Content uploaded by Mustafa Yilmaz
Author content
All content in this area was uploaded by Mustafa Yilmaz on Oct 12, 2017
Content may be subject to copyright.
Historical mosque orientation in Turkey: Central-Western Anatolia Region,
115 0e1590
Mustafa Yilmaz
Department of Geomatics Engineering, Afyon Kocatepe University, ANS Campus, TR-03200 Afyonkarahisar, Turkey
Abstract
The orientation of buildings in the ancient civilisations has been referred to the movements of several celestial bodies above the horizon on characteristics
dates (two solstices and equinoxes). However, Muslims have used a sacred direction (qibla) towards Kaaba located in the courtyard in the Sacred Mosque
in Mecca to pray and to perform various ritual acts in their daily lives since the early days of Islam. Thus, the mosques had then to orientate towards the
qibla direction, being indicated by a niche in the focal point of the qibla-wall wherever they were building on the Earth. This article focuses on the mosque
orientation in Turkey before the seventeenth century with regard to the astronomical knowledge derived from Arabic sources before Islam, mathematical
theory and spherical computation derived from Greek sources and traditions based on the early Islamic period. The mosque orientations are compared to
the qibla directions that are used in sacred geography which was determined by the producers of folk astronomy and in the application of the geometric
or trigonometric formulae in mathematical astronomy.
Ó2012 Elsevier Ltd. All rights reserved.
Keywords: Mosque orientation; Qibla;Qibla-wall; Sacred geography; Folk astronomy; Mathematical astronomy
The term ‘orientation’refers to the position and the direction of
a structure at the site of its construction. Social or religious rules
and traditions explain the relation between the orientation of the
structure and the point of sunrise or the point where a star rises on
a specific date.
1
In ancient civilisations, people tried to understand
the changing visible form of the sun, the moon and the stars by
observing and recording their movements. The obvious effects of
these movements on earthly concerns (for example, the length of
day-light, solstices and seasons) attested to their importance. There
are many historical records of cultures building markers to align
with sites on the horizon to mark the solstices. Research in the field
of archaeoastronomy has produced a considerable body of knowl-
edge about the orientation of prehistoric and historical tombs,
temples and other buildings world-wide.
2
The orientation of
historical buildings has been referred to several celestial bodies
above the horizon on characteristic dates (for example, solstices
and equinoxes).
3
Astronomical alignment in this sense is evident in
the form of monuments at Stonehenge,
4
in Egyptian and Mayan
pyramids
5
and in Chinese tombs.
6
Determining the orientation of historical buildings may yield
important information about the social character and religious
constitution of a civilisation. Several of the world’s major
E-mail address: mustafayilmaz@aku.edu.tr
1
M. Hoskin, Tombs, Temples and Their Orientations: A New Perspective on Mediterranean Prehistory, Bognor Regis, UK, 2001.
2
A.F. Aveni, Whither archaeoastronomy, in: A.F. Aveni (Ed.), World Archaeoastronomy, Cambridge, 1989, 3e12; I. Liritzis and H. Vasiliou, Further solar alignments of Greek
Byzantine churches, in: I. Liritzis (Ed.), Ancient Watching of Cosmic Space and Observation of Astronomical Phenomena, SEAC 14th 2006, Mediterranean Archaeology and
Archaeometry, Special Issue 6, 3 (2006); J.A. Belmonte, A.T. Gaspar, A.P. Betancort and R. Marrero, On the orientation of pre-islamic temples of north Africa: a re-appraisal
(new data in Africa proconsularis), in: I. Liritzis (Ed.), Ancient Watching of Cosmic Space and Observation of Astronomical Phenomena, SEAC 14th 2006, Mediterranean Archaeology
and Archaeometry, Special Issue 6, 3 (2006); G. Pantazis, R. Korakitis, E. Lambrou and D. Sinachopoulos, Researching the orientation of monuments: the church of the Great
Meteoro monastery, in M.O. Altan (Ed.), Proceedings of XIXth CIPA Symposium, Antalya, Turkey, 2003.
3
I. Liritzis and H. Vassiliou, As tronomical orientations of ancient temples at Rhodes and Attic a with a tentative interpretation, Med iterranean Archae ology &
Archaeometry 2 (20 02) 69e79; I. Liritzis and H. Vassiliou , Archaeoastronomical orientation of seven significant ancient Hellenic temples, Archaeoas tronomy Internat ional
17 (2 0 0 3) 9 4 e100; I. Liritzis and H. Vassiliou , Were Greek temples oriented towards aurorae, Astronomy & Geo physics 47 (2006) 14e18.
4
G.S. Hawkins, Stonehenge decoded, Nature 200 (1963) 306e308.
5
C.B. Smith and K.E. Parmenter, Khufu and Kukulcán, Civil Engineering 74 (2004) 38e49, 79e80.
6
I. Charvátová, J. Kloko
cník, J. Kolma
s and J. Kostelecký, Chinese tombs oriented by a compass: evidence from paleomagnetic changes versus the age of tombs, Studia
Geophysica et Geodaetica 55 (2011) 159e174.
Contents lists available at SciVerse ScienceDirect
Journal of Historical Geography
journal homepage: www.elsevier.com/locate/jhg
0305-7488/$ esee front matter Ó2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jhg.2012.06.002
Journal of Historical Geography 38 (2012) 359e371
religions eJudaism, Christianity and Islam ehave historically
observed the practice of orientation to a sacred geographic
direction. According to the tenets of Judaism, Jews have to face
the direction of Jerusalem while praying though in actual practice
Jews have had a rather flexible attitude towards the direction of
Jerusalem and have rarely been entirely precise about orienting
to Jerusalem. In fact, most synagogues from the middle ages to
the eighteenth century placed the ark along a wall that was due
eastward.
7
In early Christianity, the practice of facing the east for prayer
would have been influenced by the pagan sun-worshipping of the
time. However, Christian scholars have also offered numerous
other historical and theological explanations for facing eastward.
8
The most noticeably theological mention is a story told in the Book
of Matthew. ‘As the lightning comes from the East .so shall the
Son of Man appear.’
9
There, it is the scriptural basis for the
Christian belief that Jesus will come from the east. This belief is in
turn the foundation for the ancient Christian practice of facing
eastward while praying. The custom of facing eastward while
praying has affected the orientation of historical churches.
Christian churches have been predominantly oriented along an
EasteWest alignment.
10
Several studies have previously been
carried out for determining the orientation of churches. For
example, Wehner listed the orientation of 72 German churches.
11
Searle presented the orientation of 27 churches in England.
12
Abrahamsen investigated the orientation of 330 Romanesque
churches in Denmark.
13
Hoare and Sweet studied the orientation
of 183 medieval churches in England.
14
Ali and Cunich analysed
the orientation of 143 English churches.
15
And finally, Liritzis and
Vassiliou have investigated the orientation of twelve Byzantine
churches from Greece towards the rising sun at the saint’sname
day.
16
Among the major religions, Islam has paid the most consider-
able attention to the importance of orientation to a sacred
direction. The sacred direction in Islam, called the qibla in all
Islamic languages, has a paramount importance in the daily lives of
Muslims while praying and performing various ritual acts for
fourteen centuries.
17
Islamic tradition further prescribes that
certain acts such as reciting the Qur’
an, announcing the call to
prayer and ritual slaughtering of animals for sacrifice are also to be
performed in the qibla direction. In addition, bodily functions are to
be performed perpendicular to the qibla direction.
18
Also Muslim
graves and tombs were laid out so that the body would lie on its
right side and face the qibla to rise on the Day of Resurrection.
