Reading What Isn't Written── Building a bridge to the principles outside the rules: how cognitive science helps answer "But it's not in the creation guide, is it?"

Every time a materials review hits the question "But it isn't written anywhere, is it?", my talk of principles falls flat. When I learned that experts and novices differ in the very structure of their knowledge, I realized half the responsibility for the failure was mine. Part 22 asks how to build a bridge to someone the message isn't reaching.

01"Where does it say that?"

I was explaining why I had sent a promotional material back. The author opened the table of contents of the creation guide (= the rulebook for making promotional materials, compiled by the pharma industry association) and asked, "Which clause forbids this expression?" There was no edge in the voice. If anything, this was a diligent person: the guide bristled with sticky notes, and the list of prohibited items was nearly memorized. Still, for the point I had raised, the answer was "I can't find a matching clause."

I pointed to the "Basic Principles" page in the opening chapter. The author nodded, but kept flipping through the pages of specific prohibitions at the back. I showed the chapter of principles; the other person searched the chapter of individual clauses. The conversation was polite, yet our eyes never met. What were those parallel lines about? That is what I want to think through here.

The rules themselves offer a clue. First there are the Standards for Fair Advertising, which implement the Pharmaceuticals and Medical Devices Act (= the law regulating drug advertising in Japan). Then the "three conditions of advertising" (= intent to attract customers, naming the product, and being recognizable by the general public), which decide whether something counts as an ad. Then the Guidelines for Sales Information Provision Activities (= the code of conduct for company information provision set by the health ministry), and the industry association's creation guide. Every one of them puts a "Basic Principles" or "Purpose" chapter first. The individual prohibitions come after. In other words, the writers wanted the principles read before the prohibition list. The order itself is a design.

In practice, the order gets reversed. The busier the author, the more likely they skip the opening chapter and go straight to the list of allowed and forbidden words. It is faster, and the answers are black and white. There is no bad faith in it. But if you cannot read the principles that live outside the clauses, you are stranded every time a new expression appears that is not on the list. The parallel lines between me and that author were not a difference of character or effort. They looked like a difference in something more structural. What that is, the next section considers with the help of a psychology experiment.

02People who read the surface, people who read the structure

In 1981, the cognitive scientist Michelene Chi and her colleagues (Chi studied how people become expert) gave physics experts and novices the same deck of problem cards and asked them to sort "similar problems" into piles. The results split cleanly. Novices made piles by what the pictures showed: "pulley problems," "inclined-plane problems." Experts made piles by the law used to solve them: "conservation of energy problems," "Newton's second law problems." Looking at the same cards, they were seeing different things. Chi and her colleagues summed it up this way: novices classify problems by surface features, experts by deep structure (= the principle hidden beneath the surface).

I think something very similar was happening in that review room. Looking at the material, I read it as "the way the graph is cropped, combined with that headline, adds up to a guarantee of efficacy (= promising that the drug will definitely work — something the advertising standards forbid)." The author read it as "none of the words used are on the prohibited list." I was sorting by composition; the other person was sorting by wording. We were both looking seriously at the same material. Only our piles were built differently.

The important point is that Chi's experiment does not blame the novices. Sorting by surface is not laziness; it is a stage in knowledge that is still growing. The experts, too, once sorted by pulleys and inclined planes. Only after solving hundreds of problems did the law behind the picture become visible first. So dismissing the author who asks "where is the clause?" as under-studied is a mistake. That question is an honest sign that the knowledge is still at the level of wording.

Reframed this way, the reviewer's job changes too. Pointing at the opening chapter and saying "please read the Basic Principles" will not change how the piles are built. Experts did not come to see the laws by reciting them; they got there by repeatedly connecting concrete problems to the laws. Then what I should do, for the single card that is the material in front of us, is put both layers into words — the wording layer and the composition layer — and go back and forth between them out loud. The ability to read outside the clauses grows only from this card-by-card back-and-forth, not from lectures.

03Why talk of principles sounds like talk from far away

In a meeting room, I once said it: "We are part of an industry that deals with human life." The moment I did, I could see the light drain from the eyes of the person across the table. The gaze dropped to the material at hand; the mind, most likely, went back to next week's deadline and the face of the boss who would have to be told about the revisions. My words were not wrong. But they were not landing. For a long time I blamed that flicker of the eyes on "low awareness."

