The slicker the material looks, the sooner you check whether it can be traced to its source. Work you cannot point back to the original number or sentence does not move up, however clever its delivery. Grounding sets the ceiling of the level.

A Building With a Ceiling — the Source Sets the Roofline

When you build a house, you cannot place a tall structure on soft ground. How many floors you can stack depends on how firm the earth is. Grading material work uses the same idea. "Grounding" means being able to trace a sentence in the material back, with your finger, to the number or original text in the underlying paper or package insert. When this ability to return is weak, the work cannot reach the higher levels, however skillful its delivery. We call this the "grounding ceiling."

Levels come in four steps. L1 = does only this one assigned job as told. L2 = has learned the pattern and can reproduce it. L3 = understands why and can apply it. L4 = designs the whole system and sets the standard. The grounding-ceiling rule is simple. Work that cannot be traced to its source stays at L1, or is held for review, no matter how strong the delivery. Without a foundation, you do not build the second floor.

The Origin Label on a Dish — Taste Cannot Rescue a False Provenance

A cook serves a plate. The plating is beautiful, the explanation fluent. But if they cannot answer "where was this fish caught," the plate must not go out. Good taste cannot make up for a provenance that cannot be stated. Material is the same: skill in appeal (the craft of conveying value) cannot fill a hole in source grounding. This is the line that separates the floor of the evaluation (the necessary condition).

Recall this series' two axes. The horizontal axis is "fidelity to fact (grounding)," the vertical axis is "design power to deliver (reach)." Of the four quadrants, the most dangerous is low fidelity x high design — work that persuades but drifts from fact. Because it looks impressive, the reader believes it. So in grading we first check whether the work stands on the floor of grounding, and only then measure the overall score for excellence. The order must not be reversed.

The Trap of "Proving a Negative" — the Assertion Case

To write "no abnormality" on a health check, you need the basis of having tested and confirmed it. Writing "there is no abnormality" without testing is not fact but wish. The same risky assertion happens in making material.

In a reported case, a creator asserted in a document that "there is no increase in mortality risk" without sufficient basis. When checked, the person answered, "because it is not clear at present, I wrote it this way." Not clear — that is, no data — had been swapped for the word "none." In another case, an important potential risk was emphasized only as "a low level of risk is expected." In both, returning to the original source finds no basis for saying so. They cannot be grounded.

Why does this happen? Four psychologies work behind it. In this assertion case, "motivated reasoning" is central. The conclusion one wants to sell comes first, and the reading of the data is pulled toward it. The person does not feel they are lying. Over this layers the "sin of omission" — not honestly writing that there is no basis, staying silent until asked. The table below lines up the case, the psychology, and the power that stops it.

Reported deviationUnderlying psychologyPower that stops itSeen through grounding
Asserting "no increase in risk" without basisMotivated reasoning (conclusion first)Source-grounding powerCannot return to data = g0
Writing only "low risk is expected" for a potential riskSin of omission (not stating the inconvenient)Self-audit powerNo such statement in the source = g0
Explaining afterward "because it is not clear now"Externalizing responsibility (blaming the situation)Misreading-prediction powerUnknown swapped for none = g0

g=0 Does Not Raise the Level — Zero Multiplies Everything to Zero

Suppose we score the degree of grounding from 0 to 3. g3 = returns accurately to the source. g2 = mostly returns but partly vague. g1 = there is a trace of trying but it does not match. g0 = cannot return, or the claim is not in the source. Recall multiplication. However large the number, multiply by zero and the answer is zero. Work grounded at 0 does not rise in overall level even if appeal design is full marks.

This may look like too harsh a rule. But there is a reason. As in the earlier assertion case, an ungrounded sentence directly distorts the reader's judgment. If a doctor believes that sentence and chooses a drug, harm reaches the patient. So grounding is placed not as a bonus item but as a gate you must pass to go further (a non-compensable gate — a checkpoint no other power can make up for). The table below shows how the delivery score and the grounding score affect the final level.

Grounding (g)Appeal designFinal levelReasoning
g3 highhighL3-L4 candidateStands on the floor, with excellence too
g3 highlowaround L2Correct but hard to deliver. Room to grow upward
g0 cannot returnhighL1 / heldMost dangerous. Persuasive misreading
g0 cannot returnlowL1 / out of gradingNo foundation

The Evaluator's Steps — Try to Return First, Then Score

When reading a proof sheet (galley for pre-print checking), a veteran does not start by praising the prose. They first match numbers and proper names against the manuscript one by one. The material evaluator goes in the same order. The first thing to do is pick three prominent claims and actually try to trace them back to the source. If they return, give the grounding score, then add the delivery score on top. If they do not return, stop there once.

