The Essence of Media / Attention Economics

The pipe that isn’t neutral

The folk theory of media is that a medium is a pipe: content goes in one end and comes out the other unchanged, and all that matters is the content. On this view television, print, and the feed are interchangeable conduits, and to evaluate them you evaluate what flows through.

This is wrong, and the previous essay’s distortion is the proof. A medium is not a pipe but a filter with a grain — it makes some things easy to transmit and others nearly impossible, and over time it selects, relentlessly, for whatever fits its grain. The content that survives is not a random sample of what entered. It is the subset the medium rewards. To understand a medium, study its grain, not its catalogue.

The medium selects

Every medium has a shape that favours some messages and punishes others. Print rewards the sequential, the argued, the slow — you can hold a complex chain of reasoning across pages. A printed lie has to survive re-reading. The broadcast image rewards the vivid and the emotional over the nuanced. The feed rewards whatever stops the thumb: the immediate, the arousing, the provocative, the brief.

None of this is about the intentions of anyone using the medium. It is structural. Put a million messages into a medium and let its grain select, generation after generation, and what emerges is overwhelmingly the kind of thing that grain favours — regardless of what anyone wanted to say. The medium is a selection pressure, and content evolves to fit it the way organisms evolve to fit an environment.

The essence of a medium is the shape of message it selects for. Everything else about it is decoration on that single fact.

A functor that drops the truth

Recall the functor from the keystone topic — a map that carries structure from one domain to another. A medium is a functor of exactly this kind: it maps the world of things-worth-saying into the world of things-that-spread. The question that decides everything is what that functor preserves.

A faithful medium would preserve worth: the things that spread would be the things worth spreading. But the attention medium preserves a different property — engagement — and engagement is not worth. The feed’s functor reliably preserves whatever holds attention and reliably drops whatever doesn’t, including truth, nuance, and importance whenever they happen to be boring. It is a structure-mapping that keeps the wrong invariant. What comes out the far side has been selected for its grip, not its merit, and the two correlate only by accident.

Why the grain points away from truth

There is a reason the attention medium’s grain runs against truth specifically, and it reaches back to the very first topic. The feed selects for what captures the evolved mind — and the evolved mind, we saw, was tuned to attend to novelty, threat, outrage, and social drama, because those mattered for ancestral fitness. Truth was never what the attention system was built to detect; salience was.

So a medium optimizing for engagement is optimizing for the ancient triggers of the human attention system, and those triggers were never aligned with accuracy. The result is a medium structurally biased toward the alarming over the important, the divisive over the true, the immediate over the consequential — not by malice, but because that is the grain, and the grain is just our own instincts, turned into a market and sold back to us.

From grain to currency

Once you see the medium as a selection pressure with a grain that favours capture over worth, the next move is forced. If attention is what the medium harvests, and attention is scarce, then attention behaves like a resource that can be accumulated, exchanged, and spent — a currency. And currencies have their own logic: who issues them, who hoards them, who is enriched and who is drained. That economic structure, with attention in the place where money usually stands, is what we turn to next.