The Maneuver

An executive asked how they got a difficult decision to go their way usually tells a story of improvisation. They saw an opening, called in a favor, found that someone was stalling and went around them, and in the end the decision came out right. Each account sounds like its own thing, and the more levers and detours it contains, the less it seems that any general description could hold it.

The Lam System for Political Interaction takes the opposite view of where the variety lives. Every maneuver, from a single conversation to a campaign with several levers and a sub-maneuver two levels down, runs the same fourteen phases in the same order. A maneuver that looks complicated is a simple one whose middle ran more than once, or which ran again, whole, one level down. The order is fixed because each phase writes down what the next one reads, so it follows from when the information exists, and skill does not change it. What the actor does choose is how many times the middle runs, how the runs interleave, and how far a failed gate is followed down, and those choices are fewer than the stories suggest.

What a maneuver is

A maneuver is one actor's intervention into one target's selection. The actor wants a resource that the target's decision controls. The actor commits to one or more levers through which the target is expected to come around to the preferred position, plays each lever with something the actor holds, and reads whether the target's view moved the way the lever projected. The maneuver closes when the target selects and an outcome resolves, and what the actor obtained flows back into what the actor holds, ready for the next one.

Each phase names one act. Interest says what is wanted. Detection says whose decision it is and what that person is choosing between. The objective fixes a projection: this selection will yield this position. A lever, which the system calls an intention, says how the target's comparison of the options is expected to change. Required resources say what the lever needs, and the gate asks whether the actor has it. Execution plays the lever and spends what it used. Perception is the target's reading of the comparison once the lever has landed, which the actor sets beside the projection. Selection and outcome settle the decision, obtained utility records what the target gained, and flow back records what the actor gained. The record writes an entry of its own when a cycle opens and another when it closes, so that nobody has to infer where a cycle began or ended.

Who performs each phase is fixed as well. The actor declares the interest, detects, sets the objective, chooses the levers, names what they need, executes them and takes in what flows back. The target perceives, selects and gains. The outcome belongs to neither of them: it is the cycle's own, the point at which the decision stops being a projection and becomes a fact.

One maneuver, start to finish

Illustrative. The people and the company are invented. Lena Ortiz is in charge of customer operations at a regional freight carrier. The night dispatch desks are run from the depots, and she wants them brought under her and paid for from her budget.

A cycle opens under her agenda. Its interest is the dispatch budget and the authority over the desks. Detection names the chief operating officer, Daniel Okafor, as the target, because the decision is his. It names the preferred position, that Daniel approves the move, and two adverse alternatives: Daniel leaves dispatch with the depot managers, or he hands it to the automation program. The objective is that Daniel approves, and that the approval brings the desks and their budget to her. That is the movement the system calls Declare.

In Choose levers, Lena declares two levers at the outset. The first is to show Daniel that a single desk cuts the overtime he has been taking criticism for. The second is to get the controller, Priya Natarajan, to put her signature on the savings figure, so that the number Daniel sees carries finance's name. She names what the first lever needs: the overtime data and her standing with Daniel.

Test and act follows. The gate asks whether she holds both, and she does, so it passes. She executes the lever, putting the overtime case to Daniel, and in doing so spends the overtime data and some of her standing with him. Then comes the first part of Read: Daniel's perception. He reads the desk as cutting the overtime his critics keep raising, which is what the lever projected.

Lena turns to the second lever and names what it needs, Priya's sign-off. She does not hold it, and the gate fails. She escalates. Daniel's cycle stops at the gate and waits, unchanged, and a sub-cycle opens under the second lever. Its interest is exactly the sign-off the parent lacked. Its detection names Priya as the target, signing as the preferred position and declining as the adverse alternative. Its objective is that Priya signs. Its lever is to take Priya through how the savings figure was built, which needs Lena's standing with Priya. That gate passes, Lena executes, and the standing is spent. Priya reads the figure as one she can put her name to, selects signing, and the outcome resolves. Her obtained utility is a savings claim that finance has tested before it reaches the chief operating officer. The sub-cycle closes, and the sign-off flows back to Lena.

Daniel's cycle resumes at the gate where it stopped. The gate tests again and passes. Lena executes the second lever: the signed figure goes to Daniel, and the sign-off is spent. Daniel reads the savings as finance's number, which is what the second lever projected. With both levers read, Daniel selects. He approves the move, and the outcome resolves.

