Grace
Theological Journal 11.2 (1990) 205-220
Copyright © 1981 by Grace
Theological Seminary. Cited with permission.
A DISCOURSE ANALYSIS OF
COLOSSIANS 2:16-3:11
GREGORY T. CHRISTOPHER
A discourse analysis of Colossians
2:16-3:17 has led to three
conclusions. First, this section forms a discourse
unit (specifically,
hortatory discourse with embedded expository
discourse). Second, the
structural framework is a chiasmus. Its functions are
to provide the
structural rubric around which the argument develops
and to provide
cohesion which holds the book together. And third,
the argument of
this section builds to a climax, identified
with the imperative, "Put on "
(
ated with
the imperatives, change in word order, and use / nonuse of the
vocative) point to these conclusions.
* * *
INTRODUCTION
Jan de Waard and Eugene Nida note that translation is essentially
interpretation.l Translators, and
by extension, interpreters, should not
only be
concerned about content, but concern should also extend to
rhetorical impact
and appeal and to rhetorical structure and meaning.
Translators/interpreters must recognize patterns of selection
and ar-
rangement. Such
concerns go beyond sentence level syntax, in that
rhetorical
structures are normally large patterns and less rigidly rule-
governed. These
structures are features of discourse.2
The focus of this paper is the discourse structure of Colossians
purpose. The
chiasmus provides the cohesion which holds the book
together. And
it provides the structural rubric around which the argu-
ment of
Colossians 2:16-3:17 develops to a climax.
1 I would like to thank Robert Longacre and Daniel Wallace for their critical
comments. The content of this paper, however, is the author's
responsibility.
2 Jan de Waard
and Eugene A. Nida, From One Language to Another (
Nelson, 1987) 40, 79.80.
206 GRACE
THEOLOGICAL JOURNAL
Before
outlining the chiastic structure, the method which under-
lies this
paper is summarized. The basis for the claims is discourse
grammar. 3
METHODOLOGICAL
OVERVIEW
In a paper
entitled, "Why We Need a Vertical Revolution in
Linguistics," Robert Longacre
outlines features of language for which
sentence level
grammars are unable to account.4 Discourse grammar-
ians claim
that sentence level grammars are unable to describe informa-
tion
structure, and therefore, are unable to fully account for the
dynamics of
language.5 Part
of the problem is that English speakers are
generally
unaware of discourse features.6
Information
structure consists of several levels. For the purposes
of this
paper, two levels are outlined. The first level is main-line verbs
of
discourse. The second level is discourse peak.
Main-line
Verbs of Discourse
Not all
information is of equal value. Some information is salient
and
carries the discourse forward to its climax, while other information
is
supportive. This distinction is marked by a specific tense-aspect-
mood.
Salient information is marked by a main-line verb while suppor-
tive
information is marked by verbs which progressively depart from
the
main-line. Longacre explains:
3 0ne should not equate discourse
grammar with structuralism, though some overlap
might exist. Daniel Patte illustrates the
latter ("Method for a Structural Exegesis of
Didactic
Discourse: Analysis of I Thessalonians," Semeia 26 [1983] 85-129). For a
critique of structuralism, see Bill Stancil,
"Structuralism and New Testament Studies,"
SWJT 22 (1980) 41-59. Longacre's discourse grammar is independent of (European)
structuralism (personal communication).
4 Robert Longacre,
"Why We Need a Vertical Revolution in Linguistics," The Fifth
LACUS Forum (Columbia, SC: Hornbeam, 1978)
247-70.
s Take, for example, the work of Henk van Riemsdijk and Edwin
Williams. Their
research focus are sentences. At the same time they recognize
limitations to such a
research program. They concede, "In principle it could turn out
that it is impossible to
characterize sentences in and of themselves without reference to their roles
in various
conversations" (Introduction to
the Theory of Grammar [Cambridge: MIT, 1986] 184).
6 James Gee notes that
"English is particularly impoverished in discourse particles
and other
formal discourse markers" ("Units in the Production of Narrative
Discourse,"
Discourse Processes 9 [1983] 392). This puts speakers of English
at a disadvantage
vis-á-vis discourse analysis, in that we
are unaware of such features and consequently fail
to search
for them in language analysis. Discourse analysts such as Longacre
correct this
disadvantage. Their exposure to languages makes them aware of language
dynamics in
general and discourse strategies in particular.
CHRISTOPHER: DISCOURSE ANALYSIS 207
Discourse
grammarians are coming to recognize more and more that in
telling a story in any language, one particular tense is
favored as the
carrier of the backbone or story-line of the story while
other tenses serve
to present the background, supportive, and predictive
material in the
story. 7
To illustrate
these distinctions, note Longacre's discussion of
Hebrew narrative and hortatory
discourse. Each discourse type has its
own
constellation of verb forms (what Longacre calls
salience scheme).
