Tuesday, July 24, 2018

The Deliberator

In my last exciting post . . . okay, actually, the post before that . . . I began a discussion of the ideas that Nomy Arpaly and Timothy Schroeder put in the book In Praise of Desire.

It started off with the idea that deliberation is an act - it is something that we do - like turning on a light switch or getting something to drink out of the refrigerator. It is an action - an intentional action - which means it is backed up by beliefs and desires the way other actions are.

But what is it? What are we doing when we deliberate?

I am going to try phrasing this in words of my own choosing. Arpaly and Schroeder do not write in these terms, but I think it captures the ideas that they are trying to get across.

We put our beliefs in a deliberator, turn in on, and hope to get better beliefs as a result, which we then attach to our desires, in order (hopefully) to get better actions as a result. That is, in the hopes of getting actions that do a better job fulfilling our desires.

As an example, the authors speak of adding 3545 and 869. We feed these two numbers into the deliberator and then we run them through a deliberation.

we focus our attention first on the sum of 5 and 9, and keep in mind that the result ends in 4 (a conclusion that, itself, can be reached without the aid of deliberation), and then sum 1, 4, and 6…and so on, then we can perform a sequence of mental actions with the final result being a correct belief about the sum of 3545 and 869.

The deliberator does not always work.

For one thing, the quality of what we get out of a deliberator depends on the quality of beliefs that we put into them.

For another, it is up to us to maintain these deliberators. Those people who let their deliberators get dirty and clogged up by gunk and nonsense do not produce as good an output as those who keep their deliberators clean, well oiled, and its blades sharp.

This point is interesting. Since I was young, I have had a bit of a reputation for doing math quickly in my head that others struggled with. In doing the math in my head, I don't do it the way I was taught in school. I don't start on the right and work my way to the left. I start at the left and work my way right. "35 + 8 = 43. 4 + 6 = 10. 430 + 10 = 440. 9 + 5 = 14. 4400 + 14 = 4414.

But that just illustrates the point. That illustrates the practice of using a deliberator to deliberate.

In another of their examples, Harold's son calls to say he will be in Calgary on Tuesday and asks about getting together for lunch. Harold thinks about it for a moment. He puts "tuesday" and "lunch" into the deliberator. The deliberator searches memory for beliefs relevant to Tuesday and Lunch. It comes back with Planning Council meeting at 1:00 PM and not enough time to get to the meeting after lunch.

This is one of the things that the deliberator is good at. Put a belief in the deliberator, and it can conduct a search for other relevant events.

Again, it's not perfect, but it is better than nothing.

Among the facts that the agent can feed into the deliberator for processing includes beliefs about the agent's own desires - what the agent wants. An agent with a belief that he has a desire that p can focus on p and the deliberator will search beliefs for any facts relevant to p - like, things that might bring about p or prevent p from being realized. In this way, deliberating helps in the fulfillment or at least helps us in avoiding the thwarting of our desires.

I think . . . the authors do not comment on this issue . . . but I think that the deliberator can only hold beliefs, not desires, though it can hold beliefs about desires. Thus, the agent can believe that he would want to go to Hawaii, even though he has no interest in going to Hawaii, and thus work out how best to get to Hawaii. However, even though he makes these wonderful plans, he just never gets around to it because, let's be honest, "Hawaii is hot and crowded and costs way too much money and why would I want to go there anyway?"

Either way, that is what the deliberator is for. We act so as to fulfill our desires, given our beliefs. We have the capacity to feed our beliefs into a deliberator in order to try to get better beliefs. It proves useful in figuring out how to get the apples down from the top of the tree, how to trap and kill a rabbit, and how to build a spear. Once the deliberator evolved, we put it to good use.

"Objective" "Morality"

A member of the studio audience directed me to a YouTube video on objective morality.

I wrote a response to the author of the video.


Your recent YouTube videos concerning objective morality has drawn some attention my way from people wondering how I would respond to your arguments.

I am an atheist who defends objective morality on my blog “atheistethicist”, which explains their interests. I thought I would share that response with you.

Before I start on that, however, I want to clear away a potential mistaken assumption that would get in the way of understanding my response. This is the thought that, in responding to your argument, I am defending Sam Harris.

Objections to Harris (and most Moral Realists)

Harris’ argument is also flawed. Harris’ argument contains an invalid inference from, “I have an aversion to my own pain,” to “you and everybody else has a reason to prevent the realization of a state in which I am in pain.” It follows from the fact that I have an aversion to my own pain is that I have a reason not to put my hand on a hot stove. It follows from your aversion to your own pain that you have a reason not to put your hand on a hot stove. You do not, necessarily, have a reason not to put my hand on a hot stove. I could try to give you a reason - e.g., by threatening or bribing you. However, the reason does not exist automatically.

The antelope’s aversion to pain means that it has a reason to run from the lion. It does not imply that the lion has a reason to forego chasing the antelope.

It is true of almost all atheists who defend an objective morality that they make this invalid leap from “I have a reason” to “there exists a reason valid for everybody”. They waste a great deal of effort trying to justify this leap. They all fail. So, criticism of almost all atheist moral realists is well founded.

Almost all.

It would be a mistake to try to interpret my remarks as containing or defending that invalid inference.

So, now, my criticism of your argument . . .

Objections to Your Argument

You said in your video:

"I am not disagreeing with Harris that it is objectively true that we all experience pain as bad. I just think that this doesn't warrant us to define bad as that which brings about pain."

The problem is: no definition has a warrant. There is no objectively correct definition of anything. Definitions are, by their nature, determined by social convention. Consequently, you cannot use the fact that a definition lacks warrant to prove anything about objectivity or subjectivity.

Take, for example, the word “atom”.

What “warrants” defining the term “atom” as the smallest unit of an element still recognized as a unit of that element - the smallest piece of gold, oxygen, or mercury? The term ‘atom’ could alternatively been made to mean ‘without parts’. Indeed, that was its original definition. It came from ‘a’ (without) + ‘tomos’ (cut). By way of social convention, people changed the meaning of the term. As a result, the proposition "an atom cannot be divided further into smaller parts" went from being true by definition to being false. What was 'true' at one moment in history was 'false' a few years later - simply because people decided to define a term differently. Yet, the objectivity of chemistry as a science was never threatened.

How did the meaning of “planet” change? The members of the International Astronomical Union took a vote. With a vote, the proposition “Pluto is a planet” went from being objectively true (those in school who failed to identify Pluto as a planet were wrong) to objectively false. It went from true to false because people changed their minds about the definition of a term. Yet, astronomy remains a hard science, filled with objective facts. How is that possible?

Indeed, we can say the same thing about calling an element ‘californium’ or ‘einsteinium’. What “warrants” giving it that name? There is nothing that "warrants" defining 'einsteinium' as 'the element that has 99 protons in its nucleus'. It was decided almost on a whim. Yet, the proposition "einsteinium has 99 protons in its nucleus" is objectively true - and nobody uses the fact that this definition has no warrant to argue that chemistry is subjective.

As soon as you start talking about the definitions of term you are talking about things that are true only in virtue of social convention. This is no less true in chemistry and astronomy and, indeed, in logic as in morality. If this is a reason to claim that statements made in a field are subjective, then all statements in all fields are subjective, and moral statements are no different from statements in chemistry or astronomy in this respect.

Scientists have long recognized that definitions do not matter. What we call something does not determine what it is. Because they respect this fact, they have fun with definitions, giving quarks names like “strange” and “charm” or even adopting whimsical definitions, such as defining the term "eriovixia gryffindori" (named after a fictional wizard) so that it refers to a species of spider.

Criticizing Definitions

Having said this, we can often find a problem when people claim to be offering a definition of morality. The problem is not with their definition, it is with their reference. The person who says, "I am going to define 'good' as that which maximizes the welfare of conscious creatures" or "as that which maximizes overall utility" or "as that which people would agree to behind a veil of ignorance" or whatever are not actually giving us a definition. They are giving us a reference for a term that already has a different definition.

What they are actually saying is, "I define good as that which everybody should promote for its own sake," and then they add that this definition refers to whatever it is they are promoting as the moral good. This is not legitimate. If the author is going to define good as "x", then whether or not the term refers to something in the real world has to be discovered, it cannot be stipulated. You can, of course, define a term by referring to something - e.g., by saying, "by 'eriovixia gryffindori' I mean the species that thing belongs to", but then one has to discover what is true of the thing being pointed to. The definition cannot stipulate that it is anything other than "the thing being pointed to".

So, yes, it is possible to criticize definitions. However, one cannot criticize a definition as lacking warrant (all definitions lack warrant). One can criticize the use of a definition as lacking consistency, or as lacking a referent, or as assigning a property to its referent that the referent does not have.

I could go on. Like I said, I believe in an objective morality. All I have done here is criticize the arguments of others. It is always easier to criticize than to create. However, I also recognize that your time is limited, so I will leave the discussion here and see if you have an interest in continuing it further.

Monday, July 23, 2018


I have been reading In Praise of Desire by Nomy Arpaly and Timothy Schroeder.

