Matthew Dicks collaborated with a sociologist on a writing project. The two of them were working on a book that would combine the sociologist's research findings with Dicks's storytelling craft. The collaboration had a recurring conflict. The sociologist wanted every paragraph to open with its topic sentence — the academic-writing convention where the first sentence of each paragraph announces what the paragraph is about, and the subsequent sentences support the announced claim. Dicks resisted. The sociologist insisted; it was how she had been trained, it was how scholarly writing operated, it was how readers would expect the prose to flow.
The collaboration failed at the structural level. The topic-sentence opening killed the storytelling. Each paragraph, opening with its conclusion-announced upfront, removed the reader's discovery experience. The reader was told what to think before being shown the evidence; the discovery that produces engagement was foreclosed by the topic-sentence structure.1
Dicks's broader argument: the school-writing pedagogy of topic-sentence-followed-by-supporting-sentences is a load-bearing barrier to storytelling-writing. School pedagogy was designed for argumentative essays where the reader needs to know what claim each paragraph is supporting. Storytelling writing operates on the opposite principle — the reader should be brought through the experience without knowing in advance where the paragraph is going. Topic sentences prevent this; their elimination is one of the load-bearing skills storytelling-writers need to develop after years of school-pedagogy training.
The principle has three components.
1. School writing pedagogy teaches a specific paragraph-structure called the topic-sentence-and-support structure. The first sentence of each paragraph states the paragraph's claim; the remaining sentences support, explain, or elaborate the claim. The structure is taught from elementary school through college freshman composition. It produces the five-paragraph essay and analogous forms.
2. The topic-sentence structure is appropriate for argumentative writing and inappropriate for storytelling writing. Argumentative essays need readers to track the structural claim-and-evidence pattern; topic sentences serve this function. Storytelling needs readers to experience the narrative without pre-announcement of where each paragraph is going; topic sentences foreclose this experience by announcing each paragraph's outcome before delivering it.
3. Storytelling-writers must un-learn the topic-sentence reflex. Years of school training have made topic-sentence openings automatic for most writers. Producing storytelling-writing requires actively resisting the reflex and constructing paragraphs that develop into their content rather than announcing it upfront. The un-learning takes deliberate practice and is one of the most common bottlenecks in adult-writers' transition from school-essay competence to storytelling competence.
The cognitive mechanism: storytelling depends on the reader's prediction-investment in what is going to happen next. The reader is hypothesizing about the protagonist's next move, anticipating the scene's unfolding, building expectations the storytelling will eventually pay off (see the power-out test for the related diagnostic).
A topic sentence announces the paragraph's outcome before the paragraph develops. The reader's prediction-investment cannot operate because the prediction has been pre-empted. The next day my grandmother died (as the opening sentence of a paragraph) eliminates the reader's investment in the question what happens next? — the reader already knows. The subsequent sentences of the paragraph deliver detail about the death-event the reader is already prepared for. The detail lands without the prediction-tension that would have amplified its impact.
The same paragraph, restructured to develop into the death-revelation, produces dramatically different reader engagement. I woke up the next morning and the house was quiet in a way I had not noticed before. My mother was at the kitchen table with her hands wrapped around a coffee mug she had let go cold. She did not look up when I came in. She said: "She's gone." I knew what she meant. The reader's prediction-investment operates across the four sentences — the quietness, the cold coffee, the not-looking-up, the she's gone — each builds toward the revelation rather than landing under a pre-announced topic-sentence weight.
A second mechanism: topic sentences produce flat affect across paragraphs. Each paragraph opens with a claim; each paragraph closes having supported the claim; the structural rhythm is repetitive. Storytelling writing varies its paragraph-shape — some paragraphs open in-scene, some open in dialogue, some open with image, some open with sensory detail. The variation produces the rhythm readers experience as engaging. Topic-sentence pedagogy enforces structural uniformity that flattens the rhythm.
A third mechanism: topic sentences pull the writer toward summary and away from scene. The writer who opens each paragraph with a claim about what the paragraph contains is operating at a summary-level abstract distance from the events; the writer who opens each paragraph in-scene is operating at the sensory-experiential level the reader needs to be invited into. The topic-sentence reflex pushes prose toward summary by default; storytelling requires the opposite default.
Several alternative paragraph-opening patterns produce stronger storytelling-writing.
Open in-scene. I was standing in the kitchen by the window. The opening locates the reader in physical space and time. The reader's senses engage; the paragraph's content develops from the scene; the reader experiences the unfolding rather than receives the summary.
Open with dialogue. "Are you going to tell her?" The opening drops the reader into a conversation already underway. The reader's mind constructs the speaker, the listener, the implicit context. The paragraph develops the conversation; the meaning emerges from the dialogue rather than from a topic-sentence claim.
