Desire utilitarianism is a theory about morality.

There are many theories about morality. Non-cognitivism claims that moral sentences are not true or false at all, but are more like grunts of approval and disapproval. Preference utilitarianism claims that preference satisfaction is the ultimate good, and should be maximized. Error theory argues that although moral utterances make factual claims, all such claims are untrue because the moral values to which they refer to do not exist.

There are many other theories about morality. Desire utilitarianism is one of them – the one that I currently think is more plausible than all the others.

Some people prefer to call the theory desirism because the term ‘desire utilitarianism’ leads people to (incorrectly) assume that desire utilitarianism is similar to other utilitarian theories, like preference utilitarianism. Also, ‘desirism’ is shorter to type and say. For these reasons, I will refer to desire utilitarianism by its alternate name, desirism.

On this page, Alonzo Fyfe shares why he named the theory “desire utilitarianism.”

Alonzo Fyfe formulated desire utilitarianism during his 12 years of formal education. His story is recorded in an online book, Desire Utilitarianism: An Atheist’s Quest for Moral Truth. Since then, Fyfe has continued to refine and correct the theory. He has also explored its implications for questions of applied ethics.

Fyfe first promoted his ideas on his website, AlonzoFyfe.com, and then on his blog, Atheist Ethicist, and then in his book, A Better Place: Essays on Desire Utilitarianism. He continues to develop and defend his theory on his blog. Fyfe has also given several interviews about his theory: 1, 2, 3.

A few other writers have been partially or fully persuaded by his views, including faithlessgod (who suggested the term ‘desirism’) and myself.

Not yet. Most major philosophers are not even aware of the theory, because Fyfe has not published his work in any peer-reviewed journals. Fyfe says this is because he doesn’t want to spend his life in ivory towers – he wants to bring his ideas directly to the people.

J.C.C. Smart has seen Fyfe’s work, and said something like “This sounds about right, but I still think there’s something with the other view,” but did not explain what that something was. Peter Singer has also seen it, and basically said, “It’s been done before,” without saying where it’s been done before or – more importantly – what’s wrong with it such that he should continue advocating a different ethical theory instead. Neither philosopher was able to show what might be wrong with desirism.

The primary sources on desirism are Fyfe’s book, A Better Place, and his blog, Atheist Ethicist. Also see Fyfe’s online book, Desire Utilitarianism: An Atheist’s Quest for Moral Truth, which recounts how Fyfe developed the theory. Older articles can be found on Fyfe’s older website.

Quick summaries of desirism have been attempted by Fyfe, Richard Chappel, and faithlessgod.

I have written about desirism, mostly on this page and in my short ebook, What is Morality?, but also on other pages. I also made an index of faithlessgod’s writings on desirism.

Also see Alonzo’s podcast interview for The Infidel Guy, and his two interviews for Conversations from the Pale Blue Dot (one, two).

Fyfe has also participated in two formal debates about desirism: one, two.

After losing my faith in Christianity, I could find no arguments for moral realism that weren’t just as bad as the arguments for theism that I had rejected. So, I became a moral error theorist – I did not believe there were any true moral facts.

Later, desirism was the first moral theory I found that (1) referred only to things that existed, and (2) was logically consistent. Because of this, it’s the most plausible theory of moral realism I know of.

But I’m not sure desirism is right. If it’s wrong, I’ll revert to error theory. Here are my biggest concerns about desirism:

1. Why make desires the primary object of evaluation?
2. It may turn out that desires do not exist in any form usable by moral realism.
3. Is the notion of what qualifies as a “reason for action” clear enough to say that desires are the only kind that exist?
4. Desirism is a consequentialist theory. At what point in time do we evaluate the consequences of a desire to determine its goodness? Ten years from now? Twenty? A hundred? A million?
5. Are desires really persistent enough for desirism to work? When I’m unconscious, do I really have a desire to not be killed? What about desires that change with my daily appetites?

There are many ethical theories. You should choose among them the same way you choose a scientific theory. You pick the one that is most likely to be true.

It doesn’t matter if you like the idea of wave-particle duality, or whether it is useful to you in making decisions. What matters here is whether or not wave-particle duality is true. You choose the theory that is true.

You should choose desirism because it is more probably true than other moral theories.

Maybe. But wouldn’t it be weirder if I defended a moral theory despite thinking it was inferior to another theory? Obviously, I defend desirism because it’s the best theory of morality I know of. Otherwise, I would be defending a different moral theory.

Other moral theories postulate such exotic entities as non-natural properties, intrinsic values, hypothetical social contracts, and gods. Compared to these theories, desirism is remarkably simple. It derives morality from simple natural facts: desires, states of affairs, the effects of desires on other desires, etc. And it does not require a special epistemology but depends on a method we already know to work well: science.

But if desirism is so simple, why did nobody think of it until recently?

I think it’s because everyone was looking for the wrong thing. Most people have been convinced that morality is transcendent, intrinsic, or even literally magical. Many others thought that moral facts had to be intrinsically motivating, or “binding” in some inescapable way, or independent of desires. So, philosophers spent thousands of years looking for something that did not exist.

