Chapter 20. Holding the whole without a centre

A working draft, November 2026. This book is being written while The Grove is being built; expect it to change. Checked against the build as it stood on 1 November 2026.

In this chapter

• Why source theory, the viable system model and PowerShift each say a centre has to hold the whole
• What a centre costs, and what its absence costs
• How purpose, the root, an open flame and climbing tensions hold the whole without a centre
• How shared money is weighed by a forum with evidence, and where the gaps still are
• What happens in an emergency with no override, and the fire alarm we've designed but not built

Here is the doubt at its strongest, and it doesn't come from people who like bosses.

Something has to sit at the top. Not a tyrant, but someone or something that holds the whole: that knows what the organisation is for, sees how the parts fit, and can act when they don't. Take that away and a self-managed organisation doesn't become free. It drifts. Every part looks after its own patch, and the gaps between them grow until a customer notices.

What makes this doubt hard to wave away is who holds it. Three serious traditions, each built by people who wanted organisations more alive, not less, agree on some version of it.

Source theory, which we met in Chapter 7, says every initiative has a source, the person who first took the risk of acting, and that the whole organisation has a recognised global source who listens for what it needs and holds its boundary. Tom Nixon and Stefan Merckelbach have written about what happens without one.¹ Merckelbach's word for it is the vacuum: everyone equal, nobody leading, nothing decided.

Stafford Beer's viable system model, worked out from the 1970s by studying how living systems stay alive, says any organisation that survives has five systems in it.² The operating parts; something that stops them colliding; something that manages the here and now and shares out resources; something that watches the outside world and the future; and something that holds identity and settles the argument between today and tomorrow. Beer called the third and fifth of these System 3 and System 5, and he was clear that a viable organisation can't do without them. Martin Pfiffner, who has spent a career putting the model to work, adds an important softening: these are functions to be performed, and they need not be posts with a person's name on them.³ But performed they must be.

And PowerShift, Thomas Thomison's recent work on organisations run alongside AI, goes furthest: a permanent human Orchestrator holds the whole, with a kill switch outside governance altogether.⁴

This chapter is where we disagree with all three, and we owe them a fair account of how, and what it costs.

What a centre costs, and what its absence costs

The vacuum is real. Anyone who has spent time in a flat collective has seen it. A decision everyone knows needs making sits for months because nobody feels it's theirs to raise. A part quietly stops working and nobody says so, because saying so would feel like an accusation. The whole has no voice, so the loudest part speaks for it.

But the centre has costs too, and they're the ones the frameworks spend less time on.

A centre is a funnel. Everything the whole needs to know has to reach one place, filtered on the way by people deciding what the centre wants to hear. The person in it carries far more than one person can, and the organisation learns to wait for them. And a centre, once it exists, is where power returns whenever things get hard. A kill switch outside governance is honest about this: it says that in the end, the decisions people make together are provisional, and one hand can undo them.

For the business, a centre means everything that matters moves at the speed of one inbox. For the people, it means the quiet lesson that the whole is somebody else's job.

So the question isn't whether the whole needs holding. It does. The question is whether holding it needs a place.

Held in the open

Pfiffner's point is the crack we went through. If identity and resource sharing are functions, not posts, then the question becomes how those functions get done, and who can see them being done.

Our answer is two things, neither of them a person. The whole can be held by everyone seeing it. And it can be held by what won't stay buried: problems that climb, on a clock, until someone answers them, and that say so out loud when nobody does.

Barbara Marx Hubbard drew something like this for a whole society. Her Wheel of Co-Creation pictured its many sectors as parts of one wheel rather than layers of a pyramid: none set above the rest, each doing its own work, and each finding where it met the others. She was describing a society. We found it described an organisation too.

What holds the whole

Purpose is what closes it. An organisation needs something that makes it this organisation and not a crowd. On The Grove that's its purpose, held at the root, the first seat. It's what Beer would call identity. Only the root's own forum can change it, and a change to it is put to every member for advice before anyone decides. A branch's accountabilities change in the forum around its parent; nobody below the root can change what the whole is for.

The root is one seat, and the end of every rise. This is the honest double answer Chapter 7 gave. In decisions, the root weighs the same as each of the branches just beneath it in its forum. In structure, it's where things end: work that has no home rests there, problems nobody has answered arrive there, and only its forum touches the purpose. So the root carries more than it looks. What it doesn't carry is a private view.

Everyone can see the whole. Since 1 November 2026, the organisation's flame, the single reading of how it's doing, opens for every member, not only those at the root. Press it and it unfolds into its four readings: whether what was promised and tried has landed; how far the work arriving sits from the purpose; how widely appreciation reaches; and the pulse of the Clearing, the organisation's shared space for how people are. Each is drawn over time and opens onto what feeds it, down to the piece of work or proposal behind it. Every branch has a flame of its own too, read over itself and everything beneath it, and shown beside its siblings, never ranked. People are never scored. The readings about people are collective, shown only where enough people hold that part that nobody can be picked out. Taking the whole seriously still sits most heavily at the root, but as a responsibility people carry, not information kept from everyone else.

Problems climb until they're answered. A tension is a sensed gap between how things are and how they could be. Some the work reveals by itself: work bouncing between branches, work done from the wrong place, work nobody holds. Others are left behind when a proposal lapses, is withdrawn, or its review says it didn't work. Since 1 November, any tension left unanswered on a forum for fourteen days rises to the next forum up, and fourteen days later to the next, until it reaches the root. When it gets there, it says plainly that the whole organisation hasn't answered it. It can be answered by a proposal raised from it, or in words by someone seated where it now sits, and those words stay beside it. Nothing removes it quietly.