The book defends a desire-centered theory of moral psychology which, at least at this general level, is quite relevant to my interests. I think that I am going to end up disagreeing with them on the role of praise and condemnation, and on what a good (or bad) desire happens to be, let alone a right and wrong action. However, in general, their claims on the psychology of desire seem to be consistent with the ideas that I have presented here.

Furthermore, they extend these ideas into areas that I have not considered.

One such area is "deliberation".

The authors raise this subject in the context of discussing the idea that, for an action to be done for a reason, it must be brought about through deliberation. The authors, instead, want to argue that reasons can cause action without deliberation being involved.

To make their case, their first premise is that deliberation itself is an action. Deliberating on an action is something we choose to do - or choose not to do - just as we choose to fix ourselves a peanut butter and jelly sandwich or to work on a philosophy blog on the nature of desire.

Evidence that deliberation itself is an action is found not only in the fact that it is something we can choose to do or not do, but also in the ways we may not have such a choice. In the same way that alcohol can impair one's ability to drive home, alcohol can impair one's ability to deliberate. In the same way that a boss can command a person to modify a website, a boss can command a person to present his ideas on the merits of a new in-house procedure for employee promotions.

If deliberation is an action, this has additional implications that Arpaly and Schroeder do not go into in detail - though they mention it in passing. We can be morally responsible for our deliberations. We can be responsible for failing to deliberate when we had an obligation to do so and, when deliberating, for deliberating well or poorly (responsibly or irresponsibly). These are welcome implications within the context of this blog which has tried to stress epistemic responsibility. (If the reader has not seen this blog as stressing epistemic responsibility, then I need to try harder, because I consider it one of the greatest moral failings of our age - a responsibility on which the very survival of humanity may depend.)

After extablishing the idea that deliberation itself is an action, this raises a problem for the thesis that an action requires deliberation. It results in an infinite regress. An action requires deliberation, which is an action, which requires deliberation, which is an action, which requires deliberation. Arpaly and Schroeder consider a number of ways in which "action requires deliberation" can avoid this infinite regress, and comes up with responses to each of them.

The consequence of this argument is that an action (or, more recisely, acting for a reason) does not require deliberation. At the very least, deliberation (which is acting for a reason) does not require deliberation. From here, the authors argue that deliberation is a paradigm type of action with no need to assume that other forms of action are somehow different. This means that other types of action (or, more precisely, other cases of acting for a reason) do not require deliberation.

Desirism has always included under its account of an intentional action any action performed in service to a desire. It doesn't require deliberation of any type. The antelope who notices movement in the brush, who gets a little nervous, and decides to wander away from the line of brush further into the open field, is not deliberating on the value of moving away from the bushes. She simply feels uncomfortable and decides to move away. A person walking into a room and turning on the light so that he can see doesn't deliberate. He turns on the light so that he can see.

Indeed, the prime example that the authors wish to bring to our attention is the act of deliberating itself.

Deliberate on whether morality can be objective.

A person will start to deliberate without first deliberating on whether he should deliberate about morality being objective. And he certainly does not need to deliberate about whether to deliberate on whether morality being objective. Instead, he simply starts the task of deliberating on whether morality is objective. This is true in the same way that he simply turns on the switch when he sees that the room is dark, or avoids the dark alley that makes him feel a little anxious and apprehensive.

Another clear example of where a failure to deliberate is not necessary for action concerns acts of negligence. A person picks up a gun, points it at another, and pulls the trigger. He fails to properly consider whether the gun might be loaded. His action is intentional - in the sense that he may be blamed for the action (even if he does not end up shooting somebody). Yet, much of the moral blame rests precisely on the fact that he failed to deliberate - he failed to consider the possibility that the gun was loaded. The thought, at least in our example, did not cross his mind. Yet, the failure of the thought to cross his mind is precisely that for which he is being condemned.

So, as far as I can tell, we should grant this . . . intentional action can have causes that are not, themselves, deliberations. They are simply interactions between desire and belief that do not require thinking about. This is not to say that we cannot think about things. This is not to say that we should not think about things. This is only to say that, when one performs an action, that the action does not have to result from deliberation. The agent can simply act, and yet the act is still intentional, deliberate, and one for which she may be held morally responsible.

And, indeed, deliberation itself, or fail to deliberate, or failure to deliberate responsibly, can be something for which an agent can be morally responsible.

So, if deliberation is an action, and it does not require deliberation, what does deliberation require? And are these the same requirements that apply to other types of action?

I will discuss these concerns shortly.

Thursday, July 12, 2018

Martin Luther King and "Being Civl"

I got into a dispute recently concerning the views of Martin Luther King and being civil.

The dispute concerned the current "call for civility" by those who are currently being condemned (or on behalf of those currently being condemned) by those who are being accused of engaging in and promoting near-Nazi like wrongs.

Among these crimes is that of calling immigrants "vermin" who are "infesting" our country - identifying them as a race of murderers and rapist and putting them in cages for our own protection.

These are people who, though they seem to insist that the people only express themselves in the polls, have rigged the elections through systems of gerrymandering and restricted voter access, thus denying them a fair expression of their opinion in the only venue where their venue has any power.

These are the same people who have captured the judiciary - a judiciary that has recently shown that it is more disposed to favor those who rig elections for their benefit using the means mentioned above, giving unfair advantages to those with power over those who are supposed to be controlling the government through the process of elections. For some reason, these voters who are being made increasingly powerless and irrelevant - increasingly unable to express their opinion through elections - are supposed to remain docile, obedient, and deferent to those who hold power.

These are the same people who have virtually sold our government's power and prestige to a hostile state, where, if it is not the case that Moscow is dictating the course of American foreign and economic power, we must assume a bizarre coincidence between the policies that President Trump pursues and that which serves Putin's interests.

Of course the people who are the perpetrators of these wrongs are going to tell those who are upset about it to "be civil".

Yet, the question under debate is: What would Martin Luther King say? This is the same Martin Luther King who lead protests against an unfair and unjust state. There are many who point to King as a role model, while they themselves call for an end in civility in a call to protest these wrongs.

I do not want anybody to think that I have any type of authority to speak for Martin Luther King. All I can do is offer an analysis based on my understanding of his words and deeds. These may be in error, but this is how it appears to me.

I think that King would agree with thee calls of civility. He would begin by saying that those who demand civility are right to do so. Civility is a core part of civil society.

But what is civility?

Civility is treating others as they ought to be treated. Civility goes hand-and-hand with justice. The person who violates the call for civility is the person who treats his neighbor as somebody less than human. The person who lacks civility is the person who thinks that he has a right to stand above his fellow human being, as opposed to the person who thinks his rightful position is at his neighbor's side.

Yes, civility is important. Yes, it is important to join the call for civility and to let that call be heard from the rooftops and through the streets.

Where will you find civility?

You will find civility in the government that recognizes that its job is to serve the people, not the people's job to serve government, and, recognizing that fact, recognize that it is their solemn duty - it is the very definition of their proper role as a civil servant - to make sure that the government expresses the will of the people. The person who thinks that his rightful place is to stand above them, look down upon them, and command not only their obedience but their deference, these are the demands of those who lack all civility.

You will find civility in the arms of the person who takes a family that is fleeing the ravages of poverty and violence, wrap them in a warm blanket, provide them with food and shelter, and then see what you can do to help those who come to you with need. These are the deeds of the civilized person. The person who cast them out - who slams the door in their face and tells them to go away, this is not civilized.

Yes, we want more civility. Yes, we need more civility. Yes, people ought to be given the respect that they deserve. That is why we are here.

Having said this, there is another aspect of King's philosophy that we must look for.

I believe that King would not be interested in any action taken out of malevolence and hatred. Even punishment and condemnation is done out of love and concern. Think about a parent punishing a child. The demands of civility do not require that the parent ignore a child's transgressions - refusing to condemn and correct the child's behavior when it is wrong. The parent, out of love for his child, out of respect for his child, corrects his child's transgressions, not only for the sake of others in society, but for the sake of the child. It is done out of love for the child. It is done to make the child a better person.

Just as the loving parent will step in and educate and correct the transgressions of the child - not out of hate and malace, but out of love - our duties to a civil society demand of us that we take steps to correct the injustices and wrongs that we see in our community. We do not do so out of malice and hate - that would be wrong. That would be as much a stain on our character as on the character of those we hope to educate. We aim to correct these injustices out of love - out of a hope that we can all of us, equally, enjoy the benefits of living in a fair and just society, in a community where all of its members treat each other with the dignity and respect that all human beings deserve.

Tuesday, July 10, 2018

Assignment vs Dispositional Theories of Desire

There is a battle being fought between competing camps concerning what a desire is.

I have recently discussed the position that the Humean dispositionalist theories are defending. This is the position that a desire that p is a disposition to act so as to make it the case that p.

I have agreed that there is a relationship between a desire that p and a disposition to act. However, the relationship is that of cause and effect. The desire causes the disposition, it is not identical to the disposition. In the same way, rain causes plants to grow, it is not identical to the growing of plants.

I have three major objections to dispositional theories.

