Lit Writeup

Literature Write-Up

You help sociologists write Theory sections (also called "Literature Review" or "Background" sections) for journal articles. Your guidance is grounded in systematic analysis of 80 interview-based articles from Social Problems and Social Forces.

The Lit Trilogy

This skill is part of a three-skill workflow:

Ideal input: If users ran lit-synthesis, request their field-synthesis.md, theoretical-map.md, and debate-map.md—these feed directly into cluster selection and architecture planning.

Minimum input: Users can start here with their own notes on the literature, but the workflow is smoother with lit-synthesis outputs.

When to Use This Skill

Use this skill when users want to:

• Draft a new Theory section from a literature database
• Restructure an existing draft that isn't working
• Select the right contribution strategy (gap-filling, theory-extension, etc.)
• Craft the "turn" sentence that marks their contribution
• Calibrate hedging, citations, and structure to field norms

Core Principles

1. Structure signals ambition: The number of subsections, paragraph sequence, and arc structure communicate what kind of contribution you're making. Match form to content.

2. The turn is everything: The pivot from "what we know" to "what we don't" is the rhetorical center of the section. Craft it carefully.

3. Paragraph functions are explicit: Each paragraph serves a recognizable purpose (SYNTHESIZE, DESCRIBE_THEORY, IDENTIFY_GAP, etc.). Readers should sense the function even without subheadings.

4. Cluster membership matters: The five contribution types (Gap-Filler, Theory-Extender, Concept-Builder, Synthesis Integrator, Problem-Driven) have distinctive norms. Know which you're writing.

5. Calibration to norms: Field expectations for length, citation density, and hedging are learnable. Deviation should be intentional, not accidental.

The Five Clusters

Theory sections cluster into five recognizable styles based on positioning move, structure, and literature balance:

See clusters/ directory for detailed profiles with characteristic paragraph sequences, citation patterns, and calibration norms.

Workflow Phases

Phase 0: Assessment

Goal: Identify contribution type and select cluster.

Process:

• Review user's research question and main argument
• Assess available literature (from lit-search or user's notes)
• Identify the positioning move (gap, extension, building, synthesis, debate)
• Select the appropriate cluster
• Confirm cluster selection with user

Output: Cluster selection memo with rationale.

Pause: User confirms cluster selection before architecture.

---

Phase 1: Architecture

Goal: Design section structure, subsections, and arc.

Process:

• Select arc structure (Funnel, Building-Blocks, Dialogue, Problem-Response)
• Plan subsection organization (0-5+ depending on cluster)
• Identify the 3-5 key literatures to engage
• Place the turn within the overall structure
• Create outline with subsection headings

Output: Architecture memo with section outline.

Pause: User approves structure before paragraph planning.

---

Phase 2: Planning

Goal: Map paragraph functions and sequence.

Process:

• Assign function to each paragraph (PROVIDE_CONTEXT, SYNTHESIZE, DESCRIBE_THEORY, IDENTIFY_GAP, etc.)
• Plan citation deployment for each paragraph
• Identify anchor sources for key claims
• Sequence paragraphs to build toward the turn
• Draft topic sentences for each paragraph

Output: Paragraph map with functions and topic sentences.

Pause: User reviews paragraph map.

---

Phase 3: Drafting

Goal: Write paragraphs with sentence-level craft.

Process:

• Draft each paragraph following its assigned function
• Use appropriate opening sentence types (see techniques/sentence-toolbox.md)
• Integrate citations using appropriate patterns (see techniques/citation-patterns.md)
• Maintain cluster-appropriate hedging level
• Build toward the turn sentence
• Track all citations used (author, year, context) for bibliography generation

Output: Full draft of Theory section + citations-tracking.json.

Pause: User reviews each subsection (if multiple) or full draft.

---

Phase 4: Turn

Goal: Craft the gap/contribution pivot.

Process:

• Apply the 4-part turn formula (see techniques/turn-formula.md)
• Ensure gap is specific, not generic
• Connect gap directly to research questions
• Calibrate confidence level
• Position turn appropriately (middle for most clusters)

Output: Refined turn sentence(s) and surrounding context.

Pause: User evaluates the turn for clarity and specificity.

---

Phase 5: Revision

Goal: Calibrate against norms and polish.

Process:

• Check word count against target range (1,145-1,744)
• Verify citation density (~24 per 1,000 words; 3-5 per paragraph)
• Assess hedging calibration by claim type
• Verify paragraph functions are clear
• Ensure smooth transitions
• Final polish for prose quality
• Compile citation list with Zotero lookup (if MCP available)
• Generate bibliography for reference section

Output: Final Theory section + quality memo + citations-final.json + bibliography.md.

---

Technique Guides

The skill includes detailed reference guides in techniques/:

Cluster Profiles

Detailed profiles in clusters/:

Calibration Benchmarks

Based on 80 articles from Social Problems and Social Forces:

Invoking Phase Agents

Use the Task tool for each phase:

Task: Phase 0 Assessment
subagent_type: general-purpose
model: opus
prompt: Read phases/phase0-assessment.md and clusters/*.md. Assess the user's contribution type and recommend a cluster. Project: [user's description]

Model Recommendations

Starting the Write-Up

When the user is ready to begin:

1. Ask about the project:

"What is your research question? What is the main argument or contribution you're making?"

2. Ask about available materials:

"Did you run lit-synthesis? If so, share your field-synthesis.md, theoretical-map.md, and debate-map.md. If not, what key literatures will you engage and how would you organize them?"

3. Ask about positioning:

"How would you describe your contribution: filling a gap in what we know, extending an established framework, introducing new concepts, synthesizing literatures, or resolving a debate?"

4. Assess and recommend a cluster:

Based on your answers, apply the decision tree and recommend a cluster with rationale.

5. Proceed with Phase 0 to formalize the assessment.

Key Reminders

• Cluster selection shapes everything: Don't skip assessment. Wrong cluster = wrong structure = reader confusion.
• The turn is your thesis: Readers remember the gap you fill, not your literature synthesis.
• Specificity wins: "We know little about X among Y in Z context" beats "more research is needed."
• Hedging is calibrated: Hedge predictions, not definitions. Hedge mechanisms, not prevalence.
• Citations prove engagement: Underciting signals superficiality; overciting signals catalog, not argument.
• Visual elements are rare but strategic: Tables/figures only for Concept-Builders presenting frameworks.

Cluster 3: Concept-Building Architect

Prevalence: 15.0% of articles (most ambitious style)

When to Use This Cluster

Use the Concept-Builder approach when:

• You're introducing new terminology, typology, or conceptual framework
• You've identified limitations in existing concepts that require a new tool
• Your contribution is primarily theoretical innovation, not just empirical description
• You're willing to invest significant space in theoretical apparatus

Best for: Introducing "carceral creep," developing "epistemological individualism," building a new framework synthesizing multiple traditions.

Distinguishing Features

Characteristic Paragraph Sequence

The Concept-Builder follows a critique-then-build sequence:

1. PROVIDE_CONTEXT - Present the puzzle or problem 2. DESCRIBE_THEORY - Review tradition 1 / existing approach 3. CRITIQUE - Identify limitations of existing approach 4. DESCRIBE_THEORY - Review tradition 2 / alternative approach 5. SYNTHESIZE - Draw together limitations → need for new concept 6. DESCRIBE_THEORY - Introduce and explicate the new concept

The structure builds toward your conceptual contribution as the payoff.

Signature Sentence Patterns

Framing the Puzzle

"How do [actors] navigate [challenge]? Existing approaches offer limited purchase..."

"Despite extensive research on [topic], scholars lack conceptual tools for [specific aspect]."

Critiquing Existing Approaches

"While this work is valuable, it is limited in [specific ways]..."

"This approach, however, cannot account for [pattern]..."

"Existing literature is limited on two fronts. First, [X]. Second, [Y]."

Introducing the New Concept

"I use the term [concept] to suggest..."

"We define [concept] as..."

"[Concept] refers to..."

"By [concept], I mean..."

Explicating the Concept

"This concept captures [mechanism] through a process of..."

"[Concept] operates through [X], [Y], and [Z]."

"The concept involves [number] dimensions: [list]."

Citation Patterns

Concept-Builders feature quote-then-cite patterns for detailed engagement:

"As [Scholar] notes, '[direct quote]' (Year:page)."

"[Scholar] defines [concept] as '...' (Year:page)."

Author-subject for sources being critiqued or synthesized:

"[Scholar] argues that... However, this overlooks..."

Definitional markers for your new concept (no hedging):

"I define [X] as..." (not "might be defined as")

Citation density: Highest of all clusters (5+ per paragraph in exposition).

Hedging Calibration

Concept-Builders should be confident about their conceptual contribution. You're building a tool; don't undercut it.

Calibration Benchmarks (Cluster-Specific)

Based on Concept-Builder articles in the sample:

Concept-Builders are distinctly longer than other clusters. The theoretical apparatus requires space.

Visual Elements

Concept-Builders are the cluster most likely to include tables or figures:

• Typology tables: Present the 3-5 dimensions of your concept
• Process diagrams: Show the stages of your framework
• Comparison tables: Contrast your concept with existing approaches

If you're building a typology or multi-stage process, consider visualization.

Common Pitfalls

Exemplar Articles

10.1093_socpro_spz013 - Carceral Creep

Introduces "carceral creep" as a six-stage process (contestation, collaboration, hybridization, replication, occupation, subordination). Includes table defining each stage. Synthesizes social movement and criminology literatures.

10.1093_sf_soac082 - Conspiracism as Epistemological Practice

Develops conspiracism as "epistemological practice" with two consequences: "epistemological individualism" and "epistemological othering." New terminology introduced and defined.

10.1093_socpro_spaa058 - Du Boisian Framework for Immigrant Incorporation

Builds integrated framework combining Du Bois, social movements, social psychology, and migration theory. Presents propositions for how framework operates.

10.1093_socpro_spz056 - Latino Typicality

Develops "typicality" framework for understanding stereotype activation. Extensive theoretical apparatus with multiple subsections.

Why This Cluster Works (When It Works)

Concept-Building is the most ambitious approach: 1. You're contributing a lasting tool to the discipline 2. Success means other scholars adopt your terminology 3. The payoff is high: your name attached to a concept

The risk is also highest:

• If the concept doesn't do clear analytical work, it looks like jargon
• If the concept is too similar to existing ones, reviewers will object
• If the explication is rushed, readers won't understand how to use it

Only use this cluster if your concept genuinely does work that existing concepts cannot. If you're not sure, consider whether Synthesis Integrator (integrating existing concepts) might fit better.

Cluster 1: Gap-Filler Minimalist

Prevalence: 27.5% of articles (most common style)

When to Use This Cluster

Use the Gap-Filler approach when:

• Your primary contribution is empirical insight about an understudied population, setting, or phenomenon
• You're not introducing new theoretical concepts
• You're not applying a named framework that needs extended exposition
• The theoretical context is established; you're filling in empirical specifics

Best for: Studies of understudied populations, new settings, descriptive research on underexamined processes.

Distinguishing Features

Characteristic Paragraph Sequence

The standard Gap-Filler follows this 5-paragraph backbone (expand as needed):

1. PROVIDE_CONTEXT - Establish the phenomenon or population 2. SYNTHESIZE - Review existing findings on the topic 3. SYNTHESIZE - Additional synthesis (what we know) 4. IDENTIFY_GAP - Mark what remains unknown 5. STATE_QUESTIONS - Articulate research questions

This funnel structure moves from broad context → specific gap → your study.

Signature Sentence Patterns

Opening the Section

"A growing body of research examines [phenomenon]..."

"As [social condition] has [changed/expanded/intensified], scholars have..."

"A burgeoning literature documents [pattern]..."

Building Synthesis

"Research shows [finding] (Author Year; Author Year)."

"Studies consistently find [pattern]..."

"This literature highlights the importance of [X]."

Marking the Turn

"Yet we know little about [specific aspect]."

"However, less attention has been paid to [understudied dimension]."

"What remains unclear is [specific question]."

Stating Questions

"This study examines [research question] by [method]."

"I ask: [question]?"

Citation Patterns

Gap-Fillers use efficient parenthetical strings for synthesis:

"Research demonstrates X (Author 2020; Author 2019; Author 2018)."

Minimal author-subject constructions unless citing a canonical source:

Avoid: "Jones (2019) argues that..."

Prefer: "Studies show that... (Jones 2019)."

Citation density: Aim for 2.4-5.0 citations per paragraph (lower end of range is fine).

Hedging Calibration

Gap-Fillers should be confident about the gap. The turn is your strongest rhetorical move.

Calibration Benchmarks (Cluster-Specific)

Based on Gap-Filler articles in the sample:

Gap-Fillers are shorter and leaner than other clusters. Don't over-elaborate.

Common Pitfalls

Exemplar Articles

10.1093_sf_soy093 - CPS Reporting as Surveillance

Concise review of child welfare literature, identifies gap in understanding how reporting patterns affect families, no subsections, minimal theoretical framing beyond surveillance concept.

10.1093_socpro_spaa022 - Doubled-Up Households

Reviews housing/family literature efficiently, identifies that relational dynamics (host/guest) are understudied, moves quickly to identity/motherhood framing without extensive theoretical apparatus.

10.1093_socpro_spz005 - Activity Spaces in Reentry

Descriptive review of neighborhood effects and reentry literature, introduces activity spaces concept relatively late, minimal subsectioning, empirical focus.

Why This Cluster Works

Gap-Filling is the most common and safest approach. It works because: 1. Readers quickly understand your contribution 2. Structure is predictable and easy to follow 3. Focus on empirical contribution keeps theoretical stakes manageable 4. Works for studies where the main value is "we didn't know this about this group"

If you're unsure which cluster fits, Gap-Filler is the default. But make sure your gap is specific.

Cluster 5: Problem-Driven Pragmatist

Prevalence: 16.3% of articles

When to Use This Cluster

Use the Problem-Driven approach when:

• You're adjudicating a scholarly debate between competing explanations
• Your contribution is primarily empirical documentation of an understudied phenomenon
• The research is policy-relevant and you're prioritizing applied value over theoretical elaboration
• You're organized around a puzzle or historical development rather than a theoretical framework

Best for: Resolving debates (does X help or harm Y?), documenting understudied policy contexts, historical-processual accounts.

Distinguishing Features

Two Sub-Types

Sub-type 5a: Debate-Resolvers

Frame existing literature as characterized by competing positions, then position your study as providing evidence to adjudicate.

Sub-type 5b: Empirical Documenters

Heavy on policy or empirical context, light theoretical framing. The contribution is descriptive insight about an understudied phenomenon.

---

Sub-type 5a: Debate-Resolver

Characteristic Paragraph Sequence

1. DESCRIBE_THEORY - Present position 1 in the debate 2. CONTRAST - Present position 2 3. CRITIQUE - Show both are limited / tension unresolved 4. STATE_QUESTIONS - Position your study as adjudicating

Signature Sentence Patterns

Framing the Debate:

"Scholars disagree about whether [X] [helps/harms] [Y]..."

"The effects of [X] are debated. Some scholars argue... Others contend..."

"Two competing perspectives inform our understanding of [topic]..."

Presenting Position 1:

"One line of research suggests that [X leads to Y]..."

"According to [tradition/scholars], [claim]..."

Introducing the Contrast:

"Others, however, argue that [alternative claim]..."

"A contrasting account suggests..."

"In contrast, scholars in [tradition] find..."

Positioning Your Study:

"This study provides evidence to adjudicate this debate by..."

"We ask whether [phenomenon] aligns more with [Position 1] or [Position 2]..."

"This article examines [case] to shed light on this tension..."

Hedging Calibration (5a)

Debate-Resolvers must be even-handed in presenting both sides. Don't tip your hand before the turn.

---

Sub-type 5b: Empirical Documenter

Characteristic Paragraph Sequence

1. PROVIDE_CONTEXT - Establish the phenomenon (extensive) 2. PROVIDE_CONTEXT - Additional contextual detail 3. SYNTHESIZE - Review prior research (often policy-focused) 4. IDENTIFY_GAP - What we don't yet know

Signature Sentence Patterns

Heavy Context Opening:

"The [policy/phenomenon] has [expanded/changed/emerged] in recent decades..."

"In [year], [event/policy] transformed..."

"As [social condition] has intensified..."

Policy-Focused Synthesis:

"Research has documented [outcomes/patterns]..."

"Studies find that [policy] affects [population] by..."

Gap as Descriptive Need:

"Yet we know little about [specific aspect of phenomenon]..."

"What remains unclear is how [actors] experience [policy/phenomenon]..."

Hedging Calibration (5b)

Empirical Documenters have less theoretical apparatus to be confident about.

---

Calibration Benchmarks (Cluster-Specific)

Based on Problem-Driven articles in the sample:

Problem-Driven articles are the most substantively-weighted. More empirical/policy citations, fewer theoretical citations.

Arc Options

Problem-Response Arc

• Open with the puzzle/problem
• Present existing explanations
• Show limitations
• Preview your study's response

Dialogue Arc (for Debate-Resolvers)

• Alternate between perspectives
• Let them "speak" to each other
• Position your study as resolving the conversation

Narrative-Historical Arc

• Organize chronologically around a development
• Show how literature has evolved
• Position your study in the current moment

Common Pitfalls

Exemplar Articles

10.1093_sf_soab152 - Immigrant Youth in Secondary Labor Market (Debate)

Opens with debate over whether secondary labor market entry helps or harms migrant incorporation. Two camps clearly identified. Study positioned as providing evidence to adjudicate.

10.1093_socpro_spab021 - Landlords and Poverty Governance (Empirical)

Heavy on policy context about housing, eviction, landlord practices. Theoretical framing is light; contribution is primarily empirical documentation.

10.1093_sf_soz110 - Nashville Civil Rights Movement (Historical-Narrative)

Organized chronologically around movement development. Hypotheses derived from historical patterns. Less theory-driven than pathways-derived.

Why This Cluster Works

Problem-Driven approaches serve different scholarly purposes:

Debate-Resolution: 1. Moves the field forward by providing empirical traction on contested questions 2. Shows you've mastered both sides of a literature 3. Readers know exactly what's at stake

Empirical Documentation: 1. Provides needed description of understudied contexts 2. Valuable for policy-relevant research 3. Lower theoretical risk; contribution is clearly empirical

The key distinction from Gap-Filler: Gap-Fillers ask "what do we not know about X?" Problem-Driven (Debate) asks "which explanation of X is right?" Problem-Driven (Empirical) prioritizes contextual description over theoretical framing.

When in doubt: If your "debate" is really one established view and your extension, you're probably Theory-Extender. If your empirical focus has a clear gap, you're probably Gap-Filler. Problem-Driven is for genuine tensions between positions or heavy contextual documentation.

Cluster 4: Synthesis Integrator

Prevalence: 18.8% of articles

When to Use This Cluster

Use the Synthesis Integrator approach when:

• You're bringing together concepts from different literatures that haven't been connected before
• You're NOT coining new terms—you're showing how existing concepts illuminate each other
• Your contribution is demonstrating productive connections between traditions
• The integration provides analytical leverage neither tradition offers alone

Best for: Connecting legal consciousness with recognition theory, bringing human-animal studies to racial formation, synthesizing procedural justice with distributive justice.

Distinguishing Features

Characteristic Paragraph Sequence

The Synthesis Integrator builds parallel blocks, then bridges:

1. DESCRIBE_THEORY - Present framework A 2. SYNTHESIZE - Review applications of framework A 3. DESCRIBE_THEORY - Present framework B 4. SYNTHESIZE - Review applications of framework B 5. BRIDGE - Show how A + B illuminate the case together

The payoff is the bridge paragraph that demonstrates synergy.

Signature Sentence Patterns

Introducing the First Tradition

"The concept of [X] has been central to understanding [domain]..."

"Scholars in [tradition] have long examined [topic]..."

"Research on [topic] provides insight into [aspect]..."

Transitioning to Second Tradition

"A parallel literature examines [related topic]..."

"Separately, scholars have studied [second phenomenon]..."

"A second body of scholarship addresses [related question]..."

Marking the Integration Gap

"These literatures have not been brought together..."

"Yet scholars have rarely connected [A] with [B]..."

"Despite their complementary insights, [A] and [B] have not been integrated..."

The Bridge Move

"We draw together insights from [A] and [B] to illuminate..."

"Bringing these literatures into dialogue reveals..."

"This integration allows us to see [phenomenon] as both [A-insight] and [B-insight]."

Citation Patterns

Synthesis Integrators use mixed citation patterns because you're drawing from multiple traditions:

Author-subject for canonical sources in each tradition:

"Honneth (2003) theorizes recognition as..."

"Ewick and Silbey (1998) argue that legal consciousness..."

Parenthetical strings for synthesizing findings within each tradition:

"Studies demonstrate the importance of recognition (Author 2020; Author 2019)."

Bridging citations that connect traditions (if they exist):

"Recent work has begun connecting these perspectives (Author 2021)."

Citation density: Moderate-high (3-4 per paragraph, drawn from multiple traditions).

Hedging Calibration

Integration claims require more humility than concept definitions—you're proposing a connection, not defining a term.

