The Bedford Level

The Bedford Level is the most famous experiment in the whole flat-earth debate, and both sides tell it badly. One side says a man proved the water was flat and the scientists never answered. The other says a scientist proved it was curved and the flat-earthers were fools. The real story is longer, stranger and more honest than either version. Here it is from the records, told fairly to both sides.

A six-mile ditch

The Old Bedford River is not really a river. It is a straight drainage channel dug across the flat fenland of eastern England in the seventeenth century, running for miles without a bend. Between Welney Bridge and the Old Bedford Bridge there is a stretch of about six miles of still, straight water. If you want to test whether a surface of water curves, it is hard to imagine a better place.

On the textbook figure, the earth’s surface drops away from a straight line by about eight inches over the first mile, and the drop grows with the square of the distance. Over six miles, that adds up to many feet. Something six miles off should be partly hidden, unless the water is flat.

Rowbotham in the water

Samuel Birley Rowbotham, writing under the name “Parallax,” said he first made his tests there in the summer of 1838. In the 1865 edition of his book, Zetetic Astronomy, he describes sitting in the water “as a bather,” his eye no more than eight inches above the surface, watching through a telescope as a boat carrying a flag three feet high was rowed six miles away. His report:

The flag and the boat down to the water’s edge were clearly visible throughout the whole distance!

By the 1881 edition, the flag had become five feet high, and he worked out what the globe should have done. With his eye at eight inches, he said, the boat should have dropped sixteen feet eight inches below his line, with the top of the flag eleven feet eight inches out of sight. His conclusion:

From which it is concluded that the surface of standing water is not convex, but horizontal.

He also described a second test, with six flags set a mile apart along the bank, all five feet high, which he said lined up perfectly through his telescope. We report these as Rowbotham’s claims, in his words. Notice one detail that matters later: his eye was only eight inches above the water.

The challenge of 1870

In January 1870 a follower of Rowbotham named John Hampden published a challenge. He would pay £500 to anyone who could

exhibit, to the satisfaction of any intelligent referee, a convex railway, river, canal, or lake.

The naturalist Alfred Russel Wallace, a well-known scientist, hoped to win the money, and asked the geologist Charles Lyell whether he should take the bet. Lyell answered: “Certainly. It may stop these foolish people to have it plainly shown them.” Hampden chose the place: the Old Bedford, Rowbotham’s own ground. Each side put up £500 with J. H. Walsh, the editor of The Field, who served as stakeholder and umpire. Hampden’s referee was William Carpenter, who had written in defense of the flat earth. Wallace’s was a local surgeon, Mr. Coulcher.

March 1870: the setup

Rowbotham dates the experiment to 5 March 1870. Wallace set his line of sight on the parapet of Welney Bridge, “thirteen feet three inches above the water.” Six miles away, on the Old Bedford Bridge, he hung “a large sheet of white calico, six feet long and three feet deep, with a thick black band along the centre,” at the same height. Halfway between, he set up a pole with two red discs: the upper one at the same height as the band and the telescope, and “the second disc was four feet lower down.”

The idea was simple. If the water was flat, the upper disc would line up with the black band. If it curved, the middle of the canal would bulge up between them, and the middle disc would appear higher than the band. Wallace expected the bulge to be about five feet eight inches, “reduced a little by refraction to perhaps about five feet.”

Through his large telescope, the middle disc appeared well above the band. Writing to The Field a few weeks later, Wallace put the apparent rise at about five and a half feet. Through a surveyor’s level with a cross-hair, he reported that the middle signal fell below the cross-hair, and the far signal fell below the middle one by about as much again, a total, he said, of twenty-two feet against a theoretical figure of “something less than 24 ft.” His conclusion:

the three points equidistant from the surface of water were not in a straight line

The views were sketched on the spot. Coulcher sketched what the large telescope showed and Carpenter signed it as correct; Carpenter sketched the level’s views and Coulcher signed them.

Two readings of one view

Here the two sides parted. According to Wallace, when Carpenter first looked through the level he cried “Beautiful! Beautiful!” and said it showed “a perfect straight line.” Carpenter held that the three signals lined up, and that the large telescope

proved nothing, because the telescope was not levelled, and because it had no cross-hair!

Wallace replied that Carpenter had misread the cross-hair. Hampden himself, Wallace says, “declined to look through either telescope, saying he trusted to Mr. Carpenter.”

Rowbotham later gave the flat-earth side’s technical answer in the 1881 edition of his book. He said the magnified view of a signal already set above the line of sight showed “nothing more than simply the upward divergence,” and that the level’s readings came from “the well-known and admitted refraction inseparable from the instruments employed.” He did not hold back:

In the whole history of invention, a more hasty, ill-conceived, illogical conclusion was never drawn.

But the umpire was Walsh, and his decision is not in doubt. Wallace records it plainly:

Mr. Walsh decided without any hesitation that I had proved what I undertook to prove.

We state it just as plainly. The appointed referee ruled for Wallace, and the money was paid to him.

Fifteen years in the courts

Hampden would not accept the ruling. He called the experiment a swindle and began a campaign of letters and pamphlets against Wallace and Walsh that lasted for years. He wrote to Wallace’s wife threatening her husband, and was bound over to keep the peace. Wallace won a libel suit with £600 in damages, which he could never collect. Hampden apologized in the newspapers, then began again. For further libels he spent two months in Newgate prison in 1873, and in 1875 he was sentenced to a year’s imprisonment, of which, Wallace says, he served about six months.

In 1876 Hampden sued Walsh to get his £500 back, and won. This is the part that is most often told wrongly. The court did not rule that the water was flat or that the experiment was mistaken. Under the law of the time, a wager could not be enforced, and a man could demand his stake back from the stakeholder before it was paid over. Wallace had to return the £500 and pay costs. In the end, he said, the affair cost him more than he had ever won.

Wallace did not come out of it proud. He called the wager “the most regrettable incident in my life,” and the years that followed “a tolerably severe punishment for what I did not at the time recognize as an ethical lapse.” Gambling over the question, he admitted, had been his own fault.

The repeats

The canal kept drawing people back. In 1901 the Cambridge geographer Henry Yule Oldham repeated the test with three poles of equal height, a theodolite and photographs, and reported curvature to the British Association. (Our sources disagree on his exact figure, so we do not give one.) In 1904 Lady Elizabeth Blount, who led the Universal Zetetic Society, hired the photographer Edgar Clifton to photograph a white sheet six miles away from a camera two feet above the water. The sheet appeared in the photograph, to Clifton’s own surprise, and it was published as evidence for a flat surface. But Clifton’s own notes, as reported, describe “misty” weather and “an aqueous shimmering vapour” floating on the canal. The next year an engineer named C. E. Stretton reported a test with a theodolite on another canal that showed about eight inches of dip over a mile.

We could find no documented modern repeat on the Old Bedford itself, with lasers or otherwise, so we do not claim one.

What refraction does

The key to the whole dispute is something both sides knew about in their own time: refraction. Air near the ground or water is usually denser than the air above it, so a ray of light traveling nearly level bends slightly downward and follows the surface a little. Distant things then appear higher than plain geometry predicts, and some can be seen when they “should” be hidden. Surveyors already made a routine allowance for it in the nineteenth century. Rowbotham himself allowed “for refraction” in one of his tests, and Wallace allowed for it in 1870.

The effect is strongest, and least predictable, in the first few feet above the water, especially when cold water sits under warmer air. In those conditions the bending can grow large enough to make a surface look flat, or even dish-shaped.

What the record shows

The Bedford Level does not settle the shape of the world, and it should not be used as if it did. It is a real piece of history: a ditch, a wager, a referee’s ruling for Wallace, an angry loser, a confused court case and a century of retellings shaped by whichever side was telling. Low-level sightings over water are at the mercy of the air, in both directions.

That is why we do not rest our case on a canal. The Bible’s description of the earth stands on what God has said, and we set that out from Scripture in The Shape of the Earth. Where history is concerned, our duty is simply to tell it straight, including the parts that do not help us.

He that is first in his own cause seemeth just; but his neighbour cometh and searcheth him.

A false balance is abomination to the LORD: but a just weight is his delight.

Further reading