Method.

A discipline, Chapter VI held, is re-made at every occasion it applies to; it is a thing the institution keeps doing. That chapter did not say how the re-making is done. Method is the how. It is the practice by which a single design decision is reached, and the practice by which the discipline that governs the decision is carried past the people who hold it. A method that could reach a decision but not hand on the reaching would furnish the institution for one generation and leave the next to begin again.

Method is a practice, not a procedure. A procedure is a fixed sequence: the same steps, in the same order, producing the same result in any hand that runs them and in any conditions at all. A practice is not fixed, because the work it answers to is not. A design problem arrives in its own weather. Some arrive clear, the right move already half in view; most arrive in cloud, the decision not yet visible and several plausible ones obscuring it. A procedure runs blind through that weather and calls whatever it reaches the answer. A practice reads the weather before it moves. Unlike a procedure, the method has weather, and most of what the method does is done before that weather has cleared.

The method is also single. The institution does not keep one method for its materials and another for its sentences; the decision that places a surface and the decision that sets a clause are reached the same way. And the way begins, in every case, by refusing to begin. The method does not open a problem by reaching for an answer. It opens by holding the problem still and reading it. The first move of the method is the decision not to decide yet.

A design problem almost always arrives at the surface. A card does not read against the photograph; a line of type sits flat; a section feels crowded. The problem is stated where it is seen. The method's first working move is to decline to solve it there. A problem at the surface is, far more often than not, a symptom: the surface is faithfully reporting a decision made wrongly, or never made at all, a level above it. To answer the symptom where it shows is to treat the report and leave the cause.

So the method climbs. A mark answers to the surface it sits on; the surface answers to the page; the page answers to the kind of document it belongs to. The method ascends that order until it reaches the level at which the decision is genuinely made. Height is also visibility. At the surface a maker stands inside the weather, the obscuring possibilities close around them; from a level or two above, the same problem is seen whole, and what was a fog of options resolves into the few that are real. This is the sense in which the institution reads from altitude. The method climbs not to escape the surface but to see it.

The method does not climb higher than it must. A decision taken too high becomes a principle with no surface to land on; the right level is the lowest one at which the surface move becomes inevitable rather than chosen. And having reached that level, the method still does not decide. It holds. The work at altitude is not to pick the answer but to see the problem clearly enough that the answer, when it comes, will be the obvious one. What the method does in the time before that clarity arrives is the practice itself.

Held at altitude, with the decision deliberately deferred, the method generates. It does not work the first framing of a problem harder and harder; it opens several. A design problem has more than one true description, and the description chosen settles, in advance, which answers are even reachable. The generative work of the method is to hold a problem in more than one description at once, before any single one of them is committed to.

The strongest of those descriptions are rarely found inside the problem's own field. They are drawn in. The most generative move the method has is to take a concept from somewhere else entirely and set it against the work to see what it reshapes. Koestler, examining where new ideas come from, placed the creative act in exactly this: bisociation, the joining of two frames of reference ordinarily kept apart, so that what is held in both at once becomes something neither frame would have reached alone (Koestler, 1964). The institution's design is built from such joinings. The site is composed as a prospectus: a financial disclosure document was drawn in from another discipline and allowed to reorganise the site around itself. Its motion is drawn from the physics a reader's eye already carries, water finding its level and cloud crossing a ridge. Its stillness is read through music, where a rest is a written mark and not an empty space. None of these arrived from within web design; each was drawn in, and each, once drawn in, reshaped everything downstream of it.

The same move, made smaller, is the reframe. A design problem often resists not because its answer is hard to find but because the question, as posed, has no good answer. The clear glass card in Chapter I is the plainest case: the question how do we make this card visible produced, each time it was asked, a heavier card, and the decision came only when the question became how do we let the photograph carry through while still defining the card. Pólya, setting down the method of mathematical discovery, made this the centre of it, holding that a problem which resists solution should be restated until a related and reachable problem comes into view (Pólya, 1945). The reframe is the drawing-in turned on the question itself. It is why the institution's halo material was reached only after the documentary register had been framed more than one way and the weaker framings had been set aside: the framing, and not the effort, was what the work had been waiting on.

None of this can be proceduralised. No sequence of steps can tell a maker which field to reach into, or that the question being worked is the wrong question; the drawing-in is a leap, and a leap is not a step. This is the deepest reason the method is a practice and not a procedure. It is also why the method can reach so widely. Belonging to no single field, it is free to import from any of them, and a practice that treats music, weather, physics, and the printed document as all available to a question about a website has refused to be only a design method. It is a way of working, and design is one application of it.

With several framings open, the method waits. This is the hardest of its disciplines and the one most easily mistaken for the absence of one. A maker under pressure wants to choose, because a choice feels like progress; and a choice made early, while the framings are still equally plausible, feels as much like progress as a choice made well. It is not. When every candidate framing seems equally right, that equality is not a tie waiting to be broken. It is the air not yet cleared. The method reads it as weather, and does what weather asks: it holds, and does not move.

Premature closure is the chief failure of design method, more costly than any single wrong decision, because a decision made before the air clears is made blind and then built upon, and everything composed against it inherits the blindness. The discipline against it is slowness. Slowness here is not the absence of pace; it is a positive act, the holding of the work open past the point of comfort, on the understanding that the right framing is more often waited for than forced. The institution would rather reach a decision late and clear than early and in cloud.

