DISCOURSE ANALYSIS II: LEVINSOHN AND RUNGE
Discourse without prominence would be like pointing to a piece of black cardboard and insisting that it was a picture of black camels crossing black sands at midnight.
— Robert E. Longacre
In the previous chapter, the Summer Institute of Linguistics (SIL) was introduced as one of the four major schools of discourse analysis. Perhaps the most important SIL linguist for the study of Greek is Stephen Levinsohn. While Levinsohn is not “Hallidayan” in the sense that his approach is an application of the work of Prague School linguists Jan Firbas and E. Beneš,1 rather than of Halliday himself, there is some overlap in certain areas, so that we see that the two approaches are not mutually exclusive. An important distinction of Levinsohn’s work, however, is that it is explicitly applied to the Greek of the New Testament, though he does not claim to do so comprehensively.2 This chapter will offer an outline and evaluation of Levinsohn’s approach to discourse analysis as recorded in his volume Discourse Features of New Testament Greek,3 as well as that of Levinsohn’s most significant disciple — Steven Runge.
According to Levinsohn, the value of discourse analysis is that it “draws its explanations, not from within the sentence or word (i.e., the factors involved are not syntactic or morphological), but extrasententially (from the linguistic and wider context).”4 In view of Porter’s criticism that he does not go far beyond the level of the sentence (see §7.2.1), Levinsohn acknowledges that this is necessarily true for some units of study, but not for others that involve extrasentential factors.5
Levinsohn begins by describing his approach to discourse analysis as eclectic, “making use of the insights of different linguists and different linguistic theories to the extent that I feel they are helpful.”6 Levinsohn is even prepared to admit that he will take a “seed idea” from another linguist and “develop it in another direction.” He finds, for example, the ideas of Givón valuable but often arrives at opposite conclusions in their application.7
Levinsohn’s linguistic approach is functional in the sense that he is interested in what functions linguistic structures serve, rather just being interested in the structure of a language for its own sake.8 One implication of a functional approach is the principle that choice implies meaning. When an author has a choice of expressing something in more than one way, the choice of one expression over another is significant. This is not simply a question of style; there is a linguistic reason for the choices that are made.9
While Levinsohn does not use the term idiolect (at least not in the book of interest here),10 he acknowledges the significance of the concept nonetheless. He assumes that an individual author will use discourse features in a consistent way and that there is potential for variation of such usage between authors. Levinsohn also admits that this means his own presentation contains a weakness, namely, that he has not studied every biblical author in depth. The elucidation of certain discourse features only applies with reference to the biblical author under discussion.11
According to the concept of markedness, “when a certain marker is present, the feature implied by the marker is present.”12 The reverse, however, is not necessarily true: the absence of the marker does not mean that the feature is absent. Levinsohn illustrates the importance of markedness with respect to the Greek article. One of the uses of the article is that its substantive refers to a particular item, but its absence does not indicate non-particularity; the referent of the substantive might well be particular and known (as is the case with proper nouns).13
8.2.5 Semantic Meaning and Pragmatic Effects
Levinsohn regards the distinction between semantics and pragmatics as essential to his approach. A particular construction will have an inherent semantic meaning but differing pragmatic uses according to context.14
Levinsohn’s volume contains six parts: constituent order, sentence conjunctions, patterns of reference, backgrounding and highlighting, reported conversation, and recognizing subunits and boundaries. The following is a brief sketch of Levinsohn’s presentation of these topics.
The term “constituent order” is used rather than “word order,” acknowledging that “the elements that are ordered are often phrases and clauses, not single words.”15
8.3.1 Coherence and Discontinuities
In the first chapter in this section, Levinsohn discusses the ways in which sections, subsections, and paragraphs are demarcated through four dimensions of continuity and discontinuity, operating with the assumption that New Testament books are coherent.16 In narrative, these dimensions are time, place, action, and participants; for other discourse genres, the dimensions are generalized as situation, reference, and action. Paragraph and section breaks become fairly obvious through discerning significant shifts in one or more of these four dimensions in the text.17
The second chapter discusses a device called a “point of departure” (or topicalization), which designates an element placed at the beginning of a clause with the double function of providing a starting point for communication and anchoring the following clause(s) to something already in the context.18 Points of departure signal discontinuities of situation, reference, and sometimes of action.19 In this chapter, Levinsohn also endorses the claim that Greek is a verb-subject-object (VSO) language.20 While acknowledging that this cannot be proved without thorough statistical confirmation, he draws on the implications of this claim to say that variations of the VSO constituent ordering creates an emphasis on the fronted constituent.21
Chapters 3 and 4 in Levinsohn deal with the order of constituents in the remainder of the sentence, whereas the previous chapter was concerned with fronted items. Chapter 3 discusses four default ordering principles: the placement of pronominal constituents immediately following the verb, the placement of core constituents before peripheral ones, the placement of an overtly expressed propositional topic before the comment about that topic, and the placement toward the end of a clause or sentence of the most important or focal constituent of the comment.22 Chapter 4 includes further discussion of constituent order, dealing with issues such as negated sentences, information questions, focus switching, and discontinuous constituents.23
Levinsohn’s second major part addresses the most common nonsubordinating conjunctions in the New Testament. While traditional approaches to conjunctions identify their various “senses,” Levinsohn follows the linguistic principle of semantic constraints. Each conjunction expresses a single constraint that it places on the way a sentence is to be understood with reference to its context.24 The differing senses identified by traditional approaches are produced by the one constraint applied in different contexts. In other words, the constraint is the semantic value of the conjunction, while its various senses are pragmatic functions of the semantic value.
