See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/226630792

Άbdul-Ramān al-Şūfī and His Book of the Fixed Stars: A Journey of Re-
discovery

Chapter · January 2011

DOI: 10.1007/978-1-4419-8161-5_7

CITATIONS READS

21 3,541

3 authors, including:

Ihsan Hafez Wayne Orchiston

11 PUBLICATIONS   60 CITATIONS   
University of Southern Queensland (Toowoomba, Australia)
634 PUBLICATIONS   3,171 CITATIONS   
SEE PROFILE
SEE PROFILE

Some of the authors of this publication are also working on these related projects:

ICOA-9 Proceedings View project

Jenny Andropoulos' Research View project

All content following this page was uploaded by Wayne Orchiston on 20 May 2019.

The user has requested enhancement of the downloaded file.

Άbdul-Ra hm 
. ān al-Şūf ī and His Book of the
Fixed Stars: A Journey of Re-discovery

Ihsan Hafez, F. Richard Stephenson, and Wayne Orchiston

Abstract  The Book of the Fixed Stars was written by the Persian astronomer
Άbdul-Ra      hmān
· al-Şūf ī around AD 964, and is one of the most important medieval
Arabic treatises on astronomy. al-Şūf ī’s work contains an extensive star catalog as well
as detailed star charts for the 48 classical constellations, and draws on material in
Ptolemy’s Almagest. At present no English translation of al-Şūf ī’s treatise exists. This
paper summarizes a detailed study by the first author for a Ph.D. degree in the Centre
of Astronomy at James Cook University (Townsville, Australia) which will include for
the first time a complete English translation of the main parts of al-Şūf ī’s major work.
This paper includes a brief biography of al-Şūf ī, along with information on
the extant manuscripts of the Book of the Fixed Stars, the structure of the book
and star catalogue, and the star maps and charts. A major finding which we high-
light in this paper is al-Şūf ī’s stellar magnitude estimates which were based upon
a unique three-step intermediate magnitude system that he developed. al-Şūf ī also
identified and commented on more than 100 new stars that were not listed in the
Almagest or other early star catalogs. al-Şūf ī’s contribution to astronomy reverber-
ated throughout history, extending to recent times.

1 Introduction

al-Şūf ī’s Book of the Fixed Stars, dating from around AD 964, is one of the most
important medieval Arabic treatises on astronomy. This major work contains
detailed star charts and an extensive star catalogue, which lists star co-ordinates and
magnitude estimates. Other topics include descriptions of nebulae and Arabic folk
astronomy.

Ihsan Hafez (*)
Centre for Astronomy, James Cook University, Townsville, QLD 4811, Australia
e-mail: [email protected]

W. Orchiston et al. (eds.), Highlighting the History of Astronomy in the Asia-Pacific Region, 121
Astrophysics and Space Science Proceedings, DOI 10.1007/978-1-4419-8161-5_7,
© Springer Science+Business Media, LLC 2011
122 Ihsan Hafez et al.

Thus it is surprising that at present no English translation of al-Şūf ī’s treatise

exists. In order to remedy this, one of us (IH) is currently conducting a detailed
study of this book for a Ph.D. in the Centre for Astronomy at James Cook University,
and this will include the preparation of what will eventually be a complete English
translation.
Topics to be discussed in the present paper include a brief biography of al-Şūf ī,
the extant manuscripts of al-Şūf ī’s treatise, the structure of the book, the star
catalogue, and the star maps and charts. This paper also includes some of the major
findings of this doctoral research, such as the magnitude estimates and the unique
three-step intermediate magnitude estimate adopted by al-Şūf ī, as well as the
identification of more than 100 new stars which were mentioned by al-Şūf ī in his
constellation commentaries but were not included in the Almagest or any other
ancient star catalogues that preceded al-Şūf ī’s treatise.
The subtitle chosen for this paper and for the doctoral thesis at James Cook
University is “A Journey of Re-discovery”, as this is a journey to re-discover the
man, and the book which had an important influence on astronomy and helped
mould our understanding of the celestial sphere.

