A fox invites a stork to dinner and serves thin gruel in a shallow plate.
He laps his clean in a minute. She stands there "fishing away with her beak" and gets "not a morsel in it."1
A week later she returns the invitation. He arrives early and enthusiastic, praising everything before he has tasted anything — "What taste! What chic! And the food—done just to a turn!" — sits down hungry, and smells meat.
It is minced, and it is served in "a long-necked, narrow-mouthed urn."
She eats her fill, easily, with her long bill. His snout is "the wrong shape and size," and he goes home "empty-bellied, tail dragging, ears drooping, as red in the face as a fox who's been caught by a hen."1
That is the entire fable, printed in a marginal column of Law 44, beside the chapter's most confident claims about teaching people lessons.
And the fox learns nothing.
Judge the reflection purely as craft and it is close to perfect.
Same social frame: a dinner invitation between the same two parties. Same offence: food placed in a vessel the guest's anatomy cannot use. Same outcome for the victim: watching the host eat comfortably while getting nothing. Same duration, same setting, same relationship. If the Moral Effect were a matter of accuracy, this would be the chapter's showpiece.
The fox certainly receives it. He is not confused about what has happened — the fable's last line has him red-faced, which is a description of a man who has understood exactly.
And then the story ends. No apology. No restitution. No repair. No change. He goes home.
Compare the fable printed a few pages earlier in the same chapter, where Pilpay's merchant runs a much cruder mirror and gets his money back, an unforced confession, and the relationship intact.2 Two Moral Effect fables, same margin, same chapter, opposite results — and the more precise mirror is the one that produces nothing.
So accuracy is not the variable. Something else is.
Here is what the stork's mirror actually reflects, and it is not what she thinks she is reflecting.
The fox's offence was a choice. He chose a shallow plate. The fable is explicit that he had options and picked the cheap one — "Being habitually mean, He didn't go in for haute cuisine."1 Nothing forced him. He could have served the same gruel in a bowl she could use and lost nothing but a little smugness.
The stork's reply lands on his body. He cannot eat from a narrow-necked urn because his snout is the wrong shape, and there is no version of him that fits down that neck.
Sit with the difference, because it is the whole page. What the fox experiences at that table is I am the wrong shape for this house. What he needed to experience was I did a mean thing to someone and it was worse than I thought.
You cannot repent of a snout.
That is why the perfect mirror teaches nothing while the crude one produces a settlement. Pilpay's merchant reflected a decision — the friend chose to tell an absurd lie, and the owl story pointed straight at the choosing. There was something for the target to stop doing. The stork reflected a constraint, and a constraint offers no action, so the only content the fox can extract from the evening is humiliation with no instruction attached.
The Moral Effect's real success condition is not fidelity. It is whether the reflection lands on something the target could have done otherwise.
La Fontaine states the stork's motive plainly, and the plainness is the point:
"To pay him back for this cruel Practical joke, the stork invited The fox to dinner the following week."1
To pay him back. Not to teach him, not to reach him, not to make him see. The fable names revenge and does not dress it up.
Now set that beside the chapter housing it, where the same act is described as pedagogy: "you teach others a lesson by giving them a taste of their own medicine"; "they realize in the profoundest sense how they hurt or punish others with their unsocial behavior."3
The fable and the frame disagree about what is happening, and — this is the useful part — the fable's description predicts the outcome and the chapter's does not. Read as revenge, the stork's dinner is a complete success: she wanted him to go home hungry and he went home hungry. Read as teaching, it is a total failure.
Which suggests something about the Moral Effect generally that the chapter never entertains. The mechanism delivers reliably on payback and unreliably on instruction, and the two feel identical from the inside while you are planning them, because both involve arranging for someone to experience what they did. The satisfaction arrives on schedule either way. The lesson is the part that may or may not turn up, and there is no moment during the operation when you would notice its absence.
The honest question before running one is therefore not will they get it? It is: if they go home red-faced and unchanged, will I feel this was worth doing? If yes, it is revenge, and it may still be reasonable — but call it that.
Two words in the fable's second line pre-announce the failure, and the chapter walks past them.
"Being habitually mean."
Not the fox was in a bad mood. Not the fox was short of money. A settled disposition, named as the cause, before the dinner has even been served.
A single demonstration cannot move a habit. Whatever produced decades of meanness — temperament, upbringing, whatever a fable means by such a thing — is not going to be dislodged by one evening of being on the receiving end, and La Fontaine has told us so in advance.
Which makes this fable a scope condition for the whole Teacher's Mirror, sitting in the chapter's own margin. The technique is aimed at behaviour and works, if it works at all, on behaviour that is not yet a character. A one-off unkindness, a habit forming, a person who has not yet decided who they are — those are reachable. The habitually mean are the population most likely to earn a mirror and least likely to be altered by one.
And there is an unpleasant corollary. The people who most provoke the Moral Effect are precisely the repeat offenders, because repeat offence is what exhausts your patience with talking. So the technique gets deployed, systematically, against the targets on whom it has the worst odds.
One small detail is doing more work than it appears to.
