Commoner’s Four Laws of Ecology
An organization is a connected system: changes propagate beyond where they start, costs relocate more often than they disappear, and apparent free gains usually have an unseen payer.
What Is It?
Biologist Barry Commoner proposed four informal laws of ecology in his 1971 book The Closing Circle, meant to summarize how ecosystems actually behave rather than to function as physical laws in the strict sense:
- Everything is connected to everything else.
- Everything must go somewhere.
- Nature knows best.
- There is no such thing as a free lunch.
The first three are really different expressions of the same systems principle. Changes propagate beyond where they’re made, so a fix applied for a good local reason still ripples outward through connections nobody was tracking. Costs move rather than disappear, a risk relocated, a problem handed to whoever’s downstream, rather than actually resolved. And apparent free gains are usually paid for somewhere else in the network, even when the accounting doesn’t show where. The organizational habit this corrects is treating a system boundary, a department, a budget line, a team’s scope, as if it were also a causal boundary. It usually isn’t. The org chart says where responsibility sits. It says nothing about where an effect actually travels once it leaves the place it started.
Commoner’s fourth law, “nature knows best,” is intentionally treated differently here. His original claim was that an ecosystem shaped by a very long history of trial and error already encodes a kind of tested fit that a deliberate redesign, working from incomplete information, is unlikely to improve on without real understanding of why the system is the way it is. That’s a genuine insight, but it’s also more context-dependent than the first three, and it’s already covered more carefully elsewhere in this Field Guide, by Chesterton’s Fence and Orgel’s Rule specifically. Treating “the old way probably knows something you don’t” as a co-equal fourth law overstates how well it generalizes on its own.
Why Does It Matter?
Organizations routinely act as though a change can be evaluated by looking only at where it’s aimed, and the cost of forgetting the boundary between a team’s scope and a system’s actual reach shows up everywhere. A process fix in one department that quietly increases the workload of another isn’t a side effect, it’s the first law showing up on schedule. A team that “eliminates” a cost by pushing it into a different budget line, a different quarter, or a different team’s queue hasn’t actually removed it, it’s just moved somewhere the org chart makes harder to see. And a win that looks free, a shortcut that saves time with no apparent cost, is usually a win whose cost hasn’t been billed yet, technical debt, a skipped review, a corner cut that a future team or a future version of the same team will eventually pay for. None of this requires anyone to be careless. It’s what a connected system does by default, whether or not the people inside it are tracking the connections. Systems don’t care whether your reporting structure notices the dependency.
This is also why local improvement can coexist with global decline: the system can get better in one place and worse overall, because the improvement and the damage it causes occur in different places, measured by different people, often on different timelines.
What Changes Once You See It?
You start asking where a change’s effects land outside the decision-maker’s boundary of visibility, not just whether it worked for the team that made it.
You start treating a cost that seems to have disappeared as a cost that’s moved, not a cost that’s gone, and you go looking for where it landed before believing it’s actually resolved.
You get more skeptical of any win that looks free. Somewhere in a connected system, it usually isn’t, the question worth asking is who’s paying for it and whether they know yet.
Common Misunderstandings
- It isn’t a claim that every consequence is knowable in advance. Connections in a real organization are often invisible until something crosses them, the point is expecting them to exist, not being able to trace all of them ahead of time.
- The fourth law, “nature knows best,” isn’t included here as a standalone principle worth applying on its own. It’s a genuine insight about inherited systems, but this Field Guide already covers it more carefully under Chesterton’s Fence and Orgel’s Rule, and treating it as a fourth co-equal law overstates how well it generalizes.
- It isn’t an argument against making changes, or evidence that every change has a hidden cost large enough to matter. Some effects genuinely are contained, and some apparent free lunches genuinely are free. The claim is about where to look before assuming that’s true, not a verdict that it’s never true.
- It isn’t primarily about incentives or metrics. Goodhart’s Law and the Cobra Effect explain how optimization distorts behavior once a specific number is targeted. Commoner’s laws describe what happens because the system itself is interconnected, independent of whether anyone is optimizing for anything.
Diagnostic Question
If this change succeeds exactly as planned, who pays for the success?
Explore Further
Field Notes
None yet.
Related Field Guide
Origin
Barry Commoner, The Closing Circle: Nature, Man, and Technology (1971); borrowed here as a structural pattern for systemic interconnection and hidden cost, not as literal ecological law.