19
A
verse of Qur’
an, regarded by Muslims as the Word of Allah, enjoins
Muslims to face Kaaba in Mecca during their prayers as the qibla.
20
‘Turn your face toward the Sacred Mosque [in Mecca]. And wher-
ever you [Muslims] are, turn your faces toward it [in prayer].’
21
It is
a requirement for every Muslim to be able to face the qibla in
prayer, as accurately as possible.
The mosque is the most prominent religious building in Islam,
serving both as a place of worship and as a community centre.
Muslim scholars have emphasised the role of the mosque as
a condition for the fulfilment of an Islamic way of life.
22
For four-
teen centuries, with regard to the fundamental unifying force of the
concept of the qibla within Islam, mosques have been oriented
towards the qibla direction, being indicated by a niche in the focal
point of the qibla-wall.
23
The qibla-wall is marked as being nearest
to Kaaba and thus guiding the orientation of prayer. Within the last
several decades, the orientation of mosques has been the subject of
many scientific studies. Thus Barmore investigated 298 mosque
orientations in Turkey with the concept of magnetic declination.
24
Bonine examined the orientation of 61 mosques in Morocco
25
and 7
mosques in Tunisia
26
for the structure of traditional Islamic cities.
Avni provided the data for the orientation of 12 mosques in the
settlements in Negev highlands, Palestine.
27
Finally, Pantazis and
Lambrou investigated the orientation of 11 mosques situated in
several regions of Greece.
28
7
D.Z. Levin, Which way is Jerusalem? Which way is Mecca? The direction-facing problem in religion and geography, Journal of Geography 101 (2002) 27e37.
8
See additional details in Levin, Which way is Jerusalem? (note 7).
9
Book of Matthew [24:27].
10
J.R. Ali and P. Cunich, The orientation of churches: some new evidence, The Antiquaries Journal 81 (2001) 155e193; J.R. Ali and P. Cunich, The church east and west:
orienting the Queen Anne churches, 1711e34, Journal of the Society of Architectural Historians 64 (2005) 56e73.
11
H. Wehner, Ueber die kenntnis der magnetischen Nordweisung im fruhen Mittelalter, Das Weltall 18 (1905) 319e356.
12
S. Searle, The church points the way, New Scientist 61, 879 (1974) 10e13.
13
N. Abrahamsen, Orientation of Romanesque churches and magnetic declination in the 12th century in Denmark, GeoSkrifter 23 (1985).
14
P.G. Hoare and C.S. Sweet, The orientation of early medieval churches in England, Journal of Historical Geography 26 (2000) 162e173 .
15
Ali and Cunich, The orientation of churches (note 10).
16
I. Liritzis and H. Vassiliou, Does sunrise day correlate with eastern orientation of Byzantine Churches on significant solar dates and Saint’s days? A preliminary study,
Byzantinische Zeitscrift 99 (2007) 523e534.
17
D.A. King, The orientation of medieval Islamic religious architecture and cities, Journal for the History of Astronomy 26 (1995) 253e274.
18
D.A. King and R.P. Lorch, Qibla charts, qibla maps, and related instruments, in: J.B. Harley and D. Woodward (Eds), The History of Cartography, Vol. 2, Book 1: Cartography in
the Traditional Islamic and South Asian Societies, Chicago, 1992, 189e205; A.J. Wensinck, Kibla: ritual and legal aspects, in: C.E. Bosworth, E.J. van Donzel, B. Lewisand Ch. Pellat
(Eds.), The Encyclopaedia of Islam, new Edition, Vol. 5, Leiden, 1986.
19
D.A. King, World-maps for Finding the Direction and Distance to Mecca: Innovation and Tradition in Islamic Science, Leiden, 1999; B.L. Gordon, Sacred directions, orientation,
and the top of the map, History of Religions 10 (1971) 211e227.
20
King and Lorch, Qibla charts, qibla maps, and related instruments (note 18). Kaaba is a black granite cube-shaped edifice located in the courtyard of the Sacred Mosque
[Masjid al-Haram] in Mecca. The qibla depends on one’s proximity to Kaaba, as Ibn al-Qass (ca. 940) stated: ‘Kaaba is the qibla for the Sacred Mosque [Masjid al-Haram], the
Sacred Mosque is the qibla for the sacred precincts [of Mecca], and the sacred precincts arethe qibla for the inhabitants of the whole World from where the sun rises to where
it sets.’
21
Qur’
an [2:144]. Qur’an, is the central religious text of Islam, which Muslims consider the verbatim word of Allah.
22
Z. Aazam, The social logic of the mosque: a study in the relationships between building typology and urban morphology, in: A.S. Kubat, Y. Güney, O. Ertekin and E.
Eyuboglu (Eds.), Proceedings, 6th International Space Syntax Symposium, Istanbul, Turkey, 2007.
23
Aqibla-wall is one of the typical mosque components. See M. Frishman, Islam and the form of the mosque, in: H.U. Khan and M. Frishman (Eds), The Mosque, London, UK,
1994, 17e41.
24
F.E. Barmore, Turkish mosque orientation and the secular variation of the magnetic declination, Journal of Near Eastern Studies 44 (1985) 81e98.
25
M.E. Bonine, The sacred direction and city structure: a preliminary analysis of the Islamic cities of Morocco, Muqarnas 7 (1990) 50e72.
26
M.E. Bonine, Romans, astronomy and the qibla: urban form and orientation of Islamic cities of Tunisia, in: J.C. Holbrook, R.T. Medupe and J.O. Urama (Eds), African Cultural
Astronomy eCurrent Archaeoastronomy and Ethnoastronomy Research in Africa, Berlin, 20 08, 145e178.
27
G. Avni, Early mosques in the Negev highlands: new archaeological evidence on Islamic penetration of southern Palestine, Bulletin of the American Schools of Oriental
Research 294 (1994) 83e100.
28
G. Pantazis and E. Lambrou, Investigating the orientation of eleven mosques in Greece, Journal of Astronomical History and Heritage 12 (2009) 159e166.
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371360
The history of mosque orientation
The method of mosque orientation towards the qibla has consti-
tuted one of the most intriguing aspects in the history of Islamic
civilisation.
29
The determination of the qibla direction and orient-
ing the mosques towards Kaaba have been intensively investigated
by Muslim scholars and scientists for more than twelve centuries.
30
The qibla-wall of the mosques was intended to be parallel to one of
the walls of Kaaba in mosque orientation. This orientation is ach-
ieved by facing the mosque towards the same astronomical horizon
phenomena as one would be facing when standing in front of the
appropriate wall of Kaaba.
31
The orientation designed to face the
distant Kaaba is the most fundamental aspect of the qibla direction
in the sacred geography of Islam. The Kaaba-focused sacred geog-
raphy was developed from this notion. The world was divided into
sectors centred on Kaaba and the qibla direction in each sector was
defined in terms of astronomical horizon phenomena. Historically,
Muslims have used a number of simple, practical and reliable
techniques to orientate the mosques towards the qibla. Analysis of
Arabic historical, legal and astronomical texts has provided
evidence that several mosque orientation traditions were accepted
at various times and locations (for example, emulating the practice
of the Prophet Muhammad, cardinal or astronomical orientation to
Kaaba, determining the qibla direction along a great circle of the
terrestrial sphere by the application of trigonometric formulae).
32
The mosque orientation method could also change over time at
any particularly location, leading to rather different directions for
the qibla even within the same city.
33
In the first two centuries of Islam, when mosques were beingbuilt
from Andalusia to Central Asia, Muslims had no truly scientific means
of finding the qibla direction.