This is rather like the search for the “soul” – that thing that causes our personality and intentions. People were convinced that the soul was non-physical, that it survived the death of the body, and so on. So they looked for thousands of years and could not find it. Now we know the soul is simply the mind, which is itself a function of the brain, which is itself a system of neuron firings. The soul is wholly physical, and it does not survive the death of the body. But we finally found it, and we are beginning to understand it scientifically.

To be added…

Desirism is both a meta-ethical theory (about what moral terms mean) and a normative theory (about how to decide what is right and wrong).

Desirism is:

• cognitivist – ethical sentences like “murder is wrong” can be true or false; they assert a proposition.
• objectivist – ethical sentences refer to facts, not opinions.
• realist – some ethical sentences are true; they correspond to reality.
• naturalist – moral facts reduce to non-moral facts about the world.
• gnostic – many ethical sentences can be known to be true; moral knowledge is possible.
• consequentialist – the goodness and badness of something (of a desire, in the case of desirism) is determined by its consequences.

Okay, so what does desirism claim?

First, desirism claims that when people make moral claims, like “Killing is wrong,” they seem to be claiming that it’s an objective fact that there are reasons for action against killing. All moral talk is about reasons for action. Reasons for action to feed the poor and show kindness to others. Reasons for action to not rape and not murder.

Second, desirism claims that desires are the only reasons for action that exist. Intrinsic values, categorical imperatives, ideal observers, divine commands, social contracts, and other foundations for moral theories do not exist.

Third, desirism claims that moral value exists as a relation between desires and states of affairs.

Fourth, desirism claims that desires themselves are the primary objects of moral evaluation. A good desire is one that tends to fulfill other desires. A bad desire is one tends to thwart other desires. Thus, a right act is one that a person with good desires would perform, and a bad act is one that a person with good desires would not perform. A good law is one that a person with good desires would enact, and a bad law is one he would not enact. And so on.

Thus, morality is the practice of shaping malleable desires: promoting desires that tend to fulfill other desires, and discouraging desires that tend to thwart other desires. Another way to think about this is to say that morality is about achieving a harmony of desires. (But not because desire fulfillment or a harmony of desires have intrinsic value. Intrinsic value does not exist.)

Here is a longer summary of desirism, by Alonzo Fyfe.

To be added…

To be added…

To be added…

To be added…

To be added…

Desirism makes a huge number of empirically falsifiable claims. For example, desirism would be falsified if it could be shown:

• that non-desire reasons for action exist, for example intrinsic values or categorical imperatives
• that desires do not exist
• that beliefs do not exist
• that agents do not form intentions from their desires, given their beliefs
• that no desires are malleable
• that desires do not tend to fulfill or thwart other desires
• that people do not generally have reason to promote desires that tend to fulfill other desires, or people do not generally have reason to inhibit desires that tend to thwart other desires
• that methods like praise, reward, condemnation, and punishment do not affect malleable desires

But desirism is also a meta-ethical theory, and so it also deals with semantic issues. Regarding these issues, desirism would be defeated if it could be shown that there is another moral theory that fits better with the broad usage of moral terms and makes only true claims.

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

This view is called moral non-cognitivism. See {6.01}.

I happen to agree that if you eliminated all beings with desires, value would not exist. But this is a poor argument against the existence of objective values.

If planets didn’t exist, then moons wouldn’t exist. Does this mean that moons do not objectively exist? Of course not. Because the fact of the matter is that planets do exist, and so do moons, and we can research their properties and relations and make objectively true claims about them. The same is true of values.

Yes, moral philosophers use the word “objective” to mean “mind-independent.” Desires are mind-dependent, so desirism is not objective in this sense.

But moral philosophers do not seem to have realized that this definition of “objective” has curious consequences. For example, the recent AFP headline “Forty injured in US prison riot” would not be considered “objective.”

Why? Because “injured” is a value-laden term. It is a mind-dependent term. Injury refers to damage that is not desired. If you ask somebody to pierce your ear for an earring, that is damage but not injury. If a robber grabs you and cuts your ear to threaten you into submission, that is damage and also injury, because it was undesired. So injury is a value-laden and mind-dependent term. So the AFP headline “Forty injured in US prison riot” is not “objective,” as moral philosophers use the term.

So we need to be clearer about how we use our terms. Fyfe suggests three tiers for the objective/subjective distinction:

(1) Broad objectivism, narrow subjectivism: If a statement has truth value, it is objective. “I like ice cream” is objective, because anybody who denies it is objectively wrong. I do like ice cream, and that’s a fact. A subjective utterance might be “Ewwwww… cockroaches!”

(2) Middle objectivism, middle subjectivism: If I’m reporting my own psychological states, that’s subjective. If I’m reporting on somebody else’s psychological states, that’s objective. But this has a curious consequence. If I say “Luke likes ice cream,” that is subjective. If you say “Luke likes ice cream,” that is objective. The exact same proposition can be either an objective fact or a subjective opinion, depending on who speaks it. So here we are not classifying propositions as subjective or objective. Instead, we are classifying speech utterances. There is also the problem of saying “We like ice cream.” Is this subjective or objective or a little of both?

