W
References
Wang Chong
FABRIZIO P REGADIO
The Chinese philosopher Wang Chong was born in
Shangyu (modern Zhejiang) in AD 27. According to his
biographies, he came from a poor family and devoted
most of his life to teaching, but also held minor posts in
the state administration. Later he retired to compose the
work by which we know him today, the Lunheng
(Balanced Discussions). This extensive treatise in 85
chapters (one of which is lost) was completed in AD 82
or 83, about 15 years before the author’s death.
Wang Chong lived about one century after Confu-
cianism had emerged as imperial ideology. In this
process, the “rationalist” and socially minded philoso-
phy taught by Confucius had been integrated with
cosmological doctrines extraneous to the letter of his
teaching. The school of though: to which Wang Chong
belonged propounded, on the contrary, a reading of the
classic texts devoid of esoteric interpretations.
In his work, Wang Chong analyzes, with a strongly
skeptical and even iconoclastic spirit, ideas expounded
by earlier thinkers and beliefs shared by the people of
his time (e.g., the recourse to divination, the belief in
ghosts, the search for physical immortality, and the idea
of an individual spirit that persists after death). Wang’s
typical procedure is to bring out contradictions in the
anecdotes and accounts that he first quotes in full.
Through an exemplary logical method, he often does
not hesitate to take as true one detail that he has
previously refuted, if this may serve to invalidate a
different detail.
Philosophically, Wang Chong maintained that all
phenomena arise spontaneously, and are not expressions
of Heaven’s will. Related to this was his opposition to
the belief in prophecies and portents, through which
Heaven was deemed to legitimate or censure rulers, and
assent to or dissent from their policies. While Wang
Chong rejected the blend of these forms of esoterism with
Confucianism, he fully accepted the metaphysical and
cosmogonical doctrines traditionally placed under the
egis of Daoism. In this way, he anticipated some of the
new developments in post-Han Confucianism.
Chan Wing-tsit. A Source Book in Chinese Philosophy.
Princeton: Princeton University Press, 1963.
Forke, Alfred. Lun-heng. Philosophical Essays of Wang
Ch’ung. Vol. 1. London: Luzac and Co., 1907; Vol. 2.
Berlin: Georg Reimer, 1911.
Fung Yu-lan. A History of Chinese Philosophy, Vol. 2: The
Period of Classical Learning. Princeton: Princeton Uni-
versity Press, 1953.
Hui, Huang, ed. Lunhengjiaoshi (A Critical and Annotated
Edition of the Lunheng). Changsha: Shangwu Yinshuguan,
1938.
Needham, Joseph. Science and Civilisation in China, Vol. 2:
History of Scientific Thought. Cambridge: Cambridge
University Press, 1956.
Wang Xiaotong
A NG T IAN S E
There is no record of Wang Xiaotong’s early life nor his
year of death. We estimate that he flourished from the
second half of the sixth century to the first half of the
seventh century. The little we know of him is from a
memorial he presented to Emperor Gaozu of the early
Tang dynasty (AD 618–906) on the occasion of the
submission of his mathematical text to the throne. The
mathematical text he submitted was known as Jigu
Suanjing (Continuation of Ancient Mathematics) and
was subsequently selected as a prescribed text for
imperial examinations in AD 656. In his memorial,
which is now attached to his mathematical text, he
mentioned that he had studied mathematics from a very
young age. He studied the Jiuzhang Suanshu (Nine
Chapters in Mathematical Art) thoroughly and had
great admiration for Liu Hui’s in-depth commentary on
the text. On account of his mathematical acumen, Wang
Xiaotong was appointed as an instructor in the
Department of Mathematics, and later as a deputy
director in the Astronomical Bureau. In AD 623,
together with Zu Xiaosun, an official of the board of
Civil Office, he was appointed to re-examine the
adequacy of the current calendar.
Composed by Fu Renjun and promulgated for use
since AD 619, the calendar had on several occasions
2200 Wang Xichan
been found to be losing accuracy in the predictions of
solar and lunar eclipses. Based on the structure of the
Kaihuang calendar composed by Zhang Bin of the
previous Sui dynasty (AD 581–618), Wang Xiaotong
criticized the adoption of the ding shuo method and the
precession of equinoxes in the current calendar. The
critique sparked a 3-year debate between Fu Renjun
and Wang Xiaotong, culminating in a submission of a
proposal for rectifications consisting of more than 30
errors to the Astronomer-Royal. This does not
necessarily reflect Wang Xiaotong’s achievement in
calendrical science. On the contrary, his view of
adhering to the traditional model without taking into
consideration the uneven apparent motion of the sun
and the precession of equinoxes in calendrical calcula-
tions was a retrogressive one. His contribution lies in
mathematics.
From the late Han dynasty in the first century
onward, Chinese mathematicians were familiar with
quadratic equations and their solutions. But it was not
until the appearance of the Jigu Suanjing that equations
of the third degree were presented. They arose because
of the special needs of engineers, architects, and
surveyors of the Sui dynasty. There are 20 practical
problems in the Jigu Suanjing consisting of a problem
(No. 1) on calendrical calculation, six problems (Nos. 2–6
and 8) on engineering constructions, seven problems
(Nos. 7, 9–14) on the volume of granaries, and six
problems (Nos. 15–20) on right-angled triangles.
The solutions of most of the problems involved equa-
tions of the third degree. For example, problem No. 15
says: “There is a right-angled triangle, the product of
two sides of which is 706 50 1
and whose hypotenuse
is greater than the first side by 36 109 . Find the lengths
of the three sides.” Wang’s solution amounts to the
formation of the cubic equation as follows:
S p2
x 3 þ x2 ¼ 0
2 2S
where P is the product and S the surplus. As a matter of
fact, most of the problems in the last three categories
involved the use of the equation
x3 þAx2 þ Bx ¼ C;
where A, B, and C are positive numbers. Wang
Xiaotong provided the rules for the arrangement of
the equations in all these problems but did not explain
the procedure for arriving at such equations. He also
did not discuss the equations of higher degrees.
Numerical equations of degrees higher than the third
degree occurred first in the work of Qin Jiushao around
AD 1245.
See also: ▶Liu Hui and the Jiuzhang Suanshu,
▶Calendars, ▶Qin Jiushao
References
Guo, Shuchun. Wang Xiaotong. Zhongguo Gudai Kexuejia
Zhuanji (Biographies of Ancient Chinese Scientists).
Ed. Du Shiran. Beijing: Kexue Chubanshe, 1992. 317–9.
Mikami, Yoshio. The Development of Mathematics in China
and Japan. 2nd ed. New York: Chelsea Publishing
Company, 1974.
Qian, Baozong ed. Suanjing Shi Shu (Ten Mathematical
Classics). Beijing: Zhonghua Shuju, 1963.
Ruan, Yuan (AD 1799). Chouren Zhuan (Biographies of
Mathematicians and Astronomers). Vol. 1. Shanghai:
Shangwu Yinshuguan, 1955.
Wang Xichan
L IU D UN
Wang Xichan (July 23, 1628–October 18, 1682),
sometimes known by his literary name, Xiao’ an, was
from Wujiang, Jiangsu province, China. When he was
sixteen, the Ming Dynasty (1368–1644) collapsed and
the door to social advancement through the imperial
examination system was suddenly closed. After several
unsuccessful suicide attempts, Wang abandoned his
hopes for an official career and became one of the most
distinguished Ming loyalists in his area. He seems to
have survived by teaching literature, although his main
interest was in astronomy, which he had studied on
his own since his youth. Despite both poverty and
illness, Wang made unremitting efforts to observe the
heavens, calculate planetary positions, and write about
astronomy.