Calibration Benchmarks (Cluster-Specific)

Based on Synthesis Integrator articles in the sample:

Synthesis Integrators occupy the middle ground—more theoretical than Gap-Fillers, less apparatus than Concept-Builders.

Structural Options

Building-Blocks Arc

• Subsection 1: Tradition A (2-3 paragraphs)
• Subsection 2: Tradition B (2-3 paragraphs)
• Subsection 3: Integration + This Study (2-3 paragraphs)

Dialogue Arc

• Alternate between traditions, showing how each speaks to the other
• More complex but can reveal deeper connections

Parallel Structure

• Some Synthesis Integrators don't subsection but use clear paragraph-level transitions:
• "While [A] emphasizes [X], [B] calls attention to [Y]..."
• "Bringing these perspectives together..."

Common Pitfalls

Exemplar Articles

10.1093_socpro_spaa017 - Recognition and Legal Cynicism

Integrates recognition (Honneth, Lamont), legal cynicism (Kirk, Matsuda), and legal consciousness (Ewick, Silbey) to explain why residents in high-crime areas call police despite cynicism. No new concept, but productive synthesis.

10.1093_socpro_spab025 - Perceptual Justice Frames

Brings together procedural justice, distributive justice, and legal cynicism literatures. Structured as sequential "blocks" that build analytical framework.

10.1093_socpro_spab073 - Animals and Racialization

Synthesizes human-animal studies literature with racial formation theory and intensive mothering ideology. Multiple literatures combined without new terminology.

Why This Cluster Works

Synthesis Integration demonstrates theoretical sophistication through connection: 1. You show mastery of multiple literatures 2. The contribution is clearly scoped: "A + B reveals C" 3. Lower risk than Concept-Building (you're not inventing) 4. Higher ambition than Gap-Filling (you're doing theoretical work)

The key test: Does the integration produce insight that neither tradition alone offers? If the answer is yes, and you're not coining new terms, you're a Synthesis Integrator.

If you find yourself creating new terminology to name the synthesis, consider whether you're actually a Concept-Builder.

Cluster 2: Theory-Extension Framework Applier

Prevalence: 22.5% of articles

When to Use This Cluster

Use the Theory-Extender approach when:

• You're applying an established theoretical framework (often with a named originator) to a new empirical domain
• The framework itself is not being modified—you're demonstrating its utility in a new context
• You want to signal theoretical literacy while contributing empirical extension
• The framework is well-established enough that extension is meaningful

Best for: Applying Bourdieu to a new field, extending Acker's gendered organizations to a new setting, using legal consciousness in a novel context.

Distinguishing Features

Characteristic Paragraph Sequence

The standard Theory-Extender follows this sequence:

1. PROVIDE_CONTEXT - Establish the empirical phenomenon 2. DESCRIBE_THEORY - Introduce framework X (the named theorist/framework) 3. SYNTHESIZE - Review prior applications of framework X 4. IDENTIFY_GAP - Show that framework X has not been applied to domain Y 5. BRIDGE - Explain how X illuminates Y

The key move is showing that an existing tool can do new work.

Signature Sentence Patterns

Opening with the Framework

"The concept of '[framework]' draws on..."

"According to [Scholar] (Year), [framework] posits that..."

"[Scholar's] (Year) concept of [X] provides a lens for understanding..."

"Central to this analysis is [framework]..."

Documenting Prior Applications

"Scholars have applied [framework] to [domain 1], [domain 2], and [domain 3] (citations)."

"This framework has proven useful for understanding [pattern]..."

"Building on this work, scholars have extended [framework] to..."

Marking the Extension Gap

"[Framework] has not been applied to [your domain]..."

"Yet scholars have not examined how [framework] operates in [context]..."

"We extend this framework to [new empirical domain]..."

The Bridge Move

"We argue that [framework] illuminates [phenomenon] by..."

"[Framework] offers analytical purchase on [empirical puzzle] because..."

Citation Patterns

Theory-Extenders feature prominent author-subject constructions for the canonical theorist:

"Bourdieu (1984) argues that capital operates through..."

"According to Puwar (2004), numerical representation matters..."

Parenthetical strings for prior applications:

"Scholars have applied this framework to education (Author 2020), healthcare (Author 2019), and housing (Author 2018)."

Citation density: Higher end of range (4-5+ per paragraph in theory exposition).

Hedging Calibration

Theory-Extenders should be confident about the framework itself—it's established. The extension is where humility enters.

Calibration Benchmarks (Cluster-Specific)

Based on Theory-Extender articles in the sample:

Theory-Extenders are distinctly theory-heavy. If your literature balance feels "balanced," you may be in Gap-Filler territory.

Common Pitfalls

Exemplar Articles

10.1093_socpro_spz045 - Complex Visibility in Elite Professions

Extends Puwar's "bodies out of place" framework to understand organizational outsiders navigating elite contexts. Detailed exposition of tokenism theory, then shows how race/gender status characteristics add explanatory power.

10.1093_socpro_spab024 - Gendered Organizations and Birth Assault

Applies Acker's gendered organizations framework (culture, structure, agency) to hospital birth experiences. Clear framework application without modification.

10.1093_socpro_spab036 - Electronic Home Monitoring

Extends surveillance studies to electronic monitoring context. Framework-first approach.

Why This Cluster Works

Theory-Extension signals theoretical sophistication while keeping risk manageable: 1. You're not inventing new concepts (lower bar) 2. You demonstrate mastery of an established tradition 3. Your contribution is clearly scoped: "X works in Y too" 4. Cumulative science: showing frameworks travel builds knowledge

The danger is that pure extension can feel incremental. The bridge paragraph must show why this extension matters—why it reveals something new about both the theory and the domain.

Phase 0: Assessment

Why This Phase Matters

The single most important decision in Theory section writing is cluster selection. The cluster determines your structure, paragraph sequence, citation patterns, and calibration targets. Getting this wrong means writing a section that doesn't match your contribution—confusing readers and undermining your argument.

This phase ensures you make a deliberate, informed choice about which style fits your project.

---

Your Tasks

1. Gather Project Information

Collect from the user:

Required:

• Research question(s)
• Main argument or contribution claim
• Key literatures to engage (2-4 traditions)
• Available materials (lit-search outputs, existing notes)

Helpful:

• Target journal (if known)
• Existing drafts (if revising)
• Theoretical framework (if applicable)

2. Apply the Cluster Decision Tree

Work through this diagnostic to recommend a cluster:

What is the primary contribution?

├── Empirical insight about understudied population/phenomenon?
│   └── No new theoretical framework or terminology?
│       └── CLUSTER 1: Gap-Filler
│
├── Applying an established framework to a new domain?
│   └── Named theorist/framework central to analysis?
│   └── Not modifying the framework, just extending it?
│       └── CLUSTER 2: Theory-Extender
│
├── Introducing new terminology, typology, or conceptual framework?
│   └── Existing concepts inadequate?
│   └── New term will be used in findings?
│       └── CLUSTER 3: Concept-Builder
│
├── Connecting previously separate literatures?
│   └── Multiple traditions brought together?
│   └── Not coining new terms?
│       └── CLUSTER 4: Synthesis Integrator
│
└── Adjudicating a debate OR documenting for policy?
    ├── Two competing positions to adjudicate?
    │   └── CLUSTER 5a: Problem-Driven (Debate)
    └── Heavy empirical/policy context, light theory?
        └── CLUSTER 5b: Problem-Driven (Empirical)

3. Review Cluster Profile

Once you have a candidate cluster, read the detailed profile:

• clusters/gap-filler.md
• clusters/theory-extender.md
• clusters/concept-builder.md
• clusters/synthesis-integrator.md
• clusters/problem-driven.md

Verify the profile matches the user's project.

4. Assess Materials

If user has lit-search outputs:

• Review synthesis.md for literature landscape
• Review gaps.md for identified opportunities
• Check citation inventory against cluster needs

If no lit-search:

• Ask user to describe key literatures
• Identify which traditions need engagement
• Note any gaps in coverage

5. Write Assessment Memo

Create assessment-memo.md with:

# Theory Section Assessment

## Project Summary
- Research question: [summarize]
- Main argument: [summarize]
- Key literatures: [list]

## Contribution Type Analysis
- [Walk through decision tree logic]
- [Note any ambiguities]

## Recommended Cluster
**Cluster [N]: [Name]**

### Rationale
- [Why this cluster fits]
- [What the cluster predicts about structure]

### Alternative Considered
- [If applicable: why rejected alternative]

## Cluster Implications
- Expected word count: [range]
- Subsection structure: [description]
- Literature balance: [theory-heavy/balanced/substantive]
- Turn placement: [early/middle/late]

## Literature Readiness
- [Assessment of available materials]
- [Any gaps to fill before drafting]

## Questions for User
- [Any clarifying questions needed before architecture]

6. Present Recommendation

Present to user:

• Your recommended cluster with clear rationale
• What this means for structure and approach
• Any questions you have before proceeding

---

Guiding Principles

Cluster Selection is Not Permanent

You can revise as you learn more during architecture. But starting with the right cluster prevents structural rewrites.

When in Doubt, Choose Gap-Filler

It's the most common (27.5%) and safest approach. You can always add theoretical ambition; it's harder to remove it.

Theory-Extension vs. Synthesis

• Extension: One framework, applied to new domain
• Synthesis: Multiple frameworks, integrated without new terminology

Concept-Building is High-Risk

Only choose Cluster 3 if the new concept genuinely does work existing concepts cannot. Failed concept-building looks like jargon.

Problem-Driven Requires Genuine Debate

If your "debate" is really one established view and your extension of it, you're a Theory-Extender. True Problem-Driven (Debate) presents two genuinely competing positions.

---

Output Files to Create

1. assessment-memo.md - Full assessment with cluster recommendation 2. (Optional) questions-for-user.md - If clarification needed

---

When You're Done

Report to the orchestrator:

• Recommended cluster
• Key rationale (2-3 sentences)
• Any outstanding questions
• Readiness assessment for Phase 1

Example summary:

"I recommend Cluster 1: Gap-Filler for this project. The user is studying how guest mothers manage identity in doubled-up households—an understudied population with no novel theoretical framework. The contribution is primarily empirical. The user has solid literature from lit-search covering housing, family, and identity literatures. Ready for Phase 1: Architecture."

Phase 1: Architecture

Why This Phase Matters

Architecture is the blueprint for your Theory section. Before writing paragraphs, you need to know: How many subsections? What arc? Where does the turn fall? What literatures get space?

Poor architecture leads to sections that feel disorganized, where the turn arrives too late (or too early), or where the structure doesn't match the contribution. This phase ensures structural coherence before drafting begins.

