In this guide

A tenth-century astronomer in the Persian city of Shiraz looked up at a sky that no longer exists. The stars he charted have drifted, the constellations have shifted shape, and yet his descriptions still guide our eyes. Al-Sufi’s Book of Fixed Stars (Kitāb Ṣuwar al-Kawākib, completed around 964 CE) is the first comprehensive Arabic star catalog, describing 48 constellations with 1,018 individual stars, their magnitudes, colors, and positions. It is also the oldest surviving illustrated astronomical manuscript, a bridge between Greek science and Islamic observation.

The book matters to us because it preserved what the ancients saw. Al-Sufi worked at the court of the Buyid dynasty in Isfahan and Shiraz, and he did something no one had done before: he checked the old Greek catalog of Ptolemy against the actual sky and recorded the differences. The result is a document that is half science, half poetry, and entirely obsessed with the question of what a star looks like when you really look at it.

What exactly is the Book of Fixed Stars?

What exactly is the Book of Fixed Stars

The Book of Fixed Stars is a revision and expansion of Ptolemy’s Almagest, the second-century Greek star catalog that dominated astronomy for a millennium. Al-Sufi, whose full name was Abd al-Rahman al-Sufi (903-986 CE), did not simply translate the Greek text. He re-observed every star, corrected errors, and added his own measurements.

His catalog organized the sky into the 48 classical constellations, following the Greek tradition. For each constellation he provided:

  • A description of the figure it represented
  • The number of stars in each part of the figure
  • Each star’s longitude, latitude, and magnitude
  • Notes on color and brightness
  • Two illustrations: one as seen from outside the celestial globe, one as seen from inside

The book survives in dozens of manuscript copies, the oldest dating to around 1009 CE, now held in the Bodleian Library at Oxford. The illustrations are remarkable: each constellation appears twice, once as the sky god would see it from outside, once as the earthbound observer sees it from below. That dual perspective was al-Sufi’s innovation, and it changed how astronomers thought about mapping the heavens.

How did al-Sufi describe the stars themselves?

How did al-Sufi describe the stars themselves

The heart of the catalog is the individual star descriptions. Al-Sufi gave each star a position, a magnitude, and often a color. He used Ptolemy’s six-point magnitude scale, where first magnitude is the brightest and sixth is the faintest visible to the naked eye. But he corrected Ptolemy’s values where his own observations disagreed.

His descriptions are precise and practical. For example, he describes the stars of Ursa Major by their positions in the bear’s body: the star at the snout, the star at the eye, the three stars of the tail. He names the bright stars individually, often using the Arabic names that had developed over centuries of Bedouin astronomy.

One of his most famous observations concerns the Andromeda Galaxy. He noted a “little cloud” near the star that Ptolemy had listed as the 28th star of Andromeda. Modern astronomers identify this as M31, the Andromeda Galaxy, the nearest major galaxy to our own. Al-Sufi is the first person on record to have described it, nearly a thousand years before telescopes.

He also described the Large Magellanic Cloud, visible from the southern latitudes of the Arabian Peninsula, though he could not have known it was a satellite galaxy. He called it the “white ox” and placed it in the constellation of the southern fish. For observers in Europe, these southern stars were invisible, and al-Sufi’s descriptions were the only record of them.

What did the Arabic star names mean?

The star names al-Sufi used did not come from Greek astronomy. They came from the pre-Islamic Bedouin tradition, a parallel sky that existed alongside the formal Greek constellations. The Bedouin saw the sky not as mythological figures but as practical markers: animals, tents, camels, and the seasonal patterns that governed desert life.

The star we call Sirius was al-Shi’ra, “the bright one.” Betelgeuse was yad al-Jawza, “the hand of the central one.” The Pleiades were al-Thurayya, a name that appears in pre-Islamic poetry and in the Quran. These names carried weather lore: the heliacal rising of certain stars marked the start of the rainy season, the time to plant, the time to migrate.

Al-Sufi recorded both systems side by side. For each Greek constellation he gave the Arabic names of its principal stars, often noting which Bedouin asterism they belonged to. This dual naming preserved a tradition that would otherwise have been lost. When European astronomers later adopted Arabic star names into the modern catalog, they were inheriting this Bedouin sky through al-Sufi’s pages.

Why did he draw the constellations twice?

The dual illustrations are the most distinctive feature of the Book of Fixed Stars. Each constellation appears twice: once on a celestial globe, seen from outside, and once in the sky, seen from the earth. The difference is not cosmetic. A figure seen from outside the globe is mirror-reversed compared to how it appears from below.

This matters for practical astronomy. If you hold a star chart above your head to match the sky, you need the earthbound view. If you are constructing a celestial globe, you need the god’s-eye view. Al-Sufi provided both, and the manuscripts show the figures in full human or animal form, with stars marked at the appropriate points on their bodies.

The illustrations are also beautiful. The figures are drawn in the style of Persian manuscript painting, with gold and lapis lazuli, the constellations rendered as dignified figures in flowing robes. The dragon of Draco coils across the page; the hunter Orion strides with his club raised. These are not abstract charts but portraits of the sky as a living thing.

How does the Book of Fixed Stars compare to other star catalogs?

Ptolemy’s Almagest was the direct source, and al-Sufi never hid his debt. But the comparison is revealing. Ptolemy’s catalog was compiled in Alexandria around 150 CE, based partly on the earlier work of Hipparchus. It listed 1,022 stars. Al-Sufi listed 1,018, a difference of four, but he corrected many positions and magnitudes.

The later European catalogs, such as Tycho Brahe’s of 1598 and Bayer’s Uranometria of 1603, built on the same Greek foundation. But they had telescopes and better instruments. Al-Sufi worked with nothing but his eyes, an armillary sphere, and a plumb line. That his positions are accurate to within a degree or two is a proof of his patience.

The Chinese star catalogs of the same period, such as the Tang dynasty’s Yisizhan, organized the sky completely differently, into lunar mansions rather than constellations. The comparison shows how much of astronomy is cultural: the same stars, arranged into different stories. We have our own guide to the lunar mansions elsewhere in this library, and the contrast between those systems is instructive.

What do people get wrong about al-Sufi’s work?

The most common misconception is that al-Sufi was merely a translator. He was not. He observed, corrected, and expanded. His catalog is an original work of astronomy, not a copy.

Another error is the claim that he discovered the Andromeda Galaxy. He described it, yes, but he did not know what it was. He called it a “little cloud” and assumed it was a fixed feature of the sky. The discovery of its true nature came centuries later, with telescopes. What al-Sufi did was record what he saw without forcing it into a theory.

People also assume the star names are purely Arabic. Many are Greek in origin, transliterated into Arabic and then back into Latin. The star we call Aldebaran comes from the Arabic al-dabaran, “the follower,” because it follows the Pleiades. But the constellation figures themselves are Greek. The Book of Fixed Stars is a synthesis, not a pure Arabic creation.

How can you use the Book of Fixed Stars today?

For the practitioner, the catalog is a practical tool. The star names carry correspondences that have survived into modern astrology and magic. Knowing which star is which, and what the old observers saw in it, gives depth to any working that involves the fixed stars.

The dual illustrations also offer a meditation. To see a constellation from outside the celestial sphere is to take the god’s perspective, to understand the sky as a whole rather than from one point on earth. Al-Sufi’s book invites that shift in vision.

We recommend reading it alongside a modern star chart, preferably on a clear night. Find the stars he describes, note their colors, and compare what you see with what he recorded. The sky has shifted in a thousand years, but the bright stars are still there, and his descriptions still hold.

FAQ

Was al-Sufi the first to describe the Andromeda Galaxy? Yes, his description of a “little cloud” near the star in Andromeda, written around 964 CE, is the earliest known record of what we now call M31.

How many stars did al-Sufi catalog? He listed 1,018 stars across 48 constellations, revising Ptolemy’s earlier catalog of 1,022 stars.

What language was the Book of Fixed Stars written in? Arabic, and it was later translated into Persian and other languages. The oldest surviving manuscript dates to about 1009 CE.

Are al-Sufi’s star positions still accurate? They are remarkably accurate for naked-eye observation, within one or two degrees, though precession has shifted the stars’ positions since his time.

The Book of Fixed Stars is a record of a sky that no longer exists, but the stars themselves remain. Al-Sufi’s descriptions connect us to a thousand years of observation, from the Bedouin of the desert to the astronomers of the Buyid court. For more on how the fixed stars were organized into systems, see our guide to the lunar mansions and the related glossary of star names. The sky is still there, waiting to be read.