Criminal liability for acts committed by AI systems has recently become a hot legal topic. This paper includes three different contributions. The first contribution is an analysis of the extent to which an AI system can satisfy the requirements for criminal liability: accomplishing an actus reus, having the corresponding mens rea, possessing the cognitive capacities needed for responsibility. The second contribution is a discussion of criminal activity accomplished by an AI entity, with reference to a recent case involving an online (...) bot, the Random Darknet Shopper. This discussion will provide the context for the analysis of commonalities and differences between criminal activities by humans and by artificial systems. The third contribution concerns the evaluation of different ways of addressing criminal activities by AI systems in a regulatory perspective. (shrink)

According to many criteria, agency, intentionality, responsibility and freedom of decision, require conscious decisions. Freud already assumed that many of our decisions are influenced by dynamically unconscious motives or that we even perform unconscious actions based on completely unconscious considerations. Such actions might not be intentional, and perhaps not even actions in the narrow sense, we would not be responsible for them and freedom of decision would be missing. Recent psychological and neurophysiological research has added to this a number of (...) phenomena (the "new unconscious") in which behavior is completely unconscious or in which the decision or its execution is influenced by unconscious factors: priming, automatic behavior, habitualized behavior, actions based on plain unconscious deliberations, intrusion of information from the dorsal pathway, etc. However, since this makes up the largest part of the behavior which is generally regarded as action, intentionality, yet agency, responsibility and even compatibilist freedom of decision for the largest part of our behavior may be threatened. Such considerations have led to a lively debate, which, however, suffers from generalizations that lump all these unconscious phenomena together. In contrast, the aim of this article is to discuss individual unconscious influences on our behavior separately with respect to what extent they require changes in traditional conceptualizations. The first part (2-4) of the article outlines the "traditions" and their elaborations: the intentional causalist concept of action, an associated empirical theory of action and standard concepts of responsibility and compatibilist freedom of decision, as well as the challenges for them. In the second part (5-9), the aforementioned unconscious influences on our actions (except for automated and habitualized actions, which I discuss elsewhere) are examined: 1. unconscious priming, 2. dynamically unconscious motives, 3. dorsal pathway information influencing conscious decisions, 4. unconsciously altered execution of conscious intentions, 5. unconscious deliberations and decisions. To what extent do these phenomena C1. require a change in the concept of action, C2. curtail intentionality or agency, C3. responsibility and C4. freedom? The result is: The curtailments prove to be far less dramatic than they initially appear; they require more watchfulness but no conceptual change. (shrink)

Supervenient libertarianism maintains that indeterminism may exist at a supervening agency level, consistent with determinism at a subvening physical level. It seems as if this approach has the potential to break the longstanding deadlock in the free will debate, since it concedes to the traditional incompatibilist that agents can only do otherwise if they can do so in their actual circumstances, holding the past and the laws constant, while nonetheless arguing that this ability is compatible with physical determinism. However, we (...) argue that supervenient libertarianism faces some serious problems, and that it fails to break us free from this deadlock within the free will debate. (shrink)

Two components of lay observers’ calculus of moral judgment are proximal intent (the actor’s mind is focused on performing the action) and distal intent (the actor’s mind is focused on the broader goal). What causes observers to prioritize one form of intent over the other? The authors observed whether construal level (Studies 1-2) and beliefs about free will (Studies 3-4) would influence participants’ sensitivity to the actor’s proximal versus distal intent. In four studies, participants read scenarios in which the actor’s (...) proximal and distal intent were independently manipulated. In Study 1, when only distal intent was present in the actor’s mind, participants rated the psychologically distant actor more responsible than the psychologically near actor. In Study 2, when only distal intent was in the actor’s mind, participants with a chronic high level of action identification rated the actor more responsible than did those with a low level of action identification. In both studies, when only proximal intent was in the actor’s mind, construal level did not predict judgments of responsibility. In Study 3, when only proximal intent was present in the actor’s mind, the more participants believed in free will, the more they rated the actor responsible. When only distal intent was in the actor’s mind, free will belief did not influence ratings of responsibility. In Study 4, the same pattern emerged when free will / determinism beliefs were manipulated and the actor performed a positive (life-saving) act. The authors discuss how these results shed new light on the literatures on moral reasoning and psycho-legal theory. (shrink)

‘Actus non facit reum nisi mens sit rea.’ But when is a mens rea? In the last twenty years discussions of this question have been stimulated by controversial decisions in the English House of Lords in the cases of Smith, Morgan and Majewski.The case of Smith decided that a man might be guilty of murder if a reasonable person, knowing the circumstances, would have foreseen that death would result even if the agent himself did not so foresee. This appeal to (...) the ‘reasonable person’ imports what is known as the ‘objective test’ as opposed to the ‘subjective test’ of the agent's actual foresight or knowledge. The case of Morgan inclined per contra to the subjective test. (shrink)

