Hey, SpectrumDT! You wrote a post asking what David Chapman means by "meaningness" and "meaning". I'm quite familiar with Chapman's work, so I gave a reply. We've been having some productive back and forth in a thread over there, and I thought we should try to elevate it to a dialogue post to continue our discussion.
To bring readers up to speed, my answer is that by "meaning" Chapman means "purpose". Meaning is about orientation towards, salience, importance, and a sense of why things matter.
You then asked:
Hm... do you understand Chapman's claim that meaningness (or purpose) is neither objective nor subjective, and that existentialism is therefore false? Because to me purpose seems subjective, which suggests that existentialism is true.
I responded:
Yes. This is a place where I think David's phrasing is a bit confusing because what he actually means to say is "not only subjective".
These's two ways to get a better handle on what not only subjective means.
The first is to understand that subjectivity is really intersubjectivity. That is, we each have our own subjective experience, but we learn about others' subjective experiences and treat them as facts, and then this creates a social reality that feels objective because it contains facts that we believe are true even if we don't have direct knowledge of them ourselves.
But David is using "subjective" here to mean "only subjective" as in solipsism, which is a common view many people adopt and needs to be rejected. This happens because people correctly catch on to the subjective part and then fail to understand how their beliefs about the subjective experiences of others impact their own beliefs, sort of like they just see one level of the system and not the whole thing. That's the kind of subjectivity he's pushing back against.
Now I want to be clear that intersubjectivity is not the whole story when it comes to the complete stance. David's also rolling into it the idea that meaning is not something that comes purely from doxastic or epistemic knowledge. It involves many other ways of knowing (and not knowing) that are perhaps beyond the scope of the question here. There's a sense in which meaning creates itself and is orthogonal to the objective/subjective distinction, but I don't think I can explain that idea in a comment, and is arguably why David's writing a whole book.
Since then you've been following up trying to understand what I think intersubjectivity has to do with the search for meaning.
Any other context you want to bring over to get us going here?
Yes. Among other things, you wrote:
And why any particular person or system has the particular purpose they do is a complicated but ultimately physical question about teasing out causality to find why things are the way they are.
...
Living things generally have purposes like survive and reproduce. Tea kettles have purposes like heating whatever is inside them. Thermostats have purposes like keeping the thermometer reading the desired temperature.
As I see it, this is still compatible with existentialism. (Here I am using existentialism in Chapman's sense of "the belief that meanings are all subjective".)
The topic of the purposes of tools (as intended by the human being who made or designed the tool) seems irrelevant to the topic. I do not think these are the kinds of "meanings" or "purposes" that anyone searches for.
The existentialist would agree that living beings have been shaped by evolution, and that this could in some sense be called a "purpose". But evolution selects those beings that are best at procreating at all costs. Are you saying that we can thus prove that the "purpose" of a human being is to spread its genes as much as possible? I do not think this is what Chapman is gesturing at.
Moreover, if we look at ethics and customs, these have varied wildly across time and space. Are all values equally good as long as we can find an example of how this value has arisen as a meme for historical reasons?
Or am I completely missing what you are trying to say?
Chapman obviously thinks he knows which ethics and morals are "correct". He often judges moral claims as correct or incorrect - here, for example. Ironically, on the page I just linked to, Chapman says that Buddhist tradition is full of moral admonitions with no justification, but I have pretty much never seen Chapman justify his own moral claims, so I do not know what Chapman thinks a proper justification is supposed to look like.
As I see it, this is still compatible with existentialism. (Here I am using existentialism in Chapman's sense of "the belief that meanings are all subjective".)
My reading of Chapman is that he's using "existentialism" to mean something more like "the belief that meanings are only subjective" Maybe this is the same thing you mean, but there's an important, subtle difference that I was pointing at in the discussion of intersubjectivity, which is that subjective and not-objective aren't the same thing.
I'd agree that there are versions of existentialism that are compatible with what Chapman is saying, and my memory is that Chapman's use of "existentialism" is usually as a synonym for "nihilism".
The existentialist would agree that living beings have been shaped by evolution, and that this could in some sense be called a "purpose". But evolution selects those beings that are best at procreating at all costs. Are you saying that we can thus prove that the "purpose" of a human being is to spread its genes as much as possible? I do not think this is what Chapman is gesturing at.
I'm saying that purpose is rooted in how evolution has shaped us. Survival and reproduction are not the whole of purpose, but they are, practically speaking, the foundation of what we care about.
But I think I got ahead of myself and jumped to talking about my own ideas about purpose without actually addressing the question where Chapman argues that meaning comes from.
