Episodic Survey of the History of the Constellations

The illustrations on this page have been compiled from a variety of sources. They are reproduced in accord with 'fair use' provisions unless copyright is otherwise noted. If advised that copyright has been infringed I will immediately remove the particular illustration(s).

Return To Section Index Page

O: Modern Western Constellations

29: Al-Sufi's star catalogue

A depiction of a Medieval Islamic astronomer. Thought by some to represent Abd al-Rahman al-Sufi. The instrument is a triquetrum; an ancient astronomical instrument for determining altitudes of celestial bodies. It was one of the most popular astronomical instruments until the invention of the telescope. It is related to the back-staff/cross-staff.

The illustration above is a section of a painting (miniature) showing astronomers working at the observatory of Taqi al-Din at Istanbul circa 1577. In the entire 16th-century picture of Taqi al-Din's observatory there are 16 persons, illustrating all activities with instruments in the observatory. (None of the instruments have come down to the present-day. However, some of the manuscripts depicted on the bookshelf are now in the University Library, Leiden.) The particular painting is from the epic poem Shahinshah-nama by 'Ala ad-Din Mansur-Shiazi. It was written in honour of Sultan Murad III who reigned from 1574 to 1595. Though it is common to speak of Arabic astronomy the more correct term would be Arab-Islamic astronomy. Many of the astronomers (and peoples) were not Arabs but were from the regions of (modern-day) Iran, Iraq, and Afghanistan. Arabic was the scientific language and lingua franca for followers of the Islamic religion. The language of the religion was Arabic. It is correct to speak of Greek science being passed to the Arabs. Arab rulers of Arab states, for example the 'Umayyad dynasty (which collapsed in the 740s), funded and patronised the transmission process through Syriac sources. The 'Abbasid dynasty which followed can also be considered as an Arab regime.

A page from Al-Sufi's book on the constellations Kitab suwar al-kawakib (Book of the constellations of the Fixed Stars). The book was written for and dedicated to the Buwayhid ruler Fana Khusrau, titled Adud al-Dawla (made Emir of Iraq in 949 CE and died in 982 CE), who was a great patron of astronomy and had erected an observatory at Shiraz. Al-Sufi’s main source for his treatise is the star-catalogue compiled by the Alexandrian astronomer Ptolemy (called "BatlamiyUs" in Arabic). Ptolemy's star-catalogue was included in his major work Mathematika Syntaxis. In the preface to Kitab suwar al-kawakib, al-Sufi mentions that he referred to as many translations of the Almagest as he could find, having noticed discrepancies between them. Paul Kunitzsch has noted that al-Sufi's treatise mostly follows the wording of Isliaq b. Ijunayn's translation, but obtains star-positions from a range of versions.

Al-Sufi (al-Sûfi) was one of the greatest Islamic astronomers of the medieval period.

Al-Sufi’s best known (and most important) work is Kitab suwar al-kawakib (Book of the constellations of the Fixed Stars). "Fixed Stars" is a term describing the stars, which were thought to be fixed to the surface of a large celestial sphere, turning around the Earth. (The "Wandering Stars" are the planets, which were seen to move independently and at different paces, across the celestial sphere.) Al-Sufi's work on the Fixed Stars is one of the most important works of Islamic uranography. The book is based on Ptolemy's Mathematike Syntaxis (Almagest).

The Greek constellation set was readily adopted in the Islamic Middle Ages. (Islamic astronomy was a mixture of indigenous work, together with Indian, Persian, and Greek astronomy.) Knowledge of the fixed stars in Greek-based Arab-Islamic astronomy was derived mainly from Ptolemy's Almagest which contained a catalogue of 1025 stars arranged in 48 constellations (circa 150 CE). The first critical revision of Ptolemy's catalogue of fixed stars was carried out by al-Sufi. However, al-Sufi adopted Ptolemy's basic scheme and pattern of constellations. He did not add or subtract stars from Ptolemy's star list and neither did he re-measure their (frequently incorrect) positions. (Al Sufi’s constellation maps were based mostly on the star tables of Ptolemy's Almagest. Al Sufi’s book has quite precise coordinates and illustrations of the 48 constellations described by Ptolemy.) Also, though al-Sufi based his constellation drawings (figures) on the classical Aratean tradition his figures have a distinct oriental character.