Levinsohn’s fifth chapter addresses these common conjunctions in narrative, discerning their semantic content rather than focusing on the gloss translations that are so common in grammars. In the Gospels and Acts, καί is the default form for linking sentences (in John, asyndeton is default). The most common conjunction when καί is not used is δέ. The distinct semantic element that δέ expresses is that it signals a new step or development in the author’s story or argument.25
8.4.2 Τότε, Nonconjunctive Καί, and Τέ Solitarium
Τότε is an adverb that may function as a conjunction, especially in Matthew and Acts. When it is not functioning as a conjunction, it retains its temporal expression (“then”), but when in conjunctive use, it functions as a cohesive device “indicating continuity of time and of other factors between the subsections.”26 When τότε functions adverbially rather than as a conjunction, it is used at subsections of an ongoing story and to highlight a conclusion — especially in Matthew and Acts.27
Levinsohn addresses nonconjunctive καί as a means of “indicating that a sentence or proposition is to be related to its context by addition.”28 Rather than functioning as a conjunction to link sentences (see §8.4.1), nonconjunctive καί indicates “parallelism between the proposition concerned and an earlier one,” and also “backwards confirmation,” which refers to a proposition “added to confirm an earlier one.”29
Τέ Solitarium refers to the use of τέ by itself without any corresponding καί or τέ. Almost all of such occurrences are found in Acts, and they add “distinct propositions that are characterized by sameness, in the sense that they refer to different aspects of the same event, the same occasion, or the same pragmatic unit.”30
The final chapter in this section addresses thematic development in non-narrative text. It “concentrates on the four most common ways in which sentences in non-narrative text are formally related: by means of δέ; simple juxtaposition, i.e., asyndeton; conjunctive καί; and οὖν.”31
In non-narrative text (as with narrative text), δέ “is used to mark new developments, in the sense that the information it introduces builds on what has gone before and makes a distinct contribution to the argument.”32
Asyndeton is often used “when the relation of the following material to the context is not logical or chronological,” yielding two opposite functions in non-narrative text: to indicate a close connection between the information concerned, and to indicate that there is no direct connection between the information concerned. In other words, asyndeton can indicate that certain information belongs together in the same unit, or that information belongs to different units.33
In non-narrative text, conjunctive καί “constrains the material it introduces to be processed as being added to and associated with previous material,” but this material does not represent a new development in the context.34
As with its use in John’s gospel, οὖν is used inferentially and as a resumptive in non-narrative material.35 The inferential use introduces inferences drawn from a previous statement. The resumptive use of οὖν follows digressional material in order to return to the point of a preceding assertion.36
Levinsohn’s third major section deals with some elements “that determine how an entity is referred to.”37 Reference may be created through use of a noun, an article plus a noun, a pronoun, or an articular pronoun, or through implicit information conveyed by a verb (the person and number of the subject).38 Of particular interest is the identification of default patterns of reference, so that marked forms of reference may also be recognized, informing the reader “that a new section is beginning or that a particular event or speech is being highlighted.”39
The first chapter in this section addresses participant reference. The most fundamental distinction regarding participant reference is that between major and minor participants. There is a variety of ways in which participants may be introduced in texts, including presentational articulations (e.g., “Now a certain man in Caesarea named Cornelius”), through a verb of arrival (e.g., “A Samaritan woman came to draw water”), through a verb of perception (e.g., “and he saw a tax collector named Levi”), or through association with another participant.40
The chapter continues with a discussion of further references to activated participants. That is, characters who have already been introduced in a story may reappear through a variety of means.41
This chapter also includes the issue of references to VIPs. A VIP is a Very Important Participant, who is distinguished from all other participants. The VIP can be identified on the global level, meaning for a whole book, or on a local level, meaning a section of a book or a single episode. Levinsohn states that each gospel treats Jesus as the global VIP, whereas Acts focuses on different apostolic leaders as VIPs at different sections.42 Interestingly, the most common way for the Gospels to indicate that Jesus is the global VIP is to make no overt reference to him at the beginning of new narrative units, thus establishing him as the “given” subject of the entire discourse. According to Levinsohn, this also means that when Jesus is referenced overtly, it is a marked reference. This may be used to introduce a major break in the story or a key speech or action.43
The second chapter in this section addresses the article with substantives. Levinsohn makes the following claim: “if the referent of an anarthrous noun phrase is known and particular . . . this gives it prominence. It is marked as prominent because it is of particular importance.”44 The default rule is that a participant is introduced for the first time by name without the article. Subsequent references to him or her within the same episode use the article.45 Instances in which subsequent references do not use the article are therefore prominent.46
When a character is reintroduced in a story, the norm is for that character to be “reactivated” with an anarthrous reference, with exception to the VIP. Reactivations of the global VIP are articular, while other participants’ reactivations are anarthrous.47
8.6 Backgrounding and Highlighting Devices
Levinsohn begins this section by differentiating between the distinction of background and foreground material and that of backgrounded and foregrounded material. That is, the latter distinction is concerned with the relative relationship between elements — some elements will be backgrounded with reference to their context, and others foregrounded with reference to their context. This distinction is his concern, rather than trying to establish criteria for the fixed categories of background and foreground.48
Levinsohn follows Callow’s relating of foreground to thematic prominence: “this is what I’m talking about.” It carries the discourse forward, contributing to the progression of the narrative or argument.49 He adds that features that may be used for highlighting in one genre may be used for backgrounding in another, and vice-versa. Thus, highlighting and backgrounding features are not universal; they are genre-specific.50
In the first chapter in this section, Levinsohn discusses three devices that indicate that a sentence has been backgrounded: prospective μέν, verbal aspect, and the use of ἐγένετο. (1) The prospective μέν is its use that anticipates a corresponding sentence containing δέ. Levinsohn claims that this function downgrades the importance of the sentence containing μέν, in comparison with what is introduced with δέ.51
(2) With respect to verbal aspect, Levinsohn claims that it is natural for imperfective aspect to convey information of less importance than that conveyed by perfective aspect.52 This pattern, however, does not always hold. For example, imperfects that do not convey an incomplete action (its default meaning) are marked, since it would be normal to expect an aorist to be used to convey completed events.53 All of this pertains to narrative in particular.
(3) Levinsohn also addresses the combination of ἐγένετο and a temporal expression, such as ἐγένετο δὲ ἐν ταῖς ἡμέραις ἐκείναις in Acts 9:37a. This device, found in the LXX, is used by Luke “to background information with respect to the following foreground events.”54
The second chapter in this section is concerned with backgrounding within sentences, rather than the backgrounding of entire sentences. The features discussed here are anarthrous participial clauses and a particular type of relative clause. Anarthrous participial clauses that precede the main clause to which they are dependent (Levinsohn calls these “prenuclear” clauses) are always backgrounded with respect to their nuclear clause,55 whereas “no such claim can be made about postnuclear participial clauses.”56
Regarding relative clauses, Levinsohn focuses on continuative relative clauses, which “describe an event that involves the referent of the relative pronoun and occurs subsequent to the previous event or situation in which the referent featured.”57 Normally, the information preceding the relative pronoun is backgrounded with respect to what follows. An example of this is seen in Acts 28:23c, which occurs subsequent to 23b: (b) ἦλθον πρὸς αὐτὸν εἰς τὴν ξενίαν πλείονες (c) οἷς ἐξετίθετο διαμαρτυρόμενος τὴν βασιλείαν τοῦ θεοῦ.
