Thomas M. Powers

Sociotechnical explanations can contribute to individual failure investigations and even normative critiques of an entire system, in the event that clear descriptions of causes and conditions of failure can be specified. But in many failed systems, what has happened is complicated. At present, sociotechnical systems approaches in organizational studies, engineering, and philosophy do not meet standards of clarity that are required of explanation. Further, sociotechnical system approaches so far have not been used to predict instances of failure, though it would be highly desirable if we could develop such predictions. Complications with explanation and the dim prospects for prediction lead us in the direction of new computational technologies, but these are not without risks. Progress in understanding sociotechnical systems means, in effect, that explanations and predictions would become useful for “unenhanced” human understanding, not just for computational models of sociotechnical behavior. One byproduct of this better understanding could be guidance on when to hold certain parts of AI sociotechnical systems accountable—namely, when the algorithms by which they operate could be held accountable for damages that they cause.

This book features papers from CEPE-IACAP 2015, a joint international conference focused on the philosophy of computing. Inside, readers will discover essays that explore current issues in epistemology, philosophy of mind, logic, and philosophy of science from the lens of computation. Coverage also examines applied issues related to ethical, social, and political interest. The contributors first explore how computation has changed philosophical inquiry. Computers are now capable of joining humans in exploring foundational issues. Thus, we can ponder machine-generated explanation, thought, agency, and other quite fascinating concepts. The papers are also concerned with normative aspects of the computer and information technology revolution. They examine technology-specific analyses of key challenges, from Big Data to autonomous robots to expert systems for infrastructure control and financial services. The virtue of a collection that ranges over philosophical questions, such as this one does, lies in the prospects for a more integrated understanding of issues. These are early days in the partnership between philosophy and information technology. Philosophers and researchers are still sorting out many foundational issues. They will need to deploy all of the tools of philosophy to establish this foundation. This volume admirably showcases those tools in the hands of some excellent scholars.

This volume consists of selected papers from the 2015 joint international conference—the first-ever meeting of the Computer Ethics-Philosophical Enquiry conference series of the International Society for Ethics and Information Technology, and the International Association for Computing and Philosophy—held at the University of Delaware from June 22–25 of 2015. The organizing themes of the conference are well represented in the volume. They include theoretical topics at the intersection of computing and philosophy, including essays that explore current issues in epistemology, philosophy of mind, logic, and philosophy of science, and also normative topics on matters of ethical, social, economic, and political import. All of the essays provide views of their subject matter through the lens of computation.

This project has two connected goals: to analyze formal and non-formal aspects of Kant's ethics and thereby to offer a new understanding of his contributions to ethical theory. Philosophers generally have viewed Kant's ethical theory as the paradigm for formalism in ethics, exemplified in his universalizability test for maxims. Most of the criticisms of Kant turn on this standard interpretation. The formalist interpretation of Kant is neither entirely accurate nor charitable, given his ambivalence in the pre-Critical and Critical works and the logical difficulties inherent in universalization. Moreover, the formalist interpretation neglects Kant's subtle and compelling non-formal theory of value, for which I argue. ;The first chapter is an exegesis of the relevant texts: the pre-Critical essays, the first Critique, the Groundwork, second Critique, and Metaphysics of Morals. Kant's early works reveal both formalist and phenomenalist tendencies. This ambivalence is confronted, but not resolved, in the first two formulations of the Categorical Imperative: the formula of the Universal Law and the formula of the End-in-itself. Kant's later writings on ethics only bring tension to the sustained ambivalence of his views. Many modern commentators have wanted to resolve this tension in favor of the formalist interpretation. The second chapter considers these formalist reconstructions of Kant's views in the German- and English-language literature. The third chapter explores the possibilities for a formal theory of willing based on Kantian universalization. I suggest formal models for both monotonic and non-monotonic universalization mechanisms and criticize them in turn. The fourth chapter addresses objections to Kant based on Max Scheler's Formalismus in der Ethik. We focus on Scheler's attacks concerning Kant's moral psychology, value-empiricism, and the notion of Gesinnung. My last chapter presents a resolution of these problems and suggests a revised ethic based on the Kantian values of impartiality, freedom, and dignity, now understood in a non-formal sense. I conclude by showing that an immanent Kantian ethic obviates the need for Kant's formalistic answer to the problem of freedom in the first two Critiques.

