---
slug: research-engagement-primitives
status: settled
authored: 2026-05-19
provenance: agent-synthesis
temporal_contract: write-once-on-converge
---

# Research-engagement primitives — derived from five external methodology traditions

> Ported research position behind ARD v0.1 — the reasoning trace for a framework commitment; see [../../SPEC.md](../../SPEC.md). The attestation layer is not reproduced in this repo (it paraphrases copyrighted sources), so the load-bearing passages are quoted inline and the works are listed under [§Sources](#sources) — the reasoning stands on its own here.

## What this position is about

When an agentic research framework commits to concepts like "the engagement unit," "scope authority," or "disconfirmation," it is making claims about *how research engagement is structured*. Those claims could be designed from first principles, derived from an existing system's needs, or grounded in how established research traditions actually operate. This position records a derivation of the third kind.

Five research methodology traditions were engaged — grounded theory, qualitative data analysis and coding methods (the Miles/Huberman/Saldaña lineage), philosophical hermeneutics, systematic review and evidence synthesis, and Zettelkasten — and each was asked how its own practitioners structure multi-source research engagement: how they determine the unit of engagement, how they govern scope, what output forms they produce, how they accumulate material across sources without premature synthesis, and how they treat disconfirmation. The traditions were engaged independently, in their own vocabulary, with no target schema in view. Five primitives were derived from cross-tradition convergence at the synthesis tier.

This position is tradition-derived substrate. It is deliberately not a framework schema and carries no mapping to any implementation. That mapping is the downstream consumer's work.

## The five traditions in brief

Before stating the primitives, it helps to know what each tradition is and what kind of engagement it describes.

**Grounded theory** is a methodology for constructing theory inductively from qualitative data — interview transcripts, field notes, documents. Its defining commitment is that theory should *emerge from* systematic engagement with data rather than be tested against it: "the discovery of theory from data—systematically obtained and analyzed in social research—can be furthered" [glaser-discovery-of-grounded-theory]{35}. The two foundational techniques are *theoretical sampling* (ongoing, theory-driven decisions about what to look at next) and the *constant comparative method* (comparing every new datum against every previously established code or category). Theoretical *saturation* — the point at which fresh data no longer reveals new properties of a category — is the stopping criterion [glaser-discovery-of-grounded-theory]{35}. The tradition has an internal split: Glaser and Strauss's 1967 founding text treats coding as a continuous integrated process with no named phases; Corbin and Strauss's subsequent branch adds three named coding phases (open, axial, selective) and a structured analytic paradigm [corbin-basics-of-qualitative-research]{37}. Charmaz's constructivist rearticulation is a third voice, flexible and process-oriented [charmaz-constructing-grounded-theory]{36}.

**Qualitative data analysis and coding methods** (the Miles/Huberman/Saldaña lineage) is a practitioner-facing methodology for analyzing primarily text-based qualitative data. Its central model is an *interactive* and iterative process: data condensation (selecting, focusing, simplifying, abstracting), data display (organized visual arrays — matrices, networks), and conclusion drawing/verification [miles-qualitative-data-analysis]{38}. The tradition names an explicit spectrum of research design tightness: "A case can be made for tight, prestructured qualitative designs and for loose, emergent ones. Whether you choose one end of the continuum or the other for your particular study must be your decision" [miles-qualitative-data-analysis]{38}. Conclusion-drawing is supported by a 26-tactic apparatus for generating meaning and testing/confirming findings, including active disconfirmation: "You are actively seeking disconfirmation of what you think is true" [miles-qualitative-data-analysis]{38}.

