No public, no power? Analyzing the importance of public support for constitutional review with novel data and machine learning methods

Sternberg, Sebastian

Vorschau

PDF
Dissertation_Sebastian_Sternberg.pdf - Veröffentlichte Version
Download (4MB)

URL:	https://madoc.bib.uni-mannheim.de/52329
URN:	urn:nbn:de:bsz:180-madoc-523294
Dokumenttyp:	Dissertation
Erscheinungsjahr:	2019
Ort der Veröffentlichung:	Mannheim
Hochschule:	Universität Mannheim
Gutachter:	Gschwend, Thomas
Datum der mündl. Prüfung:	17 Juli 2019
Sprache der Veröffentlichung:	Englisch
Einrichtung:	Fakultät für Sozialwissenschaften > Politische Wissenschaft, Quantitative Sozialwissenschaftliche Methoden (Gschwend 2007-) Außerfakultäre Einrichtungen > GESS - CDSS (SOWI)
Fachgebiet:	320 Politik
Freie Schlagwörter (Englisch):	constitutional courts , public opinion , legislative noncompliance , machine learning
Abstract:	Constitutional review is a central feature of liberal democracy. However, with neither the power of the purse nor the sword, the mere presence of constitutional courts does not automatically imply the effective exercise of judicial authority. Courts must rely on elected officials for the implementation of their rulings. The ability of a court to ensure that government officials faithfully comply with judicial decisions critically depends on the existence of sufficient public support for the court and the public’s ability to monitor legislative responses to judicial decisions. In this dissertation, I study the importance of public support for the relationship between court-government and court-public. I draw on the judicial politics literature on separation of powers, public support and legislative noncompliance and extend existing theory in two regards. First, I argue that not all courts possess the sufficient level of public support necessary to ensure legislative compliance. Varying degrees of public support strongly affect the leverage that courts possess in judicial-legislative and judicial-public interactions. Second, I argue that courts actively take measures in the form of the institutional tools at their disposal when they expect legislative noncompliance. One such tool is decision language, whose strategic usage allows judges to pressure the government or hide likely noncompliance from public view, if necessary. I test these arguments empirically by combining classical inferential methods such as survey experiments with novel data on court decision-making and methodologies from the field of machine learning and computational linguistic. Throughout all chapters, I employ a comparative perspective and test my arguments using data on the German Federal Constitutional Court, a court with strong and robust levels of public support, and the less popular French Conseil Constitutionnel. My empirical evidence shows that considering varying degrees of public support and the institutional tools of judges indeed helps to generate a more accurate picture of how judges behave in judicial-legislative and judicial-public interactions. Three conclusions are drawn. First, court decisions can legitimize public policies, albeit only if the court itself is perceived as a legitimate institution. Second, courts are more attentive to the political environment of a decision than previously thought: depending on their degree of public support, they actively adapt the language of their decisions as a function of the risk of noncompliance and their institutional support. Third, public support and other political context factors are important for judicial decision-making not only from an inferential but also from a predictive perspective. The results of my analyses confirm that public support plays a crucial role for courts’ ability to effectively exercise constitutional review, as well as highlighting the benefits of increased differentiation of constitutional courts institutional tools and their diffuse support from a comparative view. Therefore, my results have implications for the growing literature on strategic courts using their institutional tools to address potential noncompliance and the general awareness of judges for their institutional reputation. Overall, this project offers new perspectives on the most important resource of judges – their public support – and has important implications not only for research on judicial politics but also for the efficacy of constitutional review in a constitutional state, and thus the sustainability of liberal democracy.