Checking the calendar too often

The calendar app was opened 41 times between waking and sleeping, and on 37 of those occasions the day went on exactly as it would have gone anyway. The number is a composite, assembled from what people describe when they are honest about an ordinary Tuesday rather than read off anybody's phone, and I would rather say so at the top than have you carry it as a measurement.

Four of the 41 did some work. One settled when to start a long drive. One moved a call by two hours. One was the reason a form went in on Wednesday instead of Friday. One was somebody checking which day a fast fell on for their mother, a use so old and so ordinary that it hardly belongs in an essay.

The other 37 are the subject.

Take one and slow it down, because at ordinary speed it goes past without being seen. It lasts about three seconds and it has three parts. Something arrives, not a thought exactly, more a small pressure, usually attached to a matter that has not been settled. The thumb finds the icon before the pressure has finished forming. Then the pressure goes down, a little, for a while.

Nothing in that sequence is foolish, and the checker is not somebody taken in by a bad idea. The checking works. It does what it is for, every time, which is why the count is 41 and not four.

What the three seconds buy

Certainty, in a small piece, on demand.

That is not a trick, and not something to apologise for wanting. It is what a calendar is: an instrument for making a bit of the future knowable before you have had to live through it. A panchang (पञ्चाङ्ग, pañcāṅga), the five-part daily sheet an Indian almanac prints, tells you what today is made of before you have spent a minute of it. Gochara (गोचर, gocara), the word the method uses for where the planets stand now against where they stood when you were born, describes the stretch you are in. Both are answers, and both arrive in under three seconds.

Now set that beside the question actually pressing on you, which is whether the thing you did in April was a mistake, or whether your mother's cough is nothing. You cannot have an answer about April. You can have one about Thursday afternoon, immediately, at no cost. So the checking is a substitution made in good faith: the nearest available answer stands in for the one that cannot be handed over at all.

Then watch what happens to the size of the piece. Early on, a look covers a month: the shape of a stretch, the boundaries of a period. Later it covers a week. Then a day, then this afternoon, then the next ninety minutes. What people notice about themselves is the frequency, which is the less interesting half. The other half is that the answer has to be about a smaller and nearer piece of time in order to feel certain at all, and a smaller, nearer piece of time runs out sooner. A month will last you a month. Ninety minutes lasts ninety minutes.

Two other things hold the habit in place, and neither is about astrology. Both have been looked at in laboratories where nobody was thinking about calendars.

The first is the shape of the payoff. Most looks give you nothing. A few give you something that feels like a find: a timing you had not noticed, a line that fits the week you are having. If the finds arrived every time, the habit would be easy to put down, since a habit built on a dependable payoff falls apart quickly once the payoff stops. It is the unreliable ones that last. That has been in the literature since Ferster and Skinner's 1957 book on reward schedules, and the finding is dull and solid. A reward that comes sometimes, unpredictably, produces behaviour that outlasts almost anything you can do to it. A calendar interesting on one morning in nine will do.

The second surprised me when I came across it. Checking a thing over and over does not make you surer of what you saw. It makes you less sure. Two researchers in the Netherlands had ordinary volunteers, not patients, repeatedly check a gas stove simulated on a screen, then asked about the last check. Accuracy held up. Confidence did not, and neither did the vividness and detail of what people remembered (van den Hout and Kindt, Behaviour Research and Therapy, 2003). Their subject was a stove and not a calendar, so I am borrowing the shape rather than the conclusion, but the shape is worth having. Look at the same page nine times and the ninth look competes with eight others for one slot in your memory. None stands out, and what is left is a sense of not quite knowing what it said. Which is an excellent reason to look again.

Where, and whether

There is using a calendar and there is being used by one, and the difference is not how often you look. A farmer, a priest and a family with a wedding coming consult a calendar constantly, and none of them is in any trouble. The difference lies in what the answer is allowed to change.

A version of that distinction is already in the essay this one grows out of, about the daily windows: one person is planning, another is asking the day for permission, and from the outside the two are indistinguishable. What that essay lacks is a way to tell them apart from the inside, in the three seconds while it is happening.

Does the answer change where a thing goes, or does it change whether it happens?

Where is planning. You have something that is going to be done, and the calendar puts it on Thursday morning rather than Wednesday evening. The thing survives the answer: you were choosing its place, and calendars are very good at places. The word remembers this better than we do. Calendar comes from the Latin calendarium, an account book. It is named for the kalends, the first day of the Roman month, when debts fell due and accounts were settled. The oldest thing inside the word is a record of what is owed and when. Not a permission.

Whether is the other kind. You have nothing to place. You have a feeling, and you take the feeling to the calendar to find out what it is entitled to, and you come back holding a verdict on it. Now the thing does not survive the answer. It happens or it does not, on three seconds of reading, and the reading never knew what you were asking.

Apply that to your own 37 and I think nearly all of them were whether. Not because you are superstitious. Because you had nothing to place. Some of them you did not even go looking for: they were pushed at you by a feed built for pushing things, which is a different subject with a different writer.

Three interruptions

None of them asks you to decide anything about astrology.

The first is a window, and it is not the hour's delay I have suggested before, which only moves the question along. One sitting a day, at a time fixed in advance, in which everything you were going to ask gets asked at once. Put it where you can still act on what you find, which for most people is not seven in the morning: late afternoon, or the evening for the day after. There is nothing modern about the shape. A printed sheet was consulted once and then went back on its nail, and it did not refresh.

Expect the first week to be mildly uncomfortable. The impulse arrives at half past ten, finds the window shut, and has nowhere to go. That discomfort is the whole of the experiment. Most impulses will not survive until the window. The few that do are what the sitting is for.

The second belongs to the page. If you keep the four lines and the three described in keeping a decision journal, give it one extra column for a week, and only a week: the time, what you wanted to know, and what you did differently. The third column is the one that matters, and it is the one that stays empty. A page with 37 rows and four entries in the last column is not an argument about whether the method works. It is a record in your own hand, and it settles in seven days what an argument would not settle in a year. Then stop keeping it, because a log becomes one more thing to check.

The third is the smallest and it does the most. At the window, copy out on paper the two or three timings that genuinely bear on tomorrow. Then shut the phone. The paper is not better than the screen at anything, and it has exactly one property the screen lacks: it cannot be asked again. It says the same thing at eleven at night as it said at five, in the same words, and after a fortnight of that something in you stops going to it for reassurance.

None of this is written for the heavier version, where the checking has taken the morning or the relief has stopped arriving at all. That is a doctor's question, not an astrologer's, and not mine either. What is in front of most people is smaller: 41, a working Tuesday, and four looks that earned their place.

So, tomorrow, at about twenty past eleven, the pressure arrives the way it does. The window is shut. The sheet is already written and says what it said at five. There is nothing to open, and for a few seconds the pressure stands there with nowhere to put itself. Then it goes. What it was about is still there when it has gone, in the condition you left it in. It was never about Thursday.