Rules as data

Whereas regulatory content has often been gauged through experts’ opinions, short summaries, or outputs, its direct coding is gaining attention as an alternative empirical ground. Ideally, a coded corpus can serve different theories and be treated with different techniques to corroborate or bridge them. Moreover, it makes technical details available for transfer and learning. These promises, however, stand depending on how rules’ content is coded. Coding organizes information into evidence of a meaningful construct. This sense-making operation entails a selection of empirics according to theoretical criteria of relevance and, hence, a purposeful disregard for the remaining information. As especially emphasized in debates on indicators, relevant information from the viewpoint of some theoretical concern may result in biases outside its original scope. The solution toward more reusable coding consists of turning to lighter schemas and organizing information by frameworks instead. The Institutional Analysis and Development framework suggests two coding strategies. The first renders the seven special functions that rules perform in structuring a specific ‘action situation’ – for instance, of selecting participants, distributing information, or making certain options available. The second understands a rule as an analytic unit whose indispensable components are grammatical instead and span from the attributes that identify the addressees to the cogency of the prescription and the consequences of non-compliance. A further strategy borrows from the longstanding logic model diffused in organizational and policy design to shape the rationale of plans. In its essential version, a logic model clarifies how certain resources and activities are deployed to produce the outputs that contribute to some expected outcome. While again it revolves around some stylized representation of an actual situation, this model embraces a sequential mindset in which each element is assumed to be necessary to the following element in producing the outcome. Applied to the rules governing special situations – from international trade to organizational policies – the logic model can unveil inconsistencies, gaps, or implausible assumptions in design. These proposals open two sets of intertwined questions: 1. What are the limits of the different coding strategies in providing some fundamental empirical basis to interpretations and concerns from regulatory theories? Are they exhaustive? Is there any claim or concern that cannot build on at least one of the three? 2. What is the relationship between different codings? Are they incompatible or complementary? Which one, if any, is more 'fundamental'? The workshop calls for papers that can contribute to answering these questions, preferably, but not exclusively, by focusing on: (a) the problems of matching regulation theories and interpretations with rules’ content; (b) the issues encountered in the construction of codebooks and datasets; (c) the limits, if any, of reusing existing data beyond their original scope; (d) the viability of mixing or merging different coding strategies. Contributions are welcome from either young and more established researchers and any disciplinary background