While my reading has necessarily been focused lately on books being discussed at the Oxford Literary Festival, it has not been exclusively in this category. I have, for instance, been dipping into Long, Long Ago, a selection of essays by the American critic Alexander Woollcott (pictured), published by the Right Book Club in 1946.

In a piece about the poet and academic A.E.Housman, I came across the following curious passage. It contains a significant error; see if you can spot it: "In 1922, Professor Housman - by this time he had grown venerable and become Kennedy Professor of Latin at Cambridge University - published thirty-one more lyrics which, with magnificent finality, he called Last Poems. Thirty-one and sixty-three, one hundred and four in all, the same number - could it have been coincidence? - as had been left behind by an earlier and somewhat more cheerful pagan named Quintus Horatius Flaccus. I think it was no coincidence."

I think it was no coincidence either, but in a different sense. If you have not spotted the mistake yet and want to keep hunting, then look away now. I invite other readers to perform a simple act of addition. They will find that 31 and 63 do not add up to 104 but to 94. Thus is Woollcott's whole point blasted utterly to smithereens.

d=3,3,1It seems astonishing that such an error should have sailed into print. Even if it had not been discovered by the book's British editors, you would have thought that it might have been detected earlier when Long, Long Ago came out in the US. Didn't any reader spot it when the article first appeared in a magazine? Amazingly, you might think, the magazine was The New Yorker, famous for its much-vaunted determination to be absolutely accurate. This was obviously one occasion when editor Harold Ross and his team of fact-checkers failed in their mission.

You know, I am beginning to wonder if this is not the first correction of Woollcott's error that has ever appeared . . . While my reading has necessarily been focused lately on books being discussed at the Oxford Literary Festival, it has not been exclusively in this category. I have, for instance, been dipping into Long, Long Ago, a selection of essays by the American critic Alexander Woollcott (pictured), published by the Right Book Club in 1946.

In a piece about the poet and academic A.E.Housman, I came across the following curious passage. It contains a significant error; see if you can spot it: "In 1922, Professor Housman - by this time he had grown venerable and become Kennedy Professor of Latin at Cambridge University - published thirty-one more lyrics which, with magnificent finality, he called Last Poems. Thirty-one and sixty-three, one hundred and four in all, the same number - could it have been coincidence? - as had been left behind by an earlier and somewhat more cheerful pagan named Quintus Horatius Flaccus. I think it was no coincidence."

I think it was no coincidence either, but in a different sense. If you have not spotted the mistake yet and want to keep hunting, then look away now. I invite other readers to perform a simple act of addition. They will find that 31 and 63 do not add up to 104 but to 94. Thus is Woollcott's whole point blasted utterly to smithereens.

It seems astonishing that such an error should have sailed into print. Even if it had not been discovered by the book's British editors, you would have thought that it might have been detected earlier when Long, Long Ago came out in the US. Didn't any reader spot it when the article first appeared in a magazine? Amazingly, you might think, the magazine was The New Yorker, famous for its much-vaunted determination to be absolutely accurate. This was obviously one occasion when editor Harold Ross and his team of fact-checkers failed in their mission.

You know, I am beginning to wonder if this is not the first correction of Woollcott's error that has ever appeared . . .