Reading the sky: telescopes, observatories and astronomies of the world

A fragment of brown clay tablet covered in small engraved wedge-shaped signs. — Clay tablet written in cuneiform in Babylon: it records an observation of Halley’s Comet, between 22 and 28 September 164 BCE (British Museum, London). Gavin.collins, Public domain, source

More than two thousand years ago, in Babylon, scribes recorded eclipses on clay tablets.

The oldest still in use date from 720 BCE.¹

Today these records make possible a measurement that no other source allows.

The Earth is slowing down

Our planet is spinning a little more slowly than it used to. It is mainly the Moon that brakes it — through the tides, as our page on the Earth and the Moon explains.

The question is how much it is slowing down. The length of the day is increasing by an average of 1.8 milliseconds per century. That is a tiny amount. But it adds up century after century, and after twenty-seven centuries the accumulated offset runs to hours — enough to become observable.²

The result is intriguing: the tides alone predict a lengthening of 2.3 milliseconds per century. What is measured is 1.8. So the Earth is slowing down markedly less than expected. Something, inside the planet and at its surface, is working in the opposite direction — the rebound of land once crushed by ice, and exchanges between the core and the mantle.³

The method of measurement

A total eclipse of the Sun is visible only along a narrow band of the Earth. A text saying that it was total at a given place therefore fixes the orientation of the Earth at that moment, to within a few minutes.⁴

If the sky of the past is recalculated by computer on the assumption that the Earth has always turned at the same speed, the eclipse falls somewhere else. The gap between where the calculation places it and where someone actually saw it measures how much the Earth has slowed down since.

Babylonian tablets have preserved only two observations of a total solar eclipse. One of them, in 136 BCE, indisputably describes a total eclipse, and its date is written in black and white in the text: on its own, it provides an independent measurement.⁵

What this means

The real slowdown cannot be measured from modern data: over a few decades, it is drowned out by much larger fluctuations. Observations from the distant past are therefore needed — however crude they may be by our standards.⁶

Now, the most precise ones available before the invention of the telescope are accounts of eclipses. They go back to about 720 BCE.⁷

These observations come from everywhere: Babylon, China, Greece, the lands of Islam, Europe. Citing them is not a matter of courtesy but an acknowledgement of dependence: without these scribes, this value would remain unknown.⁸

For centuries, Babylonian astronomers systematically measured the beginning and end of eclipses. They wrote on clay tablets, in cuneiform script — which can now be read without difficulty. The British Museum holds some two thousand of them devoted to astronomy, and it is estimated that fewer than 10% of the original records survive.⁹

The critical assessment of ancient data

These texts are treated not as mere treasures, but as data — and all data call for critical assessment.

The trap of circular reasoning. When a tablet is damaged and the date is missing, it would be tempting to recover it from the eclipse itself. But to do so, one would already need to know the very slowdown one is trying to measure. The researchers therefore deliberately set aside any observation whose date had been reconstructed in this way.
Sorting by quality. Not all observations are of equal value: the oldest give the time only to within twenty minutes, sometimes forty. They are therefore given different weights, and the least reliable are set aside.
Doubt openly acknowledged. Two ancient observations do not agree with the rest. Rather than twisting the curve to make them fit — which would require implausible variations in speed — the researchers flag them as exceptions to be studied.¹⁰

That is what respecting an ancient observation means: using it, and checking it — not marvelling at it.

The guest star of 1054

In the eleventh century, astronomers in East Asia recorded the appearance of a “guest star” in the constellation of Taurus, so bright that it was visible in broad daylight; it then slowly faded and disappeared.

It was an exploding star — a supernova. Its remains are now known as the Crab Nebula. Without these notes, we would know that the nebula exists, but we would not know when it was born.

Here again, the texts are debated: the translations have been revised several times. This in no way diminishes the value of these documents: it means treating them as historical sources in their own right.

What we see depends on the instrument

For millennia, the sky was observed with the naked eye, with the help of instruments for measuring angles. Then came refracting telescopes, reflecting telescopes, photography, and finally sensors for what is invisible to the human eye (infrared, X-rays, radio waves).

At each stage, we did not simply see further: we saw something else.

That is why the fundamental question is not only “what was seen?” but also: what was it observed with?

Instruments for seeing something else

Before the telescope, there were already observatories. In Jaipur, in India, a site built in the early eighteenth century brings together some twenty large fixed instruments in masonry, designed to measure the positions of the stars with the naked eye. Some are the largest ever built in their category.¹¹

In 1610, Galileo published what he had seen by using, for the first time, a telescope as an instrument of science: the Moon covered with mountains and valleys, the stars of the Milky Way, moons around Jupiter.¹²

A refracting telescope gathers light with a lens, which disperses it and produces false colours. Isaac Newton replaced the lens with a concave mirror. His telescope is still kept by the Royal Society in London.¹³

The atmosphere gets in the way too: shifting pockets of air blur the view — this is what makes the stars twinkle — and it blocks part of the ultraviolet and the infrared. The Hubble Space Telescope, launched on 24 April 1990, was therefore placed in orbit above the atmosphere. Its mirror is 2.4 metres across.¹⁴

The James Webb Space Telescope, launched on 25 December 2021, has a mirror 6.5 metres across, made of 18 hexagonal segments coated with gold, which reflects infrared light better. Wider than the rocket, it was folded for launch.¹⁵

Finally, we do not observe only light. Tasked with understanding the source of interference to transatlantic telephone communications, the engineer Karl Jansky traced it to the centre of the Milky Way. His discovery of 1933 laid the foundations of radio astronomy.¹⁶

A compass that exists only in a body

The task is to cross the Pacific without instruments, without a map and without a compass: not once by luck, but regularly, over thousands of kilometres.

A traditional navigator uses the rising points of the stars as a compass. This compass exists nowhere physically: it is entirely contained in his memory.¹⁷

The method

The navigator steers towards a known star; when it climbs too high, another takes over. But the stars are only one marker among others: the Sun, the Moon, the swell and the flight of birds are read too. Even after several days of completely overcast sky, a great navigator was able to keep in mind an accurate picture of his canoe’s progress.¹⁸

The demonstration

In 1976, the master navigator Mau Piailug guided the canoe Hōkūleʻa from Hawaii to Tahiti — about 4,000 kilometres, without a map or instruments. Having set out on 1 May, it arrived on 4 June.¹⁹

On the approach, it was birds that announced land: small white terns skimming the sea, a sign that a still-invisible atoll was very close. Soon, the atoll appeared.²⁰

This voyage weakened the hypothesis that the Pacific was settled by chance: these were not castaways, but navigators.²¹

The cost of knowledge

Mau Piailug came from Satawal, in Micronesia. This knowledge was passed on according to strict rules: he himself had been chosen by his grandfather, a master navigator. By sharing it, he broke the taboos surrounding this teaching — to save an art that was in danger of disappearing.²²

A sky divided up differently

In Australia, the constellation of the Emu is not made of stars: it is drawn by the dark areas of the Milky Way, the clouds of dust that hide the stars behind them. The eye is drawn to the emptiness between the stars rather than to the stars themselves.

Nothing obliges us to divide up the sky the way we do. A constellation is not in the sky: it is a figure that a culture traces there.

An essential caveat: these traditions most often reach us through old transcriptions, made by people who did not know the sky. The weak link in the chain is not the oral tradition but the way it was written down.

What images really show

To the question “Do nebulae really look like the photos?”, the answer is almost always no.

A telescope takes a series of black-and-white images, each through a filter that lets through only one wavelength. A colour is then assigned to each before they are superimposed.

Not a trick, a translation: the image shifts light that the eye does not perceive into the visible range. The colours represent real differences — but they are chosen colours.
The relevant question is the one from our page on the giant planets: not “is it real?”, but “what had to be translated?”

Note 1

Useful records of eclipses extend back to about 720 BC.

Stephenson, F. Richard, Morrison, Leslie V. and Hohenkerk, Catherine Y., “Measurement of the Earth’s Rotation: 720 BC to AD 2015”, Proceedings of the Royal Society A, vol. 472, 20160404, 2016