–appealing to Jesus. Rather, the Father would give to us on the basis of Jesus’ name.

I’ll leave it to the text critics to debate the merits of each reading, but the two need not be viewed as radically different, if one changes the traditional understanding of the NA27 reading. I’d be interested in hearing thoughts. I must admit I had never really given this issue much thought, and I do not have the time to do so at the moment.

1. Eberhard Nestle et al., Novum Testamentum Graece (27. Aufl., rev.; Stuttgart: Deutsche Bibelstiftung, 1993), 303-04.
2. Michael W. Holmes, The Greek New Testament: SBL Edition ( (Logos Bible Software, 2010; 2010
3. Eberhard Nestle et al., Novum Testamentum Graece (27. Aufl., rev.; Stuttgart: Deutsche Bibelstiftung, 1993), 304.
4. Eberhard Nestle et al., Novum Testamentum Graece (27. Aufl., rev.; Stuttgart: Deutsche Bibelstiftung, 1993).

But wait, there’s more!

Participant-driven versus Event-driven

Writers will not necessarily place equal value on each parameter. Instead these parameters are better understood as forming a continuum, with the focus on participants at one end and on events at the other. I’m not just talking here about within a given narrative, I mean within the narrative  as a whole. In other words, you could have a narrative that is almost completely focused on participants to the exclusion of events, or vice versa. How, you ask? How is it possible to tell a narrative without events, or to tell a story without really focusing on the participants? It happens a lot more than you might think! The basic difference between these two extremes can be crassly summarized as the difference between a “chick flicks” and your standard issue Chuck Norris or Arnold Schwarzenegger movie.

Participant-Driven———————————————————————- Event-Driven

Participant-driven narrative

I read Jane Austen’s Pride and Prejudice over Christmas, taking a fair amount of ribbing for getting in touch with my feminine side. In any case, it was time well spent based on what I learned. It was the most uneventful book I have ever read, literally! I mean there were virtually no events! Almost the entire book was dialogue between characters, with very little actually reported via narration from Austen. It was not just that she avoided frilly descriptions of scenes; if there was narration to be done, she used the participants to accomplish it. It took me about 2/3 of the way through the book to recognize the pattern that had been subconsciously bugging me. Here’s what I am talking about.

In chapter 49 the Bennets are awaiting word about the whereabouts of Lydia, who had run away with the unscrupulous Mr. Wickham. Instead of narrating the arrival of the expressman from London with a letter bearing news, Austen uses a servant–one of the rare instances in the book where the hired help is anything more than a prop–to convey the arrival of the letter. But even here the servant doesn’t say, “Hey look, the mailman Behold, a rider approacheth! Prithee come hither!” Nope! Instead the housekeeper recounts something that has already happened. The narration is kept to an absolute minimum:

Two days after Mr. Bennet’s return, as Jane and Elizabeth were walking in the shrubbery behind the house,¹ they saw the housekeeper coming towards them, and, concluding that she came to call them to their mother,² went forward to meet her; but instead of the expected summons, when they approached her, she said to Miss Bennet, “I beg your pardon, madam, for interrupting you, but I was in hopes you might have got some good news from town, so I took the liberty of coming to ask.”³

“What do you mean, Hill? We have heard nothing from town.”

“Dear madam,” cried Mrs. Hill, in great astonishment,⁴ “don’t you know there is an express come for master from Mr. Gardiner?⁵ He has been here this half-hour, and master has had a letter.”

Away ran the girls, too eager to get in to have time for speech.⁶

Most narratives have a fair amount of material that is attributed to the narrator, describing a scene or event. Not so in Austen’s Pride and Predjudice. In place of narration, she uses dialogue between the characters to accomplish most of what she needs conveyed. There is narration, but it’s like she avoids it at all costs.

Event-driven Narrative 

So what about the other end of the spectrum, where there is almost total focus on events instead of participants? You have your average action movie like “Rambo,” “Commando,” or may the remake of “The A-Team.” You get to know the characters mainly through what they do instead of what they say. What they say? Not very much. In most cases there is an attempt a relationship or romantic involvement, there is an effort to provide some kind of meaningful linkage between the events that passes for a plot. However, the real point of interest is the events, watching stuff get blown up or the good guys prevailing over the bad. If there is dialogue in the action movie, it is usually poorly delivered and pretty lame. I’m thinking here of the scene in “Rambo” where the Vietnamese girl is asking Rambo if he’d take her back to America with him. About the time the intimacy begins to develop, the character is killed off. There’s a “touching” moment after she is inevitably killed: Rambo ties on the Jade necklace that she has given him around his neck as he is suiting up for battle. Her death is symbolized by the necklace, becoming another motivation for rescuing the POWs and killing bad guys.

So what’s my point? Genre is not monolithic. There are times in narrative, as in the gospels and Acts, where most all of the content is delivered in dialogue rather than with extensive narration, like Jesus conversation with Nicodemus or the woman at the well in John 3 & 4.  But there are plenty of other times in John’s gospel where he interrupts the story to narrate important details. So although Austen may have a tendency toward dialogue and the action-movie writers toward, well, action, we find a mix of both even within a single author’s work as well.

Okay, quiz time: What makes following picture funny?

1. Perhaps they were looking for the Knights who say “Ni”?
2. Note that Austen uses the participants inner musings to describe what might happen (irrealis). This material does not advance the plot, but actually delays learning what actually is the reason for her coming out. It’s a great delay technique to add prominence to what follows.
3. Note that the housekeeper has assumed the Bennet girls know what she knows: an important letter has arrived from London. On the contrary, her statement causes confusion since the girls had no idea anything had transpired while they frolicked in the shrubbery!
4. Note the mid-sentence placement of the quotative frame. It might could have been eliminated, or at the very least placed at the beginning of the clause. Note it also includes a characterization of the speech–crying (out) in great astonishment. The mid-clause placement comes between the unnecessary “vocative” of address and the main proposition, another great use of delay tactics to build suspense just before some significant thing. Delay what? Learning why Hill should believe that the Bennets may have received word from London.
5. Obviously the don’t know, which is precisely the point of all these delay devices being clustered.
6. Jane Austen, Pride and Prejudice (London: C. Scribner’s Sons, 1918), 307.

Basic Genre Distinctions

Robert Longacre used two different parameters as a heuristic for better understanding what separates one genre from another. The two parameters are agent orientation and contingent temporal succession.  He essentially claims that genres can be differentiated based on how they place a higher or lower value on the particular parameter; this differs from the absolute claim that it is completely absent. Agent orientation concerns how large a role participants play in the discourse; they are essentially either primary or not in his view. Contingent temporal succession has to do with the organization of the discourse; it either plays a strong organizational role, or it does not. For more of an introduction see my earlier post on the topic.

(5) Broad categories of genre (from Robert E. Longacre, The Grammar of Discourse, Second ed., Topics in Language and Linguistics (New York: Plenum Press, 1996), Chapter 1.)

One of the most striking things about reading the discourse analysis (DA) literature has been the  treatment of genres like narrative or procedural as though they were monolithic. For instance,  since narrative is agent oriented and values contingent temporal succession, one might expect to find both in equal measure or always present. I know that if I were to question the authors on these matters they would agree that genre distinctions are not as tidy as their general assertions suggest. Therefore my criticisms should be seen as clarifying more than correcting.

What are some examples of narrative breaking from these parameters? Well, it is not uncommon for writers to report events out of temporal sequence for various reasons, such as Mark’s delayed reporting of  John the Baptist’s beheading in Mark 6. We first learn of his death in Herod’s speech in v. 14, but it is only in vv. 17-29 that what actually happened is reported.

Another case where the temporal sequence may be broken is where two parallel plots are progressing simultaneously. Written and spoken discourse preclude the “screen within a screen” option used with video. Discourse demands that one plot be disclosed before switching to the other, and then perhaps back again. We see this kind of switching back and forth in the lead up to Jesus’ passion in Judas betrayal of Jesus. Judas purposes to betray Jesus before the last supper (e.g. Jn13:2), but then he leaves in the midst of it (Jn 13:26-30). Jesus and the remaining disciples then finish the dinner and move to Gethsemane (Jn 18:1). Then comes a “meanwhile, back at the ranch” kind of switch to the Judas plot in v. 2, which is reported in many translations as a parenthetical statement. The parenthesis ends at v. 3 where the two plot lines merge again with the arrival of Judas with the soldiers to arrest Jesus.

These examples illustrate that although temporal sequence is the norm in narrative, there are regularly occurring instances that depart from this norm. This does not invalidate Longacre’s assertions, it merely illustrates that they are not absolute claims. The same holds true for the other parameter of agent orientation. In many instances, the background information that one finds in narrative quite often does not concern participants or characters, but states of affair and other non-agentive things. Consider, for example, the description of the storm in Mk 4:37 just before Jesus calms it; no agents are involved, though some might argue the storm itself is agent-like. Again, these parameters are general rather than absolute. There are many instances where agents are not directly involved, yet narrative nonetheless tends to be agent-oriented.

