Relational Realism in psychology, physics, embryogenesis, and math
Relational realism is the idea that relationships are what is real. I think of it as meaning “properties are given by relationships, and only by relationships”. This implies that properties are never inherent to the object we attribute them to, but instead are only due to the relationships the object has with other objects. For example, sugar does not taste sweet in and of itself but only in how it interacts with a typical human tongue and brain.
Relational realism is counterintuitive because it doesn’t match how we typically conceive of our first-person experiences. When we experience the world with our sense perceptions, it seems like we perceive properties as being inherent to the object rather than being relational. We perceive water as wet, stone as hard, apples as red, music as beautiful, etc. Call this view “storage realism”, where an object has a property because the property is stored in it.
Language provides a instructive example of how we seem to experience things as storage realism, yet can see that relational realism must be true. When you hear the word “tomato”, or read it, it seems to carry with it its meaning into your brain, as if the meaning of the word is imbued into the letters or phonemes themselves. But of course, this isn’t true: The word “tomato” only means what it does in a particular context, e.g., the context of the English language. In another context, the same word might carry a different meaning, or no meaning at all.
The actual meaning of the word “tomato” is relational: its meaning is a matter of understanding its position in a web of relationships. Perhaps most crucially, when you hear the word “tomato”, it’s often in the context of seeing a tomato. So over time, you learn to predict that the word “tomato” refers to tomatoes. Since we can imagine a version of the English language where we say “badoofle” when we see a tomato instead of “tomato” when we see a tomato, we can see that the meaning of the word “tomato” is not contained within the word itself but within its relationships with other phenomena.
The meaning of tomato—the concept, not the word—is also relationally real. No two tomatoes are alike, nor do they share some essence, such as a genetic signature universal to all tomatoes and is contained within nothing else. You learn to group tomatoes together not by detecting some property stored within the tomato but by finding functional similarities across a collection of instances. That is to say, you determine that it makes sense to treat the squishy red objects in your hands or on your plate as basically the same thing.
It might seem tempting to say that the essence of a tomato is contained in its relationships, but this isn’t true either. Relationships are flexible; you construct them based on your memories and your goals. Whether a tomato is a fruit or a vegetable to you depends on your history in combination with what you want to do with a tomato. There’s no essence anywhere, just a position in a system of relationships.
Relational realism is the consequence of a view that rejects the idea of natural kinds due to the complete lack of evidence as to their existence. Natural kinds are the idea that some categories exist as natural facts of the universe; there is no evidence that natural kinds exist, so the alternative is to suppose that categories are constructed.
Relational realism in psychology: psychological constructionism
When you see someone smile, this seems to carry a meaning with it, just like the word “tomato” seems to carry the meaning tomato through your ears into your brain. But in reality, facial expressions carry no inherent meaning; their meaning is constructed just like the meaning of anything else. Sometimes people smile when they’re in pain, or during a funeral, or confused. Meanwhile, sometimes people cry at a wedding. Which emotion you construct when you see a face move depends on the web of relationships that the facial movements are situated within. This includes relationships within yourself; if you just saw a horror movie where the killer has a big, friendly smile, then you might be creeped out by a smile that would otherwise strike you as friendly.
This is counterintuitive because when you see someone smile in a context where it makes sense to construct it as happiness, you just think “They are happy” rather than “Based on my previous experiences and my goals, as well as the context of this situation, it will be useful to construct a perception of them as happy”. Psychological constructionism does not match what we perceive to be our first-person subjective experiences. But there is a lot of evidence that it is true.
Nevertheless, when you see something, like a happy facial expression, it seems like you just see it, as a direct reproduction of reality, rather than something that is constructed. Perhaps the most counterintutive idea of relational realism in psychology is the idea that our senses do not perceive the meanings of objects but instead construct their meanings.
When you see or hear something, the interactions are turned into electrical signals via complex chemical reactions that are passed to the brain through nerves. These electrical signals do not carry with them the sights and sounds from the external environment. Instead, they are changes in voltage. These voltages don’t carry some external meaning into the brain. Instead, the brain has to construct their meaning by considering what the changes in voltage mean in light of a web of connections formed out of previous events and other signals the brain is processing, such as interoceptive signals.