Psychology has construal level theory (= the idea that people change the grain of how they think about something depending on its psychological distance). It was proposed by two researchers, Trope and Liberman. Distant things look abstract, in broad strokes. Near things look concrete, in fine detail. A trip a year away is "time to find myself again"; a trip tomorrow is "did I pack the charger?" When the distance changes, the very mode of thinking switches.

For that person, the material and the deadline were "near." Every phrase, the axes of the graph, the boss's sign-off — all of it visible in concrete grains. The principles of the industry and the responsibility to patients were right, but "far." Distant things are processed only at the abstract level. In other words, I was trying to judge a single sentence, in front of someone seeing the world in concrete grains, using abstract words. Inside that person's head, my words had nowhere to land.

• Far words: "Be conscious that we serve human life." True, but it does not connect to tomorrow's deadline.
• Near words: "If a doctor reads this sentence and overestimates this drug, what happens?" The same principle, restated at the grain of the material on the table.

When a sound argument comes across as a sermon, it is not because the content is wrong. It is because the distance is off. Judge a near problem in far words, and people feel: "I know, but I can't deal with that right now." There is no need to abandon the principle. Bring it down to the grain size the other person is looking at. That is not a compromise. These days I think of it as translation work.

04Half of every failure to communicate is mine

As my years in review grew, a feeling grew inside me: "Anyone who reads the opening chapter should obviously grasp the intent of this regulation." More than once, asked for the basis of a comment, I answered with just the name of a notification and left it there. I mistook the other person's silence for agreement.

There is a famous experiment. Tap out the rhythm of a well-known song on a desk with your finger. Tappers estimate that half their listeners will name the tune. In reality, about one song in forty is guessed. Inside the tapper's head the song is playing; all that reaches the listener is a dry tap-tap. The experiment was run by the psychologist Elizabeth Newton at Stanford in 1990. The gap between the tappers' prediction and reality was roughly twentyfold.

The name itself, the "curse of knowledge" (= once you know something, you can no longer recall how things looked before you knew it), had been coined the year before, in 1989, by the economist Camerer and colleagues in a different context. The linguist Steven Pinker wrote that this is the single biggest reason unreadable prose gets written. In the writer's head the context is playing. The reader hears only tap-tap. What makes the curse frightening is that the person under it cannot feel it.

On the days I inwardly fumed, "why can't they see this?", I thought I was testing the other person, when in fact I was the one under the curse. In my head, the background of the notification and the past violation cases were all playing. What I handed over was only the tap-tap of a notification number. A failure to communicate is, before it is a lack of study on the side that doesn't know, a limit of imagination on the side that does. After pulling the responsibility back to where it belongs, my review comments got a little longer. I began writing the basis as if explaining it to the self who did not yet know.

Knowledge can be added; ignorance cannot be recovered. So when something fails to get through, the first thing to question is not the other person's comprehension but my own tapping. The curse cannot be lifted. Knowing that it cannot be lifted is the only antidote.

05Why "not written" does not mean "free"

At a review meeting, a sales-side representative once said to me, "Nowhere does it say this expression is forbidden, does it?" True enough. No clause in the creation guide flatly forbade that exact wording. I was quiet for a moment, then answered, "Maybe it isn't written not because it's allowed, but because the people who wrote the rules couldn't imagine this expression." The representative looked unconvinced, but this distinction goes to the root of review.

There is a well-worn parable. The person who posted "No vehicles in the park" at the gate was picturing cars and bicycles. Electric kick scooters did not exist for them. So may an electric scooter enter? Going by the letters on the sign alone: "it isn't written." But measured against what the sign was trying to protect — the safety of people walking — the answer is plain. The legal philosopher H. L. A. Hart (a 20th-century British legal theorist) called this the "open texture" of rules (= the property that any rule's language always leaves an edge its makers could not foresee). The blank space does not exist because the rule was badly made. Writing down every future in words is something human beings simply cannot do.

So the blank space in the clauses is not a "free zone." It is the place where you must go back to the purpose of the rule and judge with your own head. The creation guide puts "Basic Principles" first in its opening chapter not as decoration, but to hand you, in advance, the point of origin to return to when the clauses fall silent. When you wander onto a road the map doesn't show, what you rely on is the direction of your destination.