What clears the air is not patience alone; it is pressure, of two kinds. The first is pushback. A framing held open is a framing to be pushed against, and the institution treats pushback as a working instrument and not as friction: a hard question, a counter-case argued with its reason attached, a refusal where a refusal is earned. Each push, taken seriously, returns something the work did not have before. The second is the test of the eye. A framing argued only in words can seem agreed when the parties to it are imagining different things; the method distrusts that agreement and puts a candidate where it can be seen, because a framing made visible can be judged rather than merely described. Pushback is the wind through the held-open work, and the visible test is the light it is read by.

Held this way, pushed and seen, the framings stop being equal. The change, when it comes, is not a decision the maker forces; it is a change in the visibility. The cloud that made several options look alike thins, and one of them is, simply and now, the framing the work can be seen to want. Nothing has been added. What has happened is that the conditions have cleared enough to show what was already true.

A framing the cleared air leaves standing has done something the others did not: it has begun to gather the work around it. The decisions still open, tested against it, stop feeling arbitrary and fall into place; the decisions that conflict with it show their conflict plainly. This is what it means, in the method, for a decision to have gravity. Gravity is not the maker's confidence in a decision, which is felt before the evidence and is the least reliable of signals. Gravity is observed, and observed downstream. It is the decision that other decisions have begun to orbit. A candidate a maker is sure of, but around which nothing else settles, has not earned its place; a candidate around which the rest of the work quietly arranges itself has.

When a decision has gathered that weight, the method locks it. To lock a decision is to commit to it as foundational: to stop holding it open, and to treat it from that point as a fixed thing the rest of the work composes against, rather than a candidate still in question. A lock is a promise the institution makes to itself, the promise to defend the decision against the ordinary drift of later opinion. It is made sparingly, because a thing locked too early calcifies the work and a thing never locked lets drift accumulate. The sign that a lock is sound is felt in everything decided after it: the later decisions feel inevitable rather than chosen, consequences of the lock and not fresh acts of will. Where the downstream work feels arbitrary still, the lock was premature.

To lock a decision is to accept a constraint, and the constraint is the point of the lock and not its price. A maker who has locked nothing meets, at every step, the whole field of the possible, and a field with no edges offers no direction; the work scatters after each fresh option, and a work that scatters never arrives. The lock draws an edge. With the alternatives it was chosen over now closed, the decisions that remain are fewer and sharper, and a few sharp decisions compose a direction where many open ones compose only a heap. This is the constraint as the method uses it: not the wall that stops the work, but the bank that gives the river its momentum.

A constraint of this kind is accepted without the fear of being wrong, because a lock can be returned to, and the return is never the defeat it appears, because it is never truly a return. Heraclitus observed that no one meets the same river twice: the water has moved on, and so has the one who would meet it again (Heraclitus, c. 500 BCE). A method that goes back to a locked decision goes back as a changed maker, to changed conditions, carrying the experience of the ground just crossed; it chooses the second time what the first choice could only have guessed at. The path that failed was not waste. It was the reconnaissance by which the right one is found. What the method cannot recover from is not the wrong lock but the refusal to lock at all: a work that constrains itself nowhere keeps no direction, and a maker with no direction is not free but lost, distracted by every alternative and arriving at none. To hold a direction, against the pull of all that was not chosen, is the form discipline takes in the method.

A locked decision is then recorded, and recorded with its reason. This is the act that completes the method, and the one most often skipped. To write down only the decision is to leave a command: do this. A command can be obeyed by a maker who does not understand it, and obeyed wrongly the moment the occasion changes, and that is the rulebook Chapter VI refused. To write down the decision together with the reason it was right is to leave something else: not a command but the reasoning itself, which a later maker can weigh against an occasion the first maker never saw. The institution records its locks this way. A card is defined by its frame, because an opaque ground would patch over the photograph that grounds every page: the decision and its reason, set down inseparably, so that the rule can be applied with judgement and not merely with obedience.

The locked decisions, each carrying its reason, accumulate. Held together they become a map: the fixed points the institution has committed to, and the gravity that pulls every later decision toward coherence with them. A new question is not answered from nothing; it is answered against the map, in the company of every decision already locked. This is how the discipline of Chapter VI is carried across the years and across the people who hold it. It is not inherited as a manual, and not as the private taste of whoever happens to care; it is inherited as a map of locked decisions, each still showing the reason it was right. The institution's many refusals are one discipline, and the method is the practice that works that discipline out, decision by decision, and writes down what it finds.

Method does not stand on its own. It is the practice beneath the elements, the voice, and the discipline; what the earlier chapters described as things the institution has, this chapter has described as the work by which it comes to have them. Positions, in Chapter VIII, sets down where that work, run across the whole of this paper, has left the institution standing, and what it leaves still in cloud. The method can be put as a single rule, and a chapter on method should end at one. A design decision is finished when it has been held open until its conditions cleared, locked once it had gathered the work around it, and written down with the reason it was right. Before that there is weather, and a maker reading it. That reading is the method, and it is the one thing the institution asks a reader to carry away.