Open Journal of Modern Linguistics, 2015, 5, 302-318
Published Online June 2015 in SciRes. http://www.scirp.org/journal/ojml
How to cite this paper: Woods, J.R. (2015). Cohesive Chains in the Transfiguration Narrative of Matthew 17:1-13. Open
Journal of Modern Linguistics, 5, 302-318. http://dx.doi.org/10.4236/ojml.2015.53027
Cohesive Chains in the Transfiguration
Narrative of Matthew 17:1-13
Justin R. Woods
McMaster Divinity College, Ontario, Canada
Received 19 July 2014; published 26 June 2015; accepted 30 June 2015
Copyright © 2015 by author and Scientific Research Publishing Inc.
This work is licensed under the Creative Commons Attribution International License (CC BY).
While much work in biblical studies has been offered in the form of theological exposition and
historical critical speculation of literary origins of the gospels, few modern biblical studies scru-
tinize the language of these gospels from rigorous linguistic criteria. This paper takes a discourse
analytic approach from the field of Systemic Functional Linguistics (SFL) founded by M. A. K. Hal-
liday. The notion of cohesive chaining is taken in order to evaluate the level of literary unity in the
original Greek of the episode of the transfiguration narrative found within the Gospel of Matthew.
Cohesive chaining will be defined; its function within the greater sociolinguistic theory of SFL will
be examined, and then the concept will be directly applied. The study concludes with a unique
contribution demonstrating how these cohesive chains are then unified via choices in verbal as-
pect as the grammar that not only solidifies semantic continuity among those chains, but contours
the discourse using a scheme of markedness that signals the prioritization of its message. The
hope of this study is to reinforce the recent paradigm shift of biblical research utilizing modern
linguistic paradigms as tools to transform biblical interpretation and exegesis into a rigorously
discourse-centered linguistic methodology.
Cohesion, Discourse Analysis, Linguistics, Markedness, Verbal Aspect
This essay will examine the cohesive ties running through the text of Matthew’s Transfiguration account (17: pp.
1-13). This analysis views thes e linguistic resources as demonstratin g that Matthew’s account is a linguistically
unified and situationally responsive use of an episode of Jesus’ life. This study is offered in the hope that fur ther
linguistic studies w ill be conducted from it because Hallidayan fr amework has already been specifically fruitful
in cohesive studies of text analysis outside of the Gospels (Van Neste, 2004; Reed, 19 99: pp. 28-46; idem., 1997:
J. R. Woods
pp. 188-218). First, a definition of cohesion will be established from a Hallidayan Systemic Functional Linguis-
tic (SFL) perspectiv e which itself has been p roven useful in recent biblical scholarship (Martin-Asensio, 2000).
Then examples of cohesion will be given. Finally, the framework of cohesion will be applied to the transfigura-
tion episode in Matthew’s Gospel. I will conclude with a sociolingu istic interpretation of Matthew’s transfigura-
tion episode as an instance of social discourse.
2. Place of Cohesion in the Linguistic System
Understanding cohesion requires understanding its placement in the general linguistic framework of SFL as a
realization of one of the semantic systems of language (Halliday & Hasan, 1976: pp. 26-27). This framework
will be examined f rom two perspectives which h ave a stratified relation : firstly, cohesion’s broad er relationship
out of the context of social situ ation and cultur e, and lastly, its d irect contribu tion within ling uistic expressio n of
2.1. Broader Stratified Relation to Social Context
There are three general areas of “human experience” which discourse seeks to construe: the “field”, or general
activities conveyed within culture, the “tenor”, or role relationships enacted between the members of a culture,
and the “mode” or channel and organization of a message within culture (Halliday, 1985: p. 30; cf., Halliday,
1978: p. 129). These categories “field, tenor, and mode” reflect the “ecological matrix” which constitute what
Halliday calls the greater “context of culture” and it is forms of discourse that use language to socially construe
the m ( Halliday & Matthie s s en, 2006: p. 321; Martin, 2001: p. 45; Halliday, 1985: pp. 29-30; cf., also Lukin et al.
2011: p. 193). Halliday also observed a tendency of social convention from culture that seeks to develop a go-
verning predictability over language by maintaining strong associations between language use and “situation
types” within the structure of society (Halliday, 1978: p. 111, 157; Halliday, 2007: pp. 231-248; O’ Donnell,
1999: pp. 251-252). This governance can be seen in Halliday’s referen ce to “code” enforcement. Halliday refers
to “code” as “governing the choice of meanings by a speaker and their interpretation by a hearer … The code
controls the semantic styles of the culture … it determines the semantic orientatio n of speakers in particular so-
cial contexts.” (Halliday, 1978 : p. 111 ; Halliday, 2007: pp. 231-248). Such cultural convention reinforces which
lexicogrammatical selections from the macro-level of field, tenor, and mode will be “appropriate” to use within
the micro-level “context o f situation” as it feeds back in to that larger cultural contex t (Halliday & Hasan, 1976:
pp. 21-28, 52). Reed offers an insightful paraphra s e of this phenomenon.
“Halliday is suggesting that changes in the context of situation contribute to changes in the use of language.
Speakers/authors conform their discourse to the context of situation, and consequently draw upon accepted
forms of language that others recognize as appropriate for that situation.” (Reed, 1997: p. 54).
This situation-specific phenomenon represents a certain “register” of speaking (Halliday, 1978: pp. 110-111;
Halliday, 2007: pp. 181-182; Halliday & Hasan, 1976: pp. 21-26). Fig ure 1 illustrates the stratif ied relationship
between the register of language as a “semantic configuration” (via metafunctions ideational, interpersonal, and
textual) and the elements of the greater social context (Lukin, 2011: p. 192; Hasan, 2009: p. 170; Halliday &
Hasan, 1989: p. 55; Hasan, 1995: p. 231).
Figure 1. Register: language reflecting situation types within a cultural (Martin & Rose, 2007: p. 297).
J. R. Woods
While registers are not produced by any one member of society, a “register analysis” of any single text does
help reveal the contribution it makes in this regard (Porter, 2004: pp. 220-233). A fu ll-scale register analysis is
not within the scope of this essay, but those elements found to contribute toward cohesive harmony of the text
may be considered as the key semantic fields and lexical selections formative for the register that gave rise to the
discourse. Cohesion, therefore, may be understood as a single dimension of register analysis (Porter, 2004: p.
224; Teich, 2009: p. 122).
Matthew’s account then represents a single linguistic contribution toward a larger cultural phenomenon. It
was this unique and central situation which forced the Jesus group to appropriate a normative linguistic expres-
sion for the identity and resurrection of Christ. Register concepts may also lead to criteria for “authenticating the
words of Jesus” from the Gospels as well (Porter, 2004: pp. 210-238). No doubt, it was thrust upon the apostles
as a peculiar and unprecedented “context of situation” that required their immediate management (Halliday,
2007: p. 180; idem, 1978: 28-29). This helps to illuminate their motivation to quickly establish so much litera-
ture and preaching (cf., Luke 1:1; Acts 5:42). Trying to attain some measure of predictability would only have
been a natural impulse. They desired the predictability which carving out a linguistic register within the larger
social structure could provide (Reed, 2000: pp. 121-153). Reed sees Paul trying to establish a dialect for much
the same functional reasons, though I would agree with Halliday that early Christianity was constructing a
“technical register” out of their unique situation type (i.e. life and resurrection of Jesus) to be able to embed
within various societies, i.e. “antilanguage.” (Wo ods, 2013; Hal l iday, 197 6).
Though a descriptive linguistic account of this contribution may be offered, one limit of this study is that it
cannot reach hard conclusions about the entire set of registers of early Christianity in this regard. Indeed, the
Gospel of Matthew portrays Jesus as engaging many arenas: rabbinic circles, prophetic activities, public ap-
pearances, family relations, etc. It may be implied then that the disciples were trained to interact among these
various “situation types” in society thr ough the activity of their leader prior to his death and resurrection. If the
cohesive ties speak strongly toward a unified discourse of this section, then this may be seen as Matthew’s
“contextual configuration” which he offers to sh ape a larger register deman ded by his context of situation (Ha-
san, 1995: 231; Halliday and Hasan, 1989: 48, 55, 88-96). It would be his way of using language to reinforce the
issues most directly influencing the “solidarity” that could maintain the “social identity” of his fellowship in
Christ within the greater Jewish and Greco-Roman context (Halliday and Hasan, 1976: 186; Hasan, 1995: 229;
Rohrbaugh, 2007: 179-181; Hudson, 2001: 106; Schmidt, 2001: 22). T his topic, however, lies beyond the scope
of this essay. Language here represents one of the foundational mediums through and by which social identity
and change may be transmitted and modified. As such, cohesive ties that unify such social discourse may be in-
terpreted as representative of the predominant issues for which it was designed to “socialize” its audiences (Hal-
liday, 1978: 99; Halliday and Hasan, 1989: 41).
Kress relates the ability of language in this regard as tending specifically to the perception of social distancia-
tion and power: “Language provides the most finely articulated means for a nuanced registration of differences
of power in social hierarchical structures, both as a static system and in process.” (Kress, 1989: 52). In light of
this, Matthew may be viewed as issuing common themes about Jesus that seek to distance itself from competing
views. Considering how quickly the Christian movement spread throughout the Mediterranean, and the kind of
resistance it was met with, it is little surprise tha t different communities wou ld quickly develop their own so cia-
lized expression of certain topics whether informed by apostolic precedent or not. Matthew’s gospel material
then may be seen as a means of codifying a linguistic constraint for such communities. See Keener who ob-
serves the situation for Matthew’s audience is “struggling with those they believe to be illegitimate spokesper-
sons” (Keener, 2009: 4 9). The cohesive ties that unify this text would function as that material which would be
most essentially formative to this goal. See Meeks for an excellent study of “resocialization” in the early Chris-
tian community (Meeks, 1986: 13).
2.2. Direct Semantic Contribution
A unified text of social discourse will have simultaneously running through it three predominant semantic
threads according to the Hallidayan framework: 1) ideational, 2) interpersonal, and 3) textual (Halliday, 1985:
29-30; Halliday, 1978: 128). While Halliday makes it clear that a sentence creates its own internal unity “by
virtue of the structure” itself, a text larger than the sentence is “not a structural unit” but a semantic one. There
are other metafunctional representations of Halliday’s theory (Halliday and Hasan, 1989: 44). noting that the
J. R. Woods
nomenclature “ideational” conflates “experiential” with “logical.” (Fawcett, 2000: 51, 72, nt. 20). Halliday as-
serts that a text as a cohesive “semantic unit” is “not composed of sen tence s but is realized in sentences” (Halli-
day, 1978: 135). Though the role of structure will be discussed later, it is clear that text or discourse requires
linguistic de vi ce s a nd re l ati ons whi c h do not rely on grammatical st ruct ure . Be c a use c ohe s ion works to achieve this
very end, it is found among the semantic systems which realize the textual metafunction of language (Halliday and
Hasan, 1976: 120). As a textual resource in the semantic system, the function of its lexical and grammatical con-
nections serve s to provi de gen er al organizati on for the dis course (Halliday, 1985: 532; cf., Stubbs, 1983: 15-19).
3. Cohesive Relations
There are five general types of cohesive relation identified by Halliday and Hasan: “reference, substitution, el-
lipsis, conjunction, and lexical cohesion” (Halliday and Hasan, 1976: 4; Halliday and Hasan, 1989: 73; Halliday,
1985: 532-538; Gregory, 1988: 311; Martin, 2001: 35. Cf., also Porter, 1999: 304-307). Cohesion setting “dis-
course relations which transcend grammatical structure” makes it possible to set these semantic relations be-
tween units of any size, direction (i.e., an aphoric, or cataphoric), or distance of separation (Halliday and Ha san,
1976: 14; Halliday and Hasan, 1989: 71; Stoddard, 1991: 15). It is, therefore, capable of connecting at almost
any level of the “syntactic rank scale” of the linguistic system, i.e., word, word group , clause, etc. (Tucker, 1998:
41; cf., Halliday, 1985: 7-10) See chart in Halliday and Hasan, 1976: 29. The relations may also incorporate ad-
jacent uses of tense as capable of cohesive tie (Gledhill, 2009: 65-84). Their relation is such that when one unit
shows to be related to another, the former is reliant on the latter for meaning or otherwise could remain too
enigmatic for interpretation (Halliday and Hasan, 1976: 4-5). Examples are provided below.
3.1. Examples of Cohesive Tie
The simplest example is when there is a relationship made between parallel units in the rank scale, as in the case
of John and he below.
John is running for office. He is expecting to win by a landslide.
After having read both clauses, understanding who the pronoun he stands for in the second clause is revealed
by a cohesive tie to John in the first. This type of relation may also be expanded to entire word groups as the
next example illustrates.
The neighbors grow ye llow Chrys anthemums.
a) I could gro w R ED ONES.
b) I could grow the SAME.
c) I could gro w some (TOO).
d) I could grow some of the SAME (Halliday and Hasan, 1976: 110).
The concept of presupposition is foundational to cohesion. Statements a) through d) establish the lexical con-
nections with the surrounding co-text to supply an interpretive constraint to meaning. It stands, therefore, that
the absence of such connections renders a corresponding loss of constituent interpretability. Here much of the
relations instantiate at the word or word group level, in this case r elating “yellow Chrysanthemums” to the capi-
talized terms in various ways. Moving further along the rank scale, the following shows r elations toward larger
linguistic structures where the parenthetical remark is assumed to remain presupposed (Halliday and Hasan,
a) You’re tired.
b) Are you tir ed?
c) You’re not tired?
d) Aren’t you tired? Yes. (I am tired.) (Halliday and Hasan, 1976: 138).
As one can see, the term “Yes” taken in the absence of presupposition renders interpretation impossible (Hal-
liday and Hasan, 1989: 75). In this case, no one reading the text would be able to say what is being agreed to
without the presupposed statement. Both what is presupposed from the linguistic context as well as the referring
item itself share a two-way relationship (Halliday and Matthiessen, 2006: 84). The original clause elicits the
production of the response, and the response relies on its presupposed clause for meaning (Halliday and Hasan,
Hence, the surrounding co-text is both probabilistically influential as well as semantically reflective toward
the utterance (Halliday and Hasan, 1989: 222).
J. R. Woods
It is easy to see how the potential for complexity with cohesion could be considered as theoretically limit- less
(Stoddard, 1991: 1). This is especially important when one considers such linguistic devices as word play, me-
taphor, poetry, “antilanguage”, and so on (Halliday, 1976: 570-584). Indeed, being able to establish relations
between any item of the linguistic system apart from structure allows for a tremendous range of possibilities to
create the “texture” that constitutes part of the unity of a text (Halliday, 1985: 579-585; Halliday and Hasan,
1976: 2-3; Halliday and Hasan, 1989: 70-96; Stoddard, 1991). There are even times when entire units of dis-
course may be the referent. Consider the potential to draw connections to an entire discourse unit.
Gary: That’s when it hit me.
My leg started jumping up a n d down. I just couldn ’t c ontrol m y nerve s.
J. R.: (That’s) Te rrible (Schiffrin, 1994: 220).
The comment which J. R. makes does not point back to any specific item in the discourse, but “presupposes
everything that precedes” in order to interpret or predicate over the entire section of text (Halliday and Hasan,
3.2. Linguistic Structure Mapping Cohesion
While the relations between cohesive ties is examined in this study as predominantly a non-structural one,
structure can, however, act like a treasure map pointing to the relations it contains. Coming back to the example
above, notice the “parallelisms” which aid in signaling and interpreting the paradigmatic alterations within the
structure, as illustrated in Figure 2 (Halliday and Hasan, 1989: 81-82). See Fawcett (2000: 56, nt. 6) for an in-
sightful discussion as to how the issue of structure was omitted from Halliday and Hasan’s work, Cohesion in
English (1976), not due to overlooking it, but because Halliday’s work was designed specifically to focus on
non-structural cohesion. Fawcett readily admits the importance of structure toward cohesion: “the fact is that
TRANSITIVITY, MOOD, THEME etc. can also contribute to the ‘cohesion’ of a text, as Martin (1992) clearly
demonstrates.” (Fawcett, 2000: 56). This was not a criticism of Halliday and Hasan’s work, only to state that
they understood this princ iple and that it was simply outside the s tated scope of the work they created. Faw cett's
criticism does, however, extend to those works which followed their work blindly producing the “unfortunate
effect” of “incomplete” works on the topic later on. I have incorporated in this essay a limited attempt toward a
more complete study of cohesion in the Greek NT (Fawcett, 2000: 56).
The structures themselves do not constitute the textual cohesion but seen from the oppositions in both the
syntagmatic and paradigmatic axes, a repetition in structure renders cohesive relationships easier to recognize
and interpret. This is because the syntagmatic axis for a text is neither haphazard nor completely autonomous
but may well be driven by the impetus to recontextualize former elements in light of new developments in the
discourse. Tucker’s comment is especially astute here: “The ‘autonomy of syntax’ is not a principle of systemic
functional linguistics. A major assumption in SFG is that the structural potential is the way it is in order to re-
flect the meaning potential” (Tucker, 1998: 61). See Fawcett ( 2000: 70-73, 152 -154) for an excellent discussion
of why Halliday may have been hasty in his early abandonment of this S-P-C-A nomenclature representing an
“integrated syntax” (Halliday uses “combined”) of structure which the Cardiff Grammar has taken up.