2 al-Şūf ī: A Biography

Surprisingly, we know relatively little about al-Şūf ī’s life and career. However, we
do know that his full name was Άbdul Ra hma - Ibn ΄Umar, Ibn
·  n, Abu΄l Husayn,
·
Mo hammad,
· al-Rāzī, al-Şūf ī.
al-Şūf ī was born in AD 903 in the city of Rayy, south east of what is now called
Tehran, the capital of Iran. He died in AD 986 but we do not know where, most
probably in Shiraz or maybe in Baghdad (which was the political capital as well as
the cultural center of that period). From his name we can deduce that al-Şūf ī’s
father’s name was ΄Umar, signifying that he was a Sunni Muslim rather than a Shiite
Muslim – which was the dominant sect of the Buwayhid Dynasty at that time.
al-Rāzī indicates that he came from the town of Rayy, while al-Şūf ī signifies that he
or his family was part of a Şūf ī religious sect.
From the introductory chapter of his book we know that al-Şūf ī lived most of his
life between the provinces of Isfahan and Fars in Iran. He also wrote that he visited
Dainaour, which is the home of the famous scholar and astronomer Abu Ḣanifa
al-Dainaouri. He also visited Isfahan to research a celestial globe constructed by
another important astronomer of that period. In his book, al-Şūf ī explains the
reasons why he wrote The Book of the Fixed Stars. He mentions that his book was
dedicated to the Buwayhid ruler, Άd. ud al-Dawla, who was a great patron of astronomy
and was himself an accomplished scholar and astronomer. In his book al-Şūf ī also
comments upon several other works, such as al-Battāni’s star catalog which was
written in AD 880, as well as al-Dainaouri’s book on old Arabic astronomical tradi-
tions, and of course Ptolemy’s Almagest.
Άbdul-Rah.m ān al-Şūf ī and His Book of the Fixed Stars: A Journey of Re-discovery 123

From other historical records we learn that al-Şūf ī also wrote a Zīj, which is an
astronomical handbook (Kennedy 1956). Unfortunately it no longer exists. He also
wrote several treatises on astrology, which we also know very little about. Available
treatises on the astrolabe and celestial globes reveal that al-Şūf ī was an accom-
plished instrument-maker (Kennedy et al. 1983) and that a celestial globe which he
constructed existed in Cairo in AD 1043 but can no longer be located. Historical
records (Sayili 1960) also show that al-Şūf ī built an observatory at Shiraz, in Iran.
Many influential astronomers in the past based their astronomical tables on
al-Şūf ī’s work, including al-Bīrūnī in AD 1030, the authors of the Alfonsine tables
in AD 1252 and the famous prince and astronomer, Ulugh Bēg in AD 1437. In
more recent times, al-Şūf ī’s work has been discussed by Ideler (1809), Argelander
(1843), Knobel (1917) and Fujiwara and Yamaoka (2005). However, al-Şūf ī’s
name was sometimes mis-spelled or mis-written. Thus, he has been referred to as
Abolfazen (which we find in an Italian translation of the Alfonsine Tables), Esophi
by Leo Africanus and Azophi by the Spanish Jewish astronomer Ibn ‘Ezra (see
Kunitzsch 1989).

3 The Book of the Fixed Stars

al-Şūf ī’s star catalog was based on Ptolemy’s classical work, the Mathematike
Syntaxis, which was written in AD 125 and was later called the Almagest by the
Arabs. al-Şūf ī updated Ptolemy’s stellar longitudes from AD 125 to AD 964 by
adjusting for precession. In his book he described the methods he used in con-
structing his catalog and especially in calculating precession. For his epoch of AD
964 he applied the most accurate Arabic precession constant available at that time
of 1° in 66 years rather than the correct value of 1° in 71.2 years, thereby adding 12°
42¢ to Ptolemy’s longitudes. Over the 839 years between the tables of Ptolemy and
al-Şūf ī, precession would actually amount to 11° 47¢. Hence by using 12° 42¢ al-Şūf ī
over-corrected Ptolemy’s stellar longitudes by 55¢. al-Şūf ī would not have been
aware of this over-correction because his calculations were based on the Almagest.
The original Arabic name for al-Şūf ī’s book was Şuwar al-Kawākib al-­Thamāniyah
wa al-Ārba'een, which is simply translated as The Forty-eight Constellations.
However, it was later known by other names, the most famous of which was Kitāb
al-Kawākib al-Thābita, or The Book of the Fixed Stars. al-Şūf ī’s original Arabic
text contained 55 astronomical tables, plus star charts for 48 constellations. al-Şūf ī
commented in detail on each constellation, and supplemented this material with star
charts. These tables and charts were written in the same order as in the Almagest,
and were divided into three main groups. The first group contained 21 northern
constellations: Ursa Minor, Ursa Major, Draco, Cepheus, Bootes, Corona Borealis,
Hercules, Lyra, Cygnus, Cassiopeia, Perseus, Auriga, Ophiuchus, Serpens, Sagitta,
Aquila, Delphinus, Equuleus, Pegasus, Andromeda and Triangulum. The second
group contained the 12 constellations of the zodiac: Aries, Taurus, Gemini, Cancer,
124 Ihsan Hafez et al.

