Position, Speed, Acceleration

Only ask this one if the last chapter came back yes. A thing with no feedback in it has no acceleration worth reading — nothing compounds, so where it is and where it’s heading tell you everything there is. This chapter is for the other kind.

You don’t need a number to know something’s turning. You need to ask where it is, which way it’s moving, and whether that movement is speeding up or slowing down — and your own body already answers the last of those, even though almost nobody thinks to ask it on purpose.

Think about a plane. Taking off, you feel it — pressed back into your seat, no doubt about it. Cruising at five hundred miles an hour, smooth and level, you feel nothing at all. You could be reading, asleep, unaware you’re moving faster than you’ll ever move on solid ground. Your body was never built to notice speed. It was built to notice a change in speed — starting, stopping, speeding up, slowing down. Steady, however fast, reads as nothing.

This has names worth learning properly, because the plain feeling and the formal idea are the same thing. Position is where something is. Speed is how fast that’s changing. Acceleration is how the speed itself is changing — the thing your body actually feels. Three descriptions of one motion, and they can disagree completely: a car can be ahead of you, slowing down, about to be passed. Its position says winning. Its acceleration says losing.

Everything that loops has these three orders, and there’s a rough ladder of attention that goes with them.

Most people look at position. Where am I? What’s in the account, what’s on the scale, how did today go. It’s the only order you can see without comparing anything, so it’s the one that feels like reality.

More careful people look at speed. Not where I am — where I’m heading, and how fast. This is already a large improvement, and it’s most of what passes for foresight.

Almost nobody looks at acceleration. And acceleration is where the future is actually being decided, because it’s what determines whether the trend you’re currently on holds, steepens, or reverses.

Here’s why it’s hard, and it isn’t a matter of effort. Position takes one observation. Speed takes two. Acceleration takes three, spaced far enough apart for the change to stand out from noise. In anything you track rather than ride, acceleration is, definitionally, the thing you cannot read from inside a single moment — which is the same claim this book keeps making about loops, arriving from a completely different direction. You feel a line. Position is one point on it.

Now the part that matters most, and it’s the reason this chapter is here at all.

You never get to touch position directly. Not once, ever. You can only apply a force — and a force produces acceleration. Time does everything after that: integrating your acceleration into speed, and your speed into position, quietly, without asking permission and without telling you it’s working.

This is the whole book stated in physics. A daily practice is a small constant force. You are not moving yourself to a destination; you are setting a curvature and letting time deliver you somewhere in a year. “We build the world by repeating what we cannot see” is a claim about acceleration specifically.

It also explains something you will otherwise experience as failure. A change in acceleration produces almost no visible change in position for a long time. Weeks of a new practice, and the numbers look identical. That is not evidence it isn’t working. That is what a second derivative looks like from inside, and mistaking the delay for failure is probably the single most common reason people abandon something that was about to work.

Acceleration is not a synonym for growth. A reinforcing loop compounds whatever it carries — a habit building and a debt deepening are the same mathematics running the same way. And deliberately slowing something down is a legitimate, often correct goal. That’s damping, not failure. Not every good force points at more.

One more thing worth knowing before you trust any reading you take: a loop has a natural period, and you need to have watched at least one full turn of it before you can honestly tell a trend from a stage. A habit’s loop closes in a day — watch a week and you’ve already seen it cycle several times over, trend included. A loop closing across generations is a different matter entirely. Watch an entire human lifetime and you may have caught less than half a single turn of it, and from where you’re standing, that half-turn is indistinguishable from a permanent fact about the world, rather than one phase of something that was always going to reverse eventually. A fast loop announces itself. A slow one disguises itself as the ground you’re standing on, precisely because you’ve personally never had the chance to stand anywhere else.

This is where a written record earns its place as more than a nice habit. Your own memory doesn’t span a slow loop’s period, and no amount of attentiveness fixes that — attention only ever covers the time you were actually there for it. A log does something memory structurally cannot: hold two points far enough apart in time to let you see the shape actually moving between them, rather than trusting a single, recent impression to stand in for the whole cycle. This is the Self-scale case of what every written record is for: it is kept because the pattern it tracks outlasts what any one sitting could perceive.

Try this: Before calling something a fixed truth about your life, your relationship, or your work, ask what its natural period might be if it turns out to be a loop rather than a fact. If you’ve only been watching for a fraction of that period, withhold the verdict — not because the trend isn’t real, but because you don’t yet have enough of the cycle in view to tell a trend from a stage.

There’s an everyday version of this that almost nobody recognizes as an acceleration problem, because it arrives disguised as a scheduling one.

The last chapter’s arithmetic about rules is also arithmetic about a life. Commitments do not add either — each one has to be carried in the presence of every other one, so the cost sits in the pairs rather than the count. Every pair is real: a conflict to resolve, a context to switch, a promise that has to stay true while a different promise is being kept.

Which is why the felt experience is never I have too many things. It is: I only added one thing, and somehow everything got harder.

That gap — between what you added and what it actually cost — is where the sense of losing control lives. Not in the size of the load. In the load being unaccountable, arriving from a decision you can no longer point at.

And it’s an acceleration problem exactly as this chapter describes one. The moment a load is set and the moment it is felt are different moments, and the first one is cheap. Saying yes costs nothing on the day you say it. The interactions arrive later, all at once, long after the decision that caused them has stopped being visible. You were applying a force. Time did the rest, quietly, the way it always does.

The practical form is a single question, asked before agreeing to anything: what will this have to be true alongside? Count the relationships, not the items.

One number, because a chapter that tells you to read acceleration and then offers no arithmetic is asking for faith. Anything growing at a steady percentage doubles in a number of periods roughly equal to seventy divided by that percentage. Three percent a month doubles in about two years. Ten percent a month doubles in about seven months.

That single division turns be patient, it’s compounding into a date you can hold the thing against. It is not the same instrument as the natural period from earlier in this chapter, and both are worth having: the period tells you how long before a trend is readable at all; doubling time tells you how long before it has grown enough to be visible in position — before the number on the outside finally moves. A practice can be exactly on schedule and show nothing, and now you can say roughly how long nothing should last.

Knowing when to stop

Acceleration is itself a position, one level up — it has its own rate of change, and that rate has its own. This is just the Nest again, and the mathematics will happily keep going forever.

The skill is knowing when not to follow it. In almost any human situation, the third derivative is real mathematics and complete noise. Stop at the level where you can actually apply a force, which for a life is nearly always acceleration. Recursing past the point of usefulness is not depth. It’s a way of staying busy in your head while nothing gets pushed.

Try this: Take one thing you’re currently tracking — a habit, a savings balance, a relationship, a project. Write down where it is. Then write down which direction it’s moving. Then, honestly, write down whether that movement is speeding up or slowing down, and what you’ve been doing lately that would explain the difference. That third line is the only one you can actually act on. It’s also, almost certainly, the one you’ve never written down before.