Beginning in the late Ming Dynasty, Western astro-
nomy had been disseminated intermittently into China,
and subsequently it was adopted by the Qing (1644–
1911) rulers. In 1646, a number of astronomical treatises,
earlier written or translated by Western missionaries
serving the Ming court, were published together by
order of the Qing emperor under the general title Xi
Yang Xin Fa Lin Shu (Astronomical Treatises of the
New Methods of the West, 1646). Unfortunately, the
“New Methods” introduced by the missionaries did not
reflect the advanced achievements of modern astronomy
in seventeenth-century Europe. Even worse, there were
some defects and internal contradictions, especially in
the parts dealing with the cosmological theory of
planetary motions.
In his Xiao An Xin Fa (New Method of Xiao’an,
completed in 1633), Wang argued that all of the
Western techniques could be reconciled with classical
Chinese schemes and therefore could be used to revive
the lost traditional astronomy of ancient China. By means
of trigonometry, which was not used in traditional
Chinese astronomy and mathematics, Wang created a

series of methods to calculate ecliptic positions and
predict planetary occultations. Forty years later, in his
Wu Xing Xing Du Jie (On the Angular Motion of the
Five Planets, completed in 1673), Wang criticized the
contradictions in the Xi Yang Xin Fa Li Shu and
proposed instead his own model of the planetary
motions which differed from both the Aristotelian–
Ptolemaic and the Tychonic geostatic models. In addi-
tion, he attributed the planetary motions to an attractive
force radiating from the outermost moving sphere (i.e.,
the primum mobile). Wang also suggested that there
were planets inside the orbit of Mercury which might
account for the appearance of sunspots. All of these
thoughts, along with some of his methods, were heu-
ristic and had considerable influence on his successors.
In general, Wang was one of a few pioneers who
responded to Western science by conscientiously study-
ing it; he was open to critical acceptance of its major
ideas when suitably reinterpreted for use in seventeenth-
century China. Nevertheless, some of Wang’s arguments
were clearly exaggerated. For instance, he assumed
that Western astronomy had in fact originated in ancient
China, but such claims were due to his radical natio-
nalism and traditional reluctance to recognize any
innovations from the Westf.
References
Chen Meidong et al. eds. Wang Xi Chan Yan Jiu Wen Ji
(Collected Papers on the Studies of Wan Xichan).
Shijiazhuang: Heibei Science and Technology Press, 2000.
Jiang, Xiaoyuan. Wang Xichan. Zhong Guo Gu Dai Ke
Xue Jia Zhuan Ji (Biography of Ancient Chinese
Scientists). Vol. 2. Ed. Shiran Du. Beijing: Science Press,
1993. 1005–15.
Sivin, Nathan. Wang Hsi-Shan (Wang Xichan). Dictionary of
Scientific Biography. Vol. 14. Ed. Charles C. Gillispie.
New York: Charles Scribner’s Sons, 1976. 159–68.
Wang, Xichan. Xiao An Xin Fa (New Methods of Xiao’an,
completed in 1663). Shanghai: Commercial Press, 1936.
Wang, Xichan. Wu Xing Xing Du Jie (On the Angular
Motions of the Five Planets, completed in 1673).
Shanghai: Commerical Press, 1939.
Xi, Zezong. Shi Lun Wang Xi Chan De Tian Wen Gong Zuo
(An Essay on the Astronomical Work of Wang Xichan).
Ke Xue Shi Ji Kan 6 (1963): 53–65.
Water in India: Spiritual and Technical
Aspects
K. N. S HARMA
The subject of water has been treated spiritually,
philosophically, cosmologically, medically, and poeti-
cally in the ancient Indian literature comprising the
Water in India: Spiritual and technical aspects 2201
Veda(s), Brāhman.a(s), Upanis.ad(s), Purān.a(s), and
Smr.ti(s). Water is regarded as the primordial substance
from which the universe came into being. Water
occupies the highest place amongst the five basic
elements of nature, called pañcamahābhūta. These are:
ākāśa (ether, substratum, space), vāyu (air), teja or agni
(radiation, energy, or fire), āpa (water), and pr.thivī
(Earth). These five bhūta constitute the physical
universe. Air is said to have been generated from
space, fire from space, water from fire, and earth from
water. Fire and water, which are said to pervade the
entire universe, have a close nexus and are believed
to possess procreative powers (Narayana, 1995). The
tripartite nature of agni has been connected with the
three forms of water – celestial, atmospheric, and
terrestrial, called by different synonyms in Sanskrit so
that they have characteristics and attributes responsible
for different cosmic, atmospheric, or terrestrial actions.
In the social and religious traditions and culture of
India since Vedic times, water has enjoyed a unique
status. Water is the single most important tool/mode for
performing daily religious rituals or social ceremonies
and a primary means for purification of body and soul
in Hindu culture.
From birth till death in a Hindu society, water
remains an essential ingredient in performing all rituals.
Vedas and Their Chronology
The content of the Vedas is astonishingly scientific
although much of it remains to be interpreted correctly.
Knowledge of the Vedas is synonymous with knowl-
edge of the science and metaphysics of creation.
The time periods of various Vedas are as follows:
R. gveda 6500–3100 BCE
Sāma Veda 3100–2500 BCE
Yajurveda 2500–2000 BCE
An organic chronological development in India from
6500 BCE has been suggested (Frawley 1994), taking
into account both Vedic literature and recent archaeo-
logical findings:
6500–3100 BCE Pre-Harappan (early Rigvedic period)
3100–1900 BCE Mature Harappan (period of four Vedas)
W
The Myth of Aryan Invasion
For the past hundred years or so, Mortimer Wheeler’s
hypothesis of “swashbuckling, horse-riding light-
skinned Sanskrit-speaking people, called Aryans,
coming to Northwest India from Central Asia or Central
Europe in about 1500 BCE and razing to the ground the
2202 Water in India: Spiritual and technical aspects
highly urbanised Harappan civilization,” was accepted.
These so-called Aryans were supposed to have
composed the Vedas over the centuries, spread Sanskrit,
and built the Ganges civilization. According to this
concocted theory, these “invading” Aryans destroyed
the Harappan civilization which had been flourishing
there for over a millennium and whose habitants were
Dravidians.
References in the R. gveda point to India’s being a
land of mixed races. The R. gveda also states, “We pray
to Indra to give glory by which the dasyu will become
āryan.” This statement confirms that to be an āryan
was not a matter of birth. While the word dasyu meant
uncultured and illiterate, āryan meant noble, well-
cultured people. It is also used in the context of
addressing a gentleman or lady (āryaputra, āryaka-
nya). Nowhere in the Vedic literature is this word used
to denote race or language. This was a notion of Max
Mueller who, in 1853, introduced the word ārya into
the English language to refer to a particular race and
language. When challenged, he refuted his own theory
later in 1888.
