Why Networks Go to War

You've probably watched it happen to someone you know. Maybe it happened to you.

A friend joins a new group — a movement, a community, a cause. At first, nothing changes. They're still the same person. They still show up to poker night. They still text you about the game.

Then, gradually, they start pulling away. The group takes up more of their time. The language shifts — new vocabulary, new ways of framing old things. And at some point, you get the sense — spoken or not — that your friendship is becoming a problem. Not because of anything you did. But because you're connected to people and ideas that their new group considers the other side.

It feels personal. It feels like a choice they made. But what if the same thing happens in ant colonies, immune systems, and fungal networks underground — systems that have no opinions, no agendas, and no feelings at all?

To understand why, we need a few concepts from network science and biology. Don't worry — we'll build them as we go, and by the end you'll have vocabulary for something you've always sensed but couldn't quite name.

Boundary Maintenance: The First Law of Networks

Let's start with something that has zero political baggage: your body.

Your immune system is a network. White blood cells, antibodies, lymph nodes — all connected, all talking to each other, all doing one job: figuring out what's you and what's not you, then attacking the not-you. Biologists call this boundary maintenance — the process by which a system defines and defends the line between inside and outside.

Your immune system doesn't care if the foreign thing is helpful. Organ transplants? Your body attacks them unless doctors basically roofie your immune system into compliance. Beneficial bacteria? Attacked until the system grudgingly learns to tolerate them, like a dog that eventually stops barking at the mailman. The default is simple: if it's not part of us, it's a threat.

This isn't smart. It's not strategic. It's just what networks do to stay intact. A network that doesn't maintain its boundaries eventually stops being a distinct network — it dissolves into the environment around it. Boundary maintenance is the price of existence.

You can see it everywhere in nature. Ant colonies identify members through chemical signatures — hydrocarbon profiles on each ant's exoskeleton that function like a chemical ID badge. Biologists call these identity markers. An ant from Colony A wanders into Colony B's territory? Dead. Not because of any grudge. Because its identity marker doesn't match. The label is wrong, and the network responds to labels, not to individuals.

Underground fungal networks — the root systems connecting trees in a forest — compete for territory. When two rival fungal networks meet, they don't merge. They form what mycologists call a zone of inhibition — a dead strip where neither network grows. Think of it as a no-man's land. Neither side will cross it. Better to waste the space than risk blending with the competition.

Cells, ants, fungi. No brains. No ideology. No podcasts about their journey. And they all do the same thing: form a group, draw a line, defend the line. That pattern has a name: boundary maintenance. Remember it. You're going to see it everywhere once you know what to look for.

Competitive Exclusion: Why Groups Fight Over People

Boundary maintenance explains why groups defend their borders. But it doesn't explain why groups go on the offensive — why they actively pressure members to cut ties with other groups. For that, we need a concept from ecology called competitive exclusion.

The principle is straightforward: two species that need the exact same resources in the exact same environment can't coexist indefinitely. One will eventually edge the other out. Not through malice. Through math. There's only so much food, water, and territory. If two species are both trying to eat the same berries from the same bush, one of them is going to go hungry.

In social networks, the scarce resource isn't berries. It's you.

Your time, attention, loyalty, and emotional energy are finite. There are 24 hours in a day, and you're already sleeping through a third of them (or lying awake worrying through a third of them — either way, those hours are gone). You can only maintain so many real relationships — anthropologist Robin Dunbar put the number at roughly 150 meaningful connections, with an inner circle of about 15 people you're genuinely close to. That's your social bandwidth. And every group you belong to is competing for a share of it.

Your bowling league competes with your church group for your Tuesday nights. Your work friends compete with your college friends for your weekends. Your marriage competes with your buddy group for your emotional bandwidth. (And if your wife and your best friend both need you on the same Saturday, you already know which one has veto power.) Most of the time, this competition is gentle. There's enough of you to go around — barely.

But when two groups occupy the same space — when they're both trying to define who you are, what you value, and how you see the world — the competition sharpens. Because now it's not about your Tuesday night. It's about your identity. And identity is the one thing you can't split between two groups that disagree on what it should be.

That's competitive exclusion at the social level. Two networks claiming the same niche — the same slice of who you are — can't coexist peacefully forever. The pressure to choose isn't coming from any individual person's meanness. It's coming from the structure of the system itself.

The Bridge Problem: Why the Middle Gets Crushed

In earlier articles, we talked about bridge nodes — people who connect otherwise separate clusters in a network. Your buddy who's friends with both your work crew and your gym crew. Your cousin who fits in at Thanksgiving despite having wildly different politics from the rest of the family.

Bridges are enormously valuable to the overall system. They transmit information between groups. They enable cooperation. They prevent the kind of total isolation that turns disagreement into dehumanization. Society runs better with more bridges, not fewer.