So while narrative tends to be agent-oriented and to value temporal succession, there will be times when one or the other will be missing. Valuing temporal succession generally translates into a greater focus on events than other genres. After all, events are the things that happen in the close temporal succession.

So to summarize, narratives prototypically focus on agents/participants and events, but even these parameters are not equally balanced. Some (portions of) narratives are more agent-oriented than event-oriented, and vice versa. They may be thought of as endpoints on a continuum. In my next posts I’ll take a look at the effects of focusing more on one than the other. This was my great epiphany, though it might be old news to you.

These two parameters also make up a continuum, in that some discourses will place a higher value on the agent orientation than on the events themselves. In my next post I’ll take a look at the effects of placing more emphasis on agents than events.

Of more immediate interest is the central focus of the post: instantaneous imperfects. This is a class of usage cataloged by Wallace that happen all to be verbs of speaking. In most cases, they end up reintroducing the same speaker within what would otherwise be a single speech rather than switching to a new speaker; this makes them even more of an oddity. For this reason, I think they are better described in the following matter, rather than as though they were normal uses of the imperfect. Here is a link to a PDF excerpt from the Discourse Grammar that provides an alternate description to that provided by Mounce or Wallace. I would have posted it directly, but the formatting  does not convert readily to HTML.

Now let’s take a look at John 16:22, here is the text:

21 ἡ γυνὴ ὅταν τίκτῃ λύπην ἔχει, ὅτι ἦλθεν ἡ ὥρα αὐτῆς· ὅταν δὲ γεννήσῃ τὸ παιδίον, οὐκέτι μνημονεύει τῆς θλίψεως διὰ τὴν χαρὰν ὅτι ἐγεννήθη ἄνθρωπος εἰς τὸν κόσμον. 22 καὶ ὑμεῖς οὖν νῦν μὲν λύπην ἔχετε· πάλιν δὲ ὄψομαι ὑμᾶς, καὶ χαρήσεται ὑμῶν ἡ καρδία, καὶ τὴν χαρὰν ὑμῶν οὐδεὶς αἴρει ἀφʼ ὑμῶν.¹

Although conjunctions occur within a single clause, they often cannot be fully understood without referencing the larger context. Let’s begin with the role of καί. Since there is another coordinating conjunction present (the postpositive οὖν) καί is doing something other than linking two clauses or phrases. It always links two things of equal status, but they may not be right next to each other. This latter function is generally called the adverbial function. Note that one Greek conjunction καί accomplishes the same tasks as three separate English words: and/but (coordinating conjunction, adding contiguous elements), also (additive adverb, adding non-contiguous elements), even (intensive adverb, adding something to itself). If we look at the preceding context of v. 21, we see there is a parallel thought. Whereas the woman experiences sorrow/pain as the hour of childbirth arrives, so too the disciples will have a similar experience. The connection between v. 22 and v. 21 would have been there with or without καί; adding it to the context constrains the reader to look for some relevant connect. The presence of the other coordinating conjunction οὖν tells you to look for a non-contiguous yet related element. One down, two more to go.

The second connective we come across is οὖν. In John’s gospel, οὖν is used to signal the next development in the dialogue or story. A development is the next stage or step; it may or may not be closely related. The use of οὖν signals +development, +close connection. The use of δέ signals +development, -close connection. So John has signaled that v. 22 is the next development, but closely related to what precedes i.e. the illustration in v. 21. Thus while καί connects “woman” to “disciples”, the οὖν is coordinating the two clauses. Two down.

What about μέν? It’s sole purpose in life is to create the expectation that some related element will follow. It is the first shoe to drop so to speak, setting the stage for the typically more salient second shoe. In most cases, μέν works in conjunction (pun unavoidable) with δέ as a paired set. We might paraphrase v. 22 using “while” or “although” to create the same kind of expectation in English. There is also the temporal adverb νύν that creates a different kind of connection. Why do we need to think about now unless there will be a temporal transition of some kind? John tends to speak slowly and use small words; here we have a doubling up of signals to keep our eyes peeled for a related element. Sure enough, it comes later in v. 22. The key is that they will see Jesus again, another connection but backpointing this time. The disciples have sorrow because Jesus has announced he is leaving them. Seeing him again at some later date will be a game-changer. Their λύπη will be changed to something else: χαρά.

