An agent is considered negligent when an agent disregards risks that a responsible agent would have made himself or herself aware of. In the case of negligence (as opposed to recklessness), it is not required that the agent actually knew that their actions put others at risk. It is sufficient that the agent should have known - or should have checked out - the possibility of harm.
Negligence is thought to be a problem for motive-based (desire-based) moral theories. This is because negligence does not spring from any bad motive or desire.
Imagine a farmer who wants to get his hay to a neighboring farm. He does not want to make two trips - a perfectly common motive - so he stacks all of the hay on the trailer. Along the way, a few bails of hay at the top of the poorly secured stack fall off. They bounce into oncoming traffic where they cause a wreck.
The farmer is considered negligent and condemned for his actions. Yet, they spring from no bad motive. The only motives at play are a desire to get the hay to the neighbor's farm and an interest in doing it in one trip.
A person seeking to defend a motive (desire) based theory needs to explain the moral category of negligence.
Henry Sidgwick, a philosopher at the start of the 20th century, used the problem of negligence to challenge the motive-based theory of James Martineau - and to defend act utilitarianism as the better theory.
Martineau argued that what matters in judging an action is not its consequences but the motives of the agent. According to Martineau, each motive comes with its own intrinsic value. We could know this value by calmly reflecting on the matter. God wrote this knowledge into the human brain at creation, and we have access to this information. Whenever we have a choice among two possible actions, our moral task is to determine the motive for each option, then choose the action that springs from the best motive.
According to Sidgwick, the moral concept of negligence proves that consequences are what matter in evaluating an action, not the value of the motive from which it sprang. Negligence springs from no bad motive. Yet, it is condemned. The reason it is judged harshly, Sidgwick tells us, is because of its consequences.
Yet, Sidgwick's response has its own problems.
One problem is that the farmer is guilty of negligence even if he does no harm. One can tell the farmer that it is wrong to stack the hay so high because it is dangerous even before the farmer leaves with the load. If the bails fall when there is no traffic, and the farmer clears the road before anybody is hurt, the lack of bad consequences does not allow the farmer to escape blame for what could have happened. Even if no bails fall and he reaches his destination safely then he is still no better than the drunk who manages to make it to the driveway without killing anybody. He can still be condemned for his negligence.
Nor is the farmer to be condemned just for creating a risk of bad consequences. Let us assume that one of the tires was made of substandard materials such that even a normal load would risk blowing the tire, causing a wreck. In this case, assume that the farmer did not overload the trailer but decided to make two trips. In this case, the farmer is creating a risk. However, he is not guilty of negligence precisely because he has no way of knowing about the risk.
Here, then we have examples of condemnation without bad consequences, and bad consequences without condemnation, both of which create problems for the act-consequentialst.
Desirism differs from Martineau's theory in a number of ways. One important difference is that desirism evaluates motives (desires) according to their malleability and the degree to which they tend to objectively satisfy or thwart other desires. Desires have no intrinsic value. They only have a power to objectively satisfy or thwart other desires. It is in virtue of this capacity that others have reason to promote or inhibit certain desires.
On this system, a right act is not necessarily an act that springs from good desires. It is the act that a person with good desires would perform. We do not say that the person who rescues a drowning child because he wants a reward should have let the child drown. We still say that rescuing the child was the right thing to do. However, we condemn the rescuer when he demands the reward or reports that, "If I had known I would not have been rewarded I would not have rescued the child." In these cases, we do not condemn the agent for the act of rescuing the child (which still remains the right thing to do) but for the attitudes he has towards that rescue.
In the case of negligence, the good person has a level of concern for the interests of others that motivates him to discover (before it is to late) and avoid undue risks to other people. This concern motivates the good person to investigate potential threats and take precautions against those threats. When we condemn the farmer for negligence, we condemn the farmer for actions that demonstrate a lack of concern for the interests of others. This farmer obviously did not care enough to protect others from harm and, for that, he is morally condemned for his actions.
We hear this in the language of condemnation. "You should be more careful." The intersts that motivate agents to be more careful are those that are targeted by condemnation in the case of negligence.
This account also handles the two cases that the act-utilitarian has problems with. The negligent person who, by luck, causes no harm still deserves condemnation because he did not care enough to prevent the risk. At the same time, the person who causes harm by putting a normal load on a trailer with a defective tire is not condemned because his actions do not demonstrate a lack of concern for the interests of others.
Our reasons for promoting a concern for the well-being of others diminish as the costs of caring go up and the benefits (the risks of harms) go down. We have no reason for an overall concern that paralyzes the community as each and every member ferrits out every possible source of harm. At some point, the costs of determining if there is a risk go beyond what we have reason to expect from the person with good desires. The case of the blown tire is an example of this.
Desirism even tells us where to draw the line between negligence and accident. We should demand a concern with the welfare of others that is strong enough to prevent easily prevented harms, but not so strong that people waste huge amounts of energy hunting down all possible sources of harm. We should draw the line at the point where additional care for the interests of others creates costs in terms of time and effort spent that are greater than than the costs in terms of harms that one can expect will be avoided. Please note that, in saying this, we are evaluating the proper level of concern, rather than making a cost-benefit analysis of each individual action.
The concept of negligence does not only apply to actions that demonstrate a lack of concern for the interests of others but also for beliefs. A person who carelessly adopts beliefs that threaten the interests of others may be shown to lack a level of concern for others that people generally have reason to promote. In these cases, an agent may be condemned for epistemic negligence.
Desirism, then, shows its superiority over act-utilitarian theories on the issue of negligence. It explains negligence without bad consequences, bad consequences without condemnation, and where to draw the line between them. At the same time, it avoids the faults of a Martineau-like motive theory. It grounds the value of motives on the reasons people have to praise and condemn, and allows the application of praise and condemnation to cases not only where an agent acts on a bad desire, but fails to act on a good desire.