What matters here is not treating the creator as an "evil person." The assertion and the omission are merely the circuit an ordinary creator falls into under the pressure of deadlines and sales. The aim of evaluation is not condemnation but enabling the creator to watch their own circuit — has the conclusion come first, am I writing "none" for the unknown. So the evaluation sheet must always leave, in words, "where the grounding broke." So that next time the same situation comes, the person can stop there. Grounding is the ceiling of the level and, at the same time, the starting point of development.

Measuring Skill from Work and Behavior ── Map of all 10 episodes

  1. Vol. 1: Measure by the Materials Actually Made, Not by Impressions or Self-Report ── A material maker's skill is measured from the actual deliverables and observable conduct, not from self-report or others' impressions.
  2. Vol. 2: Tracing the Brief, the Choices, and the Result — In Order ── Read a creator's skill from evidence by walking through one real project in order: the brief, the thinking, the actions, and the result.
  3. Vol. 3: Reading "Faithfulness to the Facts" and "Craft of Delivery" Out of the Work Itself ── This installment shows how to recode a finished piece into two axes — faithfulness to the facts and the craft of getting it across — by reading concrete clues, not impressions.
  4. Vol. 4: The Rules That Keep Measurement Honest ── Six ground rules that keep the evaluator from drifting when measuring an author's real skill.
  5. Vol. 5: Three Rulers: Accuracy, Clarity, and Balance ── Defines three rulers for grading material-making skill and scores each on a four-step scale: accuracy as the floor, clarity as the reach, and balance as the adjustment between too much and too little.
  6. Vol. 6 (this episode): How to Decide the Level — Returning to the Source Sets the Ceiling ── Work that cannot be traced back to its source cannot earn a higher level, however polished it looks. Grounding sets the ceiling.
  7. Vol. 7: What Deliverables Signal Which Level ── An anchor table that reads a creator's level (L1-L4) from visible deliverables and behavior patterns.
  8. Vol. 8: How Far Can We Trust a Judgment? ── How sure a level judgment is depends on how visible the evidence is; less observable skills produce shakier judgments, so we attach a confidence to each verdict.
  9. Vol. 9: Combine More Than Self-Assessment: Add the Reviewer's and Requester's View ── Layering four viewpoints — self, reviewer, requester, and AI — surfaces the deviations of omission that a single pair of eyes cannot see.
  10. Vol. 10 (final): Connecting the Measurement to Pass/Fail and a Development Plan ── The finale links the score to the pass floor and a plan for what to grow next.
In closing

The steps to decide a level do not start from slickness of delivery. First test whether the claim returns to its source; work that cannot return does not move up. Grounding sets the ceiling, and g=0 pulls down the total like a zero in multiplication. This is not for harshness but to stop persuasive misreading — the most dangerous low fidelity x high design — at the gate.

And evaluation must not end in condemnation. If you leave in words where the grounding broke, the creator can stop at the same point next time. Whether one can return to the source is both the ceiling of the level and the starting point of development.

Key Points ── Three to take with you
  1. Grounding sets the ceiling. Work that cannot be traced to its source does not move up, however skillful the delivery. No foundation, no second floor.
  2. g=0 is a zero in multiplication. If grounding is 0, full marks in appeal design still do not raise the total. Place grounding as a mandatory gate, not a bonus.
  3. Evaluation is the start of development. Leave in words where grounding broke, and the creator can stop before the same assertion next time.
Sources & references
  1. Ministry of Health, Labour and Welfare, Compliance and Narcotics Division (commissioned project). Report on the Monitoring of Promotional Information Activities for Prescription Drugs (March 2024 and prior years). Flagged cases are published with company names anonymized; the deviation patterns cited here are generalized from this report.
  2. Ministry of Health, Labour and Welfare. Guidelines on Promotional Information Activities for Prescription Drugs. Scope of appropriate information provision and the principle of citing sources.
  3. Japan Pharmaceutical Manufacturers Association. JPMA Code of Practice. Principles of evidence-based information provision and fairness.
  4. Standards for Fair Advertising of Drugs (related to Articles 66 and 68 of the Pharmaceuticals and Medical Devices Act). Prohibition of exaggerated advertising and of claiming unapproved efficacy; criteria for judging assertion and overstatement.