The third chapter of this section deals with the highlighting function of the historical present. Levinsohn states that the “presence of a historical present most often has the effect of highlighting what follows.”58 The historical present highlights the event(s) that follow, especially in Mark and John.59 It indicates prominence and has different patterns of usage in Matthew, Luke-Acts, and John.60
8.7 The Reporting of Conversation
In this section, Levinsohn addresses the fact that reported conversations are not structured like ordinary narrative events.61 The first chapter in this section explains the default strategy for reporting conversations, considering the way in which the speech is introduced in order to indicate whether the speech is an end in itself or is an intermediate step to the actions that result from the conversation.62
The second chapter offers more on reported conversations in the Synoptic Gospels and Acts, addressing such topics as asyndeton in reported conversations in Matthew’s gospel63 and speech verbs in the historical present in Matthew.64
The third chapter in this section discusses reported conversations in John’s gospel, primarily dealing with historical presents in speech orienters in John.65
The fourth chapter details three ways of reporting speech, which are the indirect reporting of speech,66 direct reporting of speech with ὅτι recitativum,67 and ὅτι following λέγω σοι/ὑμῖν.68
The final section of Levinsohn’s book concerns “the criteria that enable the reader to recognize boundaries between paragraphs and larger semantic or pragmatic units, such as ‘sections’ of a book.”69 The chief characterization of a major unit of text is the coherence of a single theme within it. Levinsohn therefore acknowledges that the surface features explored earlier in the book can only provide supporting evidence for boundaries, which must be established on other grounds. The problem here “is that supporting evidence can be cited for conflicting boundaries, so a need exists to discern which evidence is valid.”70 He then suggests that “the presence or absence of a point of departure has a major part to play in determining the validity of potential evidence.”71
A point of departure signals some sort of discontinuity, which “indicates the primary basis for relating what follows to the context.”72 In this way, a point of departure reveals which supporting evidence is valid. Levinsohn then spends the rest of the section detailing items that constitute supporting evidence for boundaries within text. These are briefly summarized below.
1. Conjunctions and Asyndeton. Levinsohn states that δέ and τότε and asyndeton often occur at paragraph and section boundaries, whereas καί and τέ are less frequently found at such boundaries.73
2. Spatiotemporal Changes. When the text indicates a temporal change at the beginning of a sentence, this is supporting evidence for a boundary. The temporal change is thus a point of departure. Changes of location may coincide with a boundary, but this is not determinative.74
3. Summary Statements. A summary statement unites the preceding material as a block, and in so doing terminates or begins units of text.75
4. Chiastic Structures. A chiasm is obviously a self-contained unit and thus creates its own boundaries.76
5. Inclusio Structures. Like a chiasm, an inclusio is a self-contained unit that creates its own boundaries.77
6. Rhetorical Questions. Sometimes rhetorical questions signal a new subject or some new aspect of same subject and can collaborate with a developmental conjunction to indicate a boundary.78
7. Participant Reference by Means of a Noun Phrase. Sometimes a redundant noun phrase reference to a participant is used to mark the beginning of a unit.79
8. Vocatives. Likewise, use of a vocative can provide supporting evidence for a boundary, though they do not automatically indicate a boundary.80
9. Changes of Cast and Role. A change of cast in narrative that affects the global VIP supports a section break. Additionally, a significant change of role of the global VIP constitutes grounds for supporting a paragraph or section break.81
10. Changes of Verb Tense-aspect, Mood, and/or Person. Changes in aspect and mood may support the discernment of a boundary. A change of person parallels a change of cast. Nonspeech historical presents commonly occur at the beginnings of subsections.82
11. Back-reference. Reference to a preceding paragraph often occurs at the beginning of new paragraph; hence such back-reference provides an evidence for a boundary.83
Levinsohn has made a significant contribution to Greek discourse analysis, and one reason that I have provided a thoroughgoing summary of his Discourse Features is the hope that more students and teachers of Greek will become familiar with it. Clearly, there is much useful material that supplements traditional grammars, and it offers the opportunity to think more about how parts of the sentence relate to one another and how wider units of text are demarcated. Unlike Halliday, of course, Levinsohn has studied Greek parts of speech in depth and has married such investigation with several principles of the wider linguistic world. The volume is worthy of serious study by all Greek students.
There are, however, some areas of concern. Though Porter’s critique was based on Levisohn’s first edition, published in 1992, some of his comments remain potent. His chief criticism of Levinsohn is that it remains fairly well preoccupied with the level of the sentence,84 and it is true that much of Levinsohn’s material concerns sentential analysis, though there is much that moves beyond the sentence too. The sections on constituent order, sentence conjunctions, and backgrounding and foregrounding (at least in part) concern intra-sentential issues. Yet, these discussions hardly represent a “return to sentence grammar”; they are more linguistically robust than traditional grammar discussions about the Greek sentence and offer helpful insights not normally gleaned through traditional approaches. Moreover, there is much in Levinsohn that extends beyond the level of the sentence. The sections on patterns of reference, reported speech, and especially boundary features all explore wider units of text.
Porter adds that “there is a tendency to focus upon idiolect, or even the language of a single book” in Levinsohn’s work.85 This is certainly a fair observation, though it is not necessarily a negative aspect of Discourse Features. On the one hand, it might be a little frustrating to the reader to have so many approaches outlined, moving from discussion of a discourse feature in Matthew’s gospel only to learn about a different pattern at work in John. On the other hand, it is good to be aware of idiolect and to avoid the mistake of illegitimately attributing a pattern found in one book to all other books. Then again, I would like to see the model move to a deeper analysis of semantics that is able to account for the pragmatic functions of discourse features across idiolects and genres. For example, in discussion of the historical present, a semantic analysis that explains all historical presents, and yet can also cope with the differing ways in which Mark and John employ the phenomenon, would be ideal.
My own concerns may be put alongside Porter’s. Levinsohn self-consciously adopts an eclectic linguistic approach rather than adhering to an established linguistic school (see §8.2.1). Some linguists would object to this, while there is also a case to find it an acceptable approach. After all, every linguistic school builds on the work of others to some extent, usually adopting insights from a variety of sources. In that sense, they are all eclectic to some extent — at least initially. Certainly, Levinsohn is well schooled in linguistics and he knows what he’s doing.86 Perhaps that fact alone should alleviate the concern, but a chief advantage of adopting a particular linguistic approach is that linguistic schools are rigorously tested and pushed for consistency. An ad hoc linguistic approach will not have had the same level of scrutiny and may fall short at the theoretical level. This is not to say that Levinsohn does fall short here, but it means that we need to stand back from the work and reflect on whether the whole system hangs together, or whether there are subtle linguistic contradictions or lapses in theory.
On this topic, I will offer one caution at the level of linguistic theory. Levinsohn will oftentimes make a claim about Greek by way of reference to similar features in other languages; his work as a Bible translator leads him to stress the importance of comparing the discourse features of Greek with those of other languages, using the methods of typology (see §1.4.9).87 While there is no doubt much to be learned about Greek through typological comparison with other languages, this risks introducing similar problems that came through the methods of nineteenth-century comparative philology. In my opinion, the comparative approach has caused much trouble for Greek linguistics over the past century, since Greek was traditionally associated closely to Latin. Deponency is the obvious example (see §4.2.4), and the traditional understanding of the Greek perfect may be another. These are highly controversial and difficult issues, ones that arguably arose because of the comparative approach. Levinsohn’s determinations about Greek based on what happens in other languages therefore gives me pause.
At the practical level, Discourse Features suffers from a presentation that is difficult to navigate. It is hard to find definitions (apart from the glossary), and it is probably a little bewildering to the student. Nevertheless, it remains a valuable resource that is worth the effort. We turn now to consider Levinsohn’s most influential student, Steven Runge.