Jean-Francois Jarrige, Director of the National
Museum of Asiatic Arts of Paris, the excavator of the
famous proto-historic site of Mehrgarh (Pakistan) and
member of the French Academy, carried out extensive
work over the last 30 years in the Indian subcontinent
Water in India: Spiritual and Technical Aspects. Fig. 1 Course of erstwhile Sarasvatī River (Kalyanaraman, 1998).
and showed that the excavations revealed artifacts
from a much older civilization of the Sarasvatī era
(Pushkarna, 1998). The civilization included metropoli
like Mohenjodaro, Harappa, Ganveriwala (in Pakistan),
and Dholavira and Rakhigarhi; towns like Lothal,
Surkotda, Banawali, and Kalibangan, and villages like
Kunal, in India. The excavations exposed not just these
towns or cities, but also an earlier settlement beneath it,
and an even earlier one further down. Figure 1 is a
map indicating rivers in the time of Sarasvatī, and
Fig. 2 shows the spread of Harappan and Sarasvatī
civilization sites. According to archaeologist Bisht,
before the mature Harappan stage, many regional
cultures like Amri, Kot Dirji, Kalibangan, Dholavira,
and Lothal had coalesced into the cultural umbrella
of Harappa. They were bound by common economic
compulsions and cultural ethos. The site of Dholavira
is an excellent example of a Harappan city that tells
the history of Early, Transition, Mature, Late, and
Final Phases of the Indus-Sarasvatī (Harappan) civili-
zation in India (ca. 3500–1700 BCE). The site
spreads over an area of 100 ha. This compares well
with the size of Harappa, Rakhigarhi, or Ganveriwala.
All these evidences have proved the invasion theory
to be unfounded, showing that the Aryans were the
native Indians, having an unbroken chronology of
civilization.

Water in India: Spiritual and Technical Aspects. Fig. 2 Harappan sites in Northwest India (Kalyanaraman, 1998).
Place of Water in Ancient Indian Literature
Water has enjoyed a high status in the social and
religious context of ancient Indian culture. It was an
essential medium for performing daily religious rituals
and social ceremonies and a primary means for
purification of body and soul. Even after thousands of
years, the rivers in India, especially Gangā (the Ganges)
and Yamunā are considered divine and capable of
purifying a sinful body with a few drops of water. Agni
(fire) is said to be born from the waters. Both water and
fire are said to possess procreative powers. Water is
considered a mother while fire is seen as a prolific
generator (Narayana, 1995).
The place of water as a life-giving and life-sustaining
element was very high. It was considered a cleanser of
sins and regarded as a divine protector. It was addressed
by various names – nectar, honey, ambrosia – in
prayers. Stagnant water was considered unhygienic.
The cosmic energy is the generator of the
universe, the embryo of waters the leader of
Water in India: Spiritual and technical aspects 2203
humans, most virile defender of the human race, it
remains ever illuminated by its own radiance and
it provides sustenance for its beloved progeny
(RV 3.1.12).
Water verily is arka (essence). What was there as
froth of water hardened and it became earth (the
embryonic state of the Universe).
Earth was formerly water upon the ocean of space
(Atharvaveda AV 13.6).
Water or the water element had many names in the
Vedas and other Vedic literature. It has more than 100
synonyms. These are not synonyms in a strict sense,
since they have been used to indicate their different
W
forms/states and contexts. Some names for water are:
Ambu, toyam, vāri, jalam, āpah., bhes.ajam, udakam,
salila, madhu, ambha, ghr.tam, and ks.īram. For
example, salila is a technical word in the Vedas which
is different from āpah. as mentioned in a mantra from
śatapatha brāhman.a (11.1.6.1) which says that āpah.
2204 Water in India: Spiritual and technical aspects
were indeed salila earlier. Salila is the primordial state
of the universe, when there is nothing manifest.
Nature’s Forces as Deities
In Vedic cosmology, pr.thivī (earth) symbolizes the
material base and dyāvā (heaven) symbolizes an
unmanifested immortal source. Together and between
them, they form the paryāvaran.a (environment).
The seer praises Heaven and Earth (dyāvā pr.thivī) by
saying, “You are surrounded, Heaven and Earth, by
water; you are the asylum for water; imbued with
water; the augmenters of water; vast and manifold; you
are the first propitiated in the sacrifice; the pious
(people) pray to you for happiness, that the sacrifice
may be celebrated. May Heaven and Earth, the effusers
of water, the milkers of water, dischargers of the
functions of water, divinities, the promoters of
sacrifice, the bestowers of wealth, of renown, of food,
of male posterity, combine together.” RV 6.70.4–5.
Water stands for all the elements, because it is really
a combination of water, fire, and earth, according to the
tripartite creation of the gross elements. Water is all
pervading.
Waters Regarded as Having Medicinal Powers
The Vedas also mention the medicinal qualities of
water. R. gveda hails water as the reservoir of all curative
medicines and of nectar. It invokes water which the
cows drink and offers oblations to deities presiding
over the flowing waters:
O Water, which we have drunk, become refreshing
in our body. May you be pleasant to us by driving
away diseases and pains – O divine immortal
waters (RV 63).
The Atharvaveda describes various sources of waters
and describes them as dispellers of diseases and as
more healing than any other healer. The scriptures
believed that waters avert pain that they are restorative
and curing. Wherever waters fall on earth, plants grow.
The hymns in Atharvaveda (6.23, 24, 57) hail water as
possessing medicinal qualities. A hymn in Atharvaveda
prays for waters to cure “incurable” diseases.
The scriptures, Sam . hitā(s), also regard water as
capable of alleviating pain. “O water which we have
drunk, become refreshing in our body. Be pleasant to us
by driving away diseases and pains.”
The R. gveda (1.161.9) states, “There exists no better
element other than water which is more beneficent to
the living beings. Hence waters are supreme.” Varuna is
a cosmic ruler as well as the deity that dwells in waters,
presides over them and is prayed to for granting
strength and virility from waters.
Another hymn from the R. gveda says,
Ambrosia is in the waters, in the waters are
medicinal herbs; therefore, divine (priest), be
prompt in their praise. Soma has declared to me,
‘all medicaments, as well as Agni, the benefactor
of the universe, are in the waters’: the waters
contain all healing herbs. Waters, bring to
perfection all disease-dispelling medicaments for
(the good of) my body, so that I may long behold
the Sun. Waters, take away whatever sin has been
(found) in me, whether I have (knowingly) done
wrong or have pronounced imprecations (against
holy men) or (have spoken) untruth. I have this
day entered into the waters – we have mingled
with essence – Agni, abiding in the waters,
approach, and fill me, thus (bathed), with vigour
(RV 1.23. 19–23).
According to Taittareyī Āran.yaka (7.3.2), Agni is an
antecedent form and sun the later. Water is a compound
and lightning is the joining element. Maitreyī Sam . hitā
further divides water into three places – sky, earth, and
mid-region.
Vājasaneyī Sam . hitā describes the medicinal use of
waters. “O water, which we have drunk, become
refreshing in our belly. May you be pleasant to us by
driving away diseases and pains.” The verses are
recited while touching one’s navel after drinking liquid
in a sacrificial procedure. Like Agni and Vāyu, āpah.
(water) also serves as a purifying agent.
Rivers as Goddesses
The rivers and river waters have been treated with great
reverence since ancient times. Traditionally, rivers such
as Gangā, Yamunā, or Narmadā are worshipped as
goddesses. Every morning and evening, on the banks of
the Gangā at Haridwār, there is daily “Gangā worship”
with lighted lamps in the presence of thousands of
devotees, traditional holy music, and chants of mantras.
The Sarasvatī River was one of the largest rivers in
ancient India before 3000 BCE and drained the Sutlej
and the Yamunā. By the end of Harappan culture, the
Sarasvatī went dry, bringing about the end of Harappan
civilization around 1900 BCE. The Vedic texts are
replete with references to the river Sarasvatī and the
seas.
River Sarasvatı̄: Myth or Reality?