(3) Narrow objectivism, broad subjectivism: The philosopher’s definition: any statement about psychological states is not objective but subjective. So,”5 > 3″ and “The earth is round” are objective, but “I like ice cream” and “Forty people were injured in a US prison riot” and “There is a new outbreak of disease in Kenya” are subjective (injury and disease are value-laden terms that refer to psychological states).

With these distinctions in place, we can say that:

• Desirism is objective(1), since it is entirely factual.
• Desirism is mostly objective(2), since it refers to all desires, of which the speaker’s desires make up a negligible proportion.
• Desirism is not objective(3), since it refers to desires, which are psychological states.

What people often mean when they say something is “objective, not subjective” is that it is “factual, not a matter of opinion.” Desirism is factual, not a matter of opinion.

Throughout history, our intuitions have been dead wrong about damn near everything. It’s no surprise they were wrong about morality. The fact is that intrinsic value, categorical imperatives, and the commands of God – the traditional foundations of morality – do not exist. So nearly all moral theory is in serious error.

But that doesn’t have to be the end of the story. Water was once universally considered to be an element. Now we know it’s not, but that doesn’t mean “Water does not exist.” Atoms were once universally held to be without parts – indeed, that’s what “atom” means. We’ve discovered that’s not true, but that’s doesn’t mean atoms do not exist. Likewise, morality was once almost universally held to have a transcendent quality. Does that mean naturalist moral philosophers are not even talking about morality? I think not. Our definitions of things change as we learn more about the real world.

I think desirism is a true theory about desires and certain types of value relationships. But I think it’s worth calling it a moral theory about moral values because:

1. Desirism accounts for the three categories of moral action: obligatory, forbidden, and permitted.
2. Desirism accounts for mens rea or “guilty mind.”
3. Desirism accounts for actions “above and beyond the call of duty.”
4. Desirism is a universal consideration of reasons for action, not just a prudential or institutional consideration.
5. Desirism accounts for the wrongness of negligence.
6. Desirism accounts for moral dilemmas.
7. Desirism explains why praise, condemnation, reward, and punishment are core components of morality.
8. Desirism explains what subjectivists get right about morality: that moral value depends on desires.
9. Desirism also explains what objectivists get right about morality: that moral value is independent of human opinion, and individuals and societies can be wrong about morality.

For these reasons, I think it is sensible to talk about desirism not just as a theory about desires and their relationships to states of affairs and other desires, but a theory about morality and moral value.

Desirism says you ought to “do what a person with good desires would do.” But there probably isn’t any such person with only good desires. He is a hypothetical agent.

But one of my objections to social contract theory is that the social contract is hypothetical. So how can I defend a theory that refers to hypothetical good agents?

My answer is that the “good agent” is a metaphor. When I say, “do what a person with good desires would do,” I mean “act on good desires.” The reason I conjure a hypothetical agent is that a concrete image sticks in human minds better than abstract ideas.

Now, contractarians will say the “social contract” is also a metaphor. But for what? Nothing that exists, as far as I can tell. In contrast, “act as a person with good desires would act” is a metaphor for something in the real world: namely, acting on good desires.

Here’s Fyfe:

It is widely assumed that something can either present a fact (e.g., “the earth is 4.5 billion years old”), or a non-fact value (e.g., “that is a very good story”). Facts are said to be objective while values are subjective. Nothing can be both a fact and a value, and nothing can be both objective and subjective, or so it is thought.

This looks good on the surface, until you start to ask some questions about what this really means.

For example, even though people speak of values as referring to some type of subjective non-fact entity, those same values are said to be able to influence things in the real world. They cause us to behave one way or another, or to choose one thing over another. These non-fact values somehow cause the world to be different than it would have otherwise been.

How can a non-fact entity influence the flow of matter through the physical universe?

Values are facts. They are part of the physical universe. Specifically, they are relationships between desires and states of affairs. A “good” story is one that fulfills common desires about what we want a story to be like. Both stories and desires exist in the physical world, as do the relationships between them.

Values are facts. I value ice cream. That is a fact. The subjectivist says that if values were facts, that would require us all to like the same thing. This is nonsense. Height is an objective fact but this does not require that we all be the same height. Location is an objective fact even though it is impossible that even just two of us have the same location, let alone everybody.

Definitions are not important. Language is an invention. Changing our words does not change what is true of the world.

Consider the propositions at the core of desirism:

(1) Desires exist.

(2) Desires are the only reasons for action that exist.

(3) Desires are propositional attitudes.

(4) People seek to realize states of affairs in which the propositions that are the objects of their desires are true.

(5) People act to realize states of affairs in which the propositions that are the object of their desires are true, given their beliefs – meaning that false or incomplete beliefs may thwart their desires.

(6) Some desires are malleable.

(7) Desires can, to different degrees, tend to fulfill or thwart other desires. That is, they can contribute to realizing the propositions that are the objects of other desires true, or contribute to preventing the realization of those propositions.

(8) To the degree that a malleable desire tends to fulfill other desires, to that degree people generally have reason to promote or encourage the formation and strength of that desire. To the degree that a malleable desire tends to thwart other desires, to that degree people generally have reason to inhibit or discourage the formation and strength of that desire.