**Philosophical hermeneutics** is importantly *not* a research methodology in the empirical sense. It is a philosophical account of what *understanding itself is*. Gadamer's position — the tradition's founding modern statement — is that understanding is not achieved by applying a method but is an event the interpreter participates in, governed by the interpreter's historical situatedness and by the hermeneutic circle: every reading begins with a *fore-projection* of the whole that gets revised as the parts are engaged [gadamer-truth-and-method]{30}. "The circle, then, is not formal in nature. It is neither subjective nor objective, but describes understanding as the interplay of the movement of tradition and the movement of the interpreter" [gadamer-truth-and-method]{30}. Ricoeur's variant gives the tradition more structure: interpretation moves from a naive guess at meaning, through structural explanation and validation ("a logic of uncertainty and of qualitative probability"), to an *appropriation* in which the text's world becomes part of the interpreter's own expanded horizon [ricoeur-interpretation-theory]{31}. The tradition is explicitly anti-procedural — and that is precisely what makes its contribution to this derivation a *constraint on* the other four traditions' results rather than a parallel set of claims.

**Systematic review** is a tradition from healthcare evidence synthesis. Its defining commitment is the *protocol*: before any source is searched or engaged, the review's question, inclusion criteria, search strategy, and analysis plan must be written and publicly registered. The Cochrane Handbook states: "All Cochrane reviews must have a written protocol, specifying in advance the scope and methods to be used by the review, to assist in planning and reduce the bias in the review process" [cochrane-handbook]{39}. The logic is bias-prevention: if scope can evolve after findings start coming in, then findings can influence which evidence counts, destroying the validity of the synthesis. The tradition uses the GRADE framework to grade the certainty of a body of evidence — not individual claims, but the whole synthesized pool [cochrane-handbook]{39}. Systematic review is therefore the tradition that most explicitly *opposes* emergent judgment about scope.

**Zettelkasten** is a personal knowledge management methodology derived from sociologist Niklas Luhmann's actual practice and codified by Ahrens. Its defining architecture is a layered note-type system — fleeting notes (temporary), literature notes (source-proximate, brief), and permanent notes (translated into the note-taker's own language and entered into the slip-box) [ahrens-smart-notes]{32}. The permanent notes link to each other non-hierarchically; the connections and emergent clusters are the system's intellectual product, not the notes themselves. Luhmann framed the system as a tool for productive communication: "It is useful...to create in the system of notes a competent partner of communication" [luhmann-communicating-slip-boxes]{33}. Disconfirmation is built into the note-comparison act — Ahrens urges us to "prime ourselves for seeking out disconfirming facts... disconfirming data becomes suddenly very attractive, because it opens up more possible connections and discussions within the slip-box, while mere confirming data does not" [ahrens-smart-notes]{32}. The scope of what gets added to the slip-box is set entirely by the note-taker's situational judgment; there is no protocol.

## Disconfirming analysis

Before committing each primitive, the cross-tradition counter-evidence was actively searched. The substantive disconfirming checks per primitive:

- **Primitive 1**: the first synthesis draft claimed five-of-five convergence on "the unit is purpose-determined and discovered in-engagement." Systematic review is the genuine disconfirming case — its units are fixed by protocol before engagement and do not shift as findings emerge. This was corrected to a two-part claim: claim (a) holds five-of-five; claim (b) holds four-of-five with systematic review as the named exception. The analysis also searched for a tradition treating the source material's own structure as *deciding* scope (rather than merely informing it) — none was found.
- **Primitive 2**: the analysis searched for a tradition that treats the engagement unit and scope authority as fully independent decisions. QDA came closest — it explicitly names a spectrum — but ultimately frames unit and authority as one design-time decision, not two.
- **Primitive 3**: hermeneutics is the principled abstainer on the two-class distinction; Ricoeur's hermeneutic arc adds a *phase structure* for a single engagement without introducing an accumulating-substrate-versus-converging-artifact vocabulary. QDA's data displays, which can be simultaneously accumulative and convergent within one artifact-type, were added as a qualifier after the adversarial read flagged their omission.
- **Primitive 4**: hermeneutics was the principal counter-case — the function of holding accumulated material before synthesis is present (tradition, *Wirkungsgeschichte*), but no externalized artifact is named. QDA is a second partial counter-case: the function is present; it is explicitly not formalized. The primitive is committed with explicit variance in formalization.
- **Primitive 5**: the analysis searched for a tradition with no disconfirmation commitment — none was found. The discrete-vs-continuous modal split was itself a disconfirming-search outcome: an initial single-mode framing did not survive contact with the five traditions, and a clean two-mode binary did not survive the QDA and hermeneutics straddle cases.