Whereas the famous River Gangā is mentioned only
once in the R. gveda, the River Sarasvatī is mentioned at
least sixty times. Sarasvatī is now a dry river, but
it once flowed all the way from the Himalayas to
the ocean across the desert of Rajasthan. Research by
Dr Wakankar has verified that the river Sarasvatī
changed course at least four times before going
completely dry around 1900 BCE. The latest satellite

Water in India: Spiritual and Technical Aspects.
Fig. 3 Archaeological sites along erstwhile Sarasvatī River.
data combined with field archaeological studies have
shown that the R.gvedic Sarasvatī had stopped being a
perennial river long before 3000 BCE (Giri).
Numerous archaeological sites have also been
located along the course of this river, thereby
confirming Vedic accounts. The Sarasvatī is now dated
long before 3000 BCE. This means that the R. gveda
described the geography of North India long before
3000 BCE. This shows that the R. gveda must have been
in existence no later than 3500 BCE.
Tritium (hydrogen isotope) analysis of deep water
samples taken by Bhabha Atomic Research Centre
(BARC) has provided a broad spectrum dating for the
waters of the Sarasvatī river now revealed as
groundwater sanctuaries and aquifers. The waters
range from 4,000 to 8,000 years BP (Kalyanaraman).
For over 2,000 years (6000–4000 BCE), the Sarasvatī
flowed from Bandarpunch massif (Sarasvatī-Rupin
glacier confluence at Naitwar in western Garhwal).
Figure 3 shows the location of archaeological sites
along Sarasvatī river. A remote sensing study of the
India desert reveals numerous signatures of palaeo-
channels. The LANDSAT imagery highlights the
course of the River Sarasvatī in Punjab, Haryana, and
Rajasthan. The digital enhancement studies of IRS-1C
data (1995), combined with radar imagery from
European remote sensing satellites ERS I/2, have
identified subsurface features and reorganized the
palaeochannels beneath the sands of the Thar desert
in northwestern India (Paik, 2000).
Ancient Weather Science
In the early R.gvedic period, the performance of Yajnas
(sacrifices) as described in Vedic and Brahmanic
Water in India: Spiritual and technical aspects 2205
literature has some scientific basis. These rituals were
performed to ensure timely and adequate rainfall for
abundant availability of food, thriving of animal and
plant life and for overall human prosperity (Sharma,
2002). The Yajnas are very closely related to the
evolution of the universe, the solar system, human
procreation, occurrence of the seasons, rainfall and life
on earth. Several places in the Vedic literature mention
that Yajnas produce rain which in turn produces food.
The subject of meteorology was dealt with by several
ancient sages, including Nārad, Kashyap, Garg, Parā-
shar, Vasisht.ha, Druhin, Brihaspati, Devala, Vajra,
Sahadev, and Rishiputra. Asht. ādhyāyi (500 BC) by Pān.
ini is perhaps one of the earliest post-Vedic works
giving information on rainfall measurement and
droughts. The Arthashāstra (fourth century BC) of
Kaut. ilya gives information on rainfall distribution in
the country as well as the methods of its measurement.
Several of the old treatises in the form of Purān.as, such
as Matsya Purān.a, Vāyu Purān.a, or Vishn.u Purān.a,
dealt with the role of the sun in rainfall. Water science,
which was initially based on observations, kept being
refined and in the post-Vedic period from the third
century BC to the sixth century AD, many of the earlier
hypotheses came to be perfected. Kaut. ilya even gave
an estimate of the average rainfall in various parts of
India.
In Varāhamihira’s time, the units of measurement
of rainfall were the pala, adhaka, and dron.a.
The raingauge was round with a diameter of one
hasta (18 in., 46 cm) and had marks of pala. When it
was full, it indicated one adhak of rainfall. One dron.a
means 2–1/2 in. (6.4 cm) of rain. Accordingly to
Parashar, the height and diameter of the raingauge
should be 8 angulas (6 in., 15 cm) and 20 angulas
(15 in., 38 cm), respectively, and when it is filled to the
brim it measures one adhaka. Parashar also gives a
method of measuring rainfall on the ground. If it
measures four cubits on the ground it amounts to one
dron.a.
Ramanathan (1993) described the significance of
rituals and yagnas (sacrifices) with their cosmic and
scientific interpretations in great details. Observations
and measurements of wind, clouds, lightning, thunder,
rain, solid precipitation, atmospheric optical phenome-
na, and agricultural meteorology were presented.
Just as the water vapours are carried higher in the
form of clouds and are condensed in the presence
of cold air existing in the sky, similarly one can W
reach the height of spiritual progress and can get
strengthened due to restraining the breath through
Yogic exercise (YV 23.26).
The uniform water passes upwards and down-
wards in the course of days, clouds give joy to the
earth; fires rejoice the heavens (RV 1.164.51).
2206 Water in India: Spiritual and technical aspects

Water in India: Spiritual and Technical Aspects. Fig. 4 Major river basins of India.

Water in India: Spiritual and Technical Aspects. Table 1 Basinwise surface water potential of India (cubic km or
BCM/year)

Sl. No. Name of the river basin Average annual potential in river

1. Indus (up to Border) 73.31

2. a) Ganga 525.02
b) Brahmaputra, Barak & Others 585.60
3. Godavari 110.54
4. Krishna 78.12
5. Cauvery 21.36
6. Pennar 6.32
7. East Flowing Rivers Between Mahanadi & Pennar 22.52
8. East Flowing Rivers Between Pennar and Kanyakumari 16.46
9. Mahanadi 66.88
10. Brahmani & Baitarni 28.48
11. Subernarekha 12.37
12. Sabarmati 3.81
13. Mahi 11.02
14. West Flowing Rivers of Kutch, Sabarmati including Luni 15.10
15. Narmada 45.64
16. Tapi 14.88
17. West Flowing Rivers from Tapi to Tadri 87.41
18. West Flowing Rivers from Tadri to Kanyakumari 113.53
19. Area of Inland drainage in Rajasthan desert NEG.
20. Minor River Basins Drainage into Bangladesh & Myanmar 31.00
Total 1869.35
 Water management and reservoirs in India and Sri Lanka
O all learned people, fully realise your conduct
towards different objects of the universe, know ye
the electricity that maintains all beautiful objects,
the sun, the invisible matter brought into creation,
the invigorating vital airs, thus ye become the
utilisers of all objects. Homage to him who knows
the science of clouds, and to him who knows the
science of electricity (Yajur Veda).
These hymns give insight into the cloud-seeding
phenomenon and occurrence of rains. Various activities
occurring in the firmament were well observed and
formulated in scientific terms.
Festivals Related to Water
The focal point of Hindu social, cultural, and religious
rituals has been water. Indian mythology attaches a lot
of importance to the bath (snāna) which is mandatory
for participation in any important religious occasion. A
dip in holy rivers is considered an essential part of
Hindu culture, especially on specific occasions such as
the solar and lunar eclipses or occasions specified on
the basis of specific planetary configurations, which are
considered to have a cosmobiological effect on the
human body. In a cycle of 12 years, a great Indian
festival takes place which is known as Mahā Kumbha
or Great Kumbha, where millions assemble and have a
dip in the waters of the sacred rivers. Every third year, a
smaller water festival called Ardha Kumbha (half
Kumbha) is also held, where people congregate at
specified places at the banks of the holy rivers (Sharma,
1998). In the mythological scriptures it says that the
elixir of life, ambrosia (amr.t), that emerged from the
churning of the ocean by the gods and demons,
splashed out of the pitcher and fell to earth at four
places: Haridwār, Prayāg, Ujjain, and Nāsik. These are
located on the banks of the River Gangā, at the
confluence of the rivers Gangā, Yamunā, and Sarasvatī,
on the banks of Shiprā, and on the banks of Godāvarī.
