“More punishment does not necessarily lead to less crime,” a news report on the ETH life website declared. According to the report, scientists at ETH Zurich have built a new model to simulate social interactions. The agent-based model is more realistic than previous models because it includes not only criminals and law enforcers, but also honest citizens. The model explains why tougher punishments do not eliminate crimes. Therefore, the focus of combatting crime should be on the socio-economic context. The report is picked up and redistributed by other science news outlets, e.g., the Science Daily. The original version can be found at http://www.ethlife.ethz.ch/archive_articles/131010_kriminalitaet_fb/index_EN.
Before we tear up the Penal code on the account of scientific discovery, it is probably worthwhile to read the actual paper quoted by the news report. The paper, titled “Understanding Recurrent Crime as System-Immanent Collective Behavior,” is published by Matjaz Perc, Karsten Donnay, and Dirk Helbing, and can be found at http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0076063.
The study discussed in the paper carried a much narrower focus than what the news report ostensibly suggested. The study was aimed to explain the empirical evidence that criminal activities often follow a cyclical pattern. The study was based on simulations of a spatial inspection game, which employed a number of variables that represented criminal gain, penalty, inspection cost, inspection reward, and probability of being caught. The simulation showed that different states existed, each of which was dominated by criminals, inspectors, or ordinary people, respectively. As a result, criminal activities are likely to be recurrent than evolving towards an equilibrium. The spatial interaction, i.e., the degree of people interacting with each other, is an important factor in the dynamics of these state transitions.
Unlike what the report suggested, the study did not pronounce the death of criminal punishment system. Although the simulation did show that at transitional points, increasing penalty does not eliminate crimes, this does not mean that penalty has no effects on the criminal activity in general. On the contrary, the study showed that if the penalty is too low, then the temptation for crime will be too high, which leads to a complete breakdown of social order. The inspection cost has a similar effect.
The model used in the study may continue to provide insights on how society works, and whether certain criminal prevention strategy may work better than others. But the study also showed that one has to be careful to draw quick conclusions because the human society is extremely complicated.
So keep your penal code. It probably wouldn’t disappear from the bar exam any time soon.