## The five primitives

### Primitive 1 — The engagement unit is purpose-determined, and in most traditions discovered in-engagement

**What "the engagement unit" means.** At any given moment, a researcher is engaging with some *unit* of source material — a word, a line, a section, a whole study, a body of studies. How does that unit get determined? Who or what sets its granularity?

**Claim (a): Purpose, not source structure, sets the unit — five of five traditions.** No tradition makes the engagement unit a property of the sources themselves. Grounded theory's granularity is set by where the emerging theory needs elaboration — theoretical sampling "means seeking pertinent data to develop your emerging theory" and targets gaps in the theoretical categories [charmaz-constructing-grounded-theory]{36}; QDA's coding granularity is chosen against the research questions and the analyst's developing sense of what matters, with data segments ranging from a single word to a full paragraph [miles-qualitative-data-analysis]{38}; hermeneutics governs the unit by the part-whole coherence the interpreter is seeking [gadamer-truth-and-method]{30}; Zettelkasten's "one idea per note" is set by future recombinability — "In which context will I want to stumble upon it again?" [ahrens-smart-notes]{32}; systematic review's units (the study, the outcome, the body of evidence) are set by the review's question and methodology, not by the literature's structure [cochrane-handbook]{39}. The source's own structure *informs* the unit but does not *decide* it.

**Claim (b): The unit is discovered in-engagement — four of five; systematic review is the principled exception.** In grounded theory, QDA, hermeneutics, and Zettelkasten, the unit is not fixed in advance: it is found, shifted, and re-judged during engagement. In grounded theory, granularity shifts across the arc — from word or line, to code, to category, to theory — and the theoretical sampling process itself shifts its comparative range [glaser-discovery-of-grounded-theory]{35}. Hermeneutics holds the relevant whole is genuinely discovered, never fixable in advance [gadamer-truth-and-method]{30}.

Systematic review is the principled exception. Its units — the study, the outcome, the body of evidence — are fixed by the methodology before engagement begins, as a deliberate bias-control commitment. "Reliable conclusions can only be drawn from analyses that are truly pre-specified" [cochrane-handbook]{39}. Pre-specifying units prevents post-hoc scope adjustment that lets findings influence which evidence counts.

**The unit-question is itself plural.** QDA shows that "the engagement unit" is not one decision. The *case* (a bounded context — site, participant, time period) is often defined at design time; *coding granularity within the case* is a continuous in-engagement choice [miles-qualitative-data-analysis]{38}. A tradition can fix one level of unit while leaving another emergent. This plurality is why Primitive 1 and Primitive 2 are tightly coupled but not identical.

### Primitive 2 — Scope authority sits on one axis: pre-registered protocol vs in-engagement judgment

**What "scope authority" means.** Scope authority is the question of *who* or *what* holds authority over when the engagement's scope is fixed, and how open it remains to revision during engagement. This is not about the size of the scope; it is about whether scope is committed before engagement begins or allowed to evolve as the work proceeds.

**The axis.** The cross-tradition finding is that scope authority sits on a single recurring axis:

- **Pre-registered-protocol pole**: systematic review. The protocol — written, publicly registered, fixed before any source is engaged — holds scope authority. Reviewer judgment operates only within protocol constraints. The tradition is explicitly hostile to scope evolution during engagement [cochrane-handbook]{39}.
- **In-engagement-judgment pole**: Zettelkasten and hermeneutics. Zettelkasten places scope authority entirely with the note-taker's situational judgment, and does not prescribe a predetermined reading order [ahrens-smart-notes]{32}. Hermeneutics places it with neither protocol nor arbitrary will, but with the encounter itself — scope is disclosed through engagement, through cultivated judgment, not procedure [gadamer-truth-and-method]{30}.
- **The middle, and the explicit naming of the axis**: QDA frames this as "loose vs tight research design" — "A case can be made for tight, prestructured qualitative designs and for loose, emergent ones. Whether you choose one end of the continuum or the other for your particular study must be your decision" [miles-qualitative-data-analysis]{38}. Grounded theory sits in the middle: data collection is "controlled by the emerging theory" [glaser-discovery-of-grounded-theory]{35}, meaning authority is exercised by the researcher but driven by the theoretical gaps the analysis surfaces.

**What did not survive contact with the traditions.** A candidate third authority-pole — the source material's own structure deciding scope — did not survive. No tradition locates scope authority in the source's structure. The material informs the unit; it does not hold authority over scope.

**One commitment unique to grounded theory.** The canonical instruction to delay the literature review until an independent analysis has formed — to prevent prior concepts from contaminating emergent categories [charmaz-constructing-grounded-theory]{36} — is the inverse of systematic review's stance. Prior-knowledge management is itself a scope-authority variable the traditions handle oppositely.

**Why this axis recurs.** The protocol-vs-emergent axis is a single distinction that recurs across multiple primitives: it governs what the unit is (Primitive 1(b)), who fixes it, and how disconfirmation fires (Primitive 5). Whether this means these are "really" one axis or three correlated axes is left open; the synthesis observes the recurrence without asserting a count.

### Primitive 3 — Research engagement produces two output classes: accumulating substrate and converging artifact

**What the distinction is.** Research engagement produces two structurally different kinds of analytical output. The distinction is not about the *content* of what is produced; it is about the *relationship between output and convergence horizon*:

- An **accumulating substrate** grows without a convergence horizon. Material is added without a commitment to producing a finished output at any particular moment. Examples: grounded theory's codes and memos; QDA's running code list and assertion list; systematic review's per-study evidence tables; Zettelkasten's slip-box.
- A **converging artifact** is assembled from the substrate at a chosen moment, committing to a synthesis. Examples: grounded theory's integrated theory; QDA's themes and final report; systematic review's pooled estimate and Summary-of-Findings table; Zettelkasten's drafted manuscript [ahrens-smart-notes]{32}.

Zettelkasten names this distinction most explicitly — the permanent slip-box versus the occasional manuscript [luhmann-communicating-slip-boxes]{33}. Systematic review makes it operational — per-study extraction (accumulating) versus cross-study meta-analysis (converging). Grounded theory and QDA carry it implicitly in the distinction between the data-accumulation phase and theory-writing.

**Three riders from the traditions:**

1. **Whether a converging artifact is produced at all is a variable.** Grounded theory and systematic review treat the converged artifact as the telos. QDA treats conclusion-drawing and verification as one of three concurrent analytic flows that "shuttles" iteratively throughout the study rather than mandating a single terminal *theory* [miles-qualitative-data-analysis]{38} — its pragmatic-realist, methodologically eclectic posture allows varying degrees of theoretical formalization, and its standard terminal output is a final write-up rather than a mandated theory. Zettelkasten treats the slip-box as primary and permanent; manuscripts are occasional byproducts [ahrens-smart-notes]{32}. A research engagement may thus legitimately rest in an extended accumulating substrate rather than a single theoretically-converged artifact.

2. **The accumulating substrate is internally layered by source-proximity.** The artifacts a research engagement produces are ordered from source-proximate to synthesis-proximate: a source-faithful record (Zettelkasten's literature note, systematic review's per-study extraction), the researcher's own accumulating thought (the permanent note, the memo, the running assertion list), and the converged synthesis. Zettelkasten makes this layering explicit: literature notes are brief, page-anchored, context-preserving; permanent notes are translated into the note-taker's own language to stand independently of the source [ahrens-smart-notes]{32}.

3. **The two classes are not always separate artifacts.** QDA's data displays can be simultaneously accumulative (cases added to a matrix over time) and convergent (a causal network representing a conclusion) within a single artifact-type. The two-class distinction is the dominant pattern; it is not a clean partition of every artifact.

**Hermeneutics abstains from this primitive.** Gadamer describes the event of understanding, not output forms or deliverables. Ricoeur qualifies the abstention: his hermeneutic arc gives interpretation a phase structure (naive guess → structural explanation/validation → appropriation) [ricoeur-interpretation-theory]{31}. But this is a theory of understanding's stages within a single engagement, not a two-class output typology. Hermeneutics is the principled exception here.

### Primitive 4 — There is a deferred-synthesis intermediate

**What this means.** Most traditions have a structured intermediate form whose explicit job is to hold accumulated cross-source material *without* premature synthesis. This is Primitive 3's accumulating substrate viewed not by form but by *function* — the function of preventing a researcher from synthesizing before sufficient material has accumulated.

- **Grounded theory**: the memo — "the pivotal intermediate step between data collection and writing drafts of papers. When you write memos, you stop and analyze your ideas about the codes in any-and every-way that occurs to you during the moment" [charmaz-constructing-grounded-theory]{36}.
- **Systematic review**: evidence tables held under an explicit heterogeneity gate — meta-analysis is withheld until pre-specified similarity criteria are met. The I² statistic quantifies whether heterogeneity is low enough to justify pooling [cochrane-handbook]{39}.
- **Zettelkasten**: the slip-box itself — links and emergent clusters accumulate ideas across sources with no classification scheme and no synthesis commitment until a manuscript is drawn [ahrens-smart-notes]{32}.
- **QDA** (partial): a running list of assertions and propositions, revised as evidence appears that confirms or disconfirms them, and cross-case meta-matrices that array cases for comparison before any cross-case claim [miles-qualitative-data-analysis]{38}. QDA exercises the deferred-synthesis function through these artifacts without naming a dedicated "pre-synthesis holding area" as a term of art — the tradition is more oriented toward moving iteratively among condensation, display, and conclusion-drawing than toward formalizing a distinct pre-synthesis state.
- **Hermeneutics** (abstain): tradition (*Wirkungsgeschichte*) is the accumulated cross-source record that conditions every interpreter's horizon — but it is not a discrete artifact the researcher manages. Hermeneutics does not externalize the intermediate at all.

**What the convergence finds.** A deferred-synthesis intermediate is a genuine cross-tradition primitive, but with real variance in formalization. Three traditions name a dedicated artifact; one exercises the function without naming it; one operates the function entirely inside the interpreter without externalizing it. The primitive is committed with explicit variance in formalization.

### Primitive 5 — Disconfirmation is a structural commitment, on a spectrum from checkpointed to embedded

**What "disconfirmation" means here.** All research methodologies treat some form of asking "what would falsify this?" or "what evidence opposes this conclusion?" The question for this derivation is not *whether* traditions do this, but *when* and *how* they do it.

**All five traditions treat disconfirmation as load-bearing, not optional.** This is the clearest cross-tradition convergence in the primitive set. The differences are in mode, not in whether disconfirmation is required.

**The spectrum.** The mode ranges from discrete operations fired at named checkpoints to a continuous property of the comparison act:

- **Discrete-checkpoint pole** (protocol tradition): systematic review fires disconfirmation as named operations at named points — risk-of-bias assessment per study, heterogeneity analysis at synthesis time, and sensitivity analyses after synthesis [cochrane-handbook]{39}. (GRADE, separately, grades the certainty of the synthesized evidence pool — a quality-rating step, not itself a disconfirmation operation.)
- **Continuous-embedded pole** (emergent traditions): grounded theory builds disconfirmation into the constant comparative method — the founding text contrasts grounded theory with *analytic induction*, which organizes itself around deliberate negative-case search. Grounded theory's own disconfirmation mechanism is *maximizing group differences* — "The sociologist does not merely look for negative cases bearing on a category (as do others who generate theory); he searches for maximum differences among comparative groups" [glaser-discovery-of-grounded-theory]{35}. Zettelkasten builds disconfirmation into note-comparison: contradictions surface automatically as new notes are compared to existing ones, and disconfirming material is actively primed for during reading [ahrens-smart-notes]{32}.
- **Straddle cases**: QDA has both discrete tactics ("looking for negative evidence" is one of a named set of testing tactics, alongside rival explanations and extreme cases [miles-qualitative-data-analysis]{38}) and a continuous standing orientation toward holding conclusions provisionally, captured in the instruction to keep a running list of assertions "and revise them as field work continues and evidence appears" [miles-qualitative-data-analysis]{38}. Hermeneutics, with Ricoeur, also straddles: the fundamental mechanism is continuous (the encounter with the text's resistance revises the fore-projection at every moment), but Ricoeur's validation phase gives disconfirmation a more structured moment — "a logic of uncertainty and of qualitative probability" with "criteria of relative superiority" [ricoeur-interpretation-theory]{31}.

