That Forgotten Concept in Preaching: The Paragraph

I recall some time ago listening to a conversation between two great preachers, Haddon Robinson and Warren Wiersbe (when Robinson was president of Denver Seminary he had a very helpful audio series on preaching). They were both funny, provocative, and pointed in their discussion of preaching, particularly “expository preaching.” These are two men who have studied and taught about preaching as much as anyone in the last 50 years.

In the discussion Robinson was criticizing the atomistic tendencies in preaching; that is preachers who seemingly dedicate a whole sermon expositing a colon. Wiersbe quickly adding, “or after his operation, a semi-colon.” The point was humorous and well made. As some preachers drill down so deeply into the text they (and their congregations) seemingly lose sight of both context and flow of the whole book. Even Charles Spurgeon noted this problem. In his Commenting and Commentaries he noted,

In the synagogue, it was the rule of the Rabbis that never less than twenty-two verses of the law should be read at one time, and the preaching consisted of notes upon a passage of that length. Such a rule would be a mere superstition if we were slavishly bound by it, but I could almost wish that the custom was re-established, for the present plan of preaching from short texts, together with the great neglect of commenting publicly upon the word is very unsatisfactory. We cannot expect to deliver much of the teaching of Holy Scripture by picking out verse by verse, and hold these up at random. The process resembles that of showing a house by exhibiting separate bricks.[1]

In my own little way, I want to remind pastors that the command of Paul to preach “the whole purpose of God” (Acts 20:27), is not a call to detail or depth but rather completeness. As Keener, F. F. Bruce, and virtually every commentator on Acts have pointed out, Paul is using “watcher on the wall” imagery in Acts 20 (Keener, Acts, 3:3028-29). A watcher sounding the alarm gives a clear and concise message, not an inventory and statistical analysis of the approaching enemy and their materiel.

In this regard there are a few concepts that are important to remember in preaching.

First, of all words, it is vital to remember that words convey no real fixed meaning in and of themselves. An examination of any lexicon (English, Greek, Hebrew, Latin, etc.) will demonstrate almost every word, within its semantic domain, is capable of varied meanings. D. A. Carson gives an extended discussion of the “Word Study Fallacy” (Exegetical Fallacies, Baker 1984) and concludes, “it is very doubtful if profound understanding of any text or any theme is really possible by word studies alone” (66).

Secondly, neither sentences nor parts of sentences convey any real conclusive meaning either. Like a fork in the road a sentence read in isolation may take a person down an entirely wrong path arriving at a meaning never intended by the writer (Acts 22:16 anyone?). Little better than words, clauses or other parts of a sentence are even worse places to find meaning. For instance a writer may craft a sentence:

“The reddish light shown through mist-like haze enveloping the valley.”

Great sentence, evoking an image, but an image of what? A traffic signal, sunrise or sunset, a raging fire coming down from the mountain, the brake-light of the car you are about to hit, a beacon light, or the town square Christmas Tree? The possibilities are nearly as endless as a word. This sentence, standing alone, doesn’t convey any reliable or definitive information. While I believe it is important to be able to properly diagram a sentence; it is equally important to remember that a sentence alone won’t give the reader a complete thought. In simply examining words and sentences exegetes will often find the meaning they were inclined to find anyway.[2]

So, finally, where is meaning to be found? In paragraphs! In writing, the paragraph is the vehicle for a “unit of thought.” It is what one scholar called, “a structure in which completion is a mark—completion of form and discussion.”[3] It’s my contention that paragraphs are where preachers ought to find their subject and sermon, not words and sentences. While lexical considerations and sentence structure are important; they only see finality of meaning in the paragraph. To say we preach “verse by verse” (or even word by word) is perhaps symbolically useful advertising, but beyond that it is homiletically nonsensical.