(9) The tools for promoting or inhibiting desires include praise, condemnation, reward, and punishment.

These are true propositions, and they make no reference to morality, or even “value.” If you decide that the word “morality” refers to intrinsic value or God’s commands or categorical imperatives or imaginary social contracts, then you are welcome to use the word “morality” in that way, but you will be talking about things that do not exist.

Value terms are about reasons for action, and moral value terms are usually spoken of in the context of a universal consideration of reasons for action (not just prudential reasons for action). Desirism is a theory about reasons for action, and especially about a universal consideration of reasons for action. That’s why I think it makes sense to consider it a theory about morality.

Also, desirism accounts for many common features of moral theory, which is another reason to consider it a theory about morality. See question {3.04}.

Also see: one, two, three.

To be added…

To be added…

Alonzo Fyfe thinks that:

…the greatest objection to {desirism} says that there are no desires.

{This} is because we have too many questions to answer about what a desire is and who has them. Some people predict that when we come up with a theory of intentional action that can handle all of these questions, that this theory will be so radically different from the theory of beliefs and desires that the latter will face elimination, {and this would falsify desirism.}

One of the questions is: At what point do creatures actually acquire desires. Let’s say that plants and amoeba have no desires. Do worms? Oysters? Lobsters? Sharks? Insects? Frogs? Snakes? Cattle? It seems difficult to draw a line somewhere.

Do thermostats have desires? Is it reasonable to explain what a thermostat does as having a desire for the temperature of the room to be at 72 degrees, a belief that it is at 68 degrees, and thus it forms the intentional act of heating the room.

…Another relevant question to ask is: Under what conditions do machines acquire desires?

We must wait for neuroscientists to answer these questions. Until then, there are no solid competitors to the standard belief-desire model of intentional action (BDI theory) on which desirism is founded. BDI theory can be used to successfully predict physical phenomena in the universe (in particular, the behavior of others), and it is the best theory we have for doing so.

For a very accessible introduction to this problem, see Daniel Dennett’s Kinds of Minds.

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

According to desirism, a morally good desire is one that tends to fulfill more and stronger desires than it tends to thwart. But this would require (1) counting up all the desires in the world, (2) measuring their strengths, and (3) mapping all their causal relationships. And this, we cannot do. So even if desirism is true, it is unusable, right?

No. We often make judgments even though we do not have all the information. In fact, all our judgments are that way. We even make complex judgments we are fairly certain about while lacking much information. We can use our reasoning and make “educated guesses” and also estimate how prone to error each educated guess is.

As for (1) counting desires, we can estimate these. For example, it is quite clear that the detonation of an atomic bomb in Tokyo would thwart more desires than a serial killer in Siberia. And we can use evidence and argument to approach accurate estimates in more difficult cases. But there is a need for moral experts here. We need people to weigh the arguments and study the data and take surveys and count desires. Those researching applied ethics under a consequentialist framework have been doing basically this for several decades already.

We can also (2) measure the strengths of desires. While we don’t yet know what a desire looks like in the brain or how to measure it directly, neuroscience will eventually tell us. In the meantime, the crude tools we possess will have to do. People could tell when one thing was hotter than another before they had thermometers to give precise numbers. And we can tell that Tom’s desire for warm shelter is stronger than his desire for ice cream. You might be surprised to learn how precisely we can currently measure the strengths of desires through methods such as an economist’s willingness to pay algorithms.

Finally, (3) mapping the causal relationships of desires may be the most difficult. This always requires a great deal of theorizing and argument and counter-argument. But as more evidence is accumulated and more prejudices are left behind, we may yet be able to understand how certain desires cause the fulfillment or thwarting of other desires. Many cases are not much disputed – for example, the desire to rape pretty clearly causes a great many strong desires to be thwarted around the world each year.

As explained in section {3.03}, there are many definitions of the term “objective.” Desirism is objective according to some definitions, and not objective according to others.

But desirism is objective even given the definition above: “not influenced by personal feelings, interpretations, or prejudice; based on facts; unbiased.” According to desirism, the terms “good” and “evil” refer to desires that tend to fulfill or thwart other desires, respectively. And the truth of whether a desire tends to fulfill or thwart other desires is based on the facts of the matter, and is not influenced by personal feelings, interpretations, prejudice, or bias.

For example, it may be my personal feeling that black people should not enjoy the same human rights that white people do. Even if everyone was racist – including blacks themselves – it would still be a fact that this racism thwarts a great many desires of blacks. Even if blacks believed their racism did not thwart their own desires, it would still be the case that it thwarted a great many of their desires. Even if the blacks interpreted their circumstance to say that this culture of racism fulfilled their own desires, and even if blacks themselves were prejudiced and biased against their own kind, it would still be a fact that racism thwarts a great many desires of blacks. See {3.22}.

