La Bourse / Volume VI / Nº VI.09 / Quiz, reflection, essays

A woman in linen standing at a doorway beside a young olive tree, warm light in the room behind her. — Plate VI.09 · Quiz, reflection, essaysThursday, Noon, at the Cathedral Door.A conflict system is not a document either. It is a place, a time, and a group of people who are known to be there — and the cheapest of the three to provide is the place.

ASSESSMENT · Chapter VI.09 — Conflict, Repair, and the Long Table

Three instruments: a ten-point quiz, eight reflection questions, five essay prompts. The quiz checks comprehension rather than recall. The reflections are private and first-person. The essays are arguable from more than one side.

THE QUIZ — ten points

Four on recall.

1. State Ostrom's fourth and sixth design principles, in her terms.

Four — monitoring: monitors who audit conditions and behaviour are accountable to the appropriators, or are the appropriators themselves. Six — conflict-resolution mechanisms: appropriators have rapid access to low-cost local arenas to resolve conflicts among themselves and with officials. One mark each. A half mark is lost for "monitors are appointed" — the accountability direction is the whole principle, not a detail of it.

2. Name the six rungs of a graduated sanction ladder as the chapter sets them out, and the rule for moving between them.

A word · a recorded word · a facilitated conversation · a panel hearing with a finding · suspension of a specific right · exclusion. The movement rule: a rung is used twice before the next becomes available, and nobody skips except for conduct defined in the membership agreement. Both halves of the rule are needed: without the first the ladder collapses to its top rung under pressure; without the second it becomes a shield.

3. What did McEwen and Maiman find about compliance with mediated agreements against adjudicated judgments, and what is the caveat the authors themselves state?

Full compliance with mediated agreements was 70.6 per cent; with adjudicated judgments, 33.8 per cent. The caveat is selection: these were cases that chose mediation, so the parties were not randomly assigned.

4. What population does the randomised evidence on restorative conferencing actually cover?

Cases where the offender had already admitted responsibility and both parties consented to a conference. The mark is for recognising that this excludes the person who denies everything and will not attend — which is the case a conflict system most needs a design for.

Four on application.

5. A firm proposes replacing its grievance procedure with mandatory arbitration, arguing it is faster, cheaper and less adversarial. All three claims are true. What is missing from the paper, and what one clause would you add?

Missing: the distributional finding. Colvin (2011) found an employee win rate of 21.4 per cent in AAA employment arbitration against 36.4 per cent at federal trial — about 1.70 times. Speed is real and it is not neutral; the efficiency is partly a transfer of expected value to the firm. The clause to add is both-ways invocation: either party may reach the terminal rung. It costs nothing and it is what distinguishes a fallback from a management instrument. Credit any answer naming the asymmetry; full marks require a remedy, not only a critique.

6. A colleague says: "Mediated agreements are complied with twice as often as court orders, so we do not need enforcement." What is wrong with the inference?

Two things. First, the 70.6 vs 33.8 comparison is of voluntary compliance; a court order's residual can be enforced and a mediated agreement's cannot, so the comparison is incomplete until enforcement effectiveness is specified. Second, selection: mediating parties chose to be there. The strong answer computes the break-even — consent alone wins only while enforcement recovers less than 55.6 per cent of the unpaid balance — and notes that this makes the question empirical and local rather than philosophical.

7. Your organisation's published count shows rung-three (facilitated conversation) outcomes falling and rung-six (exclusion) outcomes rising, with total complaints flat. Diagnose it.

The ladder is collapsing to its top rung. Under stress, organisations stop using the rungs that take three weeks and reach for the one that takes an afternoon. The total being flat rules out a rise in conflict and isolates the change to routing. Credit answers naming remedies: shortening rung three's elapsed time, protecting panel capacity, or making the skip rule explicit rather than customary.

8. Why does the chapter insist the first sitting of a new forum have nothing on its agenda?

Because a forum that first appears on the day it is needed is a tribunal, and a tribunal is convened about somebody. A forum that has already met four times about nothing is furniture, and furniture can be used without anyone having to be brave. The standing cost — £17.50 per member per year in the worked house — buys exactly this and nothing else.

Two that require the arithmetic to be done.

9. A cooperative of 200 members runs a panel costing £3,500 a year standing and £330 per hearing. One exclusion costs it £7,575. Without a forum, 30 per cent of disputes end in somebody leaving; with the ladder, 5 per cent do. Above what dispute rate does the ladder pay for itself? Show your working, and then state what happens to that threshold if the cooperative grows to 1,000 members.