34
There are several traditions within
Islamic heritage suggesting that the qibla used for mosque orienta-
tion was defined with some flexibility except for Mecca owing to the
absence of accurate geographic or geometric knowledge before the
late secondand the early third century ofIslam.
35
According to these
traditions, the qibla was sometimes determined by using the corre-
sponding direction of the road in which the pilgrims left the city to
Mecca for Hajj,
36
or alternatively the mosques were oriented simply
due south because the Prophet Muhammad (and his Companions)
had prayed directlysouth when he was in Medina (north of Mecca).
37
The mosque orientation with the qibla to the south is often emulated
in converting EasteWest oriented churches to mosques (in such
locations as Jerusalem and Damascus).
38
Furthermore, in other parts
of the early Islamic world, the mosques were sometimes oriented
towards south because in many Muslim lands, the word ‘qibla’has
become the name of the point of the compass, according to the
direction in which Mecca lies; thus the qibla means south in Egypt
and in Palestine.
39
On early Turkish rhomb-cards the south is thus
called al-qibla, while the wind blowing from south is named as the
qibla in Turkey. Nevertheless, in the Turkish language the qibla means
south. It was probably because of that confusion that Eastern
Muslims attached more importance to the southern end of the
compass needle than to the northern one.
40
The fact that the eighth-
century Umayyad Caliphate of Spain had its original roots in Syria e
with their qibla directly south emay have influenced a similar
mosque orientation in the Iberian Peninsula.
41
In the middle ages of Islam, there were two main approaches,
each existing alongside the other, for mosque orientation. The first
one drew from ‘folk astronomy’, ultimately derived from the astro-
nomical knowledge of the Arabs before Islam, which was devoid of
theory and innocent of any calculation. It aimed to face Kaaba and in
its interpretation of the qibla drew heavily on the work of Islamic
legal scholars. Mosque orientation based on folk astronomy was
widely practiced over the centuries. The second approach was that of
‘mathematical astronomy’, derived mainly from Greek sources and
involving both theory and computation. It was used to determine the
direction towards Mecca along a great circle of the Earth globe.
Mathematical astronomy for mosque orientation was practiced only
within the scientific community, and was largely ignored by wider
community (especially legal scholars), although it was more scien-
tifically accurate than the orientation based on folk astronomy.
42
Throughout the Islamic medieval world different mosque orienta-
tions began to be used in different regions and even cities.
43
Mosque orientation based on folk astronomy
Observing the locations and directions of the stars was very impor-
tant and almost instinctualfor Arab observers during the pre-Islamic
era when trading land caravans and sea merchants used the stars for
navigation during the night. The astronomical alignments based on
this folk astronomical knowledge (associated with the rising or
setting of the sun and prominent stars at the solstices or equinoxes)
were widely used for mosque orientation in the middle ages of Islam.
Medieval legal scholars of Islam established practical mosque
orientationtechniques for facing Kaaba. Theyinstructed worshippers
29
D.A. King, Two Iranian world maps for finding the direction and distance to Mecca, Imago Mundi 49 (1997) 62e82.
30
D.A. King, Astronomical alignments in medieval Islamic religious architecture, Annals of the New York Academy of Sciences 385 (1982) 303e312; S.K. Abdali, The Correct
Qibla, 1997, at http://patriot.net/wabdali/ftp/qibla.pdf (accessed 18 January 2012).
31
G.S. Hawkins and D.A. King, On the orientation of Ka’ba, Journal for the History of Astronomy 13 (1982) 102e109.
32
King, Astronomical alignments in medieval Islamic religious architecture (note 30); Hawkins and King, On the orientation of Ka’ba (note 31); King, The orientation of
medieval Islamic religious architecture and cities (note 17); King and Lorch, Qibla charts, qibla maps, and related instruments (note 18); King, World-maps for Finding the
Direction and Distance to Mecca (note 19).
33
King, The orientation of medieval Islamic religious architecture and cities (note 17).
34
D.A. King, Astronomy in the Service of Islam, Aldershot, UK, 1993.
35
M.S.M. Saifullah, M. Ghoniem, A.R. Squires and M. Ahmed, The Qiblah of Early Mosques: Jerusalem or Makkah,2001,athttp://www.islamic-awareness.org/History/Islam/
Dome_Of_The_Rock/qibla.html (accessed 18 January 2012).
36
King, The orientation of medieval Islamic religious architecture and cities (note 17); King, Astronomy in the Service of Islam (note 34). Hajj is the Muslim annual pilgrimage
to Mecca during the twelfth month of the lunar Islamic calendar.
37
King, Astronomy in the Service of Islam (note 34).
38
Bonine, Romans, astronomy and the qibla (note 26); King, The orientation of medieval Islamic religious architecture and cities (note 17); Saifullah, Ghoniem, Squires and
Ahmed, The Qiblah of Early Mosques (note 35).
39
Gordon, Sacred directions, orientation, and the top of the map (note 19).
40
G. Sarton, Query no. 25: orientation of the mihrab in mosques, Isis 20 (1933) 262e264; G. Sarton, Fourth answer to query no. 25: orientation of the mihrab in mosques,
Isis 38, 1e2(1947)95e96.
41
Bonine, Romans, astronomy and the qibla (note 26).
42
King and Lorch, Qibla charts, qibla maps, and related instruments (note 18); King, Two Iranian world maps for finding the direction and distance to Mecca (note 29); A.S.
Massasati, Mapping the direction to Makkah: a cartographic perspective, American Journal of Islamic Social Sciences 19 (2002) 87e94.
43
D.A. King, The sacred direction in Islam: a study of the interaction of religion and science in the Middle Ages, Interdisciplinary Science Reviews 10 (1985) 315e328.
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371 361
to face that part of Kaaba traditionally associated with the region in
which they were located.
44
The orientation of Kaaba itself has
influenced the astronomical orientation of mosques in many parts of
the Islamic world. The astronomical alignment of Kaaba was known
in the seventh century. The first generation of Muslims who were
familiar with Kaaba knew that if one stood in front of any of its four
walls one would be facing significant astronomical horizon
phenomena. The major axis of Kaaba is aligned towards the rising
point of the star Canopus and the minor axis is aligned towards
summer sunrise and winter sunset. These directions are roughly
perpendicular at the latitude of Mecca.
45
Figs. 1 and 2show the
astronomical alignment of Kaaba as implied by the MS Milan Bib-
lioteca Ambrosiana 73 Sup., a treatise authored by a Yemeni astron-
omer, Muhammad ibn Abi Bakr al-Farisi (ca.1290).
46
The astronomical alignments of the sides of Kaaba were
associated with the directions of the four cardinal winds in
certain medieval Arabic texts. Several wind schemes defined in
terms of solar or stellar risings and settings (Fig. 3)werepart
of the folk astronomy and meteorology of pre-Islamic
Arabia.
47,48
These astronomical alignments could be used for mosque
orientation anywhere in the Islamic world in order to ensure the
same axes as Kaaba itself. Hence, the first Muslims knew that, when
facing a particular wall or corner of Kaaba in Mecca, one was facing
a particular solar or stellar rising or setting point; they assumed
that away from Mecca, if one faced in that same astronomical
direction one would still be facing the same wall or corner of
Kaaba.
49
From the ninthcentury onward, a colourfulsacred geography was
developed with a view of Kaaba as the centre and navel of the world.
Every region of the Islamic world was associated with a particular
segment of the perimeter(or wall) of Kaab a andthe qibla was defined
precisely in terms of astronomical horizon phenomena, cardinal
directions or specificwinds.