**The mode tracks Primitive 2's axis.** Phased/protocol traditions checkpoint their disconfirmation; emergent traditions embed it continuously. QDA and hermeneutics straddle because they have both phased and continuous structure. The derivation does not collapse this to a clean binary.

**The five primitives are interdependent.** Primitive 2's axis governs Primitives 1(b) and 5: the same protocol-vs-emergent dimension that determines who fixes the unit (Primitive 1) also determines whether disconfirmation is checkpointed or embedded (Primitive 5). Primitive 4 is Primitive 3 by another name — the accumulating substrate viewed by function. This position does not assert a definitive count of how many truly independent primitives there are; any collapse below five is an organizing choice left to the consumer.

## The hermeneutic boundary condition

Hermeneutics' contribution is not just Primitive 5's straddle case. The tradition also contributes a *constraint on how the whole primitive-set may legitimately be operationalized* — and this constraint is the boundary condition any consumer of these primitives should apply.

The tradition resists the procedural framing the other four traditions assume. Gadamer's core position is that genuine understanding is not achieved by applying a method: hermeneutics works to clarify the conditions in which understanding takes place rather than to prescribe a method for producing it [gadamer-truth-and-method]{30}. This is not an obstacle to using the primitives; it is a constraint on what any operationalization of them can legitimately claim.

Ricoeur qualifies Gadamer without overturning him. Ricoeur gives interpretation a genuine procedural middle phase — structural explanation and validation — but is explicit that this procedure covers only the middle phase of a three-stage arc: "In the beginning, understanding is a guess. At the end, it satisfies the concept of appropriation. Explanation, then, will appear as the mediation between two stages of understanding" [ricoeur-interpretation-theory]{31}. The arc's entry (the initial naive guess at the text's meaning) and its exit (appropriation, which is the interpreter's own horizon expanded by the encounter) remain outside the methodological domain. Even the hermeneutic figure most amenable to method confines method to one phase of one engagement.

Three cautions that fall out, each qualifying a primitive:

- **On Primitive 1**: the engagement unit is genuinely discovered, never fixable in advance (in the hermeneutic account). A deployment that hard-codes the unit risks foreclosing the part-whole revision that is constitutive of understanding. Systematic review shows this is possible for its purpose; hermeneutics shows it is one option on an axis, not a default.
- **On Primitive 3**: convergence is never final closure for hermeneutics — Ricoeur's appropriation is self-enlargement, a "moment of dispossession of the egoistic and narcissistic ego," not a terminal artifact [ricoeur-interpretation-theory]{31}. A converging artifact is a provisional stopping point, not a closed terminus.
- **On Primitive 5**: disconfirmation that operates only through discrete procedural checks misses what hermeneutics names — the continuous, cultivated openness to the text's resistance that no procedure fully captures.

**The boundary condition stated plainly.** The five primitives describe the *structure* of research engagement; they are not a procedure. A deployment that hardens them into a rigid protocol — fixing the engagement unit mechanically, treating a converging artifact as a closed terminus, reducing disconfirmation to a checklist — loses what the hermeneutic tradition names as the enabling conditions of genuine understanding. Systematic review demonstrates that a fully protocol-hardened engagement is possible and, for its purpose, correct. The boundary condition is that protocol-hardening is *one position on Primitive 2's axis*, not the default any implementation slides into.

