Haute Lumière

Commerce · II.11 · MMXXVI · daylight

La Bourse  /  Volume II  /  Nº II.11  /  Workbook — the student

WORKBOOK — THE STUDENT

Chapter II.11 · The Case Against This Book

For the person studying this alone, or in a seminar, with no organisation to change yet. This is the chapter that makes the other seventy-six usable in an examination hall, and the skill it teaches is the one that will still be worth something in thirty years.

---

WHY THIS WORKBOOK IS DIFFERENT

Every other workbook in this edition asks you to apply something. This one asks you to attack something, which most students have never been taught to do well.

The reason to learn it now, rather than later, is that you will be marked on it for the rest of your education and paid for it for the rest of your career, and almost nobody is taught the mechanics. What passes for criticism in most seminars is one of three moves: doubting the author's motives, listing things the author did not mention, or restating a rival position at greater volume. None of the three is a criticism. A criticism names a claim, names the observation that would retire it, and reports whether that observation exists.

That is a procedure. Procedures can be learned in a term.

And the second reason, which matters more. The habit of writing kill conditions for your own beliefs is the single cheapest form of intellectual insurance available. You are about to spend several decades accumulating positions. The ones you write kill conditions for will cost you very little when they turn out to be wrong, because you will already have decided what wrong looks like. The ones you do not will cost you years.

---

PART ONE — DISCOVERY

Weeks 1–4: find criticism working at its best

Exercise 1.1 — The five cases, read properly (3 hours)

The chapter names five episodes of criticism working. Read one primary source for each, not a summary.

For each, write two sentences: what the critic did procedurally, and what the criticised party did that made the correction possible. The second column is the one nobody notices, and it is where the transferable skill lives. Reinhart and Rogoff sent the file. Kahneman agreed to a joint reanalysis with a man holding the opposite finding, at seventy-nine years old. Note that.

Exercise 1.2 — The appreciative interview about being wrong (45 minutes)

Find someone whose judgement you respect and ask exactly this:

"Tell me about a time you changed your mind about something that mattered. Not a small thing — something you had said out loud. What arrived? What made it possible to take it?"

Then stay quiet. Take notes on the conditions, not the conclusion.

You will hear, over and over, the same two conditions: the correction was specific, and it did not require them to be a fool for having believed the other thing. Those are design constraints on every criticism you will ever deliver, and you have just derived them yourself rather than being told.

Exercise 1.3 — The register you can already read (60 minutes)

Run this and read all twenty-four rows:

python3 lib/verify.py II.11

Pick one claim you think is misclassified — an F you would call U, or a U you think could be written as an F. Write the better version of the row. Bring it to your seminar.

This is not a comprehension exercise. It is the exercise. A register exists to be argued with line by line, and a student who can improve one row has done something a summary cannot demonstrate.

---

PART TWO — THE ARITHMETIC

Weeks 5–8: compute before you argue

Exercise 2.1 — Reproduce the chapter's figures (3 hours)

Do not take them on trust. Open lib/verify/II_11.py, read it, and compute these independently — by hand, in a spreadsheet, in whatever language you use.

1. The decoupling ratio. UK territorial 811 → 417 MtCO₂e; consumption-based fall of 15 to 20 percent. Confirm the 48.6 percent fall and the 2.43–3.24× band.
2. The required rate. 37.4 GtCO₂ against 275 GtCO₂ gives 7.35 years. Halving over 11 years requires 6.11 percent a year; at 3.0 percent output growth, intensity must fall 8.84 percent a year against an observed 1.45. Confirm 6.08×.
3. The footprint decomposition. 1.71 Earths at a 60 percent carbon share. Confirm 1.026 and 0.684, and the overshoot of 0.71.
4. The ISEW threshold. Welfare 100 at 1.5 percent a year, depletion charge 1.0 escalating at 3.0 percent. Confirm the index turns down in year 27, and that at a 1.5 percent escalation it never turns down at all.

Then do the thing that matters most: take one figure in this chapter and check it against a source the chapter does not cite. The consumption-based band is the obvious target, and it is given as a band precisely because it is revised. If you find a discrepancy, write it down and bring it. This edition wants to be checked; that is the difference between a text and a doctrine.

Exercise 2.2 — Decompose an index somebody loves (90 minutes)

Choose a composite indicator that is used in argument where you live — a university ranking, a cost-of-living index, a wellbeing score, an ESG rating, a league table.

1. Find the weights. If they are not published, that is your finding.
2. Compute what happens to the ranking if you drop the single largest component.
3. Compute what happens if you halve its weight.

Write one paragraph: is this index a measure, or is it one component wearing six? Run this test on an index whose conclusion you agree with. That is the version of the exercise that teaches something.

Exercise 2.3 — Price a disagreement (60 minutes)

Take a live argument you have with someone — political, economic, about a career choice — and convert it into a settleable proposition, the way Simon and Ehrlich did.

• What quantity?
• Measured how, by whom?
• Over what period?
• What value settles it each way?

You do not have to make the bet. Writing the terms will very often end the argument by itself, because at least one side will discover that their position does not specify anything. That discovery is the whole exercise.

Exercise 2.4 — The honest negative against yourself (45 minutes)

Take a position you personally hold and write the strongest honest case against it. Not a straw version — the version that troubles you, made by the best person who holds it, with their name on it.

Then answer two questions in writing:

• What would have to be true for them to be right?
• Has anyone looked?