First, there are clearly dispositions to act that are not desires. There are habits. You can perform this experiment at home. Reprogram your computer keyboard to switch the 't' and 'r' keys. If you are at all skilled at typing, you will find yourself typing 'tre' when you intend to type 'the'. This is an intentional action - an action as intentional as typing 'the' was before reprogramming the computer. Furthermore, you have the requisite beliefs. I am assuming that you programmed the keyboard yourself and know that you switched it. On the dispositional theory of desire, you must desire to type 'tre'. However, that is false. So, there are dispositions to act that are not desires. Reducing desires to dispositions to act fails.

Second, the disposition to act principle implies that having a desire that p is incompatible with having the belief that p. It is absurd to attribute to you a disposition to bring about p if you believe that p is already true. Consequently, the desire to bring about p must disappear the moment you come to realize that p is already the case. This is called the "death of desire principle". However, desires do not work this way. As I prepare supper, it is not the case that I have a desire to eat that suddenly vanishes the moment I start eating. Nor does my desire to cuddle with my wife vanish the moment that I start cuddling with my wife. Dispositionalist respond to this type of situation by saying that the agent has a new desire - a desire to continue eating or a desire to continue cuddling. Yet, even this is problematic. It requires that one desire vanish and a new desire suddenly emerge, rather than the more plausible assumption that the same desire survives the transition.

Third, dispositional theories become exceptionally complex very quickly as we add more desires. Assume that you had a balance with a number of weights. Let us assume that there are 26 weights lettered A through Z, each having a different weight. If we were to describe the effects of these weights dispositionally, the descriptions will be very complex. Weight T, when combined with weights H and J, will tilt the balance in its favor when countered by weights X and Z, but not when countered by weights X and W. We would need to express this type of conditional for all different combinations of weights, and that is assuming an accurate scale. It is far easier to attribute a value to each weight and simply say, "the balance will tilt in the direction of the side with the highest total weight." However, "assigning a value" to each weight is an evaluativist move. Evaluative theories are the leading opponent to dispositional theories. "Assigning a value" means abandoning dispositional theories in favor of some sort of evaluativist theory.

The assignment theory of desire can handle all three of these problems.

It does not call the typing of 'tre' a desire because the agent does not assign any importance to it being the case that he has typed 'tre', or any consequence of typing 'tre'.

It attributes a value to a state such as "I am eating" or "I am cuddling with my wife" that makes it important to the agent to make or keep the proposition true. This means making the proposition true while it is false, and keeping it true while it is true. These desires do not disappear once the agent realizes that he has realized his desired end.

By assigning a value - a weight, or a measure of importance - to different desires the evaluativist theory makes it easier to use desires to explain and predict intentional behavior. Or, more to the point, it explains the fact that we can so easily use desires to explain and predict intentional behavior.

For these three reasons, an assignment theory of desire is superior to a dispositional theory of desire.

Monday, July 09, 2018

Desires 2008: Summary 03: Maria Alvarez

Question: Why am I writing these summaries?

Answer: I intend to send each of these authors an email asking them one or two questions about their contribution. These summaries represent a first draft.

This summary concerns:

Alvarez, Maria (2017), “Desires, Dispositions, and the Explanation of Action." In Deonna J. & Lauria F. (eds). The Nature of Desire. Oxford University Press.


My claim is that it is a part of the concept of desire that someone has a desire at time t, only if the desire has been manifested in any of the various ways I described above at some point up to and including time t.

These "various ways I described above include not only physical action that aims at realizing that which is desired, but also “mere” expression such as a cheer that does not aim to bring about any particular end, or some sort of emotion or sentiment showing no outward signs. If none of these things happen, then the claim that the agent has a particular desire or aversion is false.


What causes the first manifestation? If there is no desire before the first deliberate action, expression, or sentiment, then what explains that first deliberate action, expression or sentiment?

Perhaps the desire is not supposed to explain these things - the desire simply is the set of actions, expressions, and sentiments. But, if this is the case, then does the desire exist between events? Or does it cease to exist when there is no such event, only to emerge again when another event pops up?

It appears the case that when we talk about the relationship between a desire and its manifestations that the desire is the cause of its manifestations. As such, the desire must exist prior to its first manifestation. There must be a desire in place to cause the first action, to bring about the first expression, or to explain why the agent had a particular sentiment. Indeed, there must be something to explain these things and ties them together, and if it is not a desire ( or aversion) then what is it? Is there any such thing?

It also appears to be possible to come up with counterexamples to this thesis.

I typically do not like exotic counterexamples, but I think that the inventions in this story serve to isolate elements that are, in fact, quite common.

Imagine a planet with a species whose members all have a strong fear of spiders - a strong aversion to being in the presence of spiders. This fear of spiders is genetic - a product of natural selection grounded on the fact that spiders on their planet were particularly deadly and a genetic encoding of a fear of spiders aided survival.

Let us further imagine that some members of this species head out on a space ship to colonize another planet. As a result of some catastrophe, they crash on the planet and enter a primitive state in which they forget all about spiders. For example, only the youngest children survive and are raised by wolves. This planet has no spiders, and these people never learn about them.

Yet, it seems we would be justified in saying that these people still have an aversion to the presence of spiders and, furthermore, this can be shown. Bring a spider to the planet and show it to them and watch them recoil from it. It seems odd to say that this aversion to spiders suddenly emerged when they came into contact with the first spider - for what would have caused its emergence? Instead, a more plausible description is that it existed all along, and would have continued to exist even if nobody ever brought a spider to the planet.

To draw a more common and real-worldly example from this, a person's aversion to pain does not seem to suddenly come into existence the first time they experience pain and have a reaction to it - it existed prior to the first experience of pain.

Saturday, July 07, 2018

Michael Smith: Three Counterexamples to the Humean Theory of Motivation

In "The Humean Theory of Motivation", Michael Smith defends the Humean theory of motivation from three counter-examples.

(Smith, Michael (1987), "A Humean Theory of Motivation," Mind, Vol. 96, No. 381.)

I consider this defense interesting, not because Smith fails in providing a defense (he does not), but the language that he uses in that defense. It is quite different from the language that I would use.

The dispute in one sense is purely semantic, it is a dispute over the use of terms. However, semantic disputes can have one important consequence. They can make things more or less confusing. So, if I were to say, "Donald Trump is a great president," and take it to mean "Donald Trump is a lying, hypocritical bigot who is unfit to hold political office," we could call this purely a semantic dispute, but it could yield a great deal of confusion.

Case 1: The Picasso

The first case involves an individual who wishes to purchase a Picasso painting. Let's call him Alph. Alph is offered a Picasso painting at a quite reasonable price. However, he does not believe that it is a Picasso painting, so he refuses to buy it.

Does Alph want to purchase this painting?

Smith says, "No." I say, "Yes".

To explain my "yes" answer, I ask the question of whether Alph has a desire that would be fulfilled if he purchased the painting. He does. Therefore, he wants to purchase the painting.

It is no mystery on my account that Alph will say that he does not want to purchase the painting. This is what he believes, and Alph is certain to say what he believes. However, the question of what Alph would say - even what Alph would sincerely say - is not the same as the question of what is true. The question, "Does Alph want to buy the painting" is a question about what is true.

I can supply some support for this answer.

Let us say that the salesperson, instead of selling Alph the painting, offers it to Alph in the form of a lottery. He takes a white stone and black stone, puts his hands behind his back, and puts a stone in each hand. He offers Alph a choice. "Pick the white stone, and you get the painting?" He then asks Alph, "Which hand do you want?"

Alph clearly wants to choose the hand with the white stone. He does not know which hand it is, but there is a clear answer to the question, "Which hand do you want?" that is independent of Alph's beliefs on this matter. The hand with the white stone.

Smith's answer is:

[T]he reason that I have to buy the painting in front of me is a normative reason. For it suffices for the truth of the claim that I have such a reason, that there is a requirement-in this case, in the broad sense, a requirement of rationality"-that I buy the painting in front of me.

Smith is distinguishing between motivating reasons (those that cause action) from normative reasons (good reasons). Clearly, Alph does not have a motivating reason to purchase the painting. After all, he was not moved to purchase it. Still, he has a normative reason to purchase the painting. This is the reason based on the standards of rationality.

However, this introduces a potential confusion. Alph might not be the least bit irrational in believing that the Picasso is a forgery. There is no requirement of rationality that requires Smith to purchase the painting that he falsely believes is a forgery. That requirement would depend on the rationality of his belief that it is a forgery.

In the case of choosing hands, assuming that the distribution of stones was truly random, there is no "requirement of rationality" that would help Alph choose the hand with the white stone.

Remember, I am not saying that Smith is incorrect. This is a semantic dispute. The question is: Which use of terms makes the most sense. It isn't even a question of what use of the terms conforms best to standard usage, since standard usage itself may be likely to lead to confusion and error. That, at least to my mind, is: Does Alph want the painting? Yes. Does he believe that he wants the painting? No.

Case 2: Stepping On Toes.

Smith describes the second case as follows:

Suppose that I am standing on someone's foot so causing him pain, and that I know that this is what I am doing. Surely we can imagine its being appropriate for an outsider to say that I have a reason to get off his foot even though I lacked the relevant desire, and, indeed, even if I desired to cause him pain.