But this is not intuitive. Instead, we experience our senses as if they simply play a movie of the external environment, as suggested by this image describing the idea of the homunculus argument:
In this movie-like understanding of the senses, your senses simply transmit the external state of reality to your brain encoded as electrical signals, and your brain decodes those signals back into their inherent meaning. But in reality, there is nothing to decode. If there were a person in your brain, he wouldn’t have a movie to watch. He’d have to figure out what’s going on based on a bunch of electrical messages that carry as much inherent meaning as “SOIEFJWOE” does. There is no “decoding” this message for the same reason that there is no decoding the marks in the sand left by the wind. Instead, meaning is constructed by seeing what other messages '“SOIEFJWOE” correlates with. Meaning is constructed, not deduced.
Here are three interesting lectures on relational realism and psychological constructionism by Lisa Feldman Barrett:
Economist Friedrich Hayek also argues for relational realism and psychological constructionism in his book, The Sensory Order.
Relational realism in physics
I don’t know anything about this, but it is a perspective that some people have. See Helgoland by Carlo Rovelli to learn more about it. See this Stanford Encyclopedia of Philosophy on relational quantum mechanics as well. If relational realism is true of physics, then it shouldn’t surprise us to learn that it is true of psychology as well.
Relational realism in embryogenesis
Embryogenesis is the emergence of an embryo, a complex multicellular organism, from a single cell. How can this happen? Traditionally, the answer to this question has been thought to lie in the genes, but today we know this isn’t true. Instead, the answer lies in how cells interact with each other as they construct the embryo via their relationships.
Embryogenesis is a complex, dynamic process in which cells move around, divide, and self-organize into larger structures such as tissues and organs. They do not do so by following genetic instructions or by being manipulated by some external or central controller like a brain or nervous system. Instead, as cells change where they are and what they are doing, they reshape the landscape for other cells, altering the affordances and constraints that the other cells use to determine their behavior. Thus, the cells determine each other’s behavior by adjusting how they relate to one another, leading to the creation of the familiar form of the embryo in the aggregate. As Esther Thelen and Linda B. Smith put it (p. 260),
This picture is much different from one that casts the genes as the puppeteer, pulling the right strings at the right time to control the ensuing events in the cells. In a dynamic view, we consider the marionette and the puppeteer as affecting each other equally. Or, more accurately, we do away with the puppeteer and the marionette altogether: What is important is the relationships among the strings as they pull and then become slack.
Relational realism in math
Relational realism is not predominant in science, but in mathematics, it is the mainstream view, under the heading of structuralism. According to structuralism, mathematical objects have no inherent meaning or properties but are instead defined entirely by their relationships with other mathematical objects. If math can be understood in relational realism terms, then it seems plausible that physics and therefore everything in our world can be understood in relational realism terms as well.
Relational realism in mathematics offers a very clear depiction of what it means for properties to be given by relationships, and only by relationships, instead of being stored in the object. For example, consider this element:
Above is pictured an element, call it a. Sets are the building blocks of mathematics, and elements are what exist inside a set; they determine sets and therefore can be thought of as the true primitives of set theory. Yet this element has no properties of its own. It is not actually a black circle—that’s just for visualization.
Now see what happens when we put a into a relationship with another element, as denoted by the black line:
Suddenly, the element a has gained a new property: it is a minimum element, meaning it is smaller than every other element in the order. This isn’t a property stored in a or in b. Instead, it is a property constituted of their relationship with each other. Different relationships would make a a maximum element, or neither minimum nor maximum.
Other familiar properties in math are also relational. Consider a one-element set, e.g., {!}, where the element ! is the sole member of this set. Logically, this set is different from the set {?} because they have different elements. However, this “difference” makes no difference; it does not affect anything and so doesn’t deserve to be called a property. For all intents and purposes, the sets {!} and {?} are identical. There is no functional distinction between them.
In math, the phrase “no functional distinction” takes on a literal meaning. The sets {!{ and {?} are isomorphic to each other, meaning there are a pair of functions between them that invert each other. Moreover, there is only one isomorphism between them, a unique isomorphism. The practical consequence of this is that for any argument where we are using the set {!} to do something, we could replace ! with ? and get the exact same answer.
Because all one-element sets are functionally equivalent, mathematicians think of the category of sets as having only a single one-element set, even though there are an infinite amount of them in reality. The property of being a one-element set doesn’t even have to be described in terms of having just one element stored in a set. Instead, it can be understood in purely relational terms. By definition, every set has exactly one function to a one-element set, and one-element sets are the only kind of set with this property. Since functions are relationships between sets, we can describe the property of being a one-element set in purely relational terms, as a relationship between sets, without having to mention the elements at all. This contrasts with what a reductionist view of mathematics might expect, where sets are the building blocks of mathematics, and elements exist inside sets, implying that the properties of sets should be stored inside the elements.
Here is more information about structuralism in math: