A Brief Introduction To Discourse Analysis
A Brief Introduction To Discourse Analysis
A Brief Introduction To Discourse Analysis
Noah Kelley
May 23, 2016
I. Introduction
II. Definition
Discourse analysis has been variously defined. To start with the broadest definition (and work
down to something more specific), Jeffrey T. Reed says that discourse analysis “at its broadest
level refers to the study and interpretation of both spoken and written communication of
humans.”2 He says that it is “analysis that takes seriously the role of the speaker, the text and the
listener in the communicative event.”3 Thinking about communication in light of these three
elements (also referred to here as “sender,” “message,” and “receiver,”) illuminates the one
major characteristic of the discipline, namely concern for the roles of syntax, semantics, and
pragmatics in human language.4
To take the last category first, any act of communication by definition involves a situation
external to the message that serves as the context for it. This is the facet of language use that falls
under the domain of pragmatics.5 It involves the relationship between the sender and the
receiver, the pool of shared knowledge that the sender and receiver have, the things going on in
the referential world outside the text that prompted the message, etc. In addition, the pragmatic
dimension includes what the human author is attempting to do by means of the message.6 The
1
This is adapted from a paper written for an independent study on Discourse Analysis presented to Dr.
David Alan Black in the summer of 2014.
2
Jeffrey T. Reed, “Discourse Analysis as a New Testament Hermeneutic: A Retrospective and Prospective
Appraisal,” Journal of the Evangelical Theological Society 39/2 (June 1996), 224. The older term was
“textlinguistics” and was distinguished from discourse analysis by focusing specifically on written texts. While
discourse analysis as a linguistic field is involved in the study of spoken and written language, in the context of
biblical studies it is restricted to written texts.
3
Ibid., 225.
4
Stanley E. Porter, “Discourse Analysis and New Testament Studies: An Introductory Survey,” in
Discourse Analysis and Other Topics in Biblical Greek, Stanley E. Porter and D. A. Carson, eds. Journal for the
Study of the New Testament Supplement Series 113 (Sheffield: Sheffield Academic Press, 1995), 18.
5
Gilllian Brown and George Yule, Discourse Analysis (New York: Cambridge University Press, 1983), 26;
Richard A. Young, Intermediate New Testament Greek: A Linguistic and Exegetical Approach (Nashville: B&H,
1994), 7.
6
John Beekman, John Callow, and Michael Kopesec, The Semantic Structure of Written Communication
(Dallas: SIL, 5th revision, 1981), 58. This is the domain of “Speech Act Theory.”
1
sender may attempt to motivate, command, deceive, inform, etc., but there is always a human
situation in the background that determines a message’s purpose and function. This concern for
pragmatics in discourse analysis even descends down to a concern for how particular devices in a
language function to create a “pragmatic effect” in terms of helping the reader follow the flow of
thought or identify significant elements of the discourse.7
While the human situation motivates an interest in pragmatics, the nature of language as
the medium of communication motivates an interest in syntax and semantics. As Beekman,
Callow, and Kopesec point out, language consists of a “form-meaning composite.”8 The actual
message from the sender to the receiver consists of the thoughts of the sender which are encoded
in actual language which gives perceptible form to the thoughts of the sender.9 The receiver must
then decode the message in order to try and understand the thoughts of the sender. Syntax has to
do with how the elements of a language are arranged to form the clauses, phrases, and sentences
of which the actual form of the message is constructed.10 Semantics has to do with “the
conveyance of meaning through the forms of the language.”11
In addition to the focus on syntax, semantics, and pragmatics, the concern with human
communication leads to a second, narrower characteristic of the discipline, namely that “[a]bove
all, the emphasis of discourse analysis is upon language as it is used.”12 That is to say, in contrast
with traditional grammar study which is largely prescriptive and concerned with “correct usage”
of language, discourse analysis is concerned with the descriptive analysis of actual human
communications.13
This emphasis on the descriptive analysis of actual human communications may in fact
be related to what Reed calls “the most distinguishing, if not best known, feature of the theory,”
the analysis of human communication on linguistic levels above the sentence level.14 Traditional
grammar, with its prescriptive approach that emphasizes grammar in the abstract, tends to
analyze language up to the sentence level because the sentence is seen to be the highest level at
which the grammatical features can be analyzed. Cotterell and Turner helpfully explain the logic
of discourse analysis on this point:
There does not seem to be any good reason why one should be able to analyze the
7
Steven Runge discusses this in his section “semantic meaning versus pragmatic effect” (Steven Runge,
Discourse Grammar of the Greek New Testament (Peabody: Hendrickson, 2010), 7).