Close and carry forward ends it. Daniel's obtained utility is lower overtime and a cleaner organization. The cycle closes, and Lena's obtained resources, the budget and control of the desks, flow back into what she holds. What flowed back is what she wanted at the top: the interest the cycle opened with is the resource it returned. On the normal path the two always match, and the match is how the record shows the maneuver delivered what it was opened for.

Counted, the maneuver cost Lena the overtime data, some of her standing with Daniel and her standing with Priya. The sign-off came in through the sub-cycle and went out through the second lever. She ends holding the budget, the authority and what is left of her standing with Daniel.

Cycle: Lena and Daniel

1. OpenThe cycleA new cycle opens under Lena's agenda.
2. InterestActorLena wants the dispatch budget and the authority over the desks.
3. DetectionActorDaniel will approve the move, leave dispatch with the depot managers, or hand it to the automation program.
4. ObjectiveActorDaniel approves, and the desks and their budget come to Lena.
5. IntentionActorFirst lever: show Daniel that one desk cuts the overtime he has been criticized for.
6. IntentionActorSecond lever: have Priya sign off the savings figure.
7. Required resourcesActorThe first lever needs the overtime data and Lena's standing with Daniel.
8. Gate passesThe cycleLena holds both.
9. ExecutionActorLena puts the overtime case to Daniel.Spent: the overtime dataSpent: some of her standing with Daniel
10. PerceptionTargetDaniel reads the desk as cutting the overtime his critics raise.
11. Required resourcesActorThe second lever needs Priya's sign-off.
12. Gate failsThe cycleLena does not hold the sign-off.
13. EscalateThe cycleDaniel's cycle waits at its gate. Nothing Lena holds is spent.

    Sub-cycle: Lena and Priya · 14 entries

    1. OpenThe cycleA sub-cycle opens under the second lever.
    2. InterestActorLena wants exactly the sign-off the cycle above lacked.
    3. DetectionActorPriya will sign off the figure, or decline to.
    4. ObjectiveActorPriya signs.
    5. IntentionActorTake Priya through how the savings figure was built.
    6. Required resourcesActorLena's standing with Priya.
    7. Gate passesThe cycleLena holds it.
    8. ExecutionActorLena walks Priya through the figure.Spent: her standing with Priya
    9. PerceptionTargetPriya reads the figure as one she can put her name to.
    10. SelectionTargetPriya signs.
    11. OutcomeThe cycleThe savings figure is signed off.
    12. Obtained utilityTargetPriya has tested the savings claim before it reaches the chief operating officer.
    13. CloseThe cycleThe sub-cycle closes.
    14. Flow backActorLena holds the sign-off.Returned: Priya's sign-off
14. ResumeThe cycleDaniel's cycle picks up at the gate where it stopped.
15. Gate passesThe cycleLena now holds the sign-off.
16. ExecutionActorThe signed figure goes to Daniel.Spent: Priya's sign-off
17. PerceptionTargetDaniel reads the savings as finance's number.
18. SelectionTargetBoth levers read, Daniel approves the move.
19. OutcomeThe cycleThe night dispatch desks move onto Lena's budget.
20. Obtained utilityTargetDaniel gets lower overtime and a cleaner organization.
21. CloseThe cycleThe maneuver closes.
22. Flow backActorLena holds what she opened the cycle for, beside what is left of her standing with Daniel.Returned: the dispatch budget and the authority over the desks

Opened, Priya's sub-cycle shows fourteen rows, one for each phase from the opening to the flow back, as any cycle would.

Why the phases come in this order

Each phase writes something into the record, and the phase after it reads that thing. Detection cannot ask who controls the resource until interest has said what the resource is. The objective projects that a named selection yields a named position, so it needs detection's names. A lever projects a shift in the target's weighing of the options, and the shift has to lead to the objective, so the objective comes first. Required resources belong to a lever, so the lever is declared before them. The gate tests what the actor holds against the required resources, and it has nothing to test until they are written. Execution plays a lever the gate has passed. Perception is read against the lever's projection, after the lever has landed. Selection comes only after each lever still in play has been read or dropped, because the target decides once (the essay on Selection takes the point further). The outcome is the selection made fact, obtained utility is the target's gain from that outcome, and flow back returns what the closed cycle yielded.

The figure below follows that order. On the actor's side it runs from interest to the objective, then through the intention and its required resources to the gate, and on to execution. On the target's side it runs from the matrix through perception to selection, and then to the outcome, which belongs to the cycle itself. A failed gate sends the actor back to interest, where a sub-cycle opens for what was missing, and what the outcome yields returns to what the actor holds, for the next maneuver.