λ = (F/N) / [(q − s)·X − c]. Numerator: £3,500 / 200 = £17.50 per member per year. Denominator: (0.30 − 0.05) × £7,575 = £1,893.75, less £330 = £1,563.75. λ = 17.50 / 1,563.75 = 0.01119 disputes per member per year — 1.12 per hundred members per year, or 2.24 disputes a year in a house of 200. At 1,000 members: λ = 3.50 / 1,563.75 = 0.00224 per member per year — which is still 2.24 disputes a year. The full mark is for seeing why: F is fixed, so the threshold in whole disputes does not move with size at all. The point of the question is that there is no organisation of humans below two and a quarter disputes a year, which means the ladder is not a scale decision.*

10. The Campbell review reports d = 0.155 for restorative conferencing on repeat offending. Assume a two-year reconviction base rate of 50 per cent. Convert the effect into a number needed to treat, then price one prevented repeat offender at £1,500 a conference and compare it to a £48,000 prison place.

Convert d to a log odds ratio by multiplying by π/√3 = 1.8138: exp(0.155 × 1.8138) = exp(0.2811) = 1.3246. Control odds at 50 per cent are 1.0000; treated odds are 1 / 1.3246 = 0.7549; treated rate is 0.7549 / 1.7549 = 43.0%. Absolute reduction 7.0 percentage points; NNT = 1 / 0.070 = 14.3. Cost: 14.3 × £1,500 = £21,482 per repeat offender prevented, against £48,000 for one prison place — 2.23 : 1 on avoided custody alone. Full marks require noticing that this is well below the published 3:1 to 14:1 range, and saying why: those figures count police, court, victim and subsequent-offence costs over a longer horizon. A candidate who reproduces the published ratio without rebuilding it has not done the question.

REFLECTION QUESTIONS — eight, private

Think of a disagreement you are currently carrying rather than raising. What would have to exist — not change in you, exist in the world — for you to raise it this month?

Recall a time somebody gave you a graduated response rather than the maximum one. What did that make possible that the maximum would have foreclosed?

Where have you used exclusion — ended a relationship, left a group, stopped replying — because it was the only rung you could reach? What was the third rung you did not have?

Write down the last three times you complied with something. For each, was it an agreement you made or an instruction you were given? What does your own pattern say about the 70.6 and the 33.8?

When have you refused to take part in a process, and been right to? What would a well-designed system have needed to offer you for your refusal to be heard as a signal rather than as bad faith?

Think of a forum you belong to that meets on a schedule regardless of whether there is business. What does its mere existence let you do that you would not otherwise do?

Who in your life or work has apologised to you properly? What did they do, in what order, and what did it cost them? Write the sequence down — it is the content of rung three.

If somebody published one page a year about how you handle disagreement — count, rungs used, median days to resolution — what would be on it, and which line would you least want to see?

ESSAY PROMPTS — five

Each is arguable from more than one side. Each requires at least one source the chapter cites and at least one it does not.

1. The fallback and the free process. The chapter argues that a terminal rung nobody uses is what makes the rungs beneath it work, following Mnookin and Kornhauser on bargaining in the shadow of the law. Argue either that a credible coercive fallback is a precondition of genuine voluntary repair, or that its presence contaminates the consent it is meant to protect — that a party who agrees because the alternative is worse has not agreed at all. Use Mnookin and Kornhauser, and one source on coercion, consent or procedural justice that the chapter does not cite.

2. Is a small effect size an argument for restorative justice or against it? The pooled randomised evidence gives d = 0.155 — about a 7-point absolute reduction and a number needed to treat of 14.3, with named arms where the effect was null or reversed. Take a position on whether this is a strong result for a low-cost intervention or a weak one dressed in favourable framing, and say what evidence would change your mind. Engage Strang et al. (2013) directly, and at least one published critique or replication the chapter does not cite.

3. Ostrom's principles outside the commons. The chapter applies principles four, five and six to firms, cooperatives and teams — none of which are common-pool resources in Ostrom's sense. Argue whether this transfer is legitimate or whether it strips the principles of the conditions that made them work: a bounded resource, a bounded membership, and appropriators who cannot leave. Use Ostrom (1990) and Cox et al. (2010), and one source on organisational governance or institutional transplant that the chapter does not cite.

4. The threshold that nobody is below. The chapter computes a break-even of about 2.24 disputes a year and concludes that every organisation is above it — therefore the repair ladder is always cheaper than exclusion. Write the strongest attack on that conclusion, then the strongest defence. The attack should go at q − s, the share of disputes the ladder diverts, which is the chapter's least evidenced input. Use Boushey and Glynn on replacement cost, and one empirical source on employee turnover, grievance outcomes or workplace dispute rates that the chapter does not cite.

5. Speed, and who pays for it. Arbitration is about eighteen months faster than a trial and the employee win rate is roughly 1.70 times higher at trial. Argue either that efficiency in dispute resolution is a public good whose distributional effects should be fixed by other means, or that a mechanism whose efficiency is inseparable from its asymmetry should not be described as efficient at all. Use Colvin (2011), and one source on access to justice, arbitration clauses or court funding that the chapter does not cite.