Chart 1 illustrates Hebrew narrative, while Chart 2 illustrates
Hebrew
hortatory
discourse.8
The preterite (waw-consecutive) marks the main-line in Hebrew
narrative, a
chain of (necessary verb-initial clauses). Supportive infor-
mation is
scalar, moving from action (Band 2), to static verbs (Band 4),
to irrealis (Band 5).
__________________________________________________________________
Band 1 1. Preterite
I
Storyline
__________________________________________________________________
Band 2 2.1 Perfect
Backgrounded 2.2
Noun + Perfect
Actions
__________________________________________________________________
Band 3 3.1 hinnen + participle
Backgrounded 3.2 Participle
Activities 3.3 Noun + participle
__________________________________________________________________
Band 4 4.1 Preterite
of haya,
"be"
Setting 4.2
Perfect of haya,
"be"
4.3 Nominal clause (verbless)
4.4 Existential clause with yesh
__________________________________________________________________
Band 5 5.
Negation of verb clause
Irrealis
__________________________________________________________________
Chart
1
Hebrew Verb
Rank Scheme for Narrative Discourse
Used with permission
Note the
differences between narrative and hortatory discourse
(see Chart 2). Preterite
marks main-line of narrative while the main-
line of
hortatory discourse is the imperative. Supportive information is
also
scalar.
7Robert Longacre, Joseph: A Story of Divine
brauns) 64.
8Longacre, Joseph, 81, 121.
208 GRACE
THEOLOGICAL JOURNAL
__________________________________________________________________
Band I 1.1
Imperative (2pl)
Primary line of 1.2
Cohortative (1 pl)
Exhortation 1.3
Jussive (3pl)
__________________________________________________________________
Band 2 2.1 'al
+ jussive/imperfect
Secondary line 2.2 Modal imperfect
of
exhortation
__________________________________________________________________
Band 3 3.1 waw-(consecutive) perfect
Results/consequences 3.2 lo'/pen
+ imperfect
(Motivation) 3.3 (Future) perfect
__________________________________________________________________
Band 4
4.1 Perfect (of past events)
Setting (or problem) 4.2 Participles
4.3 Nominal clauses
__________________________________________________________________
Chart
2
Hebrew Verb
Rank Scheme for Hortatory Discourse
Used with permission
Identification
of main-line verbs is important. First, main-line
verbs are
textual clues as to salient versus supportive information.
Main-line verbs carry the discourse forward. Second, main-line
versus
supportive verbs
serve to identify paragraph structure.9 Third, altering
the
main-line can mark discourse peak or climax (see next section).
And fourth, abstraction of macrostructure is related to verb
ranking.
Macrostructure is essentially the overall plan or design of a
discourse. 10
The second
level of information relative to this paper is discourse
peak. The
assumption here is that a discourse “is going somewhere in
terms of its
inner drive and development.” 11 There is movement toward
a
conclusion. An interpreter's goal, then, is to retrace an author's
progression of
thought (being conscious of the main-line), a progres-
sion which
builds to the conclusion or discourse peak. In narrative, for
example, peak
may be maximum tension (climax) or a crucial event
that
provides a resolution to a plot (denouement).
The peak of horta-
tory
discourse is the most effective attempt to change behavior.
Longacre calls
this progression the profile of a text, i.e., linguistic
reflexes of
mounting and declining tension. 12
9For a complete discussion, see Longacre, Joseph,
83-118.
10Longacre, Joseph,
42. See
also Teun van Dijk, Text and Context (
Longmans, 1977).
11Robert Longacre, "
(Ed. J. Wirth; Ann Arbor: Karoma,
1985) 84.
12Robert Longacre, "A Spectrum and Profile Approach to
Discourse Analysis,"
Text 1 (1981) 337.
CHRISTOPHER: DISCOURSE ANALYSIS 209
Longacre describes discourse peak as a "zone of
turbulence in
otherwise placid
flow of discourse." 13 Changes from the "routine"
within a
discourse serve as cues to mark the progression of a discourse.
Longacre
explains that languages possess a number of possible strate-
gies to
mark discourse peak. These strategies vary across languages.
Essentially the regular flow of the discourse is altered at
peak. A given
discourse may
employ one or several strategies. Such strategies may
include
rhetorical underlying (e.g., paraphrase), change in word order,
discourse peak
particle, change of tense-aspect-mood, change of sen-
tence
length, al.14 The
underlying assumption is that variation is not
random nor
arbitrary.15
The
identification of discourse peak is important. Longacre
explains:
The importance
of the identification of peak is that it enables us to get at
the overall grammar of the discourse. If we can identify a
discourse
peak, then we can identify pre-peak and post-peak sections.