By the actorBy the targetBy the cycle itselfA name only

The actor's side

1. Actor α / a [a.] A name only, performed by no one
   • → Agenda
2. Agenda Α / A [A] A name only, performed by no one
   • → Interest
3. Interest ι / i [A/i.] Performed by the actor
   • → Detection
4. Detection δ / d [A/i./d.] Performed by the actor
   • → Objective
   • Names the target, and moves nothing
   • Names the preferred position, and moves nothing
   • Names the adverse alternative, and moves nothing
5. Objective ω / o [A/i./o.] Performed by the actor
   • → Intention
6. Intention γ / g [A/i./o./g.] Performed by the actor
   • → Required Resources
7. Required Resources θ / q [A/i./o./g./q.] Performed by the actor
   • → The gate
8. The gate Does the actor hold every resource the lever needs?
   • If yes: → Execution
   • If no: escalate. A sub-cycle wants exactly the missing resource, and opens at Interest
9. Current Resources Θ / R [a./R.] A name only, performed by no one
   • → The gate
   • → Execution
   • → Actor. What flows back is held for the next maneuver
10. Execution ε / e [A/i./o./g./e.] Performed by the actor
    • → Decision Matrix. Changes benefit and cost
11. Obtained Resources Φ / Q [A/i./O./Q.] Performed by the actor
    • → Current Resources

The cycle's own

1. Outcome Ω / O [A/i./O.] Performed by the cycle itself
   • → Obtained Resources
   • → Obtained Utility

The target's side

1. Target τ / t [A/i./t.] A name only, performed by no one
   • → Perception
2. Decision Matrix Δ / D [A/i./t./D.] A name only, performed by no one
   • Preferred Position π / P [A/i./t./D./P.] A name only, performed by no one
   • Adverse Alternative ψ / X [A/i./t./D./X.] A name only, performed by no one
   • Benefit β / b A name only, performed by no one [A/i./t./D./P./b.] under preferred position, [A/i./t./D./X./b.] under adverse alternative
   • Cost κ / c A name only, performed by no one [A/i./t./D./P./c.] under preferred position, [A/i./t./D./X./c.] under adverse alternative
   • Utility λ / u A name only, performed by no one [A/i./t./D./P./u.] under preferred position, [A/i./t./D./X./u.] under adverse alternative
   • → Perception, from the utility row
   • → Selection, from the options
3. Perception Γ / G [A/i./t./D./G.] Performed by the target
   • → Selection
4. Selection Σ / S [A/i./t./S./P.] Performed by the target
   • → Outcome
5. Obtained Utility Λ / U [A/i./O./U.] Performed by the target

A method can be followed loosely, with steps swapped when time is short. This order cannot, because a skipped phase still leaves the next one needing its input, and the next one takes it from somewhere else, usually the actor's memory or a hope the actor never wrote down.

The cost of that substitution shows in particular places. An actor who plays a lever without first naming what it needs finds out about the shortfall at the moment of execution, in the target's office, in front of the person the lever was meant to persuade. Had Lena gone to Daniel with the savings figure before checking whether Priya would stand behind it, she would have discovered the missing sign-off when Daniel asked whose number it was.

An actor who executes without a projection has nothing to compare the target's response against. Perception in this system is a reading of the target set beside what the lever said would happen. Without the projection, all that remains is an impression of how the meeting went. The actor can say that Daniel seemed receptive, and cannot say whether the overtime case did the particular work it was chosen to do. Over many maneuvers the difference compounds, because only the second kind of reading teaches the actor anything about which levers move which people.

An actor who treats the selection as settled before every lever has been read is counting a decision the target has not made. The target selects once, on the whole comparison, and a lever still in flight can change it.

1. DeclareInterest, detection, objective
2. Choose leversIntention, required resources
3. Test and actThe gate, execution, escalation
4. ReadPerception, selection, outcome
5. Close and carry forwardObtained utility, obtained resources, flow back

Declare

The phases in it

Interest, detection, objective

What it reads

Only the agenda the cycle opens under. Inside the movement, detection reads the interest, and the objective reads the names detection wrote.

What it writes

What is wanted; whose decision it is and the options in front of them; the selection projected to yield the wanted position.

If it is skipped

A lever has no named alternatives to beat and no objective to lead to, so it is aimed at whichever options the actor already had in mind.