3.3. Cohesive Chai ns
Identifying several of these cohesive ties in a sequence which runs through a text is called a “cohesive chain”.
by a landslide.
Figure 2. Parallel structure aids recognition of cohesive relations.
(Fawcett, 2000: 70-73, 152-154).
J. R. Woods
Halliday and Hasan have categorized such chains into two “typical” groups (Halliday and Hasan, 1989: 84).
Refer to Figure 3 below for de l i ne a tion acco rding to the ir work.
These general categories will be operative in this essay. Discovering several sequences of cohesive tie within
a text demonstrates a certain level of discourse unity. According to Halliday and Hasan, the major contribution
which cohesion makes in affirming a stronger degree of “cohesive harmony” lies along the cline of the extent to
which these cohesive chains interact with and support one another (Halliday and Hasan, 1989: 84). The semantic
domains of each chain cannot remain in isolation, but must find “interaction” throughout the grammar of the
discourse to finalize their contribution to unifying the text (Halliday and Hasan, 1989: 91; Martin, 2001: 40).
Martin offers a helpful analysis of this phenomenon in his examination of a “feminist narrative” resource, Pig-
gybook (Brown, 1989, quoted in Martin, 2001: 39).
3.4. Chain Interactions
Mr Piggott lived with his two sons, Simon and Patrick, in a nice house with a nice garden, and a nice car in
the nice ga ra ge. Inside the house w a s his wife.
“Hurry up with the breakfast, dear,” he called every morning, before he went off to his very important job.
“Hurry up with the breakfast, Mum,” Simon and Patrick called every morning, before they went off to their
very important school.
After they left the house, Mrs Piggo tt washed all the breakfast th ings… made all the beds… v acuumed all the
carpets… and then she went to work.
“Hurry up with the meal, Mum,” the boys called every evening, when they came home from their very im-
“Hurry up with the meal, old girl,” Mr Piggott called every evening, when he came home from his very im-
As soon as they had eaten, Mrs Piggott washed the dishes … wa shed the c l othes … did the ironing… and then
she cooked some more.
Here, one can see a number of identity chains for each participant. See the example chains below.
• Mr Piggott-his-his-he-he-his-they,
• his two sons-Simon/Patric k -Simon/Patrick-they-their-they-the boys-they-their-they,
• his wife-dear-Mum-Mrs Piggot-she-Mum-old girl-Mrs Piggott-she.
Also notice the similarity chain s utilizin g various lexical semantic relations. Exa mples ar e below.
• house-garden-garage-beds-carpets (Martin, 2001: 40).
Analyzing the chains from a syntagmatic perspective, one can see that the repetition of the epithet nice occurs
inside a repeated syntactical pattern of “epithet-thing” for various items in the meronymy chain for house, e.g.,
nice house, nice garden, nice garage. Mrs. Piggott’s chain offers a contrastive syntax by key omissions. I will
refer back to the rhetoric involved in omissions later on in Matthew’s transfiguration account. See section 4.2.
These are in stances of cohesive tie that all refer to the same person, place,
or thing typically instantiated through either grammatical cohesion which
utilizes pronominals, demonstratives, definite article, or comparatives, or
through lexical cohesion which utilizes equivalence, naming or semblance.
Similarit y ch a in s .
These are instances of cohesive tie related by either “co-classification” or
“co- ex ten sion .” Thes e are “m ember s of non -identical but related classes of
things, or events.” These are typically realized by substitution and ellipsis,
as well as a full spectrum of general lexical relations such as repetition,
synonymy, antonymy, meronymy, equivalence, naming and semblance.
Figure 3. Types of cohesive chain. (Halliday an d Hasan, 1989).
J. R. Woods
For instance, her introduction contrasts the previous items by an absence of the epithet nice, and her going to
work contrasts the destinations of the males by its absence of the epithe t important.
Elements within chain interactions may also strategically collocate with markers of discourse development
which set the audience’s expectations and characterizations of events within the plot (Puigdollers, 2009: 34). For
example, Mrs. Piggott’s chain shows an implicit compliance to the males’ demands for food in the morning, and
then her in- volvement becomes explicit when she has to clean their messes in the evening (Martin, 2001: 42).
Here, the start of the day is marked by the explicit demands of the males, whereas the end of the day is marked
by the explicit cleaning activities of the wife.
The contrastive parallelism is manifested through the absence and presence of cohesive ties in a reasonably
predictable manner. During the morning scene, the plurality of demands by the males sets their collocation
alongside the wife’s implicit obedience to respond (Halliday and Hasan, 1989: 81). This structure expectedly
shifts in the nighttime scene where the exp licit mention of the plur ality of the wife’ s activities is collocated with
the absent response of the males. One can see that multiple dimensions of discourse function such as topic man-
agement, plot development and chronology may find their place in the textual structure of the ties that signal
their emerging boundaries (Mor ley, 2009: 5; cf., Halliday and Hasan, 1989: 94). It is not the mere existence of
cohesive chains that unifies the texture of a discourse, but the interactions they instantiate between th ese chains
through the grammar (Halliday and Hasan, 1989: 91).
4. Examination of Matt 17:1-13
This study will now move to a direct examination of the cohesive chains in the text of Matthew’s transfiguration
account according to the NA28. The chains are as follows: 4.1 Identity Chain for Jesus; 4.2 The Supernatural;
4.3 The Disciples; 4.4 Conjunctive Chain; and 4.5 Imperfective Aspectual Chain. See Table 1 below for the
identity chain for Jesus.
4.1. Identity Chain for Jesus
The first notable aspect of the identity chain for Jesus (see Table 1) is its “text exhaustive” status through the
narrative (Halliday and Hasan, 1989: 84). In other words, Jesus is referred to at every point throughout the de-
velopment of the narrative plot. The single exception to its exhaustive nature might be considered at the clause
in v. 6 where it says that the disciples became terrified and fell on th eir faces. However, when one considers that
it is their reception of the statement about the Son from the voice in the cloud which is the cause of this fear, Je-
sus carries a bit of secondary influence even within that verse as well. It the very exhaustive nature of this chain
which may supply the impetus behind commentators who feel the need to offer a full topical study for every re-
ferent embedded in the narrative’s “variety of allusions” because it would be tempting to see every referent as
predicating Jesus (Keener, 2009: 437). Davies and Allison (2004) are the quintessential example of how their
92 explicit references to “Moses” between pp. 685-741 actually distract the audience from the linguistic issues
Table 1. Chain indentifying Jesus through Matt 17:1-13.
v. 1 παραλαμβάνει *ἡμᾶς προσῆλθεν ὁ Ἰησοῦς λέγων
ὁ Ἰησοῦς εἶναι ἁψάμενος τὸ ὅραμα
ἀναφέρει Θέλεις εἶπεν ὁ υἱὸς τοῦ ἀνθρώπου
v. 2 μετεμορφώ θη σοὶ v. 8 αὐτὸν v. 10 αὐτὸν
αὐτοῦ v. 5*αὐτούς Ἰησοῦν v. 11 ὁ
αὐτοῦ οὗτός μόνον ἀποκριθεὶς
v. 3 *αὐτοῖς ἐστιν v. 9 καταβαινόντων εἶπεν
αὐτοῦ ὁ υἱός μου ὁ ἀγαπητός αὐτῶν v. 12 λέγω
v. 4 τῷ Ἰησοῦ ἐν ᾧ εὐδόκησα ἐνετείλατο ὁ υἱὸς τοῦ ἀνθρώπου
κύριε αὐτο ῦ ὁ Ἰησοῦς μέλλει
* = ambiguous reference Pronou ns by th eir n ature o ften carr y ambig uit y though a dis cours e. Here it is un certai n wheth er Jes us is included in th is col-
lective pronoun or not, as the disciples may be seen as being differentiated from Jesus (Halliday and Hasan, 1989: 89).
J. R. Woods
running through the text in place of a wealth of knowledge on a topic that is given no elaboration in this section.
Functional analysis of cohesive chains extends further than simply textual unity. As formerly demonstrated,
authors will often signal th e development of a discours e by thematic collocations wh ich are strategically paired
with each of the n od es w ithin their cohesive chains (Halliday and Has an, 1989: 50, 81 -82; Tucker, 1998: 94-96).
