Preach the Word – Discourse Analysis for Preaching

Themes in Current Homiletical Theory

Discourse Analysis for Preaching

Haddon Robinson was one of the most well-known homileticians of the late twentieth century. He was the Distinguished Professor of Preaching at Gordon-Conwell Theological Seminary, but most remember him for his homiletics textbook Biblical Preaching, which is widely used in seminaries across America. After making his case for expository preaching, chapter 2 is devoted to “The Big Idea.” Robinson laments what he calls “fragmentation” in preaching, namely, how some preachers “offer little more than scattered comments based on words and phrases from a passage, making no attempt to show how the various thoughts fit together as a whole.”¹ Robinson goes on provide biblical and communicative rationale for the importance of distilling one central theme—in other words, why “a sermon should be a bullet, not buckshot.”² In the homiletical world, Robinson has become synonymous with “The Big Idea.”

The terminology may differ—Robinson’s “Big Idea,” Chapell’s “central theme statement” that passes the “3:00 A.M. test,”³ or Gerlach and Balge’s “propositional statement”⁴—but basic homiletical theory emphasizes the importance of distilling the text into one central statement during the textual analysis stage of sermon development. But that begs the question—how do you determine “the big idea” in preaching? Homileticians frequently tell preachers to determine a central idea when they are studying the text but not necessarily how to do so. For example, in his section on “The Formation of an Idea,” Robinson says, “Because each paragraph, section, or subsection of Scripture contains an idea, we do not understand a passage until we can state its subject and complement exactly. While other questions emerge in the struggle to understand the meaning of a biblical writer, these two (“What precisely is the author talking about?” and “What is the author saying about what he is talking about?”) are fundamental.”⁵ Later he gets slightly more specific, “You must become aware of the structure of the passage and distinguish between its major and supporting assertions.”⁶ Chapell lists two simple steps: (1) “read and digest the passage” to determine the text’s main idea or an idea with enough material in the text and (2) “melt down this idea and develop it into one concise statement.”⁷ Gerlach and Balge encourage preachers to mark the major coordinate thoughts with a heavy line and then to mark the minor subordinate thoughts with a lighter line, grouping the subordinate thoughts under the coordinate thoughts.⁸

How do you determine the “big idea” in preaching?

But to ask it again, “How exactly do you determine which are the main points and which are the subpoints?” If the answer is not clear or explicit, it is no surprise that often textual analysis (if it is even done) rests on the preacher’s own intuitive instincts. Preachers do their best to group the text into various parts and then summarize what the text is all about. Because this method is based on different instincts, different preachers create different themes and parts and take their sermons in different directions. What is lacking in current homiletics is a more objective, data-driven, textual way to determine the big idea. Thankfully, a solution exists in another discipline of theology, namely, biblical studies, and more specifically, biblical linguistics.

Basic Features of Discourse Analysis

Linguistics is the formal, scholarly study of how language works. Many of us were trained in exegetical text studies that basically work like this: give an overly literal translation of the original text, analyze the verb forms, make some comments on the historical background or the doctrinal content, then go on to the next verse, and finally read some commentaries. Linguistics has shown that a text is far more than that.

Discourse analysis falls under linguistics. While the method of text study I described above is usually focused on aspects within the sentence itself,⁹ discourse analysis seeks to analyze the text on a more macro-level. It focuses on the clues an author has left embedded in the text to discern how each thought relates to one another, and which points the author intends to be the main points and which points the author intends to be the subordinate points. As David Black explains:

Generally speaking, discourse analysis is the attempt to study the organization of language above the sentence level. … Chief among the concerns of discourse analysis is to show the internal coherence or unity of a particular text. Discourse analysis involves a wholistic study of the text. It is not simply “verse by verse analysis” (the method usually taught in seminaries and employed in most commentaries), but rather an analysis of how verses fit into the structural unity of the entire text. It is critical to realize that discourse analysis is not merely an investigation into the flow of thought of a text, but is at heart an investigation into how the text produces flow of thought.¹⁰

Discourse analysis focuses on the clues an author has left embedded in the text to discern how each thought relates to one another.

The renowned Bible translator Eugene Nida quipped that exposure to linguistics “can be a positive, although sometimes difficult, learning experience.”¹¹ I need to warn pastors: the first time you are exposed to discourse analysis, it can be a rather frustrating experience filled with technical language. The problem is that formal discourse analysis, generally speaking, is not taught at the MDiv level but is reserved for postgraduate study.¹² However, even if you have never studied discourse analysis, everyone practices a basic form of discourse analysis whenever they speak English. Consider this example:

While I was busy writing my sermon, my wife called. The day was a disaster. Our tired kids were screaming all day. So I decided to quickly wrap up my sermon and go home. After all, family is so important.

What’s the main point? There are certain clues in the text itself, words like “while,” “so,” and “after all.” The hardest part of discourse analysis is when there are no connectives between sentences, but then we can make the implicit explicit by supplying the implied connectives. With that in mind, I can turn this example into a basic discourse analysis:

• My wife called
• (temporal) while I was busy writing my sermon
• (general-specific) [She said] The day was a disaster
• (grounds) [because] our tired kids were screaming all day.
• (cause-effect) So I decided to quickly wrap up my sermon and go home.
• (grounds) After all, family is so important.

