Autonomy is not independence: cultivating the strategic political work of supporters

Here’s a session I run with organizing teams. It starts with everybody standing up.

The setup is simple. A volunteer leader, somebody you developed and somebody you trust, makes a call without checking with you first. If you would be comfortable with that, step forward. If you would want to be consulted, stay where you are. There are no right answers here. I only want to see where the instincts sit.

Then I read the situations out slowly, one at a time, leaving a few seconds between them for people to move. Examples: She decides what time the event starts. She changes how she opens a conversation on a doorstep, because she thinks something else will land better in her neighbourhood. She extends the canvass into a block that was not in the plan. She gets a hostile response at a door, walks away, and does not record the contact. She recruits somebody new to the team without running it past anyone. She decides her team is not ready and cancels the action. A local journalist stops her in the street and she speaks to him, briefly, positively, on message.

The room moves together on the first two. Everybody steps forward, and somebody usually laughs. Then it starts to come apart, generally around the third, and by the sixth people are scattered across the floor at different distances looking at each other.

Cancelling is the one that surprises people. It is the most obviously legitimate leadership call on the list. She is closest to the team, she can see they are not ready, and running the action anyway would be worse for everybody. It still feels like a breach.

What I ask afterwards is not whether they were right to hesitate. It is what the hesitation was actually about. Is it that you do not trust her? Is it the risk to the organization? Or is it that nobody ever told her what she was allowed to decide, so you do not know either?

It is almost always the third. 

A large part of why we do not build autonomy into our structure is that the word autonomy feels wrong.

Say autonomy in a room of organizers and they picture independence. Separation. People off doing whatever they like with the organization's name attached to it while they are unreachable and accountable to nobody. Put that way, of course nobody wants to hand it over. I wouldn’t either.

Almost nobody takes a job in a progressive organization for the money. They take it because they are disturbed by power in the way it works in our world. How it is held, who it is held over, and what it does to the people at the wrong end of it. That instinct is correct, and it is most of the reason the progressive end of our sector exists at all - to fight injustices. It also produces a belief that turns up everywhere: that progress only happens in perfectly flat structures, where every voice carries identical weight in every decision, because fairness and democracy are the things we are fighting for.

Fighting injustice, though, is not the same as refusing to hold power or acknowledge power at all.

Pavini Moray gives this failure a name in How to Hold Power. They call it flattening. This is the well-meant move of treating power as a fixed quantity to be divided into equal parts. What happens is you defer to a weaker idea in order to seem fair, or insist that everybody here is a peer while you go on holding the actual authority. Their point is that the power does not disappear when you decline to name it. It starts operating dishonestly instead, and the differential goes on shaping outcomes whether anybody acknowledges it or not. They are direct about who does this most: people who are allergic to hierarchy.

Their starting position is that power is neutral. What matters is the kind you exercise: power over people, power with them, or the power within yourself that makes it possible to hold either without harm.

It is worth remembering what we are doing here. We are not only fighting power. We are building it, which is the entire point of organizing. If the power we build is going to be worth anything, somebody has to hold it, and holding it well is a skill rather than a betrayal.

Ignoring or refusing to hold power in your relationship to supporters, is not what autonomy means.

Autonomy is having a decision space — a well defined area inside which you do your own thinking, work the problem out yourself, and are trusted to decide. It is not freedom from the collective. It is a defined piece of the collective's work that is actually the supporter’s. (I first came across the term in a course I took at Stanford Extension, taught by Melissa Valentine, and it reorganized how I think about this - more on that below). Once you have the phrase, you notice that there are three things in play and that our sector routinely names only two of them.

The first is the collective agreement. What we are all doing, what we stand for, and what nobody does under any circumstances. That gets defined centrally, as it should be.

The second is the decision space: what is deliberately not being decided centrally. The set of questions we are leaving to the person closest to them, marked out on purpose rather than left over. That also has to be defined centrally and purposefully. This is where the step forward exercise can be helpful.

The third is autonomy, which is less a thing an organization grants than what happens in the decision space once somebody is standing in it. Her judgement. Her read of her own place. Her call.

Leave out the middle one and you get what most programs have. An agreement everybody has signed up to, no stated space, and the gap between them managed by the absence of definitions. In this somebody follows up, somebody checks, somebody reminds. That is control, and it costs exactly the capacity that nobody has.

I want to be careful not to make this sound like alignment is the compromise and autonomy is the good part, because it is closer to the other way around.

Organizing gets its power from the level of coordination. That is the equation underneath all of it, and it is why the workers standing outside the factory have power: they linked arms. The same number of people, each deciding privately that their working conditions are sub par does not create power. Talking and organizing together surrounding that consensus does. The coordination is the power. (See more about this in my post “How do we measure organizing?”) 

If you’re an organization insisting on a collective agreement, you’re right to insist. As you might be picturing a hundred local groups heading in a hundred directions. That is a surefire way to lose.

But coordination is not control. Coordination is a shared understanding of what we are doing together. Control is somebody checking that you did it. (Consent is a key. And inspiration to a shared higher purpose through transformational leadership is the magic sauce.) 

