Defending Realism from ‘Error Theory
I. Introduction
It seems to me that one of the primary reasons why certain people reject moral realism is because they suffer from a misunderstanding of what moral realism requires. They seem to think that a moral realist must accept both of the following propositions:
(1) Intrinsic prescriptivity is a part of the meaning of moral claims.
(2) Intrinsic prescriptivity exists.
However, the person asserting that moral realism must be rejected holds that (2) is false. Intrinsic prescriptivity (which is often called 'objective values', though this term proves to be a misapplication of the concept of 'objective') does not exist. Thus, we have to reject moral realism.
According to J.L. Mackie, both of these assumptions are built into the very meaning of moral terms (Mackie, 1977). Since (2) is false, it follows that all of our moral claims, from the wrongness of abortion to the wrongness of murder, are false.
However, these two propositions identify only a specific instance of moral realism. Rejecting moral realism because this specific type has problems is no different than arguing that horses do not exist because there is no such thing as Pegasus (and Pegasus was a horse).
The more general definition of moral realism says that a moral realist must defend the following two propositions:
(1') X is a part of the meaning of moral claims (for some X), and
(2') X exists
If X = intrinsic prescriptivity, then you get the type of moral realism identified at the start of this essay -- a species of moral realism that does not exist. However, it is still possible that X = something other than intrinsic prescriptivity. If it does, and if X exists, then we can have moral realism without intrinsic prescriptivity -- without what has been confusingly mislabeled as 'objective value'.
II. Overview
There are a lot of different moral terms. Here, I hope that I can clarify my view by focusing on a specific moral term -- the concept of moral obligation. We are trying to discover what X equals in propositions (1') and (2') above. I am going to suggest that X equals the following combination of propositions.
(1) An obligation aims to fulfill those desires that can be molded through positive and negative reinforcement and that tend to fulfill other desires regardless of whose they are.
(2) 'Obligation' is applied to intentional actions
(3) An obligation would fulfill the desires in question.
(4) Obligations fulfill the desires in question both directly and indirectly.
Furthermore, I am going to argue that X exists, which would classify me as a moral realist. However, I am a moral realist who denies the existence of intrinsic prescriptivity (so-called 'objective values').
This formula says that value is not an intrinsic property, but that it depends on desires. Without desires, there is no value. But, desires exist, as do the relationships between desires and the objects of evaluation. So, under this definition, both (1') and (2') are true, making this a realist moral theory.
Furthermore, because the desires being referred to are the same desires no matter who is speaking, there is a common truth to moral statements. If one person says 'People have an obligation to do P", and another says, "There is no obligation to do P", at least one of them is mistaken. They cannot both be right.
Does this sound like a contradiction? Can a person believe that there are no 'objective values', yet also believe that moral statements are objectively true or false? As I will eventually show, it only appears to be a contradiction because people misuse the concept of 'objective values'. There are a lot of ways for things to be 'objective' without being an intrinsic property.
I want to stress that, in this essay, I am not going to try to derive these four principles from some more basic foundational assumptions. Rather, this essay will look at the implications that these four criteria logically entail. I wish to argue that terms defined according to these criteria would look substantially like our conventional moral terms.
In this type of argument, the four criteria are 'given' in the sense that they are assumed true. What they logically entail is not 'given'. Nor is it a 'given' that what we can infer from then matches our common-sense moral concepts. These are the things I need to demonstrate.
The form of this argument is the same as that which would be used in comparing two pictures. The arguer says, "Look at this feature on Picture 1. It matches this feature on Picture 2. This other feature on Picture 2 matches this other feature on Picture 1. From this, we reach our grand conclusion, that we have two pictures of the same thing." I will continue to go back and forth between these two pictures, showing similarities, in order to demonstrate the four criterion above, and our common-sense concept of moral obligation, are two pictures of the same thing.
III. Genesis
I think it might help to understand these criteria if I explain how I came about them.
In offering this explanation, it is important to note that this is not the same as a defense of my thesis -- that comes later. Rather, I offer this story as an explanation of what it is I will eventually be defending.
Newton, for example, could explain how he came up with his theory of gravity with a story of an apple falling on his head. Even if his inference from that event to the conclusions that he drew were mistaken, or even if the event never happened at all, this cannot be used as an argument that Newton's theory of gravity was flawed and should be rejected. Refuting Newton would require a different type of argument that attacked the theory directly.
But, my 'apple' story starts with a study of act-utilitarianism. Act utilitarianism holds that the right act is that act which produces the greatest utility. Act utilitarianism's first major proponent, 19th century philosopher Jeremy Bentham, described the theory by saying that the best act is the act that produces "the greatest happiness for the greatest number".
Philosophers came up with a huge number of counter-examples to this theory. There are times when the 'right act' will not produce the best consequences. Famous examples in contemporary philosophical literature include a doctor who has an opportunity to kill a healthy patient to distribute his organs among five others -- thus saving five lives. Counterexamples also include a case of a sheriff who can frame an innocent person and thereby stop a vengeful mob from rioting. They include a case of a reporter in a 3rd world country given an option, by an evil dictator, to murder one villager, or to watch the dictator murder 20 others. In all of these cases, the 'right act' seems to diverge from the act having the best consequences.
To fix some of these problems, some utilitarians have migrated from act-utilitarianism to rule-utilitarianism. On this theory, we evaluate an action according to its conformity to a set of rules, and we evaluate the rules according to their ability to maximize utility. A rule against killing healthy patients, or against framing innocent people, or against killing innocent villagers, would tend to maximize utility overall, even though it does not do so in some very rare and exotic circumstances.