Chapman's answer is interaction. He uses the metaphor of a rainbow, where a rainbow exists only because of the interaction of the sun, water droplets, and us, the observer. To quote Chapman directly:
To make the analogy explicit, meanings:
- are interactions among people and circumstances
- are physical phenomena, but not physical objects
- have no definite locations (whether inside or outside heads)
- are observer-relative, to varying extents
- are usually well-understood, and 100% metaphysics-free
- are mostly not subjective, mental, illusory, or dependent on magical properties of brains
- are not inherent in objects
- mostly are publicly verifiable, so reasonable observers mostly agree about them.
His take is that purpose is a primary dimension of meaning. My argument, that I've partially made elsewhere, is that meaning arises from purpose to such an extent that I think we can just sub in "purpose" for "meaning" and call the rest obvious corollaries.
I should be clear about something, though, which is that I've equivocated on "purpose" in my answers as we've been conversing. In my original, short answer that "meaning is purpose" I meant this in a non-jargon way of offering a synonym that I thought would do a better job of explaining the concept Chapman calls "meaning" if you were coming in cold and wondering which of many definitions of "meaning" Chapman is talking about. However, in these later posts, I've slipped into talking of "purpose" as jargon from cybernetics, where "purpose" means both the goal a closed loop feedback system is trying to achieve and a more expansive definition of what the system seems to care about in service of that goal (noting that humans can be modeled as multiple, layered closed loop feedback systems with multiple, often conflicting goals, if that helps your intuitions here).
In that case, what ARE the purposes of human beings? And how do we ascertain them?
Do all humans have the same purposes, or does each individual have its own more-or-less unique purposes? If the latter, what are YOUR purposes, and how can you tell?
Since we are evolved beings, I'd argue that our ultimate purposes are survival and reproduction. Or maybe you could argue that it's the replication of our genes, which I'd see as roughly the same thing. One of the challenges of cybernetics is that it makes really clear that ontology matters: how you frame what counts as a system can change what its purpose seems to be.
We know that these are our purposes because we are evolved, biological creatures originating from Earth; like all such creatures, we have a particular form to our lives inherited from how life began on Earth. Some of these features might be Earth specific, like replicating RNA and DNA. Others seem to be features likely true of all life in our universe, like needing to consume energy to live to overcome entropy.
But I should note that this is a contingent, not an ultimate, answer to what our purpose is and where it comes from. It's contingent on our understanding of the world, and would be different if we had less scientific knowledge, and may be different in the future when we have more. I think this is why Chapman sticks to talking about meaning and interaction: it's staying more at the level of experience than trying to answer questions about how the mechanisms of that experience arose, whereas purpose requires understanding the origins of the mechanism.
Do all humans have the same purposes, or does each individual have its own more-or-less unique purposes? If the latter, what are YOUR purposes, and how can you tell?
I'd make a distinction between purpose and goals. Purposes are what we've been talking about here and are fairly universal, although to be clear this is a pattern and has to be instantiated each time in each being. Goals are the person-specific things people care about and try to achieve, and they are motivated by their purposes but also shaped by how they think they can achieve their purposes in the world they find themselves in.
For example, maybe you care about writing great posts on LessWrong. Why? Well, maybe you see writing a good post as a way to attain high status, and humans want high status because it often gives them access to resources that they need to survive and helps them attract mates to reproduce. We don't necessarily consciously think through all this, being adaptation executors; we just feel compelled to do things, and looking at those compulsions we get data about what our purposes are by inferring the pattern from the specific observations of behaviors.
As for myself, I care about a great many things. And although there's sometimes a lot of twists and turns as to why I care about them, I'm compelled by the theory that I ultimately care about survival and reproduction and that these are my purposes because I've yet to find behavior in myself that can't be explained by this theory (though I admit this is not as good as prediction since we can always rationalize anything to fit a desired explanation if we are epistemically dishonest).
Thanks for the explanation. As far as I understand, you are saying that a person's purpose manifests as a set of underlying preferences, which in turn give rise to our conscious preferences and goals.
My conscious preferences lead me to suspect that my underlying "purpose" is not survival and reproduction. When I was unhappy, I have at times felt almost suicidal, which suggests that survival is not paramount. Certainly neither is reproduction; it took my wife some work to convince me to have one child, and I do not want more. (I also suggested adoption, but my wife rejected that.)
As far as I can determine, my greatest values are my own well-being and the well-being of others.
But back to my main question: How is this not subjective? To me, all the above seems just as subjective as preferences regarding food, music, and the like. What is the difference?
My conscious preferences lead me to suspect that my underlying "purpose" is not survival and reproduction. When I was unhappy, I have at times felt almost suicidal, which suggests that survival is not paramount. Certainly neither is reproduction; it took my wife some work to convince me to have one child, and I do not want more. (I also suggested adoption, but my wife rejected that.)
