One issue that becomes clear is that there are substantial costs from using flawed paradigms. One of the articles I need to write is “Three Flawed Criminological Paradigms.” The field of Criminology has been operating under a number of flawed paradigms. Three of the most important are the “general theory” paradigm, the “criminal career” paradigm, and the “career criminal” paradigm. These paradigms ignore “the facts of life” and should be replaced. I will show that the cost of using these flawed paradigms has been in the trillions of dollars.
The general theory paradigm is flawed because the basic premise is that there is a single cause for crime. For example, “The General Theory of Crime” posits that low self control is the main cause of crime. While low self control is a risk factor, it is not the only cause of crime. There are a multitude of factors that lead to variation in crime rates. The “multiple factor” theory, which had been proposed in the early 1900s provides a more accurate paradigm for the study of crime. Sutherland had argued that the multiple factor theory is not much use. However, with the new methods of analysis provided by machine learning and artificial intelligence, a multiple factor approach can be implemented that will help reduce crime.
The criminal career paradigm proposes that there is a criminal trajectory that has a start, middle, and end. A look at risk trajectories of criminal offenders shows that this is an over simplification that is not supported by the data. Criminal careers do not follow any trajectory, except in the broadest sense. Individual trajectories are governed by “high dimensional chaos” and it is unlikely that criminologists will be able to predict which trajectory a person is following. Criminologists should focus on studying individual trajectories rather than searching for sets of life-course trajectories.
The career criminal paradigm has been used to justify the imprisonment of high numbers of people who break the law. This paradigm was based upon an econometric model developed in the 1970s. This model is obviously flawed since it is based on the premise that one can compute an average number of future crimes per offender. The model is taking the average of a nonlinear distribution and no average can come close to estimating the actual number of future crimes. If one examines the problem from a “facts of life” perspective, it becomes clear that the costs of such a model are much higher than any expected benefits.