The problem with rule utilitarianism is that it collapses into act-utilitarianism. Utilitarianism, in general, says that consequences are all that matter. What if an act that violates the rule will bring about the best consequences? If consequences are the only thing that matter, and an act that violates a rule will bring about better consequences, then we cannot justify the decision to obey the rule based on consequences. Following the rule has to have some sort of value independent of consequences -- it has to be 'good in itself'. Rule utilitarianism has a notorious difficulty explaining this second type of value. Without it, the 'right act' remains the act with the best consequences, and we return to act-utilitarianism.
However, what if the 'rules' are written into the brain in such a way that they do not allow for exceptions?
There is a principle in ethics that 'ought' implies 'can'. Another way of stating this same principle is to say that 'cannot' implies 'it is not the case that one ought'. If an agent 'cannot' violate the rule, then it follows that it is not the case that he 'ought' to violate the rule.
Desires are rules written into the brain that do not allow for exceptions. An aversion to killing healthy patients, or to framing innocent prisoners, would act as a motivating reason preventing the agent from performing these acts, even when the act would have the best consequences. So, the doctor with such an aversion -- such a rule written into his brain -- would be compelled not to kill innocent patients, even when doing so would produce the best consequences. The sheriff with such an aversion would find himself unable to frame the innocent person even while he is aware of the consequences of not doing so.
To see how this works, we need to take a closer look at the concept of a desire.
IV. Desires
What would it take for an agent to always perform the act-utilitarian best action? To do so, he would have to be a person who has only one desire -- the desire to perform the act-utilitarian best action (as well as accurate beliefs that would allow him to correctly identify that action among the options available). If he had any other desire -- any desire at all -- then there would be circumstances in which that second desire would conflict with the desire to maximize utility, and pull the agent away from performing the act that maximizes utility.
For example, assume that an individual has two desires -- a desire to maximize utility, and a desire for strong alcoholic drink. Let us make this desire for strong alcoholic drink very strong, to the point that virtually everything the agent does is governed by whether or not it will allow him to get drunk. He would kill, literally, for a drink. He has an interest in maximizing utility, but this is only going to manifest itself when the two options present nearly equal chances of obtaining alcohol.
We can easily imagine this agent pursuing alcohol at virtually all costs.
Now, let us weaken this desire in stages. As the desire weakens, the agent will find more and more cases in which he is acting to maximize utility. There will be more and more instances in which he is willing to sacrifice a drink in order to maximize utility.
However, there is only one point at which he will never sacrifice utility for the sake of a drink, and that is the point where the desire to drink does not exist. If there is any desire to drink at all, there will be some set of cases where the agent will not act to maximize utility. In other words, the only person who will always act to maximize utility, is the person who desires nothing but to maximize utility.
To say that a person always ought to maximize utility is to say that he may have no desire but to maximize utility. Act-utilitarianism has often been criticized for being "too demanding", but it actually goes so far as to demand the impossible.
A being with no other desire would have no aversion to pain, no desire for sex, no preference for chocolate ice cream over vanilla, no special affection for any friend or relative, would find no discomfort in heat or cold and would look upon a sunset and see only its utility. We are talking about a strange form of life that could not even be recognized as being human.
Of course, there is nothing of intrinsic merit in being human, but the term does describe a set of possibilities relevant in any use of the inference 'ought' implies 'can'.
So, by introducing two premises into the rule-utilitarian argument -- (1) that the rules are written into the brain in a way that does not allow for exceptions, and (2) 'ought' implies 'can', we can save rule-utilitarianism from collapsing into act-utilitarianism. The types of rules we are talking about here are desires. The type of theory we are talking about here is desire-utilitarianism.
V. Act-Utilitarian Counter-Examples
That was the story of how I came up with the idea of desire-utilitarianism. However, to defend the theory, I need to show how well it handles a number of concepts used in moral discourse -- how it makes them meaningful and sensible.
A part of the defense of this theory rests in its ability to handle the common counter-examples to act-utilitarianism. The doctor ought to have an aversion to killing healthy patients -- a rule, written into the brain that does not allow for exceptions, against performing such an act. The sheriff should have an aversion to framing innocent people. The reporter put in a dilemma by the evil dictator ought to have an aversion to killing innocent people as well.
A person with such a desire will sometimes perform actions that are not the act-utilitarian best act. The doctor with an aversion to killing healthy patients will sometimes refuse to do so even when doing so would save other lives. The sheriff with an aversion to framing innocent people will sometimes let the town riot rather than frame an innocent person.
The reporter given the option of killing one peasant by an evil dictator may refuse to act.
Desire-utilitarianism is consistent with the demand that people sometimes not perform the act-utilitarian best action.
VI. Greater and Lesser Evils
The case of the reporter illustrates another important benefit of this theory -- its ability to deal with the concept of greater and lesser evils. Our intuition regarding the reporter is not that he ought not to murder one villager to prevent the murder of 20, but only that he ought not to do so casually. If he does commit the murder, he must at least demonstrate some reluctance -- we expect him to try to find some way out of the dilemma. We expect him to commit the murder only after he has exhausted all options. And, if he decides not to -- if he decides to keep his hands clean and the dictator then keeps his promise to murder 20 villagers, we can understand that as well.
Can desire-utilitarianism handle the complexities and nuances of this type of situation?
Desires come with different weights -- different strengths. A desire for chocolate can outweigh an aversion to being fat. A desire for the benefits of a paycheck can outweigh the aversion to getting up in the morning and going to work.
Earlier, I said that these rules are "written on the brain in a way that does not allow for exceptions'. However, they do allow for the possibility of being outweighed.
What's the difference?
Take a scale. Put a weight on one side. Pick up another weight and place it on the opposite side. The scale will tilt towards the heavier weight. If the second weight is light enough, the first weight will still dictate the direction of tilt. If the second weight is heavy enough, it will dominate the first. It 'outweighs' the first.
An example of the concept of an 'exception' is one in which, the instant the second weight is added, the first weight is removed. This means that no matter how small the second weight is, it will win the battle against the first.
This illustrates how 'outweighing' differs from 'exceptions'. The difference between the two is that it takes more weight to shift the balance in the case of 'outweighing' than it does in the case of 'exception'. In the first case, the second weight must be greater than the first to tilt the balance. In the second case, any weight at all, however slight, will tilt the balance.
(A) An Example of Moral Weight
So, a parent takes his child fishing. They park at a turnoff on a seldom-used mountain road. One other car is parked there. The child gets stung by a bee, and starts to have an allergic reaction. The parent's car will not start. He looks in the other car -- the keys are in the ignition. For an anxiety-filled minute he calls out for the owner, and gets no answer. The parent ought to have an aversion to taking another person's car without his consent. But, like all aversions, it can be outweighed by a stronger desire to save his child's life.
However, the aversion to taking something without consent will still weigh on the person with good desires. He owes the owner of the car enough consideration to minimize the cost. He leaves behind a note, and makes arrangement to send the car back -- or to send somebody back -- to pick up the owner as soon as possible. He apologizes and admits the wrong in taking the car, and asks for forgiveness, and offers compensation for any inconvenience he may have caused the owner.
(B) Moral Exceptions
Both 'exceptions' and 'outweighing' exist in our moral code. 'Exceptions' depends on the degree to which our desires can be fine-tuned. An individual could, perhaps, have an aversion to lying except to wrongdoers. So, he feels the compulsion not to lie to others in most circumstances. However, when the NAZI comes to his door and demands to know of any Jews, he does not need to outweigh his aversion to lying. His aversion is to lying 'except to wrongdoers', so there is no impulse at all to tell the truth to these Nazis.
There is an important psychological difference between the outweighed wrong and the exception. The outweighed wrong consists in an aversion to X, countered by an even stronger aversion to Y, where the agent allows X in order to prevent Y. A moral exception takes the form of a desire that "not X except when Y". When Y occurs, there is no impulse at all to 'not X'.
There is nothing in this theory that says that a person cannot have a desire of the form "not X except when Y". And if he can have such a desire, an argument can be made that people, at times, should have such a desire, as in the case of lying to the NAZI. The individual in this example should not be so accustomed to lying that he will lie at the drop of a hat. However, when the NAZI officials come to his door, the aversion to lying should be outweighed by his desire to protect innocent people from harm. If he has an aversion to lying except to hide innocent victims from wrongdoers, then he can lie to the Nazis without any trace of guilt.
(C) Moral Absolutes
One of the implications of this theory is that there are no moral absolutes. Lying is not absolutely wrong under all circumstances. It may be wrong, but under circumstances it may be a lesser evil. Or, it may be a type of case where we are sophisticated enough to aim for a more complex aversion to lying -- one that allows for exceptions. Whether lying falls within the realm of exception or lesser evil, it is not the case that one should never lie.
Yet, if we have the capacity to program people into having a complex desire such as "an aversion to X except when Y", we still have not violated the claim that moral rules are written into the mind in a way that does not allow for exceptions. The full rule written into the mind in this case -- the full desire -- is a desire for "not X except when Y", and this full rule does not allow for exceptions. He cannot cast it aside on a whim when some other situation Z should spring up. It can be outweighed, but it cannot be ignored.
VII. Ought Implies Can and the Role of Praise and Condemnation
The maneuver to save rule-utilitarianism from collapsing to act utilitarianism made use of two facts. First, the rules were written into the mind in such a way that do not allow for exceptions (given a precise understanding of 'do not allow for exceptions'), and 'ought' implies 'can'.
'Ought' implies 'can' simply says that a person is not to be blamed for that which is outside of his control, or praised for that which he could not have avoided doing. 'Ought' implies 'can' is logically equivalent to 'cannot' implies 'it is not the case that one ought'. It is in this latter form that we most commonly find this principle. For example, it makes no sense to say that I 'ought' to teleport the child out of a burning building, because I cannot do such a thing.
This, then, gets related to the issue of free will. The claim is that, if determinism is true, then morality is nonsense, because there is never a case in which the agent 'could have done otherwise'. All alternatives to the action that the agent actually performs is like that of teleporting the child out of the burning building -- something that the agent could not have done, so not something that the agent should have done.
Compatibalists argue for a concept of 'can' that is compatible with determinism (hence, the name). A common compatibilist proposal is that 'could have done otherwise' means 'would have done otherwise, if he had wanted to'. It takes seriously the fact that our intentional actions are caused by our desires -- such that 'wanting to do otherwise' would determine that the agent would have done otherwise. If we change an agent's desires, we will change his actions as well.
I would like to look at this a little deeper.
If 'could have done otherwise' means 'would have, if he had wanted to', then 'should have done otherwise' gets translated into 'would have done otherwise if he had wanted to, and should have wanted to". Here, 'ought' implies 'can', but it takes something in addition to 'can' to make the trip back to 'ought'. Not everything that can be done, should be done.