Sometimes I used the Knob Metaphor to illustrate what I mean by desires that “tend to fulfill” or “tend to thwart” other desires. Imagine a knob that controls the strength of a given desire within a population. If we turn the knob to the left, the strength of that desire decreases in the population. If we turn the knob to the right, the strength of that desire increases. To say that a desire is good is to say that if we turn its knob to the right, more desires will be fulfilled (or fewer desires thwarted). To say that a desire is bad is to say that if we turn its knob to the right, more desires will be thwarted (or fewer desires fulfilled).

Since each desire is a reason for action, there are more reasons for action to increase the suffusion of a good desire in a population than there are to decrease it. (That is what it means to say it is a good desire.)

So let’s consider the knob for racism. To keep things simple, let’s define “racism” as a desire to deny blacks many privileges that whites enjoy, for example freedom to travel, freedom to choose one’s labor, freedom to vote, freedom to defend oneself in court, and freedom from physical abuse.

Is racism a good desire or a bad desire, according to desirism?

Racism is a bad desire because it tends to thwart other desires. That is, if we suffuse a society with racism by “turning its knob to the right” (or, in the real world, by praising and rewarding racism), then this racism will cause a great many desires to be thwarted. Sure, the desires of the racists will be fulfilled, but a great many desires of the blacks will be thwarted. Their desires for freedom to travel, freedom to choose one’s labor, freedom to vote, freedom to defend oneself in court, and freedom from physical abuse will be thwarted.

Now, suppose we deplete a society of racism by “turning its knob to the left” (or, in the real world, by condemning and punishing racism). In that case, fewer racist desires will be thwarted, because there will be fewer racist desires. And also, the desires of blacks will be better fulfilled – for example their desires for freedom of travel, freedom to choose one’s labor, and so on.

So there are more reasons for action to deplete a society of racism than to suffuse it with the same. Thus, racism is a bad desire.

Note that this is true even if we turn the knob all the way to the right, such that everyone is racist, even blacks. Why? It may help to call to mind comedian Dave Chappelle’s character Clayton Bigsby, a blind white supremacist who doesn’t know he’s black, and remains a racist even when he learns the truth. Even if every black person on earth desired that blacks be denied freedom of travel, freedom to choose one’s labor, and so on – and even if they believed this denial of freedoms actually fulfilled their own desires – it would still be a fact that this racist denial of freedoms greatly thwarted the desires of blacks.

The only way this could not be the case is if we also changed the desires of blacks such that they no longer desired freedom of travel, freedom to choose one’s labor, freedom to vote, freedom to defend oneself in court, and freedom from physical abuse. But morality is concerned with malleable desires and many of these desires are not very malleable. (In contrast, racism is extremely malleable, having gone from near-total suffusion to a minority desire in many societies in just a few generations.) Moreover, at least one of these desires could not be reversed: the desire for freedom from physical abuse. Abuse is defined in terms of that to which you have an aversion, and it’s not possible to have a positive desire for something to which you have an aversion (a negative desire).

To be added…

According to desirism, the question “Should I do x?” can be answered by solving another problem: “Would a person with good desires do x?”

And what is a good desire? A good desire is one that tends to fulfill other desires.

An example: should I rape someone? No, because rape is caused by a desire to rape, and a desire to rape is a bad desire, because that desire tends to thwart, rather than fulfill, other desires.

One image that can help you make moral decisions is to imagine a set of knobs that control the number and strength of desires in a population. You can “turn a desire up” so that everyone strongly has that desire, or you can “turn a desire down” so that nobody has that desire.

If your turn the rape-desire knob to the right, then more desires will be thwarted. Either the desires of the victims will be thwarted, or the desires of the rapists to rape will be thwarted (if their potential victims evade them).

But if you turn the rape-desire knob all the way to the left, then no desires are thwarted. The desires of the potential victims are not thwarted, and the rapists have no desires to rape that are being thwarted.

But why not turn the knob which controls the aversion to rape, instead? Why not make it so that everybody has the desire to rape, but as it happens everybody enjoys being raped?

In the case of rape, this is impossible because rape is defined as “undesired forced sexual intercourse.” Once somebody has lost the aversion to forced intercourse, it is no longer called “rape.”

Turning desire-knobs is an important moral practice. You ought to promote desires that tend to fulfill other desires, and discourage desires that tend to thwart other desires.

But that is not all. As Fyfe writes,

What you ought to do is that which a person with good desires would do. But a person with good desires is not always going to be spending his time turning knobs on other desires. He has other concerns in addition to this one.

The father who reads to his child before bedtime has a desire that tends to fulfill other desires. But it is not a desire to “turn knobs” on other desires. Instead, it is a desire for the welfare of his child, and a love for the child that motivates him to spend time with the child. Because of these desires, he is likely to spend time with his child instead of performing some other activity that would qualify as “turning the knob” on other desires.

In fact, {it makes no sense to turn knobs all the time.} What would it mean to spend time promoting parental affection that would cause fathers to spend more time with their children, if one is not permitted to act on that desire and spend time with one’s children for no reason other than one desires to do so? The practice of promoting desires has to be made consistent with moral permission to act on those desires.

The reason we promote good desires is so that people will act on those good desires.

In summary: you ought to do what a person with good desires – desires that tend to fulfill other desires – would do.

For answers to what you should do in a given situation, see the section on applied ethics.