Any consumer of these primitives should ask, per committed design decision: does fixing this as a schema value harden something that must remain a judgment?

## Contradictions

### Cross-tradition contradictions

1. **Protocol-predetermination vs emergent-judgment.** Systematic review fixes scope before engagement and is hostile to scope evolution; grounded theory, Zettelkasten, and hermeneutics fix scope during engagement. QDA names this as a spectrum. The poles disagree about whether letting findings reshape scope is a bias to be prevented or the basic mechanism of discovery.

2. **Is the converging artifact the telos or an occasional byproduct?** Grounded theory and systematic review treat the converged artifact as the engagement's purpose. QDA's iterative, concurrent flows — shuttling among condensation, display, and conclusion-drawing — do not mandate a single terminal theory. Zettelkasten treats the accumulating slip-box as primary and manuscripts as occasional. The traditions disagree about whether a research engagement *must* terminate in a converged artifact.

3. **Does anything ever fully converge?** Hermeneutics holds that understanding stays constitutively open; Ricoeur's appropriation is self-enlargement, not closure [ricoeur-interpretation-theory]{31}. The other four traditions have terminal artifacts. A genuine disagreement about whether "converged" is a real state or only a provisional stopping point.

4. **Disconfirmation: checkpointed vs embedded.** Systematic review checkpoints disconfirmation at named operations; grounded theory and Zettelkasten embed it as a continuous property of comparison. QDA and hermeneutics straddle. Tied to contradiction 1 — the mode follows the protocol-vs-emergent axis.

### Intra-tradition contradictions

5. **Grounded theory is internally contested.** The 1967 founding text contains no axial coding, no coding paradigm, and no named coding phases — coding is a continuous integrated process [glaser-discovery-of-grounded-theory]{35}. Corbin and Strauss's later branch adds three named coding phases (open, axial, selective) and an analytic paradigm, framed by Corbin as "a tool and not a set of directives" [corbin-basics-of-qualitative-research]{37}. A second divergence: the founding text rejects deliberate negative-case search as the mark of *analytic induction* and commits instead to maximizing group differences [glaser-discovery-of-grounded-theory]{35}; Corbin and Strauss re-prescribe negative-case search as "useful" [corbin-basics-of-qualitative-research]{37}.

6. **Hermeneutics is internally divided between Gadamer and Ricoeur.** (a) *The standing of structural explanation*: Gadamer holds structural analysis is not part of hermeneutics as such; Ricoeur holds structural explanation is a necessary internal phase of interpretation, dialectically related to understanding. (b) *The text as dialogue-partner vs autonomous object*: Gadamer holds that entering into dialogue with the text is not a metaphor but a structural description [gadamer-truth-and-method]{30}; Ricoeur holds that "[h]ermeneutics begins where dialogue ends" — writing ruptures the dialogical situation and the text is semantically autonomous [ricoeur-interpretation-theory]{31}. The two converge on appropriation and self-transformation but diverge on what hermeneutics is a theory *of*.

## What no tradition grounds

Two engagement concerns were searched for and not found in any of the five traditions:

- **A researcher validating their own prior claim** (self-as-claim-holder). There is a reflexivity theme across the traditions — QDA's checking for researcher effects, hermeneutics' interpreter-changed-by-engagement, grounded theory's constructivist reflexivity — but no tradition names a primitive for a researcher deliberately validating a claim they themselves hold. The traditions assume the claim under test is the data's or a study's, not the researcher's own.
- **Research that has converged but awaits a downstream consumption decision** (converged-but-pending). The closest is systematic review's "living review" updating obligation — a review is a provisional document, updated when new evidence arrives. But "converged-and-held-pending-a-downstream-commitment" specifically is not a state any tradition names.

Both belong in a deployment's workflow, not in a framework's architecture. A consumer that needs either should treat it as a deployment-specific extension.

