What did I send — and what did

you receive?

I have something I want you to understand. I choose some words. Perhaps I add a gesture, an image, a tone of voice or an example. I send. You receive. Communication? Hopefully. But I haven't transferred my representation directly into your head. I've produced signals from which you construct your own representation. And those aren't quite the same thing.

Put the receiver back in

We often describe communication as though meaning travels inside the message. I know what I mean. I say it clearly. You receive it. Done. But consider: ‘The man in the red hat’. That may seem like an excellent way to identify someone. Unless the receiver can't distinguish red. The signal hasn't necessarily changed. The receiver has. The same can apply to language, accent, terminology, cultural references, tone, speed, context and assumed knowledge. So instead of asking only: Was the message clear? we can ask: Was it sufficiently clear for this receiver, for this purpose, in these circumstances?

Sending is an act of judgment

Before you communicate, you make choices. Which words? How much detail? How quickly? Which channel? What does this receiver probably already know? What differences do I need to make available? What can safely be left unsaid? You cannot communicate everything. Nor would the receiver have unlimited resources to process it if you could. So effective communication involves resource allocation on both sides. The sender decides what seems worth including. The receiver decides what seems worth attending to. Good communication isn't maximum information. It is enough of the right information, made sufficiently available for the purpose.

Different receivers need different

signals

Different receivers need different signals Imagine explaining the same problem to: • a five-year-old • a colleague • an expert • someone who doesn't share your first language Same subject. Different receivers. Effective communication may require different signals. That isn't necessarily “dumbing down.” It is adaptation. A skilled communicator doesn't merely ask: How can I express this accurately? They also ask: What might help this receiver construct a sufficiently useful representation? Sometimes that means more detail. Sometimes less. Sometimes slower speech. A picture. An example. A gesture. A different word. A question. Style matters when style matters. In some communication, style may be central to what the sender is trying to achieve. In others, it may matter very little. There is no single best way to communicate. The question is whether the choices we make help this receiver construct a sufficiently useful representation for this purpose, in these circumstances.

Shared understanding doesn't

mean identical understanding

Suppose I describe something and you understand me. Does your representation now exactly match mine? Almost certainly not. And usually it doesn't need to. We need enough overlap for what we're trying to do. Ordering lunch requires one degree of shared understanding. Performing a complex task together may require another. So perhaps successful communication doesn't require: MY representation = YOUR representation It requires something more practical: Our representations overlap sufficiently for the current purpose. How much is sufficient depends on the purpose and the consequences of getting it wrong.

Misunderstanding is normal

Different receivers bring different models, experiences, expectations and resolution. So misunderstanding shouldn't surprise us. The important question may not be: How do we prevent every misunderstanding? We probably can't. It may be: How quickly can we notice that our representations have diverged? You think I mean A. I think you've understood A. You've actually constructed B. For a while, neither of us may know there is a problem. That period matters. Because once the difference becomes available: Wait a second — I don't think we're talking about the same thing. we can do something about it.

Communication is shared work

If you don't understand me, whose responsibility is that? Yours? Mine? NMF suggests that question may sometimes be too simple. The sender has work to do. The receiver has work to do. I can try to design signals that work for you. You can question, check, clarify and signal when your representation doesn't seem good enough. And we can both watch for evidence that our representations may have diverged. Effective communication is a shared responsibility between sender and receiver. Not necessarily an equal responsibility. Circumstances matter. Expertise matters. Authority matters. Capability matters. But communication is something we are constructing between us.

“Yes” isn't proof

Do you understand? Yes. That's evidence. But what does it prove? Perhaps the receiver has constructed exactly what the sender intended. Perhaps they've constructed something different that makes perfect sense to them. Perhaps they don't want to admit they didn't understand. Perhaps neither person yet knows there is a difference. Agreement can therefore provide evidence of shared understanding without guaranteeing it. Sometimes a better check is to create an opportunity for the receiver to show their representation. Explain it back. Act on it. Ask a question. Point. Demonstrate. Let the difference, if there is one, become available.

Every communication has two

subjects

When you communicate with me, I may learn something about the subject. But I may also learn something about you. Do you exaggerate? Do you listen? Do you acknowledge uncertainty? Do you adapt when I don't understand? Do you notice when we're talking past one another? Can I rely on what you say? So: Every communication has two subjects: the subject and the sender. Today's communication contributes to my model of you. And that model affects how I receive your signals tomorrow. Communication therefore doesn't only transmit information. It can change the receiver — and the relationship between receivers.

Communication changes what

comes next

Suppose I misunderstand you. You notice. We repair the difference. Something has been learned. Perhaps next time you phrase it differently. Perhaps I recognise the ambiguity sooner. Perhaps we develop a shorthand. Perhaps I become more alert to a difference that previously passed unnoticed. Communication has changed both receivers. And that can make future communication cheaper, faster or more reliable. What once required ten sentences may eventually require: That thing again. A look may become enough. Shared history has become part of both models of normal.

Could we train this?

This is where NMF becomes particularly interesting. Training communication often concentrates on what people should say or do. But perhaps we can also train further upstream. Can senders become better at representing the receiver before choosing their signals? Can receivers become better at noticing weak evidence that shared understanding has been lost? Can we increase resolution for the tiny differences that suggest: Something isn't quite right here. Can teams become better at making differences visible before they become consequential? Can we practise adaptation rather than simply teaching one “correct” way to communicate? We think at least some of this can be trained. And perhaps that changes more than the response. It changes the receiver who encounters the next communication.

Could we communicate better?

Not perfectly. Not by eliminating every misunderstanding. And not by making everyone construct identical representations. Something more modest. Can I make what matters sufficiently available to you? Can I notice when my signal isn't working? Can you make it easier for me to see what you've understood? Can we detect divergence earlier? Can we adapt? Can we build enough shared understanding for what we're trying to do? And can what happens between us make both of us a little better at communicating next time? Sometimes the useful question isn't: Did I say it clearly? It is: What did I send — and what did you receive?