Moral situations commonly involve agents and patients. Let us define the class A of moral agents as the class of all entities that can in principle qualify as sources of moral action, and the class P of moral patients as the class of all entities that can in principle qualify as receivers of moral action. There can be five logical relations between A and P. Three are unrealistic and are not consider them here.   The remaining two are: (1) A and P can be equal, or (2) A can be a proper subset of P.

Alternative (1) maintains that all entities that qualify as moral agents also qualify as moral patients and vice versa. It corresponds to a rather intuitive position, according to which the agent/inquirer plays the rôle of the moral protagonist, and is one of the most popular views in the history of ethics, shared for example by many Christian Ethicists in general and by Kant in particular. One may refer to is as the standard position.

Alternative (2) holds that all entities that qualify as moral agents also qualify as moral patients but not vice versa. Many entities, most notably animals, seem to qualify as moral patients, even if they are in principle excluded from playing the rôle of moral agents. This post-environmentalist approach requires a change in perspective, from agent orientation to patient orientation. In view of the previous label, one may refer to it as non-standard.

In recent years, non-standard ethical theories have been discussing the scope of P quite extensively.   Comparatively little work has been done in reconsidering the nature of moral agenthood and hence the extension of A. Post-environmentalist thought, in striving for a fully naturalised ethics, has implicitly rejected the relevance, if not the possibility, of supernatural agents, while the plausibility and importance of other types of moral agenthood seem to have been largely disregarded. Secularism has contracted (some would say deflated) A, while environmentalism has justifiably expanded only P, so the gap between A and P has been widening; this has been accompanied by an enormous increase in the moral responsibility of the individual.

Some efforts have been made to redress this situation. In particular, the concept of ‘moral agent’ has been expanded to include both natural and legal persons. A has then been extended to include agents like partnerships, governments or corporations, for whom legal rights and duties have been recognised. This more ecumenical approach has restored balance between A and P. A company can now be held directly accountable for what happens to the environment, for example. Yet the approach has remained unduly constrained by its anthropocentric conception of agenthood. An entity is considered a moral agent only if (i) it is an individual agent and (ii) it is human-based, in the sense that it is either human or at least reducible to an identifiable aggregation of human beings, who remain the only morally responsible sources of action, like ghosts in the legal machine.

Limiting the ethical discourse to individual agents hinders the development of a satisfactory investigation of distributed morality, a macroscopic and growing phenomenon of global moral actions and collective responsibilities resulting from the ‘invisible hand’ of systemic interactions among several agents at a local level. Insisting on the necessarily human-based nature of the agent means undermining the possibility of understanding another major transformation in the ethical field, the appearance of artificial agents (AAs) that are sufficiently informed, ‘smart’, autonomous and able to perform morally relevant actions independently of the humans who created them, causing ‘artificial good’ and ‘artificial evil’ ( Gips [1995] ). AA may play an important role in the dynamics of distributed morality. Both constraints can be eliminated by fully revising the concept of ‘moral agent’.

In Floridi and Sanders [2004] is was shown that AAs are legitimate sources of im/moral actions. It was therefore argued that A should be extended so as to include AAs, that the ethical discourse should include the analysis of their morality and, finally, that this analysis is essential in order to understand a range of new moral problems not only in Computer Ethics but also in ethics in general, especially in the case of distributed morality. The latter topic is now investigated in this new paper.

The phenomenon of distributed knowledge is well-known in epistemic logic. However, distributed morality in multiagent systems, a similar phenomenon in ethics, has been largely neglected so far. The paper is divided into three parts, each focusing on the logic, the genesis and the implications of distributed morality. More specifically:

a) in the first part, an explanation of what distributed morality is, as a feature of moral agency, is analysed. Here the central topic is the logic of distributed morality.

b) in the second part, the problem of how distributed morality can arise in multiagent systems comprising also non-human individual is investigated; and

c) in the third and concluding part, the implications of the occurrence of distributed morality in our information-based society are explored.

References

Floridi, L. 1999, "Information Ethics: On the Philosophical Foundations of Computer Ethics", Ethics and Information Technology, 1(1), 37-56.

Floridi, L. 2003, "On the Intrinsic Value of Information Objects and the Infosphere", Ethics and Information Technology, 4(4), 287-304.

Floridi, L., and Sanders, J. W. 2001 "Artificial Evil and the Foundation of Computer Ethics", Ethics and Information Technology, 3(1), 55-66.

Floridi, L., and Sanders, J. W. 2004, "On the Morality of Artificial Agents", Minds and Machines, 14(3), 349-379.

Gips, J. 1995, "Towards the Ethical Robot" in Android Epistemology, edition, edited by C. Glymour and P. Hayes K. Ford (Cambridge MA: MIT Press), 243-252.

It is possible, but utterly unrealistic, that A and P are disjoint. On the other hand, P can be a proper subset of A or A and P can intersect each other. These two alternatives are little more promising because they both require at least one moral agent that in principle could not qualify as a moral patient. Now this pure agent would be some sort of supernatural entity that, like Aristotle’s God, affects the world but can never be affected by it. But being in principle ‘unaffectable’ and irrelevant, it is unclear what kind of rôle this entity would exercise with respect to the normative guidance of human actions. So it is not surprising that most macroethics have kept away from these ‘supernatural’ speculations and implicitly adopted or even explicitly argued for one of the two remaining alternatives discussed in the text.

Environmental ethics has developed since the 1960s as the study of the moral relationships of human beings to the environment (including its nonhuman contents and inhabitants) and its (possible) values and moral status. It often represents a challenge to anthropocentric approaches embedded in traditional western ethical thinking. For an article-length introduction to environmental ethics see A. Brennan and Y-S. Lo, Environmental Ethics, The Stanford Encyclopedia of Philosophy (Summer 2002 Edition), URL = http://plato.stanford.edu/archives/sum2002/entries/ethics-environmental/.

The more inclusive P is, the ‘greener’ or ‘deeper’ the approach has been deemed. In Floridi [1999] , Floridi and Sanders [2001 ] , and Floridi [2003] I have defended a ‘deep ecology’ approach to information/computer ethics.