Metamoral Criminal Law

For centuries, legal philosophers have been trying to find answers to the question: What conduct should be criminalized? While continental European and German scholars in particular tend to focus on the protection of legal goods, theorists from common law jurisdictions traditionally favor the harm principle. There are numerous differences between and within these and other approaches, yet, their methodology remains the same. Philosophers and lawyers alike heavily draw on their own intuitions. After all, this seemed to be all what an enlightened homo sapiens could rely on in the past. However, recent research in behavioral economics, moral psychology and experimental philosophy suggests that some intuitions may be more reliable than others. In my dissertation, I examine the implications of these findings for theories of criminalization and, ultimately, the criminal law itself.

Among other things, I argue for a new theory of criminalization based on the so-called dual-process theory of learning and decision-making as popularized in Daniel Kahneman’s Thinking Fast and Slow (2011) and applied to moral decision-making at length in Joshua Greene’s Moral Tribes (2014). Dual-process theory states that there are two distinct (but related) processes of thought. These processes which are sometimes referred to as system 1 (S1) and system 2 (S2) thinking essentially have very different functions.[1] While S1 thinking has been linked to a significant number of biases, it is also described as very efficient (due to its low computational costs). Notably, S1 thinking is informed by previous experiences whereas S2 thinking carefully analyzes the environmental surroundings, their causal relations, and only then evaluates which action has the best outcome. Although often assumed, this does not mean that S2 thinking is always better from a normative perspective and all policies and laws should be evaluated accordingly. On the contrary, I argue that humans can often rely on S1 thinking. However, if humans have not made long-lasting valuable experiences, it is likely to be more promising to refer to S2 and evaluate the consequences of the conduct in question.

For instance, intuitions[2] about criminal laws such as murder, manslaughter and assault can draw on experiences of tens of thousands of years and are likely to serve us rather well. Yet, when thinking about modern technologies such as artificial intelligence and gene-editing tools, modern societal structures such as democracies and supranational organizations or modern criminal laws on incest, price-gouging and factory farming, S1 thinking is likely to be insufficiently informed. Although S2 thinking cannot guarantee morally justified outcomes, it may still be the best option available under such conditions - whether the results are intuitively appealing or not.

[1] Naturally, there is quite a bit of disagreement regarding the specific functions, origins and underlying neural mechanisms of thinking. In my work, I adopt a computational theory of cognition as defended by Fiery Cushman (2013), Joshua Greene (2017) and others.

[2] It is important to note that philosophers and psychologists use the word „intuition” differently. Here, I use the psychological definition of referring to an intuition when talking about S1 thinking.