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

---

PART THREE — DREAM AND DESIGN

Weeks 9–12: build your own register

Exercise 3.1 — The present-tense description (60 minutes)

Write 500 words describing a discipline, eight years from now, in which schools publish their kill lists. Present tense, description not aspiration.

Constraints:

• Describe a Tuesday — a seminar, a viva, an editorial meeting.
• Include one thing that is difficult and remains difficult.
• Include what happens the first time a well-liked claim is retired, and who has to say it out loud.

The last constraint defeats most people, and that is the finding. A culture of disconfirmation is easy to describe until you have to write the sentence where a specific person is publicly wrong and the room does not punish them.

Exercise 3.2 — Your own register (2 hours)

The central artifact of this workbook. Take a body of belief you actually hold — your dissertation's thesis, your view of your field, your reason for choosing this degree — and build the four-column table.

1. Write twelve claims, each in one sentence, each written so that two readers would test it identically.
2. Class each F or U. Be honest; most people's first pass is two-thirds U and they are surprised.
3. For every F, write the kill condition: the design, the sample, the result.
4. Leave status blank. You have not looked yet.

Then compute your own ratio, the way the chapter computed the edition's, and write it down: of twelve claims, ___ forbid an observation.

Exercise 3.3 — Go and look (3 hours)

Take one F-claim from your register and spend three hours trying to kill it. Search for the study that would retire it. Report one of three verdicts and nothing else:

• Standing — I looked, and the disconfirming evidence does not exist.
• Contested — I looked, and there is evidence both ways. Here is the best on each side.
• Retired — I looked, and here is the finding that retires it.

Never a score. A verdict of retired on your own claim is the best possible outcome of this exercise and should be marked as such, because it is the only one of the three that is certain to have cost you something.

---

PART FOUR — DESTINY AND DELIGHT

Making it a habit, and enjoying it

Exercise 4.1 — The second reader (this week)

Find one person who will read your register and tell you where you are wrong. Not a friend who will be encouraging. Someone who disagrees with you about something and whose reasoning you respect.

Ask them for one thing: which row is misclassified, and why?

Give them the credit, publicly, for whatever they find. That is how you recruit a second reader permanently rather than once.

Exercise 4.2 — The tightening rule (ongoing)

Put your register somewhere you will see it — the front of a notebook, a pinned file. Every time you revise a kill condition, write the previous version above the new one and leave it visible.

You will discover, within a term, that the pressure runs entirely one way: towards softening. Noticing that pressure in yourself is worth more than the register.

Exercise 4.3 — Delight, on purpose (ongoing)

There is a real pleasure available here and it is worth designing for deliberately.

Once a month, find something you believed and no longer do, and say so out loud to somebody. Not as confession — as news. Notice how light it is compared with the alternative, which is carrying the old position around in case anyone asks.

Write one sentence: the thing I most enjoy about being able to be wrong is ___. If you cannot complete it yet, do the exercise for a term and try again. A practice with no pleasure in it has a half-life.

---

THE TERM PROJECT

One claim, taken all the way

Choose a single claim made anywhere in this edition — any volume, any chapter — and take it the whole way.

Deliverables.

1. The claim, restated (200 words). In its strongest form, written so two readers would test it identically. If the edition's own wording is loose, tighten it and say what you changed.
2. The steelman against it (800 words). The best published case on the other side, with named critics and real citations. Not a summary of objections — the argument as its strongest holder would make it.
3. The kill condition (200 words). Design, sample, result.
4. The search (1,000 words plus a bibliography). Does the disconfirming evidence exist? What did you find, where did you look, and — this is marked — where did you not look?
5. The verdict (400 words). Standing, contested or retired, with the reason.
6. The reflection (400 words). What you expected to find, what you found, and what the gap tells you about your own priors.

How it is assessed. Not on whether you killed the claim. On whether the steelman is one its holders would recognise, whether the kill condition is specific enough to act on, and whether the search states its own denominator.

A project that fails to retire a claim and states honestly where it did not look is a first-class piece of work. A project that triumphantly retires a claim by attacking a version nobody holds is not.

---

SELF-ASSESSMENT

Score yourself honestly. This is for you.

The two that matter most are the second and the fourth. The first is easy and almost everyone can do it; doing it to your own writing is a different skill, and stating an opponent's case so well that they would sign it is the rarest thing on this list.

---

CARRYING IT FORWARD

You will not have an organisation to validate for some years yet. What you will have, if you do this properly, is:

• A twelve-row register of your own beliefs, with kill conditions, that you can hand to anyone.
• One claim taken all the way from steelman to verdict, which is a research method you can reuse on anything.
• Twenty interviews about how people actually change their minds, which is a genuine education in persuasion that no course teaches.
• The ability to decompose a composite index in ten minutes, which will make you the most useful person in a great many rooms.

That last one is worth more than it sounds. Most arguments about indicators are won by whoever first asks which component is doing the work, and almost nobody asks.

---

APPRECIATIVE QUESTIONS FOR YOUR SEMINAR

1. When has one of us been argued out of a position, and what did the other person do that made it possible?
2. Which claim in this edition does this group find least convincing — and can we write the kill condition for it together, right now, in ten minutes?
3. What is already working in how this seminar disagrees? Name the specific behaviour, not the atmosphere.
4. If every essay in this course arrived with the finding that would retire its thesis, what would change about how we read each other's work?