Based on the repetition of key concepts in these verse, there were natural connections present between Jesus’ illustration and his encouragement to his disciples. The connectives and adverbs simply strengthen the connections that were already present. Each one brings a unique constraint to the context; each performs the same basic function, even in the presence of connectives. This was a pretty straightforward example, but I bring in some others where commentators gravitate toward assigning some special function to the cluster rather than looking at the contribution of each component.

So when you see a conjunction–particularly one that is hanging with his homies in a cluster–be sure to ask each one present, “Conjunction-junction, what’s your function?”

If you want to learn more about this approach to connectives, see chapter 2 of A Discourse Grammar of the GNT. This chapter is part of a sample available here.

1. Michael W. Holmes, The Greek New Testament: SBL Edition ( Logos Bible Software, 2010), Jn 16:21–22.

Find an expert and carry their bags

If you want to excel at something, then find the most qualified person you possibly can to mentor you. This is the strategy I learned in Boy Scouts, where each merit badge had a counselor that was knowledgeable in the topic area. He’d help you get grounded in the topic and guide your learning. I have used the same strategy in learning about building, real estate, church health, and linguistics. Once I found someone, I took every opportunity to hang out with them that I could find. I refer to this as “carrying their bags.” This required me to research about their field and approach before contacting them so that I did not waste their time with silly questions. If you want to engage someone of this caliber, then do your homework before talking to them. Writing and saying you are a big fan will not impress most of them.

As you consider doctoral programs, you need to think long and hard about your adviser. Do they have the necessary background to supervise this topic? This may sound like a silly question, but it seems to be ignored in the case of interdisciplinary studies. If someone had only ever read about the synoptic problem but never published on the topic, would you want them advising you?  Of course not! So then why ask to pursue a linguistic topic under someone who has not published in the area? In the case of interdisciplinary study, some seem willing to take on topics which they do not have the background to supervise. In a perfect world, your adviser will be your expert.

In my own experience, it is difficult to find one person who has everything. If they have the necessary background, they may not be at an institution that grants doctoral degrees. They may also not have all of the components you need. So besides your adviser, is there some specialist from the secondary field with whom you could (in)formally interact? Their primary role would be guiding the development of your theoretical framework. If it is not properly formulated, the conclusions that come out of your model will show it. Actually, let me rephrase that. Your conclusions may impress your primary field, but they will be rejected or ignored by the secondary one. “Two are better than one…” Keep this in mind.

External readers

What are the qualifications of your external reader? Is she an expert in the secondary field, or did she just play the part of one on TV? While students have no control over such matters, institutions need to think long and hard about their selection. Although we want our students to complete their program, it should not be at the cost of quality. Just because a plausible argument is presented does not mean it is valid or correct. It is crucial to involve experts from the secondary field in order to guard against this. Only utilizing NT scholars to review NT linguistic topics has led to an inbreeding of sorts. While it may look like consensus, I consider it to be a sign of something else. The fact that such ideas are not being adopted outside our discipline should be a harbinger.

If you want to do quality work it behooves you to gather high quality criticism as early as you can in the process. This may sound masochistic, but it really is in your best interest. Ideally, this would begin at the proposal and lit review stage, not at the external review. Get feedback on how you should shape your question, on your theoretical framework, on your preliminary conclusions. If you only pitch your ideas to non-specialists, you simply defer the criticism. Get outside NT studies and see if your ideas hold water. I can assure you that if some of the NT proposals about word order or verbal aspect were presented in the SBL “Linguistics and Biblical Hebrew” section or at the Linguistic Society of America, the results would not be pretty.

Engage counterarguments

This should go without saying, but I often see strawman-(dis)engagement of opposing ideas in NT linguistic studies. “Their definitions are too imprecise, their model is not empirical enough, therefore it is to be rejected.” Unfortunately if there were significant linguistic arguments against a position, few in NT studies would know about it. But even if there are major disagreements within NT studies, then get to know them well enough to understand why they exist. If an idea has gathered a significant following, then it is probably addressing some shortcoming of the competing theory. You see this in the work of N.T. Wright, correcting inadequacies of traditional views. However, part of the process of correction is often overstating the problem in order to bring about change. Defenders of the status quo will also tend to overstate their point in response. There is a reasonable chance that BOTH are wrong at some point. Mike Bird has a great paper exemplifying such issues in the Wright/Piper debate entitled “What is there Between Minneapolis and Durham?” This is not just a “why can’t we all just get along” appeal, but a respectful critique of both sides. I wish this happened more often.