These, plus
special beginning and ending sections, give us a surface
grammar of
discourse that is not dissimilar from the recognition of
subject, verb, and
object on the clause level in a language. The verb as a
central constituent
of the clause can be compared with the peak as a central
constituent of
the discourse.16
A critical part
of understanding a discourse (or a section of a
discourse),
then, is the identification of the conclusion or discourse
peak. This
paper will illustrate the importance of this concept in terms
of
Colossians
Summary
The method which underlies the interpretive conclusions of this
paper is
discourse grammar as developed by Robert Longacre.
The
13Longacre,
"A Spectrum and Profile," 351.
14Longacre, "
Profile," 349-51.
15Longacre, Joseph , xiii.
16Longacre,
Joseph, 97. Paul Ricoeur's
comment about the conclusion of a story is
relative here. He writes: “To follow a story is to move forward in the
midst of contin-
gencies and peripeteai
under the guidance of an expectation that finds its fulfillment in
the
'conclusion' of the story. This conclusion is not logically implied by some
previous
premises. It gives the story an 'end point,' which, in turn, furnishes
the point of view from
which the story can be perceived as forming a whole. To understand
the story is to
understand how and why the
successive episodes led to this conclusion, which far from
being foreseeable, must finally be acceptable, as congruent with the
episodes brought
together by the story" (emphasis mine; Time and Narrative [3 vols;
discussion of emplotment. Essentially, emplotment is a way to describe the organization
210 GRACE
THEOLOGICAL JOURNAL
goal of
this method is to identify the information structure of a
discourse. For
the purpose of this paper, two levels were introduced,
main-line verbs
and discourse peak. An application of this method to
Colossians 2:16-17 follows.
DISCOURSE STRUCTURE OF COLOSSIANS
2:16-3:11
A discourse analysis of Colossians 2:16-3:17 (hereafter central
section)
suggests that this portion of Colossians is hortatory discourse
with
embedded exposition. The structural framework is a simple
chiasmus. It
provides the cohesion that holds the book together, and it
provides the
rubric in which the central section progresses to its dis-
course peak
or climax.
Before
outlining the chiasmus, I shall first define and illustrate this
structure within
the broader context of NT interpretation.
Definition
As commonly
accepted chiasmus is inverted parallelism.17 The
interior
consists of either a single element (e.g., C) or two comple-
mentary
elements (e.g., B B'). The exterior
consists of pairs of com-
plementary
elements forming a composite meaning (e.g., A A'). These
are
illustrated in Figure 1.
A B A B
C
B'
A' B' A'
FIGURE
1. Chiastic Structures
Colossians 1:2 is an example of a simple chiasmus. The interior ele-
ments are
"saints" and "faithful brethren." The exterior elements are
two
prepositional phrases, "in Colossea" and
"in Christ."18 Note the
following
figure.
A in
B
saints
B'
faithful brethren
A' in Christ
FIGURE 2.
Chiastic Structure of Colossians 1:2
__________________________________________________________________
of a
discourse, a concept which parallels Longacre's
macrostructure. Because of em-
plotment an interpreter is able to
"follow" a discourse from its beginning to the climax
(Time and Narrative, 1:64-10).
17de Waard and Nida, From One
Language to Another, 112-20.
18The English translation
is "to the saints in Colossea and to the
faithful brethren in
Christ"
(author's translation). The chiasmus is lost in the English.
CHRISTOPHER: DISCOURSE ANALYSIS 211
Chiasmus
and NT Interpretation
Blass-Debrunner note that the
identification of chiasmus in the
NT is controversial.19 Though chiastic
structures are more readily
associated with
the OT ,20 NT
studies have begun to recognize their
presence and
potential for interpretation.21 Some have taken a progres-
sive
position.22
Several have
identified chiastic structures over larger sections of
text. M.
Philip Scott, for example, suggests that a chiasmus is a key to
interpreting Mark's
Gospel.23 George
Rice identifies a chiasmus as the
central
section of Hebrews.24 More
germane to this paper, Steven M.
Baugh suggests that the hymn of Colossians 1:15-20 is a
chiasmus.25
Identification
of Chiastic Structure
A chiasmus marks the central section of Colossians (see
Figure 3).
The chiasmus provides the
cohesion which ties together the two halves
of the
book and provides the rubric around which the argument of the
central
section develops to a climax. I will now summarize my interpre-
tation (alternative positions are
cited in the notes).
First,
is the
shift of tense-aspect-mood and word order (see Figure 3). The
19F.Blass and A. Debrunner, A Greek Grammar of the
New Testament and Other
Early Christian Literature (Ed. Robert Funk; Chicago: The
1961) 252.
20de Waard and Nida, From One
Language to Another, 112-20.
21Ronald Man, "The
Value of Chiasm for New Testament Interpretation," BSac
141:146-57.
22John Welch suggests
that the issue is no longer whether chiasmus exists, but rather
contends that research should focus upon (1) frequency of occurrence and
(2) the
structure's significance for exegesis (Chiasmus
in Antiquity [
1981] 9). John Breck suggests that an intimate
connection exists between rhetorical form
fand thematic context
("Biblical Chiasmus: Exploring Structure for Meaning," BTB 17
[1987]
70-74.