Choose levers

The phases in it

Intention, required resources

What it reads

The objective and the options detection named. Each lever projects a change in how the target compares them.

What it writes

Each lever's projected comparison at its own address, and what each lever needs.

If it is skipped

The gate has nothing to test, and a shortfall surfaces during execution, in front of the target.

Test and act

The phases in it

The gate, execution, escalation

What it reads

Each lever's required resources, set against what the actor holds, counted.

What it writes

The gate's verdict; on a pass, the execution and the resources spent; on a failure, a sub-cycle under the lever.

If it is skipped

A lever played without the gate rests on resources the actor only assumed, and a lever never executed leaves the target nothing new to read.

Read

The phases in it

Perception, selection, outcome

What it reads

Each lever's projection and its execution. Selection reads every live lever's perception.

What it writes

The target's reading beside each projection, the target's selection, and the outcome.

If it is skipped

The actor has an impression of the meeting and no measure of whether the lever did its work, and the close has no selection to rest on.

Close and carry forward

The phases in it

Obtained utility, obtained resources, flow back

What it reads

The outcome, and the interest declared at the top of the cycle.

What it writes

The target's gain, the actor's gain, and what the actor holds after the close.

If it is skipped

What flowed back is never checked against what was wanted, and the count of what the actor holds stops matching what was spent and gained.

The failure moves leave the order alone. They branch off at a small number of points: at the gate, where a missing resource can be fetched by escalating; at the reading, where a perception that fails can be answered with a new lever, by mooting or by abandoning; and anywhere at all, for abandoning, which is always available. Each move either carries the cycle on from where it stood, as a resumed parent picks up at its gate and a failed reading is followed by a new lever, or ends the cycle, and none of them lets a phase run before the one it reads from.

Several levers, one close

With only its first lever, Lena's maneuver would be the baseline case, in which every phase fires once: one interest, one detection, one objective, one lever, one set of required resources, one gate that passes, one execution, one perception, then the selection and the close. Fourteen phases fire, and there is no question of sequencing, because there is only one of each phase to fire.

The second lever changes the count of passes and leaves the shape alone. Each lever gets its own pass through the middle of the cycle: its intention, its required resources, its gate, its execution and its perception, in that order. The order within a pass is as fixed as the order of the cycle. The number of passes is the number of levers the actor declares. Then the passes converge, and the cycle ends on one selection, one outcome, one obtained utility, one close and one flow back.

The convergence follows from the unit. A cycle has one target and one decision. Daniel does not approve the dispatch move once for the overtime case and again for the savings figure; he approves it once, on the comparison as both levers have left it. The levers are two ways of changing a single comparison, so they share its close. That is also why levers that work on the same decision belong in one cycle. Split into two maneuvers, Lena's levers would each claim Daniel's approval as their own outcome, and the record would show two wins from one decision. A lever aimed at someone else's decision is a different matter: it has another target and so another cycle.

What the shape does not fix is how the passes interleave. Lena declared both levers at the outset and worked each one through its gate and execution before turning to the next. She could have declared the second lever only after seeing how Daniel took the first, or declared both and named everything each one needed before testing either gate. All three run the same passes in the same internal order, and they differ in when each pass begins. The same is true of reading: Lena read each lever's effect as it landed, and another actor might read the whole play once every lever has been played. How the levers were framed shows which reading a cycle is using.

A sub-cycle is just a cycle

When the second lever's gate failed, the procedure stayed the same. The sub-cycle with Priya opened, declared an interest, detected, set an objective, chose a lever, named its resource, passed a gate, executed, and then Priya perceived, selected and gained, and the cycle closed and flowed back. The same fourteen phases run at every level of a maneuver, in the same order, performed by the same parties.

What distinguishes a sub-cycle is its position: where its addresses sit, how it ends, and what it blocks while it runs.

Every constituent in the system has an address read like a path from the agenda down, and a sub-cycle's addresses sit under the lever that opened it. Priya, as the target of the sub-cycle, is written:

Read aloud: the target of the interest opened under the second lever, under the objective of the agenda's only interest. The last two segments, an interest and its target, are the same two segments that would address Daniel at the top of the maneuver. Everything before them says where the cycle sits. Anyone reading the record later can tell which lever a sub-cycle was opened for from its address alone.

A top-level cycle's close ends the maneuver. A sub-cycle's close hands what it obtained back to the parent, and the parent resumes at the gate where it stopped, which then tests again. The handback is checkable because the sub-cycle wanted exactly what the parent lacked: when the sign-off has flowed back, the gate has something to find. Escalation takes up what waits while a sub-cycle runs and when going a level deeper costs more than it returns.

While any cycle is waiting on a sub-cycle, no new top-level maneuver can open. The maneuver in progress finishes or is given up before another begins. Each escalation adds one waiting cycle and each handback removes one, so the waiting cycles rise and fall as the maneuver goes deeper and returns, and the deepest always finishes first.

Because the differences are all of position, anything that holds for a cycle holds for a sub-cycle, at any depth. Priya's sign-off is a decision with its own detection, its own reading and its own gain for the person making it. A sign-off of this kind is easily treated as an errand, something to collect on the way to the real decision. The system treats it as a maneuver, which is why Priya's obtained utility appears in the record beside Daniel's, and why Lena's standing with Priya is counted as spent. The resources are Lena's at every level, and whatever a sub-cycle spends is recorded in the sub-cycle, at the lever that spent it, where it can be seen when the maneuver is read back.

Where the actor's choices are

If the order within a pass is fixed and a sub-cycle is just a cycle, the actor's choices have to be somewhere else, and they fall into two kinds.

The first kind is interpretive. Whether Lena holds enough standing with Daniel for this particular ask is a judgment about Daniel, made from what she knows of him; the gate's verdict follows from it, and the system cannot make it for her. Whether Daniel read the overtime case the way the lever projected is a judgment only someone who watched him can make. Whether Priya's reading of the figure is one she will sign on is the same kind of question one level down.

The second kind is strategic: which lever to take through its gate first, whether to declare every lever at the outset or add each after seeing the last, whether to name a pending lever's resources now or press on with the lever in hand, whether to escalate for the sign-off at all or let the overtime case carry the cycle by itself, and whether to moot a lever once another lever's outcome has made it unnecessary.

Actors answer the strategic questions in characteristic ways, and the system recognizes three stances toward the uncertainty about whether a set of levers will be enough. One actor declares each lever only after seeing what the last one did. Another declares every lever at the outset and takes each through its gate and execution before the next, which is what Lena did. A third declares every lever and names everything each one needs before testing any gate, then goes after whatever turns out to be missing. Each suits a different judgment about how much the actor knows going in, and the system receives whichever stance the actor takes and never chooses one. The stances have an essay of their own, and The Boundary of Computation takes up where the system's work stops and the actor's begins.

The two kinds of judgment are one boundary seen from two sides. The interpretive judgments decide how the tests come out: whether the gate passes, whether a perception matches its projection. The strategic judgments decide the order in which the tests are reached. Once both are given, the rest of the maneuver is determined: which entries the record gains, what the actor holds, which cycle has control, which exits remain open. The system is exact in that sense only: relative to the actor's judgments, which it leaves open on purpose.

The arrangement leaves a machine a real share of the work. It can keep the record, refuse an entry that does not fit the phase the cycle is in, say which cycle has control and what the actor holds, and replay any earlier moment. In Lena's maneuver it could say where Daniel's cycle picks up once the sign-off flows back, or what she holds after the first lever, because each follows from her judgments. A system that went further, deciding that Daniel was persuaded or that the sign-off was worth escalating for, would be writing its own guess into the record as a fact about Daniel or about Lena's priorities. An actor can learn from the record because every entry in it is either something the actor judged or something that follows from what the actor judged.

What a fixed shape is for

Because every maneuver has the same shape, any two can be laid side by side, phase by phase. Lena's maneuver with Daniel and one she runs next year with someone else can be compared at the detection, at the gate, at the reading. A stalled maneuver stalls at a place in the shape, and the place says what kind of trouble it is in. A cycle waiting at a gate is short of a resource. One with a failed perception has a lever that did not do the work it was chosen for, and one that reached its close and returned something other than what it wanted was usually misdetected at the top.

The shape also decides which questions cannot be skipped. An actor who has to write the objective before choosing a lever must say what success would look like before deciding how to pursue it, and an actor who has to name required resources before the gate finds the missing sign-off on paper, before Daniel asks for it.

The executive's story of improvisation is accurate about the judgments: whose decision it was, which lever to try first, whether a sign-off was worth the standing it cost. Everything that followed from those judgments ran in the fixed order, and that is what lets the record show, afterward, which of the judgments went wrong.