Fig. 1  The constellation “Lepus” (The Hare). At the top of the figure is an image of the constellation
(in duplicate). In the lower part is the table of stars in this constellation, including their ecliptical
coordinates and estimated magnitudes.

Leo, Virgo, Libra, Scorpio, Sagittarius, Capricorn, Aquarius and Pisces. The last
group contained 15 southern constellations: Cetus, Orion, Eridanus, Lepus, Canis
Major, Canis Minor, Argo Navis, Hydra, Crater, Corvus, Centaurus, Lupus, Ara,
Corona Australis and Piscis Austrinus. By way of illustration, Figure 1 shows the
constellation Lepus (The Hare). At the top of the figure we can see two images of
the constellation, and below these is a table of the stars in this constellation.
Άbdul-Rah.m ān al-Şūf ī and His Book of the Fixed Stars: A Journey of Re-discovery 125

3.1 The Structure of al-Şūf ī’s Book and Star Catalog

Since al-Şūf ī’s work was based on the Almagest, al-Şūf ī used the same structure
and layout. Thus, his book was divided into four main sections:
1 . An Introductory Chapter
2. The Northern Constellations (of which there are 21).
3. The Constellations of the Zodiac (of which there were 12).
4. The Southern Constellations (of which there are 15).
The description of each constellation is divided into three parts. The first part
is a detailed written commentary describing the position of the stars, their numbers
and magnitudes, as well as many other details. al-Şūf ī also tried to identify the
stars or groups of stars according to the old Arabic tradition, by giving their old
Arabic names and what the Arabs said about them. The Arabic text in Figure 2 is
from a copy of al-Şūf ī’s book which was produced by Ulugh Bēg in the fifteenth
century, and is a very well-written and clear manuscript. This was one of the

Fig. 2  The detailed

written commentary on the
constellation Ursa Minor
describing the position of
the stars, their numbers,
and magnitude as well as
many other details.
126 Ihsan Hafez et al.

Fig. 3  The table showing the

coordinate and magnitude
values for the constellation
Ursa Minor.

main manuscripts that we used for the English translation. The second part of each
constellation chapter is a table showing the coordinates and magnitudes of the
stars (see Figure 3). al-Şūf ī used ecliptical coordinates, as did Ptolemy before him.
The last part of each constellation chapter contains the star charts (e.g. see Figure 4).
There were two charts, depicting the stars as they actually appeared in the sky and – in
reverse – as they were drawn on a celestial globe.

4 Translating The Book of the Fixed Stars

The main effort to search for the hidden astronomical treasures in al-Şūf ī’s book
started with the translation of this work from Arabic to English, and especially the
constellation commentaries. For every constellation al-Şūf ī wrote a commentary
which describes in detail the number of stars, their locations and their magnitudes.
The layout of this translation is depicted in Figure 5, which shows in Figure 5a
the star catalog for the constellation of Ursa Major. Next to it, in Figure 5b, is the
Άbdul-Rah.m ān al-Şūf ī and His Book of the Fixed Stars: A Journey of Re-discovery 127

Fig. 4  The dual charts depicting the stars as they appear in the sky and as they are drawn on a
globe for the constellation Ursa Minor.

corresponding English translation. At the top of this table al-Şūf ī noted that he
added 12° 42¢ to Ptolemy’s longitude to allow for precession. The first column in
Fig. 5b lists the number of the star in the constellation. The second gives the descrip-
tion or name of the star. This sometimes includes the star’s color, Arabic name and
explanation of the position of the star in the constellation. The third group of col-
umns lists the ecliptical longitude coordinates. It was also customary to divide the
ecliptic into twelve 30° divisions. Therefore, when describing the longitude, al-Şūf ī
first wrote the number of that division then the remaining degrees and minutes in
128 Ihsan Hafez et al.

Fig. 5  The table or star catalog for the constellation Ursa Major (left) (a) along with the corresponding
English translation (b).
Άbdul-Rah.m ān al-Şūf ī and His Book of the Fixed Stars: A Journey of Re-discovery 129

order to depict the complete longitude value. The fourth column gives the latitude,
and also specifies the direction of the star north or south of the ecliptic. The fifth
group of columns lists the latitude coordinates, and the final column records al-Şūf ī’s
estimate of the apparent magnitude of each star.

5 Extant Manuscripts of al-Şūf ī’s Book

Before a detailed study of al-Şūf ī’s book and the English translation could begin,
the extant manuscript copies of the Book of the Fixed Stars had to be located, and
examined. It is a measure of the popularity of this book that many manuscripts
are still preserved in libraries throughout the world. However, the tracking down
of some of these involved extensive worldwide travel and much library research.
This resulted in a total of 35 different manuscripts being located, and copies of
the major ones, needed for this study, were acquired. Table  1 lists the existing
manuscripts of al-Şūf ī’s book known at the time of this study, and the 35 copies are
distributed among 20 libraries in 15 different countries.
These 35 extant copies merely represent the ‘tip of the iceberg’. As might be
imagined, various copies of the Book of the Fixed Stars were written by hand, time
and time again, and were passed down from one generation to the next. So there may

Table 1  Geographical distribution of extant copies of al-Şūf ī’s Book of the Fixed Stars
No. of copies of the
Country City Library Book of the Fixed Stars
Denmark Copenhagen Royal Library 1
Egypt Cairo The Egyptian Dar books 1
England London British Library 5
Oxford Bodleian Library 3
France Paris Bibliotheque Nationale 4
Germany Berlin Ahlwardt 1
India Hyderabad Asafiya 1
Iran Tehran Majles 2
Italy Bologna Collection Marsigli 1
Vatican Rossi 1
Lebanon Beirut American University of Beirut 1
Qatar Doha Museum of Islamic Art 1
Russia St. Petersburg Bibliotheque Imperiale 3
Spain Madrid Library Escurial 1
Tunisia Tunisia Bibliotheque Nationale 1
Turkey Istanbul Topkapi Sarayi 3
Istanbul Sulaymania Library 1
USA New York Metropolitan Museum of Art 2
Princeton Princeton University Library 1
Washington Library of Congress 1
130 Ihsan Hafez et al.

have been hundreds of copies in existence at any one time. Unfortunately the origi-
nal manuscript, written by al-Şūf ī, has not survived, but we do have the next best
thing – a copy made by his son.

6 Identifying the Manuscript for Translation

Unfortunately not many of the manuscripts listed in Table 1 could be used for this
study once the following criteria were considered: manuscript age, legibility, correct-
ness of the information contained and completeness. As a result of this analysis,
two manuscripts were selected to form the basis of the translation and discussion:
‘Marsh144’ in the Bodleian Library in Oxford and manuscript ‘MS5036’ in the
Bibliotheque Nationale in Paris. The Marsh144 manuscript is the oldest surviving
copy of the Book of the Fixed Stars and dates to AD 1009, just 23 years after al-Şūf ī’s
death. This manuscript was actually written by al-Şūf ī’s son. A facsimile of this manu-
script was acquired for this study and was used as the basis for the English transla-
tion. Meanwhile, MS5036 in Paris was copied much later, in AD 1430, but it is much
better written and is a more complete version. Various differences were found in the
specific contents of these two manuscripts, which were probably due to scribal
errors at the time the manuscripts were copied. Where differences in listed stellar
coordinates and magnitudes were found, the values in Marsh144 were used in the
preparing the English translation. Since this manuscript was older than the Paris
manuscript and was scribed by al-Şūf ī’s son it was presumed to be more accurate.
Although there has been no English translation of the Book of the Fixed Stars, a
French translation was published by the Danish astronomer Hans Karl Frederik
Schjellerup in 1874. Schjellerup used the Copenhagen manuscript, ‘MS83’, which
dates to AD 1601 for this translation. In 1956 an Arabic copy of al-Şūf ī’s book was
published in Hyderabad (India) by Dār al-Āthār al-΄Uthmāniyah, who made use of
several different manuscripts. This version contains many mistakes and is of limited
usefulness.

7 The Maps in al-Şūf ī’s Catalog

One of al-Şūf ī’s innovations in charting the stars was the production of dual illustrations
of each of Ptolemy’s constellations. One illustration was as portrayed on a celestial
globe, and the other as viewed directly in the night sky. By way of example, Figure 6
is a picture of the constellation Cancer. The upper figure shows the constellation
as seen on a celestial globe and the lower figure as it is seen in the sky. In this Figure
the stars drawn in gold are considered part of the constellation while stars drawn
in red are bright enough to be noticed but lie outside the constellation. The stars in every
constellation were divided into two groups. Stars in the first group formed the main
image of the constellation, while stars in the other group were outside that image.
Άbdul-Rah.m ān al-Şūf ī and His Book of the Fixed Stars: A Journey of Re-discovery 131

Fig. 6  The constellation

cancer. The upper figure
shows the constellation as
seen on a celestial globe.
The lower figure shows
the constellation as it is
seen in the sky.

At the end of the chapter on Ursa Minor, al-Şūf ī explains why he provided two
different star maps and how these should be used:
For every constellation we have drawn two pictures: one as it is projected on the globe and
the other as it is seen in the heavens. Hence we have covered both of the different cases, so
there is no confusion for anyone who sees that what is viewed on the globe is different from
what is in the heavens. When we want to see the constellation as it [really] is we lift the
book over our heads and we look at the second picture [in the book]. From beneath [the
book] we are viewing [the constellation] as it is seen in the heavens.
132 Ihsan Hafez et al.

Fig.  7  An image of the constellation Ursa Minor taken from the Book of the Fixed Stars, and
projected onto a modern chart, showing the true locations of the stars for comparison.

An important issue concerns the accuracy of these maps. Also, were they really
used as intended by al-Şūf ī? It is apparent that as an observational astronomer
and an instrument-maker al-Şūf ī was very concerned about the accuracy of the
data he had and that they should be used correctly when constructing a celestial
globe. Therefore, in order to investigate these questions one of al-Şūf ī’s star charts
for Ursa Minor is reproduced here in Figure 7 projected onto a modern chart. As can
be seen, the constellation chart in the MS5063 manuscript is fairly precise, indicat-
ing that the star charts in the Book of the Fixed Stars could have been used as al-Şūf ī
intended.

8 Star Names and Modern Designation of Stars

In another section of this doctoral study, a complete list of all the stars in the
constellations in the Book of the Fixed Stars was compiled. This included a
description of each star, as listed in al-Şūf ī’s star tables, and their HR, Bayer and
Flamsteed numbers, so that each star could be correctly identified. This list also
includes the name of the stars according to the old Arabic tradition as mentioned in
al-Şūf ī’s book, as well as the various names which have been given to these stars
throughout history.
Άbdul-Rah.m ān al-Şūf ī and His Book of the Fixed Stars: A Journey of Re-discovery 133

9 Magnitudes of the Stars in the Book of the Fixed Stars

9.1 Comparison Between al-Şūf ī and Ptolemy

The next analysis was to compare al-Şūf ī’s stellar magnitudes with those found in
the Almagest. All stars which showed a difference in magnitude or coordinate values
were identified so that an investigation of the reasons for these discrepancies
could be made. This study was mainly to confirm that al-Şūf ī actually made the
measurements himself.
The results showed that al-Şūf ī’s magnitude values for 520 stars out of the total
of 1,019 stars (i.e. 51%) were identical to those provided by Ptolemy. Therefore
one might wonder whether al-Şūf ī only re-estimated the magnitudes of about half
of the stars observed by Ptolemy. However, upon detailed comparison we found
that out of these 520 stars only 206 stars differed in value from the modern visual
magnitudes by ³0.5 magnitude and 56 stars differed by >1 magnitude. The results
also showed that of these 56 stars, 22 had magnitudes of either 5 or 6. This discrep-
ancy can be understood given how difficult it is to visually estimate the visual
magnitudes of faint stars. Therefore, a level of accuracy of no better than ±0.5
magnitude can be expected for al-Şūf ī and Ptolemy. This conclusion is confirmed
by our calculation of the standard error (see below). However, this does not prove
that al-Şūf ī personally re-estimated the magnitudes of all of the stars in his Book of
the Fixed Stars.

9.2 al-Şūf ī’s Three-Step Intermediate Magnitude System

al-Şūf ī and Ptolemy both added intermediate values to the magnitudes of some
stars. Ptolemy mentioned the words “more-bright” and “less-bright” for certain
stars, while al-Şūf ī expressed these intermediate magnitude values by the words
“Aṡghareh”, which means “less”, or “Akbareh” which means “greater”, or “A'z·  ameh”
which means “much-greater”.
Therefore, in this part of the study a complete magnitude analysis was made
in which al-Şūf ī’s magnitude values were numerically interpreted by constant
differences of 0.25 magnitudes: that is, “+0.25” for “less”, “−0.25” for “greater”
and “−0.5” for “much-greater”. Ptolemy’s two-step intermediate magnitude
difference was interpreted by a difference of −0.3 or +0.3 magnitude. In the transla-
tion of the charts the letters (s) for “less”, (k) for “greater” and (m) for “much
greater” were added. For example, the magnitude of the star 19 Ursa Major was
expressed by al-Şūf ī as “much greater than 3rd magnitude”. This can be interpreted
on the three step scale as 3 minus 0.5, which is equal to magnitude 2.5. The modern
magnitude of this star is 2.44 which is very similar. However, if we are to interpret
the magnitude estimate on a two-step scale – as Ptolemy did – then we end up with
a magnitude of 2.7. Therefore, we believe that al-Şūf ī used what we have termed a
134 Ihsan Hafez et al.

three-step intermediate magnitude system, which was more accurate than Ptolemy’s
two-step system. We think that with this system al-Şūf ī was able to express all mag-
nitude values by a constant difference of 0.25.
One of the main topics in this study was to research this three-step intermediate
magnitude system which would shed new light on the accuracy and independence
of al-Şūf ī’s work. However, in order to analyze this topic further, all the data and
information from al-Şūf ī’s book were collected according to the example shown
in Table 2. The first three columns show the number and the number sequence
of the stars and constellations. The fourth column shows the magnitudes of the
stars according to al-Şūf ī. We used the letters (s) for “less”, (k) for “greater” and
(m) for “much-greater”. The fifth column shows the magnitudes after adjust-
ment for the three-step system and the sixth column for the two-step system.
As explained above, this was done by adding the values +0.25 for “less”, −0.25 for
“greater” and −0.5 for “much-greater” for the three-step system while we added
the values +0.3 or −0.3 for the two-step system. The seventh column shows the
magnitude according to Ptolemy. Here we used the magnitudes which al-Şūf ī
attributed to Ptolemy. The eighth column shows Ptolemy’s magnitudes after adjust-
ment for the two-step system. The ninth and tenth columns show the modern visual
magnitude and the HR number for each star. Then in columns 11 to 13 we made an
accuracy analysis for the magnitudes of al-Şūf ī and Ptolemy by calculating the
difference (D) between those values and the visual magnitudes. We made two kinds
of analysis to see if al-Şūf ī had in mind a two step or three step magnitude system
(D al-Şūf ī-1 and D al-Şūf ī-2). The results of this statistical analysis of all 1,019
stars are summarized in Table 3. From the values listed in this table it would seem
that the mean for the three-step system is slightly better. The standard deviation is
the same whether we apply the three-or two-step system, whereas it is higher with
Ptolemy. The dispersion in al-Şūf ī’s data is thus significantly less than in that found
for Ptolemy’s data.
Given the above values, we cannot prove conclusively that al-Şūf ī chose to
use the three-step system in lieu of the two-step systems, although we believe that
was his intention. The main reason for this is the way al-Şūf ī expressed or described
the values of the stellar magnitudes in his book. Most scholars who have studied
al-Şūf ī’s work used the Schjellerup (1874) translation, which does not differenti-
ate between the two words “Akbareh” and “Athami”. In Schjellerup’s translation
the magnitude was always written as the middle value; for example 4–5 (i.e.
between 4 and 5 magnitude). In their work on Ptolemy, Knobel and Peters (1915)
and later Toomer (1998) and Grasshoff (1990) all relied on Schjellerup’s transla-
tion of al-Şūf ī’s data. They categorised Ptolemy’s magnitudes by the word
“greater” and “less”, and expressed these magnitudes in a two-step system. This
two-step intermediate magnitude was later numerically interpreted by a constant
difference of 0.33 magnitude, especially by Grasshoff. However, when we look
at al-Şūf ī’s text in detail it is evident that he made a clear distinction between
three intermediate magnitudes. For example, if we look at magnitude values in the
constellation Gemini, we see that he made the distinction between (m) and
(k), and he really was not concerned with word repetition or correct sentence
Table 2  Analysis of magnitudes of stars in the Book of the Fixed Stars
al-Şūf ī al-Şūf ī al-Şūf ī Ptolemy Ptolemy Visual
magni- magnitude magnitude magni- magnitude magni- HR D D D D D
Seq Recno Constallation tude adjusted-1 adjusted-2 tude (adjusted) tude number al-Şūf ī-1 al-Şūf ī-2 Ptolemy al-Şūf ī-3 al-Şūf ī-4
1 1 Ursa Minor 3 3.00 3.00 3 3.00 2.02    424 −0.98 −0.98 −0.98
2 2 Ursa Minor 4 4.00 4.00 4 4.00 4.36 6,789 0.36 0.36 0.36
3 3 Ursa Minor 4 4.00 4.00 4 4.00 4.23 6,322 0.23 0.23 0.23
4 4 Ursa Minor 4 4.00 4.00 4 4.00 4.32 5,903 0.32 0.32 0.32
5 5 Ursa Minor 5(k) 4.75 4.70 4 4.00 4.95 6,116 0.20 0.25 0.95 0.20 0.25
6 6 Ursa Minor 2 2.00 2.00 2 2.00 2.08 5,563 0.08 0.08 0.08
7 7 Ursa Minor 3 3.00 3.00 2 2.00 3.05 5,735 0.05 0.05 1.05
8 8 Ursa Minor 4 4.00 4.00 4 4.00 4.25 5,430 0.25 0.25 0.25
1 9 Ursa Major 4 4.00 4.00 4 4.00 3.36 3,323 −0.64 −0.64 −0.64
2 10 Ursa Major 5 5.00 5.00 5 5.00 5.47 3,354 0.47 0.47 0.47
3 11 Ursa Major 5 5.00 5.00 5 5.00 4.60 3,403 −0.40 −0.40 −0.40
4 12 Ursa Major 5 5.00 5.00 5 5.00 4.76 3,576 −0.24 −0.24 −0.24
5 13 Ursa Major 5 5.00 5.00 5 5.00 4.80 3,616 −0.20 −0.20 −0.20
6 14 Ursa Major 5 5.00 5.00 5 5.00 4.56 3,771 −0.44 −0.44 −0.44
7 15 Ursa Major 4(s) 4.25 4.30 4 4.00 4.67 3,624 0.42 0.37 0.67 0.42 0.37
Άbdul-Rah.m ān al-Şūf ī and His Book of the Fixed Stars: A Journey of Re-discovery
135
136 Ihsan Hafez et al.

Table 3  Statistical comparison of the magnitude systems of al-Şūf ī and Ptolemy

Astronomer Mean Standard deviation
al-Şūf ī three-step −0.06 0.59
al-Şūf ī two-step −0.09 0.59
Ptolemy +0.07 0.71

structure. This example shows that he expressed 4(m) and 4(k) consecutively then
4(k) twice. He also mentioned several successive examples of (s). Therefore, the
assumption of word repetition is not valid in this case because if al-Şūf ī was
concerned with correct grammatical structure then why did he not use other words
for (s) “Asgharih”. He repeated the word “Asgharih” (s) many times in many
locations throughout his book.

10 Stars Mentioned by al-Şūf ī that Are Not in the Almagest

In his written comments on the constellations al-Şūf ī mentioned some additional stars
that were not included in Ptolemy’s star catalog, but surprisingly he did not include
these stars in his tables even though he identified many of them in detail, describing
their locations and estimating their magnitudes. One reason why al-Şūf ī did not include
these additional stars may have been out of respect for Ptolemy, whose catalogue had
long been a standard reference work in this field. In his introductory chapter, al-Şūf ī
clearly stated that the tables he produced were based upon Ptolemy’s work.
In our study we have identified 128 of these ‘missing’ stars, 59 in the Northern
constellations, 41 in the Zodiac constellations and 28 in the Southern constellations.
al-Şūf ī specifically mentioned these stars in his constellation commentaries but not
in the tables, and he clearly said that these stars were not included in the Almagest.
For identification of these stars we included their HR numbers, along with their
magnitudes as derived by al-Şūf ī and their modern magnitudes.

11 Nebulae in al-Şūf ī’s Book

We find many references in modern history of astronomy books to al-Şūf ī and

those few nebulae which he identified. Therefore in this study all of these nebulous
objects were located and all the commentaries which were written by al-Şūf ī were
summarized. Just to give an example of the importance of al-Şūf ī’s comments
about nebulae, we would like to quote here a remark which he made more than
1,000 years ago: “The two lines (of stars) begin from the nebula which is close to
the 14th star [in the constellation of Andromeda] …”. Here he is describing the
constellation of Andromeda, and for the first time in recorded history he mentions
the location of a nebula which is now identified as M31, the spiral galaxy in
Andromeda. The location of this is shown in Figure 8.
Άbdul-Rah.m ān al-Şūf ī and His Book of the Fixed Stars: A Journey of Re-discovery 137

Fig. 8  al-Şūf ī’s image

of the constellation
Andromeda. The ‘Great
Nebula in Andromeda’ is
close to the 14th star, which
is marked by the arrow.

12 Old Arabic Astronomical Traditions in al-Şūf ī’s Book

Finally, in the last section of this doctoral study the old Arabic astronomical tradi-
tions mentioned in al-Şūf ī’s book were examined. We tried to summarize the names
of the stars and/or asterisms for the constellations that were used in Arabic folk
astronomy as per the explanations provided by al-Şūf ī. For example, the Arabs
located the star Alcor which they named al-Suhā. However, this star was not men-
tioned by Ptolemy or any of the other ancient Greek astronomers. The Arabs con-
sidered this star to be a faint star which they used to test their eyesight. The study of
old Arabic astronomical traditions is a fascinating work in its own right, and shows
the level of sophistication of the ancient Arabs and how the stars were considered a part
of their everyday lives.

13 Concluding Remarks and Publishing

al-Şūf ī and his Book of the Fixed Stars have a very important place in the history
of Arabic observational astronomy. al-Şūf ī not only corrected observational errors
in the works of his predecessors, such as the famous Arab astronomer al-Battāni,
but he also exposed many of the faulty observations found in the various versions of
 138 Ihsan Hafez et al.

the Almagest. He carefully defined the boundaries of each constellation, and

recorded magnitudes and positions of stars using new and independent observations
which he made.
After this doctoral project is completed (see Hafez 2010) the first author plans to
publish this study along with a full English translation of al-Şūf ī’s famous book.

Acknowledgements Grateful acknowledgment is made to all the libraries and museums with
al-Şūf ī manuscripts for kindly permitting us to study and reproduce these for research purposes,
but especially the Bodleian Library in Oxford, the Bibliothèque Nationale de France in Paris and
the British Library in London. We would also like to thank all those librarians and fellow research-
ers who helped and supported our work during the time we spent working on this project.

References

Argelander, F.W.A., 1843. Uranometria Nova. Berlin, Verlag von Simon Schropp.
Dar Al-Athar Al-Uthmania, 1956. Book of the Fixed Stars. Hyderabad, Dāiratu’l-Ma‘ārif-il-
‘Oṣmānia (in Arabic).
Fujiwara, T., and Yamaoka, H., 2005. Magnitude systems in old star catalogues. Journal of
Astronomical History and Heritage, 8, 39–47.
Grasshoff, Gerd, 1990. The History of Ptolemy’s Star Catalogue. New York, Springer.
Hafez, I., 2010. Abdul-Rahman al-Şūf ī and his Book of the Fixed Stars: A Journey of Re-discovery.
Unpublished Ph.D. Thesis, Centre for Astronomy, James Cook University, Townsville, Australia.
Ideler, Ludwig, 1809. Untersuchungen uber den Ursprung und die Bedeutung der Starnnamen.
Berlin (in German).
Kennedy, E.S, 1956. A Survey of Islamic Astronomical Tables. Philadelphia, Transactions of the
American Philosophical Society.
Kennedy, E.S, et al., 1983. Studies in the Islamic Exact Sciences. Beirut, American University of
Beirut.
Knobel, E.B., 1917. Ulugh Beg’s Catalogue of Stars (1437). Washington, Carnegie Institute of
Washington Publication No. 250).
Knobel, E.B., and Peters C.H.F., 1915. Ptolemy’s Catalogue of Stars. Washington, The Carnegie
Institute of Washington.
Kunitzsch, Paul, 1989. The Arabs and the Stars. London, Variorum.
Sayili, Aydin, 1960. The Observatory in Islam. New York, Arno Press.
Schjellerup, H.C.F.C., 1874. Description des Étoiles Fixes Composées au Milieu du Dixième Siècle
de Notre Ère par l’Astronome Persan Abd-al-Rahman al-Şūf ī. St Petersbourg, Commissionaires
de l’Academie Imperial des Sciences (in French).
Toomer, G.J., 1998. Ptolemy’s Almagest. Princeton, Princeton University Press.

View publication stats