Technicality in methods often disguises the normativity of methodological choices and of concept formation itself. Behind this apparent neutrality lies real normative weight, e.g., who is included in the population, what counts as a case, and whose usage of a concept counts as authoritative. This panel aims to bring that backgrounded normativity to the fore as an explicit phenomenon to be discussed from multiple angles.
We are especially interested in the normative dimensions of concept (mis)formation, in the political stakes of essentially contested concepts, and in what gets lost, or added, when concepts travel across regions and languages. Across all three, the same question recurs: who decides what a concept means, and on what authority?
On method, we ask that same question of every approach on offer: where does normativity sit in set-theoretic methods, in interpretive approaches, in computational text-as-data and machine-learning analysis, in multi-method design? Papers that treat methodological reflexivity as a substantive issue, not a caveat in the appendix, are especially welcome.
Submissions may come from any subfield of political science. We particularly welcome work that treats concepts under strain, e.g., democratic backsliding in regime research, frozen conflicts in peace studies, polarization in political psychology, crisis in international relations and public policy (from 'permacrisis' discourse to competing definitions of state emergency). Critical and decolonial perspectives on the cross-linguistic and cross-cultural travel of concepts are especially relevant here.
The panel aims to assemble four to six papers reflecting genuine diversity of methodological tradition, career stage, and region.
Suggested themes:
Concept formation and measurement: concept formation and measurement decisions, e.g., calibration thresholds in fuzzy-set QCA, coding decisions in political regime indices, or family-resemblance typologies of populist movements.
Conceptual travel as normative labor: what is gained, lost, or smuggled in when concepts cross regions and languages, e.g., legitimacy reinterpreted between the Global North and South.
Naming as a normative act: the political stakes of choosing a label (crisis, emergency, resilience, decline), who is authorized to apply it, and what follows politically from that choice.
Whose norms train the model: the normative assumptions embedded in text-as-data and machine-learning pipelines, e.g., who labels the training data for right-wing extremist rhetoric analysis.
Reflexivity in multi-method research: multi-method designs that treat researcher positionality, in every regard.
