In episode 14 of Morality in the Real World, Alonzo Fyfe and I discuss the habit of arguing about definitions, and why we don’t do it.
Every five episodes we answer audience questions, so please do post your questions and objections below. Make sure your questions address the topics of this episode only. If we plan to address the subject of your question in later episode, we will not answer it in the next Q & A episode. You can also leave your question in audio and we will play it back during the Q & A episode and respond to it: call 413-723-0175 and press 1 to leave a voicemail.
Transcript of episode 14
ALONZO: Luke! You’re still alive! Where have you been?
LUKE: Ummm… mostly reading neuroscience. I want to be able to defend what I say in this podcast, and we ran into something where we didn’t have the evidence we needed. So, I hit the books.
ALONZO: That sounds a bit suspicious – kinda like trying to find evidence to support a belief you already have?
LUKE: Well, if that’s what I was doing, I did a poor job. I actually didn’t find much evidence for what we were planning to cover in today’s episode, so later I’m going to end up claiming something else instead.
ALONZO: Does this mean that are you giving up on desirism?
LUKE: No. But I am going to propose some adjustments to what we said earlier based on some data from the fields of neuroscience and artificial intelligence.
Also, some things in the field of meta-ethics really clicked for me as I did that research, and I’m really excited to get it written up to see what other people think. But for now, we should get back to talking about different theories of desire.
ALONZO: Right. When we left off, we had just finished talking about action-based theories of desire.
Action-based theories of desire are those theories that say that to desire something means to be disposed to act to make it the case that what one desires is realized.
In this episode, we were supposed to be talking about pleasure-based theories of desire. Pleasure-based theories say that to desire something is to experience pleasure when it seems to be the case that it has been realized.
LUKE: Right. We are going to talk about pleasure-based theories of desire. But I need to distinguish two types of arguments. Well, actually, we already started distinguishing between these two types of arguments, but that distinction is going to become more important now.
ALONZO: What distinction are you talking about?
LUKE: The distinction between disputes over facts of the world versus disputes about the words we use to talk about facts of the world.
Part of the struggle in writing about pleasure-based theories of desire has been a blurring of this distinction. I find that a lot of people writing in this field agree on the facts of the matter. They just can’t agree on the words to use in describing those facts. A substantial portion of the debate is a debate over which words to use, not over which facts are true.
To me, these fights over words are a waste of time. Pick a definition and move on. I’m interested in the facts of the world, not the words.
ALONZO: That’s what you said earlier. On the idea that a lot of the dispute in morality seems to be about determining the best super-dictionary definition of a word, you said that this was not your project. Your project was to look at the world and see what the facts were.
LUKE: Right. Following Eliezer Yudkowsky, I’ve come to speak of this as replacing the symbol with the substance. If there’s a dispute over which symbol to use (which word to use), then perhaps the best way to get past that roadblock and move on is to drop the symbol and replace it with the actual facts you are intending that symbol to stand for. So if the word ‘planet’ is ambiguous, we can skip the word ‘planet’ and instead talk about ‘objects large enough to become roughly spherical because of their own gravity,’ or something.
When it comes to symbols, we should be less concerned with whether our use of them conforms to some super-dictionary definition and more concerned with whether we can use the term efficiently to communicate ideas about the world. That’s the whole point of using words in the first place, right? Efficient communication.
In fact . . .
ALONZO: . . . In fact . . . What?
LUKE: Well, Alonzo, you invented these terms, “thwarting” and “fulfilling” to refer to relationships between desires and states of affairs. A desire that P is fulfilled in any state of affairs in which P is true, and it is thwarted in any state in which P is false.
ALONZO: Well, yeah. I wanted to make it clear that desire fulfillment is not about achieving some psychological sensation like pleasure or happiness. It is about making or keeping a proposition true – about realizing states of affairs where the things one wants actually exist in the real world.
LUKE: Yep. And every time you present your idea to a new audience you have to explain how you are using those terms, because it’s not obvious. And a lot of people who encounter your writing don’t encounter the explanation of what you mean by those terms, so it confuses them.
LUKE: I have a solution to that problem. I found a couple of people who picked what I think might be better words for making this distinction. Instead of talking about ‘desire fulfillment’ versus ‘desire satisfaction’, they talk about ‘objective desire satisfaction’ versus ‘subjective desire satisfaction’.
What you call desire fulfillment – a state where the propositions that are the objects of a desire are actually true – they call objective desire satisfaction. And what you call satisfaction – the sensation of pleasure or happiness one often gets when one thinks – rightly or wrongly – that what one wants has been realized – they call subjective desire satisfaction. In the first case, we are talking about situations in which what a person wants has been realized. In the second case, we are talking about whether the agent merely believes it has been realized.
ALONZO: So, in my favorite example, a parent’s desire that their child is healthy and happy.
LUKE: Yeah. That desire is objectively satisfied if the child is healthy and happy. The desire is subjectively satisfied if the parent believes that their child is healthy and happy, even if the child lays cut and bleeding in a dark alley.
I think, with these terms, people can get an idea of what you are trying to say without wading into an explanation of how you are using your terms, because people have some idea of what ‘objective’ and ‘subjective’ mean, even before you explain them. That makes these terms better than the ones you originally suggested. They are more efficient for communication, I think.
ALONZO: So, you want me to give up talking about fulfilling or satisfying desires, and talk about objectively satisfying or subjectively satisfying desires instead.
LUKE: Yeah, I think so. And I can point out cases in the professional literature where people are already doing that. And your old distinction was kind of confusing because people in the professional literature have often been using the word ‘satisfy’ to mean objective satisfaction, whereas you were using the word ‘satisfy’ to mean subjective satisfaction.
ALONZO: And the reason for this change isn’t because you discovered that this is more consistent with the super-dictionary definition of these terms – a careful exercise in conceptual analysis. It is because you think that this would be more efficient.
LUKE: Exactly. That’s the point of using our word-tools in the first place. Efficient communication.
ALONZO: Okay. I’m convinced.
Henceforth I shall use the term ‘objective desire satisfaction’ when talking about states in which an agent desires that P and P is true. And I will use the term ‘subjective desire satisfaction’ when talking about states in which an agent desires that P, and believes that P is true.
LUKE: Right, though maybe try not to use so much algebra. Our podcast episodes are hard enough to follow as it is!
ALONZO: But our listeners have been clamoring for a detailed, accurate presentation of desirism. How else can we give it to them?
LUKE: Yeah, it’s true. That’s the reason we’ve gotten so technical and precise and scientific on this podcast, but we don’t want to lose all our listeners, so we’ve got to strike a balance.
Anyway, I want you to notice something. We just had a dispute where we resolved a question about the use of words by looking at the issue of efficient communication. A different way to debate the use of terms – the one that bogs down a large chunk of analytic philosophy – is this issue of looking for the best super-dictionary definition of a term; what philosophers call ‘conceptual analysis.’
Here, let me give you an example of what I am talking about.
Ummmm… If a tree falls in the forest and nobody hears it, does it make a sound?
ALONZO: Well, that’s kind of a dumb question.
What do you mean by ‘sound’?
Are you asking whether the tree falling sends off shock waves that a person, if present, would have perceived as a noise? The answer is “yes”.
Or are you asking whether it actually generates sensations in the brain of somebody present, in which case the answer is ‘no’, because there is nobody present?
LUKE: Well, but what’s the right answer? Yes or no?
ALONZO: Well, what’s the right question? Which definition of the word ‘sound’ did you intend? You tell me which question you are asking, and I will tell you what the answer is.
LUKE: Can’t you tell what the right question is just by listening to my question?
ALONZO: Actually, No. I can’t.
Now, imagine two people who think that we should be able to tell what the “right question” is – which definition for the word ‘sound’ is correct or best. So they go looking for all sorts of evidence about how the word “sound” is used and they get into this long detailed fight over which is the right question – which is the right definition of ‘sound.’ They propose outlandish hypothetical scenarios to test what their intuitions tell them about the one correct meaning of the word ‘sound’. One person suggests a ‘correct’ meaning of the word ‘sound’, and then another person proposes intuitive counterexamples to that definition, and someone else proposes another definition of the word ‘sound,’ but then still more people propose intuitive counterexamples to that definition, too!
And they keep debating: What question are people actually asking when they ask about the possible sound of a tree falling in the forest when nobody is present?