But here's the tension: bridges are expensive for the individual groups they connect.

Think about it from the group's perspective. Everyone's on the same page. Conversations flow easily. There's trust, shorthand, shared assumptions. Now one member is also deeply embedded in a group that sees things differently. That member brings in outside ideas. Asks uncomfortable questions. Sometimes defends the other side's perspective.

That person is introducing complexity into a system that runs most smoothly when it's simple. Every bridging connection costs the group energy to manage. It's friction.

Physics has a concept called minimum energy — systems naturally settle into states that require the least energy to maintain. A ball rolls downhill. Water finds the lowest point. And groups, left to their own devices, drift toward simplicity. Fewer contested connections. Fewer complicated members. Less friction.

"If you're still hanging out with those people, can you really be one of us?"

That's the sound of a network moving toward its minimum energy state by pressuring a bridge to break. It works — not because it's fair, but because the person in the middle eventually gets tired of being pulled in two directions and lets one side go. Network scientists call this bridge elimination, and it's one of the primary mechanisms by which competing networks become more separate over time.

The Evolutionary Backdrop: Roles That Didn't Compete

For most of human history, the competition between networks was manageable because the major networks in a person's life occupied different lanes.

Go back far enough and the division is basic. In small human groups — and we lived in small groups for 95% of our existence — roles sorted along biological lines. Not because somebody wrote a rule book. Because differential advantages made certain divisions effective.

Individuals with greater physical strength and higher risk tolerance — overwhelmingly male — gravitated toward hunting, defense, and exploration. Individuals with advantages in social bonding, long-term planning, and nurturing — overwhelmingly female — gravitated toward caregiving, resource management, and maintaining the social fabric that held the group together between hunts.

These weren't rigid rules. They were probability distributions — tendencies, not absolutes. Most men hunted; some didn't. Most women nurtured; some also hunted. (There's always been a Karen who could throw a spear better than half the guys. Every tribe had one.) But the patterns were consistent enough, across enough cultures and millennia, that they calcified into what we now call gender roles — which, in the language we developed in article six, are class definitions inherited so deeply that most people experience them as natural law rather than social convention.

Here's what matters for our discussion: these role-based networks were complementary, not competitive. In ecology, we'd say they occupied different niches. The hunting network didn't compete with the nurturing network for the same resources. They occupied different spaces in the same system. The hunter needed the planner. The protector needed the healer. The boundaries between them were porous on purpose, because the whole thing only worked when they connected.

Complementary networks don't trigger competitive exclusion. They don't demand exclusive loyalty. They don't pressure members to cut ties with the other side. Because there's no "other side" — there's just a different part of the same machine.

That arrangement worked for a very long time. It's the water most human civilizations have been swimming in.

Niche Overlap: When Complementary Becomes Competitive

Now fast-forward to the world you actually live in.

Physical strength isn't the primary driver of economic value anymore. Information is. Communication is. Cognitive skills — distributed more evenly across biological lines — are what most jobs actually require. The niches that used to be separate have begun to overlap.

When previously complementary networks start occupying the same space, the ecological dynamics shift. Networks that could coexist peacefully when they needed different things start competing when they need the same things. That's not a moral judgment — it's the competitive exclusion principle doing what it does whenever two systems converge on the same niche.

At the same time — and this is where things accelerate — new identity labels are being created at a pace with no historical precedent.

Every new label is a new class definition (article six). And every new class definition creates a new network — a new cluster of people who share that label, form connections around it, develop shared vocabulary, and begin to exhibit boundary maintenance. These new networks need members. They need nodes. And they recruit from the same population that existing networks already claim.

So now you've got the legacy networks — family, community, church, profession, gender role — and the emerging networks, built around new identity categories and new ways of defining who's in and who's out. And they're all drawing from the same finite pool of human attention and loyalty.

The competitive pressure is automatic. A node — a person — who belongs to both Network Old and Network New is a divided resource from each network's perspective. Network New, like every network before it from fungal mats to ant colonies, benefits when its members are fully committed. Your old connections represent what an economist might call resource leakage — energy flowing to a competing system.

And so the pressure comes: maybe those old connections aren't just different. Maybe they're harmful. Maybe the people in your legacy network hold views that are incompatible with who your new network says you should be. Maybe you need to distance yourself. For your own good. For the good of the group.

That's not manipulation (or it's not only manipulation). It's boundary maintenance and competitive exclusion, operating exactly as they do in every networked system in nature. The difference is that it's happening to you, in your actual life, with your actual relationships.

The Human Twist: Doing It on Purpose

Everything we've described so far happens automatically. Immune cells don't plan their attacks. Ants don't hold strategy meetings. Fungal networks don't write manifestos. The competitive behaviors just emerge from simple rules: protect the boundary, compete for resources, simplify when possible.

But humans add a layer that no other network has: intentionality.

