A common style of argument in the literature on free will and moral responsibility is the Manipulation Argument. These tend to begin with a case of an agent in a deterministic universe who is manipulated, say, via brain surgery, into performing some action. Intuitively, this agent is not responsible for that action. Yet, since there is no relevant difference, with respect to whether an agent is responsible, between the manipulated agent and a typical agent in a deterministic universe, responsibility is (...) not compatible with the truth of determinism. In response, some theorists have argued that there is a relevant difference, and have developed two sorts of accounts of that difference: bypassing views, and manipulator-focused views. Manipulator-focused views suggest that the difference concerns the presence of a manipulator, whereas bypassing views suggest that the relevant difference concerns the fact that the action issues from attitudes that the manipulated agent acquired in a way that bypassed her capacities for control over her mental life. One sort of case used to decide between these sorts of accounts is a case of machine induction, which is just like a manipulation case, yet the change in the agent is the result of some natural force. Against the received view, Xiaofei Liu argues that such cases pose problems for bypassing views, and favor manipulator-focused views instead. This paper addresses Liu’s arguments, as well as a variety of cases, concluding that cases of machine induction do not provide motivation for a bypassing theorist to adopt a manipulator-focused view. (shrink)

We propose four policies of ethical design of human-grade Artificial Intelligence. Two of our policies are precautionary. Given substantial uncertainty both about ethical theory and about the conditions under which AI would have conscious experiences, we should be cautious in our handling of cases where different moral theories or different theories of consciousness would produce very different ethical recommendations. Two of our policies concern respect and freedom. If we design AI that deserves moral consideration equivalent to that of human beings, (...) that AI should be designed with self-respect and with the freedom to explore values other than those we might impose. We are especially concerned about the temptation to create human-grade AI pre-installed with the desire to cheerfully sacrifice itself for its creators’ benefit. (shrink)

As neuroscience progresses, we will not only gain a better understanding of how our brains work, but also a better understanding of how to modify them, and as a result, our mental states. An important question we are faced with is whether the state could be justified in implementing such methods on criminal offenders, without their consent, for the purposes of rehabilitation and reduction of recidivism; a practice that is already legal in some jurisdictions. By focusing on a prominent type (...) of view of free action, which I call bypassing views, this paper evaluates how such interventions may negatively impact the freedom of their subjects. The paper concludes that there will be a tension between the goals of rehabilitation and reduction of recidivism, on the one hand, and the negative impact such interventions may have on free action, on the other. Other things equal, the better that a particular intervention is at achieving the former, the more likely it is to result in the latter. (shrink)

I argue that considerations pertaining to constitutive luck undermine historicism—the view that an agent’s history can determine whether or not she is morally responsible. The main way that historicists have motivated their view is by appealing to certain cases of manipulation. I argue, however, that since agents can be morally responsible for performing some actions from characters with respect to which they are entirely constitutively lucky, and since there is no relevant difference between these agents and agents who have been (...) manipulated into acting from a character bestowed upon them by their manipulators, we should give up historicism. After presenting this argument and defending it against some potential objections, I briefly criticize the standard structuralist alternative and propose a new structuralist position that is shaped by reflection on constitutive luck. (shrink)

We study whether robots can satisfy the conditions of an agent fit to be held morally responsible, with a focus on autonomy and self-control. An analogy between robots and human groups enables us to modify arguments concerning collective responsibility for studying questions of robot responsibility. We employ Mele’s history-sensitive account of autonomy and responsibility to argue that even if robots were to have all the capacities required of moral agency, their history would deprive them from autonomy in a responsibility-undermining way. (...) We will also study whether humans and technological artifacts like robots can form hybrid collective agents that could be morally responsible for their actions and give an argument against such a possibility. (shrink)