The concept is a synthesis of stru ctural “parallelisms” in the text which motivate the lexical co llocations. Halli-
day and Hasan speak of the “interdependence between grammatical and lexical cohesion.” (Halliday and Hasan,
1989, p. 82). Tucker observes that “each lexical item possesses its individual grammar, its unique set of collo-
cates.” (Tucker, 1998: 96). This concept is especially seen in the relationships which the syntagmatic and para-
digmatic axes have with cohesion, as aforementioned. So that, as the lexemes of each chain are used, they moti-
vate the collocation of systemically and/or semantically related sets of terms and the gra mmatical structures that
can realize their relations. This is manifested in the chain interactions that unify both the chains themselves as
well as the overall text to communicate key issues in social discourse.
See the discussion below on how the identity chain for Jesus collocates with that of the disciples oscillating
both case and function to open and close the activity section of the transfiguration. It can be seen as no coinci-
dence that the repetition of an explicit grammaticalization of Jesus’ name, ὁ Ἰησοῦς, appears at key moments
through the narrative structure.
v. 1 ὁ Ἰησοῦς τὸν Πέτρον καὶ Ἰάκωβον καὶ Ἰωάννην… κατʼ ἰδίαν
v. 4 τῷ Ἰησοῦ63
v. 7 ὁ Ἰησοῦς
v. 8 αὐτὸν Ἰησοῦν μόνον
Verses 1 and 8 help audiences to process the textual structure which surrounds the section’s primary epicenter
of activity. The no minative grammaticalization of v. 1, ὁ Ἰησοῦς, communicates to audiences that Jesus initiates
the scene prior to the start of the divine activity. Additionally, the collective use of the article covers the dis-
ciples who are appropriately in the accusative as receptors of Jesus’ activity (Porter, 1999: 88, 110). The accusa-
tive functions as “direct object” of the verb and Porter uses Matt 17:1 as an example not covered under the
Granville Sharp rule where the article “covers several personal names.” This sets an important pattern for the
structure of the episode because a similar pattern is paralleled at the conclusion of the activity where another
collective pronoun for the disciples αὐτῶν identifies their reception of Jesus’ initiative to provide his encou-
ragement in v. 7 which, by no happy coincidence, is also fully grammaticalized as ὁ Ἰησοῦς. The genitive case
of the collective pronoun for the disciples in v. 7 is accounted for by the functional use of it as the direct object
of the verb ἁψάμενος which cleverly reiterates the possessive genitive of the next verse τοὺς ὀφθαλμοὺς αὐτῶν,
thereby setting the cohesive ties through the grammatical structuring (Porter, 1999: 94, 130). The close of the
activity is signaled by the rhetorical impact of switching from the three disciples (τὸν Πέτρον καὶ Ἰάκωβον καὶ
Ἰωάννην) as grammatical objects whom Jesus leads alone κατʼ ἰδίαν and then finally arrives at the three-fold
emphatic reference to Jesus (αὐτὸν Ἰη σοῦν μόνον) who now functions as the sole (μόνον) grammatical object of
their frightened gaze. The phrase τὸν ἀδελφὸν αὐτοῦ is separated function ally because it is used to differentiate
John the disciple from “John the Baptist” Ἰωάννου τοῦ βαπτιστοῦ at the end of the episode (cf., v. 13). Mat-
thew’s choice of the article τὸν for the collective grammatical object in v. 1 both foreshadows and culminates
toward the intensive αὐτὸν for Jesus at the activity’s completion.
Therefore, rather than viewing the use of the article here as “highlighting” any one disciple in particular, the
discourse function of the article demonstrates the cohesive nature of the boundary markers framing the scene
(Nolland, 2005: 698; contra Davies and Allison, 2004: 2, 694). This analysis is corroborated by Matthew’s ref-
lective use of the summative term τὸ ὅραμα recognized by several commentators (Hagner, 1998: 498; France,
2007: 652; Newman and Stine, 1992: 541). Firstly, it occurs in the text at a point where it can cohesively tie
with the whole of the activity anaphorically. Additionally, it establishes a similar type of “point of departure”
relationship which the temporal phrase in v. 1 served to introduce the entire scene coming out of the previous
section (Levinsohn, 1995: 61; cf., Black, 2002: 211). The phrase in v. 1 Καὶ μεθʼ ἡμέρας ἓξ is a recognized
“point of departure” introducing the transfiguration section which v. 9 has much in common with and is noted
by commentators in various ways (Nolland, 2005: 699; Hagner, 1998: 484). Nolland recognizes v. 9 as a
“change of scene” and Hagner notes that v. 8 is the “final verse” for the activity of this section. Keener’s analy-
sis (Keener, 2009: 436) sections all of 16:28-17:13 and offers no linguistic justifica tion for doing so but lines up
the thematic allusions within the text instead. Davies a nd Allison (2004: 68 4, 704-705) tentatively as s ume this to
be the end of a chiasm at v. 8 though offering no linguistic justification.
J. R. Woods
4.2. The Supernatural
This is a part of a more general similarity chain connecting each of the lexical references to the supernatural
elements of the section, see Table 2 below (Louw and Nida, 1996: 33. 4488).
Since a great deal of commentary has typically focused on these elements, a few brief comments about some
of the more popular lexemes identified in this chain are in order: particularly regarding Matthew's use of "high
mountains" (v. 1 ὄρος ὑψηλὸν), and the appearance of "Moses and Elijah" (v. 3). Matthew’s use of “high moun-
tains” for special activities of God is well known, though not every reference to a mountain should be regarded
as programmatic for such activity; it is fr equent and in this case applicable. (Hagner, 1998: 492; Nolland, 2005:
699; Davies and Allison, 2004 : 695). The anaphoric reference to the same mountain in v. 9 ἐκ τοῦ ὄρους is not
included in this chain because the supernatural action has subsided and so their direction is moving away from
the epicenter of that activity. The separation may be regarded as thematic as well as geo-spatial, since demonic
activity is taken up directly subsequent to the exposition of Elijah’s coming (17:14-8).
Regarding the use of the passive in v. 3 to convey the "appearance" (v. 3 ὤφθη) of Moses and Elijah, see
“failure of concord”. (Porter, 1999: 7 5) Newman and Stine offer, “There appeared,” as an acceptable translation.
(So RSV, NRSV, NKJV) Though regarding Moses and Elijah as the actors of this verb (so ESV, LEB, ISV,
NET), they quote the TEV which conveys a very different idea: “the disciples saw Moses and Elijah.” Besides
the cohesive tie with the frequent phrase of Matthew “Law and the Prophets” which Jesus fulfills (cf. 5:17; 7:12),
the supernatural nature of their status also fits the context here (especially mentioned together in this manner)
and this is confirmed by many commentators (France, 2007: 648; Hagner, 1998: 493; Davies and Allison, 2004:
698). While much commentary is devoted to their supposed symbolism, the following linguistic approach will
shed a more well-defined light as to the possible rheotoric behind Matthew referring to their appearance in this
The primary concentration of activity in this chain lies within vv. 2 - 3, and v. 5 which surrounds the interruption
to the scene by the comments of Peter in v. 4. Then it breaks from the narrative until vv. 9 - 13 which is de-
signed to focus solely on an exposition on the arrival of Elijah who is revealed to be John the Baptist in v. 13.
This leaves open the question that the disciples have not asked in connection with the participants in the event.
While they have seen and understood the arrival of Elijah in both the transfiguration vision as well as the con-
temporary role of John the Baptist, the corresponding vision of Moses has remained explicitly untethered to any
equivalent role contemporary to their time and questioning. One can see the artistry implicit in the rhetorical af-
fect behind the text’s cohesive lopsidedness. Thus, even the absence of any explication for the role and appear-
ance of “the vision” having to do with Moses coheres with Jesus’ command for silence. The absence of other
closer referents often reveals more general semantic relations. As asserted here: “the interpretation of items in
the absence of a linguistic referent and/or any situational clues as well as the perception of semantic relation
between un-interpreted implicit devices is made possible because of the third type of tie: that which is based on
co-extension.” Jesus, then, by purposeful omission, could well be added to this list as a co-extensional tie with
Moses (Halliday & Hasan, 1989: 79). In terms of accounting for the linguistic references of this episode, Jesus,
Table 2. Chain for supernatural elements Matt 17:1-13.
v. 1 ὄρος ὑψηλὸν1 ἐπεσκίασεν
v. 2 μετεμορφώθη φωνὴ ἐκ τῆς νεφέλης
ἔλαμψεν ὁ υἱός μου ὁ ἀγαπητός … εὐδόκησα
ἥλιος v. 9 τὸ ὅραμα
λευκὰ ἐκ νεκρῶν ἐγερθῇ
v. 10 Ἠλίαν δεῖ ἐλθεῖν πρῶτον
v. 3 ὤφθη2
v. 11 Ἠλίας μὲν ἔρχεται
Μωϋσῆς κια Ἠλίας3 καὶ ἀποκαταστήσει πάντα
v. 4 κύριε v. 12 Ἠλίας ἤδη ἦλθεν
v. 5 νεφέλη φωτεινὴ v. 13 Ἰωάννου τοῦ βαπτιστοῦ
J. R. Woods
is then co-extensionally related to Moses by the role that he plays, while John the Baptist is explicitly so with
Elijah. Of course, nearly every commentator draws the thematic connection this way, but not from specifically
One might object to the missing reference in this chain to the title “Son of man” which Jesus uses for himself.
