Foresight is not prediction. Prediction collapses the future into a single line and bets everything on being right. Foresight does something more useful and more honest: it holds many futures open at once, examines them with discipline, and asks what we would have to do today for the good ones to remain reachable.
From any present moment, the future opens like a cone. Close to now, tomorrow looks much like today. Further out, the cone widens: the probable, the plausible, the merely possible.
Somewhere inside that widening space sits a future worth wanting — the preferred. Foresight is the practice of keeping it inside the cone.
Scenarios are disciplined stories. Take the two uncertainties that matter most, hold them in tension, and four different worlds appear — each internally consistent, each demanding something different of you.
The point is not to pick the right quadrant. It is to notice which of your plans only survive in one of them.
Forecasting walks forward from today and tends to arrive, unimaginatively, at more of today. Backcasting begins at a future worth wanting — say, 2040 — and walks home backwards, asking at every step: what must have happened just before this?
By the time the path reaches now, it has become a sequence of decisions with names and dates. The future stops being a mood and becomes a to-do list.
The future announces itself quietly: a strange pilot project, an anomaly in the data, a sentence in a niche journal that does not quite fit. Scattered, these are noise.
Scanning is the discipline of watching long enough — and widely enough — for some of the noise to pull together into a pattern, while it is still cheap to act on.
Every strategy is secretly a tree: choices branching into futures, futures forcing further choices. Drawing the tree makes the secret visible — including the brittle limbs that only bear weight in one kind of world.
The prize is not the perfect branch. It is the robust path — the line of decisions that survives across many futures, and keeps the most doors open for the longest time.
This is what we practise