50
The four corners of Kaaba had already
been named according to the geographical regions they faced,
regarding to the trading contacts: Syria, Iraq, Yemen and West.
Architectural details were also used to define subdivisions. Muslim
legal scholars thus devisedschemes with different number of sectors
radiating out from Kaaba according to its four walls and four corners
(Figs. 4 and 5). In this way, numerous different schemes were
formed, with the result that the various sources gave different qibla
directions for the same region.
51
Muslim legal scholars were equal to
the task of confronting different qibla directions for each region: it
was permissibleto pray in any direction whichwould be within one’s
field of vision when facing Kaaba directly ethat is, about one quad-
rant of the horizon centred on the actual direction.
52,53,54
In the medieval Islamic world, therefore, each major city would
have had a number of accepted qibla directions, although it is
Fig. 1. The alignments of Kaaba implied by the first section of the Yemeni treatise
illustrated in MS Milan Biblioteca Ambrosiana 73 Sup.
Fig. 2. The alignments of Kaaba implied by the second section of the Yemeni treatise
illustrated in MS Milan Biblioteca Ambrosiana 73 Sup.
44
King, Two Iranian world maps for finding the direction and distance to Mecca (note 29).
45
King, Astronomical alignments in medieval Islamic religious architecture (note 30); King, The orientation of medieval Islamic religious architecture and cities (note 17).
46
Hawkins and King, On the orientation of Ka’ba (note 31). Muhammad ibn Abi Bakr al-Farisi was a competent astronomer worked in Aden (ca. 1290) who was the author of
a substantial astronomical handbook with tables (zij), computed especially for Yemen.
47
King, Astronomical alignments in medieval Islamic religious architecture (note 30); King, The orientation of medieval Islamic religious architecture and cities (note 17).
The orientation with a wind theory calls to mind the classical compass winds that refers the association of geographic direction and orientation with winds in ancient Greece
and Rome. As a branch of meteorology, the proper names of the winds were used to denote general cardinal directions of the compass rose. Ancient wind roses typically had
12 winds and thus 12 points of orientation esometimes reduced to 8 or increased to24. The poet Homer (ca. 800 BC) refers to the 4 winds in his Odyssey and in the Iliad. The
philosopher Aristotle, in his Meteorology (ca.340 BC), introduced a 10-to-12wind system. The writer Vitruvius, in his De Architectura (ca. 15 BC), told us ‘Tower of the Winds’in
Athens was built as an architectural representation of an 8-wind system of the geographer Eratosthenes (ca. 200 BC). In the Early Middle Ages, Arab scholars came into
contact with the Greek works and Aristotle’sMeteorology was translated by Hunayn bin Ishaq (809e873). The winds defined in terms of astronomy are also recorded by
medieval Arab philologians and are included, for example, in E.W. Lane, ArabiceEnglish Lexicon, London, 1863.
48
King, The orientation of medieval Islamic religious architecture and cities (note 17). The relationship between the alignment of Kaaba and the wind directions was
recorded in the treatise Kitab al-tafhim li-awa’il sina’at al-tanjim (The Book of Instruction in the Elements of the Art of Astrology) by al-Biruni (973e1048).
49
King, Astronomical alignments in medieval Islamic religious architecture (note 30).
50
King, The orientation of medieval Islamic religious architecture and cities (note 17); King and Lorch, Qibla charts, qibla maps, and related instruments (note 18).
51
King, The orientation of medieval Islamic religious architecture and cities (note 17); King, World-maps for Finding the Direction and Distance to Mecca (note 19). Detailed
information about the qibla-charts can be found in King and Lorch, Qibla charts, qibla maps, and related instruments (note 18).
52
King, The sacred direction in Islam (note 43); King, The orientation of medieval Islamic religious architecture and cities (note 17). There is a prophetic tradition in Islam
that the qibla was determined with some flexibility, except for Mecca. This prophetic saying (hadith) can be found in the hadith collection Sunan al-Tirmidhi, as stated:
‘[Anywhere] between the East and the West is a Qibla.’
53
King, The orientation of medieval Islamic religious architecture and cities (note 17). The earliest known Kaaba-centred geographical scheme was recorded in the book
Kitab al-Masalik w’al-Mamalik (Book of Roads and Kingdoms) by Ibn Khurradadhbih (ca. 820e912).
54
King, The orientation of medieval Islamic religious architecture and cities (note 17). Al-Qazwini (1203e1283) was well-known for his geographical dictionary, Athar
al-bilad wa-akhbar al-ibad (Monument of Places and History of Allah’s Bondsmen).
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371362
probable that only certain directions were deemed ‘correct’during
specific periods, or by certain Muslim legal scholars.
55
Mosque orientation based on mathematical astronomy
The qibla direction varies with the place at which the direction of
Kaaba is to be determined. If the distance from Kaaba is small, its
direction may be determined by a diligent seeker, but when the
distance is great, only astronomers can determine that direction.
56
From the eighth century onwards, Muslim astronomers paid much
attention to the qibla determination (mosque orientation) of any
locality from the geographical coordinates of Mecca and of that
locality, at a time when Islamic civilisation was expanding beyond the
Arabian Peninsula.
57
The required knowledge of geographical coor-
dinates and mathematical calculations were derived mainly from
Greek geographical works that were modified to the Islamic notion of
world geography.
58
There are two options for establishing the qibla direction by
astronomical means: the great circle and the rhumb line directions
from any locality to Mecca.
59
The well-known Muslim astronomer
Ibn al-Haytham (965e1040) defines that: ‘The qibla is the direction
such that when a human observer faces it, it is as if he is looking at
the diameter of the Earth passing through Kaaba.Andtheraycoming
out of his eye in that direction is in the plane of the great circle
passing in the direction of his zenith and the point corresponding to
[the zenith of] Mecca.’
60
Ibn al-Haytham mentions the great circle on
the celestial sphere passing through the zeniths of the location and
Mecca, instead of the terrestrial great circle through the two places.
Muslim astronomers often transformed the qibla determination
problem into some familiar problem of astronomy by projecting the
points on the Earth to those on the celestial sphere. Ibn al-Haytham’s
simple and precise definition is described as the qibla is along the
line of sight to avertical point above Kaaba.Thiscanonlybethegreat
circle direction, not the rhumb line which is not the line of sight.
61
The mathematically determined qibla at any locality was defined
as the direction to Mecca along the great circle on the terrestrial
sphere (Fig. 6). It is invariably measured from the local meridian and
called inhiraf al-qibla, literally the inclination of qibla (to the
meridian). Already in the early ninth century, observations were
conducted in order to measure the coordinates of Mecca and
Baghdad as accurately as possible, with the express intention of
computing the qibla at Baghdad. Once the geographical data are
available, a mathematical procedure is necessary to determine the
qibla. The earliest Muslim astronomers who considered this problem
developed a series of approximate solutions, all adequate for most
practical purposes, but in the early ninth century, if not before, an
accurate solution by solid trigonometry was formulated. The earliest
mosque orientations based on mathematical astronomy were
cartographic, in that they can all be derived by considering plane
representations of the part of the terrestrial globe between the
locality and Mecca. The meridians are treated as parallel straight
lines as well as the parallels of latitude being parallel straight lines.
62
Some of the medieval Muslim solutions to the determination of the
qibla problem use, as an intermediary step, the computation of
Fig. 3. The alignments of the axes of Kaaba with their relationship to the four cardinal
winds as described in medieval Arabic sources. Fig. 4. A simple four-sector scheme of sacred geography associated with the published
text of the Kitab al-Masalik of the ninth-century geographer Ibn Khurradadhbih.
