Why does removing part of a living thing make it grow?

The Finnish forest

I am half Finnish, and I have seen how forests are managed. The Finns go into a young patch of trees and cut some of them out on purpose, while the stand is still growing. It looks like loss. You walk through afterward and it is barer, fewer trunks, more sky. There were always enough trees. There was never enough room and light to grow all of them at once. The same holds for pruning a tree.

When you cut the lowest branches off a tree, it does not become a smaller tree. It actually redistributes its energy. The trunk below the cut puts on less girth, the leader extends and the crown pushes up. Foresters do this on purpose, to get a tall straight knot-free log. You take wood away and you get more tree.

There is a limit though. Take more than about a third of the live crown and growth drops, because now you have cut the foliage that was feeding the tree. The difference is which branches you took, not how much.

A branch is a tax

Why would removing part of a living thing make it grow? Everywhere else, growth comes from adding. More input, more output.

A low branch is not free. It is held in place, fed water and sugar, maintained against rot, every day it exists. The tree pays for it out of a fixed finite budget. A branch deep in shade photosynthesises almost nothing. Some take more than they return. The tree is paying to keep a thing that has stopped paying it back.

So the low branches are a tax. Removing a shaded one does not add energy. It stops a leak. The budget that maintained it is now free, and the tree spends it on reaching. The growth after pruning is the freed upkeep, respent.

The tree was never short of light or water at the top. It was just short of budget, because the budget was committed to holding branches that had stopped earning their hold.

Which is why the third-of-the-crown rule is the wrong way to think about it. A shaded branch below break-even is nearly free to cut. A vigorous well-lit branch is a supplier, feeding the budget the rest of the tree spends. The question before any cut is not how much. It is whether this branch is a drain or a source. Lowest and least-productive usually travel together, which is why the rule works most of the time. Where they come apart is where pruning backfires.

The second condition

There is a second condition, easy to miss because in a healthy tree it is already there. Freed budget only becomes height if there is somewhere for it to go. A tree with a strong leader has an open channel and the budget flows up into it. A tree whose leader has also stalled throws up water sprouts instead, fast vertical regrowth low on the trunk, just as shaded and useless as what you cut. The tree replaces the tax with a new tax.

So pruning is two moves. Remove the drain, and have an open channel for the release. One without the other does nothing.

The same shape elsewhere

The shape turns up everywhere.

A research practice fills with half-started threads. None of them dead, each costing attention to keep alive. When progress stalls the instinct is to add hours. But the constraint is usually the attention spread across a dozen shaded threads. The move is to cut the ones that have stopped feeding anything and keep one question reaching, so the freed attention lands somewhere. Cut without that one live question and you start three new threads by Friday.

An organisation carries the same load. A weekly meeting that exists because something broke in 2019. A report nobody reads that still takes a day a month. Held in place, paid for daily, returning little. Cutting them frees real capacity, but only if a live initiative is ready to absorb it. Cut with nowhere for the capacity to go and you get reorganisation churn, new process bred to replace the old.

Then there is treatment. Some therapies work by removal. They take something away and the body does better. This only makes sense under the tree's logic. Removal helps when the limit is a locus draining the budget, not a missing input. Take away an exhausted, suppressive state the system has been holding, and it can respend that budget on the work that matters.

The tree's two conditions predict how such a therapy fails. Remove too much, past the parts that were supplying the system rather than draining it, and you crash it instead of redirecting it. That is the vigorous branch.

Remove the draining state but leave no open channel, and you get disordered rebound. That is the water sprout. Whether removal helps or harms is governed by what you took and whether a channel was open, more than by how much.

The sign of an intervention

Interventions have a sign. One kind adds, supplying a missing input or opening a channel. The other kind subtracts, removing a locus that drains the budget so the system can respend it. Most effort to fix things assumes the first. Something is wrong, so add. When the constraint really is a missing input, that is right. When the constraint is a budget committed to holding what no longer earns, adding makes it worse, because the budget is already spoken for.

The whole thing reduces back to one simple question:

Is the system short of input, or taxed by what it holds?

Get it right and the sign of the intervention follows. Get it wrong and you feed what needed cutting, or cut what needed feeding.

Hold and reach

Underneath that question is a balance the tree is paying for the whole time it is alive. The budget splits two ways. Some goes to holding what has been built, some goes to reaching past the edge. Hold deep and you have a lot of secure structure and a thin margin to reach with. Reach wide and you are taking ground fast with little spare to hold what you have taken. It is one pool. Width and depth trade against each other directly.

The split is never set once. It drifts, because what counts as productive hold keeps changing as the tree runs. A branch that was a well-lit earner ten years ago is a shaded drain now. Nobody moved it below break-even. The canopy grew over it. The budget stayed committed to it as if it were still paying. So the tree is always slightly out of balance, carrying depth that has quietly slipped under earning while the world it was responding to has moved on.

That is why pruning is not a one-time correction. It is rediscovering, again and again, that depth has crept up on width, that budget is held in structure the world has since shaded out. The cut shifts the balance back toward reaching.

And it has a twin, the move nobody calls by the same name. A system that has reached too wide, holding too little, thin over a base that cannot survive a shock, needs the opposite. It needs to stop reaching and pour budget back into holding what it grabbed. Same axis, other direction.

Over-held systems need cutting. Over-reached systems need consolidating. Both correct the same drift, read from opposite sides, and the skill is telling which side you are on before you act. Cutting a system that was already too thin finishes it. Consolidating one that was already overgrown just feeds the drain.

There is no stable answer to settle on, no ratio you set and leave, because holding changes the world that holding was answering. The tree that keeps its low branches shades them, which turns them from earners into drains, which means the hold that was right created the condition that makes it wrong. Reaching does the same in reverse. The system's own running keeps invalidating its own split. The balance has to be found constantly, not because we are bad at setting it, but because the act of running moves it.

The tree is not deciding any of this. It holds, it reaches, and when the balance has drifted it carries the cost until something cuts or something consolidates.

One piece of something larger

This is one piece of something larger. The hold-and-reach balance, and the way it drifts as a system runs, is the working core of Persistence Dynamics.