ALONZO: First, maybe the question is just vague. Maybe the language of the common person on the street just isn’t very clear on this subject. And, second, I still don’t see why my original answer isn’t good enough. Just tell people that the question is ambiguous and have them ask a more precise question. Ask them which definition of ‘sound’ they have in mind.
LUKE: Exactly. And I think that’s what we need to do with respect to these theories of desire. The question, “What is a desire?” is ambiguous – just like the question, “If a tree falls in the forest and there is nobody around to hear it, does it make a sound?” is ambiguous. We have to disambiguate the question before we can give an answer. And it’s not very productive to argue about what the ‘correct’ definition of ‘desire’ or ‘sound’ is – we can just propose a definition and move on to talk about the actual facts. Or we can skip the definition issue altogether – avoid using the words ‘desire’ and ‘sound’ – and just argue about what certain motivational structures in the brain are doing, or what certain shock waves in the air are doing.
LUKE: Now, back to the pleasure-based theory of desire. Tim Schroeder in the Stanford Encyclopedia of Philosophy describes pleasure-based theories of desire like this:
For an organism to desire p is for the organism to be disposed to take pleasure in it seeming that p and displeasure in it seeming that not-p.
ALONZO: And you were complaining about my use of algebra.
I would think that this is just one version of the theory. There are probably people who would substitute happiness for pleasure, for example. Something like:
For an organism to desire p is for the organism to be disposed to find happiness in it seeming that p and unhappiness in it seeming that not-p.
LUKE: Sure, but the issue I am describing will apply to all of those theories.
ALONZO: But a lot of people do think this way. Back in the first season when we were talking about Alph having a desire to gather stones, a commenter named Keith wrote:
I’m a little uncomfortable with the assertion that some desires, like Alph’s desire to gather stones, are simply ends in themselves, with no real justification. In reality, it is more likely that if someone has a desire, it is because fulfilling that desire feels good. So, if Alph has a desire to gather stones, it is probably because gathering stones is pleasurable or fulfilling to him in some way.
The thing is, psychological hedonism was widely held in the 1800s, but it was discredited over 100 years ago.
LUKE: Well, wait a minute. There is a difference between a pleasure-based theory of desire and psychological hedonism. Psychological hedonism is a theory about the world. It says that the only thing we pursue is our own pleasure and freedom from pain – or happiness and freedom from unhappiness in some versions.
A pleasure-based theory of desire is a theory about language. It says that the only thing worthy of the name “desired” is that which brings about pleasure when it seems to obtain, or displeasure when it seem not to obtain.
Somebody who holds a pleasure-based theory of desire can be perfectly comfortable with the thought that we can be motivated to pursue other ends – ends that have nothing at all to do with pleasure. But they would deny that these ends we pursue without pleasure are actually ‘desired’. They would need some other word to describe ends pursued without pleasure.
We see this in the way they argue for their theory.
One of the arguments in the debate concerns a hypothetical case of “weather watchers”. We mentioned them in our last episode. These are beings who have a preference for one type of weather over another, but are not interested in any way in acting so as to bring about one type of weather over another. They say, “Whatever comes will come, but we prefer sunshine over rain.”
ALONZO: Well, yeah. Galen Strawsen uses this to argue that our concept of desire tracks with our concept of pleasure more closely than it tracks with our concept of motivation.
He says that it makes sense to say that these weather watchers desire something that they are not motivated to bring about, which means that the concepts of desire and motivation don’t always fit together and action-based theories of desire are to be rejected for that reason.
Now, I think he’s wrong.
Let’s assume that we gave these creatures an option to pick tomorrow’s weather and they answered with a shrug, I would take that to be a shrug of indifference. They don’t care. If they did care, they would pick.
LUKE: No, Alonzo. Don’t go there. You are stepping into a tar pit from which you can never escape. You’re not discussing any difference in how the universe actually works. In fact, you’re not talking about anything real at all – these weather watchers don’t even exist. You are getting bogged down in the philosophy-of-mind equivalent of the question of whether a tree falling in the forest makes a sound when nobody is present.
Here is a place where, instead of getting bogged down in a discussion about definitions of words, we can just replace the symbol with the substance and get on with the factual investigation.