23M. Philip Scott,
"Chiastic Structure: A Key to the Interpretation of Mark's
Gospel." BTB 15 (1985) 17-26.
24George Rice, "The
Chiastic Structure of the Central Section of the Epistle to the
Hebrews," AUSemS 19 (1981) 243-46.
25Steven M. Baugh,
"The Poetic Form of Colossians
227-44.
See also Robert K. Farrell, "The Structure and Theology of Luke's Central
Section,"
Trin J 7ns (1986) 33-54.
26There are several
alternatives in the literature. Edward Lohse, for
example, makes
a major
break between
while the latter is hortatory. The conjunction XXX
"therefore" marks the transition
between sections (Colossians and
Philemon [
See
also Eduard SchweIzer, The Letter to the Colossians
(Minneapolis: Augsburg, 1982)
171; N.
W. Meyer, The Epistles to the Philippians and Colossians
and to Philemon (New
Testament (New York: Harcourt Brace
Jovanovich, 1982) 217; J. L. Houlden, Paul's
Letter from Prison (Philadelphia: The Westminster
Press, 1977) 201; Curtis Vaughan,
212 GRACE
THEOLOGICAL JOURNAL
central
section of Colossians is hortatory discourse (see discussion
below) and
is set off from the earlier expository section by the con-
junction ou#n
"therefore" in 2:16.27
Word
Paragraph Elements Tense-Aspect-Mood Order
A (
(1) Let no man,
therefore, judge you (
(2) Let no man condemn
you (
B (
If you have died with Christ (
Rhetorical question
(1) which are meant
for destruction (
(2) which is a
matter (
B' (3:1-4)
If you have been raised with Christ (3: 1)
(1) seek things
above (3: 1) Pres. 2nd plural (S)-0-V
(2) think on things
above (3:2) Pres. 2nd plural (S)-0-V
A' (3:5-17)
(1) Put to death,
therefore (3:5) Aorist 2nd plural V-(S)-O
(a) But now you
also put off (3:8) Aorist 2nd plural V-S-O
(b) Do not lie
(3:9) Pres. 2nd plural V
(2) Put on therefore,
as elect of God (
(a) Let the
peace of God rule (
(b) Let the
word of God dwell (
FIGURE 3.
A Discourse Layout of Colossians 2:16-3:17
Colossians and Philemon (Grand Rapids: Zondervan, 1980) 89; Donald Guthrie, New
Testament Introduction (Downers Grove: InterVarsity Press, 1970) 560;
Harrison,
Colossians: Christ All-Sufficient (Chicago: Moody Press, 1971) 17,74; and
Homer
Kent, Treasures of Wisdom (Winona
Lake: BMH, 1978) 25.
Peter O'Brien offers a second
position. A major break is made in 3:4. The former
section is doctrinal, while the latter is practical. The conjunction
XXX marks the transition
(Colossians and Philemon
[Waco: Word, 1982] 174). See also Robert Gromacki, Stand
Perfect in Wisdom (Grand Rapids: Baker Book
House, 1981) 131; D.
An Introduction to the New
Testament (3 vols;
Simpson
and F. F. Bruce, Commentary on the
Epistle to the Ephesians and the
Colossians (Grand Rapids: Eerdmans, 1957) 174, 175.
27George Cannon makes a
major break at
set off
from
(The Use of Traditional Materials in
Colossians [
156, 157). See also Ralph Martin, Colossians
and Philemon (
1973)
89; and T. K. Abbott, Epistle to the
Ephesians and to the Colossians (ICC;
CHRISTOPHER: DISCOURSE ANALYSIS 213
More
specifically note that the central section begins with two
imperatives, both
present tense 3rd person singular. Word order is
subject-object-verb.
Subsequent imperatives deviate along either or
both
parameters (e.g., subsequent imperatives are 2nd person plural,
there is a
shift to the aorist tense and word order changes to verb-
subject-object). The
final two imperatives, however, return to the
original form
(i.e., present tense 3rd singular). It is at these two locations
in the
central section that the TENSION of the argument is lowest. I
return to
discuss tension below. At this point it is enough to suggest
that the
Imperatives in A and A' mark the on-set and the terminus of
the
central section, forming an envelope structure.
Second, A (
the
chiasmus (see Figure 3). Recall that the imperatives of A and A'
(specifically 2a and 2b) form an
envelope structure. One should not
conclude,
however, that these imperatives balance each other to the
exclusion of the
other imperatives in A' (specially la and 1b). Rather I
take A and
the whole of A' to balance each other
as the exterior
elements
because both A and A I are hortatory
discourse. This stands
over
against Band B' which are expository
discourse.