ABSTRACT Not all crimes require mens rea, but all serious ones do. Crudely the requirement is that the defendant be able to take responsibility for the actus reus of which he is accused. What must be implied by this is essentially that the agent retain control of his actions. It is unjust to punish actions which are outside of the agent's control since such punishment cannot deter and is, arguably, pointless. Duress does not remove an agent's control of his actions. (...) This does not mean that actions performed under duress should be viewed as fully intentional. For though the agent foresees and wishes for the result of his action performed under duress, it is wished for as the best option out of a range of alternatives manipulated by the one who makes the threats which constitute the duress. The agent is responsible for his choice, but not for what is available for him to choose. He has an obligation to do what is best in the circumstances and should not be blamed if that would, in other circumstances, be accounted criminal. (shrink)

This article discusses a challenge to the traditional intentional-causalist conceptions of action and intentionality as well as to our everyday and legal conceptions of responsibility, namely the psychological discovery that the greatest part of our alleged actions are performed automatically, that is unconsciously and without a proximal intention causing and sustaining them. The main part of the article scrutinizes several mechanisms of automatic behavior, how they work, and whether the resulting behavior is an action. These mechanisms include actions caused by (...) distal implementation intentions, four types of habit and habitualization, mimicry, and semantically induced automatic behavior. According to the intentional-causalist criterion, the automatic behaviors resulting from all but one of these mechanisms turn out to be actions and to be intentional; and even the behavior resulting from the remaining mechanism is something we can be responsible for. Hence, the challenge, seen from close up, does not really call the traditional conception of action and intentionality into question. (shrink)

The most prominent argument for the incompatibility of free will and determinism is Peter van Inwagen’s consequence argument. I offer a new diagnosis of what is wrong with this argument. Proponents and critics typically accept the way the argument is framed, and only disagree on whether the premisses and rules of inference are true. I suggest that the argument involves a category mistake: it conflates two different levels of description, namely, the physical level at which we describe the world from (...) the perspective of fundamental physics and the agential level at which we describe agents and their actions. My diagnosis is based on an account of free will as a higher-level phenomenon. I call this account ‘compatibilist libertarianism’, for reasons that will become clear. Although the paper addresses a primarily metaphysical question, it uses tools similar to those employed in philosophy-of-science work on determinism and indeterminism, higher-level phenomena, and dynamical systems. (shrink)

Frequently, the just war principle of noncombatant immunity is interpreted as morally prohibiting the intentional targeting of noncombatants. Apparently, many just war theorists assume that to target means to (intend to) kill. Now that effective nonlethal weapons have been envisaged, it should be evident that there is no conceptual connection between intentionally targeting and intentionally killing. For, using nonlethal weapons, there could be intentional targeting without intentional killing. This paper explores the question of whether the noncombatant immunity principle should be (...) revised, so as to allow uses of nonlethal weapons. Preliminary to answering this question, some other questions are explored, among which are the following. Why should a noncombatant immunity principle be accepted? Why is it morally permissible to intentionally target enemy combatants? Are noncombatants grievously harmed when they are incapacitated by nonlethal weapons? Is it morally permissible to intentionally incapacitate enemy combatants with nonlethal weapons, while knowingly but not intentionally incapacitating noncombatants? In order to focus on moral questions involving nonlethal weapons, questions about their effectiveness or legality are set aside. Instead of the idea of noncombatant immunity as expressed above, a delimited principle of noncombatant immunity is proposed—namely, that, in the conduct of war, the intentional grave injuring or killing of noncombatants is morally prohibited. Also proposed is a principle of noncombatant targeting, which would allow some uses of nonlethal weapons to intentionally incapacitate noncombatants. (shrink)

The demand for 'criteria of correctness' to identify recurring particulars in Wittgenstein's private language argument favors an idealist interpretation of quantum phenomena.The indeterminacy principle in quantum physics and the logic of the private language argument share a common concern with the limitations by which microphysical or sensation particulars can be reidentified. Wittgenstein's criteria for reidentifying particular recurrent private sensations are so general as to apply with equal force to quantum particulars, and to support the idealist thesis that quantum phenomena are (...) themselves essentially mental or dependent on mental occurrences. (shrink)

A “contrastive explanation” explains not only why some event A occurred, but why A occurred as opposed to some alternative event B. Some philosophers argue that agents could only be morally responsible for their choices if those choices have contrastive explanations, since they would otherwise be “luck infested”. Assuming that contrastive explanations cannot be offered for causally undetermined events, this requirement entails that no one could be held responsible for a causally undetermined choice. Such arguments challenge incompatibilism, since they entail (...) that causal determinism is a prerequisite for moral responsibility. However, I argue that for a significant class of choices, even if we stipulate that they are determined, we will be unable to provide a relevant contrastive explanation. Hence causal determinism is no remedy for luck infestation, and compatibilists do not fare much better than incompatibilists in the face of this requirement. This should serve to weaken its philosophical appeal. (shrink)