8
Beekman, Callow, and Kopesec, Semantic Structure, 6–7.
9
“Thought can only be shared with others by giving it perceptual shape. This is what sound does—not just
any sound, but sound that is phonologically, lexically, and grammatically structured.” (Beekman, Callow, and
Kopesec, Semantic Structure, 8). They would include the written language under this description as well.
10
Porter, “Introductory Survey,” 18, cf. Chapter 4 of David Alan Black, Linguistics for Students of New
Testament Greek: A Survey of Basic concepts and Applications, Second ed. (Grand Rapids: Baker Books, 1995).
11
Porter, “Introductory Survey,” 18.
12
Ibid.
13
Cf. Brown and Yule, Discourse Analysis, 20–21, where they describe their methodology in terms of
purposefully using real communications and not sentences constructed by themselves in the abstract as the proper
data for DA.
14
Reed, “New Testament Hermeneutic,” 231. I write that it “may in fact be related” for the following
reason: while I am not aware of any direct evidence of a connection, it seems intuitive that when actual
communication is studied, new features of language become apparent that are not present in language in the abstract.
It would be like comparing the potential uses of various building supplies in a warehouse to those same materials as
they are actually used in a constructed building. The parts as they actually function in the building would have
features that are not apparent when studying the parts in abstraction.
2
structural relationships of the two nuclear sentences in ‘Give him a pay rise
because he has worked hard for it!’ and yet expect an utterance of the form ‘Give
him a rise! He has worked hard for it!’ to resist such clarification. In the first
instance the two simple sentences are admittedly linked syntactically, while in the
latter they are not, but the same semantic relation holds for both.15
This logic has led practitioners of discourse analysis to focus on the importance of
meaning as it is found above the sentence level. David Alan Black writes that “whereas
traditional grammar tended to focus on the rules a language uses to form units such as phrases,
clauses, and sentences (i.e., ‘microstructures’), textlinguistics concentrates on larger units of
language such as paragraphs, sections, and entire texts (i.e., ‘macrostructures’) [which] stand in a
hierarchical and sequential relation to each other.”16
The result of a focus on the macrostructures of a discourse is both a “top-down” and a
“bottom-up” interpretation of a text, which means that rather than expecting that the text is
interpreted exclusively by adding together the details to discover the meaning of the discourse (a
bottom-up reading), there is an understanding that the meaning of the whole discourse also
constrains the interpretation of the parts (a top-down reading). The result is an approach that
moves in both directions, discerning the whole from the parts and the parts from the whole. 17
This focus on the larger level of discourse provides a crucial help to the interpretive process, and
promises to be one of the most profitable aspects of discourse analysis. 18
Having worked from the general concern with the “study and interpretation of both
spoken and written communication of humans” down through some of the specific
characteristics of discourse analysis (namely concern for syntax, semantics, and pragmatics, and
the analysis of communication above the sentence level), a more specific definition for discourse
analysis as it is practiced in the study of the New Testament can now be stipulated. George H.
Guthrie provides one that captures the idea: “discourse analysis may be defined as ‘a process of
investigation by which one examines the form and function of all the parts and levels of a written
discourse, with the aim of better understanding both the parts and the whole of that discourse.’”19
If the goal of discourse analysis is to examine or study discourse, it is important to define what is
15
Peter Cotterell and Max Turner, Linguistics and Biblical Interpretation (Downers Grove: InterVarsity
Press, 1989), 203.