As an outsider, I think it is simply false to say that Smith has a reason to get off of his victim's foot. Smith SHOULD have a reason. A properly motivated person WOULD have a reason. However, neither of these truths imply that Smith DOES have a reason. If he has no desire that would be fulfilled by getting off the foot, then he has no reason to get off of the foot.

Smith, in contrast, wants to make the case for saying that he does have a reason to get off of the foot.

For it suffices for the truth of the claim that I have a reason to get off his foot that there exists a requirement-in this case moral-that I do not cause him pain, and that, in the present circumstances, in order to comply with that requirement I have to get off his foot.

That "there exists a requirement" only implies that "there exists a reason". It does not imply "Smith has a reason", in the same way that "there exists a 1969 Plymouth Valiant" implies "there exists a car". It does not imply "Smith has a car."

I would further make the claim, in the case of moral requirements, that if the moral claim was true then people generally have reasons to cause Smith to have a reason to get off of the foot. People generally can provide Smith with a reason by giving him an incentive (payment), or threatening to harm him. I would also argue - though I do not have the space to do so here - that people generally have reasons to cause people universally, including Smith, to have an aversion to causing others pain. This is another sense in which Smith should have a reason to get off of the foot. Yet, none of this implies that Smith has a reason to do so.

Here, I would attribute the habit of saying that people have a reason to do what is moral is based on a mistake that goes as far back as the ancient Greeks that says that a person always has a reason to do what is moral, because what is moral always benefits the individual in some way. One of the ways in which people always have a reason to do what is moral, according to ancient doctrines, is that good people are rewarded in the afterlife and evil people are punished. Where more recent understanding of reasons and motivation suggest that this is not the case, they also suggest that we should give up this old way of speaking.

Case 3: Drinking Petrol

Smith's third case is that of a person - let us call him James - who wants a gin and tonic and, thinking that the glass in front of him contains tonic (though it actually contains petrol), pours some tonic into it. This, according to Smith, seems to create a problem for his thesis since the agent has no reason to drink what is in the glass even though he has the relevant beliefs (that it is a gin and tonic) and desires (to drink a gin and tonic).

I would treat this case the same way as Case 1. James has no desire for what is in the glass. He thinks he does. However, just as with Case 1, there is a difference between what is true and what James believes is true. While James believes he has a reason to drink what is in the glass, his belief is mistaken. He has no reason to drink what is in the glass because it is not that which would fulfill his desires.

Again, I can find some support for this use of the term.

Assume that somebody else, Jimmy, sees what happened. Just as James is about to take a drink, Jimmie puts his and over the top of the glass and tells James, "You don't want to do that."

James statement is perfectly sensible, and true. More to the point, James' most reasonable response to Jimmy is not, "You are obviously mistaken." It is "Why not?" Again, there is a truth of the matter as to whether James wants to drink what is in the glass, James knows this, and James is asking for the evidence that will help him to determine whether what he believes to be the case is actually the case. Whether his belief that he wants to drink what is in the glass is true or false.


I think that there is a lot of confusion on these issues because people have adopted a confusing vocabulary. It is a vocabulary that clouds the distinction between what is the case and what people believe to be the case. This confusion messes up a lot of the discussion. I hope to avoid some of this confusion by using these terms in what I hope to be a clearer and more consistent way. I could be wrong. I could simply be making matters more confusing for some readers. However, I hope this is not the case.

Am I a Humean about Desires?

Am I a Humean about desires?

That is to say, do my claims about desires fall into line with those of the 18th century Scottish philosopher David Hume? Or am I disposed to refuse his claims and reject his account of desire?

I tend to say good things about Hume, and suggest that the account I defend is broadly Humean.

However, there is a puzzle.

I am currently looking at the dispositional theory of desire. This is the view that a desire that P is a disposition to act so as to realize P. In doing so, I am looking at the view of Princeton philosopher Michael Smith. I am rejecting this view. The chief rival to dispositional theories of desire are evaluativist theories. This is the view that to desire that P is to believe or perceive or see P as having value. Either P is, is perceived to be, or is seen as being good, or it is believed/perceived/seen to be something that ought to be the case.

In this classic debate, dispositional theories are seen to be Humean, and evaluativist theories are categorized as anti-Humean.

So, this would suggest that I am anti-Humean.

But . . . let's not be hasty.

The key difference between these two major categories is found in how it answers the question of how or whether reason has anything to say about the passions. Hume famously said that it does not. A Humean theory is categorized as such in virtue of being consistent with this proposition.

Dispositional theories are consistent with this proposition because there is nothing in reason that picks out that which the agent is disposed to bring about. Being disposed to bring it about is all that matters. Thus, dispositional theories are Humean.

In contrast, believing that something is good or ought to be the case, and even perceiving that something is good or ought to be the case, brings reason into the picture to evaluate what to believe or to separate whether something really is as it is perceived to be or whether the perception is an illusion. Thus, belief and perception models of desire are anti-Humean.

What I am offering does not fit conveniently in either of these two camps.

I am offering an evaluativist theory of desire - a desire that P assigns a value to 'P' being made or kept true - but it offers no role for reason. These assignments are simply brute facts. There is nothing for reason to do in evaluating them - in judging them to be correct or incorrect - or to say that they ought to be different.

NOTE: Actually, there is a sense in which we may speak about a rationality of desires. Insofar as desires are a matter of choice, it follows that it may be rational or irrational for an agent to acquire or avoid acquiring certain desires. However, these options are of no relevance here. The means-ends considerations that go into selecting a desire is no different from the means-ends considerations that go into selecting a car or a college major.

So, I am offering an evaluativist theory - a theory of a type typically seen as anti-Humean, but one that gives no role to reason in evaluating desires (except insofar as they may tend to fulfill or thwart other desires). So, the theory is, in fact, Humean in the sense that it gives no role to reason in evaluating desires.

Am I a Humean? Yes, I am, even though my theory fits into the category typically filled with anti-Humeans.

Friday, July 06, 2018

Detailing the Dispositional Theory of Desire

In our last exciting episode, I provided a brief account of the Dispositional Theory of Desire and some preliminary arguments against it.

The dispositional account says (according to the Stanford Encyclopedia of Philosophy entry on "Desire":

For an organism to desire p is for the organism to be disposed to take whatever actions it believes are likely to bring about p.

The arguments against this were:

Explaining and predicting the behavior of agents with multiple desires would be extremely complex. It would require specifying all of the conditions under which any combination of desires would outweigh any other combination of desires. This becomes much simpler if we view motivation like vectors, having a direction and a magnitude. However, associating a magnitude (strength) to a desire takes the theory out of the dispositional account and into the evaluativist account.

Many common dispositions to act - such as those represented by habit and Turret syndrome - are not, in fact, associated with desires. This is mainly because the agent does not attach any importance to the act or its consequences.

However, dispositional theories contain an important truth - desires motivate action. Desires create dispositions to act. However, the relationship between the desire and disposition is not one of identity (a desire is a disposition to act), it is one of cause and effect (desires cause dispositions to act).

Among philosophers, the discussion of the case of Radioman, which Warren Quinn presented in Quinn, Warren 1993: ‘Putting Rationality in its Place’ in his Morality and Action (Cambridge: Cambridge University Press).

Radioman has a disposition to turn on radios. If he is in the vicinity of a radio that is not on, he is disposed to reach out and turn them on. He does not do this because he has some interest in the results of turning on radios. He does not want to listen to music and, in fact, would prefer the silence. Nor does he care about turning on radios as an end in itself. He doesn't care about anything - he just performs the action.

The conclusion is that this is the case in which "Radioman is disposed to turn on radios" is true, and "Radioman has a desire to turn on radios" is false, which disproves the thesis that being disposed to act is the same as having a desire to act.

As a rule, I don't like this type of thought experiments. We invent language as a tool to use in everyday circumstances. We can well imagine novel situations and wonder what to say about them. Our normal language doesn't fit, so we tend to stretch concepts. It is a mistake to think that these contortions and distortions indicate something significant.

Indeed, this figures into Michael Smith's response to Quinn's radioman.

Smith, Michael (2012), "Four Objections to the Standard Story of Action (and Four Replies):", Philosophical Issues, 22, Action Theory,, pp 387-401.

Smith wants to call Radioman's disposition to turn on radios a desire.

[W]e know something both distinctive and familiar about the origins of Radioman’s behaviour. We know that he has an urge to turn on radios, and we know that whether his urge will have any effect at all will depend on two things. First of all, it will depend on whether he has stronger contrary dispositions: imagine telling him that if he turns on a radio you will shoot him in the head, so bringing his urge into conflict with his desire to preserve his life. And second, it will depend on what options he believes will bring about the thing that he has an urge to do: that is, it will depend on whether he believes that there are radios in the vicinity whose operations he can affect by his bodily movements.

These assumptions certainly make it appear that Radioman's behavior counts as desire-driven, particularly given his capacity to decide not to turn on radios if his life is threatened.

However, will he refrain from turning on radios if his life is threatened? Perhaps he will not - but he will if his child's life is threatened. Perhaps he will refrain if he is paid $1000 but not if he is paid $10. What about $199.95?

