Haute Lumière

Commerce · I.07 · MMXXVI · daylight

La Bourse  /  Volume I  /  Nº I.07  /  Workbook — the student

WORKBOOK — THE STUDENT

Chapter I.07 · The Second Circle

For the person studying this alone, or in a seminar, with no organisation to change yet. You have less authority than the executive and more access to the one thing this chapter is actually about: a small number of people who already know each other.

---

WHY THIS WORKBOOK IS DIFFERENT

The chapter is written for someone trying to move a department. You are not, yet. It would be reasonable to conclude that the diffusion arithmetic waits until you have colleagues.

It does not, and the reason is worth stating plainly: you are, right now, inside the densest network you will ever belong to. A cohort, a corridor, a society, a shared house, a team — these are clustered, high-redundancy structures where everybody has several contacts in common. That is exactly the topology the chapter says a complex contagion needs and exactly the one most organisations have engineered away. For the next few years you have the good network and no authority. Later you will have the authority and the bad network.

So use the good network to learn the mechanism, because the mechanism is the transferable part. Find the gap in a distribution. Recruit into the gap rather than at the group. Build redundancy on purpose. Write four lines. Name a successor. Split rather than widen.

---

PART ONE — DISCOVERY

Days 1–30: learn to see the shape

Exercise 1.1 — Draw your network without yourself in it (60 minutes)

Take a group you belong to — a seminar, a society, a team, a house. Write every member's name on a sheet. Draw a line between any two people who would text each other directly about something that mattered.

Now cover your own name with your thumb.

Count the lines that are left. Then count how many people have two or more lines to others, not counting lines to you.

That second number is the group's reinforced-member ratio, and it is the single most informative thing you can know about whether anything can spread in it. Below about a quarter, the group is a star and you are the centre of it, which feels like being important and is in fact a structural weakness.

Exercise 1.2 — The threshold map (45 minutes)

For a change you would actually like to see in a group you belong to — everyone reading the paper before the seminar, a rota that works, one evening a week without phones — write every member's name and beside it your honest estimate of their threshold: how many others would have to be doing this before they joined?

You will be wrong about individuals. You will be roughly right about the shape, and the shape is what you need. Sort the numbers and look for the gap — a value at which nobody sits.

Then write one sentence: the person I need is somebody at threshold ___.

That sentence is the entire method. Almost nobody, at any level of seniority, ever writes it.

Exercise 1.3 — Three appreciative interviews (90 minutes, over a week)

Find three people who have made something spread — a club that outlived them, a tradition, a study group, a campaign. Ask exactly this:

"Think of something you started that kept going after you stopped running it. Who was the second person involved, and what made them credible to the third?"

Take notes on the structure, not the story. Who knew whom. When they met. Whether the second and third people knew each other before.

You will hear the same answer more often than you expect: the thing survived because two other people got on with each other.

---

PART TWO — THE ARITHMETIC

Days 31–50: compute before you persuade

Exercise 2.1 — Reproduce the chapter's figures (2 hours)

Open lib/verify/I_07.py, read it, and then compute these independently — by hand, in a spreadsheet, or in whatever language you use.

1. Rogers' bands as expected heads for group sizes 6, 12, 25, 40 and 80. Confirm that one innovator is expected only at forty people.
2. The Bass hazard table for a circle of twelve with p = 0.03, q = 0.38. Confirm the first adopter raises the pull by 105.6 percent.
3. t* = ln(q/p)/(p+q). Confirm 6.19 periods, 46.1 percent adopted at the peak, and 17.0 percent on p alone.
4. The two threshold circles. Confirm that circle B has the lower mean and stalls at one.
5. The ring-lattice clustering coefficient at k = 2, 4, 6.

Then do the thing that matters most: take one figure in this chapter and check it against a source the chapter did not use. The Bass coefficients are a good target — they are meta-analytic means, and individual studies vary widely. If you find a discrepancy, write it down and bring it to your seminar. This edition wants to be checked.

Exercise 2.2 — Write your own threshold model (2 hours)

Write ten lines of code — any language — that takes a list of thresholds and returns how many people end up acting. The algorithm is four lines: start at zero acting, count how many have a threshold at or below that number, repeat until it stops changing.

Then play with it, because playing with it is where the understanding lives:

• Start with 0 1 2 3 4 5 6 7 8 9 10 11. Confirm twelve.
• Change one value at a time until it stalls. Notice how little it takes.
• Find a distribution with a low mean that stalls, and one with a high mean that cascades. Write down both and what they taught you.
• Add a committed parameter — people who act unconditionally — and find the smallest number that unsticks each stalled distribution.

Twenty minutes of this teaches more than the chapter does. That is not a criticism of the chapter; it is why the module exists.

Exercise 2.3 — Find the honest negative (30 minutes)

The chapter contains several cases where the approach loses: the star that cannot propagate at any size, the SDS collapse, the coordination ceiling, and Los Molinos at 5.5 percent after two years of correct advice.

Pick the one that troubles you most and write 300 words on why it is in the chapter rather than left out. Then write the strongest honest negative against something you believe about how groups work — the version that troubles you, not a straw one.

If you cannot write it, you do not yet hold your own position well enough to defend it. That is not a criticism. It is the assignment.

