Street 02: What are "Desires-as-Ends"?

This post is going to repeat some information I presented just a few days ago. I am repeating it because, in my revised revision of Sharon Street’s “Darwinian Dilemma for Realist Theories of Value,” this is where the information becomes relevant.

(And, for those who care about such things, I intend this Street series to make it into my thesis.)

In Part 01, I stated that I wanted to limit the application of Street’s argument to desires only - and, to be more specific, desires-as-ends only, excluding desires-as-means.

Note: I also objected to her assumption that value realism requires intrinsic value realism. I interpret her argument as targeting intrinsic value realism, allowing that values (including moral values) can be real even though they are not intrinsic properties.

Remember, I distinguished desires-as-ends from desires-as-means. I used the distinction between wanting to build a shelter so that one can stay warm in the winter, and wanting to stay warm for its own sake. “Desires-as-ends” refers to what we desire for its own sake.

Here, I want to state more precisely what I mean by “desires-as-ends”.

Desires - and beliefs - are propositional attitudes. They are mental states that we can express in the form:

[Agent] [attitude] that [proposition]

So, we have examples like:

Jim believes that Joan went to the movies.
Mira hopes that the weather remains good through the weekend.
Tully thinks that it is going to rain.
Zach loves that they will all be together for the holidays.

We can divide propositional attitudes into two main camps – beliefs and desires.

The primary difference between the two is the “direction of fit”. If Agent believes that P, and P is false, then Agent should change her belief. However, if Agent desires that P, and P is false, then Agent has a motivating reason to change the world.

By way of example, if Alice believes that she owns a red car, and her car is blue, then she should change her belief so that she believes that she owns a blue car. However, if Alice owns a red car and wants to own a blue car she has a motivating reason to buy a blue car, or get her car painted.

A “desire that P” is fulfilled if and only if the proposition “P” is made or kept true. In other word, the motivating reason is to make or keep the proposition true. Alice’s desire to own a blue car is fulfilled in any state of affairs where the proposition, “Alice owns a blue car” is made true – by buying a blue car, getting her car painted, or obtaining one as a gift.

If an agent has a “belief that P”, and P is true, then the belief is true. If an agent has a “desire that P” and P is true, then the desire is fulfilled.

This is fundamentally a Humean theory of motivation. It supports, for example, Hume’s thesis that:

Reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them.

By this I mean that desires pick out the ends or goals of intentional actions, and it is reason’s job to discover how to realize those ends.

Reason is not to be envied. It serves multiple masters with competing agendas, sometimes actively working to thwart other desires as the desire for health conspires with the desire to look good to thwart the desire to have more chocolate cake. And reason has to constantly be looking for ways to keep all of its masters happy.

On this Humean model, while beliefs can be true or false, desires cannot be. There are no “correct” or “incorrect” desires.

The value of Street’s argument, properly modified, is that it provides support for this part of the thesis. Her "Darwinian dilemma" provides a reason to accept the claim above - that there are no "correct" or "incorrect" desires-as-ends - is true. She shows that what we know about evolution supports the hypothesis that there is no external evaluative “truth” for desires to correspond to in the way that beliefs correspond to an external reality. The “ends” of our desires-as-ends do not have intrinsic value. They are simply those things, the desiring of which, caused our ancestors to have offspring who eventually had us.

In my next post, I will show that Sharon Street knew that she needed a distinction like the distinction I drew between desires and evaluative attitudes. However, she described the distinction she needed as between advanced evaluative attitudes and primitive, proto-evaluative attitudes. When I put desires-as-ends into the role of her proto-evaluative attitudes. They are not a precise fit. Claims she makes about these proto-evaluative attitudes are not true of desires-as-ends. However, I am going to counter that she over-intellectualizes these attitudes.

Then, I am going to draw a connection between Street's "evaluative attitudes" as she describes them and "desires-as-means". (I told you that these two sets belong together.) From this, I will show that Street's evaluative attitudes can be true or false and do correspond to an external reality. However, that external reality depends critically on desires-as-ends. Consequently, this will not be a refutation of Street's thesis. It will be a revision.

Street 01: Starting Over - Evolution and Desires-as-Ends

I am starting over.

I found my attempts to present Sharon Street's arguments in, "A Darwinian Dilemma for Realist Theories of Value" to be confusing.

Consequently, I devoted some time to coming up with a way to do a better job of presenting my position on her argument. This post is the first post in this starting-over series.

First, the major thesis:

Facts about the theory of evolution imply that there are no intrinsic values.

Street did not express her thesis this way. She expressed her argument as an objection to realism about values. Equating value realism with intrinsic value realism requires the false assumption that values must be intrinsic properties to be real. I don't think that this is right. There are a lot of properties in the real world that are not intrinsic properties and, I would argue, value is one of them. Street provides a strong argument against intrinsic value realism. However, her argument does not actually show that realist theories of value are mistaken simply because real values are not intrinsic values.

However, that is a debate for another time and place. Here, it is sufficient to note that I am applying Street's argument to the more limited subject of intrinsic value realism. You can find a more detailed writeup of my objections to Street's terminology in Moral Objectivity and Moral Realism.

Street's argument against intrinsic value realism is that evolutionary forces have shaped our evaluative attitudes. Intrinsic values, if they exist, must either be related to what produces evolutionary fitness in humans, or unrelated. The thesis that they are related to evolutionary fitness fails because scientists have been able to advance the theory of evolution quite well without adding “intrinsic value properties" to the explanation. The thesis that they are unrelated implies an unreasonable coincidence between what we have evolved to like and dislike and what has intrinsic value. Either way, there is no reason to believe that there are intrinsic value properties.

The first point where I want to add some refinements to Street’s thesis is to this concept of “evaluative attitudes” that Street says is under the influence of evolutionary pressure. She wrote:

Evaluative attitudes I understand to include states such as desires, attitudes of approval and disapproval, unreflective evaluative tendencies such as the tendency to experience X as counting in favor of or demanding Y, and consciously or unconsciously held evaluative judgements, such as judgements about what is a reason for what, about what one should or ought to do, about what is good, valuable, or worthwhile, about what is morally right or wrong, and so on.

I want to take “desires” out of this list and set the others aside and deal with them later. The thesis that I want to look at states that desires have been subjected to evolutionary pressure and, as a result, it is unreasonable to believe that what we want has intrinsic value.

The term “desires “ is ambiguous. We use it to refer to both that which we value as an end or for its own sake and what we value as a means to that end. We want a hammer so that we can use it to pound nails. We wish to pound nails because we wish to build a shelter. We wish to build a shelter so that we can stay warm in the winter. We wish to stay warm in the winter because . . . well, we just do.

The first three desires (wants, wishes) for the hammer, to pound nails, and the shelter are desires-as-means - we want something for the sake of something else. The last item on the list is a desires-as-ends. We want it for its own sake, and not for the sake of something else. The desire-as-end (comfort in winter) provides the motivation for all of the other actions. Take away the desire for comfort in winter, and you take away the reason to buy a hammer and use it to pound nails to build a shelter.

When I claim that desires are subject to evolutionary forces, I am interested only in desires-as-ends. I would like to separate these from desires-as-means and leave the latter type of desire in the same bin with the other "evaluative attitudes" discussed above. I will argue that they actually belong together.

In the mean time, the thesis that I want to defend using Street's argument is the thesis that desires-as-ends have been subject to evolutionary pressure and, consequently, are unlikely to be desires-as-ends for something that also, at the same time, has an intrinsic quality of ought-to-be-desired-ness.

Desires-as-ends include the aversion to pain, desire for sex, hunger, thirst, our preferences for particular foods, comfort (in terms of not too hot or not too cold), friendships, and the well-being of one's offspring.

However, it is not the case that all of our desires-as-ends have a genetic basis. Humans have a "plastic" mind - meaning that our interactions with our environment work to shape our beliefs and desires-as-ends, to change them. Yet, even this mechanism for acquiring new desires-as-ends and reshaping or refining existing desires-as-ends went through evolutionary refinement. Nature has built us to learn to want that which contributed to the genetic replication of our ancestors and to learn to want to avoid that which threatened their genetic replication.

So, the first part of this argument is to specify that I am only concerned with desires-as-ends and wanting to show that they are under evolutionary pressure.

The second step in my argument - tune in tomorrow - will be to specify more precisely what I mean by "desires-as-ends". On that account, desires-as-ends are incapable of being "true" or "correct" in the way that beliefs can be. In the posts that follow, Street's argument will provide a defense of that provision. Come back tomorrow for that part.

The third step in my argument - for those who like to think ahead - will be to show that Street is not adverse to a distinction like the one I am making. She recognizes the need to distinguish between complex evaluative judgments and the simpler "proto-judgments" that evolution could have acted upon. I am simply specifying that her "proto-judgments" are desires. But there are important differences between desires-as-ends and Street's proto-judgments, and we will need to look at those.

Desires vs. Evaluative Judgments: Critique of Darwinian Dilemma - Part 02

This is a Critique of Sharon Street's "Darwinian Dilemma for Moral Realism" - Part 02

You can read Critique of Sharon Street's "Darwinian Dilemma for Moral Realism" - Part 01, but it should not be needed to understand the argument below.

I know that the fact that something is not biologically useful, or could have been done much more efficiently, is not a reason to object that there is such a thing. However, when we are considering two possible explanations for a set of events, I think that there is reason to suggest that the simpler and easier explanation has merit simply in virtue of its being simpler and easier.

Sharon Street argues that evolution has had an influence on our "evaluative judgments" - causing us to "judge" as good or ought-to-be-done that which also happens to promote human genetic replication. Clearly, a being that saw "ought-to-be-done-ness" in killing its offspring will not have offspring around today to carry on that tendency.

But my question is: Why "evaluative judgements"?

Nature does an excellent job of getting us to do that which promotes our genetic replication and avoid that which is detrimental to our genetic replication by using simple desires. The pain that causes me to favor my foot when I twist my ankle has nothing to do with making an "evaluative judgment". Nor does my going for a second slice of chocolate cake as my wife and I snuggle on the couch in the evening. In fact, neither does the snuggling. With respect to the chocolate cake, I may well give a negative evaluative judgment to eating a second piece of chocolate cake, but do so anyway. I am acting on a desire - one that, at times, seems to be stronger than any motivation that may be associated with an evaluative judgment.

This becomes more obvious when we look at the behavior of animals. I throw a toy mouse for my cat, who runs off, grabs it, and brings it back to me. In explaining and predicting his behavior, I have never resorted to "evaluative judgments". Instead, all I need for my explanation are simple desires. It also explains why he eats his catfood, uses his kitty liter box, and crawls up on my lap and goes to sleep when I watch television. I suspect that his thoughts have never held an evaluative judgment. Yet, he has no trouble engaging in the type of behavior that caused his ancestors to be evolutionarily successful.

When it comes to taking care of our young, no "evaluative judgment" is needed to cause a bird to build a nest, keep her eggs warm until they hatch, and bring the fledgling birds food until they are able to fly on their own. I feel confident that, at no time, no bird ever made a judgment that taking care of her young had some type of "ought to be doneness" associated with it. The bird did what the bird wanted to do, and in wanting to do it she contributed to the genetic makeup of the next generation.

There is no reason to believe that the care of birds for their young is caused by one type of mental operation, and that human care for their young is substantially different - any more than there is to think that animal pain is substantially different from human pain. They all motivate without resorting to the complexities of "evaluative judgments".

When Street writes about "evaluative judgments", she is including desires in what she is talking about. Yet, I hope that one can recognize a difference between judging something to be good or bad, and having a desire for or an aversion to it. Street herself wrote that she was interested in "basic behavioral and motivational tendencies" and "unreflective, non-linquistic motivational tendencies." This speaks more of desires than it does of evaluative judgments. On this matter I agree. Our desires - our aversion to pain, desire for sex, hunger, thirst, comfort, the simple pleasure of taking care of a child or of the company of a friend - have been subjected to evolutionary selection. But these are desires as distinct from evaluative judgments.

Recognizing this distinction, we should recognize that we have reason to take a look at evaluative judgments and see if we can make any sense of what they are.

Critique of Sharon Street's "Darwinian Dilemma for Moral Realism" - Part 01

I intended to examine the difference between my thesis:

The right act is the act that a person with good motives and lacking bad motives would have done in the circumstances:

And Rosalind Hursthouse's thesis:

An act is right iff a virtuous person would have characteristically (acting in character) done in the circumstances.

I found that Hursthouse "virtuous person" is a person who rationally pursues certain "naturalistic ends" - these being (1) survival of the individual, (2) survival of the species, (3) experiences of pleasure, pain, and emotions characteristic of the species, and (4) actions characteristic of the species.

I, on the other hand, believe that there are no "naturalistic ends" - that there are no ends but that desire makes them so. "Good motives" are motives that tend to fulfill (other) desires, and "bad motives" tend to thwart other motives. This fulfillment or thwarting of other motives is what provides people with reasons to promote or demote other motives.

I intended to use Sharon Street's 2006 article, “A Darwinian Dilemma for Realist Theories of Value,” to argue against Hursthouse's "naturalistic ends" (Street, Sharon. (2006). "A Darwinian Dilemma for Realist Theories of Value". Philosophical Studies. 127. 109-166.). However, I discovered that, before I could use Street's argument, I needed to make some adjustments.

Specifically, Street's first premise is that our evaluative judgments are subject to evolutionary pressure. The realist about value either needs to assume an unrealistic coincidence between what evolution has caused us to value and what has intrinsic value properties, or make a scientifically unsound claim that these intrinsic value properties actually help to explain and predict the course of evolution. Neither option seems reasonable. The reasonable conclusion is that what we value does not actually have anything to do with real value properties.

On the surface, I agree with Street's argument, which is why I sought to use it against Rosalind Hursthouse. However, when I looked at the argument in detail, I noticed some problems that I need to take care of before I actually use it.

Specifically, when Street argues that evolutionary pressures have shaped our "evaluative judgments," I find that the concept of "evaluative judgments" she uses is far too broad. She includes things in this definition that are not, actually, best understood as being under evolutionary pressure.

Terminological Dispute

To complicate matters, I also have to dispute Street's use of terminology. In "Moral Objectivity and Moral Realism" I express my objections to her terminology. Street provides an argument against theories of value as intrinsic properties. However, this is an argument against realist theories of value only if we equate realist theories with intrinsic value theories. I deny this assumption. Relational properties are real, desires are real, and relationships between states of affairs and desires are real. Propositions describing those relationships are as real as any property discussed in physics and chemistry - which often deals with relational properties. For these reasons, I reject the idea that Street provides an argument against realist theories of value.

However, what matters here is her argument against the existence of intrinsic value properties. I agree with Street, they do not exist. However, I think that Street makes some mistakes in establishing that conclusion.

Street's first expression of her premise is:

the forces of natural selection have had a tremendous influence on the content of human evaluative judgements.

In defense of this, she notes how unlikely it is that a creature that acquired a dispositon to judge that it ought to kill its offspring will have Darwinian fitness, compared to an alternative creature with a disposition to judge that it ought to care for and protect its offspring.

Having said this, Street points out that she is aware of some of the complexities in this account.

Specifically:

(1) Her presentation suggests that evaluative judgments came first, and then evolution acted on those judgments to select those for evolutionary fitness.

This is not accurate. As Street admits:

the capacity for full-fledged evaluative judgement was a relatively late evolutionary add-on, superimposed on top of much more basic behavioral and motivational tendencies.

(2) The next problem is that, for evolutionary pressures to act on a judgment, they must be genetically inheritable. Yet, it is unlikely that a full-fledged evaluative judgment has this quality. Instead, she argues that these evolutionary forces have acted on:

unreflective, non-linguistic, motivational tendency to experience something as ‘‘called for’’ or ‘‘demanded’’ in itself, or to experience one thing as ‘‘calling for’’ or ‘‘counting in favor of’’ something else.

I would like to argue that the "basic behavioral and motivational tendencies" and the "unreflective, non-linquistic motivational tendencies" that evolution works on are basic desires. When one gets to more complex judgments, the idea that evolutionary pressures have influenced them becomes less plausible. However, this does not change the basic fact that our evaluative judgments are based on desires subject to evolutionary influence, and it is implausible to believe that those basic desires pick out some type of desire-independent reason for action.

Street includes in this set of evaluative judgments that are subject to evolutionary influence moral judgments. She includes the following items as examples:

1. The fact that something would promote one’s survival is a reason against it.
2. The fact that something would promote the interests of a family member is a reason not to do it.
3. We have greater obligations to help complete strangers than we do to help our own children.
4. The fact that someone has treated one well is a reason to do that individual harm in return.
5. The fact that someone is altruistic is a reason to dislike, condemn, and punish him or her.
6. The fact that someone has done one deliberate harm is a reason to seek out that person’s company and reward him or her.

My alternative thesis is this:

These "more basic behavioral and motivational tendencies" that evolution has worked on . . . these genetically inheritable "unreflective, non-linguistic, motivational tendency," are basic desires or the mechanisms by which basic desires are learned. That is all. Evolution has shaped our likes and dislikes, and the likes and dislikes we are disposed to acquire through interaction with our environment. Other evaluative judgments - particularly moral judgments - are not on this list.

It is going to take a few posts to lay out this argument.

Please bear with me.

Mill on Motives and Right Action

I have objected against Henry Sidgwick that the criterion of right action is not utility. It is whether a person with good motives and lacking bad motives would have performed that action. I combined this with the standard objection - which Sidgwick agrees with - that a person whose sole motive is utility is not a person with the right and best motives to conclude that the standard of right action is not utility.

The standard of the right and best motives may be utility. (I argue that it is not.) But the standard of right action is the right and best motives.

John Stuart Mill addressed the same issue with similar results.

Mill wrote:

[Some objectors to utilitarianism] say it is exacting too much to require that people shall always act from the inducement of promoting the general interests of society.

In response, Mill, like Sidgwick, asserted that utilitarianism does not declare that utility be the sole motive for action.

It is the more unjust to utilitarianism that this particular misapprehension should be made a ground of objection to it, inasmuch as utilitarian moralists have gone beyond almost all others in affirming that the motive has nothing to do with the morality of the action, though much with the worth of the agent. He who saves a fellow creature from drowning does what is morally right, whether his motive be duty, or the hope of being paid for his trouble: he who betrays the friend that trusts him, is guilty of a crime, even if his object be to serve another friend to whom he is under greater obligations.

Mill’s claim seems to be that the right act is the act that maximizes utility. It does not matter what the agent’s motive is, as long as the act maximizes utility.

The problem is that if an agent is motivated by something other than utility, then there are inevitable circumstances in which the agent is going to sacrifice utility for the sake of this other end. If the agent has a particularly strong affection for a friend or his child, he will choose the lesser suffering of his friend or child to the greater suffering of a stranger. If he has an aversion to lying or to taking property without consent, then he will choose to tell the truth or not to take property even where the greater utility can come from it. Sooner or Jager, the person not motivated solely by utility is going to perform the utilitarian “wrong act”.