16
David Alan Black, “The Discourse Structure of Philippians: A Study in Textlinguistics,” Novum
Testamentum 37 (1995), 18.
17
Matthew McDill describes his methodology in an excellent paper on DA: “. . . this proposed method
begins from the bottom levels of the discourse and moves to the top, which takes up the majority of the analysis.
Then the direction is reversed and the discourse is analyzed from the top down, allowing the discourse theme and
structure to govern the interpretation of the smaller units.” (Matthew D. McDill, “Methods in New Testament
Discourse Analysis,” http://static.squarespace.com/static/519b889fe4b01af80eda43d8/519b90c7e4b05d60bd
fd9fc4/519b90c9e4b05d60bdfda06a/1265888976000/NEW-TESTAMENT-DISCOURSE-
ANALYSIS.pdf?format=original accessed 7/5/14., 10).
18
Actually, good interpreters interpret this way anyway, even if the process is not called “discourse
analysis.”
19
George H. Guthrie, “Discourse Analysis,” in Interpreting the New Testament: Essays on Methods and
Issues, David Alan Black and David S. Dockery, eds. (Nashville: B&H, 2001), 255.
3
meant by “discourse.” A helpful definition is found in Cotterell and Turner: “The term
‘discourse’ is used generally for . . . any coherent stretch of language.”20 This emphasizes that a
discourse is characterized primarily by being a whole communication act that hangs together.
Reed emphasizes that it is not the length of the stretch of language that constitutes a discourse,
but to some degree the fact that it is considered to be a complete communication by language
users: “a complete discourse might involve a twenty-volume history of the world or a one-word
exchange between a parent and child . . . Speakers and listeners determine when a
communicative event begins and when it ends.”21
In addition to the primary characteristic of being a complete communication act that
hangs together, discourses have some other characteristics that allow them to be studied above
the sentence level. The first is that they are characterized by thematic grouping. This thematic
grouping is the result of what Dooley and Levinsohn call “chunking.”22 The basic idea is that the
mind cannot process and store an infinite amount of information, and so it groups together items
that are related into a unit in order to help the mind to understand and store it. These groups are
related to one another by a theme, and that theme is what ties them all together.
When an author has made one grouping and begins another, the second group will have a
distinct theme that allows the second group of information to “hang together.” This results in a
thematic discontinuity, or a “boundary” between the two units, which just means that the two
groups remain distinct on the basis of their content.23 In narratives this will be expressed in terms
of continuity or discontinuity of time, place, participants, and kinds of events.24 In non-narrative
genres it will be expressed in terms of a continuity or discontinuity in the situation, participants
and topics, and kinds of actions discussed in the discourse.25 Changes in one or more of these
dimensions, combined with certain surface structure features may indicate a change in theme and
thereby alert the reader to a boundary in the discourse.
Based on this idea of thematic grouping, another characteristic of discourse is that it is
organized into hierarchies.26 Once several groups of information have been produced, these
groups themselves will be grouped together according to thematic continuities that unite them
into a higher level unit. This process is repeated as often as necessary, resulting in levels of
groupings nestled together like Russian dolls: sentences group into paragraphs, paragraphs group
into sections, sections group into divisions, and divisions group into parts, with the highest group
constituting the discourse itself.27 This means that despite the linear nature of language, the
resulting discourse is less like a chain with links that are each connected to the previous one in a
linear sequence, and more like a single building that can be viewed as a group of rooms
constructed from a group of walls constructed from a group of bricks, etc.
Language at every level of the discourse hierarchy also shares in at least three
characteristics: coherence, cohesion, and prominence. While coherence is part of what defines a
20
Cotterell and Turner, Linguistics, 230.
21
Reed, “New Testament Hermeneutic,” 225.
22
Robert A. Dooley and Stephen H. Levinsohn, Analyzing Discourse: A Manual of Basic Concepts (Dallas:
SIL, 2001), 36. Beekman, Callow, and Kopesec refer to this as the “packaging principle” (Semantic Structure, 14).