This implication from the compatibilist understanding of 'ought' implies 'can', is that moral questions are ultimately concerned with what a person 'should have wanted'. This is fully compatible with desire-utilitarianism, which also states that moral questions are ultimately concerned with what we should and should not desire.
However, we now need to make sense of "should have wanted to."
As we try to investigate the issue of what a person should want, we are still stuck with the inference that 'ought' implies 'can'. Consequently, our analysis of "should have wanted to" has to be compatible with the implication that the agent "could have wanted to". We cannot reasonably say that an agent should want something unless it makes sense to say that he could want that thing.
Recall that in my initial set of conditions, I stated that the relevant desires were those that can be molded. That ties in here because we are looking for the desires that an agent can have.
One of the properties of desires is that they are immune to reason. Desires are brain states -- a physical property much like hair color, height, and weight. You can no more reason a person into having a particular desire than you can reason him into being 165 pounds. For example, you can take two people -- one with a preference for chocolate ice cream, and the other with a preference for vanilla -- get them to agree on all of the relevant facts, and still end up with the first person preferring chocolate and the second person preferring vanilla.
However, desires are susceptible to positive and negative reinforcement. Condemn and punish those who try to eat chocolate, and praise those who seek out vanilla, and you can end up with two people whose desires are such that they both avoid chocolate and seek vanilla.
This does not require any type of advanced cognition such as "appreciating the wrongness of chocolate-eating". It works on animals, who, we may assume, lack the capacity to "appreciate the wrongness of wetting the carpet. Rather, their behavior can more simply be explained in terms of acquiring an aversion to wetting the carpet. This aversion may have started out as a means of avoiding punishment, but it turns into something that the animal avoids for its own sake. It will govern the animal's actions even when there is no master there to punish him.
This is the same effect we try to have on our children. We hope that our moral education will create a child who will not take what belongs to others, even when the child (and the adult he becomes) knows that there is nobody around to punish him. He will not take what belongs to others, simply because he now has an aversion to taking it.
So, I am not talking about some mysterious super-natural properties here. I am talking about the simple process of affecting desires through positive and negative reinforcement -- through praise, condemnation, reward, and punishment. It should come as no surprise, then, that these are core components of the institution of morality.
This theory, now, has another significant advantage over act-utilitarianism. Act-utilitarianism cannot handle the link between 'blameworthy' and 'wrong'. To blame a person is an act, separate and distinct from the act for which the agent is assigning blame. As such, a person should blame another only if blaming has the best consequences, and refrain from blaming when it does not have the best consequences. The fact that the act in question was right or wrong turns out to be irrelevant to whether the agent should be praised or blamed.
This theory, on the other hand, draws an intimate connection between being blameworthy, and being wrong. A wrong act is an act that people generally should not want to perform, and the way to make it the case that people do not want to perform such acts is through the use of positive and negative reinforcement -- blaming the person who does wrong. Blaming not only counts as negative reinforcement for those being blamed, but also those who hear about it. In fact, simply calling the act 'wrong' generates some negative reinforcement -- because one identifies the act as one where blameworthiness (negative reinforcement) would reasonably apply.
Now, allow me to retrace my steps for a minute.
? 'Should have' (or 'ought to have') implies 'could have'.
? 'Could have' implies 'would have if he had wanted to'.
? 'Should have' implies 'would have if he had wanted to' plus 'should have wanted to'
? 'Should have wanted to' implies 'Could have wanted to'
? 'Could have wanted to' means 'positive and negative reinforcement -- praise, condemnation, reward, and punishment -- can affect desires in such a way as to create an agent who does want to'
I have reached a point where I can explain 'could have wanted to', but I still need to add something to reach the concept of 'should have wanted to'. I need a way to rank the list of options on the list of could have wanted, in order to select those that people should have wanted.
According to this theory, there is only one way to evaluate anything, and that is according to its capacity to fulfill or thwart desires. Thus, there is only one way to evaluate desires, and that is according to its capacity to fulfill or thwart other desires. Which other desires?
Well, our four criteria that we have started with suggests that it is 'all desires, regardless of whose they are'. We are going to look at the list of desires that people could have, and direct our tools of praise, condemnation, reward, and punishment to promote those that fulfill other desires, and inhibit those desires that thwart other desires.
The ultimate consequence of all of this is to justify the claim that this concept of morality does not require anything so exotic as "free will". The theory is not only compatible with determinism, but depends on determinism. It requires that acts are caused by the desires of the agent, and that the desires themselves are, at least in part, an effect of praise, condemnation, reward, and punishment. If these causal factors were not at work, moral claims would not make sense.
The mistake, leading people to believe that some sort of "free will" is required, comes from understanding the role that cause and effect has in moral judgments. It is true that 'ought' implies 'can', but 'can' only requires that the act, or the desire, fit within a particular causal chain -- that desires cause intentional action, and that positive and negative reinforcement causes changes in desires.
So, this was a long section. I would like to offer a brief summary of what is has accomplished. It has shown that the four criteria given above can handle the moral concepts of:
? 'Ought' implies 'can'
? 'Wrongdoing' implies 'blameworthiness'; 'right-doing' implies 'praiseworthiness'
? the link between moral concepts and the most reasonable theory of 'free will'
VIII. Obligation, Permission, and Prohibition
Act-utilitarianism has another important failure. Act-utilitarianism allows for only two moral categories; obligatory, and prohibited. The one act that maximizes utility is obligatory; all other options are morally prohibited. Of the prohibited acts, some are worse than others, depending on the amount of utility they contain; yet all of them ought not to be done, compared to the one utility-maximizing act that ought to be done.