Of course, Mill would allow you to perform a wrong action from a good motive if having that motive and acting on it would generally produce more utility than the motive of maximizing utility. For example, even though the motive of parental affection may motivate an agent to act to bring about the lesser happiness of one’s own child to the greater happiness of a stranger on the other side of the world, parental affection is responsible for such utility in the world that we must overlook the wrongness in these actions. On the standard of utility, these acts are still wrong, but we are not going to condemn you for them because there is negative utility in the condemnation.

But what do you say of a wrong act (buying your own child a toy instead of buying medical care for a sick child on the other side of the world) that is a utilitarian “wrong act” done from one of these utilitarian “good motives”?

The Utilitarian seems to have a convoluted answer. “You act was wrong on the Utilitarian standard, but we are going to call it permissible and expect you to treat it as permissible - to regard it in all ways as if the claim that it is permissible is true - even though it is not.”

Furthermore, we are to take this as the common meaning of moral terms.

It makes no sense - particularly the part that says that we all know as a part of the regular meaning and use of the terms “right” and “wrong” that we are to know and believe that some wrong acts are permissible, some obligatory acts are prohibited, and some permissible acts are wrong.

No . . . It makes more sense to come up with a theory of right and wrong that allows us to say that a wrong act is wrong, a permissible act is permissible, and an obligatory act is obligatory.

This means that an obligatory act is that which we have reason to condemn people for not doing.

A prohibited act is one we have reason to condemn people for doing.

A non-obligatory permissible act is one we have no particular reason to condemn a person for doing or not doing.

Which means that it is the act that a person with good motives and lacking bad motives would have done, would not have done, or might or might not have done depending on the agent’s other interests.

Moral Objectivity and Moral Realism

In the paper that I am writing on Rosalind Hursthouse's theory of right action, I am going to disagree with her on what she calls a second spring of human action - reason or rationality.

I am going to side with David Hume. Desires provide the only "spring of action" - nothing is an end except that desire makes it so. The second type of spring that she writes about does not exist.

However, the thesis that such a spring does not exist - at least in the way that I say it does not exist - has been equated with the denial of the existence of moral values altogether - moral anti-realism or moral subjectivism.

I disagree with the use of these terms.

It is not that I disagree with the thesis. I simply think that expressing the thesis in these terms generates unnecessary confusion.

So, I am going to want to say that my denial of the existence of this "second spring" is compatible with the objectivity of morality and moral realism. It is simply a denial that moral claims are claims of intrinsic prescrptivity.

Specifically . . .

In saying that a second type of “spring of action” does not exist I will be following a path much like that of J.L. Mackie in Ethics: Inventing Right and Wrong. Mackie starts his book with the proposition, “There are no objective values.” In place of Mackie’s “Argument from Queerness,” I intend to substitute Sharon Street’s, A Darwinian Dilemma for Moral Realism.”

However, in doing this, I am going to reject their terminology. What the phrase, “There are no objective values” and the term “moral anti-realism” means on the street is something quite different from what it means in the philosophy lecture hall, and I think this difference is quite destructive.

On the street, these terms imply that one cannot legitimately criticize the racist or the terrorist or anybody whose views are different from one’s own because all such views are merely differences of opinion - with no opinion being objectively better than any other. Protests on the part of philosophers that this is not what they mean are irrelevant - not unless they want to fund a massive public duration campaign to make their special meanings a part of the common use of moral language.

Morality, unlike the special and specific terms of some sciences, cannot be effectively limited to the classroom or laboratory. It is a public practice. It is unwise to use a private jargon that, when released on the public, is more likely to generate confusion and error than to serve any useful purpose.

I have described a case in which the proposition, “People generally have strong reasons to promote a universal aversion to causing pain to others” is objectively true. In that world, the proposition was true before any agent was born, remains true regardless of the beliefs and desires that the agent may acquire in life, and will remain true after the agent dies. Even if a particular agent would lose the aversion to pain, the proposition that people generally have many and strong reasons to promote a universal aversion to causing pain remains true.

This is not an appeal to some sort of fictionalism where we must all embrace a falsehood as if true to realize some social good. The proposition that a desire tends to fulfill or thwart other desires is an objectively true statement. Its truth depends on the existence and content of those other desires. It might be true in the context of one set of desires and false in the context of a different set. However, these facts do not prevent the statement from being true. The fact that, in certain circumstances, the statement, “I am in Colorado” could be false does not change the fact that it is true, at least as I write this.

In short, I am going to agree with Mackie and Street. However, while Mackie says that there are no objective values, I am going to say that there is no intrinsic prescriptivity. While Street claims to be offering a Darwinian dilemma for moral realism, I am going to say that she is offering a Darwinian dilemma for intrinsic moral prescriptivity realism. But I will hold that moral values are objective, and that some actions really are wrong. I am simply not going to ground their objectivity or reality on intrinsic prescriptive properties.

Summary of Thesis Project

I have finished Hursthouse’s book On Virtue Ethics. My next project will be to write a commentary on it, using the research notes that I wrote into the previous 18 posts.

The key difference between my proposition . . .

The right act is the act that a person with good motives and lacking bad motives would have performed in the circumstances

. . . and her thesis . . .

An act is right iff it is the act that the virtuous agent would characteristically (acting in character) do in the circumstances

. . . focuses on the proposition

There is no end but that desire makes it so.

I hold that this proposition is true.

Hursthouse holds that it is false. Hursthouse builds her theory on demonstrating the existence of four "naturalistic ends," and then adds the distinctly human trait involving the capacity to apply reason to determine if a state conforms to these ends (which no plant or animal can do).

So, this is going to become the second chapter of my Master's Thesis.

The first chapter will deal with the proposition:

The right act is the act that a person with good motives and lacking bad motives would do in the circumstances.

This will be a rewritten and more detailed version of my Ethics Proseminar paper: Sidgwick on Motives and Right Action. You can find the most recent version of that paper on the documents page of the desirism site.

But that will lead to a discussion of good and bad desires, which will have to do with ends, where . . . in Chapter 2 . . . I will defend the thesis that there are no ends but that desire makes them so. That will discuss desire as a propositional attitude - a motivational reason to make or keep true the proposition that is the object of a desire state.

More specifically, the chapter will:

(1) State the proposition to be defended: "There is no end but that desire makes it so."
(2) What is a desire? (Desire as propositional attitude.)
(3) Good desires: Desires that tend to fulfill other desires. Desires that people generally have reason to promote universally.
(4) Argument from evolution that our desires evolved, not to perceive some natural end, but to serve fitness. (Sharon Street's argument, "A Darwinian Dilemma for Moral Realism")
(5) A terminology aside - a dispute over the terms "objective morality" and "moral realism".
(6) Rosalind Husthouse's "naturalistic ends".
(7) Conclusion: My thesis of right action is better than Rosalind Hursthouse's thesis of right action

This should be a basic defense of desirism. Let's see if I can get it approved as a thesis.

Naturalistic vs. Manufactured Ends

It turns out that, in the last chapter of her book, On Virtue Ethics, Rosalind Hursthouse responded to the Darwinian objection to her theory.

In her response, she seems to see only three options:

(1) Darwinian theory underwrites the idea of a characteristic human nature that is the foundation of Aristotelian virtues.

(2) Darwinian theory provides a different set of naturalistic ends.

(3) Moral nihilism - humans evolved into an incoherent mess for whom flourishing is impossible since the different parts of our that which is characteristically human are in perpetual conflict.

In other words, either there is something in being “characteristically human” that provides the ends that we need to flourish, or we are doomed to the type of misery and suffering that has dominated human history.

This is a false dichotomy.

Our brains are “plastic” – meaning that we have the capacity to manufacture ends. Because of this, we do not need to depend on nature to provide us with harmonious ends, we can manufacture them. Of course, we have reason to manufacture ends that tend towards a harmonious whole. We certainly have no reason to create in others ends that conflict with our own. Insofar as we have an end – e.g., an aversion to pain – we have reason to manufacture in other people ends that are compatible with these natural ends – e.g., an aversion to causing pain to others.

What Darwinism gives us is not a set of naturalistic Darwinian ends. It gives us reason to believe that there are no ends but that desiring makes them so. There is reason for the belief-forming systems of our brain to aim for truth and, of these, more useful truths over those that are useless. False beliefs – e.g., that one can float when jumping off a cliff – can get one killed.

However, belief alone will not tell us what to do. To turn beliefs into actions we need desires. And there is no “truth” to desires – no natural ends to perceive. There are only desires that lead to genetic replication and those that do not. Yet, even genetic replication is not an end. It is an unintended side effect. Of course, only those whose desires produced this unintended side effect are with us today (with, perhaps, a few individual exceptions). But this fact does not make genetic replication an end.

Hursthouse does not deny this claim of plasticity in our ends. In fact, she argues for it. It plays a major role in her discussion of the moral education of children. You cannot train a child to be virtuous (or vicious, as depicted in her discussion of racism) unless there is a plasticity to the brain. She also points to cultural variation as proof of the placticity of ends. What she has apparently failed to recognize is that this plasticity of ends allows us to have a robust moral realism without the naturalistic ends of Aristotelianism or Darwinism. The plasticity itself is enough to give us morality.

The fact of plasticity, combined with our ability to determine how to use this to mold the ends of others, gives us reason to ask, “What ends should we mold?” Even accepting as true that there are no ends but that desiring makes them so, we still get the conclusion that the ends we have reason to manufacture universally are those that, when manufactured universally, tend to fulfill (rather than thwart) the desires of others. We have desire-based reasons to manufacture a harmony of ends.

It would be irrational to manufacture conflict, in the same way that it would be irrational for a person with an aversion to pain to put his hand in a hot fire.

The Neurathian Procedure for Moral Justification

The Neurathian Procedure applies to attempts to answer the question of how we can know something to be true. If "It is wrong to lie under oath" is true (or, more precisely, under conditions where it is true), the Neurathian Procedure gives us a way of demonstrating that it is true.

The example gets its name from the fact that Otto Neurath invented the following simile:

Imagine you are on a boat at sea. You depend on your boat for survival. Some of the planks on the boat are rotting and in need of repair. In addition, you can well imagine improvements that you can make. However, you cannot dismantle the boat entirely and build a new boat from scratch. Your only option is to make your improvements one step at a time - removing a plank (for example) and putting a new plank in its place, fastening it to the old frame that, itself, will someday need to be replaced.

The same is true with knowledge. We cannot build a new system of knowledge from scratch. The best that we can do is to replace one belief at a time, attaching it to our existing beliefs which, themselves, may ultimately need to be replaced. We can, over time, improve our overall system of beliefs, but we can only do so one step at a time in a way that can never ignore the fact that we are making changes to a larger system that, at each step, is left substantially intact. Eventually, over time, we may turn our rowboat into a 21st century cruise ship - but it will take time.

Rosalind Hursthouse wants to use this procedure to defend her theory of virtue ethics. She argues that we cannot build an ethics from scratch. Instead, all we can do is to take our existing ship of moral beliefs and start looking for ways to improve upon it. The objection that we are not able to completely leave an ethical system to build a new ethical system from scratch is no objection. The Neurathian Procedure represents our best way of proceeding. This admits to the fact that we are going to continue to have some moral statements in our argument. We are not going to build our ethics from pure, value-free, scientific facts.

My question, regarding this way of proceeding, is, "Why are we building two boats?"

Why are we building a science boat and a completely different ethics boat?

If this is, indeed, the correct procedure, then it seems that we should be building one boat, and we need to find some way to secure the normaltive/moral planks onto a frame of objective scientific fact.

In her defense of virtue ethics, Hursthouse defends a number of naturalistic ends. She starts with the evaluation of plants, where she states:

So, in the evaluation of individual plants, we find that we evaluate two aspects - parts and operations - in relation to two ends. A good x is one that is well fitted or endowed with respect to its parts and operations; whether it is thus well fitted or endowed is determined by whether its parts and operations serve its individual survival and the continuance of its species well,
in the way characteristic of xs.

I deny the existence of any "naturalistic ends," and I further hold that a proper understanding of evolution will provide a powerful argument to explain why no natural ends exist.

My view is that there are no ends but that desiring makes them so, and no plant desires the continuation of its species. In fact, no plant desires anything, so no plant has an end of any type. When we evaluate plants, it is our ends that we appeal to - not the plant's ends, or any "naturalistic end" that can be discovered by pure reason.

The continuation of the species, where it exists, is not an "end," it is an unintended side effect. Indeed, when humans have sex, pregnancy is not the end or goal of that activity. It is often an unintended side effect. This is even more obvious when sex results in spreading a sexually transmitted disease.

As it turns out, the plants that we come across are those that happen to have characteristics that - at least in their past environments - resulted in species survival. If they didn't, we would not be finding their species in the world today. To say that species survival is an end is to add something to this state that is quite literally false.

Hursthouse also states that if an animal or plant must act in a characteristic way for that species.

There is no sense to being attached to saying that polar bears would be better fitted to flourish in a characteristically polar bear way, to live well, as polar bears, if the males were different, or indeed, if males and females banded together to hunt. Polar bears just don't act that way and thereby cannot - unless they mutate - and that is all there is to it.

The first point to make is that we do have an account of a genetic defect. What if a genetic change happens to result in a polar bear acting in a way uncharacteristic of polar bears - forming pair-bonds with its mate, for example, the way eagles do. We must remember that every species-wide mutation began with a mutation in a single member of the species and spread over time, as those with the mutation survived and those without it died out (or those with the mutation became isolated from those that did not, etc.). In no case in evolution was the "survival of the species" a reason for a particular mutation. That mutation either brought about the survival and genetic replication of those who had it, or the mutation ceased to exist, but genetic replication never was an end.

The second point to make is that her statement could be wrong. Bear cubs learn certain aspects of their behavior from their parents. It may well be the case that, if the parents were to behave differently, so would their cubs. However, due to blind chance, bears just never happened upon a form of behavior built around males and females hunting together - because females kept driving off the males and the males, so isolated from their cubs, came to see the cubs as potential food. Hursthouse can give us no reason to insist that a characteristic polar-bear behavior could not have been different.

The third and most important point to make is that, if one understands evolution, one would see that the ends that any creature adopts is a matter of random genetic selection and contingent interactions with one's environment. There may be a characteristically polar-bear way of behaving, but you cannot get from this statistical norm to the conclusion that all polar bears ought to behave that way. If a polar bear behaves differently, even if it should fail to genetically reproduce and die off, no naturalistic moral ought has been violated. This is just a fact about how the real world works . . . nothing more.

Evolution has given us desires, and dispositions to learn new desires based on our interactions with our environment. In this sense, each of us has ends. The ends we have acquired are, in themselves, neither "correct" or "incorrect". They merely exist as they do as a result of random and contingent facts.

There is, in the real world, no ends but that desiring makes them so.

Acting Virtuously

In my last post, I promised to say something about "acting virtuously".

I have been discussing Rosalind Hursthouse's thesis that:

An action is right iff it is what a virtuous agent would characteristically (i.e. acting in character) do in the circumstances.

This definition of a right action does not look at why a person actually performed an act - so long as it was the act that a virtuous person would have done. The person who repays a debt does what he ought to do -
he ought to repay the debt (under normal circumstances). He may repay the debt because it is owed, or he may repay the debt out of malevolence - because doing so will allow the recipient to afford to purchase something that somebody I hate is also trying to purchase. My reasons for repaying the debt does not change the fact that repaying the debt is the right thing to do.

We are assuming that this is a case in which one actually should repay what is owed. There are, of case, circumstances in which one perhaps ought not to repay a debt; such as returning a borrowed car to a driver who is drunk and would likely drive. However, the interesting fact is that when repaying the debt is the right thing to do, it does not matter whether the agent acts from duty or acts from malice. And when repaying a debt is the wrong thing to do, it does not matter whether the reason is concern for the consequence or selfishness.

Yet, there is also the category of "acting virtuously" - which is "doing the right act for the right reasons".

More specifically, Hursthouse is interested in comparing the virtue-theory concept of acting virtuously with the Kantian concept of acting out of a sense of duty. She wants to claim that these two concepts are closer than people have assumed.

To "act virtuously", Hursthouse provides the following conditions:

(1) For a start, it is to do a certain sort of action. What sort? `A virtuous, good, action', we might say-and truly, but this is hardly an illuminating way to begin laying down what it is to act virtuously, or well. So we'll give examples - it is to do something such as helping someone, facing danger, telling the truth, repaying a debt, denying oneself some physical pleasure, etc.

(2) The agent must know what she is doing-that she is helping, facing danger, telling the truth, etc.

(3) The agent acts for a reason and, moreover, for `the right reason(s)'."

(4) The agent has the appropriate feeling(s) or attitude(s) when she acts.

The "right reasons" in this account actually refers to a range of reasons. Hursthouse writes:

What are reasons `typical of' a virtue? They will be the sorts of reasons for which someone with a particular virtue, V, will do a V act. So, thinking of the sorts of reasons a courageous agent might have for performing a courageous act, we can come up with such things as `I could probably save him if I climbed up there', `Someone had to volunteer', `One can't give in to tyrants', `It's worth the risk'. Thinking of the range of reasons a temperate agent might have for a temperate act, we can come up with `This is an adequate sufficiency', `I'm driving', 'I'd like you to have some', `You need it more than I do', `She said "No"'. With respect to the liberal or generous, `He needed help', `He asked me for it', `It was his twenty-first birthday', day', `She'll be so pleased'. With respect to the agent with the virtue of being a good friend, `He's my friend', `He's expecting me to', `I can't let him down'. For honesty we get such things as `It was the truth', `He asked me', `It's best to get such things out into the open straight away'. And for justice we get such things as `It's his', `I owe it to her', `She has the right to decide', `I promised'. And so on and so forth.

These are the same types of reasons, Hursthouse argues, that would also count as "acting from a sense of duty" in the Kantian sense. The Kantian does not need to actually be thinking, "I am acting from a maxim that I can will to be a universal law" to be acting from a maxim that she could will to be a universal law. She could be acting for any of the reasons given above, and still be acting from duty in the Kantian sense.

The desire-based model that I defend provides another way of viewing this question.

Desires provide the only end-reasons for intentional actions that exist. We look at the proposition that the agent is trying to make or keep true to determine why the agent acted. To refuse to make a particular claim "because it would be a lie" indicates an aversion to lying - or a "desire that I not lie". This "desire that I not lie" is not the same thing a "desire that I not say things that are untrue." To lie is to say something that the agent believes to be false; which is different from saying something that is false. An agent can say something true that she believes to be false and that would still be a lie. Similarly, an agent can say something that is false that she believes to be true, in which case she is mistaken, but not lying.

So, there is a difference between telling the truth under oath "because it is the truth" or "because the system of justice requires it of me," and doing so "because I wish to avoid being punished for committing perjury". The first two reasons provide examples of acting from a desire that people generally have many and strong reasons to promote universally. The last is acting from a self-interested desire that people naturally have.

We can see the difference as well between religious people who obey religious commandments to obtain the rewards of heaven and avoid the penalties of hell, and those who do so because it is the right thing to do. With the first sort of people, if they were to lose their beliefs in heaven and hell, they would lose their reason to refrain from these activities. They would see no reason not to engage in murder, rape, and theft. However, the person who would refrain from murder, rape, and theft anyway is somebody who not only has the desires for the pleasures of heaven and aversion to the pains of hell, but somebody who has an aversion to murder, rape, and theft for their own sake.

