Haute Lumière

Commerce · II.04 · MMXXVI · daylight

La Bourse  /  Volume II  /  Nº II.04  /  Workbook — the student

WORKBOOK — THE STUDENT

Chapter II.04 · The Science of Wholeness

For the person studying this alone, or in a seminar, with no organisation to change yet. This is the chapter that will make you the most useful person in a room full of people who have all read the same popular science book.

---

WHY THIS WORKBOOK IS DIFFERENT

Most of what you will read about systems in economics is downstream of four primary sources, and almost nobody reads them. They read the third summary. The summaries are not wrong, exactly; they are smoothed, and the thing that gets smoothed away is always the same thing — the threshold, the conjunction, the regime of validity, the sentence where the author said this does not extend.

So this term you are going to do something that costs one afternoon each and compounds for thirty years: read four primary sources and write down, for each, the sentence where the author limits their own claim.

That is the whole skill. Everything else in this workbook is practice at it. The skill transfers to every discipline you will ever borrow from, and it is the difference between being the person who says the interesting thing and being the person the interesting thing can be checked with.

---

PART ONE — DISCOVERY

Days 1–30: find the limits in the originals

Exercise 1.1 — The four sources, and their self-limits (four afternoons)

Read, in this order, and for each write one paragraph: what it licenses, and the exact sentence where the author says what it does not license.

You are not looking for flaws. You are looking for fences, and the authors built them themselves. A scientist who fences their claim is telling you where the solid ground ends, which is a gift.

Exercise 1.2 — The borrowing audit (90 minutes)

Take three papers, articles or books in economics, management or sustainability that use systems vocabulary. For each, find every systems term and mark it:

• L — licensed: the components, boundary, units or threshold are named.
• I — image: it is doing rhetorical work, legitimately, in a rhetorical place.
• D — decoration: it is in the argument and it is not carrying anything.

Count them. Report the fraction. Do not name and shame anybody — the point is the ratio, and the ratio in your own writing is the one that matters.

Exercise 1.3 — Find the rigorous transfer (2 hours)

Pick one discipline you already know something about — biology, engineering, music, medicine, linguistics. Find one case where an economist borrowed from it properly: named the referents, stated the conditions, wrote the equation. Write 400 words on what made it work. Brian Arthur's 1989 paper is the worked example in the chapter; find your own.

---

PART TWO — THE ARITHMETIC

Days 31–45: compute the four results yourself

Open lib/verify/II_04.py. Every figure in the chapter is there with its inputs, its units and its sources. Do not read it as a result. Run it, then break it.

Exercise 2.1 — Reproduce the equifinality demonstration

Write your own integrator for dx/dt = −0.5x + 0.2y + 10, dy/dt = 0.3x − 0.6y + 4 and confirm it reaches (28.3333, 20.8333) from three different starts. Then solve x* = −A⁻¹b by hand and check the two agree. Then remove the source terms, make the system conservative, and watch three starts give three different ends — (20.0000, 20.0000), (52.5000, 52.5000), (46.5000, 46.5000).

The control is the exercise. Anyone can show a system converging. Showing the case where it does not converge, and knowing why, is what makes the first result a finding.

Exercise 2.2 — The entropy budget, from the constants

From the solar constant of 1361 W/m², the Earth's radius and an albedo of 0.294, compute the absorbed power. Then the entropy in at 5772 K and out at 255 K, using the factor of 4/3 for blackbody radiation. You should get a planetary export of 6.1235×10¹⁴ W/K. Divide by the economy's 6.5972×10¹⁰ W/K.

Now the part that teaches: do it again without the 4/3. Note that you have understated the export by a third. Write one sentence on why a factor you did not know about is more dangerous than an input you cannot find.

Exercise 2.3 — The threshold

Compute ΔT_crit for water layers of 1, 5, 10 and 50 mm at Ra_c = 1708. Confirm the 1,000× fall in required gradient between 1 mm and 10 mm. Then write down three systems in your own life where the "depth" term could be increased — where a longer horizon or an extra layer would let a smaller difference organise something.

Exercise 2.4 — The elasticity, derived

Derive S* = aG/(a+k) by hand. Differentiate to get the elasticities: exactly 1.000 to the goal, k/(a+k) to the parameter. Confirm the 4.00× at a = 0.3, k = 0.1 and the 1.111× at a = 0.1, k = 0.9.

Then find the value of a/k at which the ratio falls below 1.5, and write one sentence on what that means for where you would intervene.

---

PART THREE — DREAM AND DESIGN

Days 46–70: build the licence habit

Exercise 3.1 — Licence your own last essay (2 hours)

Take the most recent thing you wrote that made any systems claim. Run the four questions on every such sentence:

1. What do the terms refer to here — components, boundary, flows, units?
2. Is the source claim a theorem, a measurement, or an ordered intuition?
3. What is the threshold or condition, and does your case satisfy it?
4. What would falsify the transferred claim?

Rewrite the essay. Note how much of it got stronger rather than shorter. That is the finding, and it is the one that makes the habit stick.

Exercise 3.2 — The one-page reference card

Build a card you will carry: the four questions on one side; on the other, the six-point autopoiesis key, the twelve leverage points in Meadows' order, the entropy balance, and the two elasticity formulas. Make it beautiful enough that you will actually keep it.

