II. Above the Ladder Essay 11
Does It Change What You Do?
A test for the thinking you use to decide things
You have already made the decision. Not recently, not this morning, not easily. The decision has been made, and everything since has been arrangement. So when somebody starts talking to you about frameworks and models and better ways of thinking, part of you is thinking: fine, but I have already decided. What would any of this change now?
That question is the right one, and most of the answers to it are given too quickly and in the wrong register. Sometimes the answer is that a piece of thinking has changed the decision before it is made, and the person who asks has not noticed. Sometimes the answer is that it has not, and the thinking was still worth having. Sometimes the answer is that it has changed the decision, and the change is unwelcome, and that is still an answer.
There is a way to tell these apart, and it is a test rather than a debate. If a way of thinking is being used to make a decision, then it has to change a decision, a prediction, or a cost. If it changes nothing, it is not being used to decide anything, and the honest conclusion is that it is something else, which is not a lesser thing but is a different thing.
Take a small workshop that makes and sells tableware. The owner has set the price of a mug once, four years ago, and has never changed it since. The cost of clay, glaze, firing and the time of two people have moved several times. The owner is aware of none of this in any precise way, and the price has nothing to do with any of it.
Now apply a system lens to the price. The system lens does not tell you what the price should be. It tells you what the price is currently made of: the cost of making the piece, the cost of the time it takes, and the constraint of what the market will bear.
The lens produces a number of things you can now do that you could not do before. You can forecast: at the current cost, this price covers the cost of making the piece, and the leftover is whatever is left after paying the two people. You can name a cost: the cost of the time, which has never been counted, because time has never appeared on an invoice. And you can predict: if the cost of firing goes up again, this price will absorb it, and the workshop will keep making the piece and keep not being paid for the days it takes to make it.
Notice that nothing has changed yet. The price is the same. But the decision is no longer closed. There is now a prediction that can be checked, a cost that can be named, and a specific question that can be asked, and each of those is a door that was shut before. On the evidence of this lens, the price decision has changed: it is no longer a decision that was made, it is a decision that is open. Whether the owner then raises the price is a separate matter, and the lens did not settle it. But the lens did something the price itself could not do, which was to make the decision available again.
The one minute test
The test has three parts, and it takes about a minute, and you can do it on a lens you already hold. Take a way of thinking you have used, or a tool, or a model, and ask three questions.
What would I do differently tomorrow? Not in principle. Tomorrow, in a specific situation, in a way somebody could observe. If the answer is a change in a specific action, the lens has passed. If the answer is a change in a description, or a change in what you would say, or a feeling that you now understand it better, then you are probably looking at the second part rather than the first.
What would I predict? A lens that changes nothing predicts nothing. If the lens cannot tell you what is going to happen, it cannot tell you whether you were right, and so it cannot be checked, and so it is a belief rather than a tool. This is the part that separates a useful model from a comfortable story, and it is the part most often skipped.
What would it cost? A lens that changes something always costs something. Time, money, a conversation you have to have, a decision you have to defend to somebody who wanted the old one. If you cannot name the cost, you have probably not understood the change. A change with no cost is not a change, it is a decoration.
What passes
Think of the system lens applied to a school timetable. Does the teacher do anything differently tomorrow? Yes: the teacher stops asking for the impossible and asks for the missing period, which is a specific request with a specific recipient. Does the lens predict? Yes: with this staffing and these rooms, the two classes that clash cannot both exist, and if you do not move one of them, one of them will lose its double period. Does it cost? Yes: a period, and a difficult conversation, and one student who will not get the option they wanted. The lens passes all three parts, and it is worth its keep in a decision.
Think of the abstract lens applied to a question of style. A writer is told to look at the first paragraph of the piece. Does the writer do anything differently tomorrow? Yes: the writer notices that the first paragraph promises an argument and the second delivers an anecdote, and rewrites the second. Does it predict? Not really, and the writer says so honestly. Does it cost? An afternoon, and the discomfort of finding out that the first paragraph was the better one. This is a partial pass, and partial passes are legitimate.
There is a version of the test that is easier to fail than it should be, and it is worth knowing about. People tend to answer the first question with something about understanding, because understanding is what they have just got. The question does not ask whether you understand it better. It asks what you would do differently tomorrow, and understanding is not an action. If the honest answer is that you understand the situation better and nothing else has changed, that is a good answer, and it belongs in the second box of the flow, not the first.
What fails
Now take a model of learning levels and apply it to a decision about a curriculum. The teacher learns about four levels of learning. Does the teacher do anything differently tomorrow? The teacher rearranges the vocabulary in the planning document. Does the teacher predict anything? Not really. Does it cost anything? No. This fails, and it fails in a specific way: the model has been used to rearrange language rather than to change a decision, and language is cheap to rearrange.