This uncovers the second-to-last thought as the main one, namely, this made-up example of how my day went is communicating that I went home early because of the importance I was placing on my family.

Discourse Analysis Applied to Homiletical Text Analysis

The most recent edition of Logos Bible Software included discourse analysis and propositional outlines at a click of a button. I recently upgraded largely for that purpose alone, especially given its utility for sermon preparation. One downside to Logos’ propositional outlines, however, is that they give little help in determining the macro-view of the pericope. G.K Beale’s approach to discourse analysis, however, does explicitly identify which propositional relationships are superordinate and which are subordinate. Unfortunately, there is no standard terminology among scholars, so for simplicity’s sake here, I am focusing on the Greek NT, and I am using G.K. Beale’s terminology.¹³

Now we are at the point where we can see how discourse analysis plays out in an actual text analysis. Figure 1 is my formal discourse analysis for Revelation 20:1-6 (Proper 5B).

By way of orientation, the brackets signal the relationship between the propositions, and a superordinate relationship is identified by *. As you go further to the left, you get to more superordinate relationships. Each * needs to be connected to another * by a bracket, and so on and so forth until you have arrived at the main idea of the entire text. In my explanation below, the relationship signaled by these brackets will be highlighted in italics.

Let’s break this text down. First, I arrange my translation of the Greek text to the right, making sure to include all the Greek connective words.¹⁴ Those are highlighted in red. The Book of Revelation includes a series of seven visions that begin with, “And I saw.” Chapter 20 introduces the next vision of an angel (which I think is Christ), holding a key and a chain. V. 2-3a is a general-specific propositional relationship that further specifies what John saw the angel doing. V. 3b is a means-end propositional relationship that explains the goal or purpose of the angel binding Satan, followed by v. 3c, which notes a temporal relationship that places Satan’s loosening at the end of the millennium. By this point, we have v. 1-3 in place, and we can move on to the next phase of John’s vision in v. 4-6, the saints who are seated on thrones. In v. 4b-5a, the saints are contrasted with “the rest of the dead” in a positive-negative propositional relationship (A, instead of B; or not B, but rather A), which emphasizes the positive all the more. In other words, the rest of the dead did not live on with Christ, but the saints did. This is the more specific thing John saw regarding the saints who were seated on thrones and were given judgment in v. 4a (which here refers more to kingly rule than judicial decrees). Finally, the apostle John interprets the meaning of all this (a fact-interpretation propositional relationship) with his benediction about those who share in “the first resurrection.” At this point, we have the propositional relationships in place for v. 1-3 and v. 4-6, but the most difficult part is determining how both relate to each other. Upon closer inspection, Satan’s binding during the millennium results in the saints’ rule with Christ (a cause-effect propositional relationship). So the “big idea” of this text is in v. 6, namely, the saints’ priestly and royal reign with Christ. This highly debated text is riddled with interpretive issues—Satan’s binding, the millennium, physical vs. spiritual “coming to life,” and so forth—but a discourse analysis helps the preacher keep the main point the main point: Christians share in Christ’s victory today.

Discourse analysis is hard. For me personally, the best part of preaching is the beginning and the end. I love text studies, and I love writing and preaching sermons. The most arduous part is the middle: the discipline of textual analysis and an extended outline. But the fruits of those painful labors will be seen when the final theme and parts and the final sermon are organically derived from the features of the text itself.

There are not a thousand sermons in every text.

A Thousand Sermons in Every Text?

This article has been interdisciplinary. I have taken one field, biblical linguistics, and applied it to another field, homiletics. Discourse analysis can seem to be quite technical (and admittedly, it is), but this article has shown the practical utility of discourse analysis or preaching. At times, I have listened to sermons—trying to go through every deductive or inductive homiletical outline or pattern I know in my head and yet still struggling to follow where the preacher is going—and then at the end I say to myself, “I’m not quite sure that was the point of the text.” I imagine many other pastors (and parishioners too!) have been in that situation. You could listen to a thousand preachers preaching the same text, and you could hear a thousand different sermons. As the old adage goes, “There are a thousand sermons in every text.” When I first studied homiletics, I found it interesting to hear so many different themes and different sermons from seminary students preaching on the same text. The more I have studied homiletics since then, the more I am convinced that was not necessarily a good thing. There are not a thousand sermons in every text. Because preachers are bound by the text, they do not have the freedom to take the text and develop the sermon in a thousand different ways. This results in atomistic preaching that often isolates key passages of a preacher’s own choosing from the text.¹⁵ At best we could say, “There are a thousand applications in every text.” Certainly, preaching on a passage like, “Love your neighbor as yourself,” will reveal a thousand different scenarios for different people in their different vocations. But even then, it would be more precise to say, “There are a thousand different examples of this one application.” Like textual exposition, applications still need to be derived from the text, and the text gives us a finite amount of applications. This is why discourse analysis is so important for preaching. If preachers can do the hard work of wading through the technical language, they will come away with a textually grounded tool for determining the main point and application of the text that is sitting right in front of them.

Written by Jacob Haag

Rev. Dr. Haag serves at Redeemer Lutheran Church, Ann Arbor, MI. His doctorate is from Westminster Theological Seminary with research in New Testament and preaching. His research project was entitled “Evangelical Exhortation: Paraenesis in the Epistles as Rhetorical Model for Preaching Sanctification.” He also serves on the Michigan District Commission on Worship.

---

¹ Haddon W. Robinson, Biblical Preaching: The Development and Delivery of Expository Messages, 2nd ed. (Grand Rapids, MI: Baker Academic, 2001), 35.
² Robinson, Biblical Preaching, 35–39, esp. 35.
³ Bryan Chapell, Christ-Centered Preaching, 2nd ed. (Grand Rapids, MI: Baker Academic, 2005), 47, 136.
⁴ Joel Gerlach and Richard Balge, Preach the Gospel (Milwaukee: Northwestern, 1982), 25.
⁵ Robinson, Biblical Preaching, 42-43.
⁶ Robinson, Biblical Preaching, 67.
⁷ Chapell, Christ-Centered Preaching, 46-47.
⁸ Gerlach and Balge, Preach the Gospel 24-25, 28-32.
⁹ J.P. Louw laments this approach in sermons, “Even Bible commentaries and sermons focus to a large extent on word meanings. Except for occasional references to historical or cultural issues, words and ‘what they mean’ have become the beginning and end of most attempts to arrive at a proper understanding of a passage.” J.P. Louw, “Reading a Text as Discourse,” in Linguistics and New Testament Interpretation: Essays on Discourse Analysis, ed. David Alan Black, Katharine G. L. Barnwell, and Stephen H. Levinsohn (Nashville: Broadman, 1992), 17. Later, he explains, “Linguists insist that the meaning of a sentence is not merely the sum total of the meanings of the words comprising the sentence and, similarly, that discourses are not a matter of sentence meanings strung together. Reading a text involves far more than reading words and sentences” (18).
¹⁰ David Alan Black, introduction to Linguistics and New Testament Interpretation, 12.
¹¹ Eugene A. Nida, forward to Linguistics and New Testament Interpretation, 9.
¹² For example, the preface of Wallace’s Greek Grammar—widely used to train pastors at seminaries across America—admits it does not treat discourse analysis. Daniel B. Wallace, Greek Grammar Beyond the Basics: An Exegetical Syntax of the New Testament (Grand Rapids, MI: Zondervan, 1996), xv.
¹³ G. K. Beale, Daniel J. Brendsel, and William A. Ross, An Interpretive Lexicon of New Testament Greek: Analysis of Prepositions, Adverbs, Particles, Relative Pronouns, and Conjunctions (Grand Rapids, MI: Zondervan Academic, 2014), esp. 7-12.
¹⁴ Unfortunately, John exhibits typical Semitic parataxis, and the numerous καὶ’s do not help the preacher. Clauses that begin with καὶ often signal a propositional relationship far deeper than “and.”
¹⁵ I actually preach significantly longer sermons than I did when I first entered the ministry. My sermons are now about 2500 words (or about 30% longer). The reason is twofold. First, my study of discourse analysis has impressed on me the importance of expositing the entire pericope (as much as possible), in order to allow the hearers to see the text’s natural flow of thought. I believe longer sermons that are focused on the entire text (as opposed to shorter sermons that isolate a few passages in the text) will result in sermons that are more similar and more textually-grounded, no matter who is preaching them. Second, my study of the theology of preaching has impressed on me the importance of a sermon fulfilling all the philosophical obligations of preaching. I have grown less convinced that a 12-15 minute sermon, generally speaking, can do everything a sermon is meant to do. For this last point, see Tim Bourman’s series of Four Branches articles on sermon length (“Sermon Length Unleashed,” December 2023; esp. “‘Law/Gospel Obsession’ and Sermon Length,” February 2024), accessible from www.wisluthsem.org/grow-in-grace-2024/the-four-branches-review.

---

For Further Study

Consider Prof. Ken Cherney’s online Summer Quarter class, “Linguistics for Exegetes” (next offered in fall 2024).

---

Devote Yourself

1 Timothy 4:13: “Until I come, devote yourself to the public reading of Scripture, to preaching and to teaching.”

Coming this fall, WELS Congregational Services will offer a new newsletter called Devote Yourself. This newsletter will focus on providing ministers of the Word with helpful articles and practical approaches to teaching, preaching, and leading worship. This new newsletter will be an email subscription-based resource that replaces the current electronic newsletter, Teach the Word, as well as the mailed newsletters of Preach the Word and Worship the Lord. Delivered on a bimonthly basis, you will find similar articles and helpful content that you have enjoyed in the past. The electronic version will not only save on printing and mailing costs but will also make it easier to include links and videos. The articles in Devote Yourself will also be posted on welscongregationalservcies.net. Watch for more information on the launch of this new newsletter resource.

---