The Law and Human Rights: Is the Law a Mere Parchment Barrier to Human Rights Abuse?

This study is the first systematic global analysis of the impact of law on human rights, analyzing the impact of twenty-three constitution provisions and an international covenant on three measures of human rights behavior, over the period of 1976-1996. Three sets of constitutional provisions are analyzed, including 1) ten provisions for individual freedoms and due process rights, 2) nine provisions for elements of judicial independence and 3) four provisions that outline procedures for states of emergency. Additionally, the impact of the International Covenant on Civil and Political Rights on actual human rights behavior is analyzed. Each of these areas of law are evaluated individually, in multiple models in which different elements vary. For example, some models control for democracy with different measures, others divide the data into the Cold War and post-Cold War eras, and some test constitutional indices. Finally, all provisions are simultaneously analyzed in integrated models. Provisions for fair and public trials are consistently shown to decrease the probability of abuse. An index of four freedoms (speech, religion, association, and assembly) decreases the probability of abuse somewhat consistently. Three of the provisions for judicial independence are most consistent in reducing the probability of abuse: the provisions for exclusive judicial authority, for the finality of judges' decisions, and banning exceptional courts. Two of four states of emergency provisions decrease abuse as international lawyers have argued: the provisions for legislative declaration of the emergency and the ban against dissolving the legislature during an emergency. However, two of the provisions are shown to hurt human rights practices: the duration and the derogation provisions. The International Covenant on Civil and Political Rights does not demonstrate a statistically significant impact. While the performance of the constitutional provisions is less than legal scholars would hope, their combined impact over time are shown to be quite large, relative to the impacts of other factors shown to affect human rights abuse