Causal Responsibility

Perhaps the root notion of responsibility concerns the sense in which we use the term to indicate mere causal efficacy as when we say things like, "The leaking value is responsible for the water in the basement," or "An infestation of locusts was responsible for poor crop production and starvation in Eastern Africa last year."

In these cases we attribute responsibility, in the sense of causal efficacy, without also imputing agency or intention­ality to the cause. The causal sense of responsibility differs from the moral senses of the term in that it carries no implication of praise or blame. Responsibility in this sense has to do with mere happenings rather than with doings. Non-agentive causal factors can be said to be 'responsible' for producing effects in this basic sense of the term.

The locust infestation case represents agent causality. As living beings, locusts are agents, that is, they act autonomously and their actions evince a certain purpose that is, in this case, built into their biological program. Locusts, like other living things, need to take in energy from their environments in order to survive and reproduce. The locusts did not intend to cause human beings to starve; they only intended to feed themselves. While it is possible to impute agency and a kind of intentionality to things like locusts, we do not treat them as moral agents, that is, we do not hold them responsible for their actions in the moral sense of having liability and being subject to moral blame. The question of how far down the tree of life agency and intentionality go is a difficult and complex one. I will discuss it later on in the sections on agency and intrinsic value. But for the time being it suffices to note that non-human agents can be responsible in the causal sense but not also in the moral sense.

When applied to human beings attributions of responsibility can sometimes take this form, that is, agentive causal efficacy, as when we say an infant was responsible for spilling some milk on the rug. The infant did not spill the milk intentionally or deliberately, but the infant is responsible for spilling it nonetheless, but only in the causal sense. This is an example of agent causality but without any imputation of intentionality or purpose.

In law, the insanity defense rests on the possibility of distinguishing among cases in which the agent acted knowingly and intentionally, in which case they are held morally responsible for the conse­quences of their actions, from cases where the actions were unknowing, unintended, or compelled, even though the defendant may have been causally responsible for some harm.

Even fully competent moral agents may sometimes be causally responsible for producing effects though they did not intend them nor deliberately act to produce them, as when we accidentally knock over the coffee mug. The idea expressed by this sense of responsibility can also be applied in cases in which the subject of responsibility is a person, however, except for cases such as those noted above, when applied to persons there is normally also an attribution of intentionality and moral agency that goes along with the assertion of causal efficacy.

The assumption of intentionality and moral agency is generally held to be necessary to ground ascriptions of moral, as opposed to mere causal responsibility, and is seen as a necessary precondition for the further ascription of "responsibil­ity" in the senses of "accountability", "culpability", or "liabil­ity."

However, it is important to note that not all attributions of responsi­bility in the senses of accountability or liability presuppose causal responsibility. We may, for instance, hold a military officer liable for the actions of a soldier under his command, even though the officer was not the cause of those actions himself. While responsibilities, in the moral sense, are limited by the range of our causal powers, that is, by the range of effects which we have produced or can causally influence, there is no necessary connection between an agent being the cause of some effect X and his being morally responsible for it.

Ascriptions of liability, culpability, and blame require the presumption of moral agency, which in turn requires the presumption of intentionality as well as further conditions such as whether the act was a voluntary one or done under duress, whether the actor did or should have foreseen the consequence, and assumptions about the agent’s moral competence. Moral agency is also a necessary condition of being the bearer of moral duties and obligations. But it is not a necessary condition of being the object of the moral duties and obligations of others. Things which can function as the direct objects of moral duties and obligations I call "moral patients." Moral agents can also be moral patients, but moral patients need not also be moral agents. I will drill down further into the meaning of these terms in other posts.

Liability Responsibility

Moral responsibility, as opposed to mere causal responsibility, for actions is applied to agents only when several psychological conditions have been fulfilled.

The main conditions usually assayed for ascriptions of moral responsibility in the sense of liability are: (1) intentionality, (2) voluntariness, (3) being able to give an account of one's actions (4) being able to have done otherwise, (5) being able to tell right from wrong, (6) foreknowledge of probable consequences, and (7) causal efficacy. Moral agents may be excused from moral responsibility for their actions if one or another of these conditions are not present at the time of the act, e.g. if they did not in fact cause an effect, or if the consequences of their acts or omissions could not have reasonably been foreseen, or if they were at the time not able to do otherwise, or not able to distinguish right from wrong.


Moral responsibility applies only if an agent possesses certain normal psychological capacities of understanding, reasoning, and control over his own behavior, the possession of which is commonly held to be a precondition of the appropriateness of moral praise or blame. Ascriptions of liability responsibility are mainly retrospective; they function to ground moral evaluations of the past actions of presumptive moral agents.