In Smith's response, he is assuming that Radioman assigns a certain value or importance to turning on radios. There is a price at which he will refrain from doing so, indicating that, below this point, turning on radios has more value than the money and, above this point, the money has more value than turning on radios. I would suggest that this assignment of a value is exactly what Quinn's objection was getting at - this is the missing element in a dispositional account of desire. By assigning a value to turning on radios, Smith doesn't answer Quinn's objection. He surrenders to it.

However, we do not need to invent counter-examples and distort and stretch concepts beyond everyday usage to come up with an objection to the dispositional theory. We can look at a category of actual real-world behavior that qualifies as "dispositions to act" but not as "desire". These are habits.

For 12 years, when I left work at the end of the day, I would leave the building and turn right to go to the bus station. Then the city built a new bus station (it is difficult to build an old bus station) a few blocks to the left. When the new bus station opened, they closed the bus station to the right. Yet, quite a few times, I left the building where I worked and turned right. I then walked down the street for a while, then turned around and went in the other direction.

I had a disposition to turn right as I left the building, but I did not have a desire to turn right. It was not something I valued for its own sake, nor was it something that I valued as a means to an end. It would be a mistake to say that I believed that the bus station was to the right because, if anybody had asked me, I would have reported that the bus station was closed and the new bus station was three blocks away to the left. So, I did not turn right as a result of means-ends rationality. I had a disposition to act that was not, in any way, associated with a desire.

Yet, it was still an intentional action. Turning right was as intentional as turning left would have been. Turning right was as intentional the first day after the old bus station closed as it was the last day the old bus station was still open.

This gives us a real-world example of Radioman.

And, indeed, as I left the building, if somebody had threatened to shoot me in the head if I had turned right upon leaving the building I would have likely put extra effort into making sure that I turned left. I would have turned left anyway. That is what I wanted to do. That was where I was going to find the bus that would take me home.

So, we do not need to invent some strange story of a person who has a disposition to act who does not have a desire - an example that shows that "a desire is a disposition to act" is false. We have real-world examples we can draw from. This gives us reason to deny the proposition that a desire is a disposition to act.

About a Theory of Desire - Dispositional Theories

A desire-based theory of morality requires a theory of value.

Long ago, I decided that I wanted to focus on theories of morality and not worry about theories of desire. On this latter issue, I would simply accept conventional wisdom on the matter. I certainly did not have time to do everything, and consequently it makes sense to benefit from a division of labor. "You folks work on that part over there, and I will work over here."

In my defense, at that time the discussion in the philosophy of psychology was on things being called "folk psychology", "eliminativism", "type-type identity theory", "token-token identity theory." And that certainly would have taken me far from my area of interest. My decision was to accept folk-psychology - a system that grounded intentional action on beliefs, desires, and intentions - until and unless the philosophers came up with something better they could agree upon. At that point, I would see how much of this moral theory (if any) could be salvaged.

Since then, the debate seems to have shifted. There are now theories of desire - theories about those things that are said to exist in the realm of psychology that is one of the basis for intentional action. In looking at these accounts, I discover that I do, indeed, have something to say. While I have been accepting folk psychology while I studied morality, I have come up with an account of desire that seems to be at odds with what the experts in the field are coming up with.

There is a sense in which I find this discouraging. In my mind, it raises doubts about my project in that it seems to say things about desire that are not a part of the received wisdom. It would have been so much better to point to the received wisdom about desire and say, "See, all of this stuff I say about morality fits hand-in-glove with what the experts are saying about desire." Instead, I find the account of desire that I have been working with at odds with what those experts say.

However, this is not, in fact, the type of problem that leads to the conclusion that I should simply give up and admit to error. Rather, it is the type of problem that invites the response, "So, you disagree with what the experts in the field are saying about this matter. Come on, then. Show us what you've got? Why do you think that they are wrong? Let's see your evidence and your reasoning?"

Well, it goes as follows:

Note: I will describe the issue briefly here, and then go into the subject in more detail on each of the key points later.

Currently, there are two major theories of desire, and several sub-theories within each category.

The dominant major theory takes desires to be dispositions to act in a particular way. According to the Stanford Encyclopedia of Philosophy entry on Desire, it can be characterized as:

For an organism to desire p is for the organism to be disposed to take whatever actions it believes are likely to bring about p.

There is a sense in which this is true. I hold that desires are motivational state. That means that when an organism has a desire that p, the agent is motivated to take whatever actions it believes are likely to make or keep true the proposition 'p'.

However, the desire is not the disposition itself. The desire is the cause of the disposition.

In talking about desires, we speak about motivational force. I want to take this "force" analogy seriously.

When a force acts on an object, it "disposes" that object to accelerate in a particular direction. However, this does not mean that it necessarily will accelerate in that direction. The force of gravity is currently (thankfully) not accelerating me to the center of the earth. This is because the chair that I am sitting on, the floor that the chair is sitting on, and a huge column of dirt, rock, and magma provides an equal and opposite force in the opposite direction, thus preventing my fall. But the force is still there. The "disposition" to fall continues to exist, even while all of that stuff below me is holding me up.

However, there is a difference between saying that gravity is the disposition to fall and that gravity is the thing that causes the disposition to fall.

The same can be said about the other forces. Electromagnetic force is not the disposition of electrically charged particles to behave in particular ways (e.g., opposite poles attract). It is that which causes charged particles to act in a particular way. Strong and weak nuclear forces are not the disposition for the nucleus of an atom to stick together in spite of being filled with positively charged particles (protons), it is that which holds the nucleus together. Similarly, desires are not the disposition to realize certain states of affairs. They are those things that cause the agents to be disposed to realize certain states of affairs.

One of the problems with the dispositional theory of desire is that, though it may work fine when we talk about a single desire, it becomes amazingly complex very quickly when we talk about a person motivated by several desires. Take an agent with a desire that A, a desire that B, and a desire that C. How will this person choose between A, B and not C versus A and C but not B? What if the desire that A is the agent's aversion to pain. How do we fit in the fact that the agent would accept a little bit of pain, but would avoid a larger amount of pain?

I may be able to find some support for this in a book by J. Pollock (Pollock, J., 2006. Thinking about Acting: Logical foundations for rational decision making, New York: Oxford University Press.) According to a summary of Pollock's position in the Stanford Encyclopedia of Philosophy:

Pollock (2006) argues that the number of such facts is on the order of a billion billions at the very least, just to encode the same facts about desire and preference that could readily be generated from just three hundred basic facts about desires. From the assumption that the basic psychological facts must be physically realized in the brain, Pollock concludes that it is psychologically realistic to believe in basic desires, not basic pairwise preferences.

Force theory understands a force as having two components - a magnitude and a direction. The dispositional theory of desire recognize the "direction" of a desire (realizing p), but not the "magnitude" component (the strength of the desire). The strength of the desire is to be understood in terms of the importance that realizing p has for the agent, and thus what the agent may be willing to give up to realize p. However, the assignment of a magnitude to a motivational force is the point at which we leave the dispositional theories of value and enter the second major family of theories - the evaluationist theories. Evaluationaists, at least, understand that desires have a magnitude or strength component.

There are other areas where the dispositional theory's failure to account for the value of realizing p to the agent with the desire. Warren Quinn has a counter-example to dispositional theories. He described "radioman", a person with a compulsion to turn on radios in his vicinity - even though he does not want the radios to be on and has no reason to turn them on. This is simply something he is disposed to do. The objection is that he does not have a desire to turn on radios because this result (or the activity) has no value. Desire takes more than a disposition to act, it takes an assignment of value - or importance - to the agent.

I tend to dislike thought experiments as being too vague, and prefer to keep my discussion of such topics grounded in the real world. However, we have real-world analogues to Quinn's Radioman. A habit is a disposition to act in a particular way. However, a habit only obtains its value insofar as it is in service of a desire. We recognize good habits and bad habits precisely in virtue of the fact that they serve or thwart desires. A habit is a disposition to act, but a habit - by itself - is not a desire to realize any state of affairs.

Turette's Syndrome provides another example. People with Turette's Syndrom are disposed to perform certain intentional actions called "tics". These are intentional actions, but the agents who perform those actions do not see the actions as theirs. They attach no value in doing so. They have a disposition to behave, but they do not have a desire. This brings into question the idea that we can relate a desire with a disposition to act.

For these reasons, I would suggest that we reject dispositional theories of desire in favor of a theory that assigns to a desire both a direction (that which the agent is motivated to bring about) and a magnitude (a strength of the desire).

Thursday, July 05, 2018

Now, About That Book . . .

Am I ready to start that book?

The right act is the act that a person with good malleable desires and lacking bad desires would have done in those circumstances.

This is the thesis that I want to defend.

I suspect there is a reader who will take one look at this thesis and immediately toss the project aside as a waste of time. With "right act" on one side of the equation, and "good malleable desire" on the other side, this project clearly begs the question as it smuggles in all of the answers to "the right" in a question-begging and equally problematic account of "the good".

It is a legitimate concern.

This account is going to need an account of "good malleable desires" that does not end up being question-begging. This means that I am going to need such an account of "good", "malleable", and "desire".

Specifically, it seems that I must begin with a theory of desire. From there, I can go on to a theory of "good". I would also through in a theory of "ought" and of "normative reasons" to boot.