A tussle ensued among the gods and the demons for
12 days. During this period the moon did not let the
amr.t fall from the pitcher, the sun did not let the pitcher
crack, Jupiter protected it from demons, and Saturn
saved it from being whisked away. Thus, these four
planets, which were responsible for saving the pitcher
of ambrosia (amr.t kumbha), have become an integral
part of the Kumbha. Various astronomical conjugations
during Kumbha represent various stages of the solar
cycle which have a direct influence on human beings
and the biosphere. The holding of Kumbha at an
interval of 12 years is symbolic of the need for
purifying the body by sublimating the inherent vices of
the 12 sense organs – five organs of action, five organs
of perception, and the mind and the intellect – thereby
arousing the six psychic centres or chakra separated
2207
from each other at distance of 12 angula (finger widths)
for attaining the amr.t kumbha or pitcher of ambrosia
(Dixit, 2001).
Water festivals are also celebrated in several other
ways in North and South India. It is customary to take a
dip in the holy waters of rivers on various auspicious
occasions.
References
Br.hadaran.yaka Upanis.ad. Ed. Swami Śivananda. P.O.
Shivanandanagar, UP, India: The Divine Life Society,
1985.
Dixit, Pankaj. The Mahakumbh: Its Sacred Significance. The
Speaking Tree. Times of India. 2001.
Frawley, David. The Myth of Aryan Invasion of India,
Published by Voice of India. 2/18 Ansari Road, New Delhi,
India, 1994.
Giri, Svami B. V. Aryan Invasion. ▶http://www.gosai.com/
chaitanya/saranagati/html/vedic-upanisads/aryan-invasion.
html.
Kalyanaraman, S. R. gveda and Sarasvatī-Sindhu Civilization-
Dates of the Sarasvatī Sindhu Civilization (CA. 3100–1400
BC): ▶http:www.hindunet.org/hindu_history/sarasvatī/html/
rvssc.htm August 1998.
Narayanan, Sampat, ed. Vedic, Buddhist and Jain Traditions.
Vol. 2. New Delhi, India: IGNCA, 1995.
Paik, Saswati. Saraswatī – Where Lies the Mystery!. Noida,
Up, India: CSDMS, 2000.
Pushkarna, Vijaya. Looking Beyond Indus Valley. ▶http://
www.appiusforum.com/week.indus.html July 1998 .
Ramanathan, A. S. Weather Science in Ancient India. Jaipur,
India: Rajasthan Patrika Limited, 1993.
R. gveda Sam . hitā. Parts 1–4. Ed. Ravi Prakash Arya and K. L.
Joshi. Delhi, India: Parimal Prakashan, 1997.
Sama Veda Sam . hitā. 1st ed. Ed. Ravi Prakash Arya. Delhi,
India: Parimal Prakashan, 1996.
Sharma, K. N. Water – The Fulcrum of Ancient Indian Socio-
Religious Traditions. Proceedings of International Con-
ference on Water Resources at the Beginning of the 21st
Century, UNESCO, Paris, 3–6 June 1998.
---. Status of Water in Ancient Indian Literature and
Mythology. Second International Conference of IWHA,
Bergen, Norway, 2002.
Yajur Veda Sam . hitā. 2nd ed. Ed. Ravi Prakash Arya. Delhi,
India: Parimal Prakashan, 1999.
Water Management and Reservoirs
in India and Sri Lanka
A NDREW M. B AUER , K ATHLEEN D. M ORRISON
W
Water storage and distribution technologies have played
an important role in the histories of southern India and Sri
Lanka. Given the variability in rainfall and the relatively
dry conditions over much of the region, it would have
been difficult for southern Asian agriculture, diet, and
2208 Water management and reservoirs in India and Sri Lanka
cuisine – particularly the heavy emphasis placed on rice –
to have taken the forms they did without the historical
development of water management techniques. More-
over, debates about the efficiency, sustainability, and
equitability of modern “big dam” projects versus
“traditional” methods of providing much-needed do-
mestic and agricultural water continue to dominate
Indian politics today. Here, we review historical forms of
water management in South India and Sri Lanka, paying
particular attention to ancient reservoir systems.
South India and Sri Lanka are dominated by a
monsoonal climate, whereby the southwest (or advanc-
ing) monsoon generally brings rains between the
months of June and October, and the northeast
(retreating) monsoon in November and December. In
addition to monsoon strength, a variety of topographic
factors relate to the distribution and concentration of
rainfall across the region. Parts of the southwestern and
western coasts of India receive between 3,000 and
3,200 mm of rainfall a year due to the orographic
(related to, or caused by, physical geography, such as
mountains or sloping terrain) effects of the Western
Ghats – South India’s most pronounced north–south
ranging mountain chain. On the eastern side of the
Ghats, however, a rain shadow is created that markedly
reduces the amount of precipitation in the central areas
of the southern peninsula. For example, Bangalore – the
capital city of the modern state of Karnataka – receives
an average of around 850 mm of rainfall per year, while
Hyderabad – the capital of Andhra Pradesh – receives
an average of only 700 mm of rainfall. Indeed, much of
the area within the orographic rain shadow of the
Western Ghats can be considered a semiarid climate,
with average rainfall levels falling as low as 400 mm
(India Meteorological Department 1981). Sri Lanka’s
topography creates a similar pattern in which the
southwestern coastal areas receive the bulk of the ad-
vancing monsoon and the northern and eastern coasts –
largely in the rain shadow of Sri Lanka’s central
highlands – receive precipitation from the shorter,
retreating monsoon. However, it is important to note
that actual rainfall in all parts of the Indian peninsula
and Sri Lanka can show significant intra- and interan-
nual variability as a result of being dependent on the
relative strength of the monsoon.
South Indian and Sri Lankan reservoirs include a range
of facilities constructed for the purposes of collecting and
storing water, generally for agricultural production.
These consist of artificial embankments built across
paths of gravity water flow, whether intermittent streams,
rivers, or simply slopes that might carry runoff after a
monsoon rain. Reservoirs may or may not involve
excavation of a basin to contain this water, but they are all
storage or storage/distribution devices built on a
relatively large-scale and meant to contain water behind
an embankment or dam, rather than within its major
construction. In this sense, we make a distinction
between cisterns (which collect and store water within
a rock-cut or other constructed facility), wells (which tap
the water table), reservoirs, and tanks. The term “tank” is
widespread in the South Asian literature, indiscriminately
used to describe almost any water-holding feature,
although the term most frequently refers either to
reservoirs or to temple tanks – large masonry structures
that hold water for ritual ablutions and other functions
associated with temple worship. Temple tanks often
derive their water from the water table. As such, temple
tanks and reservoirs are wholly different in construction,
morphology, and operation, similar only in their
capacities as water-holding devices and in certain
parallels of meaning and symbolism (Morrison in press).