Hursthouse also makes use of this distinction - and on the fact that we can tell more about what a person's reasons are by their actions than by their words. There are people who may say, "Without God, everything is permitted." Yet, even if they were to come to believe that God did not exist, would still be reluctant to put their hands in a hot flame, sell their children into slavery, betray a friend, or inflict needless suffering on somebody that they love. This is because the reasons they give for their action (to please God) simply are not true. If it were true, the reasons would vanish if the belief in God vanished. They actually do these things for their own sake - because these are, themselves, the objects of the agent's desires. whether they admit it or not.

Does Virtue Benefit the Virtuous?

A key claim of Aristotelian virtue is that virtue benefits the virtuous. Indeed, the very reason to become virtuous is to live a good life.

I find a curious inconsistency in this way of talking. Is one being virtuous for the sake of virtue: Honest for the sake of honesty? Kind for the sake of kindness? Repay a debt because it is owed? Or is one being virtuous for the sake of living a good life? The two are not the same.

I want to put that discussion aside until my next posting.

The question is whether a virtue benefits the agent.

Let's go back to the beginning and look at Rosalind Hursthouse's proposed theory of right action as compared to my own, and look at the question of whether right action benefits the agent.

Hursthouse’s thesis is:

An action is right iff it is what a virtuous agent would characteristically (i.e. acting in character) do in the circumstances.

My thesis:

An act is right if and only if it is the act that a person having good desires and lacking bad desires would do in the circumstances.

From here, the question becomes: Does a right act benefit the agent?

Hursthouse is keen to agree that this is not the case with respect to every specific action. The question is whether the disposition tends to allow the agent to live a better life. She compares virtuous behavior to smoking - or, more precisely, the absence of smoking. The person who gives up (or never starts) smoking cannot be guaranteed a long and healthy life - but does improve their odds.

Hursthouse further argues that, when parents teach their children to be virtuous, it is not with the idea that the child will then miserable. Instead, the parent realizes that the virtuous child can also expect to have a better life than a vicious child. It is with full regard for their interest in their child's well-being that they teach the child to be virtuous. This at least suggests that living virtuously benefits the agent.

Actually, Hursthouse makes a stronger claim than merely to assert that virtue benefits the virtuous. This is true nearly by definition. A virtue is a character trait that benefits the virtuous.

Now, what does my thesis say? Does a good motive benefit the person who has it?

A good motive is a motive that people generally have many and strong reasons to promote universally. They have reason to promote the motive universally because it is a motive that tends to fulfill other desires. Insofar as the desire tends to fulfill other desires, people generally benefit as the good motive becomes universal.

However, we need to be careful here about the word "benefit". A good desire is a desire that tends to full other desires. But the fulfillment of a desire is not always and automatically a benefit to the person whose desire is fulfilled. It is not the case that everything we want is a benefit to ourselves.

I have used the example of a person with one desire - a desire that the planet Pandora exist. He has a button in front of him. Pressing the button will bring the planet Pandora into existence. Pressing the button will also kill him. The agent has one reason to press the button - to bring the planet Pandora into existence. He also has no reason not to press the button. His only interest is in bringing the planet Pandora into existence. When he presses the button he will get what he wants. However, it is a mistake to say that when he gets what he wants he has obtained a benefit.

So, a good desire is a desire that, if universal, will tend to fulfill other desires. And, among humans, it happens to be the case that many of our desires are self-referencing. We have an aversion to being in pain or being uncomfortable. We have a disposition to sadness and frustration - traits to which we are also averse - when our other desires are thwarted.

So, a good desire is a desire that tends to fulfill other desires. And it happens to be the case that a desire that tends to fulfill other desires tends largely to benefit the people whose desires are fulfilled. However, it is not entirely true, and it is not necessarily true, that the universalization of a good desire benefits people.

I have a standard, simple example of such a good desire.

A community of individuals, each of which has an aversion to his or her personal pain, has a reason to promote universally an aversion to causing pain to others. As this desire becomes universal, agents within the community acquire two desires instead of one. These desires may sometimes come into conflict. An agent may find himself in a situation in which he must choose between inflicting a large amount of pain on others or a little pain on himself. The agent with this good aversion will accept the minor pain for himself.

Has he obtained a benefit?

Not really. He has realized something that he values - the absence of pain for others. But it has come at a cost - the minor pain he then experiences.

However, because of this universal aversion to causing pain to others, each agent can expect to experience less pain than he otherwise would. So, people obtain benefits from the universalization of a virtue, but not from exercising the virtue itself.

So, why become virtuous?

That's not the right question.

Why promote a virtue or good motive universally?

Because the good motive is, by definition, is one that, when universal, tends to fulfill other desires, and the fulfillment of those other desires provides the reason to promote the motive universally.

The "Morally Correct" Emotional Response

Rosalind Hursthouse devotes Chapter 5 of her book On Virtue Ethics to explaining how virtue ethics gives the best account of the moral importance of emotions.

A part of her defense rests on claiming that a particular emotional response is simply intrinsically good - or, more accurately, intrinsically correct given a particular state of affairs. A part of what it means to have a particular virtue is to be disposed to have that particular response.

We should note too that the claims in combination give some cash value to the view that the feeling of certain emotions on certain tain occasions has intrinsic moral value, rather than merely instrumental mental value or some other sort of intrinsic value. Feeling this emotion then could be said to have ‘intrinsic moral value’ simply in so far as it is the manifestation of virtue.

She further states that a particular emotional response is simply "right" or "correct". . . as in "The right answer to 'What is the capital of New Zealand?' is 'Wellington'."

So, to have a particular virtue is to have a particular emotional response to relevant states of affairs. To lack the appropriate emotional response means that one lacks the relevant virtue. There is, then, an intimate link between emotion and morality; one is not moral unless one has the "correct" moral response.

The first point that I will challenge is the claim that there is a "correct" moral response - such that those who do not respond in a particular way are "doing it wrong."

The emotions that we are disposed to have are, in part, the result of millions of years of evolution. Evolution, in selecting the emotional responses we would have, did not care about any type of intrinsic "correctness" - evolution favored reactions that produced evolutionary fitness. It is useful, in the biological evolutionary sense, for the antelope to be afraid of the lion. It is also useful for the antelope to grow anxious and alert when something happens - something it might not even have consciously noticed - that would indicate a lion in the area, such as some birds suddenly taking flight or a subtle scent in the air. A different creature - one not subject to being eaten by lions, perhaps because there are no lions, or who - because of the randomness of genetic mutation - simply never acquired such a disposition and could not have evolution select for it - would not have this reaction. They do not react "incorrectly" - just differently.

You can find a detailed account of this argument in: Street, Sharon (2005). A Darwinian dilemma for realist theories of value. Philosophical Studies 127 (1):109-166, and which I commented on previously here. While I think that Street presents a good argument, I think she presents her case in a seriously misleading way when she claims to have an objection against moral realism rather than simply intrinsic-value realism. Her theory leaves open the possibility that real moral properties exist, but not as the type of intrinsic properties that Hursthouse's "correct" emotional responses would require.

We also have the capacity to learn certain emotional responses. Hursthouse does not deny this - and goes into a long (and largely accurate) discussion of the nature of learning to be a racist and trying to unlearn racism. What she ignores is that the systems that allow us to learn and unlearn emotional responses have also been subject to evolutionary pressure, disposing us to adopt those attitudes that kept our ancestors alive and allowed them to have children. In the case of beliefs, truth is generally important, as we have reason to know whether there is a cliff ahead or whether there is a lion in the brush is true.

However, with respect to motivational force, there is no external "correctness" to latch onto. There is only that which will dispose us to act in ways that will lead to our evolutionary success.

However, if emotions are learned due to an interaction with one's environment, and each of us is a part of each others' environment, we have reasons to promote certain emotional responses in others. Insofar as each of us have an aversion to pain, we have reason to cause others to dislike - to emotionally recoil from the thought - of causing pain to others. We are, after all, the others that they would otherwise cause to be in pain. We have reason to use our power as a part of their environment to make it the case that the "correct" response to the thought of causing pain to others is revulsion.

This is not any type of intrinsic correctness. This is a correctness that comes precisely from our aversion to pain, giving each of us a reason to create an environment in which we are less likely to experience pain, which gives us reason to cause others to acquire an aversion to causing pain. We have this power to the degree that we can understand how environmental factors might cause others to acquire an aversion to causing pain (e.g., praise those who tend to avoid causing pain and condemning those who tend to cause pain). Ultimately, the value of an emotional response is found in the usefulness of promoting that response universally, not in any type of intrinsic correctness.

Having said this, we can use an agent's emotional reaction as evidence of whether or not an agent actually has a particularly useful desire or aversion. This means that we can use the presence or absence of a particular emotional response as reason to praise or condemn an agent. We can expect that if a person breaks an important promise because something more important comes up that the agent will feel some level of regret over having broken a promise to us. If they come to us and merely report that something more important came up - with total indifference to the fact that they broke a promise, that they do not have the aversion to breaking promises that people generally have reason to promote in others. This part of what Hursthouse argues for is defensible. What is not defensible is her claim that the "correctness" of a response is an intrinsic value property that we can, in some cases, learn by reason alone and is not concerned with useful ways of fulfilling certain desires.

Hursthouse's Virtue Theory and Non-Obligatory Permissions

As I read through Rosalind Hursthouse’s account of virtue ethics, I am not seeing much in the way of a theory of non-obligatory permissions.

This subject is relevant to her discussion of “positively pleasant dilemmas” - an idea she borrows from Philippe Foot.

