The GrammarofAttention.

There is an old mistake in thinking about media. We talk as though the medium were the newspaper, the screen or the feed. But those are only carriers. The medium that every message actually has to pass through is smaller and older than any of them. It is a few seconds of a human being's attention.

Attention has properties. It is finite, and it cannot be stored. It does not grow when the supply of content grows. A person alive today has the same waking hours as a person in the age of the printing press, and the same narrow channel behind the eyes. Everything has changed around that channel, and the channel has not changed at all.

Herbert Simon saw the consequence early. A wealth of information creates a poverty of attention. That observation is more than half a century old, and it has only become truer. We now live inside its full implication: the cost of producing a message has collapsed towards zero, while the cost of being heard has climbed.

The cost of producing a message has collapsed. The cost of being heard has not.

I have spent most of my working life on the receiving end of this problem, helping people, parties, institutions and peoples to be heard. Over the years, I have stopped thinking of it as a problem of volume and started to think of it as a problem of language. Any message that has ever moved a public has obeyed certain rules. Those who were good at it usually could not state them. They had an ear, in the way that a native speaker has an ear. They knew something was wrong with a sentence without being able to tell you why.

So here is the theory I want to set out. Attention has a grammar. A campaign, a brand, a speech or a national story is a sentence spoken to a public over time. It has a subject, a verb and an object. It has a tense and a mood. It obeys rules of repetition and of agreement. When a message fails, it is very often because it is ungrammatical, in a sense that can be precisely described.

Why call it grammar, and not simply strategy? Because grammar makes a stronger claim. A strategy is something you choose. A grammar is something you are already inside. You can break it, but you cannot escape it, and the listener will notice the moment you do.

Take any public communication that you have found powerful, and try reducing it to a single sentence. You will nearly always find five working parts.

The subject is who is speaking, and it is the most neglected part of the whole structure. Who is the we? When an institution says we believe, what body is doing the believing? Most corporate language has no real subject at all. It speaks in the passive voice precisely because nobody wants to be the subject. Mistakes were made. Measures have been taken. Passive construction is a way of hiding, and publics have an excellent ear for it. A subject is a place where a person stands and accepts the consequences.

The verb is the action. I would guess that most campaigns fail here. They have twelve verbs. We innovate, connect, empower and transform. A sentence with twelve verbs says nothing, because a verb is a commitment, and twelve commitments amount to none. The discipline is severe: one verb. What does this message do in the world? Does it warn, or invite, or defend, or accuse, or promise? If you cannot answer in one word, the public will not be able to either.

The object is what the action falls on. It answers the question of who or what will be different. Vague objects produce vague politics. A strong message has an object you could point to.

Then tense, which is the most sophisticated part and the least discussed. Every message lives in a time. A record speaks in the past. A situation speaks in the present. A promise speaks in the future. Those are very different kinds of claim, and they carry different kinds of proof.

Mood completes the sentence. In grammar, mood tells you whether something is stated as a fact, a command, a wish or a possibility. Publics are acutely sensitive to mood. They will forgive a wish that is declared as a wish, and they will never forgive a wish disguised as a fact.

Publics forgive a wish declared as a wish. They never forgive a wish disguised as a fact.

Once you can see these five parts, you can diagnose almost any communication failure in a few minutes. I do this with new clients as a first exercise, and it is rarely comfortable. I ask them to write their message as a single grammatical sentence, with one subject and one verb. The silence that follows is usually the most useful moment of the engagement. An organisation that cannot produce that sentence does not have a communications problem. It has not yet decided what it is.

Let me stay with tense and mood, because I believe more public failures come from here than from anywhere else.

The record, top left, is the strongest ground available: a fact about the past. We built this. It happened. Nobody can take it from you, and it needs no belief, only checking. Yet institutions tend to under-use it, because the past seems dull beside the future.

The promise is a fact claimed about the future, and it is the most dangerous quadrant on the map. Grammatically, it is a contradiction. The future is not yet a fact. Politics has lived here for centuries, and every promise spends a little of the speaker's credibility in advance, in the hope of earning it back. When the bill arrives, it is paid in trust.

The vision sits to its side: a wish about the future, honestly marked as a wish. This is what great speeches do. Their force comes from admitting that the thing does not exist yet. A vision says: this is what we want, and here is why it is worth wanting. It risks nothing false, and so it can be carried for decades.

The bottom left is nostalgia, a wish about the past. It is immensely warm, and it should be handled with care. It tells people that the best days are behind them, and that the task is to go back. Identity movements around the world are full of it. I come from a culture with a very long memory, and I know how powerful and how paralysing it can be.

Now consider the most common mistake. A campaign takes a vision, and in the excitement of launch it slides quietly into the grammar of a promise. We dream of a better city becomes we will deliver a better city. Nothing about the world has changed, but the mood has shifted from wish to fact, and the speaker has signed a contract without noticing.

Institutions make a similar error in reverse. They speak of their record as though it were a wish, with such caution that it sounds doubtful. The result is a strange inversion: the things they could prove sound uncertain, and the things they cannot prove sound sure.

Institutions speak of what they can prove as if it were doubtful, and of what they cannot prove as if it were sure.

A good message moves through the quadrants on purpose. It begins with the record, to earn the right to speak. It declares a vision, marked as one. It makes very few promises, and it keeps them. This order, past before future, is one of the deep rules of the grammar. People will follow you into the future if you have first shown them that you can tell the truth about the past.