Early Forms of Water Management in South India
and Sri Lanka
The earliest culturally significant water catchment
features in South India were not constructed reservoirs
(in the terminology outlined above), but seasonal
pooling basins, or cisterns, which developed naturally
from the differential weathering of bedrock. In
geomorphological terms, these are known variably as
gnammas, rock pools, or weathering pits, and are
considered to be characteristic features of residual hills
and inselbergs – isolated hills composed of resistant
rocks (e.g., granite or gneiss) that express pronounced
topographic relief from a surrounding plain – through-
out the heavily weathered terrain of the tropics and
subtropics (Thomas 1994; see also Porembski and
Barthlott 2000). In South Asia, such basins occur on the
granitic gneiss hills that characterize much of the central
and southern portions of the Indian peninsula, as well as
parts of Sri Lanka (e.g., Fernando 1976). Indeed,
because granite and gneiss are particularly impermeable
rock types, the bare hills of South India generate large
volumes of runoff water during the heavy monsoon
months that collects in such depressions (Fig. 1).
Water retaining rock pools appear to have taken on
cultural significance as early as the Iron Age (1000–
500 BCE) in several regions of South India. During this
period, mortuary and ritual sites were often marked by
the construction of megalith monuments, and a clear
cultural association between such ritual constructions
and seasonal water basins can be established. Large
concentrations of elaborately constructed megaliths –
ranging from dolmen cists, stone circles, rock cairns,
platform enclosures, stone spirals, and more – appear
to have been deliberately placed adjacent to water
basins in hilltop locations. Perhaps the most striking
example of such an association occurs at the site of Hire
Benkal in northern Karnataka, where hundreds of
megaliths are found near a broad shallow water basin
that likely began as a rock pool and was subsequently
 Water management and reservoirs in India and Sri Lanka
Water Management and Reservoirs in India and
Sri Lanka. Fig. 1 A rock pool at VMS 579 – an Iron
Age occupational site in the Koppal district of northern
Karnataka. Notice the artifact debris in the foreground
(photo by Andrew M. Bauer).
Water Management and Reservoirs in India and
Sri Lanka. Fig. 2 The site of Hire Benkal, showing
dolmen cists on the quarried banks of a basin feature that
likely began as a natural rock pool (photo by Andrew
M. Bauer).
expanded by quarrying activities for the construction of
monuments (Fig. 2).
Additional sites also demonstrate associations be-
tween water and culturally significant ritual places. For
example, a brick platform structure enclosed by granite
boulders at Bandibassapa Camp, also a large megalithic
complex in northern Karnataka, appears to have been
situated adjacent to a rock pool that was later modified.
Moreover, Gordon and Allchin (1955) reported 80
megaliths at a site near Bilebhavi where they identified
two “tanks” (cisterns), one of which was “lined with
stone slabs.” They also recorded a similar construction
on a hilltop megalith site near Koppal (Gordon and
Allchin 1955: 99). The association between megaliths
2209
and water basins has also been noticed in Tamil Nadu
and Sri Lanka (cf. Seneviratne 1984; Myrdal-Runebjer
1996), suggesting a ritual importance to water, and
possibly a sacred dimension to early water management
throughout much of the region.
Although rock pools likely served as the earliest
water retention features in the region, it is clear that
during the Iron Age humans began deliberately
expanding pooling basins and creating them through
activities of quarrying, excavation, and the construction
of embankments, or bunds. This is evidenced not only
at megalithic ritual sites, such as Hire Benkal, but also
at settlement sites, where both cobble lined basins and
constructed bunds are present. At the Iron Age
habitation site of Kadebakele (northern Karnataka),
for example, inhabitants modified the drainage pattern
on top of a granitic hill to form a water catchment basin.
Excavations in this reservoir show that it collected and
held water for only part of the year, partially a
consequence of the reservoir’s relatively small catch-
ment area of .027 km2. The facility also experienced
major drying episodes as well as significant siltation.
Nevertheless, it certainly provided much-needed water
to local residents at certain times.
It is difficult to say to what extent these early water
retention features may have supported cultivation.
Most are quite small and lack provisions for water
distribution necessary for large-scale agriculture.
Moreover, they are often perched atop high hills with
little cultivable land. However, some early reservoirs
do occur within natural drainage ways, occasionally in
association with check dams and at the base of
topographic features where seasonal water could
potentially be pooled for crop production. In fact,
Devaraj et al. (1995: 66) report “interlaced, hydrauli-
cally laminated” deposits behind a “rammed” earthen
construction near the base of a granite outcrop at
Watgal – a Neolithic (3000–1000 BCE) to Iron Age
(1000–500 BCE) period settlement site in northern
Karnataka.1 It is difficult to characterize the entire
range of variability in form and function of early water
management constructions without more systematic
work. Indeed, few regions have been systematically
studied, and the patterns described above may not hold
true across the entire region. Nevertheless, it is clear
that water retention techniques began to be practiced in
a variety of settings during the Iron Age (see also
Allchin 1954). In addition, the development of water
management technology during this period generally W
coincided with the introduction of new cultigens –
including rice cultivation – suggesting that water
retaining features became increasingly important to
1
However, it is important to note that the authors do not
identify this feature as a reservoir, or attribute it with any
water retaining “function.”
2210 Water management and reservoirs in India and Sri Lanka
agricultural production by the end of the first millenni-
um BCE.
In the first millennium AD there is stronger evidence
for the construction of larger reservoirs that were used to
meet the hydrological requirements of cultivation. A
series of reservoir walls have been reported in the
environs surrounding Sanchi – a well-known Buddhist
monastic site in west-central Madhya Pradesh spanning
the third century BCE to the twelfth century AD. These
reservoir features are built of earthen embankments
reinforced by stone masonry, with dams reaching nearly
6 m in height and expanding across drainage valleys, in
some cases exceeding 1 km in length. Moreover,
catchment areas range from 0.74 to 17 km2 (Shaw and
Sutcliffe 2001), potentially damming considerably
more water than the Iron Age reservoirs discussed
above. Artifact associations, as well as the proximity of
the embankments to the site of Sanchi, have allowed
scholars to suggest that several of these features were
constructed as early as the second century BCE, while
others were built throughout the first millennium AD
(Marshall 1940: 13; Shaw and Sutcliffe 2001).
Although dating features such as reservoir walls
remain problematic without direct geochronological
assessment (e.g., radiocarbon, optically stimulated
luminescence, etc.) or inscriptional data, textual
references from the Early Historic Period suggest that
reservoir construction was prevalent in some parts of
India, and in South Asia more generally, during this
time. For example, Chakravarti’s (1998) analysis of the
Arthaśāstra – an economic and political treatise
composed sometime in the late centuries BCE or early
centuries AD – suggests that Mauryan rulers (ca. 324–
185 BCE) were concerned with establishing irrigation
works. Moreover, numerous inscriptions and textual
references indicate that the construction of reservoirs
and irrigation facilities underwent remarkable develop-
ment in Sri Lanka during the Early Historic Period,
particularly on the island’s drier north and east sides.
Northeast of Sigiriya, the Minneriya reservoir built
during the reign of Mahasena (AD 276–303) is
particularly noteworthy. This reservoir – fed by a
canal as well as surface runoff – consisted of an
embankment nearly 2 km in length and at places
exceeded over 13 m in height. Inscriptions dated to the
reign of Mahasena’s successor speak of “three harvests
of [rice] paddy per year,” suggesting a marked impact
on agricultural production (Gunawardana 1971: 8).
Middle Period Reservoirs: The “Traditional”
System
It should be clear from the above discussion that
reservoirs and water storage features have a long history
in South India, and South Asia more generally. How-
ever, the proliferation of large reservoir construction for
the purposes of agricultural intensification is most
salient in South India after the Early Historic Period, and
in Sri Lanka only slightly earlier.