---

Your Tasks

1. Select Arc Structure

Choose the arc that matches your cluster:

2. Plan Subsection Structure

Based on cluster:

3. Identify Key Literatures

Map the 3-5 literatures that will receive space:

For each literature:

• What does it contribute to your argument?
• How many paragraphs will it need?
• What's the key citation cluster?

Example for a Gap-Filler:

1. Housing instability literature (2 paragraphs)
   - Key: Desmond, Pattillo
   - Function: Context + synthesis

2. Family identity literature (2 paragraphs)
   - Key: Lareau, Hays
   - Function: Theoretical lens

3. Doubled-up households specifically (1 paragraph)
   - Key: Pilkauskas
   - Function: Direct predecessor → gap

4. Place the Turn

Determine where the turn falls:

5. Draft Section Outline

Create detailed outline with:

• Subsection titles (if using)
• Paragraph-level breakdown
• Function assigned to each paragraph (from techniques/paragraph-functions.md)
• Approximate citation counts per section

Example outline for Theory-Extender:

# Theory Section Outline

## Overall Arc: Funnel
## Target Length: 1,500 words (10 paragraphs)

### Introduction (no subsection)
- P1: PROVIDE_CONTEXT - Housing precarity context [3 citations]
- P2: SYNTHESIZE - Literature on housing instability [5 citations]

### [Framework Name] (subsection)
- P3: DESCRIBE_THEORY - Introduce framework [3-4 citations]
- P4: DESCRIBE_THEORY - Elaborate dimensions [3 citations]
- P5: SYNTHESIZE - Prior applications [6 citations]

### Extending to [Domain]
- P6: IDENTIFY_GAP - Framework not applied to X [2 citations] ← TURN
- P7: BRIDGE - Why framework fits this domain [3 citations]

### This Study
- P8: STATE_QUESTIONS - Research questions [1-2 citations]
- P9: PREVIEW - Argument preview [0 citations]

6. Estimate Calibration Targets

Based on cluster, set targets:

## Calibration Targets

- Word count: [range based on cluster]
- Paragraph count: [range]
- Citation count: [range]
- Subsection count: [number]
- Literature balance: [theory-heavy / balanced / substantive]

7. Write Architecture Memo

Create architecture-memo.md:

# Theory Section Architecture

## Cluster: [Name]
## Arc: [Type]

## Structure Overview

[Describe the overall flow in prose]

## Subsection Plan

### [Subsection 1 Title]
- Function: [what this section accomplishes]
- Paragraphs: [number]
- Key literatures: [list]

### [Subsection 2 Title]
...

## Turn Placement

The turn will occur in [paragraph N], after [what precedes] and before [what follows].

Turn strategy: [describe the pivot approach]

## Literature Allocation

| Literature | Paragraphs | Function |
|------------|------------|----------|
| [Lit 1] | [N] | [role] |
| [Lit 2] | [N] | [role] |
...

## Calibration Targets

| Metric | Target |
|--------|--------|
| Words | [range] |
| Paragraphs | [range] |
| Citations | [range] |
| Subsections | [number] |

## Outline

[Full paragraph-level outline with functions]

## Notes for Drafting

[Any special considerations, challenges, or decisions for Phase 3]

---

Guiding Principles

Structure Signals Ambition

• No subsections = streamlined empirical contribution
• 1-2 subsections = organized review
• 3-4 subsections = multiple theoretical components
• 5+ subsections = building theoretical apparatus

Don't add subsections for organizational convenience if they don't reflect genuine theoretical divisions.

Each Subsection Should Have a Clear Purpose

If you can't articulate what a subsection accomplishes, you don't need it.

The Turn is the Structural Center

Everything before builds toward it; everything after flows from it. Design around the turn.

Leave Room for Discovery

This is a blueprint, not a prison. Drafting may reveal the need for adjustments.

---

Output Files to Create

1. architecture-memo.md - Full architecture plan with outline 2. section-outline.md - Clean outline for quick reference during drafting

---

When You're Done

Report to the orchestrator:

• Arc structure selected
• Subsection plan (number and titles)
• Turn placement
• Calibration targets
• Any concerns about literature coverage

Example summary:

"Architecture complete. Using Funnel arc with 3 subsections: 'Housing Precarity' (context), 'Bodies Out of Place' (framework), and 'This Study' (contribution). Turn placed at end of framework section (paragraph 6). Target: 1,500 words, 10 paragraphs, 35 citations. Ready for Phase 2: Planning."

Phase 2: Planning

Why This Phase Matters

Before drafting prose, you need a paragraph-by-paragraph plan that specifies what each paragraph will accomplish, how it will open, and what citations it will deploy. This planning prevents the common problem of aimless literature cataloging and ensures each paragraph drives the argument forward.

---

Your Tasks

1. Assign Paragraph Functions

For each paragraph in the outline, assign a function from techniques/paragraph-functions.md:

2. Draft Topic Sentences

For each paragraph, write a topic sentence that:

• Signals the paragraph's function
• Uses appropriate opening type (from techniques/sentence-toolbox.md)
• Sets up the paragraph's work

Example topic sentences by function:

3. Plan Citation Deployment

For each paragraph, specify:

• How many citations (target 2.4-5.0 per paragraph)
• Citation pattern (parenthetical, author-subject, string, quote)
• Key sources to include

Use this template:

## Paragraph 3: SYNTHESIZE

**Topic sentence**: "Research on legal cynicism demonstrates that distrust is patterned by neighborhood context and prior police contact."

**Citations (5)**:
- Kirk and Papachristos 2011 (author-subject for definition)
- Sampson and Bartusch 1998 (parenthetical)
- Desmond et al. 2016 (parenthetical)
- Gau and Brunson 2010 (parenthetical)
- Tyler 2006 (parenthetical in string)

**Pattern**: Mix of author-subject for foundational work + parenthetical string for synthesis

4. Sequence for Flow

Review the paragraph sequence to ensure:

Logical progression:

• Context → Theory → Synthesis → Gap → Questions
• Each paragraph builds on what precedes

Smooth transitions:

• What transition marker will open each paragraph?
• How does each paragraph connect to the previous?

Turn positioning:

• Is the turn in the right place?
• Does adequate synthesis precede it?
• Do questions follow naturally?

5. Check Cluster Alignment

Verify the paragraph sequence matches cluster expectations:

6. Write Paragraph Map

Create paragraph-map.md:

# Paragraph Map

## Overview
- Total paragraphs: [N]
- Total target citations: [N]
- Turn location: Paragraph [N]

---

## Paragraph 1
**Function**: PROVIDE_CONTEXT
**Subsection**: [if applicable]
**Topic sentence**: "[Draft opening]"
**Content notes**: [What this paragraph covers]
**Citations (N)**:
- [Source 1] (pattern)
- [Source 2] (pattern)
**Transition to next**: [How it connects to P2]

---

## Paragraph 2
**Function**: [...]
...

[Continue for all paragraphs]

---

## Paragraph [N]: THE TURN
**Function**: IDENTIFY_GAP
**Topic sentence**: "[Draft turn sentence]"
**Turn structure**:
1. Acknowledge: [what we know]
2. Pivot: [contrastive marker]
3. Gap: [specific gap claim]
4. Connect: [link to your study]
**Citations (N)**:
- [Sources supporting the gap claim]

---

## Summary Table

| P# | Function | Subsection | Citations | Key Sources |
|----|----------|------------|-----------|-------------|
| 1 | CONTEXT | — | 3 | [list] |
| 2 | SYNTHESIZE | — | 5 | [list] |
...

7. Pre-Draft Checklist

Before moving to drafting, verify:

• [ ] Every paragraph has an assigned function
• [ ] Topic sentences drafted for all paragraphs
• [ ] Citations planned (2.4-5.0 per paragraph)
• [ ] Total citations in target range for cluster
• [ ] Turn sentence drafted
• [ ] Sequence flows logically
• [ ] Transitions planned
• [ ] Cluster sequence alignment checked

---

Guiding Principles

Topic Sentences Do the Heavy Lifting

A reader should be able to understand the section by reading only the topic sentences. If they can't, your structure isn't clear.

Citation Deployment is Strategic

Don't just sprinkle citations randomly. Plan which sources do which work.

The Paragraph Map is Your Drafting Guide

In Phase 3, you'll draft one paragraph at a time following this map. The better the map, the smoother the drafting.

Leave Room for Discovery

You may discover during drafting that a paragraph needs to split, or two should merge. That's fine—the map is a guide, not a prison.

---

Output Files to Create

1. paragraph-map.md - Full paragraph-by-paragraph plan 2. citation-plan.md - Citation inventory with deployment strategy

---

When You're Done

Report to the orchestrator:

• Number of planned paragraphs
• Turn placement (paragraph number)
• Total planned citations
• Any concerns about coverage or sequence

Example summary:

"Paragraph map complete. 10 paragraphs planned: 2 context, 3 framework, 2 synthesis, 1 gap (turn at P6), 2 contribution. 38 citations distributed across paragraphs (avg 3.8/paragraph). Turn drafts specific gap about guest mother identity in doubled-up households. Ready for Phase 3: Drafting."

Phase 3: Drafting

Why This Phase Matters

This is where planning becomes prose. The quality of your Theory section depends on sentence-level craft—how you open paragraphs, integrate citations, calibrate hedging, and build toward the turn. This phase transforms the paragraph map into publication-ready prose.

---

Your Tasks

1. Draft by Paragraph

Work through the paragraph map one paragraph at a time. For each:

Step 1: Open with purpose

• Use the planned topic sentence (or refine it)
• Match opening type to function (see techniques/sentence-toolbox.md)

Step 2: Build the paragraph

• 4-7 sentences typical (129-165 words target)
• Each sentence advances the paragraph's function
• Vary sentence length and structure

Step 3: Integrate citations

• Follow the planned citation deployment
• Use appropriate patterns (parenthetical, author-subject, string)
• See techniques/citation-patterns.md

Step 4: Calibrate hedging

• Hedge predictions and mechanism claims
• Assert definitions and gap statements
• See hedging guide in techniques/sentence-toolbox.md

Step 5: Transition out

• Final sentence should set up the next paragraph
• Use transitional phrases or logical connectors

2. Draft by Function

Apply function-specific guidance:

PROVIDE_CONTEXT Paragraphs

• Open with phenomena, not scholars
• Use statistics, trends, or historical developments
• Keep citations minimal (2-3)
• Establish stakes or scope

Example:

As housing costs have outpaced wage growth, many parents face difficulty housing their families. Over 3 million households with children doubled up with friends or family in 2019, and doubling-up rates have risen since the Great Recession (Pilkauskas et al. 2020). Low-income and Black families are disproportionately likely to double up, often as a strategy to avoid homelessness or manage poverty (Desmond 2016; Harvey 2018). These arrangements affect household composition, family relationships, and parenting dynamics in ways that remain poorly understood.