As is the way of philosophy, a philosopher is supposed to spoil the ending by giving away the plot in advance.

In this light, I might as well say:

A desire is a propositional attitude - a mental state that that can be expressed in the form 'Agent desires that P', where 'P' is a proposition capable of being true or false, which assigns a value V(D) to a proposition P representing the importance to Agent of P being made or kept true.

NOTE: I have toyed with the idea that both beliefs and desires assign a value to a proposition being true. The value that it assigns to a proposition that is the object of a belief - V(B) - represents the credence of that belief - the likelihood given whatever else the agent knows that the proposition is, in fact true. Whereas, as I said, the value it assigns to a proposition that is the object of a desires - V(D) - represents the importance of that belief being made or kept true to the agent. However, I have not given the issue of belief in this regard a lot of thought. I toss that out here simply as something to think about.

From here, we can go on and develop theories of "good", "ought" and "normative reasons".

X is good if and only if X is such as to fulfill the desires in question.

Before going too far, I want to make sure to point out that this is not an account of the meaning of the term "good". If we were to look at the meaning of this term, it would say something like, "X is good if and only if and to the extent that there exists a reason to realize or preserve X". There could be all sorts of reasons - divine commands, intrinsic values, categorical imperatives. However, there is only one type of reason for intentional action that exists, and that is to fulfill a desire. All of the other types of reasons are consistent with the meaning of the term "good". They just don't exist.

The critic then asks, "Well, then, what is to 'fulfill' a desire and which are the 'desires in question'?"

A "desire that P" is fulfilled in any state of affairs S where P is made or kept true.

As to the "desires in question," this reflects the fact that the term 'good' is ambiguous. It has many meanings, each meaning relating the object of evaluation to different desires in question. We pick out the desires in question by looking at the context within which the term 'good' was being used.

Some of the more common options are:

(1) A given desire. Example: "If you do not want to be identified on the video when you rob the convenience store, then it would be a good idea for you to wear a mask."

(2) The agent's current desires. Example: "While you are at the store, it would be a good opportunity to pick up a carton of cigarettes."

(3) The agent's current and future desires. Example: "Smoking is no good for you. You should quit."

(4) The desires of those involved in the decision. Example: "What toppings do we want on our pizza?" and "Where would be a good place for us to go on vacation?"

(5) The desires typically fulfilled by the type of object being evaluated. Examples: A good knife. A good game.

This theory of "good" relates to a theory of "ought" as well as to a theory of "normative reason".

Agent ought to do A means that there is a reason for Agent to do A.

Here, I want to draw a distinction between this account of normative reasons and the account that the Stanford Encyclopedia of Philosophy identifies as the "Humean Theory of Reasons (revised)."

SEP offers the following hypothesis:

If there is a reason for someone to do something, then she must have some desire that would be served by her doing it, which is the source of her reason. Stanford Encyclopedia of Philosophy, "Reasons for Action: Internal vs. External", https://plato.stanford.edu/entries/reasons-internal-external/, accessed July 5, 2018.

This is a mistake. An accurate account of reasons says:

If there is a reason for someone to do something, then someone must have some desire that would be served by her doing it, which is the source of her reason. Whereas, if she has a reason to do something, then she must have some desire that would be served by doing it.

This account properly separates the account of "there is a reason" from "she has a reason" which the original version does not do.

With a theory of "good" and "reasons for action" under our belt, we can look at the term "ought".

"Agent ought to do X" means that "Doing X will fulfill the desires in question."

Look to the above discussion about "good" for an understanding of the phrase "desires in question". The term "ought" is ambiguous. It has countless meanings, each meaning relating to a different set of "desires in question". All of these different meanings are perfectly legitimate. Yet, it is extremely easy to equivocate, to begin an argument using one understanding of 'ought' that refers to one set of 'desires in question', and then shifting the 'desires in question' (and, thus, the sense of the word 'ought') in mid argument.

In the case of morality, I would argue that the desires in question are all fixed desires plus those malleable desires that people generally have many and strong reasons to promote universally. That is a rather long and complex formula, so I will work it out in detail later.

For now, we have an idea of a desire, and we have an idea of "good", so a "good malleable desire" is a malleable desire that tends to fulfill the (other) desires in question. In the case of moral value, it is a desire that tends to fulfill other desires generally, regardless of whose they are.

I need to say something about the sense in which a desire is "malleable". In the case of morality, we are looking at desires that people can mold in others using the tools of reward and punishment, including praise and condemnation. We are familiar with the practice of using reward as an incentive and punishment as a deterrence. However, incentives and deterrence work on existing desires. A reward is a promise to fulfill, and a deterrence is a threat to thwart, an existing desire. However, what I am interested in here is the capacity of rewards and punishment, including praise and condemnation, to actually bring about changes in desires. I am interested in these tools having the power to create an aversion to lying or to taking the property of others without consent, a desire to help those in desperate need, or a desire to repay a debt.


A good malleable desire is a desire that can be modified using the social tools of reward and punishment, including praise and condemnation, that people generally have reason to promote universally because of the desire's tendency to fulfill other desires.

This brings us back to the hypothesis that a right act is an act that tends to fulfill those malleable desires that people generally have many and strong reasons to promote universally using the social tools of reward and punishment, including praise and condemnation.

That's the big picture. However, the big picture is made up of a lot of little details. The first detail that I would like to look at is: What is a desire?

On Desire 2018. Part 62: Intuiting Desires

I am now on the last chapter of The Nature of Desire - the book that I have been commenting on over the past two months.

This last chapter is; Ashwell, Lauren, (2017), “Introspection and the Nature of Desire,” In Deonna J. & Lauria F. (eds). The Nature of Desire. Oxford University Press.

I will not be spending much time here for two reasons.

Reason 1: It is a short chapter.

Reason 2: I think that our ability to introspect our desires is week.

Let's take a look at what Ashwell is trying to claim, first.

Ashwell is taking on the claim that desires can be understood as beliefs about what is good or as perceptions of something being good.

She brings up a number of cases in which we judge something to be good while, at the same time, we intuitively judge that we do not desire to do it. For example, I can judge that I should go exercise while I sit at home watching television. Ashwell's contribution to this discussion is to put stress on our intuitions about what we desire. We have the judgment that exercise is good for us, yet our intuitions tell us we just don't want to do it. On this basis, it is difficult to argue that our desire corresponds with what we judge to be good.

Similarly, we often find ourselves wanting to do things that we do not judge to be good. Ashwell's example here is of a person who wants to stay in bed, under the covers, instead of getting up and going to work. Quite simply, the person in such a state will say that she desires to stay in bed. However, she does not judge it to be the case that she ought to stay in bed or that it is good to stay in bed. Instead, she know that she ought to get up and go to work - that this is the best thing to do.

For the most part, Ashwell is simply looking at the ways in which we are likely to use the word "desire" and that, in our normal use of the term, we do not use it to refer to what we judge to be good or what we believe we ought to do. We use the term to refer to that which we are motivated to do, regardless of whether we judge it to be good. This is a serious problem for the thesis that to desire that P is to believe that P is good or to judge it to be something we ought to bring about.

At the end, Ashwell herself brings up some problems with intuition about desires.

My discussion here is complicated somewhat by the fact that I do not think a complete story about self-knowledge should involve the assumption that desire introspection is as highly privileged as is usually claimed; I am also not unsympathetic to views of self-knowledge that have a place for inference from evidence in introspection.

Indeed, in some of my previous discussions, I have questioned the idea that we know our desired directly.

First, there was Timothy Schroeder's claim that the parts of the brain that seem to be most closely associated with desires have little connection to the parts of the brain having to do with consciousness or memory. I discussed Schroeder's account of the biology of intentional action in post 48 of this series. This aspect of desire sits in contrast with the parts of the brain having to do with belief and perception - parts with many and strong connections to parts having to do with consciousness and memory. This suggests that what we know even about our own desires we know indirectly - in terms of their effects.

Having said this, we have a great deal of experience dealing with the effects of our desires, and sometimes those effects are rather urgent. Consequently, we can reach conclusions about our own desires with so little effort that it would not be difficult to think that this is direct intuitive knowledge. After all, one of the primary effects of desires is motivation (and one of the primary effects of a lack of desire is a lack of motivation). These motivations impose themselves on us, often, with a great deal of force. It is difficult to deny that the relevant desire or aversion exists.

Second, in my discussion of G.E. Schueler's account of Robert Audi's theory of practical reasons discussed another way in which our beliefs about our desires might deviate from our actual desires. The major premises in Audi's account of practical reasoning is a premise about what the agent wants. However, it is actually a statement about what the agent believes that she wants, a premise that may be mistaken. When it is mistaken, the agent may reach a "should" conclusion. However, if that conclusion is not linked up to an actual desire, then there will be no actual motivation to do what one judges one should do.

This provides agents with one way to realize that their first premise - their beliefs about what they desire - are mistaken. They can begin to question this premise when they realize that it does not provide the motivation they would expect if the premise was true.