The fox arrives and "at once began Praising everything: 'What taste! What chic! And the food—done just to a turn!'"1
He praises the meal before eating it. Which means two things.
First, he walks in with no suspicion whatsoever. A week after humiliating someone at his own table, invited back by that person, he arrives "punctually," delighted, expecting a good time. This is the credulity gate the chapter describes elsewhere — the suspicions not raised at the first glance stay suppressed — and it is wide open.
Second, and more useful: he has publicly committed to the meal being excellent before discovering he cannot eat it. Having declared the food done just to a turn, he has no available complaint. He cannot say this vessel is unusable without contradicting the enthusiasm he arrived with.
The praise seals his own exit. He must sit through it, hungry, agreeing that the dinner is wonderful.
That is craft on the stork's part, if she anticipated it, and luck if she did not — the fable does not say. Either way it is the mechanism that turns an inconvenience into humiliation, and humiliation is exactly the output that guarantees nothing will be learned.
Worth stating precisely, because it is an absence and absences have to be scoped.
Within the marginal as printed (l.6541–6543, read in full), the fable ends at the fox's departure. There is no moral, no closing couplet drawing a lesson, no line reporting any change in the fox's conduct, and no further mention of either animal.1
That matters twice over.
It matters for this page's central claim: the failure is not my inference laid over a story that implies success. The story simply stops, with a hungry fox and a fed stork, and offers no reformation to point at.
And it matters for the chapter. Greene has selected, to sit in the margin of a section arguing that mirrored behaviour makes people "realize in the profoundest sense how they hurt or punish others," a fable in which nobody realizes anything and the author declines to claim they did.
⚠ The corpus has now recorded three instances in this build of a marginal fable's own moral being either wrong about its story or absent where the chapter needs one. [n≥3 PATTERN] — the margins are not corroborating the body; they are running a quieter and more accurate account beside it.
Someone has done the thing to you again and you are done talking about it.
They schedule over you. They reply to your careful message with one line, three days late. Whatever it is, you have raised it, they have agreed it was unfortunate, and it has continued.
The mirror is right there and it is very satisfying to plan. Schedule over them. Reply in one line, three days late. Let them sit in it.
Before you do, run the fox's test.
Ask what your mirror will land on: something they chose, or something they are?
If they schedule over you because they decide their time is worth more, that is a choice, and a reflection can point at it. If they do it because they are chaotic and have always been chaotic and their calendar is a disaster in every direction including their own, then your mirror lands on a snout. They will experience being messed about by a disorganised person — which is their whole life — and extract nothing from it except that you are now one of the people messing them about.
Then ask the harder question, which is about you. If they go away stung and behave identically next week, will you be glad you did it?
Answer honestly, because the answer is often yes, and that is worth knowing. It means what you want is payback. Payback is a real want and there is no particular virtue in pretending otherwise — but it should be chosen with its own name on it, because you will spend the same effort and get a different thing than you told yourself you were getting.
If what you actually want is the behaviour changed, the fox says the mirror is the wrong instrument here and the merchant says what the right one looks like: aim at the decision, leave them a way to concede that does not require confessing to a character, and stop talking the moment they see it.
And if you have already decided to run it, do not let them arrive praising the food. The stork's dinner became a humiliation because the fox committed to enjoying it before he understood — which is entertaining, and is exactly the ingredient that guarantees nothing will be learned. Humiliation and instruction do not travel together.
Strongest evidence is the pairing rather than the fable alone. Two Moral Effect fables sit in the same chapter's margins with opposite outcomes, and the one with the more precise mirror fails. That rules out fidelity as the operative variable and forces the question of what else differs — which is the choice/constraint distinction and the exit.
Tension one — the fable's stated motive and the chapter's stated purpose are different things. "To pay him back"1 against "they realize in the profoundest sense how they hurt or punish others."3 The fable's description predicts its own outcome; the chapter's does not. Not resolvable — both are printed, and the chapter never acknowledges the gap.
Tension two — habitually mean names a disposition, and the technique is a single event. The scope condition is stated in the source, in the fable's second line, and the chapter draws nothing from it.
⚠ Absence, scoped: no moral, no reformation, no further mention of either animal within l.6541–6543, read in full. Stated as an absence in the read range and nothing stronger.
🚩 SECONDARY WITHOUT PRIMARY / TRANSLATION DISTORTION. Attributed as "SELECTED FABLES, JEAN DE LA FONTAINE, 1621-1695."4 No translator, no edition, no translation date. The English is a modern rhyming version — "haute cuisine," "What chic!", "a fox who's been caught by a hen" — with idiom and jokes that cannot all be La Fontaine's, and the fox's pre-meal flattery in particular is the kind of detail translators of verse fable add or amplify. [LOW CONFIDENCE] on any specific wording; the structure is stable across versions.
Open question — did the stork intend the humiliation or only the deprivation? The fable gives her a motive (pay him back) and an outcome (he is humiliated) and never says whether the second was the plan. The distinction decides whether this is a case of the technique working as designed or overshooting.