The history of English Bible shows that if you make this claim you are really claiming to be preaching through a man-made grid laid on top of the text (sometimes with terrible awkwardness), for purposes that had nothing to do with preaching. This grid wasn’t even created until the 16th century, so before that no one preached “verse by verse.” Some publishers have begun to notice this and recently a few Bibles printed free of chapter and verse divisions have made a splash. While I doubt the practical utility of a six-volume Bible with high-end binding and slipcase; the idea of the text freed from Robert Estienne’s grid, set in a natural literary style format is appealing.

The advantage of preaching from paragraph to paragraph is clear: you are traversing from thought to thought, just as the original writer intended to convey their material to thehist_sp-cathedral-01 original audience.[4] Whatever the genre of the text, the paragraph still conveys a whole thought; or, using Spurgeon’s example you are looking at the walls that will hold up the structure, not the individual bricks. Good expository preaching is a presentation of architecture, not a lecture about construction.

Naturally there is a place to teach construction, but that’s the classroom, not the pulpit. Spurgeon said this about some of the preachers in his era,

It makes a man an inch and a-half taller by a foolometer, if he everlastingly lets fall bit of Greek and Hebrew, and even tell the people the tense of the verb and the case of the noun as I have known some to do. . . Brethren, the whole process of interpretation is to be carried on in your study; you are not to show the congregation the process, but to give them the result.[6] 

In cobricksntemporary expository preaching there has been, as Robinson, Wiersbe, and others have pointed out, this atomizing concept, a tendency to plod along not only “verse by verse” but almost clause by clause and word by word. The result is sermon series’ often lasting for years. The “whole counsel” is hard to present if couples get married, have children, raise and get them into high school and all the while the pastor has been preaching the same book that entire time. Since the average pastorate in the United States still only lasts around 3½ years the “drilling down” approach to preaching is probably doomed for most pastors before it even starts.

Aanother downside has been sermons which are simply too long. When I began in ministry, about 30 years ago, the average sermon length in evangelical churches was about 30 to 45 minutes and 45 was the top end. That length has crept up increased by probably 15 minutes to the 45 to 60 minute range, with the top end often being viewed as more preferable. I’ve noticed some discussion where those who view themselves as “real” expositors view sermons of 75 to 90 minutes favorably. Personally, I doubt the effectiveness sermons over the 45 minute mark on a regular basis. I used to teach students that long sermons are easy; anyone can get up and blather on for an hour or so; and frankly most who preach that long are often masking a undisciplined approach to sermon preparation. It takes discipline and hard work to trim all your material down to the essential points, create a discernible outline, present the text in a compelling manner, within a specified time that people will remember. As Hemingway advised writers that’s it’s OK to be drunk when you are writing, but to “edit when you are sober.”[5]

But, you may ask, doesn’t inspiration and inerrancy trump this sort of thinking? All the words are inspired and therefore all the words should be preached, right? Of course the Bible is inspired and inerrant; but again words by themselves, and sentences (or verses if you will) by themselves don’t mean anything apart from the paragraph structure that they are placed. This is the whole problem with the “proof-text” (or the so-called “Analogy of Scripture”) approach is citing random verses, excised from their paragraphical context in an attempt to prove a point, while the thrust thought in that paragraph may have noting to do with the point in question. Words and sentences are only important within the context of the paragraph, not the other way around. In studying languages too often grammar and syntax are somehow thought to stand alone and paragraphing is forgotten.

The concept of “verbal-plenary” inspiration is the truth that the inspiration of the text goes down to each word. But as Ryken points out, however, this is singularly important in the arena of translation. “It is my belief that an essentially literal translation is congruent with the doctrine of verbal or plenary inspiration. Contrariwise, the preoccupation with dynamic equivalent Bibles is with the thoughts of Scripture, with no priority assigned to the words” (Ryken, The Word of God in English, 131).

