All democratic organisations operate under a particular set of rules. Such procedures are implemented by the very individuals that create and maintain them, usually under a majority voting rule. This research project engages with the question of why and how members of parliament "abdicate" procedural power, focusing on the evolution of the rules of debate in the UK House of Commons. Working from newly collected data on the reform of Standing Orders of the House spanning 205 years (1811 - 2015), as well as records of over six million speeches, it provides a new perspective on procedural choice. Framing debate as a platform for speech-as-filibuster behaviour, I develop a formal model where the decision to support an anti-dilatory reform is primarily a function of polarisation. I show that legislators adopt restrictive rules when they are more likely to share policy preferences with colleagues within their party. The presence of shared views, then, motivates MPs to prioritise responsible use of the common resource of plenary time over individual policy influence. Both empirically and theoretically, my research offers new insights into the process of parliamentary reform in the absence of party discipline, and studies how the dynamics of procedural choice change as political parties enter the stage. Methodologically, it makes a contribution to the text-as-data field, exploring the use of novel machine-learning techniques in the measurement of political preferences.