The two
features which characterize hortatory discourse-agent
orientation and
non-chronological linkage-are present in A and A',
suggesting why
these sections are hortatory discourse. With regard to
the former
feature, note that the imperatives in A and A' are not
embedded (cf. B
and B'; see below). The expectation of A and A' is that
believers (i.e.,
agents) will behave in a certain manner. Behavior, not
exposition (of a
topic), is the focus.
Logical
progression is also evident. In A' (3:5-17), for example,
the
imperative, "Put your members to death" (3:5), is followed by a
causal
prepositional phrase, "because of which (di ] a{) the
wrath of God
is
come" (3:6). The prepositional phrase provides the MOTIVATION to
obey the
imperative. Note also the aorist participles which follow the
imperative,
"do not lie to one another" (3:9): "since you have put off
(a]pekdusame<noi) the
old man. ..(3:9) [and] since you have put on
(e]ndusame<noi) the
new man" (
The logical
progression in A (
Nonetheless the fact that the imperatives carry the reader
forward
through this
paragraph suggests that A is also hortatory discourse.
Here also believers (i.e., agents) are expected to behave in a certain
manner. The
lack of tension in A is further explained below.
And third, B (
of the
chiasmus (see Figure 3). Note that Band B' are introduced with
28Whether one understands these aorist participles as causal or as
attending circum-
stances following the imperative is not important here. In either case,
both carry a logical
progression of thought.
214 GRACE
THEOLOGICAL JOURNAL
conditional
clauses.29 2:20
reads, "If (ei]) you have died with Christ
from the
elements of the world, why then, while living in the world, are
you
subject to ordinances?" (author's translation).
3:1 reads, "If (ei]),
therefore, you
have died with Christ, seek things above" (author's
translation). Also
both sections refer to the earlier doctrinal exposition.
B (
Note also that
the apodosis of both conditional sentences are
imperatives (
imperative (i.e.,
"you should not be subject to ordinances"). But at the
same time
these imperatives are embedded within expository discourse.
Recall that the imperatives of A and A'
are not embedded.
The expository nature of the central
elements is more evident in B.
The imperative of
conditional
sentence. Note also that following the conditional sentence
the
elements are explained (i.e., "which are meant for destruction
[
(non-chronological) but the focus is on
a topic (non-agent orientation).
Recall that A and A' also shared logical progression but the
purpose
seemed quite
different. In Band B' the logical progression is a conse-
quence of explaining a topic, while in A and A'
the logical progression
is used
to provide motivation for believers
to behave in a certain
manner.
Consequently, the focus of Band B' is a topic
while the focus
of A and
A' is behavior.
Though B'
(3:1-4) begins with a conditional sentence, its exposi-
tory
character is not as clear. In B' the imperatives are not mitigated,
but are
typical 2nd person plural. The first, however, is embedded
within the
conditional sentence. These (and the other imperatives) are
discussed below.
At this time it is sufficient to say that a transition to
discourse peak
domain occurs in this section.
In summary it
was suggested that the central section of Colossians
is a
chiasmus. Evidence was offered to suggest that
that
elements, and
that
29It is at this point in the
passage that the dynamic of interpretation becomes
apparent. The interpreter must explain the semantics of the first class
condition at the
sentence level and the parallel sequence of the conditions at the
discourse level. Typically,
a
discussion of the conditions is limited to sentence level. Gromacki,
for example, notes
that these are first class conditions, where the protasis
is assumed to be true (Stand
Perfect in Wisdom, 123).
cause doubt, and are, therefore, translated, "since" (Treasures of Wisdom, 104). The
problem with this explanation is that it overlooks possible larger discourse
patterns.
Recognition
of the chiasmus provides a possible explanation as to why the conditional
sentences appear in a balanced sequence and in this specific location in
the text.
CHRISTOPHER: DISCOURSE ANALYSIS 215
central
elements.30 The
basis for this interpretation was that A and A'
are hortatory discourse and that Band B' are
expository discourse. I
shall now
turn to discuss the progression of thought through the
chiasmus. This
will clear up several issues left unanswered.
Progression
of Argument
Discourse
progresses to a climax or discourse peak. Main-line
verbs carry
the text forward, while specific text features mark discourse
peak. It is
proposed that the central section of Colossians is hortatory
discourse with
embedded exposition. The basis for this conclusion is
the
identification of discourse peak. The imperatives carry the argument
forward.
Changes in tense-aspect-mood and word order and use of the
vocatives mark
the progression that leads to the discourse peak.31
In his
discussion of Hebrew hortatory discourse, Longacre
suggests
that
imperatives (2nd pl), cohortatives (1st pl) and
jussives (3rd pl) are
30The chiasmus can
account for an additional feature of the text, the distribution of
ou#n. It was noted above that the
conjunction ou#n is that basis upon which
interpreters
mark major breaks in the text (see note 27). The distribution of ou#n
complements the
interpretation proposed in this paper. Although ou#n introduces B' and A', it does
not
occur in B (note that there is a textual variant, but the evidence
overwhelmingly favors
its
absence).