55
Bonine, Romans, astronomy and the qibla (note 26).
56
al-Biruni Abu Reyhan, Muhammad ibn Ahmad, Kitab Tahdid Nihayat al-Amakin li-Tash
ıh Masafat al-Masakin Kitab al-Amakin (The Determination of the Coordinates of
Positions for the Correction of Distances Between Cities), trans. Jamil Ali, Beirut, 1967.
57
King, World-maps for Finding the Direction and Distance to Mecca (note 19).
58
The Greek scientist Ptolemy (ca. 90e168) had a great influence on Islamic astronomy and geography. See A.J. Kimerling, Cartographic methods for determining the qibla,
Journal of Geography 101, 1 (2002) 20e26.
59
The great circle (or orthodrome) is the intersection of the sphere and a plane whichpasses through the centre point of the sphere. For any two distinct points on the surface of
a sphere thereis only one great circlethrough the two points.An arc on a great circle represents the shortest path betweentwo points on a sphere. Therhumb line (or loxodrome)is
a straight line crossing all meridians of longitude at a constant angle. The rhumb line’s property of always having the same bearing is important in navigation.
60
Quoted and translated into English by S.K. Abdali from al-Khattabi and Muhammad al-Arabi, Ilm al-mawaqit: Usuluhu wa manahijuhu, Library Congress # 89-968009,
1986 in Abdali, The Correct Qibla (note 30).
61
Abdali, The Correct Qibla (note 30).
62
King, Astronomy in the Service of Islam (note 34); King, World-maps for Finding the Direction and Distance to Mecca (note 19 ); D.A. King, The earliest Islamic mathematical
methods and tables for finding the directions of Mecca, Zeitschrift für Geschichte der Arabisch-Islamischen Wissenschaften 3 (1986) 82e149.
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371 363
the great arc distance to Mecca. A number of analemma solutions to
the qibla problem have been found in the medieval Arabic sources.
63
The earliest method to determine the qibla direction astronomi-
cally was defined by al-Khwarizmi (780e850) and al-Battani
(858e929), based on the difference of longitudes between Mecca
and a givenplace.
64
Due tothe simplicityof its geometricconstruction,
al-Battani’s geometric method remained inwide use even after more
accurate methods became available.
65
Abu al-Wafa (940e998) proved
the tangent rule in spherical trigonometry and used for the qibla
determination.
66
More exact methods were developed by Habash al-
Hasib (ca. 850) and Ibn al-Haytham (965e1040) using graphical
construction.
67
al-Nayrizi (ca. 897), Ibn Yunus (ca. 985), and al-Biruni
(973e1048) improved the previous methods by spherical trigono-
metric computations.
68
The most comprehensive medieval discus-
sion of mathematical qibla determination is in the Kitab Tahdid
Nihayat al-Amakin li-Tash
ıh Masafat al-Masakin Kitab al-Amakin (‘The
Determination of the Coordinates of Positions for the Correction of
Distances Between Cities’) of al-Biruni. In this account,surveys of the
different mathematical methods developed for the qibla direction,
based either on plane trigonometry or on spherical trigonometry
applied to terrestrial/celestial sphere, were provided. Al-Biruni fixed
scientifically the direction of Mecca from any point on the globe and
described a single method for architects and artisans for the mosque
orientation.
69
In addition to these qibla determinations, practical methods
involving astronomical instruments such as astrolabes and various
types of quadrants were devised to serve well both for performing
astronomical observations and for practically solving spherical
triangles without trigonometric tables or the labour of arithmetic
by numerous researchers.
70
Ibn al-Shatir (1306e1375) made
a much smaller sundial forms part of a compendium to establish
the local meridian and the qibla direction.
71
The qibla was also
determined by solar observations directly at certain times, and
derived from observations using spherical trigonometric calcula-
tions at other times by al-Tusi (1201e1274) and by constructing
a magnetic compass.
72
Fig. 5. A simplified twelve-sector scheme of sacred geography associated with the
geographical work Athar al-bilad by al-Qazwini.
Fig. 6. The qibla direction was considered as a great circle arc between P and Mecca
over the terrestrial sphere.
63
Analemma (the name coming from the title of a book by Ptolemy containing similar procedures) is a projection procedure for reducing the problems on the surface on
a sphere to the problems in a plane and known from Greek mathematical astronomy. See detailed information in J.L. Berggren, A comparison of four analemmas for
determining the azimuth of the qibla, Journal for the History of Arabic Science 4 (1980) 69e80.
64
M. Iqbal, Science and Islam, Connecticut and London, 2007.
65
See King, Astronomy in the Service of Islam (note 34); Abdali, The Correct Qibla (note 30).
66
A. Moussa, Mathematical methods in Abu al-Wafa’sAlmagest and the qibla determinations, Arabic Sciences and Philosophy 21 (2011) 1e56.
67
E.S. Kennedy and Y. Id, A letter of al-Biruni: Habash al-Hasib’s analemma for the qibla, Historia Mathematica 1(1974)3e11; M.T. Debarnot, The zij of Habash al-Hasib:
a survey of MS Istanbul Yeni Cami 784/2, Annals of the New York Academy of Sciences 500 (1987) 35e69; C. Schoy, Abhandlung des al-Hasan ibn al-Hasan ibn al-Haitam
(Alhazen) über die bestimmung der richtung der qibla, Zeitschrift der Deutschen Morgenländischen Gesellschaft 75 (1921) 242e258. See King, World-maps for Finding the
Direction and Distance to Mecca (note 19).
68
G. De Young, Nayrizi: Abu al-Abbas al-Fadl ibn Hatim al-Nayrizi, in: T. Hockey et al. (Eds), The Biograp hical Enc yclopedi a of Astronomers,NewYork,2007;D.A.King,
Ibn Yunus: Abu al-Hasan Ali ibn Abd al-Rahman ibn Ahmad ibn Yunus al-Sadafi, in: T. Hockey et al. (Eds), The Biographical Encyclopedia of Astronomers, New York, 20 07;
M. Yano, Biruni: Abu al-Rayhan Muhammad ibn Ahmad al-Biruni, in: T. Hockey et al. (Eds), The Biographical Encyclopedia of Astronomers,NewYork,2007.
69
King, World-maps for Finding the Direction and Distance to Mecca (note 19); Iqbal, Science and Islam (note 64).
70
An astrolabe is an elaborate inclinometer, historically used to calculate the altitude and azimuth of the sun, moon, planets, and stars and was the most important
astronomical observational device before the invention of the telescope. Its many uses include astronomy, navigation and surveying. An early astrolabe was invented in the
Hellenistic world in 150 BC but was further developedin the medieval Islamic world, where Muslim astronomers introduced angular scales to the astrolabe and added circles
indicating azimuths on the horizon. It was widely used throughout the Muslim world, chiefly as an aid to navigation and as a way of determining the qibla direction. A
quadrant is essentially a graduated quarter of a circle, set up to measure the altitude of celestial objects above the horizon. It was originally proposed by Ptolemy as a better
kind of astrolabe. Several different variations of the instrument were later produced by medieval Muslim astronomers.
71
L. Janin and D.A. King, The Sanduq al-yawaqit of Ibn al-Shatir: an astronomical compendium, Journal for the History of Arabic Science 1 (1977) 187e256.