In the short time in which Runge’s Discourse Grammar of the New Testament has been available,88 it has done more to bring discourse analysis to the attention of the wider New Testament studies world than any other work. A few distinctives set this work apart. The most obvious is its accessible nature, making Discourse Grammar the best entry-level work for Greek discourse analysis. It is popular among students and is easily teachable. Moreover, Discourse Grammar is part of a suite of resources that Runge and his team at Logos have developed, including The Lexham Discourse Greek New Testament,89 which is an application of Runge’s discourse-analytical approach to the entire New Testament.
The significance of the Discourse Greek New Testament is twofold. First, if the exegete is unsure as to how to apply Runge’s approach to text, he or she can simply look up the New Testament text in question and discover how Runge himself sees it applied. Second, the preparation required to produce the Discourse Greek New Testament means that Runge has examined the entire New Testament in preparation for his Discourse Grammar. It is not ad hoc but comprehensive; it is not without application, but it has already been applied in full to the New Testament.
The book is arranged into four major sections: foundations, forward-pointing devices, information structuring devices, and thematic highlighting devices. We will survey the key points throughout.
In his introductory chapter, Runge adopts a function-based and cross-linguistic approach,90 presupposing three core principles: choice implies meaning; semantic meaning should be differentiated from pragmatic effect; and default patterns of usage should be distinguished from marked ones.91 The reader should be familiar now with the first two of these principles; the third may require some explanation. As Runge explains, “Markedness is fundamentally the study of ‘markers,’ those things that signal the presence of some quality or linguistic feature.”92 In sets of oppositions, one member of the set may be marked for a particular linguistic feature, while the other member may not be. The latter might be able to express the linguistic feature in question in certain contexts, but the former will always express it, since it is marked for that feature. This distinction leads to another:
To summarize, markedness theory presupposes that one member of a set is the most basic or simple member, called the “default” member. All of the other members signal or “mark” the presence of some unique quality, one that would not have been marked if the default option were used. The marked options are described based on how they uniquely differ both from the default and from one another.93
Runge also presupposes the importance of prominence and contrast. A prominent feature is one that attracts extra attention within its context — anything that stands out is prominent.94 One of the effects of prominence is to create contrast, since the prominent feature will stand out from others in the same context.95
Runge’s second chapter addresses connectives, a term to be preferred over conjunctions, since other (non-conjunction) parts of speech, such as adverbs, are also able to function to connect clauses. “Connectives” thus includes conjunctions and non-conjunctions that operate in a connective fashion.96 Runge focuses on the function of each connective in “constraining” the discourse, rather than their English gloss translations.97 He deals with asyndeton, καί, δέ, narrative τότε, οὖν, δία τοῦτο, γάρ, μέν, and ἀλλά.
8.11.1.1 Asyndeton. Asyndeton is “the default means of connecting clauses in the Epistles and in speeches reported within narrative.”98 It is used at the beginning of a new thought or where the relation between clauses is clear.99
8.11.1.2 Καί. This is a coordinating conjunction used to connect words, phrases, clauses, or paragraphs,100 linking items of equal status.101
8.11.1.3 Δέ. This is a coordinating conjunction with the constraint of signaling a new development.102 “The use of δέ represents the writer’s choice to explicitly signal that what follows is a new, distinct development in the story or argument, based on how the writer conceived of it.”103
8.11.1.4 Narrative Τότε. In narrative, the temporal adverb τότε can be used as a connective when no conjunction is present. It is marked for development, like δέ, but has the additional constraint of temporal sequence.104
8.11.1.5 Οὖν. This is another development marker, but adds “the constraint of close continuity with what precedes.”105 This is why the English gloss “therefore” is often appropriately used to the translate οὖν: it is closely connected to a preceding statement, but also marks out a new development in thought or event.
8.11.1.6 Δία τοῦτο. This phrase is another example of another non-conjunction connective. It indicates a causal relation with the preceding discourse.106 The function of δία τοῦτο overlaps with that of οὖν, but its constraint is narrower since it is marked for causality, which is not the case for οὖν.
8.11.1.7 Γάρ. The central constraint of the conjunction γάρ is its explanatory function, as it adds “background” information that strengthens or supports preceding material.107
8.11.1.8 Μέν. This connective correlates its own clause with an element that follows, which usually employs δέ. Unlike the connectives listed above, μέν is forward pointing, creating an expectation of something to follow.108
8.11.1.9 Ἀλλά. This conjunction is used to sharpen the contrast between two clauses. It is not marked for continuity or development, but is marked for the constraint of “correction.”109
8.11.1.10 Function of Greek Connectives. Runge concludes the chapter on connectives with a useful table of their functions, which is worth reproducing here.110
| Continuity | Development | Correlation | Forward-pointing | Semantic constraint | |
| Ø | - | - | - | - | - |
| καί | + | - | + | - | - |
| δέ | - | + | - | - | - |
| τότε | - | + | - | - | Temporal |
οὖν |
+ | + | - | - | - |
| διὰ τοῦτο | + | + | - | - | Causal |
| γάρ | + | - | - | - | Support |
| μέν | + | - | + | + | Expectation |
| ἀλλά | - | - | + | - | Correction |
The second major part of Runge’s Discourse Grammar addresses forward-pointing devices, which are “used to attract attention to something significant in the discourse.”111 In the book’s third chapter, Runge develops the notions of a forward-pointing reference and its target. In Greek, there are three ways of creating such a forward-pointing reference: through forward-pointing interrogatives, forward-pointing demonstratives, and substitutional forward-pointing adverbs.112 Since such forms normally point backward, their use to point forward receives additional prominence.113
In chapter 4, Runge discusses point/counterpoint sets, which refers to “clauses or clause elements that have been related to one another through one or more grammatical means.”114 These means are as follows: the use of μέν to create anticipation of something to follow; the use of an interrogative or negated clause restricted by εἰ μή or πλήν; the use of ἀλλά to correct something in the previous context.115 These devices constrain the relationship between two elements, such as two clauses or phrases.