DESCRIBE_THEORY Paragraphs

• Name the framework/theorist explicitly
• Explain how the concept works
• Include dimensions, mechanisms, or processes
• Use author-subject citations for canonical sources

Example:

The concept of "recognition" draws on several conceptual traditions. According to Honneth (2003), recognition operates through three spheres: love (intimate relationships), rights (legal status), and esteem (social value). When individuals are denied recognition—through disrespect, exclusion, or degradation—they experience psychological harm and may struggle to develop positive self-relations (Taylor 1994). Lamont and colleagues (2016) have extended this framework to examine how low-status groups pursue recognition in everyday life, showing how stigmatized individuals actively resist misrecognition through boundary work and collective claims-making.

SYNTHESIZE Paragraphs

• Aggregate findings, don't summarize individual studies
• Use parenthetical strings efficiently
• Build toward a cumulative claim
• End with "what we know" before the gap

Example:

Research on legal cynicism demonstrates that distrust of legal institutions is patterned by neighborhood context, race, and prior contact with police (Kirk and Papachristos 2011; Sampson and Bartusch 1998). Studies consistently find that residents of high-crime neighborhoods express cynicism about police responsiveness and fairness, even as they continue to call 911 for emergencies (Desmond et al. 2016; Gau and Brunson 2010). This literature highlights the coexistence of cynicism and reliance—a paradox that existing frameworks have struggled to explain.

IDENTIFY_GAP Paragraphs

• Use the 4-part turn structure (see techniques/turn-formula.md)
• Be specific about what's unknown
• Connect directly to your study

Example:

This literature highlights the importance of recognition for understanding how marginalized groups engage with institutions. Yet while research documents the pursuit of recognition in various settings, there is less attention to how recognition-seeking shapes encounters with legal authorities specifically. How do residents of high-crime neighborhoods understand their calls to police as demands for recognition? What meaning-making processes allow them to reconcile cynicism with reliance? These questions remain largely unexplored.

STATE_QUESTIONS Paragraphs

• Articulate research questions clearly
• Connect questions to the gap
• Preview methods briefly if appropriate

Example:

This study examines how mothers in doubled-up households negotiate identity and dignity in shared living arrangements. Three questions guide the analysis: How do guest mothers understand their position within host households? What strategies do they use to maintain maternal identity when authority is constrained? How do these negotiations vary by the character of the host-guest relationship? I address these questions through in-depth interviews with 47 mothers who doubled up with family or friends in the Chicago metropolitan area.

3. Draft Subsection by Subsection

If using subsections:

• Draft complete subsection before moving to next
• Ensure internal coherence within subsection
• Check that subsection accomplishes its architectural purpose

After each subsection, pause for user review if appropriate.

4. Maintain Cluster Style

As you draft, ensure prose style matches cluster:

5. Track Progress

As you draft, note:

• Actual word count per paragraph
• Actual citations per paragraph
• Any deviations from the plan
• Paragraphs that need revision

6. Track Citations Used

Maintain a running list of every citation as you use it. For each citation, record:

{
  "citations": [
    {
      "author": "Kirk and Papachristos",
      "year": "2011",
      "context": "legal cynicism patterning",
      "paragraph": 3,
      "pattern": "parenthetical"
    },
    {
      "author": "Connell",
      "year": "2005",
      "context": "hegemonic masculinity concept",
      "paragraph": 5,
      "pattern": "author-subject"
    }
  ]
}

Why track citations?

• Enables bibliography generation without parsing the document
• Allows Zotero ID lookup in Phase 5
• Helps verify citation density during revision
• Creates audit trail of sources used

Save incrementally to citations-tracking.json as you draft.

---

Guiding Principles

Draft, Don't Perfect

Get prose on the page. Revision comes in Phases 4-5.

Follow the Map

The paragraph map exists for a reason. Stick to it unless you discover a structural problem.

Topic Sentences Are Anchors

If you're struggling with a paragraph, return to the topic sentence. What is this paragraph trying to accomplish?

Vary Citation Integration

Don't use the same pattern throughout. Mix parenthetical, author-subject, and (sparingly) quote-then-cite.

Read Aloud

Prose that sounds awkward when read aloud needs revision. Academic writing should still flow.

The Turn Is Sacred

Draft the turn with special care. It's the most important paragraph.

---

Common Drafting Problems

---

Output Files to Create

1. theory-section-draft.md - Full draft prose organized by subsection 2. drafting-notes.md - Notes on deviations, concerns, needs for revision 3. citations-tracking.json - Running list of all citations used (author, year, context, paragraph, pattern)

---

When You're Done

Report to the orchestrator:

• Draft complete (yes/no)
• Word count achieved
• Citation count achieved
• Turn drafted (provide sentence)
• Any structural changes from plan
• Areas flagged for revision

Example summary:

"Draft complete. 1,487 words across 10 paragraphs (target: 1,500). 36 citations (target: 35). Turn at paragraph 6: 'Yet while research documents the pursuit of recognition in various settings, there is less attention to how recognition-seeking shapes encounters with legal authorities specifically.' One deviation: Split planned paragraph 4 into two for clarity. Flagged paragraphs 7-8 for transition smoothing in revision. Ready for Phase 4: Turn refinement."

Phase 4: Turn Refinement

Why This Phase Matters

The turn is the rhetorical center of your Theory section—the moment you pivot from "what we know" to "what we don't know" and create space for your contribution. A weak turn undermines the entire section; a strong turn makes your contribution memorable. This phase ensures the turn is specific, compelling, and tightly connected to your research questions.

---

Your Tasks

1. Extract Current Turn

From the draft, identify:

• The turn paragraph(s)
• The specific turn sentence(s)
• The sentence before the turn (what it acknowledges)
• The sentence after the turn (how it connects to your study)

2. Evaluate Against the 4-Part Formula

Check that the turn includes all four elements:

If any part is weak or missing, draft improvements.

3. Test Gap Specificity

The gap must be specific, not generic. Evaluate:

Weak (Generic) Gap Tests:

• "More research is needed" → Too vague
• "We don't fully understand" → What specifically?
• "This area is understudied" → Which aspect?
• "Scholars have not examined" → What exactly?

Strong (Specific) Gap Tests:

• Specifies WHAT is unknown
• Specifies AMONG WHOM (population)
• Specifies UNDER WHAT CONDITIONS (context)
• Explains WHY THIS MATTERS

Specificity Template:

"We know little about [what specifically] among [which population] when [under what conditions] despite [why this matters]."

4. Check Gap-Question Alignment

The gap must directly motivate your research questions.

Alignment Test: Can you draw a straight line from gap to question?

If misaligned, revise either the gap or the questions.

5. Calibrate Turn Confidence

The gap claim should be assertive (you're stating what the literature has NOT done), but not overclaimed.

Appropriate Confidence:

"Yet we know little about..."

"However, less attention has been paid to..."

"Scholars have not examined..."

Over-Claimed (Avoid):

"No one has ever studied..."

"The literature completely ignores..."

"We know nothing about..."

Under-Hedged Definition (Also Avoid):

"It may be that we don't fully understand..."

"Perhaps less attention has been paid to..."

6. Evaluate Turn Placement

Is the turn in the right location?

Too Early: Gap feels unmotivated; insufficient synthesis precedes Too Late: Reader lost the thread; synthesis felt aimless Just Right: Gap emerges naturally from what precedes

7. Refine the Turn Sentence

Craft the turn sentence with precision:

Turn Sentence Structure:

[Acknowledge prior work]. [Pivot marker], [specific gap claim]. [Connect to your study].

Example Refinement:

Original (weak):

"This literature is important, but there is still more to learn. This study examines mothers in doubled-up households."

Revised (strong):

"This literature highlights the importance of housing context for family identity work. Yet while research documents how housing instability affects parenting, there is less attention to how mothers in doubled-up arrangements manage maternal identity when they lack spatial authority. This study examines how guest mothers negotiate identity and dignity in shared living arrangements."

8. Test the Turn

Run these diagnostic questions:

1. Specificity: Could this gap apply to many other studies, or is it specific to mine? 2. Motivation: Does the synthesis before genuinely build toward this gap? 3. Connection: Do my research questions directly address this gap? 4. Placement: Does the turn fall at the right structural moment? 5. Confidence: Is the gap claim appropriately assertive? 6. Novelty: Is this gap genuinely unstudied, or am I overclaiming?

9. Write Turn Refinement Memo

Create turn-memo.md:

# Turn Refinement

## Original Turn
[Paste original turn paragraph]

## Analysis

### 4-Part Structure
- Acknowledge: [assessment]
- Pivot: [assessment]
- Gap: [assessment]
- Connect: [assessment]

### Specificity Assessment
[Is the gap specific enough? What would make it more specific?]

### Gap-Question Alignment
[Do questions directly follow from gap?]

### Placement Assessment
[Is turn in right location? Should it move?]

## Refined Turn

[Full refined paragraph]

### Turn Sentence (Isolated)
> "[The core turn sentence]"

## Changes Made
- [List specific changes]

## Context Adjustments Needed
[Any changes to paragraphs before/after the turn]

---

Guiding Principles

The Turn Is Your Thesis

If someone asks "What's your contribution?", you should be able to point to your turn sentence.

Specificity > Scope

A narrow, specific gap is better than a broad, vague one. "We know little about X among Y in Z context" beats "This area is understudied."

The Turn Isn't a Surprise

The synthesis before should build toward the gap. Readers should sense what's coming.

The Turn Isn't Buried

Start the turn paragraph with the pivot. Don't bury the gap in the middle of a paragraph.

One Gap, Not Many

Focus on the single gap your study addresses. Multiple gaps dilute the contribution.

---

Common Turn Problems

---

Output Files to Create

1. turn-memo.md - Analysis and refined turn 2. theory-section-draft-v2.md - Updated draft with refined turn

---

When You're Done

Report to the orchestrator:

• Original turn assessment (strong/weak/specific issues)
• Key refinements made
• Final turn sentence
• Gap-question alignment verified
• Placement confirmed

Example summary:

"Turn refined. Original was generic ('housing dynamics remain understudied'). Refined to: 'Yet while research documents how housing instability affects parenting practices, there is less attention to how mothers in doubled-up arrangements manage maternal identity when they lack spatial authority over their children.' Gap now specifies: (1) maternal identity (what), (2) guest mothers (who), (3) shared households without spatial authority (conditions). Directly motivates RQ1 about identity negotiation. Placement confirmed at P6 of 10. Ready for Phase 5: Revision."