23
Dooley and Levinsohn, Analyzing Discourse, 37.
24
Stephen H. Levinsohn, Discourse Features of New Testament Greek, 2Nd ed. (Dallas: SIL, 2000), 3.
25
Ibid., 5.
26
Beekman, Callow, and Kopesec, Semantic Structure, 15.
27
The names of these divisions come from Beekman, Callow, and Kopesec, Semantic Structure, 131. For
narrative, the terminology is as follows: the paragraphs group into episodes, episodes into scenes, and scenes into
acts (133).
4
discourse as a whole, it is also a characteristic of each level within the discourse hierarchy.
Coherence is the “hanging-togetherness” of a text or a unit. Dooley and Levinsohn say that “A
text is said to be coherent if, for a certain hearer on a certain hearing, he or she is able to fit its
different elements into a single overall mental representation.”28 They stress that this coherence
is a matter of “conceptual unity.”29 In other words, a text or a unit of a text is coherent if all of its
parts are able to be fit together as a whole in the mind of the reader.30
Another characteristic that is closely related to coherence is cohesion. Cohesion is
defined by Dooley and Levinsohn as “the use of linguistic means to signal coherence.”31 It is the
way that language is used to show the reader where and how the information should be grouped.
Richard A. Young refers to cohesion as the way parts of a discourse are “connected or woven
together.”32 There are syntactical devices used in every language to signal thematic groupings of
information into sentences, paragraphs, sections, etc.
Finally, prominence is “simply making one or more parts of a unit more important than
the other parts.”33 Reed refers to this when he says that “language may be used to set apart (i.e. to
disassociate) certain entities from other entities in the discourse.”34 This is a necessary
characteristic of language without which communication is not possible. A statement by Robert
E. Longacre that is often quoted says that “If all parts of a discourse are equally prominent, total
unintelligibility results. The result is like being presented with a piece of black construction
paper and being told, ‘This is a picture of black camels crossing black sands at midnight.’”35
Again, this is a characteristic at every level of the hierarchy. There is something in every
sentence, paragraph, section, etc. that is more important than other things, and this is usually
related to the theme of that unit.
IV. Conclusion
Discourse Analysis attempts to analyze actual human language in terms of whole discourses. It
understands human language in terms of syntax, semantics, and pragmatics. In contrast with
traditional grammars, it is especially characterized by examining language use above the
sentence level. Discourse Analysis is concerned with interpreting all of the parts of a discourse in
light of the whole, and the whole in light of its parts, or in other words, both “top-down” and
“bottom-up” interpretation.
According to Discourse Analysis, a discourse is a whole act of communication of any
length. A major characteristic of discourse is thematic grouping which is organized hierarchically
like Russian dolls into various levels (sentence, paragraph, section, etc.). Each thematic group
28
Dooley and Levinsohn, Analyzing Discourse, 23. Beekman, Callow, and Kopesec use the term “unity” to
express this characteristic (Semantic Structure, 20).
29
Dooley and Levinsohn, Analyzing Discourse, 32.
30
Some texts are more coherent than others. Someone who struggles to maintain coherence in their
thoughts, whether through laziness or mental illness, would not produce communication characterized by coherence.
31
Dooley and Levinsohn, Analyzing Discourse, 27.
32
Young, Intermediate New Testament Greek, 254.
33
Beekman, Callow, and Kopesec, Semantic Structure, 24.
34
Reed, “New Testament Hermeneutic,” 239.
35
From Robert E. Longacre, “Discourse Peak as Zone of Turbulence,” Beyond the Sentence: Discourse and
Sentential Form, J. R. Wirth, ed. (Ann Arbor: Karoma, 1985), 83; quoted in Reed, “New Testament Hermeneutic,”
239.
5
has boundaries that mark it off from other thematic groups. Furthermore, language is
characterized at every level by cohesion, coherence, and prominence.
Such an understanding of language has the potential to help those who study Scripture
better understand how language works, and thereby make them more sensitive, accurate
interpreters.