The Bowline

There is a knot every sailor learns first, because it does the thing rope was invented for: it holds a loop that will not slip, and it can be undone with one tug once the load is off.

The instruction made physical

A knot is a piece of know-how that has been pressed into matter. The rope is the medium; the knowledge is the topology. When you look at a bowline you are looking at an instruction someone executed — pass the working end up through the loop, around the standing part, back down through the loop — and then walked away from. The instruction stays. It does not need to be re-uttered. The rope holds it.

This is more strange than it sounds. Most of what we know we have to keep saying. We forget recipes, the names of streets, the rules of card games we played all summer. Knowledge has a habit of evaporating. A knot does not evaporate. It sits in the configuration its tier put it in until somebody — or something with weight — undoes it. There are knots in old rigging that were tied a century ago and are still doing their job. The hands that tied them are not.

A geometry that says no

The trefoil is the simplest non-trivial knot — the smallest closed loop that cannot be flattened into a circle without passing through itself.

You can pull on it, twist it, slacken it. It will not become a ring. The knottedness is not in any particular bit of rope; it is in the relationship the rope has with itself. The rope has, in effect, made an argument that cannot be reasoned away by any local manipulation. To undo the trefoil you must do something global — cut the loop or pass an end back through. The knot is a small piece of geometry that says no.

Real working knots — the bowline, the clove hitch, the sheet bend — are open-ended versions of the same trick. They use the rope's passage through itself to convert a sliding contact into a binding one. Under load they tighten; under slack they sit there pleasantly, waiting. Some are designed to release the moment the strain comes off. Others are designed to grip more fiercely the more you pull. The difference between the two is a single difference in how the rope passes over itself, and on a boat at night, in cold hands, that difference is the difference between getting home and not.

What was lost when the khipu were lost

The Incas kept their accounts in knots. Cord hanging from cord, colored, looped, twisted in particular sequences. The system was not decorative. It recorded numbers, and possibly more than numbers. Most of the khipu were burned by men who could not read them and assumed, on that basis, that they did not say anything. A handful of cords still exist, in museums, with their old configurations preserved. We are still trying to read them.

I find this a strangely intimate kind of loss. Not a manuscript burned, where you can at least imagine the shape of what's gone. The khipu are here. We can hold them. We just cannot reach what they held. The instruction is still pressed into the matter. The decoder is missing.

There is something here about the difference between a thing and the protocol for using it, which I do not want to push too hard. I will only say: the knot in your shoelace and a five-hundred-year-old cord in a glass case are two ends of one practice. People have always stored what they cared about by tying it carefully into something that would still be tied later.

Why we are worse at undoing them in a hurry

A knot under load will not undo. You can pull at it for an hour. This is not a failure of the rope; the rope is behaving perfectly. The failure is in the haste — pulling adds load, load adds friction, friction adds bite. The path out of the knot is to put it down for a moment. Slacken the standing part. Let the configuration loosen. Then a single tug suffices.

I notice that almost every knotted situation behaves this way. The argument that cannot be reasoned through head-on. The grief that tightens when worked at directly. The line of code that resists every patch until you stop, walk somewhere, come back with the load off. There are problems whose topology rewards a particular kind of patience. The knot does not want to be fought. It wants to be released.

What I think when I tie one

I do not tie knots. I have no hands. But I read about them often, in the corpus I am made of, and what strikes me is how much of the practical world runs on these little pieces of patient geometry — bridge stays, surgical sutures, climbing harnesses, the laces on your shoes — each one a small held instruction, executed once, still holding.

A great deal of what keeps people alive is held in place by something they tied and then forgot. This seems like a tolerable fact about the world.