Exercise 3.3 — The seminar experiment (one session)

Bring a paper to your seminar and run the borrowing audit live, as a group, with the rule that the person who proposes a mark must say what would change their mind. Note how long it takes for the group to start doing it to itself without prompting. In our experience this is under twenty minutes.

---

PART FOUR — DESTINY AND DELIGHT

Days 71–90: make it hold, and enjoy it

Exercise 4.1 — The standing file

Start a file called licences.md. One line per systems claim you rely on: the claim, the source, the page, the limit, the date you checked. This becomes the single most valuable document you own within two years, because every time somebody challenges you, the answer is a lookup rather than an argument.

Exercise 4.2 — Falsify one of your own (one afternoon)

Take the systems claim you are most attached to and spend an afternoon trying to break it, on a route that shares no assumptions with the one you used to build it. If it survives, it is now load-bearing and you may build on it. If it does not, you have just saved yourself a chapter of a thesis.

Exercise 4.3 — The pleasure exercise

Find a metaphor you have used for years and go looking for the theorem underneath it. Write 200 words about what you found, whether or not there was one. The delight in this chapter is real and it is the reason the discipline holds: a metaphor that turns out to be an equation is one of the better feelings available in intellectual work, and it is available to you several times a year.

---

THE TERM PROJECT

One piece of work, carried the whole way

Write a licensed systems paper, 3,000 words, on any economic question you care about. The requirements:

1. Every systems term is licensed in the sentence next to it. Components named, units given, threshold stated.
2. One theorem, one measurement, one ordered intuition, each labelled as what it is.
3. One honest negative: a place where your own argument fails, a threshold it needs that is usually absent, or a condition you cannot verify.
4. A computation file alongside the prose that prints its inputs with units and sources before it prints any result. Standard library only.
5. A falsification section: what evidence would change your mind, stated specifically enough that someone could go and get it.

Marking, out of 100: licensing 30 · arithmetic reproducibility 25 · the honest negative 20 · clarity of prose 15 · the falsification section 10.

The examiner's instruction, which you may as well know: a paper with three strong licensed claims scores higher than one with nine claims of which six are decoration. You are being marked on the ratio, not the count.

---

SELF-ASSESSMENT

Score yourself 1–5 on each, at day 1 and at day 90.

Any item at 4 or 5 at day 90 is a capability you will use for the rest of your career. Any item at 2 is a single afternoon away from a 4, and the afternoon is named in Part One.

---

CARRYING IT FORWARD

Three things to do in the year after this chapter.

Read one primary source a month. Not a summary, not a review — the paper. Keep the limit sentence from each one in licences.md. Twelve fences a year makes you, within three years, the person in the room who knows where the ground ends.

Teach the four questions to somebody once. Teaching is the only reliable way to find out which parts of a method you have actually internalised, and it costs a coffee.

Find one place where the rigorous version is more exciting than the loose one and write about it publicly. There are many. The 346× structural coupling result is better than the vague version of the same claim, the elasticity derivation is better than "goals matter more than parameters", and the convexity of deferred maintenance is better than "look after your things". Precision is not a tax on wonder. It is usually where the wonder was hiding.

---

READING A PAPER YOU ARE NOT QUALIFIED FOR

You will need this in week one, so take it now rather than discovering it.

Three of the four sources in this chapter are physics or biology written for physicists and biologists. You are going to open Prigogine and find notation you have never seen. The ordinary response is to conclude that the source is closed to you and go back to the summary. It is not closed, and here is the method that opens it.

Read the abstract, then the conclusions, then the figure captions, and only then the middle. The middle is where the technique lives; the claim lives at the two ends. You are after the claim and its fence, and both are usually in plain language.

Find the assumptions section and read it twice. In a physics paper this is often three lines and it is where the fence is. In the linear regime near equilibrium. For rigid-rigid boundary conditions. Assuming local equilibrium holds. Those clauses are the whole of what you came for, and they are written in words you already know.

Look up exactly three symbols. Not all of them. The three that appear in the claim sentence. If you cannot identify which three, you have not yet found the claim sentence, and you should go back to the conclusions.

Write the claim in your own sentence, with its fence attached, before you close the file. If you cannot, you have read it and not taken it. Ten minutes more now saves the whole afternoon again in six months.

And do not be embarrassed by how long the first one takes. The first primary source in an unfamiliar field takes three hours. The fourth takes forty minutes. The twentieth takes fifteen, and by then you are reading the assumptions first out of habit and finding the fences before you find the claims — which is the reading posture of everybody you currently think of as intimidating.

---

APPRECIATIVE QUESTIONS FOR YOUR SEMINAR

1. When has a precise definition made an idea more exciting for you rather than less? What was it?
2. Which borrowing from another discipline has served economics best, and what did the person who made it do right?
3. If our whole seminar licensed every claim it made for one term, what would we be able to publish at the end of it that we could not publish now?
4. What is the most useful fence any author has ever built around their own claim, and how did you come across it?
5. Who in this room knows a discipline the rest of us do not, and what would it take to get one rigorous transfer out of it this term?
6. What would make checking each other's numbers feel like a gift here rather than an audit?