But, in the context of the abstract discussion cussion of whether there are such things as irresolvable dilemmas, Foot has nicely raised the possibility that there may be positively pleasant ones. We may be faced with a choice between goods where not having either is no loss, and `there are no moral grounds for favouring doing x over y'.

She creates an example where:

Suppose I must give my daughter a birthday present; it would certainly be very mean not to, given our relationship, her age and hopes, my financial circumstances, and so on. But I am faced with an embarras de richesse; giving any one of a whole range of things is equally desirable and acceptable. So there is an irresolvable dilemma-not one that worries us, not one where the final decision matters, but there all the same-providing a clear case where practical rationality simply runs out of determining moral grounds.

Why is this an irresolvable dilemma?

She describes it as follows:

Virtue ethics directs me to find the answer to this question by finding the answer to another: 'What would a virtuous agent characteristically do in my circumstances?' But the supposition tion that the dilemma is truly irresolvable is tantamount to supposing ing the possibility of the following. We have two virtuous agents, each of whom (let us suppose, rather unrealistically) can give her daughter just one of two things, a or b, for her birthday; there are no moral grounds for favouring one over the other (for if there were, each agent, being virtuous, would go for the one that the grounds favoured). And one does x, giving her daughter a, and the other does y, giving her b. So virtue ethics does not give me action guidance here-which is just what we want, if we want our normative mative ethics to embody the fact that there are such irresolvable pleasant dilemmas, in which there is nothing that counts as the morally right decision.

But this is exactly what we expect in the case of non-obligatory permission. Different agents can choose different actions.

Assume that I had to make a choice regarding which of several $10 bills to give to a co-worker who paid for my lunch the day before. Different virtuous agents may choose a different $10 bill. “There are no moral grounds for favoring one over the other.” However, it is odd at best to call this as”positively pleasant dilemma”. This is a case of non-obligatory permission.

The alternative that I am seeking to defend has a place not only for the morally obligatory and morally prohibited, but also for non-obligatory permission.

An act is morally obligatory if and only if it is the act that a person with good desires and lacking bad desires would have performed in the circumstances.

An act is morally prohibited if and only if it is the act that a person with good desires and lacking bad desires would not have performed in the circumstances.

An act is morally obligatory if and only if it is the act that a person with good desires and lacking bad desires may or may not have performed in the circumstances.

The category of “morally permissible” concerns those area where people generally do not have good reason to promote a universal common motive. What to eat. What to wear. Where to live. What profession to go into.

Which bill to use to pay a debt.

Which present to buy one’s daughter.

The verdict of morality is, “Pick one, for Pete’s sake. It doesn’t matter.”

The ultimate point here is that, on those matters where people generally have little or no reason to promote a universal common motive using the tools of praise and condemnation, they have little or no reason to call one option “right” and another “wrong” (since these are, in fact, statements of praise and condemnation generally used to promote universal common motives).

The Role of Rules in Motive-Based Morality

I have extensively argued that morality is primarily concerned with the evaluation of motives - and of molding motives (desires and aversions) through the application of praise and condemnation.

I have shunned the idea of moral rules.

Yet, a morality of motives has a couple of significant weaknesses over a morality of rules.

(1) Motives cannot be as complex as rules. Indeed, rules can have nearly infinite complexity (e.g., no parking from 9:00 am to 5:00 pm except on weekends).

(2) We can easily change the rules as situations change (e.g., Football 2017 Rule’s change: 4-2-2l Specifies that the ball is declared dead if a prosthetic limb comes completely off the runner.) Motives, on the other hand, once learned, are difficult to unlearn.

However, rules do not come with their own motivational force. We still need to answer the question, “Why follow the rules?” How do we get people to obey the rules?

We accomplish this by adding a motive that says, “Follow the rules.” This allows us to combine the complexity and ease of change we have with rules with the motivational force of the morality of motives.

There are three primary areas of morality where we combine these elements of rules and motives.

(1) We have the rules of a game, combined with a moral prohibition on cheating. This moral prohibition is an aversion - taught using praise and condemnation as moral rules are generally taught - against breaking the rules even when one can get away with it.

(2) We have the rule of law, combined with a moral aversion to breaking the law. We understand this moral aversion in terms of an obligation to obey the law - to be a law-abiding citizen.

(3) We have a system of duties and obligations, combined with a desire to do one’s duty. This distinguishes the person who has a desire that her neighbor be better off from the person who cares nothing about her neighbor’s well-being but helps “because it is the right thing to do”.

I am going to set aside the first example for now. Though cheating is immoral (in that people generally have many and strong reasons to promote a universal aversion to cheating), it does not raise as many serious moral issues as the other two cases. The examples it does raise will be easier to present when we discuss the relevant and more important issues of the other two systems.

The Obligation to Obey the Law

When we turn to the second system - the system of law - this way of framing the issue gives us a way of addressing the question of whether, and to what existent, there is an obligation to obey the law.

The answer to this question is going to be the answer to the question, "Do people generally have many and strong reasons to promote, universally, a desire to obey the law (or, correspondingly, an aversion to breaking the law). Unsurprisingly, this is going to depend greatly on whether the law is one that fulfills the desires (or prevents the thwarting of desires) generally, or whether it instead thwarts the fulfillment of many and strong desires in order to fulfill the desires of the few. The Fugitive Slave Law of 1850 (or any law endorsing chattel slavery for that matter) is not a law that people generally have many and strong reasons to promote a universal desire to obey. In fact, people generally (though not universally - there is a reason why I use the term 'generally' in this context) have many and strong reasons to oppose this law. The same can be said about the Jim Crow laws that followed slavery. The same can be said for laws against homosexual relationships.

The desire to obey (the aversion to breaking) the law can allow a limited form of complexity. We can argue for disregarding a set of bad laws (e.g., Jim Crow laws, the Fugitive Slave law) while still promoting "the rule of law" as a general virtue. So, the question of whether or not there is an obligation to obey the law does not come with a simple yes/no answer. The answer may well be, "'Yes' for the vast majority of laws, but 'no' to those laws over there." Yet, the fact that this list cannot be infinitely complex - that it must, in fact, being founded in a motive, be relatively simple - does not allow us to create a long and complicated assessment of laws to obey or disobey.

One possible position to take is to say that all of the laws worthy of obeying are also moral restrictions, so there is no obligation to obey the law that is distinct from the obligation to do that which is right. However this, as I stated above, ignores the advantages of law in being potentially complex and easily changed. We cannot expect people to adopt a separate desire or aversion respecting every one of the traffic laws - e.g., a desire to use one's headlights from one-half hour before sunset to one-half hour after sunrise or when visibility is less than 1000 feet or when using windshield wipers to clear rain, snow, or sleet. In fact, it is not even possible to create such a desire using the social tools of praise and condemnation. What is possible is to create a general desire to obey the law (aversion to breaking the law) combined with the belief that traffic law has these requirements.

Doing the Right Thing

We also have the capacity to take advantage of the complexity and the ability to make changes to rules by giving people a desire to do that which is right (aversion to doing that which is wrong) as well as a complex set of principles respecting what is right and wrong. This represents the difference between refraining from lying because it is lying, and refraining from lying because one has an aversion to doing that which is wrong and a belief that lying is wrong. In this system, if we wish to have people alter their behavior, we simply alter their beliefs about what is right and wrong. The desire to do that which (the agent believes) is right and aversion to doing that which (the agent believes) is wrong then takes care of the motivation to obey the updated rules.

This leaves open the question, "What does it take for the proposition 'this action is right' and 'that action is wrong'" to be true. If we can convince a person with an aversion to doing that which is wrong that writing with the left hand is wrong, we can motivate that person to refrain from writing with his left hand. But what does it take for the proposition, "it is wrong to write with the left hand" to be true?

However, we are still going to have to answer the question, "Why create a rule against writing with the left hand? What reason is there to make such a rule?" This is going to go back to the question of motives - and it will have to refer to motives other than the motive of obeying the rule (that is to say, the motive to do that which is right). We still have to ask whether those motives for making an action right or wrong in this sense are motives that people generally have reason to promote universally. So, "right action" in this sense is still "the act that a person with good motives and lacking bad motives would have done in the circumstances."

There is nothing particularly noteworthy about acting from duty in this sense. It is merely a backup system - a reserve motivation for when the regular good motives are absent, or bad motives are present, that may still provide the motivational power to get the agent to do the right thing. An agent might not have the concern for the well-being of her child that she should have, but an aversion to doing what she ought not to do, and a belief that she ought not to neglect the health of her child, may motivate her to do what a properly motivated mother would do out of natural inclination.

Conclusion

In short, the idea that morality is primarily concerned with promoting good motives does not eliminate the possibility of a rules-based morality. We can harvest some of the advantages of rules - their complexity and the ease with which they may be changed - by combining them with a motive to follow the rules. These motives include an aversion to cheating (where cheating is defined by the rules of the game), an aversion to breaking the law (where the rules in question are the rules of law), and the desire to do one's duty (where one's duty is still that which a properly motivated person would have done from inclination).

Next, we can take this distinction between acting from duty and acting from inclination and apply it to the ongoing discussion of Rosalind Hursthouse's theory of right action. It supports the conclusion that "acting from duty" does not have the moral priority that Hursthouse, Kant, and Aristotle attribute to it, and that right action still follows the Humean model of that which a person with good motives and lacking bad motives would have done in the circumstances.

Children, Animals, and Adult Virtues

Rosalind Hursthouse defends her theory of right action in part on the basis that it provides an account of the difference between the application of moral concepts to children and animals as distinct from adults.

I wish to argue that desires provides a better account.

I want to start with a lengthy quote from Hursthouse giving her view.