Our common-sense moral concepts, however, allow for three moral categories; obligatory, prohibited, and permissible. There are some acts, such as my writing this essay or the person sitting next to me reading the newspaper, that a person may decide to do, or not, as pleases them, without morality having anything to say about it, one way or another.
Desire utilitarianism allows for all three moral categories.
A good desire is a desire that it makes sense to encourage (using the tools of positive and negative reinforcement) in everybody. A desire to repay debts (or an aversion to not repaying debts) would be an example. There are other desires that we generally have reason to want nobody to have, such as a desire to commit rape or to torture small children.
There are also desires that it would be good for some people to have, but not for everybody to have. This is the set of desires that defines what is permissible.
For example, it is good to have a few people around whose desires were such that they would choose to be teachers. However, we do not want everybody to choose to be teachers. If they did, we would have no construction workers, no doctors, no firefighters, and no bus drivers, and we would all suffer for the loss. It is more desire-fulfilling, all things considered, if some people had desires that would cause them to choose to be teachers, others had desires that would inspire them to be doctors, others seek to be firefighters, and still others enjoy driving enough to drive busses.
A person with good desires would have a desire to repay debts, and an aversion to committing rape and to torturing young children. In addition, he may or may not have the desires that would inspire him to be a teacher or bus driver. These desires define options that are neither obligatory nor prohibited, but permissible options. "Pick one. The choice is yours."
I have no moral obligation to write this essay; nor am I under a moral prohibition not to write it. A person with good desires would feel no impulse to sit and write this essay; nor would he have a compelling aversion to writing it. So, he would be at liberty to decide for himself if he wanted to write the essay or, perhaps, do something else -- sleep on the bus, read a newspaper, play a computer game, or compose an email to his mother.
IX. Supererogatory
'Supererogatory' is a philosophers' word for 'above and beyond the call of duty.' We recognize that people have obligations. However, every once in a while, we hear of a person doing more than he was obligated to do. These people go 'above and beyond' and, as a result, obtain extra praise for their efforts. At the same time, their actions were not those that we would have expected them to perform or demanded from them. They had a moral permission to refuse. Their courage on the battlefield, their exceedingly generous contribution to charity, and their extra effort in encouraging their children by attending every game or school event, are praiseworthy in the extreme, partially because it is something we do not demand of every person.
Act-utilitarianism has no place for an action 'above and beyond the call of duty', because the call of duty is always to maximize utility. This is just another way that utilitarianism is seen as being too demanding.
Desire-utilitarianism has an answer.
That answer draws on 'ought' implies 'can', and that it makes no sense to demand of a person something that the person cannot do. Obligations and prohibitions are concerned with universal desires, which is tied to the effects that we can reasonably expect praise and blame to have on people generally. A stronger desire may be even more beneficial; however, we do not think it is possible (or worthwhile) to make this stronger desire universal. So, for the average person, we are content with the weaker desire and the actions that they cause.
Yet, every once in a while, we encounter a person who does have this good desire in a super-abundance; at a level we cannot reasonably expect the average person to have. When we encounter it, we give the agent particularly strong praise and recognition. We identify these acts as 'above and beyond the call of duty' because that is what they literally are; evidence of a good desire that is stronger than we could expect a common person to have.
X. Mens Rea and the Concept of 'Excuse'
Mens Rea means 'guilty mind'. It refers to a mental state that an individual must have to be considered responsible for a wrongdoing. Before I can be considered blameworthy for driving my car over a pedestrian, it must not only be established that I was driving my car and ran over the pedestrian, but that I also had a particular mental attitude that makes me responsible for running over the pedestrian. If it is discovered that my brakes failed, even though I had them inspected on a regular basis, I may have run over the pedestrian, but I do not have the 'mens rea' necessary to be blameworthy. Accordingly, if the pedestrian I ran over was a hit man that was trying to kill me, I would also not be considered guilty of any wrongdoing, even though running over pedestrians is generally considered something that I ought not to be doing.
As the examples in the previous paragraph illustrate, the concept of 'mens rea' is closely linked to the concept of 'excuse'. An excuse is a state of affairs that can shield a person from blameworthiness for doing something for which blameworthiness is the default state. The paragraph above uses the examples of accident and self-defense. There are others.
Desire-utilitarianism easily handles the concepts of 'mens rea' and 'excuse'.
According to desire-utilitarianism, what we are actually looking for in determining if somebody is blameworthy is whether failed to act as a person with good desires would act. Or, more precisely, whether a person with good desires would have done something differently.
There are five major categories of excuse. All of them work to break a connection between an act that we can assume that a person with good desires would not do, and the desires of the agent that performed the action.
(A) Accident
The first example, which I already mentioned above, is that of 'accident'. Running over a pedestrian is something that, at least on first glance, a person with good desires would generally avoid doing. Where we have a case of a person running over a pedestrian, we have good prima facie evidence of somebody failing to act as a person with good desires would act.
When the agent offers the defense of 'accident', she is saying, 'You cannot draw those inferences about my desires from this act. Even a person with good desires, in my situation, would have ended up running over the pedestrian." It would have happened because the mechanical failure resulting in the act could just as easily have happened to the person with good desires, and would have still resulted in the pedestrian being run over.