If a tree falls in the forest and there is nobody around, does it set off a series of compression waves propagating through the air such that, if there was a normally functioning human present, it would register those waves as a sound? Yes. End of the discussion.
If a tree falls in the forest and there is nobody around, does it set off a series of compression waves propagating through the air such that it actually does register in a brain as a sound? No. End of the discussion.
We don’t need to endlessly debate what the ‘best’ definition of the word ‘sound’ is, or what the ‘correct’ definition of the word ‘sound’ is. We don’t need to do conceptual analysis on the word ‘sound.’ We can just replace the word ‘sound’ with whatever facts of the world we intend to bring to mind with the word ‘sound’, and then answer the question about those facts of the world.
And, rather than debate whether our intuitive concept of desire tracks with motivation or with pleasure, one of the ways around this question is to just discuss motivation and pleasure and skip the question of which one is called ‘desire’.
Now, desirism is built on motivational states – not pleasure-states. Admittedly, we called it ‘desirism’ because we took for granted an action-based theory of desire. But if somebody wants to challenge us on that, we can tell them to call the theory motivationism instead. We don’t care. We are not going to fight over definitions. I cannot tell you how many oceans of ink philosophers have spilled arguing over definitions. I cannot tell you how much incredible brain power philosophers have wasted arguing over definitions instead of arguing over facts.
ALONZO: Okay, let me give you another argument I’ve heard when people are debating different theories of desire – one that doesn’t involve anything like imaginary weather watchers.
Here it is, a nice Saturday morning – a great day for hiking in the mountains – and I am stuck here washing my neighbor’s car. Why am I washing his car? It’s a nice car and I said that – if he would let me borrow it on Friday night when my wife and I were celebrating our anniversary – then I would wash it Saturday morning. So, here I am on a Saturday morning stuck washing my neighbor’s car. I don’t want to be washing his car. I want to be hiking through the mountains. But, here I am.
Here, I am obviously motivated to wash my neighbor’s car. That’s what I am doing, after all. But what I want to be doing is hiking through the mountains. The argument here is that this is a case in which people use words like ‘desire’ or ‘want’ to refer to the option that gives them pleasure, not the option that motivates them.
Now, here’s the answer I would have typically given to this kind of objection.
The word ‘want’ isn’t tracking with what gives pleasure, but what is desired as an end.
We used this idea when we talked about Alph gathering stones. Without ever mentioning pleasure, we talked about Alph’s motivating desire to gather stones. Because of the limited number of stones, sometimes Alph had to scatter stones so that he would have stones to gather again. We admitted that he did not want – that is, did not desire-as-an-end – to scatter stones. It was a chore. It was work he had to do in order to return to a state where he was gathering stones again.
When it comes to the washing the car example, we can say that I don’t ‘want’ to be washing the car in the sense that it’s not what I desire as an end, though it is what I desire as a means. So the action-based theory of desire can accommodate this use of our language just fine.
LUKE: Okay, but, Alonzo: listen to me.
It doesn’t matter.
It . . . does . . .not . . . matter.
ALONZO: What? You come back from your reading and not only have me change the words I have been using to describe desirism, but you would have me just toss these arguments aside as well? I spent a lot of time on these arguments.
LUKE: You wasted a lot of time on those arguments.
First: notice that, once again, this is not a dispute over how the universe actually works. The first question you should ask is whether these different theories make different predictions about what happens in the world – other than predictions about how people will use a term.
If it makes different predictions, we should be looking to discover which predictions are more accurate. If it’s a dispute about words, we shouldn’t be trying to discover the best super-dictionary definition of the term. We should be asking which definitions will be more efficient in communicating ideas about the world.
And if it’s not clear which definitions will be more efficient for the purposes of communication, then we can get around the problem by jumping past that particular symbol – that particular word – and talking directly about the facts in the world you mean to call out by using that word in the first place. You can talk about motivational states or pleasure states or whatever instead of talking about the word ‘desire’ and arguing about which definition of ‘desire’ to use.
ALONZO: Well, I suppose.
It does sound like another Planet Pluto debate now that you put it that way. In the case of Pluto, one group wants to define the word “planet” so that it includes Pluto. Another group wants to define “planet” so that it excludes Pluto. These people have no real dispute about what Pluto is. They only disagree about what language to use when talking about Pluto.