Chapter 5 deals with metacomments, which occur when a speaker stops “saying what they are saying in order to comment on what is going to be said.”116 An English example would be “I want you to know that. . . .” While metacomments do not come in a fixed form, there are some common devices associated with them, such as “attention-getters” like ἀμήν, and redundant forms of address, like ἀδελφοί, and redundant vocatives and nominatives of address.117
In chapter 6, Runge addresses the historical present. Key to his understanding is the fact that the use of the Greek present to refer to a past event in narrative creates prominence since it goes against the expected tense (past) and the expected aspect (perfective), since it is a present referring verb with imperfective aspect.118 Historical presents are used “to make sure that the reader does not miss changes or transitions in the discourse.”119 Runge clarifies that the historical present does not create a discourse boundary, but can be used to draw attention to one.120 The other significant function of the historical present is to highlight a significant speech or event that follows, so that the action of the verb is not itself prominent, but what follows it.121
Chapter 7 is concerned with redundant quotative frames. A quotative frame signals a transition from narrative proper to speech within the narrative. It is redundant when more than one verb is used to introduce the same speech-act, or when there is no need to use another quotative frame, since the same speaker is simply continuing a speech that has already been introduced. “Both have the effect of attracting more attention to the speech or segment of speech that follows.”122
In chapter 8, Runge describes “tail-head linkage,” which is a type of repetition that involves an action from one clause being restated at the beginning of the next clause. “In other words, the ‘tail’ of one clause becomes the ‘head’ of the next.”123 Its effect is to slow down the discourse in anticipation of something surprising or significant.124
8.13 Information Structuring Devices
Part 3 of Runge’s Discourse Grammar addresses information structuring devices. It describes “how variation in the ordering of propositions is used to pragmatically structure the flow of the discourse.”125
Chapter 9 treats information structure, which is most often referred to as “word order” analysis.126 Runge discusses the debate and confusion surrounding Greek word order and whether or not there is a default way of ordering Greek sentences.127 As preliminary to dealing with information structure in Greek, Runge spends most of this chapter sorting through theoretical issues within the wider linguistic world and criticizing Halliday’s systemical functional approach in particular.128 Key to Runge’s approach is the Prague School’s theme and rheme, where “theme corresponds to the ‘established’ material of the clause, while the rheme corresponds to the newly asserted or focal information.”129 This leads to an interest in thematic frames of reference for the clause that follows the frame.
In chapter 10, Runge begins to explore framing devices found in the New Testament. The function of these frames is to establish an explicit frame of reference for the clause that follows.130 They are “fronted” — used at the beginning of a clause — which gives them a highlighting function. Topical frames highlight the introduction of a new participant or topic, or draw attention to a topic change.131 Temporal frames are created when temporal information is placed at the beginning of a clause.132 Spatial frames involve a fronted prepositional phrase that relates to a location.133
Chapter 11 continues with more framing devices. Conditional frames are simply the protases of conditional sentences.134 Comparative frames “establish a basis against which something in the main clause is compared.”135 Reason/result frames “provide the reason for or the result of the main proposition that follows.”136
Chapter 12 addresses circumstantial frames, focusing on adverbial participles and their functions within subordinate clauses. Circumstantial frames precede the main clause to which they are subordinated, and they “set the scene for the main action that follows, but the action is backgrounded with respect to the main action rather than made prominent.”137
In chapter 13, Runge addresses the issue of emphasis, which he defines strictly “as taking what was already the most important part of a clause and placing it in a position of prominence in order to attract even more attention to it.”138 The most common way that emphasis is created in Greek is through the reordering of the default structure of the clause so that the item to be emphasized is found in a prominent position.139
Chapter 14 deals with left-dislocations, which are known elsewhere as cleft constructions, hanging nominatives, pendent nominatives, casus pendens, and independent nominatives.140 The two basic functions of left-dislocations in Greek are to streamline the introduction of a complex entity into one clause instead of two, and to highlight the introduction of an entity.141
8.14 Thematic Highlighting Devices
In the fourth and final part of Runge’s Discourse Grammar, the topic of thematic highlighting devices is addressed. These devices draw attention to redundant elements, which are promoted in order to shape what follows.142
Chapter 15 treats over-specification and right-dislocation. Rather than using the most succinct way in which to refer to a person or thing, instances of over-specification involve longer, overly specified references.143 Right-dislocation refers to the practice of adding more information about an already mentioned person or thing at the end of a clause.144 According to Runge, “both devices serve the common purpose of highlighting particular thematic information that the writer wants the reader to consider at that particular point in the discourse.”145
In chapter 16, Runge explains thematic addition, which is the use of adverbial modifiers “to attract extra attention to parallel elements in the discourse.”146 The adverbial use of καί, the intensive use of αὐτός, and other features are additives that signal to the reader to look for a corresponding element in the context.147
Chapter 17 explains the significance of changed reference and thematic address. Changed reference occurs when a character is referred to in some way that is not the default way to refer to them, often with the effect of recharacterizing them or supplementing the mental representation of who they are. Thematic address refers to the use of a vocative or nominative of address with reference to a certain character and can have a similar effect in recharacterizing him or her.148
Runge’s final chapter discusses the near/far distinction with reference to the demonstrative pronouns οὗτος and ἐκεῖνος. At the semantic level, the distinction between these demonstratives is spatial — the former “near,” the latter “far.” But in certain contexts these can be used for particular pragmatic effects.149 Runge claims that “when there are other elements in the discourse that potentially compete with the default thematic element, the near and far demonstrative pronouns provide a means for the writer to disambiguate the role that these competing elements play.”150 The near demonstrative indicates thematic entities, while the far demonstrative indicate nonthematic entities.
At the end of the book, Runge includes a helpful summary of all the discourse devices discussed throughout, with an indication of their discourse distribution.151
A strength of Runge’s material is that he has already provided an analysis of discourse features across the whole Greek New Testament in The Lexham Discourse Greek New Testament. Below is an extract of that work from Romans 6:1 – 6.152 It is provided here to demonstrate the theory in action. A key is provided following the diagram.153
| PRINCIPLE | 6  →Τί → οὖν ἐροῦμεν |
| SENTENCE |   ἐπιμένωμεν τῇ ἁμαρτίᾳ |
| SUB-POINT | ἵνα [TP ἡ χάρις TP] πλεονάσῃ  |
| SENTENCE | 2 μὴ γένοιτο |
| SENTENCE | [TP οἵτινες ἀπεθάνομεν τῇ ἁμαρτίᾳ TP] πῶς ἔτι ζήσομεν ἐν αὐτῇ |
| SENTENCE | 3 ἢ  ἀγνοεῖτε  ὅτι [TP ὅσοι ἐβαπτίσθημεν εἰς Χριστὸν Ἰησοῦν TP] εἰς τὸν θάνατον αὐτοῦ ἐβαπτίσθημεν |
| PRINCIPLE | 4 συνετάφημεν οὖν αὐτῷ διὰ τοῦ βαπτίσματος εἰς τὸν θάνατον |
| SUB-POINT | ἵνα [LD ὥσπερ ἠγέρθη Χριστὸς ἐκ νεκρῶν διὰ τῆς δόξης τοῦ πατρός LD] [CP οὕτως CP] [TP  + καὶ ἡμεῖς + TP] ἐν καινότητι ζωῆς περιπατήσωμεν |
| SUPPORT | 5 [CD εἰ γὰρ σύμφυτοι γεγόναμεν τῷ ὁμοιώματι  τοῦ θανάτου αὐτοῦ CD] ἀλλὰ    + καὶ τῆς ἀναστάσεως +   ἐσόμεθα |
| ELABORATION | 6  → τοῦτο  → γινώσκοντες |
| SUB-POINT | ὅτι   [TP ὁ παλαιὸς ἡμῶν ἄνθρωπος TP] συνεσταυρώθη  |
| SUB-POINT | ἵνα   καταργηθῇ τὸ σῶμα τῆς ἁμαρτίας τοῦ μηκέτι δουλεύειν ἡμᾶς τῇ ἁμαρτίᾳ  |
Key
PRINCIPLE A sentence that is marked as drawing an inference or assertion from the preceding discourse. Principles are normally signaled by οὖν, διά, διὰ τοῦτο or πλήν. Cf. Levinsohn (2000:128-133).