Fyfe writes:

A person with good desires has it made. When she asks the question, “What should I do?” the answer comes back, “Whatever you want to do.” The person with good desires is somebody who wants to find a cure for diabetes, or wants to prevent the suffering caused by malaria, or wants to teach children about the real world. This is their passion. In pursuing their passion, they also tend to fulfill the desires of others.

…It’s the rest of us who have to worry… about the gap between what we want to do and what we should do.

Our desires, especially when we’re young, are shaped by biology and other people. But as we grow, we gain some ability to choose our own desires. And we can choose to foster in ourselves good desires – desires that tend to fulfill other desires.

As a young man, I knew that intense desires for addictive drugs would probably thwart many of my future desires (for success, health, good relationships, etc.), and so I have avoided addictive drugs my whole life. To a great extent, I chose not to desire addictive drugs. As it happens, this was was probably a morally good aversion for me to develop, since addiction to drugs tends to thwart lots of desires, not just those of the addict.

Another way to choose good desires – to become a better person – is to acquire a good habit. Exercise is a good habit, and one can actually acquire a desire to exercise through various psychological techniques. For example, make exercise pleasurable by doing it with your friends or by rewarding yourself after a job well done.

To become a better person, study desirism to learn which desires are good and which desires are bad. Then, choose the desires in yourself you can most successfully change from bad to good, and change them.

Science has a lot to say about how you can change your desires. The most trustworthy collection of these techniques I have found so far is Psychological Self-Help by Clayton E. Tucker-Ladd.

Yes, it’s possible to accept desirism and theism without being inconsistent.

Many people believe in a god or gods without thinking they are the source of moral values. And if God existed, it could still be true that desires are the only reasons for action that exist. God’s desires would affect the moral landscape about as little as adding one more human and their desires into the world, unless God’s desires were as strong as those of a billion people combined (and it’s a bit hard to think of God so intensely desirous – like a horny teenager times a million).

And if God was anything like the violent, misogynistic despot of Judaism, Christianity, and Islam, then the people of earth would have very strong reasons to change God’s desires such that they fulfilled billions of desires rather than thwarting them.

To be added. For now, see here.

As far as we know, desires are a product of sufficiently complex brains, and many “higher” animals certainly seem to have desires. A plant probably does not have desires, since it lacks a brain.

But wherever desires are, there is no reason to exclude them from moral consideration. It is not the case that human desires have intrinsic value and animal desires do not. Intrinsic value does not exist.

The implications of desirism for our interaction with animals are complex, but it is a fair guess that our annual thwarting of billions of animal desires is morally evil. Desirism suggests that we ought to radically change how we interact with animals, but nobody is quite sure how. The philosophical work on this issue just hasn’t been done yet.

But here’s a start: one, two, three.

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

To be added…

Non-cognitivism is the view that moral sentences like “Murder is wrong” do not actually assert a proposition, but are in fact a kind of expressive utterance, like “Murder… yuck!” In this way, moral sentences are held to be like questions or commands, which cannot be true or false:

True or false: What is your favorite color?

True or false: Close the door.

The first problem with non-cognitivism is that it fails its most basic test. Hand somebody a list of questions with True and False checkboxes for each one and they will just be confused. Hand them a list of commands with True and False checkboxes and they will stare at you blankly. But hand them a list of moral sentences like “Murder is wrong” and “Stealing is good” and they will start checking boxes.

Let’s say somebody shouts “Murder is wrong!” and you ask them “Wait, do you just mean that you have a ‘yuck’ reaction to murder, or do you mean that murder has the property of ‘being morally wrong’?” They will choose the latter. They will tell you they meant that murder really is wrong, as a matter of fact.

Here is some further evidence² that moral sentences are meant to assert propositions:

1. Moral sentences are expressed in the indicative mood.
2. They can be transformed into questions.
3. They are embedded in propositional attitude contexts.
4. They are considered to have an objective, impersonal character.
5. Moral predicates (“good”) are transformed into abstract terms (“goodness”), suggesting they are meant to name properties.
6. Moral sentences are subject to debate that looks just like factual debate.
7. They appear as premises in arguments considered valid.

So non-cognitivism just isn’t true about the way 99% of people use moral sentences. It also suffers from other problems.

That’s why I favor a cognitivist theory. Moral sentences really do assert something. Now, it could be the case that these assertions always refer to something that does not exist (moral value), just as theological assertions always refer to something that does not exist (God), but it would still be the case that moral sentences assert something.

Moral relativism is the position that moral statements are not objectively factual, but instead are relative to individual and cultural concerns; there is no universal standard by which we can assess the truth of a moral claim.

But if moral statements are relative to something, does this mean they are not objective and factual? Certainly not. Consider the following statements:

1. I live in California.
2. I’m taller than most people.
3. Seven is greater than two.
4. The earth is 93 million miles from the sun.

Relationships and relative states are facts. They are objectively true or false.

So maybe relativism is just a descriptive claim about human societies: that different cultures consider different things to be right and wrong. That’s obviously true. But it doesn’t mean all those cultures are right. Different cultures believe different things about the origins of the universe, too, but there is an objective fact of the matter, and some cultures believe something closer to the truth than others.