Here, in the case of desire, instead of debating whether non-motivating pleasure, or motivating non-pleasure, gets to count as a desire, we can talk about non-motivating pleasure or motivating non-pleasure directly, if it exists at all.
LUKE: Right. It doesn’t really matter who wins the “pleasure is necessary for desire” debate any more than it matters who wins the “Pluto is a planet” debate. The former will not change the way the brain works, and the latter will not change the orbit, size, or composition of Pluto.
ALONZO: Looking at it this way, I do have one more point to raise. I have a practical consideration to raise for why the word ‘desire’ should, for the sake of efficient communication, track with motivation rather than with non-motivational pleasure.
LUKE: Good. Let’s hear it.
ALONZO: We have reasons to be concerned with what motivates people to act in different ways. Your actions have an effect on my world – and my actions can have an effect on yours – for better or worse. You have reason to worry about me flying airplanes into skyscrapers – I have reason to worry about whether you will give me help in an emergency.
But I have no reason to be concerned with your non-motivational pleasures. And you have no reason to be concerned with mine. I don’t even have a reason to want you to help me to realize the experience of non-motivational pleasure. By definition, such states are non-motivational. A non-motivational pleasure doesn’t give me any reason to act any differently, and it certainly does not give me a reason to cause you to act differently.
So, they’re pointless to talk about. It’s more useful to use the word ‘desire’ to talk only about brain states that motivate people to perform actions.
LUKE: Sure. That’s the type of practical consideration that actually makes sense to argue about.
It’s like the practical considerations that went into the debate over whether to call Pluto a planet. It is efficient, in science, to lump similar objects under the same name. Pluto simply has more in common with trans-Neptune objects than it does with planets. So, for the sake of efficiency, we no longer use the word ‘planet’ to talk about Pluto.
Some scientists still say that things that are big enough to be round because of their own gravity need to be lumped together because they are alike in many ways, and they want to use the word ‘planet’. So, for the sake of convenience, Pluto and a few other objects in our solar system get lumped under the name “dwarf planets”.
Notice that astronomers didn’t have to look at how the word was used in the past or whether Pluto fit our intuitions about the meaning of the word ‘planet.’ In fact, those who supported this change were ready to admit that they are breaking with tradition on how these terms are used because they value communication efficiency over tradition.
And when we discuss desire, we value communication efficiency over tradition.
But: I have to stress that if somebody comes back and argues that using the word ‘desire’ to talk about something other than motivational states is more efficient for communication, and this debate is not easily resolved, then I’m not going to waste my time arguing over definitions. I’m going to say, “Fine. Whatever. I’m not going to argue over definitions. We can just replace the symbol with the substance – replace the word with the facts we meant to call up by using the word – and then we can argue over the facts instead of arguing about the word. It’s really not worth my time to argue about definitions.”
These tree-falling-in-a-forest arguments are a huge distraction from talking about how the world actually works.
ALONZO: Okay, so when we come back, we’ll be arguing not about our definitions for words like ‘pleasure’ and ‘desire’, but about the facts in the world we might mean to call to mind when we use those words.
LUKE: Great! I’m looking forward to it.
(in order of appearance)
- “Hour Five” from Somnium by Robert Rich
- “Envelop” from The Magic Place by Julianna Barwick
- “Theory of Machines” from Theory of Machines by Ben Frost
- “Krovinka Moya” from Kasha Iz Topora by Gultskra Artikler
- “The Bath of Stars” from Winter Songs by Art Bears
- “The Oh of Pleasure” from Deep Breakfast by Ray Lynch
* marks royalty-free music. With copyrighted music, we use only short clips and hope this qualifies as Fair Use. Fair Use is defined in the courts, but please note that we make no profit from this podcast, and we hope to bring profit to the copyright owners by linking listeners to somewhere they can purchase the music. If you are a copyright owner and have a complaint, please contact us and we will respond immediately. The text and the recordings of Luke and Alonzo for this podcast are licensed with Creative Commons license Attribution-Noncommercial-Share Alike 3.0, which means you are welcome to republish or remix this work as long as you (1) cite the original source, and (2) share your remix using the same license, and (3) do not use it for commercial purposes.