Given the unwittingness of negligence, personal responsibility for negligent conduct is puzzling. After all, how is it that one is responsible for what one did not intend to do or was unaware that one was doing? How, therefore, is one’s agency involved with one’s negligence so as to ground one’s responsibility for it? Negligence is an unwitting failure in agency to meet a standard requiring conduct that falls within one’s competency. Accordingly, negligent conduct involves agency in that negligence is a (...) manifestation of agency failure. Now, nobody’s perfect. Human agency is innately fallible, and a measure of agency failure is, therefore, unavoidable. The more one’s negligence manifests failure in one’s agency as an individual, the more one is responsible for it. In contrast, the more one’s negligence involves the shortcomings innate to all human agency the less responsible one becomes, because one’s agency as an individual is less and less involved in one’s failure. Determinative of the measure of individual and of human failings mixed into an instance of negligent phi-ing is the background quality of one’s agency at meeting one’s competency at phi-ing. That is, how able one is at delivering on what one is able to competently do. The more able, the less one’s occasional instances of negligence involve manifestations of failures of one’s agency as an individual – nobody’s perfect – and are more manifestations of one’s agency’s innate human fallibility, making one less and less responsible for one’s negligence. (shrink)

In this paper, I propose and defend a structural ownership condition on moral responsibility. According to the condition I propose, an agent owns a mental item if and only if it is part of or is partly grounded by a coherent set of psychological states. As I discuss, other theorists have proposed or alluded to conditions like psychological coherence, but each proposal is unsatisfactory in some way. My account appeals to narrative explanation to elucidate the relevant sense of psychological coherence.

This article examines the nonhistoricist higher-order compatibilist theory of moral responsibility devised and defended by Harry G. Frankfurt. Intuitions about certain kinds of cases of moral responsibility cast significant doubt on the wide irrelevancy clause of the nonhistoricist feature of Frankfurt’s theory. It will be argued that, while the questions of the nature and ascription of moral responsibility must be separated in doing moral responsibility theory, the questions of whether or not and the extent to which an agent is morally (...) responsible for performing certain kinds of actions cannot be settled without also answering the question of the history of that agent’s attitudes, preferences, etc. which may have contributed to the agent’s performing such actions. Thus Frankfurt’s repeated claim that the history of an agent’s action is “irrelevant” to the question of that agent’s moral responsibility for it is rendered dubious, thereby rendering problematic the strongly nonhistoricist component of his theory. The methodological difference between my challenge to Frankfurt’s nonhistoricism and those of others is that, while others have challenged Frankfurt’s nonhistoriciams by way of thought experiments involving hyper-manipulation, mine presents a set of counter-examples to his nonhistoricism which emanate from morality and the law. Indeed, if my counter-example is plausible, then Frankfurt’s nonhistoricism cannot make good sense of some standard instances of racism. Furthermore, while many believe that racist hate crimes ought to be punished more heavily than crimes committed absent hate, the argument herein provides reasons for mitigating, rather than increasing, the responsibility and punishments of many such crimes. What matters in such cases is the extent to which the hatred accompanying said crimes has a history in the agent’s life, one which makes it significantly difficult for her to do other than what she did in committing the racist crime. (shrink)

In discussions on moral responsibility for actions, a commonly discussed case is one in which an agent is manipulated into performing some action. On some views, such agents lack responsibility for those actions partly because they issue from attitudes that were acquired in an inappropriate way. In this paper, it is argued that such views are in need of revision. After introducing a new problematic case of a manipulated agent, revisions are offered for specific views. The paper concludes with a (...) discussion of the views in a broader context, as well as some potential implications of the revisions. (shrink)

In this paper I explore various themes in Alfred Mele's Manipulated Agents: A Window to Moral Responsibility (OUP 2019). I develop four points. First, I argue that Mele's historical requirement for moral responsibility for developed morally responsible agents should be coupled with a nonhistorical theory of initially developing agents (like toddlers). Second, I argue that one might resist Mele's negative historical requirement (wherein agents must lack certain responsibility-defeating histories) with a minimal positive historical requirement according to which an agent has (...) a history wherein she did not undergo any responsibility-defeating events, like being severely manipulated.Third, I also explore the idea that one who defended a nonhistorical view, such as Harry Frankfurt's, might rely on a different conception of what moral responsibility is. This might explain why some resisting Mele would not have the intuition that in certain cases a manipulated agent is not responsible. Finally, I question how we should think of the role of intuition in thought experiments figuring centrally in Mele's work. (shrink)