It seems reasonable that the ou#n of
2:16 and 3:1 join A and B' with the preceding
expository section of Colossians. Recall that B' makes a back reference to
2:13. it also
seems reasonable to suggest that the ou#n of 3:5 joins B' and A'. Since
believers have been
raised with Christ, they have a new life and consequently should put
to death their
members (3:5) and put on godly character (3:12). The logic, then, is
that the theoretical
exposition (B') proceeds the consequences of that theological truth (A').
That logical
progression, however, is inverted vis-á-vis A and B. In
terms of A and B, the conse-
quences of the theological truth (A)
precede the theological implication (B). Therefore, if
my
analysis is correct that A and B are inverted because of the chiasmus, one
would
expect ou#n to be absent at 2:20. This
interpretation is summarized as follows.
A (oun: br) Theological implication (2:16-19)
B
Theological exposition (2:20-23)
B'
(oun: br) Theological exposition (3: 1-4)
A'
(oun)
Theological implication (3:5-17)
where br means back reference to
earlier position of Colossians.
FIGURE 6.
Logical Development of the Central Section
31The progression that leads to
the discourse peak is a statement based upon
Ricoeur's notion
of emplotment. At this point the reader should note that
the focus of
his
discussion is narrative. The concept has been borrowed in this paper with the
assumption that hortatory discourse, like narrative, is going somewhere. A
progressive
research
program will further validate the expandability of the notion, emplotment,
across
non-narrative discourse types.
216 GRACE
THEOLOGICAL JOURNAL
unranked (see
Chart 2 above). Each form of imperative marks the
main-line (or
primary line of exhortation). Longacre explains that
the
basis to
choose from one of the above forms is the sociological context.
For example, if a speaker is sociologically dominant, then the impera-
tive is
used (e.g., Joseph [incognito] speaks to his brothers in Gen
42:14-16 in the imperative).32
The imperatives
of the central section are interpreted differently. It
seems clear
that the author maintains an authoritative position through-
out
Colossians. There does not appear to be a sociological basis for
variation.
Rather the imperatives carry the exhortation forward, mark-
ing the
progression to the discourse peak. This interpretation can
account for
changes in tense-aspect-mood, changes in word order and
the use of
the vocatives.33
The imperatives
in A (2:16-19) are present tense 3rd singular.
Word order is subject-object-verb. Recall that A is hortatory
discourse
though it
lacks the tension characteristic of A' (3:5-17). This lack of
tension
suggests that A is PRE-PEAK. The underlying claim is that at
peak,
tension is highest.34
Recall that B'
(3:1-4) balances B in that both are expository
discourse. The
expository character of B' is not obvious. The impera-
tives are
typical present tense 2rd plural. The first
imperative is em-
bedded within
the conditional sentence, while the second is not.
It is in B'
that a transition occurs. The shift from 3rd person to 2nd
person
imperatives increases the tension. The author moves from a
mild
exhortation in A to a firm command in B'. The fact that the
second
command, "Think on things above" (3:2), is not embedded,
marks the
transition from expository to hortatory discourse. Again
32Longacre, Joseph, 119-23.
33Ricoeur argues that emplotment (or the organization of events) means that an
event in a story receives a definition from its contribution to the
plot's development. It
follows that a story "must organize [the events] into an
intelligent whole, of a sort such
that we can always ask what is the 'thought' of this story. In
short, emplotment is the
operation that draws a configuration out of a simple succession" (Time and Narrative,
1:65). This means that content selection or exclusion is controlled
by the "thought" or
the
macrostructure. Longacre further explains this
relationship: "Macrostructure analysis
attempts to make explicit how the overall plan and global purpose of a
story exercises a
selective control on the incidents that are included and the relative
elaboration of detail
that characterize the presentation of each incident" (Joseph, 42). Part of the interpretive
process, then, is to account for the text features (e.g., tense change)
and content. It is
necessary to relate the details of the text to its general ideas or
argument (Longacre,
Joseph, x). To account for these, then,
provides a reasonable basis to claim that the
interpreter has a viable understanding of a given discourse.
34Longacre's comment
about the relationship between peak and tension is helpful.
He
writes, "In describing a text we can draw its profile once we identify its
peak(s) [note
that the discourse as a whole has a peak and that individual
episodes or sections can have
a peak].
The profile attempts to represent diagrammatically the rising and falling
tension
of the
text with the beginning, peak, and end as reference" (Joseph, 19).
CHRISTOPHER: DISCOURSE ANALYSIS 217
tension
increases. At the same time note that the word order remains
object-verb. The
subject is understood, a characteristic of unmarked
(i.e., regular) imperatives. The
fact that the word order remains con-
stant
suggests that B' is also part of PRE-PEAK (the fact that B is
embedded
exposition suggests that B is part of PRE-PEAK).