In the Eudemian Ethics Aristotle says, with the other animals the action on compulsion is simple (just as in the inanimate), for they have not inclination and reason (logos) opposing one another, but live by inclination; but man has both, that is at a certain age, to which we attribute also the power of action; for we do not say that a child acts, or a brute either, but only a man who does things from reasoning'. So, in Aristotelian terms, we could say that the happy philanthropists, supposing them to have 'Humean' benevolence as described, do not act in the strict sense of the term at all. They live kata pathos, by inclination, like an animal or a child; their `doings' issue from passion or emotion (pathe) not 'choice' (prohairesis). And here is the sense an Aristotelian may attach to the Kantian claim that their 'actions' (in the broad sense) lack genuine moral worth because they act from inclination not from duty. It is actions proper, which issue from reason, that are to be assessed as virtuous or vicious), but their 'doings' are not actions, and thereby cannot be said to be, and to be esteemed as, virtuous ones.

In short, the difference between children and animals on the one hand and adults on the other is that children and animals act on inclination (passion) alone, and morally responsible adults act on both passion and reason.

In contrast, I have argued for a Humean conception of motivation whereby the passions alone provide motivational force and reason simply selects the means to these ends. The passions are sovereign - determining the ends or goals of intentional action. Reason, in Hume’s words, is the slave of the passions. Charged with realizing an end or goal, reason determines the best way to reach or hold this objective.

How can this Humean account handle the difference between children and adults?

First, the adult has a better capacity to select the appropriate means to ends. A child can carelessly inflict unnecessary harm - for example, with brutal honesty or choosing ineffective or even counter-productive ways to help. Animals lack the capacity to even understand the distant effects of its behavior.

Second, the passions and of a child are not the passions of a mature adult. An element of growing maturity is the adoption of cultural norms. Among these are @ willingness to share, an aversion to the use of violent assault, an aversion to taking the Roberts of others without their consent, and aversions to damaging the property and to inflicting unnecessary distress on others. In shot, the adult, like the child, continues to act on passions. However, through social conditioning, the adult has acquired better passions - passions that, in Hume’s terms - are pleasing and/or useful to self and/or others.

F course, we combine these two elements to give the agent the capacity to select better means to the ends established by these improved improved

We can combine this with the fact that adults also have an improved capacity to recognize that a passion is or is not a passion that people generally have reason to promote universally. It is not the case that adults somehow magically acquire these new and improved passions. Instead, we are taught these new and improved passions by others, who have reasons (based on their own passions) to cause us to have these interests - aversions to deception, aversions to taking property without consent, desire to help others in need. So, in addition to having adult as opposed to childish passions, and an improved capacity to recognize the relationships between actions and the fulfillment of those passions, there is a recognitioin of the passions that people generally have reason to promote universally (in virtue of being useful and agreeable to self and others).

This view is quite different from the view that Hursthouse seems to be arguing against - the idea that the passions of the child (or the animal) are the only possible passions, and that the alternative to her theory is one that adds means-ends rationality to the fulfillment of childish (or animal) passions. That view definitely deserves criticism, but it is not a view that fully appreciates the complexities of Humean morality.

I have included in this "passions that people generally have reason to promote universally" a desire to do one's duty. This is still a special kind new and improved passion. I want to talk about it in my next post.

Acting from Inclination

In recent posts, I have been examining a distinction that Rosalind Hursthouse draws from Aristotle that she describes in terms of "acting from inclination" versus "acting from duty".

"Acting from inclination" can best be understood as doing what one wants.

"Acting from duty" often means going against one's desires to do that which is right.

In recent posts, I have been arguing that all actions are examples of "acting from inclination". The distinction is actually between those who have an inclination to do that which is right (a desire to do that which is right) and those who have other desires. I draw a distinction between, for example, an aversion to taking that which belongs to other people without their consent, and an aversion to doing that which is wrong accompanied by the belief that it is wrong to take the property of another without consent.

Here, the issue gets a little confusing because Hursthouse does not talk about acting from desire - but acting from emotion. While it is the case that emotions can be connected to action, I would argue that this happens when the emotion is associated with a desire - which provides a reason for action.

Hursthouse writes:

In short, the emotions of sympathy, compassion, and love, viewed simply as psychological phenomena, are no guarantee of right action, or acting well. (Rosalind Hursthouse. On Virtue Ethics (Kindle Locations 1254-1255). Kindle Edition.)

Of course, this is true. It is also true in the desirism account. There is nothing in the desire to tell the truth, or the aversion to causing harm, or the aversion to taking the property of others without consent that guarantees that one will perform a right action. One of the reasons for this is precisely the issue that Hursthouse discusses - the act that one seeks to perform is unreasonable. She brings up the example of a person for whom, "compassion misguided by a misconception of 'good' may prompt someone to lie rather than tell the hurtful truth that another needs to know."

A more stark example would be the case of a parent who kills a child to protect the child from a demon, or who tries to "beat the devil" out of the child. We may include in this the person who, apparently out of charity, opposes globalization - unappreciative of the fact that globalization has resulted in a substantial decrease in extreme global poverty.

The examples listed here are examples of failure of means-ends rationality. The failure to tell a truth that a person "needs to know" is a failure to select the appropriate means to acting compassionately - assuming that the person "needs to know" this truth because it would ultimately be best if she knew it. Killing a child to protect it from (non-existent) demons or destroying the best tool for reducing global extreme poverty out of "compassion" are examples of selecting - due to ignorance - an inappropriate means to a good end.

So, Hursthouse is correct when she writes:

There is nothing about [the emotions of sympathy, compassion, and love, viewed simply as psychological phenomena], qua natural inclinations, which guarantees that they occur 'in complete harmony with reason', that is, that they occur when, and only when, they should, towards the people whose circumstances should occasion them, consistently, on reasonable grounds and to an appropriate degree, as Aristotelian virtue requires. (Rosalind Hursthouse. On Virtue Ethics (Kindle Locations 1254-1256). Kindle Edition.)

When a person selects the inappropriate means to an end, we often have reason to suspect that the agent does not have the particular end in question. A parent who does a poor job of protecting her child from harm - who does not take reasonable steps to determine what threats are real and which are imaginary, and whether a certain action will cause harm or prevent harm - we may suspect does not really care about whether the child is safe. A parent who refuses to vaccinate a child or to seek medical care for an easily treatable disease, we may suspect, cares about things other than the health of the child.

We have no direct access to a person's ends. We only have a means to infer the ends that best explain and predict observed behavior. If observed behavior is behavior that puts a child at significant risk of harm or cause actual harm, then this is to be taken as evidence that the agent does not have a sufficiently strong aversion to her child suffering this harm. This gives us reason for moral condemnation.

The implication that "having good desires and lacking bad desires" has on beliefs implies that, even with the standard of right action in use - an act is right iff it is the act that a person with good desires and lacking bad desires would do in the circumstances - we are not going to simply accept an agent's beliefs without question. An agent with good desires and lacking bad desires are going to be inclined to do research and to form responsible beliefs. Consequently, desires alone do not determine right action. Desires combined with the beliefs that an agent who has good desires and lacks bad desires would adopt are what determines right action.

Types of Reasons vs Degree of Difficulty

In her comparison of Aristotle and Kant on virtuous action, Rosalind Hursthouse seems to confuse a pair of distinctions. One is the distinction between acting from inclination and acting from duty. The other is the distinction between performing an act easily or performing it with difficulty, the idea seems to be that performing an act from inclination is easy or pleasurable, while acting from duty is difficult and painful.

I described this distinction in Humean terms since I hold that Davis Hume’s account of motivation is basically correct. We act in accordance with our beliefs to fulfill our desires.

So, the first distinction is a distinction between the desires one acts on. The person who visits a sick friend in the hospital because “you are my friend; I care about you,” acts from inclination. He gives only a passing thought to duty or right action. Instead, he does that which he values for its own sake.

This is to be held in contrast to the person who visits a sick friend from a desire to do that which is right, and a belief that visiting a sick friend is the right thing to do.

Hursthouse would object to this way of drawing the distinction. She would count the first as acting from desires, and the second type as acting from reasons other than desires - acting contrary to desires. This may be why she confuses this distinction with the distinction between easily performing a right action (acting according to desire) and performing it with difficulty (contrary to desire).

Consider the case of the student trying to decide between pursuing a passion - e.g., philosophy - versus learning a useful skill such as computer programming or database management. These are matters to of inclination, but that does not imply that it is an easy decision.

On the other side of the coin, consider the person who follows a religious commandment. He chooses what to eat based on an interpretation of scripture - foregoing bacon, for example. This is a case of acting from duty (though the belief that duty prohibits eating bacon may be a false belief), yet the devout individual may have no difficulty at all following this prescription.

The phenomenon of ease versus difficulty has nothing to do with the type of reason one is acting on, but on the presence or absence of counter-weighing desires - with reasons to do otherwise. The person who is struggling to decide on a career has conflicting desires - pulling her in two directions. The person who refuses bacon without difficulty has a interest in doing what her religion commands of her and no particularly strong interest in eating bacon. After all, there are a great many other foods that taste just as good.

This confusion colors Hursthouse's discussion of the virtues of courage and charity.

Courage, it seems, is a term that is only applicable when there is a conflict between competing desires. There must be a threat to the life or well-being of the individual being outweighed by a more important concern. The parent's aversion to pain, injury, or death - however important they may be - become outweighed by their interest in saving her child from a fire. The soldier's devotion to his country (though, in fact, soldiers tend to be more concerned with protecting their brothers in arms than their country) outweighs his own interest in life and limb. These are the standard examples of courage. If no self-regarding counter-weighing desires are in play, then the concept of 'courage' does not apply.