Open with sensory detail. The smell of bleach hit me as soon as I opened the door. The opening engages the reader's senses; the paragraph develops what the sensory detail means within the moment; the reader experiences the smell before knowing what to make of it.
Open with action. She picked up the photo and turned it face-down. The opening shows the protagonist doing something; the reader's attention engages around why is she doing that?; the paragraph develops the action's meaning rather than announcing it upfront.
Open with question. Was I supposed to apologize? The opening puts the reader inside the protagonist's uncertainty; the paragraph develops the protagonist's reasoning through the question; the reader experiences the deliberation rather than receives a conclusion.
The pattern across all alternatives: the paragraph develops into its content rather than announcing its content. The reader's experience is one of being brought through the material rather than being told about it.
The principle is operationally adjacent to start with location and action at the paragraph level. The location-and-action principle says story-openings should drop the reader into a specific place and a specific action; the against-topic-sentence principle says every paragraph should similarly drop the reader into specific material rather than announcing the material upfront. The two principles operate at different scales (whole-story opening vs. each-paragraph opening) but share the underlying logic.
It pairs with leverage audience imagination, do not describe: topic sentences are a form of telling-not-showing. They announce what the paragraph means; they do not let the reader construct the meaning. The against-topic-sentence principle is the paragraph-structural analog of the leverage-imagination principle.
It connects to cardinal sins: topic-sentence openings frequently commit Sin 2 (telling the audience how to feel) and Sin 3 (explaining the moral) by announcing the paragraph's conclusion before delivering its content. Eliminating topic sentences eliminates several cardinal sins as a by-product.
It contrasts with academic writing tradition, where topic-sentence pedagogy serves a real function (helping readers track complex argumentation). The contrast is medium-and-purpose-specific: academic writing benefits from topic sentences; storytelling writing does not. Writers who work in both forms must learn to switch paragraph-structure by context.
Consider the same hypothetical content rendered in two paragraph-structures.
Version A (topic-sentence structure): The 1970s economic transformation created widespread anxiety in working-class American households. Across the country, families experienced wage stagnation and job insecurity for the first time in a generation. My father's experience exemplified this pattern. He worked at a manufacturing plant in Connecticut that began layoffs in 1974. He came home one Friday to tell my mother he had been laid off. The conversation that followed shaped my childhood.
The paragraph opens with a topic-sentence (the 1970s economic transformation created widespread anxiety); the subsequent sentences support the claim with the family-specific instance. The reader knows from the first sentence what the paragraph is going to deliver. The protagonist's father appears as an example; the conversation is announced as shaping without being rendered.
Version B (storytelling structure): My father came home one Friday in 1974 and stood in the kitchen doorway without taking his coat off. My mother saw him and stopped slicing the onions she had been working on. He said the words I had been pretending I would not hear: "I got laid off today." My mother nodded slowly. She put down the knife. She walked over to where he was standing and put her hand on his arm. They stayed there for a long time, neither of them saying anything. I was eight years old. I learned what adult fear looked like in that minute.
The paragraph opens in-scene; the father's appearance, the mother's stopped knife, the specific words, the gesture of hand-on-arm. The reader experiences the moment rather than receives the topic-sentence summary. The sociological context (1970s manufacturing layoffs) is implicit in the specifics rather than announced as the paragraph's claim.
Version B is dramatically more engaging despite covering identical content. The difference is paragraph-structure. Topic sentences buried the experiential material in Version A; the absence of topic sentences let the experiential material operate in Version B.
The instructive part of the case is that the same writer could produce either version. The choice is structural, not content-level. Writers trained in school-essay pedagogy default to Version A; writers trained in storytelling pedagogy default to Version B. The un-learning of the topic-sentence reflex is what enables the shift.
You have written drafts that read flat. You suspect topic-sentence pedagogy is operating in the background. You want to un-learn it.
It is Sunday morning, 9:30am. You're at the kitchen table with a draft you wrote yesterday. You read each paragraph's first sentence in isolation. You notice: most first sentences are claims or summaries about what the paragraph contains. That was the day I decided to leave. / The conversation with my mother was difficult. / My grandmother had always been the calm one in the family. Each is a topic-sentence in disguise — announcing the paragraph's content before delivering it.
You rewrite the first sentence of each paragraph using one of the five storytelling alternatives — in-scene, dialogue, sensory detail, action, or question. That was the day I decided to leave becomes I was packing the last of my books when my sister knocked on the open door of my bedroom. The conversation with my mother was difficult becomes "I don't think I can do this anymore," I said. My grandmother had always been the calm one becomes The kitchen smelled like the cinnamon-and-butter she had been baking for forty years.
You re-read the revised draft. The reading is different — you are brought through the material rather than told about it. The paragraphs feel like scenes rather than like expository segments. The reader's prediction-investment operates across paragraphs rather than being foreclosed at each paragraph's opening.