Two major
shifts occur in A'. First, note that the imperatives shift
from
present to aorist tense.35 And second, word -order shifts from
(subject)-object-verb
to verb-subject-object.36 It seems, then, that 3:5
35Typically the basis to
distinguish between the aorist and the present imperative is
as
follows. The aorist imperative means "start to do X," while the
present imperative
means "continue to do X." McKay questions this
understanding. Instead, he defines the
aorist as "representing an activity as a total action, in its
entirety without dwelling on its
internal details," while the present "represents an activity
as a process going on, with the
focus on its progress or development." The aorist imperative,
therefore, urges activity as
a whole
action, while the present imperative urges activity as an ongoing process.
McKay
cites the aorist imperative, "Put to death" (
in which
a specific complete action is to be performed ("Aspect in Imperatival Construc-
tions in New
Testament Greek," Nov Test 27
[1985] 203,204,207,208).
Though McKay's paper is very
helpful, this paper takes the discussion one step
farther, that is a definition of tense vis-a-vis
a higher level discourse concern. It is
proposed that the shift from present tense to aorist tense and back to
present tense go
beyond a simple aspectual shift which, McKay would propose. Rather
changes in tense-
aspect-mood mark the information structure. Figure 7 summarizes that
structure vis-a-
vis the imperatives.
______________________________________________________________________________
Band 1 1.1
Aorist
Primary line of exhortation 1.2 Present (2pl)
1.3 Present (3s)
__________________________________________________________________
Band 2 2.1 Embedded present (2pl)
Secondary line of exhortation
2.2 Mitigated imperative
__________________________________________________________________
FIGURE 7.
Imperative Rank Scheme of the Central Section of Colossians
The scheme only reflects the
information structure of the central section of Colos-
sians. Band 1 imperatives carry
higher levels of tension. It is suggested
in this paper that
the use of
the vocative with the aorist tense marks the on-set of the domain of peak (3:5)
and the
discourse peak of the central section (3:12). Additional research will test
this
rank scheme's expandability to other portions of (NT) Greek
hortatory discourse.
36Blass-Debrunner suggest that Greek word order tends to be
verb-subject-object.
They do
observe, however, that this word order is characteristic of narrative (Grammar
of the New Testament, 428; see also Harold Greenlee, A Concise Exegetical Grammar of
the New Testament Greek [
quite important. This paper has suggested that the typical word order
for imperatives in
the
central section of Colossians is subject-object-verb. The word order does shift
to
verb-subject-object at the on-set of peak domain. It is at this location that the
tension
begins to peak. One should not conclude, however, that this
interpretation conflicts with
the
observations of Blass-Debrunner. The word order that
they have observed is char-
acteristic of narrative while the suggested word order in this paper is
characteristic of
hortatory discourse. Longacre
writes that different features which characterize narrative
will differ from those which characterize hortatory discourse
("Verticle Revolution in
218 GRACE
THEOLOGICAL JOURNAL
marks the
on-set of PEAK-DOMAIN. A set of text features-change in
tense-aspect-mood and
word order change-point to a shift in the
argument as
tension increases.
Even within the
domain of peak, an additional feature is present,
the use of
the vocative. Note that the imperative, "Put to death" (3:5) is
unmarked that
is, the subject is understood (cf. the imperative in B').
Note, however, the imperative, "Put off" (3:8). The subject
is overt, the
plural
pronoun u[mei?j
"you" is present. This increases the tension.
The imperative
"Put on" (3:12) marks the
central
section. Note that the imperative is aorist and the word order is
verb-subject-object. But
at the same time the imperative is highly
marked and
stands apart from all the imperatives. The use of the
extended
vocative, "elect of God, saints and beloved," brings the
tension to a
climax. The text features of 3:12, taken as a set, mark this
imperative as
unique. Consequently, this imperative is taken as the
final and
most effective attempt to change the behavior of the readers.
The imperative, "Let the peace of God rule in your
hearts" (3:15),
marks
POST-PEAK. Tension drops off as the author returns to use the
present tense
3rd singular imperatives. Word order also returns to
subject-verb (note
that the verbs are intransitive).
Figure 4
illustrates the interpretation of the argument of the
central
section of Colossians as it progresses from pre-peak, to peak
domain, to
peak and to post-peak.
Let no man judge you (2:16)
Let no man condemn you (2:18)
Mitigated imperative
[rhetorical question] (2:20)
Seek things above [embedded imperative] (3:1)
Think on things above (3:2)
Put to death (3:5)
But now you
also put off (3:8)
Put on as elect of God, saints and beloved
(3:12)
Let the peace of God rule (3:15)
Let the word of God dwell (3:16)
FIGURE 4. Tension and Argument Structure of the Central Section
of Colossians
Function
of the Chiasmus
Typically the
central element(s) of a chiasmus is the emphatic
focus.37
Augustine Stock calls this the climactic centrality.38 Ronald
__________________________________________________________________
Linguistics," 247-70). This suggests that there are two areas
which need additional
attention: (1) the relationship between word order and discourse type and
(2) the
relationship between word order and information structure. It may no longer
be advisable
to
discuss word order in terms of emphasis alone.