Open question — can a mirror be aimed at a choice when the choice and the constraint produce the same behaviour? The fox chose a shallow plate; he could not use a narrow urn. The stork had no vessel available that would have targeted his meanness rather than his anatomy. Whether such an instrument generally exists, this fable cannot say.
This marginal's real partner is The Merchant and the Iron, printed in the same chapter, illustrating the same effect, and reaching the opposite result. Greene prints both and draws no conclusion from either.
The controlled comparison is unusually clean for this book. Pilpay's merchant reflected a decision and left an exit; La Fontaine's stork reflected a constraint and left none. The merchant got restitution, an unforced apology and a surviving friendship. The stork got a hungry fox. Both mirrors were competently executed; only one was aimed at something the target could have done differently.
Against The Moral Effect itself, this fable is the counter-case the definition needs and does not have. The effect is defined by its intended output — "you teach others a lesson" — with no account of when the teaching fails, and the chapter's Reversal slot, which is where a limit would normally live, is occupied by a warning about a different phenomenon entirely.
And against Ivan and the Tartar Czar, the Observance the effect was built on, the fox supplies the missing half of the recognition problem. Ivan's mirror failed on everyone not "old enough to remember" the original — an audience limit. The fox remembers his own dinner perfectly and still learns nothing, which shows recognition is necessary and nowhere near sufficient. He sees it, and seeing it is precisely what makes him red in the face.
Toxic Shame vs Healthy Shame — two children spill water. One flushes and is playing again in three minutes; the feeling was about the event. The other's flush does not pass, because what rose in her was not a feeling attached to a spill but "a verdict about herself" — evidence of something already true.
That distinction is the fox's dinner, and it explains a failure the fable can only report.
The stork intended, presumably, the first kind: a bad evening, attached to an event, absorbed and learned from. What she delivered was the second. You are the wrong shape for this table is a verdict, not a feeling about a thing that happened — and the fable's closing image is a verdict landing: not annoyed, not thoughtful, red in the face, the whole body reporting exposure.
What the pairing produces is a mechanism for the choice/constraint rule that this page arrived at structurally. A reflection aimed at conduct produces a feeling about an act, and feelings about acts point at what to do differently — they come with an instruction attached. A reflection aimed at an unalterable attribute produces a statement about the self, and statements about the self come with nothing attached but concealment. The shame page's second child does not learn to carry cups more carefully; she learns that she is clumsy, and there is no next action in that.
Which reverses the intuition that a sharper mirror teaches better. Precision aimed at conduct sharpens the lesson. Precision aimed at a person sharpens the verdict — and the fox's evening was very precise indeed.
Criminal Justice, Neurobiology, and the Limits of Punishment — the deterrence assumption stated and then examined: that the threat of punishment inhibits the harmful impulse, and that the convicted person internalises the experience and refrains afterwards. The page's finding is that this is "incomplete at best, contradicted by evidence at worst."
That is the Moral Effect's core claim, at scale, with data.
Greene's version and the retributive version rest on identical reasoning: the offender must be made to feel the consequence, and the feeling will convert into altered future conduct. Both treat the delivery of the experience as the whole of the intervention — arrange for them to undergo it, and the change follows automatically.
The fox is that assumption failing in miniature and the criminal-justice literature is it failing at population scale, and the two together do something neither does alone: they identify the shared unexamined step. In both cases the experience is delivered successfully. The fox definitely goes hungry; the sentence is definitely served. What is assumed and never demonstrated is the conversion — that undergoing a consequence produces a revised model of one's own conduct, rather than simply producing a person who has been hurt.
And the fable supplies, in three words, the variable the institutional literature spends far more effort establishing: habitually mean. Deterrence has its worst record precisely on entrenched, repeated behaviour, and its best on the marginal case that has not yet hardened — which is the same population split this fable draws between a fox who has been like this for years and a friend who told one lie about a rat.
So the handshake yields a caution with unusual reach. Every institution built on the Moral Effect's logic — sentencing, school discipline, the performance-management conversation designed to make someone feel the impact — inherits the fox's problem: it is easy to verify that the experience was delivered, and nearly impossible to verify that anything converted. The measurable part is the punishment. The part that was supposed to matter is not measured at all, which is why systems built this way persist through decades of not working.
Sharpest implication. The Moral Effect fails when the reflection lands on something the target could not have done otherwise — and it fails harder the more precise it is, because precision aimed at an attribute rather than an act converts a lesson into a verdict about the self, and verdicts come with no next action attached. The fox understood everything and changed nothing; his understanding is exactly what made him red in the face. Which means the technique's satisfying part and its useful part come apart cleanly: payback is delivered on schedule every time, instruction is contingent on a conversion nobody checks, and both feel the same while you are planning them.
Generative questions.
If humiliation and instruction cannot travel together, is there any version of giving someone a taste of their own medicine that reliably avoids the first — or is the sting the whole reason anyone reaches for it?
The technique is worst-suited to entrenched behaviour and most provoked by it. What would you use instead on someone whose conduct has already become a character?
The measurable part of a punishment is the delivery and the part that matters is the conversion. Where else does a system survive on the strength of the half it can see?