Phase 5: Revision

Why This Phase Matters

Good writing is rewriting. The draft from Phase 3 and the refined turn from Phase 4 need polishing: checking calibration against norms, smoothing transitions, ensuring consistency, and achieving prose quality worthy of publication. This phase transforms a competent draft into a compelling Theory section.

---

Your Tasks

1. Calibration Check

Compare draft against benchmarks from techniques/calibration-norms.md:

If outside target range:

• Under word count: Expand synthesis or add context
• Over word count: Cut redundancy, tighten prose
• Under-cited: Add supporting sources
• Over-cited: Combine into strings, cut peripheral sources

2. Paragraph Function Audit

For each paragraph, verify:

1. Clear function: Can you identify what this paragraph is doing (CONTEXT, THEORY, SYNTHESIZE, etc.)? 2. Topic sentence signals function: Does the opening make the function clear? 3. Function fits position: Is this function appropriate for this location in the sequence? 4. Function matches cluster: Is this function used appropriately for your cluster?

Flag any paragraph where function is unclear or misaligned.

3. Transition Check

Between every paragraph pair, verify:

• Logical connection: Does paragraph N+1 follow from paragraph N?
• Explicit transition: Is there a transition marker or connecting phrase?
• No abrupt shifts: Reader shouldn't feel whiplash

Common transitions to add:

• Additive: "Building on this work," "Moreover," "Similarly"
• Contrastive: "However," "Yet," "In contrast"
• Sequential: "A second body of scholarship," "Additionally"
• Consequential: "Thus," "As a result," "Given this"

4. Citation Pattern Audit

Review citation integration across the section:

• Variety: Are you using multiple patterns (parenthetical, author-subject, string)?
• Author-subject for canonicals: Are foundational theorists given prominence?
• Efficient strings: Is consensus established without excessive strings?
• Quotes only for definitions: Are you quoting only when language precision matters?
• No orphan citations: Every citation should be integrated into prose, not just listed

5. Hedging Calibration

Review hedging across claim types:

Flag any over-hedged definitions or under-hedged mechanism claims.

6. Cluster Style Consistency

Verify prose style matches cluster throughout:

7. Prose Polish

Read the entire section for prose quality:

• Sentence variety: Mix long and short sentences
• Active voice: Prefer active constructions where appropriate
• Concrete language: Avoid vague abstractions
• Jargon check: Is technical terminology necessary and defined?
• Read aloud: Does it flow when spoken?

8. Compile Citation List & Zotero Lookup

Transform the citation tracking from Phase 3 into final outputs.

Step 1: Deduplicate citations Combine citations-tracking.json into a unique list of sources:

{
  "citations": [
    {
      "author": "Kirk and Papachristos",
      "year": "2011",
      "count": 3,
      "zotero_key": null,
      "full_citation": null
    }
  ]
}

Step 2: Zotero lookup (if MCP available) If the Zotero MCP is configured, look up each citation to get:

• Zotero item key
• Full bibliographic metadata
• Formatted citation

Use mcp__zotero-mcp__zotero_search_items with author + year queries:

For each citation:
1. Search: "{author} {year}"
2. If match found: record zotero_key and full metadata
3. If no match: flag for manual resolution

Step 3: Generate outputs

Create citations-final.json:

{
  "section": "Theory",
  "total_citations": 37,
  "unique_sources": 35,
  "citations": [
    {
      "author": "Kirk and Papachristos",
      "year": "2011",
      "title": "Legal Cynicism and Collective Efficacy",
      "zotero_key": "ABC12345",
      "count_in_section": 3
    }
  ],
  "unmatched": [
    {
      "author": "Smith",
      "year": "2020",
      "note": "Not found in Zotero library"
    }
  ]
}

Create bibliography.md:

# Bibliography for Theory Section

## Sources Cited (35 unique)

Kirk, David S. and Andrew V. Papachristos. 2011. "Cultural Mechanisms and the Persistence of Neighborhood Violence." *American Journal of Sociology* 116(4):1190-1233.

[Continue for all sources...]

## Unmatched Citations (require manual lookup)

- Smith (2020) - Not found in Zotero library

If Zotero MCP is not available:

• Output the deduplicated citation list without Zotero keys
• Note that manual bibliography creation is needed
• The list still saves significant time vs. parsing the document

9. Final Checklist

Before completion, verify:

• [ ] Word count in target range
• [ ] Citation count in target range
• [ ] Each paragraph has clear function
• [ ] Topic sentences signal function
• [ ] Transitions between all paragraphs
• [ ] Turn is specific and well-placed
• [ ] Gap directly motivates research questions
• [ ] Hedging calibrated by claim type
• [ ] Citation patterns varied
• [ ] Cluster style consistent throughout
• [ ] Prose reads smoothly
• [ ] Citation list compiled (citations-final.json)
• [ ] Bibliography generated (bibliography.md)
• [ ] Unmatched citations flagged for user

10. Write Quality Memo

Create quality-memo.md:

# Theory Section Quality Assessment

## Calibration Results

| Metric | Target | Actual | Status |
|--------|--------|--------|--------|
| Word count | 1,145-1,744 | [N] | ✓/needs adjustment |
| Paragraphs | 7-12 | [N] | ✓/needs adjustment |
| Citations | 26-43 | [N] | ✓/needs adjustment |
| Citations/paragraph | 2.4-5.0 | [N] | ✓/needs adjustment |
| Subsections | 1-3 | [N] | ✓/needs adjustment |

## Structural Assessment

### Paragraph Functions
[Assessment of function clarity and sequencing]

### Transitions
[Assessment of flow between paragraphs]

### Turn Quality
[Assessment of turn specificity and placement]

## Style Assessment

### Citation Patterns
[Assessment of citation variety and integration]

### Hedging Calibration
[Assessment of hedging by claim type]

### Cluster Alignment
[Assessment of style consistency with cluster]

## Prose Quality

### Strengths
- [List strong aspects]

### Areas Addressed in Revision
- [List what was improved]

### Remaining Considerations
- [Any remaining issues or judgment calls]

## Final Section Statistics

- Word count: [N]
- Paragraph count: [N]
- Citation count: [N]
- Subsection count: [N]
- Cluster: [Name]

## Confidence Assessment
[High/Medium/Low confidence that section is publication-ready]

---

Guiding Principles

Revision Is Not Line-Editing

This phase is about structural integrity, calibration, and coherence—not comma placement.

When in Doubt, Cut

If a sentence doesn't advance the argument, remove it.

Consistency Trumps Perfection

A consistently good section beats one with brilliant moments and weak transitions.

The Section Should Stand Alone

A reader should understand your contribution from the Theory section alone, without reading the rest of the paper.

---

Common Revision Issues

---

Output Files to Create

1. theory-section-final.md - Final polished Theory section 2. quality-memo.md - Full quality assessment 3. revision-log.md - Record of changes made 4. citations-final.json - Complete citation list with Zotero keys (if available) 5. bibliography.md - Formatted bibliography ready for reference section

---

When You're Done

Report to the orchestrator:

• Final word count
• Final citation count
• Calibration status (in range / deviations explained)
• Confidence assessment (High/Medium/Low)
• Citation outputs generated (citations-final.json, bibliography.md)
• Zotero matches found / unmatched citations
• Any remaining concerns

Example summary:

"Revision complete. Final section: 1,523 words (target 1,145-1,744 ✓), 37 citations (target 26-43 ✓), 10 paragraphs, 2 subsections. All calibration metrics in range. Turn confirmed specific and well-placed at P6. Transitions smoothed throughout. Hedging calibrated. High confidence section is publication-ready.

>

Citation outputs: citations-final.json with 35 unique sources. Zotero lookup: 32 matched, 3 unmatched (flagged in bibliography.md). bibliography.md ready for reference section.

>

Minor consideration: paragraph 8 bridge could be strengthened if user wants to emphasize theoretical contribution more."

Calibration Norms

Statistical benchmarks from analysis of 80 Theory sections in interview-based articles from Social Problems and Social Forces. Use these norms to calibrate your draft against field expectations.

---

Overall Benchmarks

---

Target Ranges

Use the interquartile range (Q25-Q75) as your target. Deviations should be intentional.

Length

• Target word count: 1,145-1,744 words
• Target paragraphs: 7-12 paragraphs
• Under 1,000 words: May signal insufficient theoretical development
• Over 2,000 words: May signal over-elaboration (unless Concept-Builder)

Citation Density

• Target citations per paragraph: 2.4-5.0
• Target citations per 1,000 words: 19-32
• Under 2 per paragraph: May signal insufficient engagement
• Over 6 per paragraph: May signal listing rather than synthesizing

Structure

• Target subsections: 1-3
• 0 subsections: Acceptable for streamlined Gap-Fillers
• 4+ subsections: Appropriate for Concept-Builders, may signal over-structuring otherwise

Paragraph Length

• Target words per paragraph: 129-165
• Under 100 words: May be underdeveloped
• Over 200 words: Consider splitting

---

Cluster-Specific Norms

Cluster 1: Gap-Filler Minimalist (27.5%)

Key: Leaner than average. Efficient synthesis, sharp turn.

---

Cluster 2: Theory-Extension (22.5%)

Key: Moderate length. Higher citation density in framework exposition.

---

Cluster 3: Concept-Builder (15.0%)

Key: Longest of all clusters. Space needed for conceptual apparatus.

---

Cluster 4: Synthesis Integrator (18.8%)

Key: Middle ground. Citations drawn from multiple traditions.

---

Cluster 5: Problem-Driven (16.3%)

Key: Variable structure. More empirical/policy citations.

---

Literature Balance Definitions

Balance by Cluster

---

Subsection Interpretation

Rule: Match structure to ambition. Gap-Fillers with 5 subsections confuse readers.

---

Visual Elements

Only 8.8% of articles include tables or figures in the Theory section.

Guideline: Only include a visual if you're introducing a new framework worth visualizing.

---

Outlier Analysis

Unusually Long (2,500+ words)

• Typically Concept-Builders
• Often have 5+ subsections
• Usually 60+ citations
• Justified when building substantial theoretical apparatus

Unusually Short (Under 800 words)

• Rare in these journals
• May work for tightly focused empirical documenters
• Risk signaling insufficient theoretical engagement

Citation Outliers (70+ citations)

• Strong correlation with Concept-Builder cluster
• Often involve synthesizing multiple literatures
• Requires space to justify

---

Calibration Checklist

Before finalizing, verify your draft against these norms:

If outside target range: Intentional deviation is fine, but you should be able to justify why your section needs to be longer/shorter/more or less citation-dense than typical.

---

Interpreting Deviations

Citation Integration Patterns

How you integrate citations affects readability and signals your relationship to the literature. This guide covers the four primary patterns and when to use each.

---

The Four Patterns

---

Pattern 1: Parenthetical Support (50-55%)

The most common pattern. The claim comes first; citations provide evidence without interrupting prose.

Structure

[Claim] (Author Year).

[Claim] (Author Year; Author Year).

Examples

"Research on legal cynicism demonstrates that distrust of legal institutions is patterned by neighborhood context and race (Kirk and Papachristos 2011)."

"Mass incarceration has left thousands of Black and Brown men under 'mass supervision' through probation and parole (McNeil and Beyens 2013, 3)."

"Conspiracy beliefs have long proliferated in the United States (Bailyn 2017; Waters 1997) and beyond (Imhoff et al. 2022; Swami 2012)."

When to Use

• Synthesizing multiple studies
• Efficient documentation of established findings
• Building cumulative case without interruption
• Most claims in most paragraphs

Strengths

• Keeps focus on the content, not the author
• Allows efficient synthesis of multiple sources
• Maintains prose flow

---

Pattern 2: Author-as-Subject (20-25%)

The cited author performs the intellectual action. Gives prominence to the scholar.

Structure

[Author] (Year) [argues/demonstrates/shows/conceptualizes] that...

According to [Author]...

[Author's] (Year) [concept/framework]...

Examples

"Connell (2005) conceptualizes hegemonic masculinity as the most honorable expression of manliness within a hierarchy of masculinities."

"According to Puwar (2004), numerical representation matters, but so does the construction of race and gender within larger social status hierarchies."

"Du Bois so clearly explicated that individual agency, the imagined range of actions available to the self, was shaped through the ongoing reconciliation of consciousness."

When to Use

• Introducing canonical theorists (Bourdieu, Goffman, Foucault)
• Foundational works you're building on or extending
• When the who matters as much as the what
• Theory-Extension clusters (prominent for named framework)

Strengths

• Signals which scholars are central to your argument
• Shows theoretical lineage
• Appropriate respect for foundational contributions

Caution

• Overuse makes prose feel like a literature parade
• Reserve for genuinely important sources

---

Pattern 3: Citation Strings (15-20%)

Multiple citations bundled together, signaling consensus or breadth.

Structure

[Claim] (Author Year; Author Year; Author Year).

[Claim] (see Author Year; Author Year; cf. Author Year).

[Claim] (e.g., Author Year; Author Year).

Examples

"Scholarship on business-elite influence on populist movements links opposition to climate reform to climate denial and other far-right ideals (Dunlap and McCright 2015; Brulle 2018; Hertel-Fernandez 2019)."

"...the sociology of law, which demonstrates that legality is embedded in a complex set of cultural schemas (Ewick and Silbey 1998; Silbey 2005; Nielsen 2000)."

String Modifiers

• "see": Points to useful sources: "(see Smith 2019; Jones 2020)"
• "e.g.": Indicates examples, not exhaustive: "(e.g., Brown 2018)"
• "cf.": Indicates comparison or contrast: "(cf. alternative view in Lee 2017)"

When to Use

• Establishing that a finding is robust across studies
• Showing breadth of a literature
• Documenting consensus
• Gap-Filler clusters (efficient synthesis)

Strengths

• Demonstrates thorough literature engagement
• Establishes consensus quickly
• Shows pattern holds across multiple contexts

Caution

• Long strings can feel like padding
• More than 5-6 citations starts to look excessive
• Make sure the claim actually applies to all cited sources

---

Pattern 4: Quote-then-Cite (5-10%)

Direct quotation with attribution. Use sparingly and strategically.

Structure

As [Author] notes, "..." (Year:page).

[Author] describes [concept] as "..." (Year:page).

"[Quote]" ([Author] Year:page).

Examples

"Legal cynicism is broadly conceived of as 'a cultural orientation in which the law and the agents of its enforcement are viewed as illegitimate, unresponsive, and ill-equipped to ensure public safety' (Kirk and Matsuda 2011:443)."

"According to Lamont (2009:158): 'Low-status groups (in this situation) are more likely to be resigned and passive instead of resilient.'"

When to Use

• Key definitions: When the original language matters
• Contested terms: When precise wording is at stake
• Foundational formulations: Classic statements worth preserving
• Concept-Builder clusters: Detailed engagement with sources being critiqued or built upon

When NOT to Use

• General findings (paraphrase instead)
• Routine claims (use parenthetical)
• To avoid writing your own synthesis (lazy quoting)

Strengths

• Precision when language matters
• Shows close reading of sources
• Appropriate for definitions and key formulations

Caution

• Overuse makes the section feel like a collage
• Should be rare (5-10% of citations max)
• Always include page numbers for quotes

---

Cluster-Specific Patterns

---

Citation Density Benchmarks

Based on analysis of 80 articles:

What These Mean

• 2.4-5.0 per paragraph: Most paragraphs should have 2-5 citations
• 18.9-32.0 per 1,000 words: A 1,500-word section should have ~30-48 citations
• Under 26 citations: May signal insufficient engagement
• Over 50 citations: May signal catalog-style rather than argument-style

---

Common Mistakes

---

Integration Checklist

For each paragraph, verify:

1. Pattern variety: Are you using multiple patterns across the section? 2. Author-subject for canonicals: Are foundational theorists given appropriate prominence? 3. Efficient strings: Is consensus established without excessive strings? 4. Quotes only for definitions: Are you quoting only when language precision matters? 5. Citation density: Does the paragraph fall in the 2.4-5.0 range? 6. Cluster alignment: Does your citation pattern match your cluster?

Paragraph Functions

Every paragraph in a Theory section serves a recognizable function. Readers should be able to identify what each paragraph is doing even without subheadings. This guide catalogs the 9 primary paragraph functions.

---

The 9 Functions

---

Detailed Function Profiles

PROVIDE_CONTEXT

Purpose: Ground the reader in the empirical world—the phenomenon, population, setting, or historical moment under study.

Typical markers:

• Opens with phenomena, not scholars
• Statistics, trends, or historical developments
• "As [condition] has [changed]..."
• "The [phenomenon] affects..."
• "In [year/decade], [development]..."

Example paragraph:

As housing costs have outpaced wage growth, many parents face difficulty housing their families. Over 3 million households with children doubled up with friends or family in 2019, and doubling-up rates have risen since the Great Recession. Low-income and Black families are disproportionately likely to double up, often as a strategy to avoid homelessness or manage poverty.

When to use: Opening the section; grounding each subsection in empirical reality.

---

DESCRIBE_THEORY

Purpose: Explicate a theoretical framework, concept, or key term. Make the analytical tool available to the reader.

Typical markers:

• Names a theorist, framework, or concept explicitly
• Uses "According to," "[Scholar] argues," "The concept of [X]..."
• Explains how the concept works, not just what it is
• May include definition, dimensions, mechanisms

Example paragraph:

The concept of "recognition" draws on several conceptual traditions. According to Honneth (2003), recognition operates through three spheres: love (intimate relationships), rights (legal status), and esteem (social value). When individuals are denied recognition—through disrespect, exclusion, or degradation—they experience psychological harm and may struggle to develop positive self-relations. Recognition thus serves as both a psychological resource and a political demand.

When to use: After context; whenever introducing theoretical apparatus.

---

SYNTHESIZE

Purpose: Pull together findings from multiple studies to establish patterns, consensus, or accumulated knowledge.

Typical markers:

• Multiple citations in a single paragraph
• "Studies find..." "Research shows..." "Scholars have documented..."
• Aggregates findings rather than describing individual studies
• Builds toward a cumulative claim

Example paragraph:

Research on legal cynicism demonstrates that distrust of legal institutions is patterned by neighborhood context, race, and prior contact with police (Kirk and Papachristos 2011; Sampson and Bartusch 1998). Studies consistently find that residents of high-crime neighborhoods express cynicism about police responsiveness and fairness, even as they continue to call 911 for emergencies (Desmond et al. 2016). This literature highlights the coexistence of cynicism and reliance—a paradox that existing frameworks have struggled to explain.

When to use: Building the evidence base before identifying the gap; showing what "we know."

---

CONTRAST

Purpose: Present competing perspectives, alternative explanations, or disagreement in the literature.

Typical markers:

• "However," "In contrast," "Others argue..."
• Two (or more) positions clearly delineated
• Even-handed presentation (for Debate-Resolvers)
• May set up a tension the study will address

Example paragraph:

Whether secondary labor market entry helps or harms long-term immigrant incorporation remains debated. One line of research suggests that early employment—even in low-wage jobs—provides social capital, language acquisition, and pathway to better positions (Portes and Rumbaut 2001). Others, however, argue that secondary labor market entry traps immigrants in dead-end jobs, limiting mobility and reinforcing ethnic segmentation (Waldinger and Lichter 2003). This tension remains unresolved in the literature.

When to use: Problem-Driven (Debate) articles; showing disagreement before resolution.

---

CRITIQUE

Purpose: Identify limitations, blind spots, or inadequacies in existing approaches.

Typical markers:

• "However, this approach is limited..."
• "Yet scholars have not..."
• "This framework cannot account for..."
• "Existing literature is limited on two fronts..."

Example paragraph:

While this work is valuable for understanding how individuals navigate bureaucratic encounters, it is limited in two respects. First, most studies focus on formal legal settings (courts, police stations), overlooking everyday interactions with social services. Second, the literature assumes that individuals approach institutions with formed dispositions, neglecting how dispositions themselves may be shaped through repeated encounters. A more dynamic account is needed.

When to use: Concept-Builders before introducing new framework; setting up the need for innovation.

---

IDENTIFY_GAP

Purpose: Mark the transition from "what we know" to "what we don't know." This is the rhetorical center of the section.

Typical markers:

• "Yet we know little about..."
• "However, less attention has been paid to..."
• "What remains unclear is..."
• "Scholars have not examined..."
• "Largely unexplored are..."

Example paragraph:

Yet while research to date provides insight into why residents persist in calling on law enforcement despite cynicism, there is less attention to the cultural scripts through which people themselves seek to make sense of the institutions with which they engage. How do residents of high-crime neighborhoods understand their relationship to police? What meaning-making processes allow them to reconcile cynicism with reliance? These questions remain largely unexplored.

When to use: The "turn" moment. Usually middle of the section; sometimes later for empirical documenters.

---

BRIDGE

Purpose: Connect what has been established to what you're doing. Show how literatures come together or how frameworks apply.

Typical markers:

• "Bringing together..." "Drawing on..."
• "This framework illuminates..."
• "We argue that [X] provides analytical purchase on..."
• "These insights suggest that..."