37Welch,
Chiasmus in Antiquity, 10.
38Augustine Stock,
"Chiastic Awareness and Education in Antiquity," BTB 4 (1984)
23.
CHRISTOPHER: DISCOURSE ANALYSIS 219
Man expands upon this and further explains the possible usages
of the
chiasmus. He
suggests that a chiasmus might point to the following:
emphasis of a
passage (Luke 1:6-25), the point of a passage (John
1:1-18) or the purpose of a book (Luke 10:25-18:18).39
The chiasmus
proposed in this paper, however, seems to have a
different
function. The central elements of the chiasmus (2:20-3:4) do
not mark
the emphatic focus of the central section (i.e., discourse
peak).
Rather, as noted above, both B (2:20-24) and B'(3:1-4) make a
back
reference to the earlier exposition (1:9ff). B specifically refers
back to
2:11, 12 and B' refers back to 2:13. This back reference, then,
provides the
cohesion which ties the expository section of Colossians
(1:9ff) to the hortatory section or the central section of
Colossians
(2:16-3:17). Discourse peak is not identified with the central
elements,
but is
identified with the imperative "Put on" (3:12) in A'. Figure 5
summarizes this
interpretation.
Colossians
Exposition
Exhortation
(1:9ff)
(2:16-3:17)
|
Theological implication:
| Hortatory discourse
| (2:16-19)
/
| /
|
Theological exposition:
Death Embedded expository
(2:11, 12) Discourse (2:20-24)
|
|
| Theological exposition:
Resurrection Embedded expository
(2:13) Discourse (3:1-4)
\
Theological implication:
Hortatory discourse
(3:5-17)
FIGURE 5.
Cohesion of the Book of Colossians
Summary
In summary,
this paper proposed that a chiastic structure marks
the
central section of Colossians. 2:16-3:17 forms a unit. A simple
chiasmus serves
as the structural rubric. A (2:16-19) and A' (3:5-17)
balance each
other as the external elements. Each are taken as hortatory
discourse. B
(2:20-23) and B' (3:1-4) balance each other as the central
elements. Each are taken as expository discourse.
39Man, "The Value of
Chiasm," 146-57.
220 GRACE
THEOLOGICAL JOURNAL
The central
section of Colossians is hortatory discourse with
embedded
exposition. The imperatives mark the main-line which car-
ries the
argument forward. Specific changes in tense-aspect-mood and
word order
and the use of the vocative serve to mark pre-peak, peak
domain, peak
and post-peak. The imperative, "Put on, therefore, as
elect of
God, saints, beloved, ..." (3:12), is identified as discourse
peak.
IMPLICATIONS
Several
implications follow from this study. The first implication
is the
potential value discourse grammar has for the interpretation of
the NT.
The value of such a research program is that the interpreter is
provided with a
method by which one can ask different types of
questions and by
which one can address those questions. Such questions
might
include, what are the text-based features that point to the
discourse peak?
How do the text features mark the progression to that
peak? How
is a given section of text to be divided? What are the
text-based
features that point to that conclusion? What is the main
point of a
given section of text? How does the identification of peak
contribute to
understand that main point?
The second implication
is the matter of evidence and argumenta-
tion. This
study showed that a number of grammatical changes occurred
in the
central section of Colossians--tense-aspect-mood, word order
changes and
the use versus non-use of the vocative. In turn a set of
interpretive
conclusions were offered to account for those text-based
features. This
should not imply that this set of conclusions are the
final
answer. This
paper simply provides an interpretation that can account
for these
features. It seems to follow, however, that an alternative
interpretation must
also be able to account for the same features.
And the third
implication is the value of identifying discourse
types. More
specifically, the basis for positing a chiasmus as the
structural
framework for the central section is the identification of
discourse types.
Recall that the exterior elements are hortatory dis-
course while
the interior elements are expository discourse. Typically,
however, a
chiasmus is identified by content. An implied consequence
is that
the parallel content must be of relatively equal length (i.e.,
number of
words). Constructing a chiasmus in such a manner would
call into
question the proposed chiasmus of this paper, for A and A'
are not of
relatively equal length (A' is about twice the length). How-
ever, if it
is possible to posit a chiasmus based upon discourse types,
length may no
longer be an a priori factor of evaluation.
This
material is cited with gracious permission from:
Grace
Theological Seminary
www.grace.edu
Please
report any errors to Ted Hildebrandt at:
thildebrandt@gordon.edu