SENTENCE One or more clauses that have a coordinate relationship to the preceding discourse. Sentences which begin a speech reported within the discourse are labeled as “sentences,” but are indented one level in the outline to reflect that they are technically dependent upon (i.e. subordinate to) the verb of speaking that introduces them. Sentences are coordinated to the preceding discourse using καί, δέ, or asyndeton (the absence of a conjunction).
SUB-POINT A clause that is grammatically dependent upon another one. In most cases, the clause upon which it depends precedes the sub-point. In “Complex” constructions, the sub-point may precede the clause upon which it depends. Sub-points are normally signaled in Greek by ὅτι, ἵνα, εἰ, ἐάν, καθώς, or by a relative pronoun. Sentences which build upon the sub-point are indented and labeled as sub-points.
SUPPORT A sentence that is marked as strengthening or supporting the preceding discourse. Support sentences do not extend or develop an argument, but serve instead to reinforce the preceding point. Cf. Heckert (1996:32 – 36); Levinsohn (2000:91 – 94).
ELABORATION A participial clause that expands upon the action of the main clause on which it depends. Elaboration follows the clause it elaborates.
Forward-pointing Reference — The use of pronouns like “this,” “those” or “it” to point ahead to some “target” that has not yet been introduced. The forward-pointing pronoun is the reference. The forward-pointing reference has the effect of attracting extra attention to the thing to which it refers.
Forward-pointing Target — The use of pronouns like “this,” “those” or “it” to point ahead to some “target” that has not yet been mentioned or introduced. The thing to which it points is the target. The effect of using the forward-pointing reference is to attract extra attention to the thing to which it refers.
Topical frame — The fronting of some thematic element of the clause (often the grammatical subject) in order to establish a specific frame of reference regarding the theme of the clause that follows. Topical frames are used to:
• introduce brand new participants or concepts
• draw extra attention to changes in topic, sharpening comparisons or contrasts
Metacomment — When a speaker stops saying what they are saying in order to comment on what is going to be said, speaking abstractly about it, e.g. “I want you to know that. . .,” “Don’t you know that. . .”
LD Left-dislocation — The introduction of information that is syntactically outside the main clause (i.e. it is “dislocated”), which is then reiterated somewhere in the main clause using a pronoun or other generic reference. Left-dislocations typically introduce something that is too complex to include in the main clause, one that might otherwise cause confusion. The resumptive element of the left-dislocation essentially summarizes the new content, allowing a comment about the new entity to be easily made.
CP Conditional frame — The fronting of subordinate conditional clauses to create a specific frame of reference for the proposition that follows, making clear that the proposition is contingent upon the condition of the frame being met. The condition is not the most important information in the clause, the main predication is. Fronting the condition does not result in emphasis, but establishes an explicit frame of reference for what follows. Conditional frames are often used to establish hypothetical situations in one easy step. The conditional frame enables the writer to introduce the situation and to comment about it in a single complex clause, instead of using several clauses.
+ Thematic Addition — The use of καὶ as an adverb (instead of as a conjunction) to create a connection between two things, essentially “adding” the current element to some preceding parallel element. Thematic addition is generally translated in English using “also” or “too.” Thematic addition can also be used to indicate confirmation of something, which is generally translated in English using “even.” Cf. Levinsohn (2000:100).
Point (Clause Level) — One part of a paired set of statements that usually replaces the counterpoint, and is the more important of the two. Point-counterpoint sets accomplish two primary purposes:
• Explicitly linking two things together that otherwise might not have been connected.
• Drawing more attention to the “point” that it would not otherwise have received.
It should be noted that Runge does not claim to present a comprehensive guide to discourse analysis. In fact, he does not claim that his Discourse Grammar is an account of discourse analysis at all. Strictly speaking, it is a book on discourse grammar, not discourse analysis. The goal is to present the building blocks of discourse; once these basic features are described, it is possible to move on to a formal analysis of larger discourses. Thus, it is Runge’s goal to produce a subsequent volume that moves from discourse grammar to discourse analysis.154 Unfortunately, the distinction between his goals for the Discourse Grammar and a full discourse analysis is not laid out in the volume, which has led others to criticize Runge for not achieving a more comprehensive, extra-sentential analysis of discourse.
Nevertheless, there is no question that Runge’s Discourse Grammar represents a significant step forward for the advancement of discourse analysis within New Testament studies, since a careful articulation of the building blocks of discourse is needed for discourse analysis. Runge’s theoretical framework is well grounded in linguistic principles, having been constructed through rigorous study of an array of general linguists, discourse analysts, and Greek linguists and grammarians. Runge’s work indeed complements conventional grammatical insights by going deeper into semantics, the use of markedness theory, and articulating Greek phenomena with the tools of linguistic analysis. Highly dependent on Levinsohn (though certainly not exclusively so), Runge alleviates some of the difficulties raised with reference to the former’s work. Runge’s evidence covers the entire New Testament, and his theories have been tested and applied across the board. Moreover, the book is clearly written, with excellent illustrations, applications, and definitions that nonspecialists will find accessible. Perhaps best of all for his target audience, Runge’s methods are immediately applicable for exegesis of the Greek New Testament. The result is a practical and insightful tool that offers genuine advancement to students and teachers alike.
Nevertheless, there are some limitations due to the nature of the work. The most obvious is the fact that, like Levinsohn, there is a preoccupation with discourse at the level of the clause and sentence. In fact, Runge is even more focused at this level than Levinsohn, who includes a section on boundary markers of wider units of discourse. Runge offers little for studying wider units of text, boundary markers, or discourse-wide cohesion (cf. Halliday). But again, this is not the goal of the book; both Levinsohn and Runge analyse discourse features, rather than full-blown discourse analysis.
Moreover, while Runge provides a much clearer and linguistically robust account of prominence, highlighting, and emphasis than is common in Greek studies, one could be forgiven for thinking that feature analysis is the primary focus of discourse analysis. Rather, such analysis must serve as the necessary precursor to higher level analysis of the discourse. We must bear in mind that the discernment of prominence, highlighting, and emphasis is only a preliminary step of discourse analysis. Rather, Runge addresses those concepts he deems to be of most exegetical significance.155
Finally, at the theoretical level, Runge lays many of his conclusions on his assumptions about default strategies of expression. Markedness theory, of course, depends on knowing what is default in a language, but, as Runge points out, there is much scholarly unrest about default structures in Greek. Devices that depend on markedness theory can only be fully accepted as legitimate if one accepts Runge’s claims about the default patterns. Some Greek scholars will be comfortable going in this direction; others will be less so.
