Discussion Guide — Chapter 43

Six prompts, with what to listen for. Each is designed to fail productively — the wrong answers are more informative than the right ones, and the note under each says which wrong answer to expect.

1. "Walk me through what changes when the same molecule moves from a clinic to a gym. Nothing chemical changes. So what does?"

What to listen for. Whether the group reaches the four elements without prompting, and in particular whether they get past "no doctor" to no diagnosis, no indication, and a duty running to somebody else or to nobody. Push on the third: expertise can be found in a book; a fiduciary obligation cannot.

Expect to hear "it's less safe." That is a conclusion, not an analysis, and it is also not quite the chapter's claim. Redirect: the compound is equally safe or unsafe in both places. What changed is every mechanism by which safety would be established, detected, and acted on.

2. "Somebody in a competitive field adopts an enhancement. You do nothing. Have you been harmed?"

What to listen for. This is step 3 and it is the load-bearing step. Strong groups will say yes and then struggle to name the wrongdoer — which is exactly right, and the struggle is the point. Let it run before resolving it.

Expect to hear "no, people out-compete each other all the time." Take this seriously rather than correcting it; it is a real objection, and it is exercise 2 in case study 2. The productive follow-up: is there a difference between being out-competed by someone's effort and by someone's pharmacology, and if so, is the difference about fairness or about the risk you are now being pushed toward?

3. "At step 5, everyone has adopted and the rankings are roughly what they were. Who gained?"

What to listen for. Whether the group reaches dissipation on their own: nobody gained, everybody paid. This is usually the moment the chapter clicks, and the visible reaction in the room is a good diagnostic of whether §43.5 landed.

Expect to hear "but everyone's better now, so society gained." Excellent objection — take it seriously. It is true where the improvement has absolute value (a surgeon who is genuinely more accurate) and false where the improvement only shifts rank. That distinction is the weak/strong grid, and letting the class derive it from this question works better than presenting it.

4. "Suppose the compound does nothing at all. Does the coercion problem go away?"

What to listen for. The realization that steps 3 through 5 run on belief, not on efficacy, and that the false-belief version is worse because the benefit column is empty. This is case study 2's concluding move.

Expect to hear "yes, if it doesn't work then nobody's really at a disadvantage." The counter is that the disadvantage is believed, and belief produces adoption, spending, and risk exposure regardless. Then push further: at step 5 there is no unenhanced comparison group anyone can observe, so the belief has become unfalsifiable from inside. The structure has destroyed its own control arm. That sentence usually lands.

5. "An institution offers an optional program in complete good faith, with a genuine right of refusal and no record kept. Is the consent real?"

What to listen for. Whether the group distinguishes procedural safeguards (which an institution can deliver) from social ones (which it cannot). Peers still see who participated. Reputation is not administrable.

Expect to hear "yes, if it's genuinely optional." Push on what "optional" requires. Then push the other way, because the opposite error is just as common: if a hierarchy makes consent impossible, does that prove too much? Employees consent to many things inside hierarchies. The productive landing is that voluntariness is a matter of degree, that the degree is measurable in principle, and that nobody measures it.

6. "Four of the five conditions in §43.9 could be met by research, policy, or better information. One cannot. Which, and what follows?"

What to listen for. Condition 3 — absence of coercive structure — and the reason: it is a property of the contest, not of the person, the compound, or the seller. Then the harder second half: what follows is that no individual can solve it by being careful, and no amount of safety data touches it.

Expect to hear "so nothing can be done." Push back with the sport case: something was done, at a cost, with an incomplete result. And if you have time, this is where Ostrom belongs — collective action problems are sometimes solved, by specific institutional arrangements, and the honest position is that we do not know whether any of those arrangements apply to a competitive labor market.

End the session here rather than on a resolution. The chapter does not resolve, the part introduction says the social questions are unresolved, and a session that closes on a tidy conclusion teaches the opposite of what the chapter teaches.