So maybe relativism is just moral nihilism, the view that objective moral values do not exist. That’s a reasonable default position. After all, if someone claims that objective moral values do exist, he ought to show some evidence for it. In this case, the problem with moral relativism is that I do provide good evidence that objective moral values exist.

There are many types of moral subjectivism.

One type claims that moral sentences like “Murder is wrong” are statements not about some fact “out there in the world” but about the speaker’s opinion. “Murder is wrong” really means “I have the opinion that murder is wrong.”

Let’s test this theory. Ask people what they think about murder. They will say, “Murder is wrong.” Then ask them if they really just mean that they have the opinion that murder is wrong. Most of them will say, “No, it’s not just my opinion. Murder is wrong!”

Now these people may be incorrect that murder is wrong. But it is false to claim that they’re just saying something about their own opinion. So that kind of subjectivism fails its most obvious test.

Another type of subjectivism claims that there are facts about people’s moral opinions, but that there are no facts about whether murder really is wrong or whether charity really is right. And this is a sensible default position. Someone who wants to say that certain things really are wrong should give some evidence that “rightness” and “wrongness” exist. The problem with this type of subjectivism is that I do provide good evidence that these properties exist.

A final type of subjectivism claims that moral facts exist, but they are relative to states of mind. Murder is wrong because God’s mind disapproves of it, or because an ideal observer would disapprove of it, or because it would thwart lots of desires. Desirism agrees with this type of subjectivism, because desirism states that moral values are relative to desires, which are brain states.

But that’s not usually what people mean by “subjectivism.” Subjectivists often try to say that someone cannot be wrong about morality. We can each invent our own morality, because everything is “subjective.” This is where desirism disagrees with subjectivism. Desirism claims that moral values exist as relations between desires and states of affairs. And it is certainly possible to be wrong about those relations. For example, it is false to say that the desire to rape tends to fulfill more and stronger desires than it fulfills.

Also see: one, two, three, four.

Error theory argues that (1) morality is centrally committed to X, and (2) X does not exist.

For example, Mackie argued³ that morality is centrally committed to intrinsic prescriptivity, and that intrinsic prescriptivity does not exist. I agree that intrinsic prescriptivity does not exist, but I don’t think morality is centrally committed to it.

Another example. Joyce argues that morality is committed to the queer notion of moral inescapability, that people are bound by morality even if they make no moral judgments at all. He then argues that moral inescapability does not exist. Depending on what it means to be “bound” by morality, I might or might not agree that moral inescapability is false. But I do not agree that morality is committed to moral inescapability.

In The Error in Error Theory, Finlay writes that:

For centuries, water was almost universally assumed to be an element rather than the compound it actually is. But we do not take seriously the analogous proposal that until Lavoisier, thought and talk about ‘water’ was systematically false because there was no such stuff–and we would not even if the assumption had been universal.

Fyfe offers another example, atoms:

…the fact that the word ‘atom’ means ‘without parts’ {plus the fact that atoms were once universally thought to be without parts} does not, in fact, force us to view the universe is one in which these smallest pieces of gold, carbon, lead, and oxygen are not made up of parts. Whether or not these smallest pieces of each element can be further split into parts is not a matter for the linguist to decide. And when it is discovered that these smallest pieces can be divided into parts, we simply need to revise our language.

And I’ll give one more example. It seems plausible that at one time most moral systems were intimately tied to the idea of transcendent obligations – obligations to a god, for example. At that time it would have seemed sensible to argue that morality is “centrally committed” to transcendent obligation. Does this mean that ethical naturalists – whether their theories are correct or not – are not even talking about morality, because their theories do not include the transcendent? I think not.

Desirism agrees with error theory that the vast majority of our moral discourse – about God’s commands, intrinsic values, categorical imperatives, and so on – is in error. But desirism says, “Just as with water and atoms, we’ve discovered what morality really refers to, and we don’t have to abandon moral terms just because they don’t actually refer to God’s commands or intrinsic values.”

Furthermore, desirism accounts for a great many things that appear regularly in moral discourse: the three categories of moral action (obligatory, forbidden, permissible), actions “above and beyond the call of duty,” mens rea or “guilty mind,” moral dilemmas, and so on. That’s why it makes sense to keep using moral terms even though they don’t actually refer to God’s commands or intrinsic values or categorical imperatives – just as water isn’t actually an element, and atoms aren’t actually indivisible.

To be added…

Common utilitarian ethical theories claim that X has intrinsic value, and thus ought to be maximized. Bentham argued that pleasure has intrinsic value, and should be maximized. Mill argued that happiness has intrinsic value. Hare argued that preference satisfaction is what has intrinsic value. These theories promote “the greatest good for the greatest number.”

Many people object that these utilitarian theories fail because they give surprising results. For example, they fail the 1000 Sadists Problem. Imagine a world in which 1000 sadists want to rape a child. The greatest good for the greatest number is for the sadists to rape the child. The child will endure great suffering, but her suffering is not as great as the pleasure / happiness / preference satisfaction experienced by the 1000 sadists. The objection goes like this: Surely we all know that it would be wrong for the sadists to rape the child, so therefore utilitarianism cannot be right.