Pain reaches upward. Beer gave the viable system an alarm he called the algedonic signal, from the Greek for pain and pleasure: a channel that lets distress in an operating part reach identity without passing through every layer of management. Ours is gentler. Work from outside that nobody has taken up for fourteen days on a branch shows in the morning heads-up of whoever holds the branch above, as work risen from below, with how much and how long the oldest has waited. It goes to the branch above, never to a named person, and it isn't pushed: it waits for when they come.

Resources are shared by a forum, not a post. Money stays in the organisation's own bank. What The Grove keeps is a mirror of it, and income that's recorded lands in a shared purse, the commons. Budgets come out of the commons only by proposal, heard in the root's forum like any change to the shared purse. Since 1 November a budget proposal carries its own evidence: what it's for, whether the organisation's read says that branch's work wants it to grow, its load of work over the last four weeks, and what's left of its budget. That's Beer's System 3 done by a forum with the evidence in front of it, rather than by someone with a spreadsheet. Spending within an agreed budget needs no further proposal. A record of money with its evidence counts on the word of whoever made it and is open to question by any member for a week: something like the audit Beer added to System 3, held by everyone rather than an auditor.

Mapped lightly, the rest follows. The branches are the operating parts. What stops them colliding is the routing of work to where it belongs, the boards each branch keeps, and the promises that cross between them. Looking outward and ahead is spread across the branches' bets, Chapter 14's vision held as hypotheses and tested against the world, and the Field's sensings. Identity is the purpose at the root.

The gaps are real and we'd rather name them. There's no single place that weighs the outside world and the future against today for the whole organisation; that function is spread thin across many bets, and nothing yet asks which part of the organisation is thin. Whether a tension that reaches the root should ever fade there, if the whole decides to live with it, is open. And a rise of one forum every fourteen days is slow beside Beer's alarm, which was meant to jump the layers. A tension starting three forums below the root takes six weeks to arrive.

When the building is on fire

There is no override anywhere. No veto, no emergency path, no switch that one person can pull. The people who keep the software running can change the software, as with any software, but nothing in it lets them, or anyone, step into an organisation's decisions.

That has a cost, and it's sharpest in an emergency, so here is what happens in one today.

Anyone can act on work at once. A piece of work can be marked as needing someone now. A seat holder can spend from their branch's budget and record the payment, with its receipt, on their own word. People can reply, take things up and hand them on without asking anyone. What can't move fast is structure and the shared purse. A new branch, a changed seat, money from the commons: each is a proposal on the clock Chapter 11 described. Advice runs at least three days, unless everyone asked answers sooner. The vote stays open for two days unless everyone deciding has consented sooner. A high-impact change waits a further three days, and a concern raised then stops it. In a real crisis, people will act in the world first and tell the organisation afterwards. That's what happens now.

We've designed a way through this and haven't built it. We call it the fire alarm. Someone pulls it, naming the condition, the part it covers, and who takes command. For a few hours that person acts alone within that scope. It can never change the purpose, move money to a person, take away anyone's right to leave, rewrite how decisions are made, or extend itself. It expires on its own, and calling it off is easier than any ordinary decision, because the danger is a fake emergency. When it ends, the commander accounts for everything they did and the forum takes each act as agreed or not. Until it's built, the honest answer to "what if the building's on fire?" is that the people in it act, and The Grove catches up.

The founder alone

One case deserves plainness. A founder alone holds the root, decides alone, and is, for now, the whole. The same is true while others have joined but nobody else holds a seat: decisions route up to the root, and the root is one person. This is exactly where a centre grows back by default.

What pushes against it is that the flame, the tensions, the risen work and the budgets are open to whoever joins, from the first day. The rules about who decides start to bite the moment a second person takes a seat in a forum. Until then, the founder holds the whole the way source theory says they would. The difference is that nothing in the design keeps it that way.

How we decided

We started closer to the frameworks than we ended. Early on, the readings of the whole were the root's to see and act on. We came to see that as a centre by the back door. In November, outside voices pressed on two points: that a lapsed proposal could quietly bury a real problem, and that a reading only the root could see made the root a funnel. So we opened the flame to every member and made unanswered tensions climb.

We weighed a permanent orchestrating role, as PowerShift keeps, and a recognised global source. We kept neither. Both put the whole in a person, and a person is where power returns. We weighed a kill switch outside governance and refused it for the same reason we refused an expert veto in Chapter 11: an organisation that can always be stopped from outside never quite owns its decisions.

We weighed making tensions jump straight to the root, as Beer's alarm does. We chose one forum at a time, so the nearest forum always has the first chance to answer, at the cost of speed. That choice is the one we're least sure of. Whether fourteen days is the right stretch, whether people will treat a tension arriving at the root as a call or as noise, and whether everyone seeing the whole means everyone holding it, or no one: these we'll learn from real organisations, not from argument.

What you might take from this

The frameworks are right that the whole needs holding, and right that its absence is a vacuum. Where we part from them is on whether holding needs a place. Make the whole visible to everyone, let what's unanswered climb until it's answered, and have a forum, not a post, share out what's held in common.

The whole can be held by everyone seeing it, and by what won't stay buried.

Notes and further reading

On distributing leadership in practice: Laloux, F. (2014), Reinventing Organizations, Nelson Parker. On co-creation, and a society pictured as a wheel rather than a pyramid: Hubbard, B. M. (1998), Conscious Evolution: Awakening the Power of Our Social Potential, New World Library.