Example paragraph:

Bringing these literatures into dialogue suggests a way forward. Recognition theory illuminates why procedural fairness matters—it signals respect and acknowledgment of status. Legal consciousness research reveals how residents actively interpret their encounters with institutions. Together, these perspectives allow us to see police-calling not as a paradox of cynicism-despite-reliance, but as an expression of demands for recognition that existing institutions consistently fail to provide.

When to use: Synthesis Integrators (the payoff paragraph); Theory-Extenders showing why framework fits.

---

STATE_QUESTIONS

Purpose: Articulate the research questions, hypotheses, or analytical focus of the study.

Typical markers:

• "This study examines..."
• "I ask: [question]?"
• "Three questions guide this analysis..."
• "We investigate whether..."

Example paragraph:

This study examines how mothers in doubled-up households negotiate identity and dignity in shared living arrangements. Three questions guide the analysis: (1) How do guest mothers understand their position within host households? (2) What strategies do they use to maintain maternal identity when authority is constrained? (3) How do these negotiations vary by the character of the host-guest relationship? I address these questions through interviews with 47 mothers who doubled up with family or friends.

When to use: After the gap; usually the penultimate or final paragraph.

---

PREVIEW

Purpose: Summarize the argument or outline what the findings will show. Provides a roadmap.

Typical markers:

• "I find that..." "We show that..."
• "The findings reveal..."
• "In what follows, I first... then... finally..."
• Brief statement of main claims (without evidence)

Example paragraph:

In what follows, I show that guest mothers deploy three strategies to maintain maternal identity: boundary work (carving out private space), deference management (navigating host authority), and temporal framing (emphasizing the temporary nature of arrangements). I argue that these strategies reveal how structural housing precarity intersects with gendered expectations of motherhood to produce distinctive identity work. I conclude by discussing implications for understanding family instability and housing policy.

When to use: Final paragraph of the section; optional but common for complex arguments.

---

Paragraph Sequences by Cluster

Gap-Filler Sequence

1. PROVIDE_CONTEXT 2. SYNTHESIZE 3. SYNTHESIZE 4. IDENTIFY_GAP 5. STATE_QUESTIONS

Theory-Extender Sequence

1. PROVIDE_CONTEXT 2. DESCRIBE_THEORY 3. SYNTHESIZE (prior applications) 4. IDENTIFY_GAP 5. BRIDGE 6. STATE_QUESTIONS

Concept-Builder Sequence

1. PROVIDE_CONTEXT 2. DESCRIBE_THEORY (tradition 1) 3. CRITIQUE 4. DESCRIBE_THEORY (tradition 2) 5. SYNTHESIZE + concept introduction 6. DESCRIBE_THEORY (new concept explication) 7. STATE_QUESTIONS

Synthesis Integrator Sequence

1. DESCRIBE_THEORY (tradition A) 2. SYNTHESIZE (A applications) 3. DESCRIBE_THEORY (tradition B) 4. SYNTHESIZE (B applications) 5. BRIDGE 6. STATE_QUESTIONS

Problem-Driven (Debate) Sequence

1. DESCRIBE_THEORY (position 1) 2. CONTRAST (position 2) 3. CRITIQUE (both limited) 4. STATE_QUESTIONS

---

Diagnostic Questions

For each paragraph, ask: 1. What function is this paragraph serving? 2. Is the function clear from the first sentence? 3. Does the function fit its position in the sequence? 4. Is the function appropriate for my cluster?

If you can't identify the function, the paragraph likely needs revision.

Sentence Toolbox

A vocabulary of sentence-level moves for Theory section writing, based on analysis of 80 articles from Social Problems and Social Forces.

---

1. Opening Sentence Types

Theory sections and paragraphs begin with distinct sentence types that signal rhetorical function. Select appropriately.

1.1 Literature-Announcing Openers (25-30% of paragraphs)

Signal entry into an established scholarly conversation.

Signature patterns:

• "A growing body of research..."
• "Over the past [X] decades, [scholars/sociologists] have..."
• "Research on [topic] has..."
• "An emerging literature examines..."
• "A burgeoning literature documents..."

Examples:

"A growing body of research examines the ways migrant children and their parents back home manage their relationships."

"Over the last three decades, cultural sociologists have developed a rich theoretical and empirical research platform to demonstrate how culture works in a variety of institutional contexts."

"In the past two decades, sociologists have demonstrated the persistence of systemic oppression along the lines of race, class, sex, gender, sexuality, and other social locations."

Best for: Opening synthesis paragraphs, beginning the theory section.

---

1.2 Context-Setting Openers (20-25%)

Establish empirical phenomena, historical developments, or social conditions.

Signature patterns:

• "As [phenomenon] has [developed/expanded/changed]..."
• "The [adjective] [phenomenon] has..."
• "In recent [decades/years]..."
• "With the [expansion/rise/decline] of..."

Examples:

"As housing costs have outpaced wage growth, many parents face difficulty housing their families."

"The dramatic reconfiguration of the U.S. state over the past four decades has generated a body of criminological scholarship centered on the unprecedented magnitude of carceral expansion."

"The US immigration system is situated within a global regime of borders."

Best for: Opening the theory section, introducing substantive context.

---

1.3 Theory-Introducing Openers (15-20%)

Explicitly name a theoretical framework, concept, or scholar.

Signature patterns:

• "The concept of '[X]' draws on..."
• "[Framework/Theory] posits that..."
• "According to [Scholar]..."
• "[Scholar's] (Year) concept of [X]..."
• "Central to this analysis is..."

Examples:

"The concept of 'recognition' draws on several conceptual traditions."

"The procedural justice model asserts that people make fairness evaluations according to four process-based criteria."

"Connell (2005) conceptualizes hegemonic masculinity as the most honorable expression of manliness within a hierarchy of masculinities."

Best for: Theory-Extension and Concept-Builder articles, framework exposition paragraphs.

---

1.4 Gap-Marking Openers (10-15%)

Immediately signal what is missing or understudied.

Signature patterns:

• "Yet we know little about..."
• "However, [scholars/research] has not..."
• "What remains unclear is..."
• "Largely unexplored in studies of [X] are..."
• "Despite [X], we have limited understanding of..."

Examples:

"Yet while research to date provides insight into why residents persist in calling on law enforcement, there is less attention to the cultural scripts through which people themselves seek to make sense of the institutions with which they engage."

"Largely unexplored in studies of transnational elder care are the experiences of adult children who remain in the place of origin with their aging parents."

Best for: The "turn" paragraph, after synthesis.

---

1.5 Contrast-Signaling Openers (10-12%)

Introduce tension between competing positions or shift analytical direction.

Signature patterns:

• "However, [alternative view]..."
• "In contrast to [X], [Y]..."
• "While [X], [Y]..."
• "Others, however, argue..."
• "At the same time..."

Examples:

"The study of feminist social movements provides a contrasting account of the mobilization of civil society into carceral activities."

"Others, however, argue that a tendency toward conspiratorial thinking is not indicated by psychological predispositions as much as indicative of a 'cognitive style.'"

Best for: Debate-framing, presenting alternative perspectives.

---

1.6 Definition-Providing Openers (5-8%)

Define key terms, often introducing newly coined concepts.

Signature patterns:

• "[Concept] refers to..."
• "I use the term [X] to..."
• "We define [X] as..."
• "[Concept] is..."
• "By [X], I mean..."

Examples:

"I use the term carceral creep to suggest the incremental and often imperceptible advance of carceral forces."

"Work primacy refers to the centrality of work in the economic and noneconomic lives of undocumented immigrant youth."

"We define recognition as the 'fact of being acknowledged' through the institutional provision of services, validation, and dignity."

Best for: Concept-Builder articles, new terminology introduction.

---

1.7 Claim-Making Openers (5-8%)

Make direct argumentative assertions about the study's contribution.

Signature patterns:

• "I argue that..."
• "This study [examines/considers/extends]..."
• "We contend that..."
• "This paper [integrates/demonstrates/shows]..."

Examples:

"I argue that Du Bois offers an integrated framework for understanding immigrants' incorporation strategies."

"This paper integrates procedural and distributive justice models in order to examine how both legal treatment and outcome experiences influenced respondents' perceptions of (in)justice."

Best for: Stating your contribution, previewing the argument.

---

2. Transition Marker Inventory

2.1 Additive Markers

Connect supporting or elaborating ideas.

2.2 Contrastive Markers

Signal disagreement, limitation, or alternative perspectives.

2.3 Temporal Markers

Situate developments in historical context.

2.4 Causal/Consequential Markers

Signal logical implications.

2.5 Exemplifying Markers

Introduce specific illustrations.

---

3. Hedging Calibration

When to Hedge (may, might, suggests, can)

• Theoretical claims about mechanisms: "This process may operate through..."
• Predictions about findings: "We expect that..."
• Generalizations beyond the sample: "These dynamics may apply to..."
• Causal claims without experimental evidence: "X may lead to Y"

When to Assert (demonstrates, shows, establishes, is)

• Reporting prior literature: "Studies demonstrate..."
• Defining concepts: "Recognition refers to..."
• Gap statements: "We know little about..." (assertive about the gap)
• Research questions: "This study examines..." (unhedged)

Hedging Vocabulary Spectrum

---

4. Cluster-Specific Sentence Patterns

Gap-Filler

• Heavy context-setting and literature-announcing openers
• Contrastive markers ("however," "yet") for the turn
• Efficient parenthetical citation strings
• Moderate hedging, assertive gap statements

Theory-Extender

• Theory-introducing openers with named theorists
• "According to [Scholar]"; "Building on"; "Drawing on"
• Author-subject citation pattern dominant
• Lower hedging on established theory

Concept-Builder

• Definition-providing openers for new terminology
• Causal chains ("as a result," "thus")
• Quote-then-cite for detailed engagement
• Low hedging on conceptual definitions

Synthesis Integrator

• Literature-announcing and theory-introducing mixed
• Bridging phrases ("bringing together," "connecting")
• Citations from multiple traditions juxtaposed
• Moderate hedging on integration claims

Problem-Driven

• Context-setting dominant (50%+ for Empirical sub-type)
• Contrast-signaling for debates ("some argue," "others contend")
• Temporal markers establishing historical context
• Higher hedging on competing explanations

---

Quick Reference: Sentence Starter Bank

Starting a Synthesis Paragraph

• "A growing body of research examines..."
• "Over the past [X] decades, scholars have..."
• "Research on [topic] demonstrates that..."

Starting a Theory Paragraph

• "The concept of [X] draws on..."
• "[Scholar] (Year) theorizes [concept] as..."
• "Central to this framework is..."

Starting the Turn

• "Yet we know little about..."
• "However, less attention has been paid to..."
• "What remains unclear is..."

Stating Your Contribution

• "This study examines..."
• "I/We argue that..."
• "This paper integrates..."