The reasons for leaving it out of this chain have to do with the larger chains of discourse within which the two
titles “Son of Man” and “beloved Son” of God are embedded in Matthew’s Gospel. The two titles used of Jesus
create a “paradox” with which the disciples must grapple throughout the narrative. (France, 2007: 652) The
“beloved Son” title already has glory directly associated with God through different scenes (1:18-25; 2:15; 3:17).
(Luz, 2001 : 398) These were “theophanic” episodes that were not difficult to understand. (France, 2007: 12 1; cf.,
Duling, 2012: 84) The title, “Son of Man,” however, this study would argue carries an as yet unfulfilled glory
that relies on the death and resurrection. (Nolland, 2005: 695) Regarding their cohe- sive contributions, the title
“Son of Man” corresponds well with Jesus’ statements about his needing to die, which by no coincidence di-
rectly collocates with both this section and those surrounding. (cf., vv. 16:21; 17:9, 12, 22-23) Hence, it is Jesus,
and not God, who is the one to declare his death. Moreover, it is Jesus’ announce- ment of the “Son of Man’s
coming in his kingdom” which would have challenged the disciples with a mystery in need of resolving (16:28).
This identifies the very cognitive dissonance they associated between the concepts: how the death of their leader
could be construed as bringing about the kingdom of God. This supplies the impe- tus for ordering of the trans-
figuration first with two dead historical figures, Moses and Elijah, and corresponding exposition which harbors
the resuming implicit references to the arrival of the kingdom. The disciples’ persis- tent misunderstanding of
the significance of Jesus’ death prevents further elaboration, until they put all the pieces together at the resurrec-
tion event itself. (Davies & Allison, 2004: 713) Contrastively, the title which God confirms about his “beloved
Son” coheres well with the chains of supernatural activity which the disciples show less strug gle understanding
but about which are commanded to remain silent until they perceive its purpose clearly. (France, 2007: 648)
4.3. The Disciples
This identity chain for the disciples (see Tab l e 3 ) encompasses both the references to “Peter” and “the disciples”
for two reasons. Firstly, Peter is represented as a leading figure for the disciples as a whole. ( Blomberg, 1992:
270; cf., Davies &Allison, 2004: 695, nt. 56. France, 2007 : 645) Secondly, the three participants of the event are
referred within a “collective” reference throughout much of this episode. (Porter, 1999: 131, 134) Peter and the
disciples also have a text exhaustive status in the action. One can perceive the contrast between the types of role
portrayed by this chain and that of Jesus. The examples are numerous. Their passivity into the events con- trasts
Jesus’ leadership. They do no t receive the glistening changes that J esus does when the cloud arrives. The y are
not identified by the voice as is Jesus and they receive commands from both Jesus and the voice from the cloud.
In fact, besides the narrator’s mention of Peter’s ignored comment in v. 4, this is the last time Peter, James, and
John are explicitly named. They only appear by grammatical anaphor, a summary term οἱ μαθ ητ α ὶ, or a collec-
tive pronoun. The lexemes used to record the disciples’ reaction to the event do juxtapose Jesus’ nonchalant
reciprocation with the ancestors on a semantic level, but in terms of narrative continuity is closely connected as
signaled by Matthew’s grammatical usage, as will be demonstrated below. The semantic contrast is seen
Table 3. Chain identifying the disciples through Matt 17:1-13.
v. 1 τὸν Πέτρον
τὸν ἀδελφὸν αὐτοῦ v. 7 αὐτῶν v. 10 ἐπηρώτησαν
αὐτοὺς v. 5 αὐτοῦ ἐγέρθητε οἱ μαθηταὶ
κατʼἰδίαν λαλοῦντος μὴ φοβεῖσθε Λέγοντες
v. 2 ἔμπροσθεν αὐτῶν αὐτούς
ἀκούετε v. 8 ἐπάραντες
αὐτῶν v. 12 ὑμῖν
v. 3 αὐτοῖς v. 13 συνῆκαν
v. 4 ἀποκριθεὶς v. 6 ἀκούσαντες εἶδον οἱ μαθηταὶ
ὁ Πέτρος οἱ μαθηταὶ v. 9 καταβαινόντων αὐτοῖς
J. R. Woods
between the lexemes used to describe their activities. Jesus is recorded as συλλαλοῦντες μετʼ Moses and Elijah,
and this carries “a clear implication as to reciprocal response.” (Louw &Nida 1996: 33.157) In confirmation,
Luke mirrors this mutual reciprocity in his version of the transfiguration which he portrays as everyday conver-
sation (cf., Luke 4:36 ; 20:5; Acts 10:27). Semantically this is a clear contrast to the disciples’ reaction which is
both repeated φοβέομαι (vv. 6-7) and intensified σφόδρα (v. 6) and is usually associated with the fear of death.
(cf., Matt 10:28; Acts 5:26). Grammatically, however, it makes narrative sense to provide continuity between
them in an effort to directly associate the contrast (see next section on conjunctive chain). The only comment
Peter makes receives no acknowledgment and this is contrasted by the discourse of Jesus whose words the dis-
ciples learn to cling to as an explanation which helps them understand a lesson they presumably needed to know.
Clearly, they are there to receive and witness the vision to change their perceived relationship and reaction to-
4.4. Conjunctive Chain
Conjunctions (see Table 4) do more than simply connect individual words. As Porter and Brook O’Donnell
have demonstrated, they help to construct all levels of discourse: word, clause, clause complex, and para- graphs.
(Porter & Broo k O’Donnell, 2007: 9 ) The analysis they offer for the “three axes” of conjunctive function is of-
fered below in pa raphrase.
Three Axes of Conjunctive Function Operate in Discourse.
1) The “vertical” dimension represents clines of discourse (e.g. sentence, paragraph, etc.)
2) Two “horizontal” dimensions represent:
a) continuity/discontin uity
b) logical-semantic relations (Porter &Brook O’Donnell, 2007: 8)
This analysis will be adopted here. The “vertica l” pericope boundaries of this section repres ent the first order
In this regard, the conjunctive chain through this section of Matthew begins and ends the action sequence of
the narrative, i.e. vv. 1 - 8, incorporating the “highest level of discourse continuity” (i.e., και) placed between the
expositional frames on either side, cf., Matt 16:24-28; 17:9-12. (Porter & B rook O’Don n e l l, 2007: 10; cf., Black,
2002: 112) This is for good interpretive reason. The section prior to the start of the action has Jesus elaborate
and enhance the conditions involved with taking up one’s cross, ἀράτω τὸν σταυρὸν αὐτοῦ and following him to
whatever end with the exhortation that their soul is what needs preserving. Having grammaticalized the continu-
ity, it is little surprise that when the scene of the transfiguration arrives, this principle of the soul’s survival is
witnessed in living action. (Porter & Brook O’Donnell, 2007: 10)
The souls of the great prophets Moses and Elijah are seen not only as living despite having died many years
prior but also are seen as engaging fellowship with Jesus. Here lies the motivation behind Matthew’s placement
of the word group καὶ ἰδοὺ in v. 3 because it is an event that calls the readers to take special notice. καί ἰδού
represents an interpersonal theme which “does not add to or restrict the content of what is happ ening in the sen-
tence.” Therefore, Black asserts that its sole reason for inclusion in Matthew’s narrative framework is to com-
municate, “Hey, look! Something important or unexpected is happening here!” (Black, 2002:134-135) This is
also why Peter’s comment is sectioned off with the use of δὲ becaus e Matthew communicates that althoug h it is
Table 4. Conjunctions through Matt 17:1-13*.
v. 1 kαὶ … καὶ … καὶ v. 8 δὲ
v. 2 kαὶ … καὶ … δὲ v. 9 Καὶ καταβαινόντων αὐτῶν ἐκ τοῦ ὄρους
v. 3 καὶ ἰδοὺ … καὶ v. 10 kαὶ … οὖν… ὅτι
v. 4 δὲ … εἰ … καὶ … καὶ v. 11 δὲ … μὲν
v. 5 //ἔτι αὐτοῦ λαλοῦντος ἰδοὺ85 … καὶ ἰδοὺ … //ἀκούετε αὐτοῦ v. 12 δὲ … ὅτι … καὶ … ἀλλʼ… καὶ
v. 6 καὶ … καὶ v. 13 τότε … ὅτι
v. 7 καὶ … καὶ … καὶ
*// = asyndeton italicized = Genitive absolute. A genitive absolute construction in the absence of another conjunction is considered asyndetic. (Black,
J. R. Woods
chronologically alongside the action, it is characteristically out of sync with it. (Louw and Nida, 1996: 89.87;
Black, 2002:154) While Louw and Nida have correctly seen this as in one sense a “marker of a sequence of
closely related events” what is notable is the unusual Predicator-Subj ect (P-S) construction, e specially when the
construction includes a grammaticalized subject when S-P structure is the usual collocation with the δέ conjunc-
tion. (Black, 2002 :155) This is because it is not the presence of Peter that repr esents the discontinuity, since he
was already accessed on either side adjacent to this episode. What directly represents the point of discontinuity
is the reaction which Peter gives toward the event: ἀποκριθεὶς, i.e. Peter showing eagerness to give his “reply”.