## Revisit if

- A consumer's mapping of these primitives against a concrete schema surfaces a gap — the pressure returns here rather than being absorbed silently downstream.
- Glaser's post-1990 polemic writings are acquired — they would sharpen Primitive 5's intra-grounded-theory contradiction (the founding text's rejection of negative-case search vs Corbin's re-prescription) by adding Glaser's specific objections to the Straussian additions [charmaz-constructing-grounded-theory]{36}.
- A sixth methodology tradition is engaged that grounds either of the two un-grounded concerns (action research for self-as-claim-holder; a "living evidence" methodology for converged-but-pending).
- A consumer's operationalization forces a primitive to split or merge — the question of whether the five collapse to a smaller independent set is deliberately left for the consumer and should return here if resolved.

## Sources

- **[glaser-discovery-of-grounded-theory]{35}** — Barney G. Glaser and Anselm L. Strauss, *The Discovery of Grounded Theory: Strategies for Qualitative Research* (1967; AldineTransaction reprint 1999, ISBN 0-202-30260-1). Engaged: Chapter I (the founding claim, pp. 1–19), Chapter III (Theoretical Sampling, pp. 45–79), Chapter V (The Constant Comparative Method, pp. 101–125).
- **[charmaz-constructing-grounded-theory]{36}** — Kathy Charmaz, *Constructing Grounded Theory: A Practical Guide Through Qualitative Analysis* (SAGE Publications, 2006, ISBN 978-0-7619-7353-9). Engaged: Chapters 1–8.
- **[corbin-basics-of-qualitative-research]{37}** — Juliet Corbin and Anselm Strauss, *Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory*, Fourth Edition (SAGE Publications, 2015, ISBN 978-1-4129-9746-1). Engaged: Chapters 1, 2, 5, 7, 8.
- **[miles-qualitative-data-analysis]{38}** — Matthew B. Miles, A. Michael Huberman, and Johnny Saldaña, *Qualitative Data Analysis: A Methods Sourcebook*, Fourth Edition (SAGE Publications, 2020, ISBN 978-1-5443-7185-6). Engaged: Chapters 1, 2, 4, 11.
- **[gadamer-truth-and-method]{30}** — Hans-Georg Gadamer, *Truth and Method* (*Wahrheit und Methode*, 1960), revised second edition, trans. Joel Weinsheimer and Donald G. Marshall (Bloomsbury Academic, 2013). Engaged: Part Two, Chapter 4 ("Elements of a Theory of Hermeneutic Experience"), pp. 278–376.
- **[ricoeur-interpretation-theory]{31}** — Paul Ricoeur, *Interpretation Theory: Discourse and the Surplus of Meaning* (Texas Christian University Press, 1976). Engaged: all four essays plus Introduction and Conclusion, with primary focus on Essay 4 ("Explanation and Understanding," pp. 71–88) and Conclusion (pp. 89–95).
- **[cochrane-handbook]{39}** — Julian Higgins and James Thomas (senior editors), *Cochrane Handbook for Systematic Reviews of Interventions*, version 6.5 (Cochrane, 2024). Available at: <https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current>. Engaged: Chapters I (Introduction), II (Planning), X (Meta-analysis), XII (Alternative synthesis methods), XIII (Publication bias), XIV (GRADE/Summary of findings).
- **[ahrens-smart-notes]{32}** — Sönke Ahrens, *How to Take Smart Notes: One Simple Technique to Boost Writing, Learning and Thinking*, 2nd edition (2022). Engaged: Introduction (§§1–4), Chapter 2 (§§5–8), Chapter 3 (§§9–14), Appendix (§15).
- **[luhmann-communicating-slip-boxes]{33}** — Niklas Luhmann, "Communicating with Slip Boxes" (1992; German original c. 1980/81), trans. Manfred Kuehn. Available at: <https://luhmann.surge.sh/communicating-with-slip-boxes>. Engaged as cited in Ahrens [ahrens-smart-notes]{32} and separately.