---

PART THREE — DESIGN

Days 51–75: build a real one

Exercise 3.1 — Recruit into the gap (two weeks)

Take the sentence you wrote in Exercise 1.2 — the person I need is somebody at threshold ___ — and find them. The criteria, in order:

1. They are checked with. Other people ask their opinion unprompted. In a cohort this is obvious within a fortnight and almost nobody notices it.
2. Their threshold is low but not zero. Somebody who agrees instantly is another innovator and their agreement carries no information.
3. They are visible to your next recruit. If your second and third people do not already know each other, introducing them is the work.
4. They have something to lose. Participation from somebody with no standing at risk tells observers nothing.

Ask them directly, in person, for something specific and small. Not "would you be interested in" — a task, with a date.

Exercise 3.2 — Manufacture the redundancy (one week)

Put your second and third people on a shared piece of work without you present. An hour, a real output, both names on it.

This will feel like giving something away and it is the highest-leverage hour in the workbook. Until they have a relationship that does not route through you, you have a star, and a star cannot carry anything.

Afterwards, ask each of them separately what the other is like to work with. The answer tells you whether the tie took.

Exercise 3.3 — The four lines (one meeting)

Write them, for your group, on one page:

1. Who is in, by name, and how somebody joins.
2. How we decide. Quorum, and what happens on a tie.
3. Who watches the shared thing — the rota, the money, the document — how often, and reports to whom.
4. What happens the first time somebody does not do what they said. Make it small. The point of a graduated sanction is that it is usable.

Then hold the meeting where everybody says yes out loud, together. Do not circulate it for comment. The simultaneous assent is the mechanism — it is the only moment when every member receives reinforcement from every other member at once — and an email thread cannot produce it.

---

PART FOUR — DESTINY AND DELIGHT

Days 76–90: make it outlive you

Exercise 4.1 — Name your successor and start the handover now (this week)

Choose the person who would run this if you vanished. Then, immediately, give them something that is genuinely theirs: they chair the next meeting, their name goes first on the next output, they answer the next question that comes to you.

Succession that begins at handover has already failed. The network has to be rewired months in advance, and the rewiring is visible: people have to start going to them instead of you, which they will only do if they see it happen.

Exercise 4.2 — Write the split rule (20 minutes)

One sentence: when this group reaches ___ people, it divides into two, and ___ decides how.

Write it while the group is small and nobody has anything to lose by it. Stating it in advance removes the only conversation that reliably turns a growing group into a faction fight.

Exercise 4.3 — Delight, designed (ongoing)

Delight is the adoption mechanism, not the reward. A group people enjoy being in recruits itself; a group that is merely correct requires you to keep pushing.

Choose one element and make it genuinely good. The room. The hour. The food. The ritual at the start. The thing somebody looks forward to.

Then write one sentence: the part of this people look forward to is ___. If you cannot complete it, redesign until you can.

---

THE TERM PROJECT

One circle, carried the whole way

Take one group you have real access to and run the full ninety days on it.

Deliverables.

1. The network map (1 page). Before and after, with yourself removed from both. State the reinforced-member ratio at each point and how you computed it.
2. The threshold map (1 page). Every member, your estimate, the gap identified, and the sentence naming who you needed.
3. The arithmetic (600 words plus workings). Your own threshold model, the distributions you tested, and the smallest committed number that unsticks your group. Include at least one honest negative: a condition under which your circle fails.
4. The four lines, assented to in a meeting, with the date and who was present.
5. A decision made without you (documented). One real decision the group made in your absence. This is the deliverable that cannot be faked.
6. The reflection (600 words). What you expected, what happened, and what the gap between them tells you about your model of people.

How it is assessed. Not on whether the group thrived. On whether the structure changed measurably — whether the reinforced-member ratio rose, whether a decision happened without you, and whether your numbers can be reproduced by a reader.

A circle that did not grow but became independent of you is a first-class piece of work. A circle that doubled around a single centre is not.

---

SELF-ASSESSMENT

Score yourself honestly. This is for you.

The two that matter most are the third and the fifth. The rest can be learned in an afternoon. Finding a gap and manufacturing a tie are habits, and habits take a term.

---

CARRYING IT FORWARD

What you will have, if you do this properly, is:

• One group that runs without you, which you can describe in ninety seconds.
• A threshold model you wrote yourself and understand from the inside.
• The habit of asking who is connected to whom before asking who agrees.
• Three appreciative interviews about how things actually survive their founders, which is a subject almost nobody studies deliberately.

That last one compounds strangely. Most people spend a career learning to start things and never learn to make one that continues. You can learn the second skill first, in a student society with nothing at stake, and it will transfer completely to a business unit with a great deal at stake — because the arithmetic does not know the difference.

---

APPRECIATIVE QUESTIONS FOR YOUR SEMINAR

1. When has one of you joined something because you saw two people you respected doing it? Who were they, and did they know each other?
2. Which group any of us belongs to would survive its most active member leaving tomorrow — and what makes that true?
3. What is already working about how this seminar makes decisions, and what would we have to write down for it to keep working with different people in the room?
4. If each of us built one circle this term that ran without us, what would this cohort look like in a year?