However, it is a transference fallacy to propose that because each word is inspired therefore each word is somehow a stand alone expression of the mind of God. Each word is inspired, but each inspired word only has a definite meaning within the paragraphical structure it finds itself. I wonder how many preachers have confused the act of translation with the act of exposition?[7]

The Bible is inspired and thusly inerrant and our expository preaching is an attempt to expound what those inspired writers were trying to convey as the Holy Spirit moved them. Words are the bricks, sentences are perhaps the framing, but it is the paragraph that actually conveys thought. The paragraphs are the walls we should be admiring and preaching as we help our congregations see the overall architecture of the structure.



[1] Charles Spurgeon, Commenting and Commentaries (London: Passmore and Alabaster, 1893), 21–22.

[2] Here I am always reminded of the excavations of Qumran in the 1950’s by Roland DeVaux. DeVaux determined that the Qumran community was a monastic group. Of course DeVaux was a Catholic priest, a member of a monastic group. Later archaeologists, reëxamining the site, came away with a near opposite conclusion determining the site was a resort area.

[3] Cited in Edwin Herbert Lewis, The History of the English Paragraph (Chicago: University of Chicago Press, 1894), 21.

[4] As Lewis notes in his introduction the earliest punctuation mark in Greek literature was the paragraph mark. Daniel Aiken and David Lewis Allen, in their Text Driven Preaching (B&H, 2010) make this point, particularly in Allen’s chapter 5.

[5] This is a problem in some publishing circles. Without the cost constraints of printing, some online or e-book publishers have removed some of the page length limits on their writers. The result, in my view, has not been a happy one. Books that are an extra 100 pages long, containing material that a skilled editor would have stripped away, help to confuse and bog down readers.

[6] Spurgeon, Commenting and Commentaries, 30.  Also see Michael Kruger’s post at The Gospel Coalition blog, where he says many of the same things as Spurgeon.

[6] I have always wondered at the practical necessity of a pastor to personally translate each passage they are preaching each week. Frankly for the average pastor this seems a “burden which neither we nor are fathers were able to bear.” Utility in the original languages I think is vital, but how that skill is actually utilized is another matter. So, I wonder if the act of translating each week doesn’t make a pastor so emotionally invested in the effort that the compulsion is to demonstrate that effort in the pulpit, even if homiletically, it is clearly a bad idea.

This entry was posted in Church Ministry, Preaching, Theology by Dennis Swanson. Bookmark the permalink.
Dennis Swanson

About Dennis Swanson

Dr. Dennis M. Swanson is currently the Dean of Libraries at the University of North Carolina at Pembroke. Formerly was the VP of The Master's Seminary Library, Accreditation, and Operations for 24 years. He also oversaw the production of The Master's Seminary Journal, and is an experienced writer and editor. Prior to the he was an officer with the Los Angeles Police Department.
  • Thank you for posting. Excellent article. I had Dr. Robinson as a professor and Dr. Wiersbe became a friend later in life. Your opening mention of these two made the post all the better for me on a personal level.

  • Pingback: Context: The Key to Unlocking the Gospels - ParkingSpace23ParkingSpace23()

  • Pingback: The Daily Discovery (November 30, 2016) - Entreating Favor()

  • chrisleduc1

    “… if you make this claim you are really claiming to be preaching through a man-made grid laid on top of the text (sometimes with terrible awkwardness), for purposes that had nothing to do with preaching…”

    Does not the above cited quote apply equally to the so-called “paragraph”? Especially the final clause, given that most modern, arbitrary “paragraphs” have been made with little or no consideration whatsoever for the rhetorical structure of texts that were meant to be read aloud and heard, rather than read in silence to oneself?

    • Dr. Dennis Swanson

      No, paragraphs go back to classical Greek structure. The paragraphs are rather easily discernible in the text.

      • chrisleduc1

        So there are inspired paragraph divisions?

        • Jason

          Preaching through Matthew, it seems extremely easy to discern Matthew’s paragraph breaks, when he begins and ends thoughts, and when he changes subjects. I’d say those paragraph changes are obviously inspired and yet, there are times I find the chapter / verse given to the verse to be off (14:1-12 makes more sense connected to 13:44-14:12, but I’d not die on that hill).