Arabic translations of Ptolemy’s Mathematika Syntaxis were made during the 9th-century, including those of Sahi al-Tabari, al-Hajjãj b. Yusuf b. Matar, and Isaq b. llunayn (died 877 CE), whose 827/828 CE version was later corrected by Thäbit b. Qurra (died 90 l CE).

Abd al-Rahman al-Sufi (Al-Sufi; known in Western Europe by the Latinized name Azophi) was born in 903 CE in Rayy (near modern Tehran), Persia and died in 986 CE. Al-Sufi lived and worked mostly in Isfahan (Persia) at the court of Emir Adud ad-Daula. (He worked in both Isfahan (Iran) and Baghdad (Iraq).) He is most widely known, and became most influential, for his translation and partial revision of Ptolemy's star catalogue as the Book of the constellations of the Fixed Stars (Kitab suwar al-kawakib). It was published circa 964 CE and it was frequently copied and translated. It had considerable influence on European astronomy.

Al-Sufi was both a translator and author. He was involved in translating Hellenistic astronomical works (that had been centred in Alexandria) into Arabic, especially Ptolemy's Almagest. He wrote numerous works on astronomical, astrological, and mathematical subjects. His most outstanding work is his illustrated book on the constellations (Kitab suwar al-kawakib). In this work he comprehensively describes the 48 classical constellations, which were established by Ptolemy and transmitted to the Islamic world in translations of Ptolemy's Almagest. Like all other Islamic scholars of the period he wrote all his works in Arabic (the scientific language of the Arab-Islamic world). In producing his own version of the star catalogue in Ptolemy's Almagest al-Sufi introduced many traditional Arabic star names. (Most star names used by al-Sufi (and his contemporaries) were direct translations of Ptolemy's descriptions.) It was the first attempt to relate (integrate/synthesise) the Greek stars of Ptolemy's star catalogue with the indigenous (traditional) Arabic star names and constellations (the Arab anwa tradition). As a result of Greek translations into Arabic in the 8th and 9th centuries the Greek constellation traditions and the old Arabian traditions became mixed together. In the 10th-century al-Sufi made a special effort to separate the 2 star traditions. In his book, al-Sufi usually gave the name and position of each star in the Greek tradition first and then supplemented that with Arabian traditions about the particular star. (Information concerning the traditional Arabic stars were appended by al-Sufi to his descriptions of the first 5 of the 48 Ptolemaic constellations.) The Bedouin Arabs regarded single stars as representing people and animals. Many of the the original meanings of the named stars had been forgotten by the time of al-Sufi. Because some of the Arabic star names were centuries old their meanings were lost to al-Sufi and his contemporaries, and they remain unknown today. When al-Sufi published Book of the constellations of the Fixed Stars, his own version of Ptolemy's star catalogue in the Almagest (not its original Greek title), he introduced many individual star names. Another problem is that al-Sufi used anwa texts from the Islamic period as his sources. This somewhat limits their connection with indigenous pre-Islamic Arabic anwa traditions.

Al-Sufi was a friend and teacher of the Buyid sultan 'Adud al-Dawlah to whom he dedicated the book. In the introductory chapter al-Sufi states it was the desire of this prince to know the fixed stars, their position within the constellation figures and in relation to the zodiac, which induced him to write the book. The original copy of al-Sufi's manuscript is no longer extant.

At least one later copyist included an index of the constellations at the beginning of the manuscript. An index of the constellations was not originally part of al-Sufi's work.

In Kitab suwar al-kawakib the description of each constellation comprises the following four sections: (1) A general discussion of the constellation and its individual stars. Also included in this section is al-Sufi's criticism of the Ptolemaic tradition and also notices of al-Sufi's own observations. (He described all the stars catalogued by Ptolemy and added his own criticism in each individual case.) (2) A record of the indigenous Arabic star names (predating Arabic contact with Greek astronomical science) falling within each constellation, and the exact identification of each of these stars with the corresponding Ptolemaic stars. (3) Two drawings (paintings) of the constellations, one depicting the constellation as it is seen on the celestial globe (i.e., as seen by an observer looking inwards towards earth), and the other one as it is seen in the sky (i.e., as seen by an observer looking outwards from earth). (Elly Dekker labels the different views "sky view" versus "globe view.") (4) A table of the stars making up each of the constellations, including a verbal description of each star's location and its longitude, latitude, and magnitude. (The magnitudes given were according to al-Sufi's own observations.) This table closely follows the arrangement of Ptolemy's star catalogue in the Almagest. In this book al-Sufi also described the boundaries of the constellations.