I don’t think this is a good objection. The objection assumes that we are born with some kind of highly accurate morality-detector, or that our pre-philosophical moral intuitions are usually correct. But there is no evidence for either of these assumptions. So, the fact that most of us feel that something is wrong provides no evidence that it is wrong.

So, I don’t think the 1000 sadists problem, stated this way, is a real problem. (But, desirism “solves” this problem anyway. See {3.15}.)

Instead, I think common utilitarian theories fail because they depend on the existence of intrinsic values, and intrinsic values do not exist. We can’t explain any facts about the world better by postulating intrinsic values than by postulating less exotic entities, for example desires. To quote Alonzo Fyfe: “By the powers vested in me by Occam’s Razor, intrinsic values do not exist.”

To be added…

Virtue ethics says that right acts are those done out of a virtuous character. But what is virtuous? Virtue is a habit or quality that allows a something to fulfill its purpose. The virtues for an axe are sharpness and durability. The virtues for a hunting dog are a sensitive nose, stealth, and obedience.

But notice that the virtues of a hunting dog are assigned by someone outside the dog: its master. So are the virtues for an axe. An axe’s purpose is to chop wood only because we gave it that purpose. If a hurricane lodged a sharp, slender piece of rock into the end of a short stick, that stick-and-stone would have no intrinsic purpose. It would only have a purpose if somebody came along and decided to use it to chop wood, or pound in a tent peg, or prop up a stool, or set a trap to catch a mouse.

If humans have a purpose, it must be intrinsic to humans or else assigned from the outside, for example by God. The first option fails because God does not exist. The second option fails because intrinsic purpose does not exist. Nobody has ever shown me evidence that intrinsic purpose exists. (And if you think evolution provides intrinsic purpose, please see question {6.10}.)

Actually, desirism can be considered a kind of virtue ethics, but with no commitment to intrinsic purpose or externally-assigned purpose, and a foundation for virtue in consequentialism. See here.

Many people believe that:

1. All action is aimed at the agent’s happiness.
2. Therefore, happiness is the sole good.

Ignoring the legitimacy of inference from (1) to (2) for now, let me explain why (1) is false.

All action is not aimed at the agent’s happiness.

Let’s do a thought experiment. A mad scientist has captured you and a close friend you care about deeply. He gives you two options, and you must choose one:

1. Your friend will have his memory erased, but will be set free in good health. He will be given reason to believe that you are safe and happy. Your memory will also be erased, and you’ll be given reason to believe that your friend is being tortured endlessly. You will hear his screams. But you will be fed and cared for and given as much freedom as possible.
2. Your friend will be tortured endlessly on a remote island. Your memory will be erased, and you will be given reason to believe your friend is healthy and happy. You will be fed and cared for and given as much freedom as possible.

Many people, perhaps most people, will choose Option 1 – even though this does not maximize their own happiness. (There are also other examples, ones we encounter in real life.) Thus, it is false to say that people always aim toward their own happiness.

Happiness is not the only project that people aim at. We have other projects.

But it is true that we always aim at what we desire. We may desire our own happiness, but we sometimes desire other things as well. BDI theory, the most widely-accepted model of intentional action, claims that:

Belief + desire -> intention -> intentional action

Happiness theories claim something else, which is false:

Belief + desire that “I am happy” > intention -> intentional action

Happiness is not the only thing we desire, and it is not the only thing that motivates us toward intentional action.

Also see: one, two, three, four.

It’s probably true that many of our feelings about what is right and wrong (tasteful and distasteful) have evolved.

But how can we leap from “I get a feeling of distaste when I think about rape” to the claim “Rape is morally wrong”? A mere feeling justifies no such conclusion.

Some say that evolution programmed us with moral opinions in favor of cooperation and altruism because these enable the formation of a functional society.

That may be true, but there is no valid inference from “X is an evolved trait” to “X is morally good.” If men had evolved, like male lions, to kill our step-children upon taking a new mate, would this make killing step-children moral? Or, consider this: It seems we have evolved a disposition to seek dominance over others whenever possible. Does this make the pursuit (and use) of dominance over others moral? These are invalid inferences.

Another problem is that not all our moral sentiments are evolved. Many are socially programmed. Slavery went from near-universal acceptance to wide disdain in the space of three generations. That is not the work of gradual evolution.

Evolutionary ethics also suffers from a modified Euthyphro dilemma. The evolutionary ethicist says, “What is good is that which is loved by our genes.”

To which I respond with a paraphrase of Socrates: “Is it good because it is loved by our genes, or is it loved by our genes because it is good?” Fyfe explains the dilemma:

If it is good because it is loved by our genes, then anything that comes to be loved by the genes can become good. If humans, like lions, had a disposition to slaughter their step children, or to behead their mates and eat them, or to attack neighboring tribes and tear their members to bits (all of which occurs in the natural kingdom), then these things would be good. We could not brag that humans evolved a disposition to be moral because morality would then be whatever humans evolved a disposition to do.

If, instead, it is loved by our genes because it is good, then we have not yet answered the question of what goodness is.