We can observe the patterns. And once we can see them, we can use them.

A leader who understands — even intuitively — that groups get stronger when members are fully committed can design loyalty tests. "If you really believe in X, you must reject Y." That's not a natural emergence. That's someone recognizing the boundary maintenance instinct and engineering a trigger for it. It's the social equivalent of knowing which button on the remote starts the TV — you didn't build the electronics, but you figured out how to work them.

A movement that understands the bridge problem can target bridges strategically. Not by attacking the person directly, but by reframing their bridge status as a moral failing. "You can't claim to support us while staying friends with people who think that way." That's the ant colony's chemical warfare, translated into language and deployed with precision.

New vocabulary gets introduced — labels that create new class definitions, new identity markers, new in-group/out-group boundaries. Old vocabulary gets redefined or stigmatized — disrupting legacy class definitions and the networks built around them. Each linguistic shift is a small rewiring of the network graph. New edges form. Old edges dissolve. Clusters tighten. Bridges break.

None of this requires a conspiracy. (Conspiracies require competence, and have you met people?) Some of it is deliberate strategy by people who understand network dynamics. Some is semi-conscious pattern-matching by people who've learned what works through trial and error. And some is just the natural behavior we've been describing, moving faster and more visibly because intelligent nodes amplify the signal.

The key insight: the natural forces and the deliberate tactics are the same mechanisms operating at different levels of awareness. Boundary maintenance happens automatically in ant colonies and intentionally in political movements. Competitive exclusion happens passively in ecosystems and actively in culture wars. The physics doesn't change. What changes is that humans can see the physics and steer it.

Network Dominance: What Winning Looks Like

There's one more concept worth learning: network dominance. In network science, a dominant network is one that has successfully captured enough nodes — and eliminated enough bridges to rival networks — that it controls the flow of information, resources, and identity within its territory.

Dominant networks don't have to fight anymore. They've won the competitive exclusion game. Their class definitions become the default. Their vocabulary becomes the standard language. Their boundaries become the boundaries of acceptable thought. Rival networks still exist, but they're marginalized — pushed to the periphery where they lack the connections to influence the center.

If this sounds familiar, it should. It's what every dominant network in history has looked like — from the Catholic Church in medieval Europe to British colonial networks across the globe to corporate monocultures in industry. Dominance isn't a modern invention. It's the natural end state of unchecked competitive exclusion.

What's different today is the speed. Digital communication lets networks recruit, enforce boundaries, eliminate bridges, and establish dominance at a pace that would have taken previous networks generations. A movement that would have needed decades to achieve cultural influence can now reach critical mass in months. Martin Luther had to nail his theses to a door and wait for the printing press to do its thing. Today he'd just need a TikTok account and a grievance. The physics hasn't changed. The clock speed has.

What This Gives You

If you've read this far, you might be waiting for the verdict. Which networks are the good ones? Which loyalty demands are legitimate? Which bridges should you keep and which should you burn?

That's not what this article is for.

What it's for is giving you vocabulary — the real terms, from real science, for dynamics you've been feeling but couldn't articulate. Let's review what you now have:

• Boundary maintenance — the process by which any network defends the line between inside and outside. Natural, universal, and morally neutral. It's not good or bad. It's structural.
• Competitive exclusion — the principle that two networks occupying the same niche can't coexist indefinitely. Explains why groups that compete for your identity eventually force a choice.
• Bridge nodes — people connected to multiple, potentially rival networks. Valuable to the system, costly to individual groups, and always the first to feel the pressure.
• Minimum energy — systems drift toward states that are easiest to maintain. Groups simplify by eliminating complexity, which often means eliminating bridges.
• Network dominance — what happens when one network wins the exclusion game and its definitions become the default.
• Intentionality — the uniquely human ability to see these natural patterns and deploy them strategically through language, loyalty tests, and identity engineering.

These aren't opinions. They're observations about how networked systems behave, confirmed in biology, ecology, physics, and social science. Every side of every culture war is subject to these forces. Every group you've ever belonged to exhibits these behaviors to some degree. Including the ones you agree with.

What you do with this vocabulary is up to you. But here's what it makes possible:

When a group asks you to cut ties, you can ask: is this because those ties are genuinely harmful to me, or because the network is doing what networks do — simplifying itself at my expense?

When a new label is offered to you, you can ask: does the class definition that comes with it serve me, or does it mostly serve the network that created it?

When you feel pressured to pick a side, you can ask: are these sides actually incompatible, or is the incompatibility being amplified because that's how competitive exclusion works?

You're not above the system. You're a node in it, subject to the same forces as every ant, every antibody, and every guy who ever got a guilt trip for missing book club. But you're a node with vocabulary — which means you can name what's happening to you instead of just being moved by it.

In a network full of nodes that are reacting on instinct, that's the closest thing to freedom you're going to get.