72
al-Tusi’smethod is valid when the declination of sun is equalto the latitude of Mecca. So, when the sun at noontime is above Mecca (at its zenith), the direction of the sun is the
direction of Mecca. But this method works only forthe Northern hemisphere, because on the Southern hemisphere the sun will not be visible. See Moussa, Mathematical methods
in Abu al-Wafa’sAlmagest and the qibla determinations (note 66); King and Lorch, Qibla charts, qibla maps, and related instruments (note 18). See detailed information about the
use of compass to determine the qibla in P.G. Schmidl, Two early Arabic sources on the magnetic compass, Journal of Arabic and Islamic Studies 1(1997)81e132.
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371364
There were many astronomical handbooks with tables (zijes)
containing sections on determining the qibla by specific mathe-
matical procedures.
73
Tables containing the qibla angle as a func-
tion of latitude and longitude difference from Mecca werecompiled
by Ibn Yunus
74
and al-Khalili (ca. 1365).
75
Al-Khalili’s qibla table
Over the centuries, numerous Muslim scientists discussed the
qibla problem, presenting solutions by spherical trigonometry, or
reducing the three-dimensional situation to two dimensions and
Fig. 7. An extract from al-Khalili’sqibla table, displaying values of the qibla for each degree of longitude difference from Mecca up to 60, entered vertically, and each degree of
latitude from 10to 56(refer note 29).
73
Zij is the generic name applied to Islamic astronomical handbooks that tabulate parameters used for astronomical calculations of the positions of the sun, moon, planets
and stars. See detailed information in D.A. King, J. Samsó and B.R. Goldstein, Astronomical handbooks and tables from the Islamic World (750e1900): an interim report,
Suhayl 2 (2001) 9e106.
74
D.A. King, Ibn Yunus’very useful tables for reckoning time by the sun, Archive for History of Exact Sciences 10 (1973) 342e394. Ibn Yunus computed the altitude of the sun
when it was in the qibla direction of Cairo for each degree of solar longitude. See King, World-maps for Finding the Direction and Distance to Mecca (note 19); King, Samsó and
Goldstein, Astronomical handbooks and tables from the Islamic World (750e1900) (note 73).
75
D.A. King, Al-Khalili’s qibla table, Journal of Near Eastern Studies 34 (1975) 81e122; D.A. King, al-Khalili’s auxiliary tables for solving problems of spherical astronomy,
Journal for the History of Astronomy 4(1973)99e110; G. van Brummelen, The numerical structure of al-Khalili’s auxiliary tables, Physis 28 (1991) 667e698. See King, Samsó
and Goldstein, Astronomical handbooks and tables from the Islamic World (750e1900) (note 73).
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371 365
solving by geometry or plane trigonometry. They also formulated
solutions using calculating devices. But one of the finest medieval
mathematical solutions to the qibla problem was reached in
Damascus for the entire Islamic world. A fourteenth-century
Syrian timekeeper (muwaqqit), Shams al-Din al-Khalili, tabulated
the qibla as a function for each degree of latitudes from 10
to 56
North, as well as for 33
30
0
(the latitude of Damascus) and each
degree of longitudes between 60
East and 60
West of Mecca
with remarkable accuracy.
76
Al-Khalili tabulated virtually all the
spherical functions tabulated by Ibn Yunus, but for the latitude of
Damascus, 33
30
0
N and for the obliquity of ecliptic 23
31
0
and
devised a set of auxiliary functions which would be used to solve
all of the standard problems of spherical astronomy for all the
latitudes, contain over 13,000 entries. However, his most
outstanding achievement in computational mathematics was his
qibla table.
77
The 2880 entries in the qibla table were expressed by al-Khalili
in standard Arabic sexagesimal (abjad) notation.
78
For each degree
of latitude there is a sub-table in which longitude difference was
entered vertically. The entries were arranged in two columns of
thirty entries for a degree of latitude, and two facing pages of both
manuscripts serve six consecutive latitudes (Fig. 7). Al-Khalili
mentioned that the qibla is southerly if the entry in the table is
in red ink and northerly if it is in black ink.
79
The latitude and
longitude of Mecca were used as 21
30
0
N and 67
in al-Khalili’s
qibla table. This longitude was obtained when considering the
prime meridian of the Canaries which was the reference adopted
by Ptolemy (ca. 90e168) in his Geography.
80
Al-Khalili’sqibla table
was computed only for the latitudes between 10
and 56
N,
because this was the range of terrestrial latitudes containing his
intended audience.
81
Al-Khalili’sqibla table is the only one
referred to by the later Syrian, Egyptian, and Turkish
astronomers.
82,83
Mosque orientation data
In this study, five cities and their districts, located in the
Central-Western Anatolia Region, were surveyed: Afyonkar-
ahisar, Ankara, Bilecik, Eskisehir and Konya. 82 mosques and
chapels,
84
dating from 1150 to 1590, were selected within the
following geographical limits: 37.4
N<4<40.2
N; 29.9
E<
l
<34.1
E. Fig. 8 shows the spatial distribution of the study
mosques.
Construction dates of the mosques
The date on which mosque construction started, obtained from the
mosque’s inscription, was used as a base reference. When the exact
date was not known and if a time interval was given, the mean time
(the difference between mean and interval borders) was assigned.
For example, a dating of, the second half of the fourteenth century
became 137525, the half of thirteenth century became 1250 10,
the first quarter of fifteenth century became 1410 10 and dating in
the form ‘about 1440’became 1440 5.These are not statistical error
estimates. In general, however, they should be close to the maximum
dating uncertainties.
85
Measurement of mosque orientation
The mosque orientations were measured from the large scale
(1:1000) topographic base maps produced by associated Munici-
palities. The azimuth (horizontal angle measured clockwise from
the grid north) of the main axis of the qibla-wall (including niche)
was taken from the maps. After a magnetic declination correc-
tion,
86
the perpendicular direction to the azimuth of the qibla-wall
was used as the mosque orientation (Fig. 9). The azimuth values,
directions and all angular quantities were given in degrees with the
positive sense as East of North. The azimuth of the qibla-wall was
examined merely to avoid the errors at construction stages. The
orientations of the other walls of the mosque were measured only
as a check against map reading errors.
Four mosque orientations obtained from the maps were
compared to the mosque orientations determined by ‘Real Time
Kinematic’technique based on the use of GPS measurements in
a previous study.
87
The comparison showed that the mosque
orientations were measured from the maps with accuracy on the
order of 0.73
.
Accepted qibla directions
There were several accepted qibla directions which could have been
used to orientate the mosques. In this study the evaluation proce-
dures were based on the following qibla directions:
1) First visible point of the star
a
Sco (Antares): The major axis of
Kaaba was aligned towards the rising point of the star Canopus
based on the folk astronomy as mentioned above. Nevertheless,
76
King, World-maps for Finding the Direction and Distance to Mecca (note 19); King, Astronomy in the Service of Islam (note 34). The muwaqqit was a professional astronomer
associated with a major mosque whose particular concern was timekeeping, the regulation of the five daily prayer times, the determination of the visibility of the lunar
crescent for regulating the lunar Islamic calendar with its holy months of fasting and religious festivals and determining the qibla direction. See D.A. King, In Synchrony with
the Heavens eStudies in Astronomical Timekeeping and Instrumentation in Medieval Islamic Civilization, Vol. 1: The Call of the Muezzin, Leiden, 2004.
77
King, Al-Khalili’s qibla table (note 75).
78
King, Al-Khalili’s qibla table (note 75). See detailed information about abjad notation in R.A.K. Irani, Arabic numeral forms, Centaurus 4 (1955) 1e12.
79
King, Al-Khalili’s qibla table (note 75).
80
D. Roegel, An Extension of Al-Khalili’s Qibla Table to the Entire World, 2008, at http://locomat.loria.fr/khalili/khalili-ext.pdf (accessed 18 January 2012). In Ptolemy’s
Geography, latitude was measured from the equator as it is today but he put the prime meridian at the most western land he knew, the Canaries.
81
G. van Brummelen and K. Butler, Determining the interdependence of historical astronomical tables, Journal of the American Statistical Association 92, 437 (1997) 41e48.
82
D.A. King, The astronomy of the Mamluks, Isis 74, 4 (1983) 531e555.
83
King, Two Iranian world maps for finding the direction and distance to Mecca (note 29).
84
The mosques and chapels are referred to as mosques in this study.
85
M. Korte, M. Mandea and J. Matzka, A historical declination curve for Munich from different data sources, Physics of the Earth and Planetary Interiors 177 (2009) 161e172 .
86
There are three norths commonly in use. Grid north is the direction of a grid line which is parallel to the central meridian on the national grid. True north is the direction
of a meridian of longitude which converges on the North Pole. Magnetic north is the direction indicated by a magnetic compass. The horizontal angular difference between
true north and magnetic north is called magnetic variation or declination. The horizontal angular difference between grid north and magnetic north is called grid magnetic
angle which needs to be applied when converting between magnetic and grid bearings. Differences between the three norths for the centre of the map and mean annual
changes are given on the lower-right corner of the 1:25,000 scale topographic maps produced by General Command of Mapping in Turkey. The magnetic declination
correction for the azimuth of the qibla-wall was determined from the associated 1:25,000 maps for the year 2012.
87
I. Yilmaz et al., The study of qibla precision with RTK in Afyonkarahisar’s mosques which belong four different ages, Electronic Journal of Constructional Technology 3 (2007)
45e49 [in Turkish].
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371366
Canopus was beneath the current horizon for the study mosque
locations as were most parts of Anatolia. An observable star was
needed for a stellar orientation of the mosques. The ninth-
century legal scholar of Cordoba, Ibn Habib (ca. 790e853),
stated that the qibla at Cordova was towards the rising point of
a
Sco (Antares) because that star rises at the eastern corner of
Kaaba.
88
The first visible point of Antares at summer solstice was
used for mosque orientation.
89
2) Winter sunrise and sunset: The azimuth of the sunrise and
the perpendicular direction to the azimuth of the sunset at
winter solstice were used for mosque orientation (see Figs. 1
and 2).
90
3) The wind ‘Janub’direction: The Arab theory of winds centred on
four cardinal winds (Qabul, Janub, Dabur and Shamal) and the
Fig. 8. The geographical distribution of mosques in the study.
88
Hawkins and King, On the orientation of Ka’ba (note 31). Antares (
a
Sco) is the sixteenth brightest star in the night-time sky.
89
The azimuth of Antares was measured using Starry Night Backyard 3.1 software.
90
King, World-maps for Finding the Direction and Distance to Mecca (note 19). The azimuths of the sunrise and sunset were calculated by an Excel workbook with VBA
functions for sunrise, sunset, solar noon, twilight (dawn and dusk) and solar position (azimuth and elevation) based on the calculation procedure by NOAA, at http://www.
ecy.wa.gov/programs/eap/models/twilight.zip (accessed 18 January 2012).
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371 367
cardinal wind directions were astronomically defined. The
qabûl was defined as being from the direction of the summer
sunrise, and the janub as being in the direction of the rising of
the star Canopus. These four cardinal winds are roughly
perpendicular but are not equivalent to north, south, east and
west: They are offset by about 25
.
91
The direction of the
cardinal wind ‘Janub’was used in mosque orientation as a qibla
of 155
(see Fig. 3).
4) South-east direction: A common qibla direction towards exactly
south-east for throughout the country was used as a qibla of 135
for mosque orientation.
92
5) South direction: The wind blowing from south is named as qibla in
Turkeyand in the Turkish language qiblameans south.
93
Therefore
the geographic southern direction was used as a qibla of 180
in
mosque orientation.
6) Qibla table values: The qibla angles based on geographical coor-
dinates interpolated from the extended qibla table of al-Khalili
were used for mosque orientation.
94
The stellar and solar azimuths and qibla table values were
calculated according to the mean of geographical coordinates of the
mosques in each city for this study.
Magnetic declination
Schmidl presented two Arabic treatises from the medieval Islamic
world that constituted the earliest known evidence attesting the
use of the magnetic compass for the qibla determination: the first
was by the Yemeni Sultan al-Ashraf (ca.1290) and the second by the
Cairene astronomer Ibn Simun (ca. 1300).
95
The use of the compass
was already widespread on the sea route between Syria and Egypt
in the thirteenth century.
96
If the compass was in widespread use in
the thirteenth century it can be considered that it was available
even earlier for orientation purposes.
97
If architects or artisans who had no knowledge about magnetic
declination used a magnetic compass to orientate the mosque
towards the qibla direction determined by the methods mentioned
above, the mosques would be oriented towards where they thought
the qibla was, though in actuality it would be a few degrees out. The
deviation between the actual mosque orientation and the desired
Fig. 9. The measurement of mosque orientation.
91
K. Eagleton, Islamic Astrolabes and the Weather, 1999, at http://www.hps.cam.ac.uk/starry/isaslabeweather.html (accessed 18 January 2012).
92
Barmore, Turkish mosque orientation and the secular variation of the magnetic declination (note 24).
93
Sarton, Fourth answer to query no. 25 (note 40).
94
Roegel, An Extension of Al-Khalili’s Qibla Table to the Entire World (note 80).
95
Schmidl, Two early Arabic sources on the magnetic compass (note 72).
96
Janin and King, The Sanduq al-yawaqit of Ibn al-Shatir (note 71).
97
Barmore, Turkish mosque orientation and the secular variation of the magnetic declination (note 24). The earliest reference to the knowledge of a magnetic compass is
found in a Persian anthology from 1232 where a sea captain who re-establishes his course by means of a mysterious iron fish which, when rotated in a bowl of water, settles
to point south, during a voyage in the Red Sea. See B.M. Kreutz, Mediterranean contributions to the medieval mariner’s compass, Technology and Culture 14 (1973) 367e383;
Schmidl, Two early Arabic sources on the magnetic compass (note 72).
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371368
mosque orientation by magnetic compass can potentially thus be
considered for the purposes of geomagnetic research as magnetic
declination data for the mosque’s location at the time of
construction.
98
In this study, the mosques dated before 1600 were
selected, taking into account the sparseness of historical magnetic
declination measurements before the seventeenth century and the
earliest individual declination measurement that was made at
1600, in Istanbul.
99
For the evaluation of the angular differences between the actual
mosque orientations and the accepted qibla directions, the historical
magnetic declination values for the mosques’location were ob-
tained at the construction dates from a global geomagnetic field
model, based on only archeomagnetic data, for 0e3ka
(ARCH3K.1).
100
The magnetic declination values based on ARCH3K.1
were calculated at the construction dates for the mean geographical
location of the mosques for each city. When the exact construction
date was not known, ARCH3K.1 magnetic declination values were
calculated for the mean time and the interval borders to account the
uncertainty in the construction date and the mean value of these
declinations was used for the study. A minimum error of 4.1
was
assigned to magnetic declination data for
a
95
confidence limit.
101
Results and conclusions
Fig. 10 shows the range of mosque orientations determined in this
study, displayed on a circular horizon, alongside the accepted qibla
directions for each of the cities. The mosque orientations
measured from the maps for this study ranged from 120.03
(in
Konya) to 210.57
(also in Konya) and showed a large amount of
scatter.
102
The differences between the actual orientation of the mosques
and the accepted qibla directions and also the magnetic declination
values are illustrated in Fig. 11. The statistical analysis of the actual
mosque orientations, the mean of the accepted qibla directions and
declination values are summarised in Table 1 for each city.
The wide range of mosque orientations illustrated in Fig. 10
may have reflected the different standards for determining the
qibla directions over time within each city. Approximately, 63% (52
of 82) of the differences in mosque orientation lay inside the 4.1
declination bounds of 95% confidence (Fig. 12). It could be
concluded that while the compass may have been generally used
in mosque orientation, it was not used for all mosques. The
distribution of orientation differences between 2
and 2
(20 of
82) provides valuable information about the knowledge and
competence to orient the mosques correctly in most periods in
Turkey (Fig. 13). Furthermore, it could be concluded that the
accepted qibla directions selected for this study were appropriate
for Turkey.
When the results presented in Ta ble 1 are evaluated, it can be seen
from Figs. 10e13 that the mosque orientations towards the star
Antares or southern direction (180
) were in widespread use in the
study cities in Turkey. 20 mosque orientations based on Antares and 11
orientations based on southern direction lied inside the 4.1
declina-
tion bounds 95 % confidence (Fig. 12). There were 7 orientations
associated with Antares and 5 orientations associated with a qibla of
180
with orientationdifferences between2
and 2
(Fig.13). It could
be concluded that the main mosque orientation method was towards
Antares and the following was based onsouthern direction, especially
Fig. 10. The range of the actual mosque orientations (coloured section) and accepted
qibla directions.
98
The angular difference between the geographic and the magnetic north is called magnetic declination. The magnetic declination changes with regard to location and
time. See I. Yilmaz, M. Gullu, M. Yilmaz and M.A Dereli, Compass roses on the Book of Navigation (Kitab-ıBahriye): declination data source for geomagnetic field models,
Physics of the Earth and Planetary Interiors 182 (2010) 170e174.
99
F. Ozcep and N. Orbay, Historical and modern geomagnetic data sources in Turkey, in: Proceedings of AGU 2000 Spring Meeting, Washington DC, USA, 2000.
100
M. Korte, F. Donadini and C.G. Constable, Geomagnetic field for 0e3 ka: 2. A new series of time-varying global models, Geochemistry, Geophysics, Geosystems 10 (2009)
Q06008, doi:10.1029/2008GC002297; M. Korte and C. Constable, Improving geomagnetic field reconstructions for 0e3 ka, Physics of the Earth and Planetary Interiors 188
(2011) 247e259. ARCH3K.1 is recommended for earth surface geomagnetic field investigations in Europe and Asia. The smoothed magnetic declinations are cautiously
considered due to sharp variations in the non-dipole geomagnetic field evidenced at least during the last 1500 years. See Y. Liritzis and M. Kovacheva, Some evidence for
sharp changes in the archaeomagnetic intensity variation during the last 2000 years, Physics of the Earth and Planetary Interiors 70 (1992) 85e89.
101
F. Donadini, M. Korte and C.G. Constable, Geomagnetic field for 0e3 ka: 1. New data sets for globa l modeling, Geochemistry, Geophysics, Geosystems 10 (2009) Q06007,
doi:10.1029/2008GC002295.
102
A list of construction dates, actual orientations, accepted qibla directions and magnetic declination values for the study mosques is available from the author.
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371 369
for the cities, Afyonkarahisar, Ankara and Konya. The mosque orien-
tation towards the direction of the wind ‘janub’(155
) in Eskisehir
should be mentioned. In Bilecik, the mosque orientation approach
could not be evaluated because of the insufficient sample size.
Thus far, in evaluating these results, no knowledge about the
impact of magnetic declination on the use of a magnetic compass to
orientate the mosque towards the qibla direction has been used. To
allow for the possibility that Mosque architects or artisans were
aware of the magnetic declination but not its direction (or became
confused as to which direction is declined relative to the other),
a deviation of twice the declination (including both the positive and
negative senses) was introduced.
103
Approximately, 61% (23 of 38)
of the mosque orientations dated between 1280 and 1500 lay inside
the 4.1
declination bounds at 95 % confidence. The distribution of
these mosque orientations is given according to the accepted qibla
directions in Fig. 14. In this context, it can be concluded that most of
the mosques dated between 1280 and 1500 were oriented towards
the first visible point of the star Antares or the southern direction
with the use of a compass. The mosques from the date range
150 0 e1590 were oriented by solar observations. The mosques
dated before 1280 had no obvious orientation with regard to the
accepted qibla directions in this study. These mosques may have
been converted from the earlier religious buildings (churches)
Fig. 11. Mosque orientation differences from the accepted qibla directions (declinations were given with 4.1error bounds for
a
95
confidence).
Table 1
The statistical values of mosque orientations, qibla directions and declinations for each city
City Actual orientation Accepted qibla Declination
Min Max Mean Std dev Antares Sunrise Sunset Janub S-E South Table Min Max Mean
Afyonkarahisar 126.05 207.04 171.51 19.04 158.57 119.90 149.94 155 135 180 152.71 3.55 12.33 8.34
Ankara 168.14 185.19 177.18 6.63 160.00 120.51 149.37 155 135 180 160.21 2.15 12.43 8.35
Bilecik 157.20 166.97 162.08 4.88 159.65 120.54 149.31 155 135 180 152.94 4.98 9.46 7.22
Eskisehir 143.16 208.42 153.16 34.39 159.96 120.26 149.60 155 135 180 155.41 4.64 12.62 9.99
Konya 120.03 210.57 168.02 22.16 158.38 119.60 150.23 155 135 180 156.27 0.95 12.19 9.72
Fig. 12. The mosque orientation differences inside the declination bounds. Fig. 13. Mosque orientation differences between 2and 2.
103
Barmore, Turkish mosque orientation and the secular variation of the magnetic declination (note 24).
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371370
without modification
104
or the pre-existing city structure may have
affected the orientations of these mosques. Perhaps, they were
oriented towards a qibla direction based on folk astronomy that was
not examined in this study. Yet the mathematically calculated qibla
directions based on al-Khalili’sqibla table were not generally used
in mosque orientation before 1600.
In order to evaluate the pattern of historical mosque orientation
more fully, a wider geographical coverage is needed. It can be
concluded from this study of the Turkish data that the information
on changes in mosque orientation between the fourteenth and
sixteenth centuries can be used as a magnetic declination data
source for global or regional geomagnetic field models of the Earth.
To test this against other data sets would be an interesting objective
for future research.
Acknowledgements
I would like to acknowledge Monika Korte from Helmholtz-
Zentrum Potsdam, Deutsches GeoForschungsZentrum and Fabio
Donadini from the Institute of Geophysics and Planetary Physics,
Scripps Institution of Oceanography, University of California, for
their helpful advice on error estimates of the ARCH3K.1 model.
Fig. 14. Mosque orientation differences (dated between 1280 and 1500) inside the
declination bounds.
104
In formerly Christian regions conquered by Islam, various churches were converted into mosques for Muslim prayers. The oldest surviving monumental mosque, the
Great Mosque of Damascus (709e715) was built over the church of John the Baptist. The great church of Hagia Sophia in Istanbul was adapted to a mosque almost 1000 years
after it was built without violating its architectural qualities. See S. Le Roux, Church to mosque: a short account of the recycling of the Pretoria West Dutch Reformed Church,
South African Journal of Art History 22 (2007) 98e114 .
M. Yilmaz / Journal of Historical Geography 38 (2012) 359e371 371