Numerous textual sources clearly indicate that
reservoirs played an important role in the Early Middle
Period (AD 500–1300). Although small dam-and-basin
facilities for water impoundment continued to be built
and used, Middle Period reservoirs typically consist of
masonry-faced earthen dams thrown up across valleys,
at the base of hills, and in other locations where seasonal
runoff and small streams could be captured (Fig. 3)
(Morrison 1993, in press). Like the Minneriya reservoir
in Sri Lanka, some were supplied by canals, which took
off via diversion weirs or anicuts, from perennial rivers
or intermittent streams. Water was moved downstream
from reservoirs to agricultural fields through masonry-
lined tunnels under the embankments, which were
regulated by sluice gates (Figs. 4 and 5). Some water
was also released over specially constructed waste
weirs, facilities which range from boulder-filled cuts to
elaborately built spillways (Fig. 6). Although the focus
was clearly on the storage and downstream distribution
of water, reservoir beds were also sometimes used for
cultivation and reservoirs served as important sources of
fish, silt and clay, and water for livestock. Reservoirs
were also used to raise the water table around them, an
important function even when the bed failed to fill. In
fact, we have documented a consistent (but not
universal) pattern of wells down-slope from sluice
gates (Morrison in press).
Reservoirs were particularly important in the far
south, in present-day Tamil Nadu, where many of them
were supplied by river-fed canals (Ludden 1999).
Water Management and Reservoirs in India and
Sri Lanka. Fig. 3 A Middle Period reservoir embankment
and sluice gate used to dam a valley below the Iron Age site
of Hire Benkal (photo by Andrew M. Bauer).
 Water management and reservoirs in India and Sri Lanka
Water Management and Reservoirs in India and Sri Lanka. Fig. 4 Diagrammatic cross-section of a reservoir embankment
and sluice gate.
Water Management and Reservoirs in India and
Sri Lanka. Fig. 5 The northern sluice of the Daroji
reservoir, south of Vijayanagara. This “double sluice”
construction became prominent in the sixteenth century,
when the Daroji embankment was constructed. This sluice
and reservoir have been maintained and continue to operate
(photo by Kathleen D. Morrison).
There it is possible to see perhaps the greatest
elaboration of the so-called “system reservoirs” – long
chains of reservoir facilities that flow one into the other,
linking large areas into tightly knit watersheds (Sharma
and Sharma 1990; see also Mosse 2003). Unfortunately,
none of these systems has been specifically analyzed
on the ground to determine precise construction
sequences, so although we know of many specific
single-reservoir projects dating as early as the seventh
century AD, we cannot say exactly how the overall
system functioned at this time or even how much of the
landscape was under reservoir irrigation. It should be
noted, however, that Early Middle Period reservoirs,
2211
Water Management and Reservoirs in India and
Sri Lanka. Fig. 6 Masonry waste weir, or spillway, to
regulate high water levels at the edge of a Middle Period
reservoir embankment near the village of Kurugodu (photo
by Kathleen D. Morrison).
“traditional” by any reckoning, ranged widely in size
from very small ponds to vast “seas,” the latter falling
well within the contemporary definition of a “large
dam.”2
Similar to the “systems reservoirs” of Tamil Nadu,
large Sri Lankan Early Middle Period reservoirs were
linked through vast networks of canals. The Minneriya
reservoir, for example, was connected to other facilities
via the construction of a “great canal” during the reign
of Aggabodhi I (ca. AD 571–604), and composite
maps of the irrigation facilities around Sigiriya show
that it fed at least two large reservoirs – the Kaudulla
and Kantalai reservoirs. Moreover, detailed survey has
shown that smaller dam-and-basin features were also
probably constructed and maintained throughout the W
period (Myrdal-Runebjer 1996). Again, it is difficult to
know how the overall systems functioned simulta-
2
Crest length ≥ 500 m, maximum flood discharge ≥ 2,000
m3 s−1, “especially difficult foundation problems" or “unusual
design" (ICOLD World Register of Dams 1998).
2212 Water management and reservoirs in India and Sri Lanka
neously. However, it is clear that Middle Period rulers
and elite made demonstrative efforts to construct new
irrigation facilities and repair older ones. In fact, the
Minneriya inflow and outflow canals were repaired at
least several times in the early centuries of the second
millennium AD (Myrdal-Runebjer 1996; see also
Brohier 1934; Gunawardana 1971).
This pattern of extensive reservoir use in Sri Lanka
and the far south of the Indian peninsula contrasts with
that of drier regions in the northern interior of the
subcontinent (Karnataka and parts of Andhra Pradesh).
In these northern regions, reservoirs were (and are)
almost exclusively runoff-fed and, given lower rainfall,
they are generally not as closely spaced as those of the
southern Tamil country. Still, many regions saw the use
of both system and isolated reservoirs. In the area we
have studied in northern Karnataka, reservoirs seem to
have been only a minor component of Early Middle
Period agricultural strategies. However, by the Late
Middle Period (1300–1700 AD), and especially with
the expansion of the large but loosely knit empire of
Vijayanagara (ca. AD 1330–1600) across much of the
peninsula, reservoir irrigation expanded considerably,
especially in the drier zones where it had previously
been limited. In and around the eponymous capital city
of this empire, urban foundations in the early fourteenth
century and the subsequent expansion of settlement and
population explosion in the region propelled reservoirs
into increasingly important components of larger
agrarian and political strategies (Fig. 7). Important
from the start of the Vijayanagara period, reservoirs
also constituted a key form of agricultural intensifica-
tion in the sixteenth century, or Late Vijayanagara
period, especially in regions where canal irrigation was
not feasible (Morrison 1995, 2001).
Reservoirs played variable roles in the processes of
Vijayanagara agricultural intensification and change;
this variation was structured as much by political
Water Management and Reservoirs in India and
Sri Lanka. Fig. 7 Small Middle Period reservoir and
sluice gate in the Daroji Valley, south of Vijayanagara (photo
by Kathleen D. Morrison).
factors and settlement dynamics as by environmental
variables such as runoff and soil conditions. What is
common to most parts of the urban hinterland, however,
is the way in which the vast majority of reservoirs fell
out of use after (in some cases, during) the Vijayanagara
period. Very few of the reservoirs from the original
system still effectively function, though there are a
few notable “living” reservoirs with long histories of
maintenance and reconstruction (Morrison 1993, 1995).
For example, the Daroji reservoir, the terminus of one
of the largest systems in our study area, continues to
collect runoff from a catchment area of 955 km2
and provides water for downstream agricultural fields
(Fig. 8).
Research on Middle Period reservoirs includes
analyses of pollen and charcoal from reservoir sedi-
ments (which allow the reconstruction of fire and
vegetation histories), sedimentological studies of reser-
voir fill and changes to erosional regimes and local
hydrology, and stylistic analyses of sluice and em-
bankment construction (e.g., Morrison 1994, 1995;
Myrdal-Runebjer 1994, 1996). In addition, we have
also considered tens of thousands of textual and in-
scriptional records describing facility construction and
Water Management and Reservoirs in India and
Sri Lanka. Fig. 8 ASTER satellite image of the Daroji
reservoir first built in the sixteenth century, south
of Vijayanagara. Notice the strong growth of agricultural
crops indicated by the reflection of near-infrared
wavelengths (shown here in red) downstream from the
embankment. Vegetation growth is also occurring on
the edges of the reservoir as receding water exposes
moisture retaining sediments (data originally obtained
from NASA).