In al-Sufi's Kitab suwar al-kawakib the constellation figures and the individual stars comprising them are shown separately (i.e., separated from each other) without any information on their relative positions being given. No sky map (with all the constellations charted) appears in the book. An important development was al-Sufi numbered the individual stars in each of his constellation drawings. The catalogue of star coordinates given by al-Sufi is believed to be Ptolemy's star catalogue (coordinates) updated for precession.

Al-Sufi had the intention to integrate Ptolemy's star catalogue with Arab star tradition and terminology, and also to define boundaries for the constellations. The constellation drawings in Al-Sufi's catalogue comprising Kitab suwar al-kawakib quickly became canonical. This influence was later introduced into Europe.

Al-Sufi's book on the constellations (and the constellation drawings contained in it) served as models for further work on the fixed stars in the Arab-Islamic world for many centuries. His description of the constellations became the basis for all later studies. (Al-Sufi's drawings of the constellation figures established a standard typology for the constellations - in Europe too.) Islamic constellation figures were introduced into Europe as least as early as the 13th-century. It is stated by some sources that al-Sufi's Kitab suwar al-kawakib was never translated into Latin. This is incorrect. A fully illustrated translation was (anonymously?) made in Palermo, Sicily, in the 12th-century at the instigation of Guillaumine II (1166-1189).the Norman king of Sicily. (A Latin manuscript of it, titled Liber de locis stellarum fixarum, now resides at the Arsenal Library in Paris as part of MS # 1036 (a collection of astronomical manuscripts with the general title Liber de stellis stellarum). The Liber de locis stellarum fixarum contains 49 illustrations representing the constellations and the signs of the zodiac. The Arabic-Persian iconography, based on characters in the One Thousand and One Nights story, is kept. The Parisian copy, made anonymously, is the oldest existing Latin copy. It is dated to the third quarter of the 13th-century and is believed to be a copy of the translation made earlier in Sicily. (It is sometimes identified as being made from a manuscript in Spain.) (The Holy Roman Emperor Frederick II, at his court in Palermo, Sicily, financed translations of Arabic works into Latin.) Al-Sufi's book was also fully translated into Spanish by Alfonso X ("Alfonso the Wise") of Leon and Castile, as Libros del Saber de Astronomia. It was through these translations that it influenced the star names of used in western Europe. Its contents were transmitted into Europe and in medieval Europe its constellation drawings were imitated in numerous Latin astronomical manuscripts.

Petrus Apianus (Peter Apian) (1495-1552) the 16th-century German astronomer and geographer took star names from al-Sufi's book on the constellations and placed them on his star charts and mentioned them in his writings. (See the chapter on the constellations in his Astronomicum Caesareum (1540).)

It has been long believed that al-Sufi's book on the constellations was the (exclusive) key source for the establishment of star names in western Europe. (Note: A few proper star names, such as Polaris (North Star), are not Arabic.) However, this now appears to be over-simplified and somewhat incorrect. The science historian Owen Gingerich writes ("Islamic Astronomy," Scientific American, Volume 254, April, 1986, Pages 68-?): "It now seems that his [i.e., Ptolemy's] 14th- and 15th-century Latin translators went to a Latin version of the Arabic edition of Ptolemy himself for the star descriptions, which they combined with al-Sufi's splendid pictorial representations of the constellations. Meanwhile the Arabic star nomenclature trickled into the West by another route: the making of astrolabes." (Some of these astrolabes have distorted Arabic names for stars.)

The astrolabe, essentially a two-dimensional model of the sky, was originally a Greek invention (dating circa 3rd-century BCE) to enable the problems of spherical astronomy (i.e., the prediction of star positions) to be solved. It moved with the spread of Islam through North Africa into Spain (Andalusia). It would appear that England, due to the scientific activity centred at Oxford, was the conduit for the introduction of the astrolabe from Spain into western Europe in the late 13th-century and the 14th-century.