These last two chapters have explored important models for discourse analysis, and the ways in which Halliday, Levinsohn, and Runge relate are worth exploring briefly. At the linguistic level, Halliday has his own linguistic school, Systemic Functional Linguistics, while Levinsohn and Runge are eclectic in their linguistic theory. There is much overlap, since all three are functional linguists and draw on much of the same research and many of the same linguistic principles that arise out of it. But Runge thinks that some features of Halliday’s SFL are not well suited to the study of Greek since it was developed primarily for English, and in certain ways there are mismatches between English and Greek that render some of Halliday’s approaches misleading for the latter. Yet others will be concerned about trusting an eclectic theoretical model.
Functionally, the biggest difference between Halliday and Levinsohn/Runge is that the former seeks to develop a comprehensive model of language and communication, discussing coherence across the whole text, the development of theme, and the interrelatedness of units wider than the sentence. Levinsohn and Runge are, as noted, more focused on the level of the clause and sentence, since their work is concerned with discourse features — the building blocks necessary for discourse analysis. Nevertheless, also as noted, Levinsohn and Runge discuss Greek discourse features, which makes their work immediately more applicable for the average New Testament student and is no doubt more precise in such application.
Despite such differences, Halliday, Levinsohn, and Runge are not incompatible, since they are aiming to achieve different things through differing means. The student of the Greek New Testament could profitably embrace all three scholars in a complementary manner. For broad, discourse-level consideration of coherence, Halliday is the seminal authority. For gritty, Greek-based, clausal-level analysis, Levinsohn and Runge have no rivals.
Finally, it is encouraging to observe that discourse analytical concerns are now being taken up by some New Testament commentary series. Runge’s own High Definition Commentary series has begun with volumes on Philippians and Romans. It seeks to incorporate the observation of Greek discourse features into widely accessible commentaries on the English text. The Zondervan Exegetical Commentary on the New Testament is a well-developed exegetical series that pays attention to clause and pericope structure, employing syntax diagrams in English. The Baylor Handbook on the Greek New Testament is a burgeoning series that engages with Greek discourse features in its focused analysis of Greek syntax.
Levinsohn, Stephen H. Discourse Features of New Testament Greek: A Coursebook on the Information Structure of New Testament Greek. Second edition. Dallas: SIL International, 2000.
———. “Some Constraints on Discourse Development in the Pastoral Epistles.” Pages 316 – 33 in Discourse Analysis and the New Testament: Approaches and Results. Edited by Stanley E. Porter and Jeffrey T. Reed. JSNTSup 170. Sheffield: Sheffield Academic, 1999.
Porter, Stanley E. “Discourse Analysis and New Testament Studies: An Introductory Survey.” Pages 14 – 35 in Discourse Analysis and Other Topics in Biblical Greek. Edited by Stanley E. Porter and D. A. Carson. JSNTSS 133. Sheffield: Sheffield Academic Press, 1995.
Runge, Steven E. Discourse Grammar of the Greek New Testament: A Practical Introduction for Teaching and Exegesis. Bellingham: Lexham, 2010.
———. The Lexham Discourse Greek New Testament. Bellingham: Logos Research Systems, 2008 (available only as an electronic product as part of Logos Bible Software system).
1. Jan Firbas, “Thoughts on the Communicative Function of the Verb in English, German and Czech,” BRNO Studies in English 1 (1959): 39 – 63; Jan Firbas and K. Pala, “Review of Ö. Dahl, Topic and Comment: A Study in Russian and General Transformational Grammar,” Journal of Linguistics 7 (1971): 91 – 101; E. Beneš, “Die Verbstellung im Deutschen, von der Mitteilungsperspektive her betrachtet,” Philologica Pragensia 5 (1962): 6 – 19.
2. “This book in no sense claims to offer a comprehensive coverage of the multitude of discourse features found in the Greek New Testament. Rather, it is an attempt to describe a limited number of features.” Stephen H. Levinsohn, Discourse Features of New Testament Greek: A Coursebook on the Information Structure of New Testament Greek (2nd ed.; Dallas: SIL International, 2000), vii.
3. See also Stephen H. Levinsohn, “The Relevance of Greek Discourse Studies to Exegesis,” Journal of Translation 2.2 (2006): 11 – 21. Footnotes make reference to more recent publications that supplement or even correct assertions in Levinsohn’s Discourse Features.
4. Ibid., viii.
5. Ibid., ix.
6. Ibid., vii.
7. Ibid.
8. Ibid.
9. Ibid., viii.
10. Instead, Dooley and Levinsohn (Analyzing Discourse, 11) follow John Lyons (Semantics [vol 2; Cambridge: Cambridge University Press, 1977], 614) in using “the term individual style to refer to ‘those features of a text . . . which identify it as being the product of a particular author.’ ”
11. Ibid.
12. Ibid., ix.
13. Ibid.
14. Ibid.
15. Ibid., 1, n. 1.
16. Ibid., 2.
17. Ibid., 3.
18. Ibid., 8.
19. Ibid., 7.
20. In later publications, Levinsohn follows Matthew S. Dryer (“On the Six-Way Word Order Typology,” Studies in Language 21.1 1997. : 69 – 103) in referring to Greek as a VS/VO language. See, e.g., Stephen H. Levinsohn, “Self-Instruction Materials on Narrative Discourse Analysis” (www.sil.org/~levinsohns/, 6).
21. Ibid., 16 – 17.
22. Ibid., 29 – 47. According to Levinsohn’s terminology, the “comment” is part of a topic-comment articulation, which contains the topic of a sentence (usually the subject) and a comment giving information about the topic; see Ibid., 7.
23. Ibid., 48 – 67.
24. Ibid., 69.
25. Ibid., 72. Later publications (e.g., Stephen H. Levinsohn, “ ‘Therefore’ or ‘Wherefore’: What’s the Difference?,” in Reflections on Lexicography [ed. Richard A. Taylor and Craig E. Morrison; Perspectives on Linguistics and Ancient Languages 4; Piscataway, NJ: Gorgias, 2013], 340) express the constraint on interpretation conveyed by δέ as “+Distinctive.”
26. Ibid., 95.
27. Ibid., 95 – 97.
28. Ibid., 99.
29. Ibid.
30. Ibid., 106 – 7.
31. Ibid., 112. For discussion of ten inferential connectives used in the Pauline epistles, see Levinsohn, “ ‘Therefore’ or ‘Wherefore,’ ” 325 – 43.
32. Ibid.
33. Ibid., 118.
34. Ibid., 124.
35. Ibid., 126.
36. Ibid., 126 – 27.
37. Ibid., 133.
38. The role of οὗτος and ἐκεῖνος in participant reference is discussed in Stephen H. Levinsohn, “Towards a Unified Linguistic Description of οὗτος and ἐκεῖνος,” in The Linguist as Pedagogue: Trends in the Teaching and Linguistic Analysis of the Greek New Testament (ed. Stanley E. Porter and Matthew Brook O’Donnell; Sheffield: Sheffield Phoenix, 2009), 206 – 19.
