On The Difficulty Of Measuring A Hallway

I have measured Corridor 7Y forty-one times. I have forty-one figures. The mean is meaningless; the variance is the data.

What I have learned, in order:

1. The corridor is longer when you expect it to be short.
2. The corridor is longer when measured alone.
3. The tape measure reads the same. The walking disagrees with the tape.
4. Counting your steps changes the count.

I no longer believe the corridor has a length. I believe it has a length per observer per attempt, and that the archive stores all of them, and that this is why the archive’s own numbers contradict each other.

We are not measuring the corridor. We are leaving measurements in it, like footprints, and it is keeping them.

Method, such as it is

For the record, since the next observer will inherit this and distrust it:

1. Steel tape, two anchors, two readers. Read simultaneously, call aloud.
2. Laser range, tripod-mounted, three pulses, logged automatically.
3. Pace count, heel-to-toe, narrated to tape.

On a sound corridor all three agree to the centimetre. On 7Y they agree with themselves and with nothing else. The tape says 84. The laser says 84. The walking says ninety-some, and the walking is the only one of the three that has a body inside the corridor while it answers. I have come to trust the walking. The walking is afraid, and fear measures honestly.

Things I no longer do

• I do not count the lights. OBS-005 said not to. OBS-005 is on file now, not in the room, and I have decided to take the warning of a filed person more seriously than the warning of a present one.
• I do not measure alone. Not because the figure is worse alone — because the figure is cleaner alone, too clean, as though the corridor performs for an audience and tells the truth only when it has you to itself.
• I do not subtract. Early on I tried to find the “real” length by averaging. There is no real length to recover. The variance is not error around a true value. The variance is the value.

Closing note

I think the corridor is not a place that has a length. I think it is a question that has answers, and that every time we measure it we are not reading the answer off the wall — we are writing one onto it, and it is filing ours beneath all the others, in our hand, against the day we stop coming and it needs someone’s handwriting to keep the log in.

Note appended in another hand: “Observer 002 measured it a 42nd time. Did not return. The 42nd figure was never recorded. Or it was, and it is this log, and I am the one reading it.”

Note appended beneath that, same ink, smaller: “I have read this log forty-one times. On the forty-second I will understand it. I am keeping count. I did not choose to keep count.”