Determining whether or not a moral agent is liable for a past action is related to the notion of accountability. Moral agents can be asked to give an account of why they have done or not done a certain thing. In some cases, the agent may be able to present an acceptable excuse or justification for their action which may have the effect of reducing or removing his moral liability to blame or punishment. Thus, one can be accountable for one's actions without also being liable for them. When an agent is found to be liable for some kind of moral wrong we say that she is culpable. Moral agents who are culpable for wrongful actions, however, may still not be punished or blamed, because they may be forgiven.

To be forgiven is not the same thing as to be excused. Forgiveness presupposes that the agent was iiable to be blamed and was in fact culpable for some wrong. To excuse someone from a moral obligation, on the other hand, means that there is some reason why the agent should not be held liable or culpable for not doing what they ought to have done. Many kinds of reasons can be exculpatory, for instance, the agent may have lacked the power to do what was morally required, or may have had other, stronger moral obligations to fulfill that prevented her from doing it. The ability to give an account of the reasons for one's actions, then, plays a crucial role in ascriptions of liability responsibility. When we use moral language in an evaluative, backward-looking way, then we want to know why an presumptive moral agent did or failed to do something that might ordinarily be considered something she ought or ought not do.

This is one reason why we do not generally ascribe liability responsibility to nonhuman animals, nor to human infants and young children -- they lack the ability to give an account of their actions. Some other categories of moral patients, for instance, those who are psychotic or in comas, also lack this capacity, and so are not held liable for their actions in the same way in which competent moral agents, generally speaking, human adults, are. Since most human adults are competent moral agents, we generally expect them to be able to give an account of why they have acted in the ways that they have, and those reasons are then factored into our moral evaluation of their behavior.

A great deal of the literature of moral philosophy and of law deals with understanding the precise contours of the notion of liability responsibility. The notion of liability responsibility is connected with metaphysical questions, for instance, the notion of free will, and is vital for understanding the legal notions such as culpability for harms or liability for damages produced by one's action on others.

The philosopher Iris Marion Young contrasts the notion of liability responsibility with a different idea of responsibility that she calls 'political responsibility' but which I prefer to call 'social responsibility'. In drawing this distinction she notes that, The most common model of assigning responsibility derives from legal reasoning and is used to find guilt or fault for harm.

Under the fault model, one assigns liability responsibility to particular moral agents whose actions can be shown as causally connected to the circumstances for which responsibility is sought The agent that produces harm can be a person or a collective agent, such as a corporation (Young 2004, 368).

Liability responsibility can be further divided into fault liability and strict liability. In the former case one assumes that the agent knowingly and voluntarily engaged in behaviors that caused harm to others, while on the strict liability interpretation a person or collective agent can be held to be liable for action that caused harm even if they did not specifically intend the harm.

Both the fault and strict notions of liability responsibility are essentially backward-looking. They function in moral discourse as a mean of assigning blame for past actions, and in judicial proceedings for determining punishment or compensation for past crimes or torts.

Furthermore, according to Young, liability responsibility “has the function of absolving other agents who might have been candidates for fault…. and usually implies that others who were accused are not guilty” (368). This feature of liability responsibility is not also a feature of the notion of social or political responsibilities which can be shared without diminishing the responsibility of other agents. For political, or what am calling social responsibility, my having these kinds of responsibilities does not diminish or excuse you of your similar but perhaps not identical social responsibilities.

While it is necessary to understand the notion of liability responsibility, is it only of marginal interest to us here. Rather, the primary sense in which I shall be using the term “responsibility” is in the forward-looking sense in which moral agents are said have certain "responsibilities", in the sense of moral obligations, that direct them to the attainment of certain ends concerning other moral agents and towards moral patients. This is the substantive and prospective sense of the term in which moral responsibilities guide the actions of competent moral agents towards the attainment of certain ends or direct them to perform certain duties.

While the conditions for competent moral agency used in the retrospective, evaluative sense of "responsibility" can also be employed as a criterion for who can qualify as a bearer of substan­tive moral responsibilities, moral agency alone does not tell us in particular what these obligations are, or to whom they are owed. Moral agency, which is the precondition of both liability responsibility and normative moral responsibility thus lacks substantive normative content since it does not specify the ends of moral action, nor its proper objects, nor how moral responsibilities are assigned to particular moral agents. All it tells us is that the agent in question can be presumed to be a moral agent, that is, the sort of thing to which moral obligations and duties, 'responsibilities' in the sense, can be meaningfully ascribed.