You note which paragraphs were hardest to convert. The hardest are usually the ones where the topic-sentence reflex was most deeply locked in — moments where you, as a writer, felt the need to explain the content rather than render it. Those paragraphs are the leverage-points for further un-learning.
You apply the protocol to subsequent drafts. After several drafts the topic-sentence reflex weakens. Your default first-sentence shifts from announce-mode to in-scene-mode. The un-learning gradually completes.
The pattern is: identify the topic-sentence reflex by reading first-sentences in isolation, replace each topic-sentence with one of the storytelling alternatives, observe the paragraph's improved engagement, repeat across drafts until the reflex weakens to the point that storytelling-paragraph-opening becomes default.
Your prose reads like an essay even when you intended a story. The essayistic quality is usually the topic-sentence pedagogy operating. Readers report your stories feel explanatory rather than experiential. The fix is the un-learning protocol above.
Your first sentences are interchangeable across drafts. Topic-sentence openings tend toward generic claim-construction. If your opening sentences would work equally well swapped between drafts, you are operating in topic-sentence mode. Storytelling openings are specific — they could not be swapped because each is rooted in a specific scene.
Your beta readers cannot remember specific scenes. Topic-sentence-structured prose conveys information; storytelling-structured prose conveys experiences that readers carry as memories. If readers remember your work as information about the events rather than as the events themselves, the topic-sentence structure may be operating.
You feel resistance to the alternatives. The five storytelling-alternative openings feel uncomfortable at first because they violate years of school-pedagogy training. The discomfort is diagnostic — it indicates the topic-sentence reflex is deeply locked in. The fix is to push through the discomfort with deliberate practice.
Watch the conflict in operation: school-trained writing-pedagogy reflexes actively interfere with storytelling-craft requirements, and the storyteller must deliberately un-learn the former to enable the latter. Several other domains have analogous un-learning challenges.
Music education / classical-to-jazz transition: Classical music training emphasizes precise execution of notated material; jazz requires improvisation within chord-progression constraints. Classically-trained musicians transitioning to jazz often struggle because their classical training reflexes (look at the page, execute exactly as written) actively interfere with jazz requirements (do not look at the page, generate within the constraints). The un-learning is well-documented in jazz pedagogy as one of the central barriers for classically-trained converts. The structural parallel to the topic-sentence un-learning is direct: prior-training reflexes that served a previous discipline actively interfere with the new discipline's requirements. Pairing music tradition with writing tradition: jazz pedagogy has developed substantial methodologies for the un-learning (improvisation-against-changes exercises, melodic-fragment recomposition, deliberate avoidance of sight-reading); writing-craft could productively borrow these methodologies for un-learning topic-sentence reflexes.
Therapy / clinical training of medical professionals: Medical training emphasizes clinical mode communication — professional distance, technical vocabulary, asymmetric authority. Medical professionals transitioning to therapeutic work must un-learn clinical-mode and learn therapeutic-presence mode. The un-learning is one of the most documented challenges in clinical-psychology training of medical converts. The parallel to topic-sentence un-learning is structural: prior-training mode interferes with new-mode requirements; the un-learning takes deliberate practice across years. Pairing therapeutic tradition with writing tradition shows that un-learning prior-mode communication reflexes is a domain-independent professional-development challenge.
Software engineering / functional-to-object-oriented or vice-versa: Programmers trained in one paradigm (functional programming, object-oriented programming, procedural programming) often struggle when transitioning to another paradigm because the trained reflexes interfere with the new paradigm's requirements. Senior programmers who have specialized in one paradigm for decades often report the transition is harder than learning programming from scratch would be — the un-learning is more difficult than fresh-learning. Put programming-paradigm transitions alongside writing-pedagogy transitions and you see writing's un-learning challenges are part of a broader pattern: deeply-internalized first-discipline reflexes interfere with second-discipline acquisition more than absence of any prior discipline would. The implication for writing-craft pedagogy: storytelling-writing pedagogy should explicitly address the un-learning of school-pedagogy reflexes as a load-bearing curricular element, not assume the un-learning will happen automatically.
The Sharpest Implication The principle implies that the largest single barrier between adult-writers' current skill and storytelling-craft mastery is years of school-pedagogy reflexes that need to be deliberately un-learned. Most adult writers underestimate how much of their default writing-mode is school-pedagogy operating in the background. The topic-sentence reflex is one specific case; the broader pattern includes the five-paragraph-essay structure, the thesis-statement-in-introduction reflex, the conclusion-restates-thesis reflex, the formal-vocabulary preference. Each is the school-pedagogy operating below conscious awareness. The implication: storytelling-writing development requires not just learning new techniques but un-learning the embedded school-pedagogy patterns that block the new techniques from operating. Writers who skip the un-learning step plateau at a ceiling set by their school-pedagogy reflexes regardless of how much storytelling-craft they study.
Generative Questions