Now back to my former professor Melissa Valentine, and her book “Flash Teams” - which studies new opportunities for working as teams through our new AI technologies. Valentine spent years studying emergency rooms, and what she found there is the cleanest demonstration of the difference I’ve seen.

A patient arrives in crisis. A group comes running. One person secures the airway. A second starts chest compressions. A third runs an IV. A fourth stands at the head of the bed and calls out the plan to move the patient to the hospital bed. In this coordinated effort there is no time to ask who is doing what. Once the patient is stable, communication happens. And builds trust for when the next patient walks through the door. 

In many cases, this emergency response is coordinated between people who have never met. They work together, and then they exchange names.

In Valentine’s research, before they even start talking, they know who is responsible for what work, who has the authority to make which decisions, and who hands off work to whom. She calls it role clarity, and she found the same thing in film crews and disaster response teams — strangers who assemble and coordinate immediately.

That room is more tightly coordinated than any campaign I have ever worked on. And nobody on the response team is being supervised at the moment.

Here is the sentence from her book that I think really is key for those of us who work in political organizing:

Structuring the roles does not mean structuring the tasks. Almost the opposite: the point is that, if someone understands their role, they can take direct ownership of coming up with the tasks that they need to complete. In fact, you don't need to micromanage someone who understands their role and their relationship to other roles.

It is our problem described by someone who doesn’t work in politics. Look at what we do. We leave the role vague. She is a group leader, which means whatever it turns out to mean here, and then we structure the task down to the last detail. The graphic, the caption, the date, the wording of the ask. We have it precisely the wrong way round, and then we are puzzled that the groups do not take ownership of anything.

Valentine has a name for what people do when the role is not clear. They forage. They go hunting through whatever they can find for some signal of what they are supposed to be doing. This is why the hardest part of starting a new job is usually working out what the job is. A volunteer leader does the same thing, with much less to go on and much less standing to ask.

The over-structured version fails in the opposite direction. Her finding is that a team locked into an inflexible workflow cannot respond to contingencies or pivot when something better appears. Two failures, produced by the same missing definition, which is why we so seldom recognize them as one problem.

One more thing about defined decision space for supporters: they have to remain true on the bad weeks too. If boundaries shift without explanation, precisely at the moment when the temptation to override them is strongest real damage to trust, autonomy, and relationships can happen. Building autonomy frameworks you can stand by.

What sits inside the space matters more than how big it is, and this is where programs stop short even when they mean well. Choosing between three tactics on a menu is not deciding. The decision space that makes something organizing rather than mobilizing is the one where your supporter is working out what her group should take on this month, against what she knows about her own town…who is angry about what, which councillor is wobbling, what happened at the school last year and how to maneuver these roadblocks she knows so well.

That is real political work. It’s not just for professionals like us, for good reason. None of your supporter’s intricate knowledge of her community exists at head office, and none of it can be sent there. It is not a pleasant side effect of distributed work. It is the strategic advantage over a centralized operation.

I have watched what happens when the space is never drawn. Organizations build local groups that look like distributed organizing from the outside. They are well attended and active, with real work going on. The leaders are engaged. But then something arrives that the framing does not cover. For example: there’s a split in the coalition, a story that breaks locally, or a  moment where the answer isn’t obvious.

And the groups freeze.

Not because the people in them are bad leaders, but rather because the organization has ever asked them to make this kind of decision., So the first one they face is a hard one, with no practice behind it, no stated authority, and no idea whether they will be backed afterwards.

Which is why the space on its own is not enough.  Infrastructure is needed to coach her through the first hard call, and then the second. Not a training session in March, but support alongside her while it is happening, to walk her through the decision she makes. Marshall Ganz, who has spent a lifetime on this, from the farm workers' movement to a classroom at Harvard, puts the purpose of the whole exercise plainly. The point of leadership is to make more leaders.

Historically holding alignment across a lot of people has always required human attention: somebody reminding, somebody correcting, somebody keeping everyone pointed the same way. As we know, time and attention is the thing an organization has least of. Time is money so they don’t even look into it. 

Valentine's book Flash Teams is about what changes when we try. Her proposal, in general terms, is that work gets divided into defined pieces. Clarity is given around each of these delegated pieces: who holds it, the inputs, the outputs, and when to escalate a problem.. Once the division is that explicit, software can help compose the structure while the people do the work on their delegated pieces. Her interest is teams of experts.. Mine is a hundred groups of everyday people who, like the ER scenario, have not met each other yet.. The mechanism is the same one.

What all of it requires feels easier said than done.  But it’s really three simple steps. The collective agreement written down rather than assumed. The decision space drawn, so that both of you and the supporters know where they can lead. Coaching and support as a supporter becomes a leader. 

Now stand up and read those questions from the beginning of the chapter. Did your answers change? My hope is that it stopped being whether you would be comfortable with your supporter making the call.

Instead it becomes a question of if she knew it was hers to make.