          • chrisleduc1

            Thematic changes, periscopes, and rhetorical makers are no doubt there. But the question is whether “paragraphs” as we see them today are a “man-made grid” or divinely inspired and therefore absolute. Especially in texts like Ephesians, James, 1 John etc.

          • Jason

            As Farnell basically said, There are often many possible solutions, each with valid arguments, pray and do your best. . . . But he and most Greek grammars I’ve read will assert, the unit of thought is the paragraph in Greek.

          • chrisleduc1

            “…the unit of thought is the paragraph in Greek” agreed.

            But that is not my point, and it’s not helpful to continue to move the flag pole.

            Either there are inspired and objective markers that begin and end paragraphs, or there are not. If there are not, then we are back to “preaching through a man-made grid laid on top of the text (sometimes with terrible awkwardness), for purposes that had nothing to do with preaching…” Its just a different “man-made grid” no matter how much evidence is greek literature for various forms.

          • Dr. Dennis Swanson

            Chris, the mark for a paragraph is, according to Allen, the oldest punctuation mark in Greek manuscripts. That the NT writers used such marks must be assumed since we don’t have originals. But in subsequent copies that paragraphical structure remains consistent. You’re arguing in the same manner of the critics who say we can never really know what the text is because we don’t have the original copies.

          • chrisleduc1

            My understanding is that the evidence we have from the earliest manuscripts varies drastically and is in no way in comparison to the limited number/types of variants we have for the words. By all appearances, they are later additions as is evidenced by the drastic variences. If there are some other published works stating otherwise, I’d love to read them, so please let me know.

          • Dr. Dennis Swanson

            Could you possibly restate your question, I’m not entirely sure what your objection at this point is.

          • chrisleduc1

            Going back to my first post, I’m simply asking for the evidence that the modern paragraph divisions are not a man-made grid laid on top of the text, similar to the modern verse divisions. The manuscript evidence I’m aware of, and conclusions reached that I’ve read, indicate the various and diverging divisions seen in various manuscripts appear to be later additions based on varying and differing opinions.

            If you are aware of evidence to the contrary I’d love to see it. Otherwise your solution seems to be guilty of the same thing you’re critiquing in regards to verse divisions.

          • Dr. Dennis Swanson

            Then we are at an impass. Paragraphing is part of writing and in Greek Literature predates the NT. Was the normal paragraph markers in the original manuscripts? It’s impossible to know, but they appear in other manuscripts, they are clearly marked in the Vulgate, etc. and they are consistent. Tyndale put his NT into paragraphical format (verse numbers didn’t exist yet) in his translation. Certainly paragraphs aren’t “required” in terms of syntax but they exist and they’ve been part of prose as far back as prose itself. Are grammar and syntax a man-made grid? Of course not. Again, context is what drives meaning and the paragraph is what defines close context. But if you want to preach one verse at a time and spend, for Ephesians for instance, 155 weeks, knock yourself out; but don’t fool yourself into thinking that is the only or even the best way to preach expositioanlly.

          • chrisleduc1

            If you can direct me to evidence for the consistency in the manuscripts that you are referring to, I’d greatly appreciate it.

          • Dr. Dennis Swanson

            You do believe in inerrancy don’t you? From your responses I’m sort of questioning that.

          • Dr. Dennis Swanson

            I think to ponder “absolute” is to set up a false tension. Without context words are imprecise; the paragraph creates context for the words. What I’m arguing for more than anything is that a unit of text for preaching should be at a minimum a paragraphical unit. If a sermon preaches a “verse” as though it represents “truth” or the thought the writer was trying to convey, I think the preacher is mistaken.

        • Dr. Dennis Swanson

          Well, what would the point of “inspired words” be apart from the structure in which the are placed? It’s a paragraph that creates context.

  • Pingback: Wednesday Link List | Thinking Out Loud()