This is why it is important that the user performed regular maintenance on the vehicle. A person with good desires -- a person with desires that tend to fulfill, and no thwart, the more and the stronger of other desires -- would have done so, just to help make sure that nothing like this happened in the future. We must assume that the agent had regular maintenance perform for his claim that "the same could have happened to a person with good desires" to stick. Otherwise, the agent is still blameworthy for the lack of maintenance that contributed to the accident.
(B) Greater Good
I have also already discussed the excuse of 'greater good'. This description applies to the father who must take somebody else's car without consent in order to get his dying child to the hospital. We are assuming a case in which time is working against the father, and the owner is nowhere in sight.
When the agent offers the defense of 'greater good', he is saying, "You cannot draw those inferences about my desires from this act. Even a person whose aversion to using another's property is at the appropriate strength would have taken the car under these circumstances. This is because such a person would also have had a stronger aversion to watching his child die. My actions were those that a person with good desires would have performed." This allows an agent to get away with a wrongful act (something to which he should have had an aversion) when there is a greater good (something to which he should have had an even stronger aversion) at stake.
(C) Mistake of Fact
A police officer steps into an alley and discovers a body laying in a fresh pool of blood. He draws his gun. From the shadows, a man quickly steps out and points a gun at the officer. The officer shoots "in self defense" he would later say. Only, he discovers that the gun his assailant held was a toy gun. He was not in any danger. He only, falsely, believed he was in danger. So he falsely believed that he acted in self-defense.
However, the officer was in a position where it was reasonable to believe that the victim was aiming a real gun at him. The officer can reasonably claim, " You cannot draw those inferences about my desires from this act. Even a person with good desires would have reasonably believed, under these circumstances, that the victim was likely going to kill him, and would have acted on that belief. I did nothing that a person with good desires would not have done."
An important component of this excuse is the claim about what a person with good desires would have believed. This brings up the issue of epistemic negligence. If the police officer cannot provide evidence that his belief was reasonable, then he would show himself guilty of another moral failing, that of negligence. The negligent person does not have a sufficiently strong concern for the welfare of others. A greater concern would have caused the agent to double-check his facts and his reasoning, as time permitted, before taking an action that risked thwarting those other desires. The concept applies to physical acts -- such as failure to secure a load on a vehicle that puts others at risk of harm. It also applies to mental acts -- believing something without taking sufficient precautious to verify or falsity that belief.
Similarly, the person who leaves the airport with somebody else's luggage may need to show that it was easy to infer that the luggage was his own. The person who leaves a restaurant without paying can avoid blame if he can show a reasonable belief that the meal was free or that somebody else had already paid for it.
(D) Diminished Capacity
A young child takes a gun from his parent's closet and starts playing with it. The gun goes off, killing the child's friend.
One of the features of diminished capacity is that it is rare for an agent to claim the capacity for himself. If he can be aware of the fact that he has diminished capacity, then he can be expected to take precautions to make sure that it poses less of a threat to others. However, others can argue 'diminished capacity' in defense of the accused.
In this case, the defender of the accused is justified in saying, " You cannot draw those inferences about the desires of the accused from this act. Yes, a person with good desires would not have shot his friend. But the shooter, in this case, lacked the ability to understand that the gun would cause harm. Even a person with good desires might have performed this action, if he lacked the capacity to understand the potential consequences of his action. We cannot even charge the accused with epistemic negligence for his belief, because he lacked the capacity to question the a belief."
(E) Consent
An explosives expert sits ready to set off an explosion. As he does so, he knows that there is a man standing near the explosion. Yet, he sets off the charge anyway. The blast covers the victim with shrapnel, and the victim later dies of his wounds. However, we discover that these two people were a part of a team working on a movie. The victim was a stunt man connected to a wire that was timed to pull him away from the explosion, giving the appearance that the explosion himself had blown him away. However, the machinery did not function, so the stunt man took the full force of the blast.
The explosives expert can defend himself from blameworthiness by saying, " You cannot draw those inferences about my desires from this act. The victim obviously knew that there was a risk, but thought the risk worth taking. We can reasonably assume that, except in the case of diminished capacity, each person is the best judge of what is most likely to fulfill his or her desires. If I were to substitute the victim's judgment for my own, then there is a good chance that I would do a poor job. I would fail to best fulfill the victim's desires in most circumstances, would only exceed in cases of extreme luck, and would be at risk of sacrificing the victim's well being for the fulfillment of my own desires. So, even a person with good desires knows that, in determining what fulfills the desires of others, to seek the consent of those others."
(F) Summary
All of these categories of excuse are elements of the same formula. They all block an attempt to infer from the fact that an act-type was performed that a person with good desires would have had an aversion to performing, to the conclusion that the agent lacked the desires he should have had. If there is nothing blocking this inference, if we can reliably conclude from the fact that a particular type of act was performed that the agent did not have the desires that a good person would have had, then we have established 'mens rea' and the accused is held to be blameworthy.
XI. Act/Motive Divergence
Desire, intention, and motive-based theories have traditionally suffered from two major sets of counter-examples. One set of counter-examples concerns negligence, where a person performs a bad action, but his action is caused by rather common motives. The other is the case of a person who does what he should (or who does not do what he should not), but is acting on a bad motive.
(A) Negligence
I have already discussed the concept of negligence above, in the context of "mistake of fact". There is more to be said on this issue.