This distinguishes courage from honesty or charity. Though honesty and charity often will require overriding self-regarding interests, this is not necessarily the case. A person can be honest when no self-regarding interests are at stake, and she can be charitable even when she has pleanty - or she has little but also does not want much. The person who is content to get by with some small amount of wealth and give the rest to charity - simply because she wants nothing more than to help others (once her own basic needs are taken care of) is a prime example of a charitable person. With courage, conflicting self-regarding desires are a necessity. With other virtues, they may exist, but not necessarily.

At this point, I need to mention that I have difficulty with the idea that courage is a virtue. Crimes from terrorism to rape to armed robbery require courage. Whether courage is a virtue depends on whether the ends that one is courageous in advancing are those that a virtuous person would advance. The person who advances a good cause where a less timid person would not is the virtuous-courageous person. The reason for praise is because "courage," in this case, demonstrates the strength of her good motives . It is for the sake of strengthening those good motives that we praise her courage. Whereas courage in the heart of an immoral person - a person who fails to pursue good motives or who does so in ways that require courage but which a good person would not pursue - warrants no praise.

So, let's look at charity instead. Who is the most charitable - the person who contributes to helping others out of inclination (a desire that others be well), or out of duty (a desire to do what is right combined with a belief that helping others who are disadvantaged is the right thing to do)?

Hursthouse compares several types of charitable actions, in order to intuitively measure their moral worth.

The first is the person who helps others out of a desire to help others - an interest in their well-being. This is a person who acts "from inclination" as it were - in the Aristotelian sense (though Aristotle did not argue that charity was a virtue).

The second type is the person who helps others out of a sense of duty. She doesn't care about others. However, she cares about doing the right thing and believes that donations to charity are the right thing to do. Perhaps she is following a religious tradition and mechanically - unthinkingly - gives 10% of her income to charity as a rule. However, when she hears about the suffering of others due to a natural disaster, poverty, disease, or injury, she really could not care less.

Among these two, it seems that the first person better embodies the virtue of charity. Charity is concerned with having an interest in the well-being of others, not an interest in "doing the right thing".

Then, Hursthouse brings in the Kantian philanthrope - the person who is suffering her own set-backs and is in a state of depression or sadness. She is more focused on her own problems - an ailing parent, the death of a child, or her own life-threatening illness - which fixes her attention so that she "has nothing left emotionally" to spare for others. Yet, she continues to help out. Hursthouse argues that the Aristotelian should agree that this person displays the virtue of charity.

Yet, I see this as confusing the question of whether the charitable act is difficult with the question of whether it is done from inclination or duty. It may well be that the sorrowful philanthrope had gotten into a habit of making a contribution to charity out of duty as described above, and the change in her situation simply does not motivate her to change her habits. Or she could have been somebody who gave to others because she cares about their welfare and, when her own situation turns bad, says, "It's not their fault that my situation has changed - and they should not suffer the lack of charity because of it. At least some good will come from my actions." The change in the agent's circumstances may give us reason to evaluate the strength of the agent's concerns, but it does not eliminate or override the distinction between acting from concern for others and acting from duty.

Hursthouse also confuses the distinction between types of reasons and degree of difficulty when talking about the merit of returning a purse one has found full of money. In her evaluation of this case, Hursthouse's conclusions are correct. The wealthy person who returns the money does not display as much virtue as the poor person who returns the money. This is because we are looking at whether the action displays that the desires are present in the required strength. The rich person who returns the money does not give us much evidence of this (though the rich person who refuses to return the money does tell us a great deal about her level of averice). The poor person who returns the money - who is struggling to feed her family, provide medical care, and the like - does show us that her interest in returning the lost money is sufficiently strong to outweigh these other concerns. Yet, even if she fails to return the money, and blows it instead on some personal luxuries, this would give us added reason for condemnation.

All of this is perfectly compatible with the desire-based theory of value mentioned above. It still provides no evidence for people acting on different types of reasons - desires versus some type of rationality-based reason. The desire-based theory can handle these types of cases without complicating the metaphysics.

Acting from Duty or Acting from Inclination

In our previous postings

For those who do not care to go back and read what came before (and I do not blame you - your time is precious), I offer a brief context for this discussion. Those who have read the earlier posts can skip to the section on morality and the emotions.

In my previous post, Virtues, Right Actions, and Reasons, I identified what I see as a key difference between my own thesis and Rosalind Hursthouse's virtue theory. It is also a key difference between my own thesis and Kantian deontology.

Hursthouse identified it in the following quote:

We should not forget that Kant and Aristotle significantly share a strongly anti-Humean premise about the principles or springs of movement (or `action' in the broad sense of the term). According to Hume, there is only one principle of action, the one we share with animals, namely passion or desire; according to both Aristotle and Kant there are two, one which we share with the other animals, and one which we have in virtue of being rational.

In this dispute, I am a Humean. That is, I side with the Humean model of intentional action. Desires provide the motivating force behind intentional action, and beliefs merely select the means. There is no "second reason" for intentional action.

In the previous post I defended the thesis on the grounds that it makes the most sense given human evolution. We evolved a disposition to learn to like that which - in liking it - our ancestors were disposed to have fit offspring. The aversion to pain, tastes for food, thirst, the desire for sex, parental affection - we came to like these things because it allowed our ancestors to successfully reproduce.

We also inherited a "plastic" mind that can learn new desires based on our interactions with the environment. However, even this learning process became molded by evolution to dispose us to acquire preferences that made us fit for survival.

And what makes a population fit for survival is a contingent fact - based in part on the environment in which one's ancestors evolved, including the other types of animals, plants, and diseases that inhabited the region. Whatever it is we have come to value - or are disposed to learn to value - it could have been otherwise.

That is one line of argument. The other is that such a theory can explain and predict how people actually behave.

To make good on this idea, I would like to look at some of the claims that Hursthouse made in defense of Aristotle and Kant (in defense of the virtue theory and deontology account of reasons for intentional action) and show that the Humean system can handle these just fine without all of the metaphysical complications.

Acting from Duty or Acting from Inclination

Hursthouse presents a dispute between Kant and Aristotle regarding the moral value of acting from duty.

She begins by presenting an interpretation of Aristotle that goes as follows:

At the end of Book I of the Nicomachean Ethics, Aristotle introduces a distinction between the `continent' or `self-controlled' type of human being, (who has enkrateia) and the one who has full virtue (arete). Simply, the continent character is the one who, typically, knowing what she should do, does it, contrary to her desires,' and the fully virtuous character is the one who, typically, knowing what she should do, does it, desiring to do it. Her desires are in `complete harmony' with her reason; hence, when she does what she should, she does what she desires to do, and reaps the reward of satisfied desire. Hence, `virtuous conduct gives pleasure to the lover of virtue' (io99aiz); the fully virtuous do what they (characteristically) do, gladly. (Rosalind Hursthouse. On Virtue Ethics (Kindle Locations 1131-1136). Kindle Edition.)

Her interest in discussing this topic is to investigate an apparent contrast between Aristotle and Kant. According to the standard view, Aristotle believes that acting from "full virtue" (arete) is the morally superior individual. This is contrasted to Kant's view that moral merit is to be assigned the 'continent' or 'self-controlled' person - the agent who acts from duty.

It is in the context of this discussion that Hursthouse brings up the idea that on one point where Aristotle and Kant agree is that it is possible for an agent to act from something other than desire - that there is a second type of reason for intentional action, namely, rationality itself.

What I want to do is to show how desirism can handle this distinction without mentioning a "second type of reason" for intentional action.

Before going on, I would like to present what desirism says about this distinction.

Let us take an example of a person who finds a wallet with some money laying in the curb laying in a parking lot. One who "has full virtue" in the sense described above has a weak interest in keeping the wallet. She has desires, and knows that the money in the wallet would help to fulfill those desires, but the desire to return to others what they have lost is sufficiently strong that she does not give the idea of keeping the money a second thought. She would not even accept a reward from the owner of the wallet when she returns it for she has done nothing other than what any decent person would have done. Her desire is to return the property to its rightful owner.

I would like to compare this to a person with a different desire. This person has no interest in returning things to their rightful owner. Instead, she has a desire to do the right thing, accompanied by a belief that returning the wallet is "the right thing". For her, the question, "How can it be the case that 'returning lost property to its rightful owner if possible' is true?" can be a perplexing question - one that might drive her study moral philosophy.

Both of these agents are acting from desire. In one case, it is a desire to return lost property. In the other case, it is a desire to do that which is right (that which is in accordance with duty) and a belief that one has a duty to return lost property.

However, the relevant difference here is that, when returning the wallet "because it is the right thing to do" is different from returning the wallet from a simple desire to return property to its rightful owner.

In the terms that Hursthouse uses, the first person "acts from inclination". The second person "acts from duty". Neither agent acts from duty alone. If not for the motivational force provided by the relevant desire, neither agent would act.

It is possible for each agent to act rationally or irrationally. Each agent, wanting to return the wallet (either to fulfill a desire to return the lost property or the desire to do what is right) could decide to go up and down the street asking each person they met if they lost a wallet. Or they could open the wallet, find some identification, search for a phone number for the person identified in the wallet, and try to contact the owner that one. One option is more rational than the other. But they are more or less rational in terms of matching means to ends, not in terms of selecting ends by reason alone.

So, using the Humean system we have an account of the two types of character - distinguished by having to different desires, and a concept of more-or-less rational action given those desires. We have no reason to over-complicate things by adding a second type of reason for action.

Those other types of reasons do not exist. And those who call an action right or wrong - or an agent virtuous or lacking virtue - based on these "other types of reasons" are grounding their conclusions on a false premise. Their system for determining which moral attitudes are true or false is unsound.