Abstract The criminal process can be seen as a system of classification, akin to a search engine. I consider the impact on it of poor measurements of the rates of false positives and false negatives. With ordinal rather than interval measurements, we lose some but not all control. I identify a novel form of indeterminacy, ‘directional aimlessness’, that leads to a behaviour of ‘shoving’. This can take a form, ‘naked shoving’, that looks purposeful but isn’t. I empirically establish the presence of naked shoving in the USA. This methodological work comprises Part I of a novel theory of the criminal process. In Part II, I use it to mathematize Herbert Packer’s influential ‘two models’ theory in criminology, both a form of validation and a way of giving an abstract theory traction in the real world. We acquire a general ‘epistemic’ alternative to economic, organizational, legal or political forms of analysis and I use it to critique Richard Posner’s law and economics. Given the lack of purpose, I conclude that policymaking in the criminal process is usually a form of ritual and, now that we have the capacity for interval measurements, contemplate the end of the old order.
Read full abstract