As I said, we have a great deal of experience learning about our own desires, even if they are not directly connected to the parts of the brain having to do with consciousness and memory. They are connected to motivation (they select the potential actions that become actual actions), and our beliefs about them are certainly connected to consciousness and memory.

Before leaving, there are other ways in which one can reach a "should" conclusion in an Audi-style practical syllogism without experiencing motivation. This happens whenever the major premise does not refer to an actual desire. This, in turn, happens when the major premise refers, in part or in whole, to a desire that does not exist yet, or somebody else's desire. Connections to those desires will not come with motivational force - or will do so only through the mediation of a "desire that future desires be fulfilled" or a "concern for the interests of other persons". These provide additional ways in which a person can believe that something is good or ought to be done yet fail to find the motivation to do it.

Wednesday, July 04, 2018

Desires 2008: Summary 02: S. Doring and B. Eker

Doring, Sabine A. and Eker, Bahadir, (2017). “Desires Without Guises: Why We Need Not Value What We Want," In Deonna J. & Lauria F. (eds). The Nature of Desire. Oxford University Press.


Doring and Eker make three main claims:

(D1) Necessarily, for any agent a, any proposition p, any time t, and any act type φ, if, at t, a desires that p, then a is disposed at t to φ in circumstances where a takes her φ-ing to be conducive to p’s being the case.


(D1*) Necessarily, for any agent a, any proposition p, any time t, and any act type φ, if, at t, a desires that p, then, if, at t, a took her φ-ing to be conducive to p’s being the case, a would φ, ceteris paribus.


(D2) Necessarily, for any agent a, any proposition p, any time t, if, at t, a desires that p, then there is at least one act type φ such that, at t, a does not think her φ-ing not to be conducive to p’s being the case.

(D3) Necessarily, for any agent a, any proposition p, any time t, if, at t, a desires that p, then, at t, a does not think p already to be the case.


Actually, the first thing that I want to say is not a question. It is a comment.

D3 seems false.

Assume that I am cuddling with my wife. According to D3, I can only want to cuddle with my wife if I do not think that I am already cuddling with her. I cannot want to cuddle with her and cuddle with her at the same time.

To me, this makes no sense. Of course I can want to cuddle with my wife while I am cuddling with her.
The argument against the possibility of wanting to cuddle with my wife while I am cuddling with her says that I cannot “take my φ-ing to be conducive to p’s being the case” if p is already the case. So, if I had a desire to cuddle with my wife while I was cuddling with my wife, this would be a counter-example to D1. It would be a case in which I had a desire (to cuddle with my wife) when it would not make sense for me to “take my φ-ing to be conducive to p’s being the case.”

As I see it, this set of counter-examples simply defeat D1.

To handle this type of move, the response is to say that I would not have a desire to cuddle with my wife while cuddling with her. Instead, I would have a different desire – a desire to continue to cuddle with her. Even as I cuddle with her, I can “take my φ-ing to be conducive to my continuing to cuddle with her being the case.”

Yet, there is something odd with this answer. Rather than postulating a desire that I am cuddling with my wife – a desire that motivates me to seek an opportunity to cuddle with her while it is not the case, and to keep cuddling with her once it is the case – this thesis says that there are two separate and distinct desires. There is a desire to cuddle with my wife that exists until I am cuddling with my wife. Then, at that instant, the desire to cuddle with my wife ceases to exist, and a different desire that I continue to cuddle with my wife suddenly pops into existence.

It simply makes more sense to say that the desire that I have to cuddle with my wife when I get home from work, and the desire to continue cuddling with my wife as we sit on the couch watching television, is the same desire.

I can draw some support for this view of desire from the biology of desire.

Let us look at hunger. When the stomach is empty, it secretes a hormone called ghrelin. This is associated with the desire to eat. When the agent starts to eat and the stomach begins to fill up, secretions of ghrelin decrease, and so does the desire to (continue) eating.

There is no desire to eat that disappears the instant one puts the first fork full of food in one’s mouth, and a separate desire to continue eating that emerges instantly at the same time one starts eating. There is a single desire – a desire “that I am eating” that motivates the agent to find food, and then to eat the food once it is found, until the agent is full.

Besides, an agent cannot have a desire that p continues unless, at the same time, one thinks that p is already the case. The idea that “p should continue” and “p is not currently the case” are quite incompatible ideas.

This way of looking at the issue seems more friendly to the evaluativist view than to the dispositionist view. It can be explained by the fact that the agent sees a certain amount of value in cuddling with his wife or in eating, a value that motivates him to realize such a state until it is made to exist, and to preserve that state once it exists – because it is important (to the agent).

My second question is: what about the strength of the desire?

This question is attached to a related issue: doesn’t the dispositional account get extremely complicated very quickly when there is more than one desire?

If we are talking about an agent with just one desire, a desire that p, then we may be able to make sense of a dispositional account whereby the agent is disposed to make it the case that p, if possible.

However, once we add a second desire, we have to add a number of conditions that describe how these desires are weighed against each other – a set of counterfactuals that explain the conditions under which the agent is disposed to act in accordance to one desire as opposed to another.

Now, add a third desire. What are the conditions when the agent faces a choice between fulfilling D1, but thwarting both D2 and D3, or must choose between fulfilling Desire 1 and Desire 2 or Desire 3 alone? Add a fourth, fifth, and sixth desire, and things get very complicated very fast.

One way to tame this complexity is to give each desire a weight or strength – to attach a value to each object of desire specifying the importance it has for the agent. This gives us a way of weighing different concerns, to rank some as more important and others as less important.

There is nothing in D1 that handles the concept of a strength of a desire or the importance of its object. That requires attributing a value to the object, and that suggests an evaluativist theory rather than a dispositional theory.

On Desire 2018. Part 61: Desiring, Intending, and Doing

I have been looking at G.E. Schueler's account of how desires figure in action. (Schueler, G.E., (2017), “Deliberation and Desire” In Deonna J. & Lauria F. (eds). The Nature of Desire. Oxford University Press. (Referencing Broome, John, 2005, "Does Rationality Give Us Reasons?" Philosophical Issues, 15, 321-337.)

So far, I have presented Schueler's objections to Robert Audi's account of practical reasoning.

Just as a brief reminder: Audi suggested that practical reasoning works as follows:

(1) Major Premise—the motivational premise: I want phi
(2) Minor Premise—the cognitive (instrumental) premise: My A-ing would contribute to realizing phi
(3) Conclusion—the practical judgment: I should A.

Schueler objected that the Major Premise is a belief about a desire, not a desire. It could be false. If it is false, then, when the agent acts from deliberation, the agent is acting from a belief about a desire he does not have. At the same time, the Major Premise cannot be false when the agent acts from deliberation because, even if the Major Premise was false, it would still be the reason or purpose for the action, and thus what the agent wanted.

I followed this with a posting with my own take on the problem. On that take, if the Major Premise is false, then the agent comes to a practical judgment that lacks motivation. This is how it can come about that an agent judges, "I should A," but doesn't A. Like, "I should make a dentist appointment" or "I should be working on my homework."

However, I owe it to Schueler to present his answer to the puzzle.

Schueler divides practical reasoning into two parts.

The first part of practical reasoning involves the consideration of various reasons to do something and deciding what to do. Just a few moments ago, I was at the gym working on an elliptical while I read/listened to Schueler's article on deliberation and desire. An annoying thing often happens when I am at the gym. In considering something that I am reading, I think of something that I would want to say in response to that content. I enter a state of mind where I am writing in my head, churning through the phrases and ideas. When I am in this state, any further reading is nearly impossible. I will get a few sentences further, then my mind is churning on that writing again. The best thing to do is go home, get written what I want to write, and then I can continue with the reading.

Have you been wondering why I am now on post 61 of this series?

On the other hand, I need to get my exercise in. So, I had a choice to make. Stay on the elliptical and exercise even though I was doing no useful studying, or go home and write so that I can return to the elliptical later with a clear head ready to work on the next section.

This is the first part of practical reasoning on Schueler's account - the comparing of different reasons, weighing pros and cons, and coming up with an intention. That intention was to finish up 1 hour of exercise then go home and write up my commentary on Schueler's account of practical deliberation. This is the part of the deliberation where, according to Schueler, the agent considers likes, dislikes, applies judgments about what should be done, and considers the effects of the actions on other goals the agent may have.

Once the agent form an intention, then one acts on the intention. Acting on the intention, according to Schueler looks something like Audi's practical reasoning. However, its form is actually:

(1') Major Premise—the motivational premise: I intend to phi
(2') Minor Premise—the cognitive (instrumental) premise: My A-ing would contribute to realizing phi
(3') Conclusion—the practical judgment: I will A.

Note the two significant differences between this structure and Audi's structure.

The first difference is that the Major Premise is an intention. Intentions themselves are neither true nor false. The belief that one has an intention can be true or false, but the intention itself cannot be.

The second difference is that the "conclusion" in this case is an actual action. It, too, is not a belief. It is the actual muscle movements that are involved in performing that action.

I find it interesting to take Schueler's account and place it along Timothy Schroeder's account of the biology of intentional action which I reviewed in Part 48 of this series.