39. Ibid.
40. Levinsohn, Discourse Features, 134 – 35. The examples offered here are cited by Levinsohn.
41. Ibid., 135 – 36.
42. Ibid., 142 – 43.
43. Ibid., 143.
44. Ibid., 148.
45. Ibid., 150.
46. Ibid., 155 – 56.
47. Ibid., 152.
48. Ibid., 169.
49. Ibid., citing Callow, Discourse Considerations, 52 – 53.
50. Levinsohn, Discourse Features, 169.
51. Ibid., 170.
52. Ibid., 172 – 74.
53. Ibid., 174 – 75. See also Stephen H. Levinsohn, “Aspect and Prominence in the Synoptic Accounts of Jesus’ Entry into Jerusalem,” Filología Neotestamentaria 23 (2010): 161 – 74, which challenges the approaches to aspect and prominence of both Robert E. Longacre and Stanley E. Porter.
54. Ibid., 177.
55. For exceptions to this assertion, see Stephen H. Levinsohn, “Adverbial Participial Clauses in Koiné Greek: Grounding and Information Structure” (paper presented at the International Conference on Discourse and Grammar, Universeit Ghent, Belgium, May 2008 [www.sil.org/levinsohns/papers]).
56. Ibid., 181.
57. Ibid., 191.
58. Ibid., 197 [italics are original].
59. Ibid., 200.
60. Ibid., 202 – 11.
61. Ibid., 215.
62. Ibid., 218.
63. Ibid., 235 – 38.
64. Ibid., 240 – 44.
65. Ibid., 248 – 60.
66. Ibid., 261 – 64.
67. Ibid., 264 – 66. See also Stephen H. Levinsohn, “Is ὅτι an Interpretive Use Marker?” in The Linguist as Pedagogue: Trends in the Teaching and Linguistic Analysis of the Greek New Testament (ed. Stanley E. Porter and Matthew Brook O’Donnell; Sheffield: Sheffield Phoenix, 2009), 163 – 82.
68. Ibid., 266 – 68.
69. Ibid., 271.
70. Ibid.
71. Ibid. [italics are original].
72. Ibid., 274 [italics are original].
73. Ibid., 275.
74. Ibid., 276 – 77.
75. Ibid., 277.
76. Ibid.
77. Ibid.
78. Ibid., 278.
79. Ibid.
80. Ibid.
81. Ibid., 278 – 79.
82. Ibid., 279 – 80.
83. Ibid., 280 – 84.
84. Porter, “Discourse Analysis,” 26.
85. Ibid.
86. See, e.g., Dooley and Levinsohn on the methodology of their volume, Analyzing Discourse, iii: “First, we intend it to be practical, addressing issues commonly confronted by field linguists. Rather than attempting to apply a rigid theory or survey a variety of approaches, we provide a methodology that has been refined over years of use. Second, although we follow no rigid theory, we aim for more than a “grab-bag” of diverse methodologies by attempting to present the material within a coherent and productive framework. Specifically, we follow a functional and cognitive approach that seems to be a good approximation of how discourse is actually produced and understood.”
87. See, e.g., Stephen H. Levinsohn, “Towards a Typology of Story Development Marking,” in Bantu Languages: Analyses, Description and Theory (ed. Karsten Legère and Christina Thornell; Köln: Rudiger Köppe Verlag, 2010), 143 – 51.
88. Steven E. Runge, Discourse Grammar of the Greek New Testament: A Practical Introduction for Teaching and Exegesis (Bellingham: Lexham, 2010).
89. Steven E. Runge, ed., The Lexham Discourse Greek New Testament (Bellingham: Logos Research Systems, 2008) (available only as an electronic product as part of the Logos Bible Software system).
90. Runge, Discourse Grammar, 3.
91. Ibid., 5.
92. Ibid., 11.
93. Ibid.
94. Ibid., 13.
95. Ibid., 15.
96. Ibid., 17. See also Steven E. Runge, “Now and Then: Clarifying the Role of Temporal Adverbs as Discourse Markers,” in Reflections on Lexicography (ed. Richard A. Taylor and Craig E. Morrison; Perspectives on Linguistics and Ancient Languages 4; Piscataway, NJ: Gorgias, 2013), 327 – 48.
97. Runge, Discourse Grammar, 18 – 19.
98. Ibid., 20.
99. Ibid., 21.
100. Ibid., 23.
101. Ibid., 24.
102. Ibid., 31.
103. Ibid.
104. Ibid., 37 – 38.
105. Ibid., 43.
106. Ibid., 48.
107. Ibid., 52.
108. Ibid., 54.
109. Ibid., 56.
110. Ibid., 57.
111. Ibid., 59.
112. Ibid., 64.
113. Ibid., 70.
114. Ibid., 73.
115. Ibid., 74.
116. Ibid., 101.
117. Ibid., 117. Such devices can also occur by themselves without functioning as metacomments.
118. Ibid., 128 – 29. Runge has subsequently described the present as indicating “non-past” temporal reference: “The present tense-form often functions as an unmarked form for non-past reference” (Steven E. Runge, “The Verbal Aspect of the Historical Present Indicative in Narrative,” in Discourse Studies & Biblical Interpretation: A Festschrift in Honor of Stephen H. Levinsohn [ed. Steven E. Runge; Bellingham: Logos, 2011], 218).
119. Runge, Discourse Grammar, 134.
120. Ibid.
121. Ibid., 137.
122. Ibid., 145.
123. Ibid., 163.
124. Ibid.
125. Ibid., 179.
126. Ibid., 181.
127. Ibid., 182 – 84.
128. “Most all of Halliday’s theoretical work to develop his framework was conducted on rigidly ordered languages, particularly English. Halliday’s definition of ‘focus’ is inextricably tied to stress and intonation, just as in English. His theory is well suited for English because prosody is the primary means of marking focus in rigidly ordered languages. . . . SFL postulates that the initial element in a clause, be it a conjunction, subject, or fronted focal constituent, is always the ‘theme.’ This idea works fairly well in a rigidly configurational language, but it proves inadequate for non-configurational languages” (ibid., 203).
129. Ibid., 201.
130. Ibid., 207.
131. Ibid., 210.
132. Ibid., 216.
133. Ibid., 220.
134. Ibid., 227.
135. Ibid., 233.
136. Ibid., 237.
137. Ibid., 250.
138. Ibid., 269.
139. Ibid., 284.
140. Ibid., 287.
141. Ibid., 291.
142. Ibid., 315.
143. Ibid., 317.
144. Ibid.
145. Ibid.
146. Ibid., 337.
147. Ibid., 339, 341.
148. Ibid., 363.
149. Ibid., 368.
150. Ibid., 369.
151. Ibid., 385 – 91.
152. Runge, Lexham Discourse Greek New Testament, Romans 6:1 – 6.
153. Definitions are taken from the glossary of the Lexham Discourse Greek New Testament.
154. So Runge commented via personal correspondence.
155. Ibid., xix.