(Louw & Nida, 1996 : 33.184) So then Matthew’s communication is two-fold: 1) that Peter’s rea ction is chrono-
logically sequenced alongside the event, and 2) that it is also disruptive to it. The corresponding genitive abso-
lute conveys well the need to squelch Peter’s attempt to show off his veneration. Peter’s response is characte-
rized as “self-assured forwardness.” Hence, Matthew’s motivation for portraying it as a development in response
to, but at discontinuity with the narrative. This then could be the motivation behind the textual variant here at v.
4 which some have in the plural ποιήσωμεν to soften Peter’s remark. The single reading, however, has earned
the “almost certain” judgment by the committee. (Metzger, 1994 : 34). Rather than contrasting the principle that
Matthew previously mentioned, i.e., “repay a man for what he has done” (16:27), it all the more embodies it
with an ironic twist. While Jesus is rewarded with God’s endorsement, Peter is rewarded, as it were, with the
command to listen to Jesus. This is because Jesus has already warned Peter against having his mind set on
earthly things (16:23) and here Matthew shows us that the rebuke from Jesus has not produced the impact that
was intended. Hence, God himself arrives in the cloud in order to override P ter’s attempt to establish an earthly
mark of glo ry to drive into him a silent receptive state.
At this stage of the episode Matthew provides his readers a quotation by God which both identifies Jesus
(οὗτός ἐστιν ὁ υἱός μου ὁ ἀγαπητός ἐν ᾧ εὐδόκησα) and commands the disciples on Jesus’ behalf (ἀκούετε).
The fear they experience proves successful in producing its intended results, i.e. silence. Jesus’ encouragement
is painted as having an “unmarked continuity” in direct response to their fear. Unfortunately, many translations
have transposed the semantic relations between the disciples’ fear and Jesus reaction into the grammar by
choosing to place a contrastive English connective but, cf., ESV, ISV, NIV, NKJV, RSV, NRSV, TNIV. The
contrast lies in the semantics of the lexemes rather than the grammar which Matthew chooses to cohere them
(Porter & Brook O’Donnell 2007: 11; cf., also Black, 2002: 137-140) The practice of distinguishing between
conjunctive meaning and semantic context must be maintained. Other translations keep the continuity, such as
the ASV, LEB, NASB95 which all translate “And.” The NLT, however, cleverly uses the conjunction “then”
which may more properly signal the sort of immediate reactionary connection intended. The best translation
strategy may lie with finding the intersection between narrative continuity and immediate personal agency using
either “and then,” or “and so” since “and” communicates the continuity, while either “then” or “so” communi-
cates the latter by either chronological or causal means.
Note Matthew’s continuous structure: v. 6 καὶ ἐφοβήθησαν σφόδρα v. 7 καὶ προσῆλθεν… καὶ ἁψάμενος be-
sides being assigned a different verse number, the Greek text shows no grammatical break between these events.
Where the connective “mid-level” discontinuity appears is in the situation that follows (v. 8 ἐπάραντες δὲ).
(Black, 2002: 153) This grammatical strategy communicates another development that the disciples’ (now cor-
porately referenced) reception of Jesus’ exhortation “not to fear” is no longer met with either insolent resistance
as the episode prior, or concern for earthly glory like Peter’s most recent response, but now at long last it is met
with both silent watchfulness, i.e., τοὺς ὀφθαλμοὺς αὐτῶν… εἶδον, and reciprocal obedience, i.e., Καὶ
καταβαινόν των αὐτῶν.
Having learned their lesson, Matthew proceeds with Jesus’ projected frames to culminate the episode using
two additional discontinuities within the conversation. The first one is low- level because the appearance in v. 11
has many expected features. Firstly, it utilizes an expected S-P constituent structure with a grammaticalized
subject, i.e., ὁ. Secondly, as is his usual collocation with δὲ, Matthew pairs it with an aorist verb-form
ἀποκριθεὶς that functions as the “background” choice for narrative. (Black, 2002: 117, 154; Porter, 1999: 23,
302) Lastly, much of what is fronted in the projected clause is not new information, but in fact repeats what the
disciples have already asked for: that Elijah must come first. In setting up the coming mid-level discontinuity,
however, the statement does instantiate three elements that leave the reader in expectation. The inclusion of μὲν
which anticipates a corresponding δέ statement is the first clue. (Porter, 1999: 112; Black, 2002: 160) Next is
changing the disciples’ verb form from a background tense (ἐλθεῖν) to a “foreground” one (ἔρχεται) and finally
the future form of the newly submitted ἀποκαταστήσει whose tense form semantically conveys “a higher degree
J. R. Woods
of expectation for fulfillment.” (Porter, 1999: 45). The mid-level discontinuity is also what Matthew has Jesus
use in his next statement: the pairing of a present tense form which itself is “foreground” tense (λέγω) alongside
a nongrammaticalized subject which is an unusual collocation with δὲ. Moreover, the statement includes many
new items which, in an unexpected turn, are then directly connected to Jesus’ fate (Black 2002: 173).
This transition comes by an unexpectedly clever switch of actor and process. The fronted participant, Jesus, is
the actual recipient of the suffering. There are several reasons discussed in the next section that point to why this
may be considered the culminating issue of the section. While using verbs that keep him in the active voice, the
actual inflictors of the suffering are only collectively referenced (αὐτῶν) (Newman and Stine, 1992: 543; Porter
et al., 2006: Matt 17:12). Conjunctions help unveil the rhetoric involved here. The entire statement is a compar-
ative frame reflected by οὕτως καὶ that comes off the heels of a projected statement using the “adversative”
conjunction ἀλλά to polarize οὐκ ἐπέγνωσαν and ἐποίησαν ἐν αὐτῷ so as to convict the guilt of the unnamed
actors (Porter, 1999: 23, 205; Eriksson, 2002: 336).
4.5. Aspectual Chain
The markedness paradigm used by Porter helps solidify the cohesive flow of the section boundaries through as-
pectual contouring (Porter, 1999: 23, 303, 305; Por ter and Broo k O’Donnell, 2007: 8; Comrie, 1976: ch. 6; Bas-
set, 2009: 205-220). See Table 5 below. This section identifies the prominent sections of discourse as signaled
first according to aspectual theory, and then analyze how that section utilizes collocating features alongside as-
pect to accomplish this. Imperfective aspect according to Porter is realized in the Greek through two verbal tense
forms: the present and the imperfect. ειμι verbs are left out due to their “aspectually vague” status. Porter
stresses this point: “Aspectually vague verbs may be used in any verbal context since they do not carry the se-
mantic weight of perfective, imperfective or stative verbal aspect.” (Porter, 1999: 25; cf., Porter, 1989: ch. 10)
emphasis original ).
From an aspectual perspective, both v. 1 and v. 9 signal between “planes of discourse” which Porter has iden-
tified as one function of verbal aspect (Porter, 1999: 23). Each of these verses utilizes their own scene initiations
through various means. Verse 1, for example, reinforces the aforementioned “point of departure” (Καὶ μεθʼ
ἡμέρας ἓξ) by collocating the discontinuity with two finite imperfective verbs (παραλαμβάνει and ἀναφέρει).
This foregrounds Jesus leading his disciples to a new location: εἰς ὄρος ὑψηλὸν (Porte r, 1999 : 23). Matt hew has
chosen a similar grammatical sequence for the break in v. 9 as well. Here he has chosen two imperfective as-
pectual forms - καταβαινόντων and λέγων - the former to initiate the spatial discontinuity, and the latter the
conversa t ional one.