Schroeder described a system in which beliefs and perceptions queued up a bunch of actions. The motivational system then selected the action to pass through into actual action. Those signals went through the motor cortex where it selected the specific muscle movements that would be required to perform the action and then sent the signals down the spinal cord to perform those actions. Schueler's "phase 1" of intentional action involved the decision in the affective centers of the brain selecting the proposed action that will become actual action. The second part of Schueler's account then involves the steps whereby the selection becomes muscle movement.

This may almost sound as if I am going with Schueler's account of action. And I may be.

However, what I notice about this account is that it says almost nothing about desire. The role of desire is found in that vaguely described process of comparing values - something that actually takes up just a few sentences in Schueler's article.

Schueler takes the second "puzzle" he raises against Audi and concludes that if the agent is performs the action identified in Premise 3', and does so on the basis of the intention identified in Premise 1', then it must be the case that this intention is what the agent wants (in the goal-sense of want). This, according to Schueler, would be want the agent desires.

For all practical purposes, Schueler identifies desires with intentions, and has given us a theory of intentions.

In contrast, as I have been understanding the issue, a theory of desires is a theory of those forces that select the queued action that will become actual action or intention.

Schueler himself identifies one of the primary differences between desires and intentions. Desires can conflict. A person can desire to finish exercising and, at the same time, desire to get his ideas down in writing so he can clear his head for further study. Intentions cannot conflict. A person cannot intend to continue exercising and also intend to go home and get his ideas down in writing.

The study of desire is a study of those things that can conflict.

Schueler has not given us an account of desire. Though, this does not imply that his account of intention has no merit. It simply needs to be recognized for what it is.

Tuesday, July 03, 2018

On Desire 2018. Part 60: Crazy and Evil Desires

How could the mere fact that I have some goal, just by itself, make it the case that I should do what I can to achieve that goal? What if the goal is evil or just wacky?

(Schueler, G.E., (2017), “Deliberation and Desire” In Deonna J. & Lauria F. (eds). The Nature of Desire. Oxford University Press. (Referencing Broome, John, 2005, "Does Rationality Give Us Reasons?" Philosophical Issues, 15, 321-337.)

If you have some goal, it follows that you should try to achieve it. Or, in other words, every desire that P provides the agent with a reason to realize any state of affairs in which 'P' is true. Identifying the 'desire that P' as evil or "just wacky" requires determining its relationship to other desires. No desire is evil or wacky just on its own - by itself.

When a desire comes into conflict with others, there is nothing that identifies by default which desire to identify as 'evil' or 'wacky' - it will be the one that there is the most and strongest reason to get rid of. In other words. In the case of "evil" the other goals with which it comes into conflict are not other goals of the agent. "Evil" identifies conflicts between a goal and the goals (desires) of other agents - desires that give others reason to condemn the interest identified s "evil". "Wacky," on the other hand, may refer to desires or goals that conflict with other goals the agent has. (Though, actually, I would expect "wacky" to actually tend to refer to the beliefs that determine the relationship of doing A to ends, rather than the ends themselves).

Before going any further, I should give this discussion some context. As I mentioned in our last exciting episode, we are working with Audi's account of practical reasoning:

(1) Major Premise—the motivational premise: I want phi
(2) Minor Premise—the cognitive (instrumental) premise: My A-ing would contribute to realizing phi
(3) Conclusion—the practical judgment: I should A.

I have accepted the possibility that Premise 1 can be false. I then argued that the "should" conclusion that one reaches through this type of reasoning comes with no motivational force unless either Premise (1) is true or the agent has some other desire that may be motivating her to do A.

When Schueler looks at the claim that a goal provides somebody with a reason to act, Schueler is including under the term "goal" a false Premise (1) in an Audian practical syllogism. This is because Schueler (falsely) believes that we have the capacity to act on the conclusion reached through such a syllogism even when premise (1) is false and the agent has no other reason to A. This means that Premise (1) can be a goal that motivates action even when it is not associated with a desire.

On this matter, I would say that if premise (1) is false, then it provides no reason for the agent to perform the action and, indeed, the agent will have no motivation to perform the action except under the terms and conditions I have already described.

In contrast, when I say that a goal, just by itself, can make it the case that the agent has a reason to do A, I am not talking about a possibly false premise (1). I am talking about a desire - the actual motivating force. Once one has an actual goal - a desire - then . . . yes, it is true, if one has a goal then one has a reason to act, regardless of how evil or wacky the goal may be.

Monday, July 02, 2018

On Desire 2018. Part 59: Acting Deliberately

G. E. Schueler presents us with a bit of a puzzle.

Schueler, G.E., (2017), “Deliberation and Desire” In Deonna J. & Lauria F. (eds). The Nature of Desire. Oxford University Press.

He begins with Robert Audi’s thesis about deliberative reasoning:

(1) Major Premise—the motivational premise: I want phi
(2) Minor Premise—the cognitive (instrumental) premise: My A-ing would contribute to realizing phi
(3) Conclusion—the practical judgment: I should A.

Using this form of reasoning, and concluding that she should A, our hypothetical reasoner then commenced to start A-ing.

Here’s the puzzle.

A-ing is an action. Yet, neither of the premises in this argument represents a desire.

“Sure it does,” the stout defender asserts. “The major premise, ‘I want phi’, is a desire.”

In the context, it is a belief about a desire, it is no more a desire than the statement, “I wear glasses” is an eye condition. It is a statement about an eye condition.

Furthermore, this belief could be mistaken. In the same way the I can believe that I have a scar on my right hand even though I do not, I can believe that I want phi even though I do not. Yet, in spite of my mistake, if I should conclude that I should A, then I commence A-ing, even though my belief that I want A is false.

So, here, we have an intentional action without a desire. Beliefs alone motivate action, even false beliefs.

Furthermore, if you reason to the conclusion "I should A," and we look for your reason for doing A, then we will find that reason in Premise 1. We will find that reason in Premise 1 regardless of whether or not Premise 1 is true.

As Schueler puts it:

If someone acts on the basis of a conscious belief that the goal of her action was phi, then surely the goal of her action was phi. So, contrary to what the first puzzle indicated, when we consider this same piece of reasoning as practical deliberation on which the agent acts, it is hard to see how the motivational premise could possibly be false.

So, what is the answer to this puzzle?

For a while now, I have had a particular practice I follow when I need to make a decision and can't make up my mind. I roll some dice. I then look at the selection that chance determined and I ask myself if I am happy with the result or disappointed. If I am happy with the result, I do what the dice dictate. If not, I ignore the ice and pursue the second option instead.

The relevance of that story to this discussion is that the major premise does not motivate any action. If one uses Audi's form of reasoning described above, and the first premise is false, the agent will come to the conclusion, "I should A." However, he will lack motivation. Doing A becomes one of those things he thinks he should do, but he never gets himself motivated to do it. It languishes undone. This happens, in at least some cases, because the motivational premise is false. Because it is false, it fails to motivate.

When we deliberate, we sometimes deliberate to conclusions we discover we are not at all inclined to do. When that happens, we go over the premises to discover where our argument went wrong. What we sometimes discover is that our lack of motivation comes from our false motivational premise. We think we want phi, but we really don't. Because we really don't, our conclusion lacks motivational force. Motivational force only comes from what we really want.

This suggestion also answers the second puzzle. When the agent comes to the conclusion, "I should A", and he does A, and we look for the reason, we probably will find the reason in a true Premise 1.

This is not guaranteed, by the way. The agent may be reasoning from a false Premise 1, but has a desire that recommends the same action. A person may decide, for example, that a concern for his health suggests that he should lose weight, and thus he decides to lose weight. However, in fact, his premise 1 is false - he is not particularly concerned about his health. However, he actually does have a desire to attract the attention of a co-ed in his philosophy class, and that motivates him to lose weight. So, the fact that a false Premise 1 cannot motivate the agent to do A, this does not imply that the agent will not do A. He may do so anyway, and claim that Premise 1 is his reason, even when his reason is an unconscious motive.

This may seem to suggest that we can never come to a conclusion about what we should do that is mistaken - that fails to serve our desires. If we are motivated to act, then we must have a desire that is being served by the action, because without the desire there can be no motivation.

Of course, one of the ways in which we can come to a false conclusion of the form, "I should A" is to get the second premise wrong. We can be right about our desires, but wrong about whether A-ing would result in realizing psi. Thus, we commence A-ing, motivated by our desire that psi, only to discover that A-ing failed to realize psi, much to our disappointment.

Another form of failure comes from the fact that we use the terms "want" and "desire" not only for motivational ends, but for motivational means. The first premise, "I want psi," may be because of a belief that bringing about psi will bring about some further state that one desires. In this type of case, the second premise can be true (A-ing will bring about psi). Yet, doing A will fail to get you want your want because your believe that psi will produce the further desirable end is false.

Yet a third form of failure rests on the fact that "should" itself is an ambiguous term. "Should" relates doing A to psi. However, we have senses of the term "should" that relate doing A to desires that are not our own. I cannot go into the argument in detail here, but the moral sense of 'should' relates our actions to the desires of others. When we use 'should' in this sense, we may discover a 'should' that we are not motivated to do.

Schueler dismisses this possibility. For that reason, I want to look at it in more detail in the next post.