Neuroethics Journal Club: Judging Mens Rea

The National Core for Neuroethics had a lively journal club discussion recently on a paper by Bertram Malle and Sarah Nelson that dealt with “the tension between folk concepts and legal concepts of intentionality.” As I was presenting the paper and facilitating the discussion, I decided to blog about it to share some of the highlights with our readers and crystallize my own thoughts on the matter, stirred up as they were by the proceedings.

The basic gist of the paper is as follows. Malle and Nelson identify “the valid and precise use of the concepts of mental states in reasoning about the defendant’s actions and in assigning responsibility, blame, and punishment” as a central challenge in creating a system of criminal adjudication. (One interesting point to consider going forward is how these same issues might apply to the context of torts, where instead of the epistemic criterion being “beyond a reasonable doubt” one is instead prompted to consider “the balance of evidence.”) In legal contexts, the term used to refer to the mental states in question is mens rea, Latin for “guilty/sinful mind.”

The specific mental state that the paper is concerned with is intention, especially as it relates to intentional action. In the grand tradition of experimental philosophy (though it really wasn’t yet a tradition in 2003!), Malle and Nelson find the by-now familiar faults with how these concepts have been developed in legal theory and philosophy – with theories of intentional action checked primarily against the intuitions of a small, non-representative group of participants in the debate, leading to a confusing mismatch between how the law asks us to use concepts, and how we (generally) are inclined to actually use them.

The paper explores the possibility of replacing the contorted and awkward legal concept of intentionality (here used as a synonym for intentional action) with a “folk” theory, one which Malle and a then-lesser-known Joshua Knobe had already undertaken to construct via several studies. Malle’s rationale is that, should the folk concept of intentionality prove to be consistent in its use and appropriate to the goals of law, it will be a natural successor to the status quo.

The model of intentional action that the authors supply is an amalgam of three components: skill, awareness, and intention; in turn, an intention is formed when a belief and a desire come together in the right way. A significant portion of our group’s discussion was devoted to scrutinizing this model. We felt that Malle & Nelson’s treatment of the “skill” and “awareness” components left a little to be desired; while it is of course possible to have an intention to do something, but then actually do that thing accidentally, the role of skill and awareness in bridging the gap between intention and acting intentionally seemed to many of us to need more specification. A few discussants noted that “awareness” could be better specified, perhaps as “knowledge that the action one is about to perform will fulfill a particular intention.” I would also suggest that the component of “skill” can itself then be cashed out in terms of knowledge or awareness: if a first-time, unskilled darts player has an intention to hit the bullseye and does indeed hit it, the reason we might be inclined to say she didn’t hit the bullseye intentionally is because she had to settle on one particular set of bodily motions from amongst many choices, and she did not know that her choice would lead to her hitting the bullseye.

An even bigger issue that many discussants pinpointed was Malle and Nelson’s handling of countervailing evidence to their claim that the folk concept of intentional action is “used consistently.” By now, many readers will be familiar with the so-called “Knobe Effect,” referring to an experimental finding by Josh Knobe. The finding is generally taken to have established that people’s judgments as to whether an action was intentional can be made to differ solely by changing the side-effect of the action in question from “helping” to “hurting.” It remains a point of contention exactly what about the change in the side-effect causes the intentionality judgment to shift – is it attributable to implicit expectations? Norm violation? The moral valence of the outcome? But whatever the root cause, it struck me and others as prima facie obvious that this is a way in which the folk concept of intentional action is not used consistently. And from a legal perspective, it is particularly problematic that the variation seems to trace back to a moral or normative distinction: it is at least possible that the most suitable way to describe the phenomenon in question is roughly:

“How do I know that X’s action Y was wrong? Because X did Y intentionally. How do I know that X did Y intentionally? Because X’s action Y was wrong.”

That lovely little circularity, in turn, could easily underlie: “How do I know that X is culpable for Y? Because I have good evidence for attributing mens rea to X when he did Y. But that same evidence wouldn’t move me to attribute mens rea to X if Y were not the sort of action one would generally be culpable for.” Of course, nobody ever actually thinks that, because presumably no juror or judge ever runs that counterfactual. (Preliminary recommendation for legal types: maybe consider doing so!) Now, of course, for the Knobe effect to facilitate circular juridical reasoning, it would have to be the case that people actually use their folk concept of intentionality, rather than the (apparently artificially) self-consistent legal concept, when making such attributions … and the paper’s authors, in fact, present good evidence for this being the case. Namely, their experimental participants answered questions no differently when given a thorough definition of intentional action to work with, relative to those who were given no definition. So supplying even a clear, simple theory (to say nothing of the law’s self-admittedly byzantine concept) appears to be wholly ineffectual.

To our surprise, Malle and Nelson were aware of the Knobe Effect but pithily bracketed it off as a “limitation” of their approach, framing it as an example of “the power of extraneous factors that can bias judgments,” comparable to factors like “time pressure, tiredness, [and] distraction” and ultimately attributable to “negative affect” rather than conceptual inconsistency. This seems to me a bizarre move: even if the inconsistency is attributable to negative affect, it is nonetheless a deep and inextricable feature of the conceptual structure of intention, and cannot just be separated out by fiat! In other words, if it is the case that we cannot mitigate the Knobe Effect by controlling for affect (and to my knowledge, it is not the case that we can), then it can make no difference whether the inconsistency arises due to affective factors or not; the affective factors are themselves effectively intrinsic. And if this is the case, the authors can hardly deal with a direct contradiction of one of their central claims (that the folk concept of intentional action is used consistently) by setting it aside.

For these reasons and others, we felt that Malle and Knobe’s folk model of intentional action was not yet ready for legal “prime time.” Nonetheless, many agreed that the findings presented in the paper amounted to good justification for some of Malle and Nelson’s more critical claims: the law’s tendencies to muddle together intention and intentional action, as well as belief, desire, and intention, are well-documented and convincingly challenged in the paper.

For me, the issues posed for the law by experimental-philosophical work on concepts of knowledge and intention will likely remain a topic of interest and inquiry. I am especially curious as to how people’s concept usage and mental-state attributions may be shaped by an individual’s understanding of the aims of criminal justice (restitution, deterrence, retribution, et cetera), and perhaps for jurists one’s preferred legal philosophy (positivism, natural law, et cetera). I anticipate many fascinating issues to ponder!

Link to the full text of Malle & Nelson’s paper for those with access.

Link to a helpful introductory article on the Knobe effect and experimental philosophy.