This has a practical implication for anyone advising leaders. When you review a speech, check the tenses before you check the adjectives. Mark every sentence as record, situation, vision or promise. Then count the promises. In my experience, that one exercise improves a text more than any amount of polishing.

Now to the part of the grammar that creative people hate most.

Hermann Ebbinghaus measured his own forgetting in the 1880s, and gave us the curve that bears his name: memory falls away steeply and then levels off, and a well-timed repetition lifts it again. That finding is about nonsense syllables and one patient man. But I think the shape holds for publics as well, as a model. A single exposure, however brilliant, decays quickly. A launch is a spike. Almost every institution spends its budget on the spike and then wonders why nothing remains.

Repetition is often regarded as a failure of imagination. Creative teams grow bored of a line within weeks, and they start to vary it. That is where a great deal of money goes to die. The boredom of the author arrives long before the understanding of the audience. By the time you are sick of saying something, a small fraction of your public has heard it for the first time.

The boredom of the author arrives long before the understanding of the audience.

In linguistic terms, repetition is not decoration. It is syntax. It is what tells the listener which part of the sentence is the main clause. A message said once is an aside. A message said a hundred times is a position. A public infers importance from recurrence, and it is quite right to do so, because recurrence is expensive and costly signals are hard to fake.

There is an art in it, of course. The rule is to repeat the sentence and vary the evidence. The thesis stays fixed, while the proof keeps changing: new people, new facts, new places. The audience receives something fresh each time, and it is also the same thing each time. All enduring brands do this. All long political careers do too.

The final stage of repetition is the most satisfying one, and it is the true sign of success: the sentence leaves you. Other people begin to say it, in their own words, without crediting you. That is the moment a message stops being a campaign and becomes part of the language.

Notice how much of that calendar is patient. Stating the sentence is the shortest bar on the chart. Most of the year is proof and repetition. Most organisations reverse those proportions: they spend most of their energy on finding the words, and then they leave the words alone.

Does a grammar survive changes in technology? I believe it does, in the way that human grammar survived the telephone. But the eras are different, and the differences matter.

In the broadcast era, airtime was the scarce resource and the gate was held by editors. A message had to be judged worthy by a professional. This favoured institutions, and it produced a measured, authoritative register. Being ignored was the main risk.

The platform era moved the gate to ranking systems. Reach became the scarce resource, and it was allocated by reaction. Anything that provoked an immediate response travelled; anything that merely deserved one did not. Over about fifteen years, this rewrote public language. It favoured fragments over sentences, and subjects over verbs. Everyone was shouting their identity, and very little action was being described. The main risk was no longer silence. It was to become the object of a crowd's anger.

The era now beginning is different again, and I think we have barely understood it. More and more, a person does not go to a source at all. They ask a machine, and the machine answers. The unit of communication is no longer the programme or the post. It is the answer. The gatekeeper is a model, and increasingly an agent acting on someone's behalf.

Consider what this does to the grammar. A model reads everything, so volume no longer has the same effect. It compresses, so only what survives compression will reach the person. It looks for structure, for clear claims with clear sources. A message with a clean subject, one verb and evidence in the past tense survives this process. A message built from adjectives and mood lighting dissolves.

A model compresses. Only what survives compression reaches the person.

This is a remarkable reversal. After fifteen years in which outrage and fragments were rewarded, the new gatekeeper may reward something closer to good old-fashioned prose. The sentence is back. The main risk now is no longer being ignored or being attacked. It is being summarised wrongly: reduced by a machine to a version of yourself that you did not write.

I should be careful here, since nobody yet knows how this settles. These systems can be gamed, and they will be. But the direction seems clear enough to act upon. If you want to exist in the answers of the next decade, you need to have stated plainly, somewhere permanent, who you are, what you do and what you have done. That means sentences. It means records. In a strange way, it means grammar.

Trust becomes the scarce resource here, since a machine will weigh sources, and a person will decide which machine to believe. Institutions that have spent years saying one true thing clearly will have an asset that cannot be manufactured quickly.

A theory is only worth having if it changes what you do on Monday. So here is the practice, in the order I use it.

Start with the subject. Write down exactly who is speaking. If the honest answer is a committee, then find a person. Institutions cannot be trusted. People representing institutions can.

Choose one verb. Put every candidate on a wall and remove them until one remains. It will hurt. Everything you remove was somebody's priority.

Name the object, in terms you could point to. If you cannot draw it, it is too abstract.

Then set the tense. Decide what share of what you say will be record, and what share vision. Be stingy with promises. I would suggest that you write down every promise your organisation is currently making, on one page. Most leaders have never seen that page, and it often frightens them.

Check the mood. Go through your materials and mark every wish that is dressed as a fact.

Write the one sentence. A real sentence, of the kind a person would say aloud. Then build the year around it, and do not touch it, however bored you become.

Finally, write it down somewhere permanent. Publish the plain version. Give the machines and the historians something solid.

I want to end with a warning, because a grammar can be used by anyone. Everything in this essay can be used to make a lie more effective. A false sentence with a strong subject, one verb and steady repetition is the basic mechanism of propaganda, and the last century is full of examples. The grammar is not a moral system. It tells you how sentences work, not which ones deserve to be said.

So the last rule stands outside the grammar, and it is the only one I would call absolute. The sentence has to be true. It has to be true of the past and honest about the future. I do not say this only out of virtue. I say it because attention can be seized by a lie, but trust cannot be held by one. In the long run, the only message that survives endless repetition is one that the world keeps confirming.

Attention is the last scarce medium. Those who speak into it are taking something from another person that can never be given back: a few seconds of a finite life. I think that deserves a little grammar.