Moving deeper into the scene, two prominent “discourse peaks” materialize on the foundation of aspectual
dis- tinctions, which fence off a “domain of prominence” at verses 5 and 12, the former an “action peak,” the
latter a “didactic peak.” (Reed, 1995: 81; Porter, 1999: 302 ; Longacre, 1999: 142, 156) . Imperfective aspect ap-
pears from the beginning λαλοῦντος, middle λέγουσα, and end ἀκούετε of v. 5. This assures that every stage
leading into this point of the narrative is received as cohesively bound within “foreground” material (Porter,
1999: 302). The next marked dimension to notice is the fact no participant at this point in the narrative receives
explicit grammaticalization, though every one of the active participants for this episode is addressed. Indeed,
apart from any presupposition, this entire verse would remain noninterpretable. This creates the highest level of
Table 5. Imperfective aspectual chain through Matt 17:1-13 (Porter, 1999: 21; idem, 1989: 92-93).
v. 1 παραλαμβάνει v. 9 καταβαινόντων
v. 3 συλλαλοῦντες v. 10 λέγοντες
v. 4 θέλεις λέγουσιν
v. 5 λαλοῦντος v. 11 ἔρχεται
λέγουσα v. 12 λέγω
ἀκούετε μέλλει πάσχειν
v. 7 μὴφοβεῖσθε
J. R. Woods
participant density in the entire episode and might be characterized as a kind of “4-way-stop” between all the
chains discussed: Jesus, the supernatural, the disciples, and conjunctions.
I) Jesus represents the entire quoted material (οὗτός… ὁ υἱός… ᾧ… αὐτοῦ)
II) Peter and the disciples begin and end the whole verse (αὐτοῦ … ἀκούετε)
III) God the Fa t he r is re ferred to in a variet y of ways
a) the cloud νεφέλη (twice)
b) the voice φωνὴ
c) the possessive pronoun μου
d) the “monolectic” verb εὐδόκησα (Porter, 1999: 30 2; cf., Black, 20 02: 118)
e) implicitly from the imperative ἀκούετε
IV) Three internal instances of conjunction
a) Two instances of asyndeton, the first at ἔτι αὐτοῦ λαλοῦντος (adjacent to ιδου) and the second at ἀκούετε.
b) One instance of καὶ ιδου
The narrative returns to a chain of backgrounding aorist forms only to break at v. 7 with “have no fear” (μὴ
φοβεῖσθε). Verse 8 concludes the action using Jesus within the complementing structure aforementioned.
The didactic peak of the section realizes similar climactic features to that of the action peak. Matthew fully
prepares his audience for it. For example, the narrative has already changed scenery (καταβαινόντων… ἐκ τοῦ
ὄρους) with the return to background aspectual forms (ἐνετείλατο, εἴπητε, ἐγερθῇ). Then the disciples’ question
supplies Jesus with the material about Elijah which he ties back to and uses a μὲν … δὲ format that concludes by
culminating many cohesive ties with a dual foregrounding verbal aspect (μέλλει πάσχειν). Notice once again, the
full assumption of participants in the final clause of v. 12 similar to the discourse strategy of the action peak
(Porter et al., 2006: Matt 17:12).
… οὕτως καὶ ὁ υἱὸς τοῦ ἀνθρώ που μέλλει πάσχειν ὑπʼ αὐτῶν. (NA28)
The fronted term οὕτως assumes the entire previous statement of what happened to the Elijah figure which
καὶ “emphasizes.” (Louw & Nida, 1996: pp. 89, 93). Furthermore, the phrase ὁ υἱὸς τοῦ ἀνθρώ που cohesively
ties back to v. 6 modifying the syntax for the paradigmatic changes. The structure maps out the interpretive na-
ture of the cohesive ties well. See Figure 4 below.
-ὁ υἱὸς τοῦ ἀνθρώπου sets lexical repetition to signal the connection to the precedented S-A-P structure.
-μέλλει πάσχειν uses double-foregrounding in an overturn of the expected structural precedent which
rightfully served as a backgrounded clause. It reinterprets into an active form for the death (A: ἐκ νεκρῶν ) that
he must suffer (P: μέλλει πάσχειν)
-ὑπʼ αὐτῶν recontextualizes the necessitated death in light of the new participants which are here assumed (οἱ
γραμματεῖς) as the ones causing the suffering. It also effectively contrasts the grammatically assumed agents of
both actions by what they cause: the first as God who gives life by raising (ἐγερθῇ), the second as the scribes
who take life through causing suffering (πάσχειν).
-the syntagmatic switch not only justifies the strong foregrounding, but aids the audience to process the final-
ity of the expository sectio n by recalling the same syntagmatic switching s trategy witnessed with the activity of
the episode. (vv. 1, 8) This strategy is further corroborated by the same conjunction δε which is used to finalize
each section. (cf., vv. 8, 12) This strategy of syntagmatic modification may be the impetus behind commentators’
mistaken need to make a “chiasm” out of the material (Hagner, 1998: p. 491; Davies &Alli son, 2004: p. 684).
Figure 4. Structure Mapping Cohesion: parallel linguistic structure helps
interpret cohesive ties (Porter et al., 2006: Matt 17:9, 12).
J. R. Woods
It is only here at v. 13 that the distinctive Matthean way of “introducing the concluding … speech to which an
episode has been building up” is used, τοτε (Black, 2002: p. 247; cf., Porter, 19 99: p. 217).
There Mathew’s episode uses John the Baptist to close out both as a semantic tie for the expository section
regarding Elijah’s role, as well as a textual connection back to v. 1 differentiating between John the disciple and
John the Baptist, while leaving his readers to conclude the unspo- ken rhetorical connection between Jesus and
Moses as that connection to which all of the action and teaching seeks to demonstrate and cohere with the com-
mand to silence until the resurre ction.
The linguistic stud y from a SFL perspective of cohesive ties and their respective chaining techniques represents
a fundamental but singular component of a larger, fuller discourse analysis. This study has attempted to demon-
strate the immanent value of using rigorous linguistic methodology tow ard the interpretation of scripture as any
other human communication—as a register of speech which the author has used to reflect the situation within
the context of that cultur e. Theological commentary simply will not deliver th is. And the days of linguistic am-
bivalence have passed. This study has demonstrated the exegetical and interpretive benefits of committing to a
specific linguistic framework. It follows a precedent set by modern biblical scholars for the methods required in
comparative studies in the future, both inside and outside the biblical material, paving the way to unfold the
greater context in which these documents are embedded.
Basset, L. (2009). The Use of the Imperfect to Express Completed States of Affairs: The Imperfect as a Marker of Narrative
Cohesion. In S. Bakker, & G. Wakker (Eds.), Discourse Cohesion in Ancient Greek (pp. 205-220). Leiden: Brill.
Black, S.L. (2002). Sentence Conjunctions in the Gospel of Matthew: Kai, de, Tote, Gar, Syn and Asyndeton in Narrative
Discourse (Volume 216). London: Sheffield Academic.
Blomberg, C. (1992). Matthew. NAC 22. Nashville: Broadman & Holman.
Brown, A. (1989). Piggybook. London: Little Mammoth.
Comrie, B. (1976). Aspect (Cambridge Textbooks in Linguistics). Cambridge: Cambridge University.
Davies, W. D., and Allison Jr., D . C. (2004). A Critical and Exegetical Commentary on the Gospel According to Saint Mat-
thew (Volume 2). ICC. New York: T&T Clark.
Duling, D.C. (2012). A Marginal Scribe: Studies of the Gospel of Matthew in Social-Scientific Perspective (Matrix: The Bi-
ble in Mediterranean Context 7). Eugene, OR: Cascade Books.
Eriksson, Anders. (2002). Contrary Arguments in Paul’s Letters. In S. E. Porter, & D. Stamps (Eds.), Rhetorical Criticism
and the Bible (pp. 336-354). JSNTS 195. London: Sheffield Academic.
Fawcett, R. P. (2000). A Theory of Syntax for Systemic Functional Linguistics. Amsterdam: Benjamins.
France, R. T. (2007). The Gospel of Matthew (NICNT). Grand Rapids, MI: Wm. B. Eerdmans.
Gledhill, C. (2009). Colligation and the Cohesive Function of Present and Past Tense in the Scientific Research Article. In D.
Banks (Ed.), Les Temps et les Textes de Specialite (pp. 65-84). Paris: L’Harmattan.
Gregory, M. (1988). Generic Situation and Register: A Functional View of Communication. In J. D. Benson, & W. S.
Greaves (Eds.), Linguistics in a Systemic Perspective (pp. 301-330). Amsterdam: John Benjamins.
Hagner, D. A. (1998). Matthew 14-28. WBC 33B, Dallas: Word.
Halliday, M. A. K. (1976). Anti-Languages. American Anthropologist, 78, 570-584.