A rule-based ethical theory is a good candidate for the practical reasoning of machine ethics because it generates duties or rules for action, and rules are computationally tractable. Among principle- or rule-based theories, the first formulation of Kant's categorical imperative offers a formalizable procedure. We explore a version of machine ethics along the lines of Kantian formalist ethics, both to suggest what computational structures such a view would require and to see what challenges remain for its successful implementation. In reformulating Kant for the purposes of machine ethics, we consider three views of how the categorical imperative works: mere consistency, commonsense practical reasoning, and coherency. The first view envisions straightforward deductions of actions from facts. The second view incorporates recent work in nonmonotonic logic and commonsense reasoning. The last view takes ethical deliberation to follow a logic similar to that of belief revision.

Approaches to programming ethical behavior for computer systems face challenges that are both technical and philosophical in nature. In response, an incrementalist account of machine ethics is developed: a successive adaptation of programmed constraints to new, morally relevant abilities in computers. This approach allows progress under conditions of limited knowledge in both ethics and computer systems engineering and suggests reasons that we can circumvent broader philosophical questions about computer intelligence and autonomy.

The moral permissibility of organ transplantation is taken for granted by most biomedical ethicists and practitioners. Of contemporary concern is not whether, but by what arrangements, we ought to allow organ transplantation. Should we institute markets for organs, thereby increasing their availability and saving many lives? Should organs be sold to the highest bidder? Should we allow the post mortem taking of organs without prior consent? Among moral theorists, the Kantians are suspected of being the least enthusiastic with respect to these and similar questions. I will show the elements in Kant's theory that account for this lack of enthusiasm and how contemporary Kantians might answer questions about the permissibility of various arrangements for organ transplantation. I will argue that Kant would have had a permissive position on organ transplantation post mortem and a restrictive position on organ markets. These results will be based on a broader Kantian view of obligations to the body which stem from a non-formal theory of value. This non-formal theory, I will argue, is at the foundation of Kantian ethics.

Beginning with the well-knowncyber-rape in LambdaMOO, I argue that it ispossible to have real moral wrongs in virtualcommunities. I then generalize the account toshow how it applies to interactions in gamingand discussion communities. My account issupported by a view of moral realism thatacknowledges entities like intentions andcausal properties of actions. Austin's speechact theory is used to show that real people canact in virtual communities in ways that bothestablish practices and moral expectations, andwarrant strong identifications betweenthemselves and their online identities. Rawls'conception of a social practice is used toanalyze the nature of the wrong and thestage-setting aspect of engaging in a practice

Rule-based ethical theories like Kant's appear to be promising for machine ethics because of the computational structure of their judgments. On one formalist interpretation of Kant's categorical imperative, for instance, a machine could place prospective actions into the traditional deontic categories (forbidden, permissible, obligatory) by a simple consistency test on the maxim of action. We might enhance this test by adding a declarative set of subsidiary maxims and other "buttressing" rules. The ethical judgment is then an outcome of the consistency test. While this kind of test can generate results, it may be vacuous in the sense that it would do no more than forbid obviously contradictory maxims of action. It is also possible that the kind of inference in such a rule-based system may be non-monotonic. I discuss these challenges to a rule-based machine ethics, starting from the framework of Kantian ethics.

Can computer systems ever be considered moral agents? This paper considers two factors that are explored in the recent philosophical literature. First, there are the important domains in which computers are allowed to act, made possible by their greater functional capacities. Second, there is the claim that these functional capacities appear to embody relevant human abilities, such as autonomy and responsibility. I argue that neither the first (Domain-Function) factor nor the second (Simulacrum) factor gets at the central issue in the case for computer moral agency: whether they can have the kinds of intentional states that cause their decisions and actions. I give an account that builds on traditional action theory and allows us to conceive of computers as genuine moral agents in virtue of their own causally efficacious intentional states. These states can cause harm or benefit to moral patients, but do not depend on computer consciousness or intelligence

In this paper, we focus attention on the role of computer system complexity in ascribing responsibility. We begin by introducing the notion of technological moral action (TMA). TMA is carried out by the combination of a computer system user, a system designer (developers, programmers, and testers), and a computer system (hardware and software). We discuss three sometimes overlapping types of responsibility: causal responsibility, moral responsibility, and role responsibility. Our analysis is informed by the well-known accounts provided by Hart and Hart and Honoré. While these accounts are helpful, they have misled philosophers and others by presupposing that responsibility can be ascribed in all cases of action simply by paying attention to the free and intended actions of human beings. Such accounts neglect the part played by technology in ascriptions of responsibility in cases of moral action with technology. For both moral and role responsibility, we argue that ascriptions of both causal and role responsibility depend on seeing action as complex in the sense described by TMA. We conclude by showing how our analysis enriches moral discourse about responsibility for TMA.