Case Study 2 — How an Option Becomes an Expectation

The coercion structure in a non-sport setting, traced step by step

Why this case is abstract on purpose

The temptation with this material is to write a scene: a named profession, a recognizable firm, a character with a decision to make. Resist it, because a scene invites you to argue about the scene. People respond to a specific illustration by disputing whether that industry is really like that, or whether that person had other options, and the structure — which is the actual subject — disappears under the anecdote.

So this case describes a generic positional contest with the properties §43.5 identified as making the coercion argument bite hard. No profession, employer, institution, or individual is named. Nothing here is a claim about any real workplace. The claim is about the structure, and the structure is what transfers.

Read it as you would read a worked physics problem: the numbers are illustrative, the mechanism is the point.

The setting

Consider a competitive field with these five properties.

THE SETTING — five properties, all of them ordinary

  P1  ADVANCEMENT IS POSITIONAL.
      A fixed number of senior positions. Selection is comparative.

  P2  PERFORMANCE IS OBSERVABLE AND COMPARABLE.
      A common metric that everyone can see: throughput, output,
      accuracy, endurance, hours sustained.

  P3  THE DECISIVE WINDOW IS SHORT.
      Selection turns on performance during a defined period. What
      happens outside that window matters much less.

  P4  THERE IS NO TESTING AND NO POLICY.
      Nobody prohibits anything. Nobody asks. Nobody discloses.

  P5  THE PARTICIPANTS ARE EARLY IN THEIR CAREERS.
      Future health is discounted against present selection, which is
      not irrational — it is what a steep discount rate looks like.

  None of these five is unusual. Most competitive fields have at least four.

Now introduce an enhancement. For this exercise it does not matter what it is or whether it works — we will return to that, and the return is the most important move in the case.

The five steps, traced

Step 1 — the contest is positional. Nothing has happened yet. This is the baseline: a field where one person's improvement necessarily lowers someone else's standing, because standing is defined comparatively.

Step 2 — early adoption. A small number of participants adopt. Their reason is straightforward: they want an advantage. They obtain a real one, or believe they do — and at this stage the difference between those two situations is invisible to everyone, including them. Nobody else's behavior has changed.

Note the moral character of this step, because it is the one everyone argues about and it is not where the interesting problem lives. These participants made a free choice to accept a risk in exchange for a benefit. Whatever you think of that choice, it is a choice.

Step 3 — non-adopters decline without doing anything. This is the step people miss, and it is where the harm actually enters. Non-adopters have not changed. Their output is what it was. Their effort is what it was. And their standing has fallen, because standing is relative and the distribution moved. They have been made worse off by other people's private decisions, without anyone acting on them at all.

There is no mechanism here for them to object to. Nobody wronged them. No rule was broken. No policy was violated. The harm is entirely real and entirely unattributable to any act.

Step 4 — defensive adoption. Some non-adopters adopt. Their reason is different from step 2's, and the difference is the hinge of the whole argument.

The step-2 adopter accepted a risk to gain something. The step-4 adopter accepts the same risk to stop losing something they already had. From outside, the two are indistinguishable — same compound, same route, same exposure, same silence. From inside, they are different decisions, and only one of them is well described by the word choice.

This is what the chapter means by structural coercion. Nobody compelled the step-4 adopter. The compulsion, such as it is, is a property of the distribution.

Step 5 — the new baseline. Adoption becomes common. The performance distribution has shifted upward. Relative rank is roughly what it was at step 1 — the same people are ahead, mostly, for the same reasons. The advantage has been competed away.

What has changed is that the entire population now carries the cost, the risk, and the concealment, in exchange for a competitive position they already occupied. And the enhancement has stopped being something a participant chooses and become something a participant is presumed to have. At that point declining is not neutral; it is a decision to underperform relative to a standard everyone else is meeting.

WHAT CHANGED BETWEEN STEP 1 AND STEP 5

                          STEP 1        STEP 5
  Relative rank           baseline      ≈ baseline
  Absolute performance    baseline      higher
  Population risk         none          borne by nearly everyone
  Population cost         none          borne by nearly everyone
  The question facing     "do I want    "can I afford to be the
    a new entrant           an edge?"     one who doesn't?"

  The competitive benefit was dissipated. The risk was not.

Three things this structure is not

It is not a conspiracy. Nobody organized anything. No employer required anything. Every individual decision in the sequence was made independently and defensibly.

It is not a claim that anyone behaved badly. The step-2 adopters did something many people regard as permissible. The step-4 adopters did something most people would regard as understandable. The institution did nothing at all — which is precisely how P4 was specified.

It is not solved by telling people to stop. This is a collective action problem, and the defining property of collective action problems is that everyone is behaving sensibly given what everyone else is doing. Exhortation fails against arithmetic.

The move that makes this uncomfortable

Now return to the question we set aside. Suppose the enhancement does not work.

Suppose it produces a visible, measurable change that participants can see and feel — a change on a scale, a change in a mirror — with no effect whatsoever on the performance metric that determines selection. Chapter 16's mass without function finding is exactly this shape, and §43.3 rates the performance claim ❌ for the relevant compounds.

Trace the five steps again with that assumption. Almost nothing changes.

Step 2 still happens: adopters believe they gained an advantage, and personal experience confirms the belief because the visible effect is real. Step 3 still happens: non-adopters believe they are falling behind, and a belief about the distribution produces the same pressure as an actual shift in it. Step 4 still happens, for the same reason. Step 5 still arrives, with the population carrying the full cost and risk.

The only thing that changes is the benefit column, which is now empty.

This is the most instructive result in the entire chapter and it is worth stating without softening: a coercive structure built on a false belief about efficacy is not less harmful than one built on a true belief. It is more harmful, because the population pays the same price for nothing at all.

It also explains something otherwise puzzling: why belief in the efficacy of these compounds is so resistant to evidence. At step 5, everyone has adopted, and the visible effect is present in everyone. The absence of functional benefit is invisible from inside the system, because there is no unenhanced comparison group anyone can observe. The structure has destroyed its own control arm.

Discussion questions

1. Identify a positional contest you have personally been in — professional, academic, athletic, or otherwise — and map it against P1 to P5. How many properties did it have? Then, without naming anyone, describe whether you observed anything resembling step 3, and how you would have known if you had.

2. The case argues that step 3 is where the harm enters, even though nothing happens in it. Defend that claim against the objection that a person who is merely out-competed has not been harmed at all — people lose contests all the time. What, if anything, distinguishes being out-competed by another person's effort from being out-competed by another person's pharmacology?

3. The step-2 adopter and the step-4 adopter perform identical actions for different reasons. Does the difference in reason change what either of them is owed — by the institution, by each other, by anyone? Argue both sides. Then say whether your answer would change if the two were indistinguishable to any observer, which by construction they are.

4. P4 specifies that the institution does nothing: no policy, no testing, no inquiry. Is doing nothing a neutral act here? Enumerate what the institution gains from the arrangement described in step 5 and what it bears. Then connect your answer to §43.8's separation of risk and benefit.

5. Suppose you could change exactly one of the five properties P1 to P5. Which would you change, and how far would the change propagate through the five steps? Be specific about which step your intervention interrupts — and note that an intervention that acts after step 4 is managing a symptom.

6. † The case concludes that a coercive structure built on a false belief is worse than one built on a true belief, because the benefit column is empty. Reconstruct that argument, then attack it as hard as you can: is there any reading on which the false-belief version is genuinely less bad — for example, because a compound that does nothing may also do less harm? State your final position, and identify explicitly which parts of it depend on the efficacy evidence and which depend on how you weigh risk against autonomy. The separation is the exercise.