Halliday, M. A. K. (1985). An Introduction to Functional Grammar (3rd ed.). London: E. Arnold.
Halliday, M. A. K. (2007). Language and Society. London: Continuum.
Halliday, M. A. K. (1978). Language as Social Semiotic: The Social Interpretation of Language and Meaning. London: Ed-
Halliday, M. A. K., & Matthiessen , C. M. I. M. (2006). Construing Experience through Meaning: A Language-Based Ap-
proach to Cognition. Open Linguistics Series, London: Continuum.
J. R. Woods
Halliday, M. A. K., & Ha san, R. (1976). Cohesion in English. English Language Series, London: Longman.
Halliday, M. A. K., & Hasan, R. (1989). Language, Context, and Text: Aspects of Language in a Social-Semiotic Perspective
(2nd ed.). Oxford: Oxford University.
Hasan, R. (2009). The Place of Context in a Systemic Functional Model. In J. Webster, & M. A. K. Halliday (Eds.), Conti-
nuum Companion to Systemic Functional Linguistics (pp. 166-189). London: Continuum.
Hasan, R. (1995). The Conception of Context in Text. In P. H. Fries, & M. Gregory (Eds.), Discourse in Society: Systemic
Functional Perspectives (pp. 183-284). Norwood, NJ: Ablex. http://dx.doi.org/10.1075/cilt.118
Hudson, R. A. (2001). Sociolinguistics (2nd ed. ). Ca mbridge Textbooks in Linguistics, Cambridge: Cambridge University.
Keener, C. S. (2009). The Gospel of Matthew: A Socio-Rhetorical Commentary. Grand Rapids, MI: Wm. B. Eerdmans.
Kress, G. (1989). Linguistic Processes in Sociocultural Practice. Oxford: Oxford University.
Levinsohn, S. H. (1995). A Discourse Study of Constituent Order and the Article in Philippians. In S. E. Porter, & D. A.
Carson (Eds.), Discourse Analysis and Other Topics in Biblical Greek (pp. 60-74). JSNTS 113, Sheffield: Sheffield Aca-
Longacre, R. E. (1999). A Top-Down, Template-Driven Narrative Analysis, Illustrated by Application to Mark’s Gospel. In
S. E. Porter, & J. T. Reed (Eds.), Discourse Analysis and the New Testament: Approaches and Results (pp. 140-196).
JSNTS 170, Sheffield: Sheffield Academic.
Louw, J. P., & Nida, E. A. (1996). Greek-English Lexicon of the New Testament: Based on Semantic Domains (2 Vols.).
New York: United Bible Societies.
Lukin, A., Moore, A. R., Herke, M., Wegener, R., & Wu, C. Z. (2011). Halliday’s Model of Register Revisited and Explored.
Linguistics and the Human Sciences, 4, 187-213. http://dx.doi.org/10.1558/lhs.v4i2.187
Luz, U. (2001). Matthew: A Commentary. Minneapolis, MN: Augsburg.
Martin, J. R. (2001). Cohesion and Texture. In D. Schiffrin, D. Tannen, & H. E. Hamilton (Eds.), The Handbook of Dis-
course Analysis (pp. 35-53). Malden, MA: Blackwell.
Martin, J. R. (1992). English Text: System and Structure. Amsterdam: Benjamins. http://dx.doi.org/10.1075/z.59
Martin-Asensio, G. (2000). Transitivity-Based Foregrounding in the Acts of the Apostles: A Functional-Grammatical Ap-
proach to the Lukan Perspective. JSNTS 202, Sheffield: Sheffield Academic.
Martin, J. R., & Rose, D. (2007). Working with Discourse: Meaning Beyond the Clause (2nd e d.). Open Linguistics Series,
Meeks, W. (1986). The Moral World of the First Christians. Library of E arly Christianity 6, Philadelphia, PA: Westminster.
Metzger, B. M. (1994). A Textual Commentary on the Greek New Testament (2nd ed.). a Companion Volume to the United
Bible Societies’ Greek New Testament (4th Rev. Ed.). New York: United Bible Societies.
Morley, J. (2009). Lexical Cohesion and Rhetorical Structure. In J. Flowerdew, & M. Mahlberg (Eds.), Lexical Cohesion and
Corpus Linguistics (pp. 5-22). Amsterdam: John Benjamins. http://dx.doi.org/10.1075/bct.17.02mor
Newman, B. M., & Stine, P. C. (1992). A Handbook on the Gospel of Matthew. UBS Handbook Series, New York: United
Nolland, J. (2005). The Gospel of Matthew: A Commentary on the Greek Text. NIGTC, Grand Rapids, MI: W.B. Eerdmans.
O’Donnell, M. B. (1999). Linguistic Fingerprints or Style by Numbers? The Use of Statistics in the Discussion of Author-
ship of New Testament Documents. In S. E. Porter, & D. A. Carson (Eds.), Linguistics and the New Testament: Critical
Junctures (pp. 206-255). JSNTS 168, Sheffield: Sheffield Academic.
Porter, S. E. (1999). Idioms of the Greek New Testame n t. Sheffield: JSOT.
Porter, S. E. (2004). The Criteria for Authenticity in Historical-Jesus Research: Previous Discussion and New Proposals.
JSNTS 191, London: T & T Clark.
Porter, S. E. (1989). Verbal Aspect in the Greek of the New Testament with Reference to Tense and Mood. New York: Peter
Porter, S. E., & O’Donnell, M. B. (2007). Conjunctions, Clines and Levels of Discourse. Filologia Neotestamentaria, 20, 3-
Porter, S. E., O’Donnell, M. B., & Reed, J. (2006). The OpenText.org Syntactically Analyzed Greek New Testament. Bel-
lingham, WA: Logos Bible Software.
Puigdollers, A. R. R. (2009). The Particles ΑΥ and ΑΥΤΕ in Ancient Greek as Topicalizing Devices. In S. Bakker, & G.
Wakker (Eds.), Discourse Cohesion in Ancient Greek (pp. 83-110). Leiden: Brill.
Reed, J. T. (1999). The Cohesiveness of Discourse: Towards a Model of Linguistic Criteria for Analyzing New Testament
Discourse. In S . E. Porter, & J. T. Reed (Eds.), Discourse Analysis and the New Testament: Approaches and Results (pp.
28-46). JSNTS 170, Sheffield: Sheffield Academic.
J. R. Woods
Reed, J. T. (1997). A Discourse Analysis of Philippians: Method and Rhetoric in the Debate over Literary Integrity. Shef-
field: Sheffield Academic.
Reed, J. T. (1997). Discourse Analysis. In S. E. Porter (Ed.), Handbook to Exegesis of the New Testament (pp. 188-218).
New Testament Tools and Studies 25, New York: Brill.
Reed, J. T. (1995). Identifying Theme in the New Testament: Insights from Discourse Analysis. In S. E. Porter, & D. A.
Carson (Eds.), Discourse Analysis and Other Topics in Biblical Greek (pp. 75-101). JSNTS 113, Sheffield: Sheffield
Reed, J. T. (2000). Language of Change and the Changing of Language: A Sociolinguistic Approach to Pauline Discourse. In
S. E. Porter (Ed.), Diglossia and Other Topics in New Testament Linguistics (pp. 121-53). JSNTS 193, Sheffield: Shef-
Rohrbaugh, R. L. (2007). The New Testament in Cross-Cultural Perspective/Matrix: The Bible in Mediterranean Context 1.
Eugene, OR: Cascade Books.
Schiffrin, D. (1994). Approaches to Discourse. Blackwell Textbooks in Linguistics, Oxford: Blackwell.
Schmidt, F. (2001). How the Temple Thinks: Identity and Social Cohesion in Ancient Judaism. Sheffield: Sheffield Academ-
Stoddard, S. (1991). Text and Texture: Patterns of Cohesion. Norwood, NJ: Ablex.
Stubbs, M. (1983). Discourse Analysis: The Sociolinguistic Analysis of Natural Language. Language in Society 4, Chicago,
IL: University of Chicago.
Teich, E. (2009). Linguistic Computing. In J. Webster, & M. A. K. Halliday (Eds.), Continuum Companion to Systemic
Functional Linguistics (pp. 113-127). London: Continuum.
Tucker, G. H. (1998). The Lexicogrammar of Adjectives: A Systemic Functional Approach to Lexis. Functional Descriptions
of Language Series, London: Cassell.
Van Neste, R. (2004). Cohesion and Structure in the Pastoral Epistles. JSNTS 280, London: T & T Clark International.
Woods, J. (2013). Antilanguage in the Synoptic Gospels: A Sociolinguistic Inquiry. Master’s Thesis, Virginia Beach, VA: