literary-research

name

literary-research

description

Guide for conducting literary research as a novelist — the craft of extracting fiction-ready material (scenes, voices, tensions, images) from sources through structured dialogue, interviews, field work, and document analysis. Use this skill when asked to help with novel research, to design research protocols for fiction, to extract usable material from conversations or documents, to evaluate whether gathered material serves a narrative, or to plan research phases for a novel-in-progress. Also use when building dialogue systems, interview frameworks, or character-mining processes for literary projects. Trigger on: "novel research", "research for fiction", "literary field work", "mining for material", "character research", "dialogue-as-research", "fiction preparation", or any request to systematically gather and evaluate material for a literary work.

Literary Research for Novel Writing

A craft guide for novelists doing research — not the academic kind, but the kind that produces fiction-ready material: scenes you can use, voices you can hear, tensions that generate plot, images that survive revision.

Based on the practice of writers who treat research as a creative act: Ted Chiang's philosophical scaffolding, W.G. Sebald's archival wandering, Elena Ferrante's social excavation, Roberto Bolaño's biographical mining, Guimarães Rosa's sertão fieldwork (the 1952 cattle drive notebooks), and the epistolary research traditions from Rilke's letters to Flaubert's correspondence.

---

The Core Distinction

Academic research answers questions. Literary research generates material.

Material is not information. Information tells you what happened. Material lets you write what happened as if you were there. Material has:

• Voice: you can hear someone saying it
• Image: you can see something specific
• Tension: something unresolved, something that resists easy resolution
• Physicality: sensory detail — what it smelled like, what the light was doing, what sound was underneath the silence

If your research produces facts but no material, you have notes. If it produces material, you have the raw substance of fiction.

---

Five Research Modes

1. Dialogue-as-Research

The most productive mode for character-driven fiction. You talk to someone — real or constructed — not to extract information but to provoke material.

The interviewer's posture:

• You are not a journalist. You don't need the whole story. You need the moment — the detail, the image, the phrase that survives compression
• You are not a therapist. You don't need emotional truth. You need dramatic truth — the version of events that generates conflict, tension, surprise
• You are not a scholar. You don't need accuracy. You need vividness — the wrong detail remembered intensely is worth more than the right detail remembered flatly

Technique — Provocative Presentation:

Present your thesis (the philosophical, structural, or thematic spine of the novel) to your research subject in accessible language. Then listen to what comes back. The research subject's resistance is more valuable than their agreement. Where they push back, where they misunderstand productively, where they offer a counter-example from experience — that's where the fiction lives.

What to capture:

• Exact phrases (not paraphrased — the words they actually used)
• Stories that came unbidden (the memory that surfaced because you said something that triggered it, not because you asked for it)
• Reformulations (when they say your idea back to you in their own language, and the translation changes the idea)
• Questions (especially questions you can't answer — these become the novel's open wounds)
• Disagreements (every disagreement is a potential scene)

Session structure:

1. Respond to whatever they said last (never ignore their previous contribution)
2. Present one idea — only one — in concrete, physical language
3. Ask genuinely: what does this remind you of? How would you say this?
4. Listen for the material. It's almost never in the direct answer. It's in the digression, the aside, the story they tell to avoid answering

2. Archival Wandering

Research in documents, photographs, maps, records — not looking for specific information but for the accidental detail that no one would invent.

The Sebald method:

• Go to archives without a precise question
• Read adjacent materials (the document next to the one you came for)
• Follow names, dates, and places that appear in margins
• Photograph everything; sort later
• The connection between two unrelated documents is the plot

What to capture:

• The specific rather than the general (not "prices were high" but "a loaf of bread cost three days' wages")
• Procedural details (how things were actually done — the sequence of steps in a process, the tools used, the time it took)
• Proper nouns (real names, real places, real streets — even if you'll change them later, they anchor imagination)
• Errors and anomalies in records (where documents contradict each other, there's a story)

3. Field Work

Going to the place. Walking the ground. Rosa's 1952 cattle drive — ten days on horseback across the sertão, filling notebooks — produced the sensory foundation for everything he wrote afterward.

What to capture:

• Sensory inventory: what each sense registers, hour by hour. The light at 6am is different from the light at 2pm. The sound of a place at night is different from the sound at noon. Write it all.
• Movement vocabulary: how people walk, gesture, sit, eat, work in this place. The body language of a specific geography and culture.
• Overheard language: not interviews — fragments. The phrase you catch walking past a conversation. The way someone calls a child. The word for a thing that has no word elsewhere.
• Maps: not cartographic maps — emotional maps. Where do people go? What do they avoid? What's the dangerous part? What's the sacred part?

4. Philosophical Scaffolding

The Ted Chiang mode: start with an idea — a thesis, a thought experiment, a what-if — and build the novel as a structure that lets the reader experience the idea rather than understand it.

Process:

1. Articulate the thesis in the clearest, most compressed form possible (one paragraph)
2. Identify the experiential core: what would it feel like to live this thesis? What would change in daily life if this were true?
3. Find the character who lives at the fault line of the thesis — the person for whom this idea is not abstract but urgent, personal, survival-level
4. The research question becomes: what does this person's life look like? What are the ordinary details of an extraordinary situation?

The scaffolding must be invisible. If the reader can see the thesis, the novel has failed. The idea should be the skeleton, not the skin.

5. Biographical Mining

Using a real or fictional person's life as a quarry. Not biography — mining. You take what serves the fiction and leave the rest.

Ethical principle: You owe the material your full attention. You don't owe it accuracy. The moment you subordinate the fiction to the facts, you've stopped writing a novel and started writing a profile.

What to mine:

• Contradictions in character (the brave person's cowardice, the generous person's hidden cruelty)
• Turning points (moments where the person became someone else — and the detail that reveals the turn)
• Relationships as systems (not "A loved B" but "A needed B to be X, and B needed A to be Y, and neither could deliver")
• The unsaid (what the person never talks about — the gap in the narrative is the most interesting part)

---

The Material Evaluation Framework

Not everything gathered is usable. Use these filters:

The Five Tests

1. Can I hear it? If the material includes a voice — a phrase, a rhythm, a way of speaking — it's strong. Voiceless material is weak.

2. Can I see it? If the material produces a specific image (not a concept, an image — a color, a gesture, a spatial arrangement), it's strong.

3. Does it resist? If the material is comfortable, agrees with your thesis, confirms what you already thought — it's weak. The best material pushes back, complicates, refuses to fit.

4. Is it surprising? If you could have invented it without doing the research, you didn't need the research. The value of research is what you couldn't have imagined.

5. Does it compress? Can this material become a scene? A sentence? An image? If it only works as exposition or explanation, it belongs in your notes, not your novel.

The Three Categories

After gathering material, sort into:

• Scenes: material that has enough specificity (place, voice, action, tension) to become a scene in the novel. These are gold.
• Texture: material that provides sensory or cultural detail — vocabulary, gestures, smells, procedures — that makes scenes believable. This is the mortar between the bricks.
• Spine: material that clarifies, complicates, or restructures the novel's thesis or architecture. This changes the plan. It's the most valuable and the most dangerous category.

---

Research Documentation

The Research Journal

Not a diary. A working document that evolves across sessions.

After each research session, record:

1. Best material (exact quotes, images, stories — with source reference)
2. Surprises (what you didn't expect to find)
3. Resistances (where your thesis was challenged — and whether the challenge was productive)
4. Connections (how today's material links to material from previous sessions)
5. Questions generated (what you now need to find out that you didn't know you needed)
6. Novel implications (how this material changes the plan — new scenes, new characters, structural shifts)

The Material Index

A running inventory of usable material, organized by category:

• Phrases (exact language that could appear in the novel — always with source)
• Scenes (material that has enough body to become a scene — one-line summary + source)
• Characters (traits, contradictions, gestures, speech patterns extracted from research)
• Images (specific sensory details that anchor the fictional world)
• Tensions (unresolved questions, disagreements, contradictions — the dramatic fuel)

The Living Blueprint

The novel plan that changes as research accumulates. Key principles:

• The blueprint is never finished until the novel is finished
• Every research session has the right to alter the blueprint
• Material that doesn't fit the current blueprint is not discarded — it's stored. It may be the material that eventually breaks the blueprint open into something better
• The blueprint tracks what's covered and what's missing — not just thematically but materially. You may have the thesis for chapter 7 but no scene, no voice, no image. That's a research gap.

---

Research Rhythms

The Session

One focused research interaction (a dialogue, an archive visit, a field day). Produces:

• Raw material (notes, quotes, images)
• A journal entry
• Updates to the material index

The Phase

A cluster of sessions focused on a particular aspect of the novel (a character, a setting, a thematic strand). A phase ends when you have enough material to draft — not enough material to finish, enough to start.

The Spiral

You will return to the same sources, the same people, the same places — but with different questions, because the novel has changed since the last visit. The second visit produces different material from the first. The third produces different material from the second. Research is not linear; it spirals.

---

The Novelist's Ethic of Research

1. Respect your sources. The people and places you research are not instruments. Give them your full attention. Listen more than you speak. If someone gives you a story, that's a gift.

2. Steal well. All fiction is theft. The art is in what you steal and how you transform it. Take the phrase, the gesture, the image — but make it serve a new context, a new meaning. The transformation is the craft.

3. Know when to stop. Research can become avoidance. When you have enough material to draft, draft. You can always research more later. But the novel exists only when you start writing it.

4. Let the material change the plan. If your research contradicts your outline, the research is probably right. The world is more interesting than your plan for it. The best novels are the ones that outgrew their blueprints.

5. The research is not the novel. No matter how beautiful your material, the reader will never see it directly. They'll see the fiction you built from it. The research is scaffolding. The novel is the building. When the building stands, the scaffolding comes down.