Negligence is thought to create a problem for motive-based theories because there is nothing particularly wrong with the motives behind the wrongdoer's actions. A farmer goes to a neighbor to pick up some hay. The trip involves driving a few miles on interstate highway. He goes to his neighbor, piles the hay high on his truck, then drives back home without securing the load. On the way home, while taking a turn, several bundles of hay fall off of the truck, causing a car in the opposite lane to wreck.
The agent, in this case, has performed a wrong action. However, his desires or motives were common desires that can be expected of all farmers. He wanted some hay to feed his animals; farmers do this all the time.
Consequently, if the value of an action is determined by its motives, then there is nothing to account for the wrongness of the negligent farmer's actions. There are no bad motives that can be used to assign a negative value to his actions.
The flaw in this line of reasoning rests in the assumption that the value of an action is determined by the value of the desires or motives which caused it. These counter-examples prove that there is something wrong with such a theory. However, that is not the theory that I am defending here.
The four criteria for an obligation that I am defending evaluates an action, not according to the actual motives of the agent, but according to whether the action is that action which a person with good motives (desires) would perform. A person with good desires would have a concern for the well-being of others that would have compelled him to secure the load better, or not piled the hay so high to start with, precisely because a high, unsecured load puts other people in danger. The farmer in question did not act as a person with good desires would act, and in this his actions are wrong.
The wrongness of the action, then, does not come from the badness of the desires that gave rise to it. It comes from the fact that it is not an action that a person with good desires would take. Moral evaluation is concerned just as much with the absence of a good desire, as it is with the presence of a bad desire.
(B) Grudge Justice
Another set of counter-examples against evaluating actions according to the motives can be illustrated by the following story. Frank is very jealous of his brother, James. James has a nice house, a wonderful family, and a good job. However, one day Frank discovers that James is embezzling money from the company. Frank, also, is in the habit of taking things from his company without paying for them. In addition, some of Frank's friends do the same thing, and he does not care. However, simply to spite his brother, he reports James to the authorities.
In this case, we would have to say that Frank did what he should have done in reporting James. At the very least, we have to admit that he did not do anything wrong. Yet, Frank's actions were not caused by a particularly admirable motive. He acted out of spite and with an intention to do harm -- motives we have a reason to discourage.
It would seem that a motive-based theory of action would have trouble handling cases such as this, where a person does what he should have done, but does so for a bad motive. The value of the act, then, seems to be independent of the value of the motive.
However, a slight shift in the way we compare actions to the motives gives us a way of avoiding these types of counter-examples. Here, again, I am going to deny that actions are to be evaluated according to the motives that cause them. Instead, I am going to assert that acts are to be evaluated according to whether a person with good desires would have performed that action.
In this case, yes. A person with good desires would report the embezzlement. Perhaps if only a small sum was embezzled and the harm inflicted on James' family would have been great, the person with good desires would simply say, "Pay back the money, and make sure it does not happen again." But, at some level, the person with good desires would report James. So, if Frank does so, for whatever motive, Frank has done what a person with good desires would have done. Frank did the right thing.
This time.
Because Frank does not, in fact, have good desires, Frank does other things that a person with good desires would not do. Those are things we can morally condemn Frank for.
XII. Proportionality
Another concept intrinsic to our moral system is the concept of <I>proportionality</I>. The punishment must fit the crime. But what does this mean?
Lex Talionis (or "an eye for an eye") was one crude example of proportionality, but it does not work in detail. If a person embezzles $100 from a company, we typically seek a punishment that is harsher than taking $100 from him. And is a 20-year jail sentence truly proportional to a 20-minute rape?
So, what does actually "fit" the crime?
This theory offers a suggestion -- that the punishment be measured according to the value of inhibiting the desires that tend to lie behind the act being punished, not only in the accused, but in society in general.
On this model, to call something 'wrong' is to say 'Hey, people, this is something that it makes sense for us to treat with our powers of positive and negative reinforcement". The greater wrong, then, is the wrong that it makes sense to treat with stronger negative reinforcement.
Some desires tend to thwart a great many other desires. For example, with respect to rape, we need an aversion to nonconsensual sex that is strong enough to override and deflect a very strong natural urge to procreate. It takes a great deal of negative reinforcement to effect this change, and thus a great deal of punishment and condemnation against those who commit rape.
In contrast, on the issue of telling the truth, we are all well aware of instances in which lies have some benefit. The practical joke, the surprise, the "white lie" are all instances of lying with few ill -- and some good -- consequences. We have no cause to create an overwhelming aversion to concealing the truth, so the punishment for lying is less severe.
Though we put little emphasis on deception in general, we do seek to promote an aversion to interfering with police officials in finding a criminal -- because of the harm that the criminal will do. So, 'interfering with a police investigation' gets a harsher penalty than lying in general, as does perjury and breaking a promise.
So, this becomes our criteria for proportionality. The negative reinforcement is proportional to the amount of work that it needs to do. The more widespread and stronger the desire or aversion being reinforced needs to be to fulfill the more and the stronger desires, the greater the wrong.
XIII. Internalism
One aspect of moral discourse that many people defend, which this theory cannot handle, is what philosophers call internalism. Internalism holds that if a person believes that "X is good", then that person has a corresponding desire for X.
This is certainly true for some types of 'good' even under this theory. All types of 'good' relate the object of evaluation to some set of desires, and some types of 'good' relate the object of evaluation to the desires of the speaker. For these types of 'good', a person who says 'X is good' literally does mean 'I like X'.