Is the disagreement about presuppositions? About definitions? Understanding your detractors will strengthen your model, so long as you are willing to learn from them. If people like Dan Wallace and Peter Gentry hold to differing views on a language issue, then try to figure out why. Both scholars know their stuff; neither is stupid or naive.   They each see some key ingredient that leads them to disagree; figure out what it is and you’ll be in a better position to advance the discussion. Doing so may also more support for your proposal if the two camps see that you have meaningfully and respectfully engaged their arguments. Engage counterarguments, do not ignore them.

Primary versus secondary

If you are interested in interdisciplinary studies involving linguistics and biblical studies, I am coming to believe that it would be better to swap primary and secondary disciplines: major in linguistics and simply apply it within biblical studies. The complexities of the verbal system (in both Greek and Hebrew) demand a high caliber background; so too understanding information structure or connectives. The people whose work I most respect are linguists first and exegetes second.

Think long and hard about what you want to do when you grow up. If you want to specialize in a single book, then major in NT studies and minor in linguistics or whatever. If you want to contribute to linguistic discussions, then major in linguistics. There are a few programs where you can have it both ways; Nick Bailey’s experience at the Vrije Universiteit in Amsterdam is a great example. The Classics program there has world-class Greek specialists doing some mighty fine linguistic work. University of Edinburgh also has a great reputation for interdepartmental projects between theology and linguistic sciences, exemplified in the work of Margaret Sim. On the other hand, two of my heroes–Stephen Levinsohn and Randall Buth–both pursued PhDs in linguistics proper, at University of Reading and UCLA respectively.

Moving forward

Interdisciplinary studies hold great potential only if they are properly pursued. I wish more NT programs said “No” to topics they were not qualified to supervise. This would greatly reduce the publication of findings that have not been adequately vetted. They may sound impressive to non-specialists, but specialists would recognize the weaknesses.

Students themselves can also affect the outcome by going outside NT studies to gain the background they need. This may mean an MA in linguistics or doctoral focus in linguistics. The latter is a risky move if you intend to teach in NT, one that several scholars tried to talk me out of. You see, I have a DLitt, not a PhD. I am neither OT nor NT, but specialized in biblical languages. These scholars told me that if I chose a non-traditional path like this, then schools would not consider hiring me because my degree was not from within the discipline. To be honest, they were right! Of all the teaching positions for which I have applied over the years, only once did I make it beyond the initial cut. The advice of these scholars has been prophetic.

So why risk pursuing a linguistically-focused degree? Because I knew that a traditional NT doctoral program would not have equipped me for my present research program. The projects I am pursuing are precisely the ones I set out to do ten years ago. In 2000, I decided I wanted to do the kind of research that I saw Levinsohn, Buth and others pursuing. The key distinction was to do it in a more approachable manner and focused on exegesis.

One of my career goals is to see students not have to take the same kind of risk I did. At some point it will be possible to help supervise doctoral students. In the meantime, I have a number of research projects that are best pursued in my present position as scholar-in-residence. Besides that, I am reluctant to give up my cool job title! But seriously, I think it will be another 5-10 years before a research-based theological institution would consider hiring a linguistic specialist. NT linguistics has over-promised and under-delivered long enough that schools are legitimately skeptical about its potential. I am not optimistic about how long it will take to change this perception.

I fully support interdisciplinary research, but only so long as it is properly pursued. It is high time we in NT studies do a better job of policing interdisciplinary study, particularly when it comes to linguistics. Being more honest about institutional and personal limitations would be a great place to start. If you need a specialist on your faculty then hire one, even if they are not NT scholars. If you insist on allowing such work, then tap into the resources of SIL. There you will find gifted, well-qualified linguists who are passionate and knowledgeable about biblical languages. You might also be surprised at what a linguist could contribute in areas like hermeneutics or homiletics. In the meantime, students must take responsibility for shaping their education based on their research goals. If that means a non-traditional path or involving an informal adviser to help guide your program, then do it!

Interdisciplinary studies are not a silver bullet, and cannot just be adopted in name-only. Like we used to say at scout camp: Be for-real, don’t Beef-a-Roni. If you want to specialize in linguistics, then do everyone a favor and really specialize. If you don’t really want to specialize there, that’s fine too. Just pick a topic that does not require it.

[Update: an image has been changed from the original post at the request of a reader.]