Historians have not settled the debate over who was responsible for the transmission of the astrolabe from Muslim Spain into Europe and when and where the astrolabe first appeared in Europe. However, by 1030 CE at the latest some European scholars possessed astrolabes and were teaching their use. Early Christian recipients of Arab astronomy (including the astrolabe) included Gerbert of Aurillac and Hermannus Contractus. Gerbert of Aurillac (circa 946-1003 CE) (later to become Pope Sylvester II (999-1003 CE)) spent several years (967-969 CE) studying in Spain in the Christian-held city of Barcelona and also possibly in the Moorish-held cities of Córdoba and Seville. (He originally went to the cathedral school of Vic, in the province of Catalonia which was on the frontier of Moorish Spain. As a result there was considerable communication between Catalunya and the Muslims of al-Andalus to the south.) It is thought that Gerbert of Aurillac may have been the author of a description of the astrolabe ( The Book of the Astrolabe which was the first Latin text explaining the astrolabe and providing instructions for the construction of an astrolabe) that was edited by the Benedictine monk Hermannus Contractus (1013-1054 CE) some 50 years later.

Many of these early astrolabes that were introduced into Europe carried both Arabic and Latin star nomenclature. It has been noted by Paul Kunitzsch that star names that appear on medieval astrolabes in Europe are often quite different from star names that appear in star lists in medieval manuscripts. The astrolabe was widely used in Europe during the late Middle Ages and the Renaissance with its popularity peaking in the 15th- and 16th-centuries. (The astrolabes of the 11th-16th centuries were an important instrument for predicting star positions.) It became one of the basic astronomical education tools. Europeans eventually began to manufacture astrolabes. In the 15th-century European astrolabe manufacturing was centred in Augsberg and Nuremberg in Germany, with some manufacturing also in France. By the middle of the 17th-century astrolabes were being manufactured all over Europe.

(The astrolabe was a 'flattened' and more portable version of the armillary sphere. It was a two-dimensional representation of the celestial sphere and was used for solving problems in celestial geography.)

Islamic astronomical globes also became highly prized in medieval western Europe. Some were purchased and used without translation of Arabic terms as Arabic was a scientific language in medieval Europe. Some were purchased and copied with the names of stars and constellations being translated into Latin versions of Arabic. Arabic was frequently used on European globes along with other scientific languages. Globes were manufactured and used with astronomical terms in Latin, Greek, and Arabic. (Astronomical globes were popular in the Arab-Islamic world.)

European astronomers and celestial map makers began to use Arabic star names in preference to Latin names circa 12th-century CE. This practice kept on increasing with the increasing ease of European access to Islamic texts and instruments. By the end of the 15th-century the process of European adoption of Arabic star names was essentially complete. (According to Emilie Savage-Smith it has been established that a nearly complete Arabic version of al-Sufi's treatise on the constellations must have reached Germany by the 1530's, for information in it was employed in a limited way by Peter Apian, who from 1527 to 1552 was professor of mathematics at the University of Ingolstadt.) The "Arabic" names were retained in the formal, scientific nomenclature until the end of the 19th-century.

Latinised Arabic star names were well established in Europe by the time Johann Bayer published his Uranometria in 1603. Bayer did not make any attempt at reforming this practice of denoting star names. The most relevant feature of Bayer's atlas was his recording of popular names for important stars, drawn from the works of Ptolemy and his successors, to assure that all known stars could be identified with those he had listed in his atlas. Bayer also introduced the use of lower case Greek letters to name and organise the stars. However, the Bayer system is full of inaccuracies.

Al-Sufi's star catalogue was in turn revised by the the 15th-century Timurid governor (of Transoxiana and Turkestan) and astronomer Ulugh Beg, at his observatory in Samarkand, Uzbekistan. The most important star catalogue of Islamic astronomy is that of Ulugh Beg in the early 15th-century CE.   

Appendix 1:

Richard Allen in his highly influential book Star-Names and Their Meanings (1899) stated that European star names came chiefly from the Arabs. Allen, who had no real understanding of Arabic, also concluded that many Arabic star-names were actually translations of Greek descriptive terms transmitted through Arabic into Latin (and from Latin into English and other languages). When the linguist Maio Pei made a check of 183 English star-names he concluded that 125 were from Arabic, and 9 were from Arabic-Latin. (See: Story of the English Language by Mario Pei (1967; Page 225).) Paul Kunitzsch and Tim Smart (A Dictionary of Modern Star Names (2006; Page 11) write: "A statistical analysis of the 254 star names here presented reveals that (counting five double entries only once) 175 names (= 70%) are Arabic and 47 (= 19%) are are Greek or Latin." The modern authority on such matters is Paul Kunitzsch.

Appendix 2:

"The star names used in the classical Islamic world were derived from two distinct sources: (1) the various (non-standardised) names originated by pre-Islamic groups of Bedouins (the nomadic desert Arabs of the Arabic Peninsula) (older body), and the main body (younger group) of indigenous Arabic star/asterism names were probably formed in the period 500-700 CE (prior to the introduction of Islam in the 7th-century CE); and (2) those transmitted from the Greek world. As Greek astronomy and astrology were accepted and elaborated, primarily through the Arabic translation of Ptolemy's Almagest, the indigenous Bedouin star groupings were overlaid with the Ptolemaic constellations that we recognize today."  (Islamicate Celestial Globes by Emilie Savage-Smith (1985)  Page 114.) "A third set of names derived from the Arabic were bestowals, often ill-based, by early modern Western astronomers even though they had never been used by Arabian astronomers. Most of these names have disappeared. Thuban, alpha Draconis, is an exception." (Early Astronomy by William O'Neill (1986) Page 162.) (Some European astronomers inventing their own constellations also invented their own Arabic star names. The earliest likely example is the Dutch orientalist and mathematician Jacob Golius (1596-1667).) Both Emilie Savage-Smith and William O'Neill are reliant on the fundamental studies of Paul Kunitzsch. An example of the first category of star names of Arabic origin is Aldebaran from Al-Dabaran. An example of the second category of star names of Arabic origin is Fomalhaut from Fam al-Hut. An example of the third category of star names derived from Arabic is Thuban, alpha Draconis.

Appendix 3:

There have been very few translations from the Arabic of Al-Sufi's Kitab suwar al-kawakib. In 1831, the preface was published in Arabic with accompanying French translation, by Michel Caussin, who referred to Ar2488 (13th-century CE), Ar2489 (dated 1266 CE), and Ar2490 (dated 1516 CE), of which all three are in the Bibliothèque Nationale de France. In 1874, H. C. F. Schjellerup published a full translation of the text, also into French, using Ms83 (dated 1601 CE) in the Royal Library of Copenhagen, and Ms191 (dated 1606 CE) in the Institute of Oriental Studies, St Petersburg. A critical edition of the Arabic text was published in Hyderabad, in 1954 by the Dãiratu'1-Marif-i1-Osmania, based mainly on Ar5036 (circa 1437 CE), in the Bibliothèque Nationale de France, but also collated with Bodleian Marshl44 (dated 1009-1010 CE), Topkapi A.3493 (dated 1131 CE), Vatican Ross. 1033 (dated 1224 CE), Berlin Staatsbibliothek A.5658 (dated 1233 CE) and Asafiya, Hyderabad (date ?). The edition included the text of the Urjüza on the constellations (purportedly written by a1-Sufi's son), translated extracts from the Preface to Schjellerup's translation of 1874, and plates from 8 al-Sufi manuscripts, including British Library Or.5323. Paul Kunitzsch perceptively noted the 1954 publication is "deplorably rich in misprints and other errors." In 1969, the 1250 CE Persian translation of al-Sufi's treatise was published in facsimile, as Tarjama-i suwar-i kawakib. In 1986, the earliest known copy of Kitab suwar al-kawakib, (dated 1009-1010 CE, Marshl44, Bodleian Library Oxford) was also published in facsimile by the Institute for the History of Arabic-Islamic Science in Frankfurt. In 2010, Ihsan Hafez completed his PhD thesis, Abd al-Rahman al-Sufi and his book of the fixed stars: a journey of re-discovery (the first English translation of al-Sufi's star atlas).

Appendix 4:

Al-Sufi's Kitab suwar al-kawakib is a catalogue of the names of individual stars and star-groups from the system of Arabian folk astronomy. He includes mention of the  Anwã' constellations, a group of 28 individual stars and small constellations along the ecliptic, which constituted a calendar and meteorological system. Paul Kunitzsch proposed a new assessment of al-Sufi's account of Arab star-names pointing out that al-Sufi uncritically depended on the premise that the Arabian astronomy literature he consulted was indeed a correct account of the names used by the Bedouins. The only literature available to him consisted of these imprecise sources (in terms of astronomical information) as poetry and lexicography, in which al-Sufi simply assumed that all this material made genuine references to actual stars and star-groups. Some of the lore did derive from authentic tradition. However, according to Paul Kunizsch this may only be said for one sixth of the recorded material. The rest consisted mainly of invention or distortion by poets, who sometimes recorded variants of star-names with poetic license. A star-name might appear in a poem as a simile, and then also be distorted to fit the meter. Unfortunately, al-Sufi took all references to star-names as read, and introduced them uncritically to scientific literature, "in many cases arbitrarily fixing a rather fluid tradition." His account was thereafter accepted as authoritative, and became a standard reference for the subject. Johann Bayer's star atlas Uranometria published in 1603 CE referred to both al-Sufi and Ptolemy, and established modem usage.

Appendix 5:

Al-Sufi also clarifies previous scholarship on native Arab astronomy. Al-Sufi cites the names of individual stars and constellations in native Arabic astronomical tradition, which lie within (or across) each given classical constellation. This pre-Islamic tradition mapped the sky into constellations, some of which are considerably larger than the Greek ones, and are earlier versions of Mesopotamian constellations that were later to form part of the 12-constellation zodiac. Specific identification is given for both individual stars, and those arranged in collective groups. Also, explanations are offered for unusual names, and common errors are pointed out. Occasionally al-Sufi repeats folk proverbs or beliefs around the star-names and he also explains that accounts from folk astronomy are often contradictory as to the exact identity of certain stars. He also remarks that recorded information on folk astronomy has been set down by people unfamiliar with both Arab folk astronomy and "scientific" classical astronomy.

Appendix 6:

Various similarities between the Phaenomena and Kitab suwar al-kawakib show that the latter is in many ways the Islamic world's answer to the Western classic of popular astronomy, presented in a more scientific manner, and strengthened with the updated, revised versions of Ptolemy's star-tables. The Phaenomena was translated into Arabic in the early 9th-century CE, probably by the Jewish astronomer Sahi b. Bishr, at the court of Abu'l-Tayyib Tahir b. al-Ijusayn. (died 822/823 CE), (general of Caliph al-Ma'mUn, and later the governor of Khurasän). Aratus was well-known to Severus Sebokht, the bishop of Kennesrin (one of the leading figures of ecclesiastical, philosophical, and scientific culture of late antique Syria), who quotes liberally from the Phaenomena in his treatise on the constellations, written in 661 CE.

"Severus made two major contributions to astronomy. The first (written in 660 CE), a Treatise on the Astrolabe, is based on a lost work by Theon of Alexandria, the contents of which Severus preserved in his own work. ... Severus's other astronomical work (generally entitled Treatise on the Constellations ) was written in 660, subsequent to that on the astrolabe. Eighteen original chapters are extant. The work begins with five chapters forming a scientific critique of astrological and poetic claims about the origins and significance of the constellations. In them, Severus shows that the figures of the constellations are not arranged in the heavens through natural means but rather are a result of human imagination. Importantly, Chapter 4 features extracts from the Phaenomena of Aratus concerning many of the constellations. The remaining 13 chapters (6–18) are devoted to a scientific analysis of the heavens and the Earth. (Article: Severus Sebokht by John McMahon, in The Biographical Encyclopedia of Astronomers edited by Thomas Hockey et. al. (2007, Pages 1044-1045).)"


Copyright © 2006-2018 by Gary D. Thompson

Return to top of page.

This Web Page was last updated on: Saturday, November 4, 2017, 9:00 pm.

This Web Page was created using Arachnophilia 4.0 and FrontPage 2003.

You can reach me here by email (but first delete the obvious attempt in the email address to foil the spammers): garyREMOVE.thompson10@bigpond.com

Return To Site Contents Page