And take a diagram of a system, beautiful and complete, made in a workshop. Does anything change tomorrow? The diagram is on the wall. Does it predict? It does not say what happens next. Does it cost? The afternoon it took to draw. This fails, and it is the commonest failure of all, because a diagram is the most satisfying object a person can make and the least likely to be about anything.
"If it does not change a decision it is not worth learning." This is the mistake this essay is arguing against, and it is a mistake because it confuses thinking with a tool for deciding. Most of what is worth knowing is not a tool for deciding anything. The doctor who knows what appendicitis feels like in five hundred patients is not about to decide anything about appendicitis today. The knowledge is not idle. It is the thing that makes tomorrow's decision fast and correct, and it is not valuable because it changes a decision but because it is what the decision is made out of.
The other misconception: "the test is too high a bar, everything changes something." No, plenty of thinking changes nothing, and that is what this test is for. The test is not a demand that every idea be productive. It is a way of telling, in a given situation, which of these you are holding.
What if the lens changes the decision and you do not like the change? This is the case that gets talked about least and matters most. A system lens applied to a small business might tell you that the founder is the bottleneck, because every decision waits for her. That is true, and it is uncomfortable, and the natural response is to set the lens down and go back to the old way of working, where the bottleneck is at least a familiar problem.
Notice what has happened. The lens failed the test and passed it at the same time. It changed the prediction, it named a cost, and it changed what you would do tomorrow, and what it changed was to make you dislike it. The honest reading is not that the lens is wrong. It is that the lens is right and the decision is now hard. A model that never makes anything hard is not a model of anything.
Now change the condition. What if the lens changes nothing, but you keep it anyway because it makes the work more interesting? That is a completely legitimate answer, and the test does not condemn it. The test only insists that you know which kind you are holding. A lens kept for pleasure is a different tool from a lens kept for decisions, and pretending the second is the first is how people end up defending tools instead of using them.
Where the test applies
The test applies to thinking used in decisions. That is a narrower claim than it sounds, and the narrowing matters. Most thinking is not decision-making. Reading a poem is not a decision. Understanding why a cup is the shape it is, for its own sake, is not a decision. Doing a careful drawing of a door because the drawing is worth doing is not a decision.
Some looking is its own reward, and the test has nothing to say about it. This is not a small concession. The whole ladder on this site is built on the idea that looking carefully at one thing is the ground floor of everything else, and if the test made looking conditional on producing action, it would be dismantling its own foundation. Looking is worth doing because it is what looking is for, and because the person who looks carefully is a different person afterwards, even if nothing changes on Tuesday.
The test is for the part of the ladder that claims authority over action. A principle that tells you what to do in a case you have not met has to be checked against cases. A system model that explains why something happened has to be checked against what happens next. A loop has to be checked against a mechanism. These are claims about the world, and claims can be wrong, and a test that asks what a claim would change is a reasonable way to find out whether it is worth the space in your head.
Keeping the test short
The one minute test is short on purpose. A longer test gets skipped, and a skipped test is worse than no test, because it gives you the feeling of having checked. The three questions fit in a minute because each of them has a short answer or a blank, and a blank is a useful answer.
There is also a version for teams, and it is more brutal than the individual one. Ask: what is different about what we do on Monday? Not what do we think about it, not what have we learned. If the answer is a description of a new understanding, the meeting has produced understanding and no work, and that is fine, and somebody should say so out loud so that the meeting is allowed to end.
Limits
The test cannot tell you whether a change was right. It tells you that something changed. Whether the change was an improvement is a separate question, and it is often harder, and no one minute answer settles it. A lens can change a decision for the worse, and the test will not object, because the test is about whether thinking is connected to action and not about whether the connection is good.
Second limit: the test measures the lens, not the thinker. A person with a good lens who does not apply it will pass every version of this test, because the test is applied to the lens and not to the person. The test is a filter for what to keep in your toolkit, not a measurement of anybody.
Third limit: it rewards the decisive over the slow, and there are situations where looking is the only honest response and no decision is available at all. Some questions genuinely have to be held open. A test that is applied there will make you act to avoid feeling indecisive, and acting is not always better than waiting.
Stuck
You are stuck when you are holding a way of thinking that is not being used for anything. The tell is that you cannot finish the sentence "so I would do X differently". Not because the answer is hard, but because there is no X, because the way of thinking is not connected to a decision and never was.
The way out is to find a decision. Not a big one. A small, real, near-term decision that this way of thinking could bear on, and then to apply it to that. Any way of thinking that has a genuine use has at least one place it could be used, and the place is usually smaller and nearer than the one you have in mind. A system lens can be used on the next two hours, not only on the five year plan.
Sources
- The three part test is the author's own, and is not borrowed from any named framework.
- The workshop example is hypothetical. No specific prices, costs, or figures from any real business are used.
- No specific studies, names, or dates are cited in this essay.