--- name: govern-theory-building description: Use when building the argument and theoretical contribution of a Governance: An International Journal of Policy, Administration, and Institutions manuscript — to theories of the state, institutions, the policy process, bureaucracy, governance networks, and accountability, with portability across cases and disciplined concept formation. Builds the argument; it does not test it. --- # Theory Building (govern-theory-building) A *Governance* contribution is judged by what it adds to how we understand **governing** — the state, institutions, the policy process, bureaucracy, governance networks, accountability. The argument must be sharp enough to test (`govern-research-design`) and **portable** enough to matter beyond one case. This skill builds the argument and its concepts; it does not run the empirics. ## When to trigger - Stating the core argument, mechanism, and scope conditions - A reviewer said the theory is "ad hoc," "just a description," or "won't travel" - Defining or refining a concept for comparative use - Specifying what the theory predicts that rivals do not ## Build a portable argument 1. **Name the mechanism, not just the correlation.** State *why* governing produces the outcome — the causal logic (delegation incentives, blame avoidance, institutional layering, policy feedback, network coordination). A mechanism is what lets the claim travel. 2. **State scope conditions explicitly.** Where does the argument hold — which regime types, levels of government, policy domains, capacity levels? Scope conditions are a feature, not a hedge; they make portability honest. 3. **Specify observable implications in advance.** What should we see in the cases/countries if the argument is right — and what would we see instead if a rival institutional account were right? 4. **Connect to a theory others can use.** Tie the contribution to an existing family (historical institutionalism, principal-agent, the regulatory state, multilevel governance) so the field can build on it — even when you revise that family. ## Concept formation for comparative work (avoid concept stretching) Comparative governance lives or dies on concepts. Apply Sartori-style discipline: - **Define by attributes, then by cases.** Fix the defining attributes (intension) before deciding which cases fall under the concept (extension); do not stretch a concept to fit a convenient case. - **Mind the ladder of abstraction.** To travel to more cases, drop attributes (move up); to gain precision, add them (move down). State where on the ladder your concept sits. - **Establish equivalence across cases.** "Accountability," "agency," "reform," "capacity" do not mean the same thing in every system — show functional/contextual equivalence before you compare. - **Avoid the degreeism / dichotomy trap.** Be explicit about whether the concept is categorical or graded, and justify the choice for the comparison at hand. ## Contribution types that land at Governance - A **new or revised mechanism** for a governing phenomenon, with scope conditions. - A **reconceptualization** that resolves a measurement or classification problem comparatively. - A **theory portability claim** — an account known in one context shown to (not) travel, and why. - A **scope-condition correction** — a leading theory holds *only* under conditions the field missed. ## The strongest-rival test (Governance-specific) For the **single strongest rival institutional explanation**, write one sentence: *"If the rival account of governing were right rather than mine, we would observe ___ across cases; my argument instead implies ___."* If you cannot write it, the theory is not yet a testable contribution — it is a description waiting for a mechanism. ## Anti-patterns - A "framework" that organizes cases but predicts nothing - Concept stretching: relabeling a familiar concept to claim novelty, or applying it where attributes don't hold - Scope conditions left implicit, so the claim silently over-generalizes - A mechanism that cannot distinguish your argument from the leading institutional rival - Theory and case so fused that the argument cannot travel to a second country ## Output format ``` 【Core claim】one sentence 【Mechanism】why governing produces the outcome 【Scope conditions】where it holds (regimes / levels / domains / capacity) 【Key concept(s)】definition + position on ladder of abstraction + equivalence note 【Rival ruled out】the strongest-rival sentence 【Observable implications】what we should/should not see across cases 【Next】govern-research-design ``` ## Supplementary resources - [`../../resources/external_tools.md`](../../resources/external_tools.md) — concept-formation and theory-mapping references - [`../../resources/official-source-map.md`](../../resources/official-source-map.md) — *Governance* scope and contribution expectations