BB #78: Relational Theories

I read Three Roads to Quantum Gravity (2001), by Lee Smolin, a theoretical physicist whose thoughtful style I’ve always appreciated. I don’t always agree with his ideas, though. This book is about Loop Quantum Gravity, in which Smolin has invested considerable effort, and that idea I’m utterly neutral on. It does seem to make more sense than string theory.

One notion I have a lot of trouble swallowing (like a cup of coffee with eight lumps of sugar) is the relational view. (As a philosophy, relationism. Al stayed home.) It’s a fundamental aspect of LQG.

But I (and apparently Kant agrees) think Leibniz was wrong.

Gottfried Leibniz (1646–1716) was a German polymath back when that actually meant something. The guy covered a lot of ground. Depending on who you talk to, either he, Newton, or both, discovered the calculus so many high school students hate.

Relevant here is that he’s also famous for holding relationist views about space and time. These in contrast to Newton’s view of absolute space and time. The relationist view makes relations the primary ontology.

We learned Newton was wrong about space and time when Einstein showed that spacetime is a unified fabric, and space, time, simultaneity, and the idea of “now” are strictly relative and local. Newton being wrong about this doesn’t necessarily mean Leibniz was right, though.

Einstein showed that space and time are relative, which is not the same thing as relational. Special Relativity is a theory of objects. (One could call the view-dependent nature of relativity a relation between perspective and view, but it’s a secondary property, as I’ll explain below.)


I’ve seen references to Immanuel Kant (1724–1804) debunking both Leibniz and Newton on their time and space views. It’s one of those things on my long list of things to check out if I ever get around to it. Kant is heavy going (most serious philosophers are), so I’ve put it off. Math and quantum mechanics are eating up my CPU cycles these days.

Kant did feel that time (foremost) and space were our most basic intuitions, and that they framed our every perception and thought. That only means we’re “wired” to comprehend reality in their context. It doesn’t require that time and/or space are fundamental. (I’ve always found it suggestive and compelling, though.)

§ §

Smolin agrees with Leibniz and promotes Loop Quantum Gravity as a relationist physical theory.

In fact, that’s related to the theory’s biggest sales point, that it’s a background independent theory. Like General Relativity, it creates the background. Space, and everything else, emerge from these relationships.


Except time. Smolin doesn’t touch on time much in Three Roads, but I just started his 2013 book, Time Reborn, and there he says plenty in just the Preface:

“More to the point, I no longer believe that time is unreal. In fact, I have swung to the opposite view: Not only is time real, but nothing we know or experience gets closer to the heart of nature than the reality of time.”

I laughed out loud when I read it. I’ve been saying the same thing forever (see these posts). Here it’s not that I agree with the expert in the book, but that the expert, after long consideration, now agrees with me. Ha!

But back to LQG and relationism.

§ §

The loops in Loop Quantum Gravity are connections between minimum-sized volumes of… well, not volumes of space, because the whole point is this theory defines space, so… volumes of volume, I guess.

Smolin says the connections, the relations, are the main ontology, but his own text contradicts that. Like, constantly. Every example he gives of relations has objects that anchor those relations.

(It reminds me of how when strict determinists say free will doesn’t exist they can’t avoid constantly using words like “choose”. Some say this is because language is inadequate. Perhaps, but it can also, I think, speak to incoherency in the idea.)

Maybe it’s a limit on my part, but I’m not sympathetic to relational theories . They always seem to imply objects for the relations to be between. I’m not sure relations as the only, or even primary, ontology is a coherent notion. I think relations are always secondary and contingent on (at least) two objects.


Look at it this way: Does the idea of a relation between nothing make sense? Just a relation as a primary ontological object.

The general notion of a relation has a certain Platonic weight, but notice what the notion entails: “A relation is a set of properties describing a link between…” Oops.

A link between what? The definition can’t continue until we supply objects for a relation to be between.

A naked relation doesn’t seem a complete notion. It needs something to complete it. Here’s a simple example:


Without an x and a y, the relation is incomplete.


On the other hand, the idea of two objects with no relation makes perfect sense. Ontologies begin with things. Once you have things, you can have relations between them.


Now that makes sense. It’s not true, but it makes sense. It can be parsed, understood, and processed.

Bottom line, I just don’t get relationism. To me it’s cart before horse.

§ §

As an aside, Smolin suggests that string theory might be a background dependent approximation where LQG is the underlying theory providing that background. According to him, the strings would ride on the spin network (the “loops” in LQG).

He also compares string theory with the simplified Newtonian view (which was also background dependent). That Newtonian view was necessary for progress in understanding motion. Einstein later updated it, but Newton still works great for lots of stuff. We sent New Horizons to Pluto with it.

Stay looped, my friends! Go forth and spread beauty and light.

About Wyrd Smythe

The canonical fool on the hill watching the sunset and the rotation of the planet and thinking what he imagines are large thoughts. View all posts by Wyrd Smythe

7 responses to “BB #78: Relational Theories

  • Wyrd Smythe

    I should perhaps be clear that I don’t have a problem with ontologies that make relations important or even the only thing that really matters (although I don’t entirely agree with that view). I’d even go along with making relations fundamental.

    But they’re not primary in my view. Thing are primary.

  • Lee Roetcisoender

    If you ever do get into Kant and actually see what Kant saw, I think you will find that his ontology is essentially a reality/appearance metaphysics similar to RAM. It places our physical universe and every “thing” that is in that universe into the ontological category of Appearances (phenomena) whereas the “thing-in-itself” is the actual Reality (noumena)…….

    Caution: getting into Kant will require some serious personal capital.

    • Wyrd Smythe

      I’ve waded in enough times to know how deep those waters run. (Apparently it’s worse in the original German due to Kant’s long long sentences and how German piles the verbs on at the end. Just parsing the syntax is a chore, let alone parsing the semantics.)

      I’ve noticed the similarity of the RAM to Kant’s transcendental idealism since you started talking about it. And while I’m not a fan of philosophical idealism, I don’t see his TI as anything like that, but merely a recognition of what we now take for granted: the only reality we know is the model in our heads. The difference might be that Kant was ultimately a realist — noumena are real even if we can never truly access them.

      Smolin’s approach, especially in Time Reborn, also has some flavor of the RAM (although none of the MOQ). You might be fascinated by his static-dynamic dualism. Reminded me a bit of what we discussed recently.

  • SelfAwarePatterns

    When thinking about relational ontologies, a question to ask about any specific thing, is what is that thing composed of? In most cases, it seems like they’re composed of component things, relations, and processes. But then, what are those things composed of?

    Eventually we get to elementary particles, which turn out to be field excitations (relations and processes). Of course, a field could be said to be composed of value points in a matrix, which could be considered things, I guess. But maybe only until we find out what they’re made of?

    The turtles await. 🥴

    • Wyrd Smythe

      Yeah, it’s the age-old game of “What’s at the root of things?” For me, the question I posed about the sense of a naked relation verses two unrelated objects is the razor.

      The fields of QFT are themselves objects, and I think we can view disturbances in that field as (sub-)objects. Certainly they act like objects — “particles” — in terms of our experience of them. If QFT is a correct theory (rather than yet another approximation), then we would probably have to accept the fields and how they behave as ontological primitives. If Smolin is right, there’s an even deeper theory, but we’re always faced with accepting something as axiomatic.

      I’ve heard turtle soup is delicious. 😉

And what do you think?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: