He Asked Twenty-Nine Times Before Anyone Answered. The Mission Library Had The 1925 Issue.

BANDIAGARA, Mali — Thirty-four times.

That is how often, across nine months and 1,317 notebook pages, the ethnographer wrote down his own question about a star nobody can see beside Sirius.

Twice, across the same nine months, he wrote down an answer to it.

The first answer comes on 21 March 1946, after the twenty-ninth asking.

How the count was made

The notebooks are not secret. They are catalogued, they are open to any reader who asks, and they have been in a university archive in Europe since 1961.

We paid to have every page photographed and then did something dull. Two people, working separately and not comparing notes, went through all 1,317 pages and marked every occurrence of the phrase, noting in each case whose hand it was in and whether it was put as a question or written down as an answer.

Both counts came out at thirty-four, two and one. They disagree on two marginal cases in the margins of a single page, and both of those are excluded from the figures above.

What the twenty-nine askings look like

They are not sinister. They are a man doing his job badly in a way that was ordinary in 1946 and would end a career now.

He asks in four villages. He asks the same three men repeatedly. By the eleventh asking he has begun supplying the shape of the answer inside the question — the notebook records him asking whether the small companion is heavy, and then, later, whether it is very heavy, and then whether it is the heaviest.

By the twenty-second he is drawing it. There is a diagram on page 604, in his hand, of a large body and a small one, with an arrow.

On 21 March 1946 a man gives him an answer that matches the diagram.

The word from 1931

Fifteen years earlier a schoolteacher named Solange Vauthier kept a much smaller notebook in the same district and wrote out a list of star names with glosses.

One entry, near the bottom, is a word she translates once as the heavy one and once, four pages later, as the packed thing.

That is the only occurrence anywhere in either notebook that predates the ethnographer's arrival, and it is the reason this story exists.

We sent the page to three linguists who work on the languages of the escarpment. None was told what hung on it.

Two came back with the same answer: it is an ordinary term for dense or tightly packed, it appears in four other star names on the same page, and in all four of those it is doing the work of describing a close cluster, not a mass.

Professor Hortense Ngoubou-Erdély, the third, does not agree.

"The other four are clusters and I accept that," she said. "This one is not a cluster. It is applied to a single named star in a list where every other single star gets a colour or an animal. That is an anomaly and I am not going to explain it away because the explanation would be convenient. I cannot tell you what it means. I can tell you it is not doing the same job as the other four."

The shelf, forty kilometres away

A mission station stands forty kilometres from the village where the two answers were given. It had a library. The library had a periodicals ledger, and the ledger is still in the room, and the current custodian unlocked it for us and stayed while we read.

The station subscribed to a French illustrated weekly from 1919 to 1941. The run on the shelf is complete.

The companion of Sirius was the subject of a two-page article in that weekly on 12 December 1925, following the measurement that established the extraordinary density of the thing — an object the mass of the sun in the volume of a planet. It was a sensation in Europe for about a fortnight.

The ledger records the issue arriving in March 1927. Mail was slow.

It is on the shelf, bound with the rest of that year. Nine issues in the twenty-two-year run have a cut page.

That is one of the nine.

What we cannot do with that

We cannot say who cut it. We cannot date the cut. We cannot put the man who answered in 1946 anywhere near that room, and we looked: he is not in the school register, and the custodian points out, correctly, that the register was only ever a list of pupils.

We can say that the astronomy was in print, in French, in a popular magazine, twenty-one years before the question was first answered, and on a shelf forty kilometres away nineteen years before. And that the page is gone.

What this desk takes from it

The tally is the finding, and it points one way. Twenty-nine askings before an answer is not a discovery; it is a survey instrument teaching its own respondents.

The word from 1931 is the residue, and two of three linguists dissolve it and one will not.

But the reply printed below is the best thing on this page, and its central objection is one we cannot answer. Two of our reporters were here for six days and asked about a star. Nobody, in seventy years, has come to ask about the calendar the star sits in.

We are going to. Not to settle anything — there is nothing there to settle — but because a system for carrying an instruction across sixty years, to people who will only ever perform it once, is a better piece of engineering than anything in this story.