Choosing "it isn't written, so let it through" here is not a rule violation. No one may ever call you out for it. Even so, I draw the line — as a matter of conscience. I could imagine the possibility that a doctor reading this material would make a judgment that harms a patient. Whoever can imagine it carries responsibility in proportion to what was imagined. Using the silence of the clauses as a shield to let go of that responsibility is something I do not want to permit myself. That is the line I will not give up, in the places where nothing is written.

06Build scaffolding one step above where they stand

Then how do you hand someone the thinking in that opening chapter, the "basic principles" behind the clauses? When I was young, I failed at it. To a representative fixated on the letter of the clauses, I abruptly preached, "let's think from the purpose of the regulation." The person went blank, and the conversation did not move an inch. Now I understand: it was like telling someone standing on the first step of a staircase to jump to the second floor in a single bound.

The developmental psychologist Vygotsky (a 20th-century Russian psychologist) held that a child's learning has two regions: what they can do alone, and what they cannot do alone but can do with help. The latter he called the "zone of proximal development" (= the range one step beyond current reach). Learning happens only inside that band; a task too far away simply bounces off. Later, Wood, Bruner, and Ross (1976) named the help that supports that one step ahead "scaffolding" (= a temporary support, like scaffolding at a building site, removed once the person can manage). The support is not a permanent crutch. The premise is that you take it down once the crossing is made.

Translated into a review conversation, it goes like this. Where the other person stands right now is "how to read the clauses." Then the question I offer needs to be only half a step above that.

• "What do you think this clause was put there to protect?" Anyone who can read the clause can reach this question.
• If the answer is "to prevent inflated expectations, I suppose," the next half step: "Then wouldn't this expression create exactly the thing it wants to prevent?"
• By this point, even on a case the clauses never mention, the person begins to walk the path of judgment on their own feet.

Recently, a representative I had talked through in exactly this order came to me, months later and about a different material, saying: "This isn't in the clauses, but thinking from the purpose, something about it bothers me." The scaffolding had already come down. Abstract principles do not transfer through sermons. You build them a question, one plank at a time, half a step above the stair where the other person stands. It looks like the long way around, and it is the only route by which a principle gets handed over.

07Show not one concrete example, but two side by side

There was a time when I explained the reason for a rejection and the other person just went quiet. A young author. I pointed at the passage, spoke about the intent of the provision, and explained carefully — or so I thought — why the expression went too far. He nodded and left, and with the next material brought me the same problem in a different outfit. My explanation had conveyed only the surface, "fix this expression," and failed to hand over the structure underneath.

These days I do it differently. I do not show only the material in question. I place a second case beside it — one that looks completely different — and ask the person to compare: "What do these two have in common?" For example: a material whose graph truncates the vertical axis to exaggerate a difference, and a material that shrinks the side-effect information into small type at the end. On the surface they are nothing alike, but both share the same skeleton: they tilt the reader's impression in a more favorable direction than the data supports. Shown one example, people remember the surface. Made to compare two, the shared skeleton is what rises into view.

This is not my invention. The cognitive scientist Dedre Gentner studied what happens in the mind when people reason by analogy (= thinking about one thing by likening it to another). What gets carried over, she argued, is not surface resemblance but the shape of the relations among the parts. This is called structure-mapping theory (= an account of analogy as the overlaying of two cases' "relational skeletons"). And in 2003, Gentner, Loewenstein, and Thompson tested it in corporate negotiation training. One group studied negotiation cases one at a time; another was made to compare two side by side. The comparison group extracted the underlying principle better and actually used it in real negotiations later. This way of teaching — imprinting a principle through the comparison of paired cases — is called analogical encoding (= a way of learning that fixes a principle in memory through comparison). It is not magic. It is just a plain, reliable finding: transfer (= being able to use what you learned in a different situation) happens more readily than with single examples.

• Show one example → what sticks is the surface: "this expression is bad."
• Have them compare two → what sticks is the structure: "don't tilt the impression beyond what the data supports."
• Once the structure sticks, they can spot the third case (a brand-new material no one has seen) on their own.

When teaching what the clauses do not say, we are tempted to speak in abstractions: intent, spirit, principle. But abstractions often make people feel they understand in the moment and give them nothing to hold onto in the field. So I have half given up on preaching the abstract. Instead, I build a bridge between one concrete case and another. I set two examples side by side and have the person say the shared skeleton in their own words. A principle you extracted yourself lasts longer than one handed to you. Teaching "what isn't written" turns out to be the work of letting the unwritten thing rise up from between two written ones.