To be clear, my theory is not that people feel like they are driven to survive and reproduce, but that these two purposes are the purposes we are given by the evolutionary process that created us. Some people do feel directly a desire to survive and reproduce, but most, like you, consciously care about well-being. The question is, "why care about well-being at all?", and what I offer is that survival and reproduction give us one good explanation for why minds would care about well-being (and many other things): well-being is correlated with us surviving and finding good conditions under which to reproduce and have healthy offspring.
Depression and suicidal thoughts, by my theory, happen when purpose (as mediated by goals and desires) seems unattainable and thus some part of the mind reckons that there's no point in trying or even going on living. My theory predicts that the most reliable way to treat depression is to help people by making their lives better, specifically in ways that address whatever critical beliefs are causing them to believe they cannot achieve their basic goals and fulfill their purpose.
But back to my main question: How is this not subjective? To me, all the above seems just as subjective as preferences regarding food, music, and the like. What is the difference?
It is subjective, it's just not only subjective.
The difference is that even your preferences are not just something you make up. What you like or dislike is influenced by factors outside your control. You didn't decide to like sugar and fat, for example. We evolved an innate preference for foods high in sugar and fat because they were rare in the past and we needed strong motivations to seek them out because that conveyed a survival and reproduction advantage. Similarly, all of your preferences are at least in part influenced by something other than what's going on when you think and feel.
But of course these preferences aren't exactly objective, either, as in there's no right answer about what preferences to have, just as there's no right answer about what you should care about. It's just that you like and care about things as the result of causes and conditions outside your control, making them not purely subjective. That's why I bring up the idea of intersubjectivity: there's a subjective element because it's all known through experience, and then there seem to be forces outside our experience acting to create it.
Thanks for the explanation. It makes sense, but:
1. When you explain why goals that seem unrelated to survival and reproduction are actually caused by a drive towards survival and reproduction... that makes your "theory" seem suspiciously unfalsifiable. Does your "theory" pay rent in terms of expected outcomes? Does it make any nontrivial testable predictions?
Yes, I admit, I'm already pretty sold on the fact that survival and reproduction are fundamental to purpose because of evolution. I don't have a detailed gears-level theory of how purpose works in all cases. You should read my arguments there as looking for plausible explanations in light of proven facts that they must fit, but like many similar evo-psych theories, it's hard to prove that the explanation is correct just yet because we don't understand enough about how the brain works. Nevertheless, I'm pretty certain about the survival and reproduction part because I'm very certain about evolution.
2. You seem to be moving the goalpost for what counts as subjective. According to your explanation here, taste in food and entertainment (often considered the "archetypal" subjective thing) is also not only subjective. What is purely subjective?
Nothing. Everything is interconnected in a way that "true" subjectivity doesn't exist. It only seems like some things are only subjective because we can imagine ourselves wanting or believing other things. This is roughly equivalent to how we don't have free will, but it feels to many people like they do because they can imagine counterfactuals.
3. I do not understand how this is supposed to make anything feel more meaningful (which, I believe, is what Chapman is after). Apparently you are saying that our goals and interests are shaped by the forces of (biological and cultural) evolution, and that is why they are meaningful. The nihilist would say that our goals and interests are shaped by the forces of (biological and cultural) evolution, and that is why they are meaningless.
I don't know if the theory should, on its own, make anything feel more meaningful. The theory just explains how meaning works. For things to actually feel more meaningful, you have to live a life full of meaning, and if someone strongly believes that life is meaningless, they will interpret their experiences to be meaningless instead of meaningful.
Chapman and I have both solved this problem in our own lives with Buddhism (though in very different lineages). It's why I'm so pro-religion despite all the epistemic issues with many religions, especially Abrahamic ones: religions consistently help people live lives of meaning.
I look at the world and see that it's shaped by forces beyond my control and say "ah, yes, everything is exactly as it was always going to be, including myself and everything I care about" and find contentment and meaning. Others find this disorienting—and my past self was included among those others!—because, I would argue, they are fundamentally dualists who believe that their self is separate from the rest of the world and therefore special, and feel sadness when they look at the world and see evidence that maybe they aren't so special and things aren't how they "should" be.
I'm not sure if Chapman's project can actually work. He explains a lot of theory, but very few people find their way to meaningness through theory alone (including, to the best of my knowledge, Chapman, and certainly myself). There's a kind of missing referent problem, where an idea can be explained clearly, but in terms of a referent that the reader is missing, and so they end up not getting the message. I think this problem is prevalent in a lot of Chapman's writing (and, I freely admit, in my own).
4. Your evolution argument here looks nothing like anything in Chapman's book. How can you be sure that the meaningness you have in mind is the same as what Chapman has in mind?
I don't know if Chapman would endorse that part of my argument. It's how I make sense of the theory and ground it. Chapman provides a different presentation that avoids getting into these mechanics. I think his arguments are weaker than mine in some ways for not addressing it, but avoiding arguments about evolution and survival and reproduction avoids having to address a large domain of concerns, which I think is a reasonable choice given the scope of Chapman's work.