But there are other types of 'good' that relate objects of evaluation to desires not of the speaker, and moral 'good' is one of these. Moral 'good' relates objects of evaluation to desires that, if universal, would tend to fulfill all other desires regardless of whose they are. It would not be inconsistent for a person to say that X is (morally) good even though that person does not have the desires that would tend to fulfill the desires of others.
So, this theory is not consistent with internalism.
Yet, even here, the question remains as to whether the fault rests in this theory, or with internalism.
Within this theory, there are four strong associations between what a person claims is right and wrong and what they desire that allow us to explain the appearance of internalism, without embracing it.
First, there is the disposition that people have to believe that they want to believe. There may be a connection between a person's believe that "Other people have an obligation to do X" and the fact that he wants it to be the case that "Other people have an obligation to do X". However, this no more warrants the claim that there is a valid inference between the two than "I hope that there is a heaven" implies "Heaven exists."
Second, the objectives of morality on this theory is to use the tools of praise, blame, reward, and punishment to create in everybody the desires that they ought to have -- desires that would tend to fulfill the desires of others. To the degree that the moral project is successful, then, to that degree, people will, in fact, have a desire to do that which they ought to do. This does not imply that there is a logical link between what it is good for a person to do and what he desires, but there is a relationship that is far stronger than straight chance.
Third, the purpose of morality is to identify desires that "tend to fulfill all desires, regardless of whose they are." From this, it is quite likely, in most cases, that morality will identify desires that will, if universal, actually fulfill the desires of the speaker. So, again, there is more than random chance relating what is right and wrong and what will fulfill the desires of the speaker.
Fourth, even though the truth of a proposition "X is wrong" is independent of what the agent desires, the act of asserting "X is wrong" has to obey the formula behind all intentional actions -- it must fulfill the more and the stronger of the agent's desires given his beliefs. This is as true in ethics as it is in science where, for example, the proposition "The T-Rex was a scavenger" is true or false independent of what the paleontologist desires. But the fact that the paleontologist is engaging in the intentional act of publishing and defending this thesis depends very much on what the agent desires.
Between these, we have all that we need to cover an apparent link between what people say is right and wrong and what they desire to be right and wrong, without arguing that it is logically necessary that what is right and wrong in fact is what the person claiming it actually desires.
XIV. Conclusion
The test for these various moral theories depends on whether they can account for the way that moral language is actually used. There is a simple formula to follow in order to show the cracks in any theory of morality. Take the theory, and run it through the following steps:
1. You say that moral terms mean X.
2. A phrase using a moral term implies Y
3. That same phrase substituting X for the moral term does not imply Y
4 Therefore, the moral term cannot be logically identical to X.
Act utilitarianism fails this test. One of the tests that act utilitarianism fails rests in the fact that it implies that there are only two moral categories -- obligatory and prohibited. In fact, there are three moral categories -- obligatory, prohibited, and permissible.
Intention-based theories also fail this test. These theories cannot handle the wrongness of negligence and recklessness, where there is no malice or 'bad intention' motivating the action. They also cannot handle the case of the person doing 'the right thing for the wrong reason'.
Common subjectivism also fails this test. Subjectivism does not allow for second- and third-person moral claims. Yet, morality is filled with second- and third-person moral claims. It also fails to account for the concepts of reform and hypocrisy, is inconsistent with the way we learn moral terms as children, cannot account for the phenomenon of debate, explain the types of evidence people use in those debates, or handle the fact that people generally, when presented with a moral statement, will assert that it is 'true' or 'false'. It cannot handle the fact that a person can wave an obligation (such as an obligation to pay a debt) by simply saying, "I don't want to do it." Nor is it consistent with the fact that wanting somebody else to do something is not sufficient to create an obligation for others to do that thing.
With respect to this theory, I would like to challenge anybody who might read this to identify the break between the implications of this theory and what is true of our moral practices. It can handle negligence and recklessness, mens rea, the concept of excuse, the counter-examples to act utilitarianism, and 'ought' implies 'can', all without postulating 'queer' entities such as intrinsic prescriptivity or free will.
In a related article, I have argued that it can handle the is/ought distinction, the naturalistic fallacy, and can be incorporated into a 'unified theory of value' that explains all value-laden terms, such as 'illness', 'injury', 'usefulness', and 'beauty'.
A Mackian error theory, which holds that our moral language contains an assumption of 'intrinsic prescriptivity', which is false, has one significant hurdle to overcome. Philosophers of language recognize an important principle of translation called the Principle of Charity. This principle states that, of all the possible translations of an utterance in the language, the translation that maximizes its truth is the best.
This principle works like Ockham's Razor does in science. With Occham's Razor, we are not to postulate entities beyond those strictly necessary to explain the phenomena in question. This does not mean that we cannot postulate any entities. However, we must somehow be forced into doing so by the phenomena that we are trying to explain. There must be no simpler explanation available.
In the philosophy of language, we are to assume that statements are true. This does not mean that we must interpret every statement in a language as true. It only means that the complexity of categorizing some statements as false is something that we must be forced into, because no more truth-maximizing interpretation is available.
Our understanding of moral claims may force us into an error theory -- a theory in which a false assumption that intrinsic prescriptivity exists that is written into every moral claim. But it is something that we have to be forced into. If we have an alternative that explains the phenomena associated with moral claims that does not require this false assumption, that alternative wins.
We have such an alternative. The four criteria listed at the start of this paper fully account for a substantial portion of our use of moral terms. Furthermore, it does not presuppose any type of 'intrinsic prescriptivity' -- it contains no error. By application of the Principle of Charity, I claim that it is Mackie’s theory that is in error.