In this chapter I present a first-order explanation for the outcomes described in the previous one. In Chapter 1, I argued that to effectively protect a given right, a system must consistently apply the “right” rule to the “right” facts. In the course of interviews and conversations about this project, I often heard people attribute impunity for the police to the routine application by judges of an informal rule that simply gives the police a free hand in executing socially marginalized people (CELS 2004 is one example of this). If this diagnosis is correct, then the first-order explanation for low conviction rates is a repeated normative failure – judges who consistently fail to follow the law. The alternative first-order explanation is that judges are applying the right rule, but the system repeatedly produces an insufficient factual record – what I have called an informational failure. If this second hypothesis is true, then the blame for failure rests primarily with those charged with producing the “right” facts: the investigative police and litigants, including prosecutors, rather than judges. The solution would then be not more legal training for judges, but a more effective and more independent investigative force.

Throughout this book I argue that each system produces its particular pattern of effectiveness and inequality owing to a combination of these two failures. The legal system in Salvador da Bahia fails predominantly because it consistently applies a rule of impunity, producing both low effectiveness and (relatively) low inequality.