Hm. All right. Thanks. Let us try to go another route. I would like to go back to the topic of ethics. Earlier I asked this question, which I do not believe you have given a direct answer to:
Moreover, if we look at ethics and customs, these have varied wildly across time and space. Are all values equally good as long as we can find an example of how this value has arisen as a meme for historical reasons?
In other words: Are some ethics better than others, and if so, why? (Other than the most contrived edge cases.)
I'd say yes.
Although we each may care about different things to different degrees, all humans share a common set of concerns. We desire safety, well-being, love, esteem, understanding, autonomy, and meaning. We want to be happy, to express ourselves, and to contribute. We only find these concerns missing or distorted in extreme outlier societies that have proven unstable because people yearned for what they were missing.
So when we think about norms, we can judge them by how well they help us live the lives we want to live. For example, suppose we want to adopt a norm against gossip. We can check to see if it works. We can look at other societies with such a norm and assess if that norm contributes or detracts from fulfilling the people's needs. Or we could run an experiment, adopt the norm for a while, then assess how it's going after a set period of time. Are people happier or sadder when they are forbidden to gossip? Are they safer or less safe? And so on.
The ability of norms to serve humans is, I'd argue, a major theme within Chapman's work even if it's not always his main point. Many systems of norms are designed not for the benefit of humanity, but for the benefit of one person's or group's idea of what would be best for humanity. The more wrong that idea is and the less well that idea maps the territory, the worse the norms it produces are likely to be. This is why lots of attempts to replace old systems of meaning have failed: they have a wrong idea about humans—or just don't really care about humans much at all!—and thus fail to provide real benefits to humanity, making them bad norms.
All right. Thanks. I have two questions:
The ethics you describe sounds like a variant of utilitarianism. Do you agree? If not, how does it differ from variants of utilitarianism?
I'd say that this is not quite a utilitarian stance. Utilitarianism proposes that we can measure goodness and we can measure it precisely enough to maximize it. I think utilitarianism doesn't work because we can't actually measure goodness precisely enough to effectively optimize for it directly, nor do I even think it's necessarily correct to collapse things we care about into a single measure of goodness, as this doesn't seem to reflect the way preferences of individual humans work.
Do you know the moral foundations theory of Jonathan Haidt and others? Do you agree that the 6 foundations are all important goals in themselves? (Care, fairness, loyalty, authority, sanctity, liberty.) What would you say to someone who argues that loyalty, authority, and sanctity are at best instrumental goals, not goals in themselves?
Yes. In fact, as just hinted, I think a big challenge for utilitarianism is that people care about multiple things and their concern along different dimensions of care is not fungible.
I wouldn't say that the foundations are goals. They are cares or concerns. Not specific outcomes to be achieved, but directions in which to go that may motivate specific goals. For example, I care about loyalty, so I have a goal of supporting my friends through tough times, and think a good friend would support me in my times of need.
I honestly don't know exactly how foundational moral foundations are. My theory is that they exist as adaptations that serve the deeper concerns or purposes of survival and reproduction, and so that would make them instrumental to our evolved purposes. But psychologically we're not fitness maximizers, we're adaptation executors, so we don't actually need to directly know or care about survival or reproduction. Moral foundation theory offers a mechanism for us to be motivated to survive and reproduce: we have psychological motivations to carry out actions, and those motivations have evolved because they helped our ancestors survive and reproduce.
As far as I can tell, your model does not really help us "ground" morality in anything solid, so as to help us resolve any nontrivial moral dilemmas. It seems to me that we can boil your explanation down to the following conclusions:
That is not much of a conclusion.
As I mentioned, Chapman clearly thinks he knows and understands a lot about morality, but I remember nothing from his writings that says anything more useful about morality than my summary above.
It's perhaps not the desired conclusion, but I think "we will have to wing it" is correct because it's what we're already doing. Perhaps in the future, with the help of AI and a long reflection or something similar, we'll be able to wing it less, but I see this as a fundamental challenge because we'd like certainty about morals but epistemic uncertainty limits our ability to know what is moral.
That said, we know more than nothing. I think it would be foolish to ignore our moral intuitions because they provide real information about how to make the world a good place to live, but they are also not the end of the story in determining what we should do.
I don't know if Chapman is right about what he thinks are good norms. I generally like his takes, but that's because they agree with my own thinking, and Chapman and I have a lot of shared values. I'm sure lots of people disagree, and alas I think we'll keep disagreeing so long as we remain so uncertain about what norms would be best.
All right. I think we can end this exchange here. I still feel that neither Chapman nor you have taught me anything substantial about meaningness, but thanks for the discussion!