 Water management and reservoirs in India and Sri Lanka
maintenance as well as conflicts over water, land, labor,
and rule (Kotraiah 1995; Morrison 1995; Morrison and
Lycett 1994, 1997). All of these diverse lines of
evidence suggest that Middle Period reservoirs, like
their contemporary and colonial counterparts (Mosse
2003: 45–46), were highly unreliable sources of
irrigation. Runoff-fed reservoirs, in particular, may fail
to fill in dry years; in the drier districts, this meant not
only that reservoirs could usually not support wet crops
such as rice, but also that dry crops such as millets might
not be assisted by the facility. The situation was
somewhat better in areas of higher rainfall, but
everywhere in southern India reservoirs are marked by
high evaporation rates, high siltation rates, and ongoing
maintenance challenges (Fig. 9).
Reservoirs, Politics, and Contemporary
Development
Across much of South Asia, contemporary “big dam”
projects have been cast as the legacies of high modernist
social and environmental engineering (cf. Gadgil and
Guha 2000; Scott 1998). In India, well-organized and
highly visible social protests have been made against
specific projects such as the Sardar Sarovar project and
others along the Narmada River. Often, antidam groups
suggest that the answer to sustainable and equable
development lies in a return to a “traditional” system of
technology and management. It is not our purpose to
argue the case for “modern” or “traditional” forms of
water distribution. Rather, we wish to suggest that
arguments that hinge on the dichotomies of large/small,
equitable/inequitable, and political/apolitical that often
accompany distinctions between “modern” and “tradi-
tional” forms of water management are inappropriate.
The notion that large political dam projects are
entirely a product of modernity is decidedly incorrect.
Water Management and Reservoirs in India and
Sri Lanka. Fig. 9 Photo showing the accumulation of an
estimated 2–3 m of silt and clay, to near the top of a sluice gate.
This reservoir is adjacent to the village of Avinamodugu, south
of Vijayanagara (photo by Kathleen D. Morrison).
2213
This is no clearer than in South India, where some
Middle Period reservoirs were created with embank-
ments more than 3.5 km in length and over 15 m in
height. As already noted, the Daroji reservoir in northern
Karnataka pooled runoff over a total catchment area of
nearly 1,000 km2; however, size alone is not the issue.
Both small and large reservoirs were deeply political
projects, tied to networks of patronage and power.
Middle Period reservoir construction was sponsored by
a wide range of political leaders from kings (rarely) to
local chiefs (commonly) and was also connected with
Hindu temples in a number of ways (Morrison 1995;
Morrison and Lycett 1994, 1997). Moreover, even
during the Iron Age and Early Historic Periods, shifts
from a reliance on rain-fed agriculture to reservoir
irrigation would have produced a new material order on
the landscape that was necessarily related to sociopoliti-
cal fields. Indeed, the questions – How was labor
mobilized? Who had access to irrigation water? And
how was this decided? – are entirely appropriate in
examining the entire history of water management. The
answers to these questions undoubtedly had sociopoliti-
cal ramifications for ancient inhabitants. This may have
especially been the case when irrigation accompanied
the introduction of new cultigens and cultural values of
cuisine shifted. It is not difficult to imagine the profound
social consequences of being a dry-farming millet
producer/consumer versus a wet-farming rice produc-
er/consumer when rice was a high status cultigen and the
preferred donation to Hindu temple gods during the
Middle Period, for example.
The above considerations of the (un)reliability and
the political nature of historic reservoirs of South India
and Sri Lanka are not to suggest that these impressive
systems have no contemporary value with regard to (re)
developing water management strategies. Quite the
opposite is true: work on the 3,000-year history of
irrigation in southern India shows both success and
failure in equal measure, portents for a reasonably
hopeful future. Thus, although there is no simple
solution to the water problems of the dry tropics of
South Asia, surely an informed perspective on the
actual historical experiences of the region must provide
a more secure basis for future planning than either a
romantic and unrealistic view of “tradition” or a blind
faith in “modern” science and technology.
References
W
Allchin, F. R. The Development of Early Culture in the
Raichur District of Hyderabad. Ph.D. Dissertation,
University of London, 1954.
Brohier, R. L. Ancient Irrigation Works in Ceylon. Colombo:
Ceylon Government Press, 1934.
Chakravarti, R. The Creation and Expansion of Settlements
and Management of Hydraulic Resources in Ancient India.
Nature and the Orient: The Environmental History of
2214 Water management in Peru
South and Southeast Asia. Ed. R. H. Grove, V. Damodaran,
S. Sangwan. Delhi: Oxford University Press, 1998.
87–105.
Devaraj, D. V., J. G. Shaffer, C. S. Patil, and Balasubramanya.
The Watgal Excavations: An Interim Report. Man and
Environment 20 (1995): 57–74.
Fernando, A. D. N. Resource Maps of Sri Lanka. Colombo:
Integrated Development Association of Sri Lanka, 1976.
Gadgil, M. and R. Guha. The Use and Abuse of Nature. Delhi:
Oxford University Press, 2000.
Gordon, D. H. and F. R. Allchin. Rock Painting and
Engravings in Raichur, Hyderabad. Man 55 (1955): 97–9.
Gunawardana, R. A. L. H. Irrigation and Hydraulic Society in
Early Medieval Ceylon. Past and Present 53 (1971): 3–27.
India Meteorological Department. Climatological Atlas of
India: Part A (Rain Fall). New Delhi: Government of
India, 1981.
Kotraiah, C. T. M. Irrigation Systems Under Vijayanagara
Empire. Mysore: Directorate of Archaeology and Museums,
1995.
Ludden, D. An Agrarian History of South Asia. Cambridge:
Cambridge University Press, 1999.
Marshall, J. The Monuments of Sanchi. London: Probsthain,
1940.
Morrison, K. D. Supplying the City: The Role of Reservoirs
in an Indian Urban Landscape. Asian Perspectives 32
(1993): 133–51.
---. Monitoring Regional Fire History Through Size-Specific
Analysis of Microscopic Charcoal: The Last 600 Years in
South India. Journal of Archaeological Science 21 (1994):
675–85.
---. Fields of Victory: Vijayanagara and the Course of
Intensification. Berkeley: Contributions of the University
of California Archaeological Research Facility, No. 53,
1995 (Rpt. New Delhi: Munshiram Manoharlal, 2000).
---. Coercion, Resistance, and Hierarchy: Local Processes and
Imperial Strategies in the Vijayanagara Empire. Empires:
Perspectives from Archaeology and History. Ed. S. Alcock,
T. D’Altroy, K. Morrison, and C. Sinopoli. Cambridge:
Cambridge University Press, 2001. 253–78.
---. The Daroji Valley: Landscape History, Place, and the
Making of a Dryland Reservoir System. Vijayanagara
Research Project Monographs. Delhi: Manohar Press, in
press.
Morrison, K. D. and M. T. Lycett. Centralized Power,
Centralized Authority? Ideological Claims and Arch-
aeological Patterns. Asian Perspectives 33 (1994): 312–53.
---. Inscriptions as Artifacts: Precolonial South India and the
Analysis of Texts. Journal of Archaeological Method and
Theory 3 (1997): 215–37.
Mosse, D. The Rule of Water: Statecraft, Ecology, and
Collective Action in South India. Delhi: Oxford University
Press, 2003.
Myrdal-Runebjer, E. Vävala väva – Sigiri Mahaväva Irrigation
System: Preliminary Results from an Archaeological Case
Study. South Asian Archaeology, Series B. 271. Ed. Asko
Parpola, and Petteri Koskikallio. Helsinki: Annales Acade-
miae Scientiarum Fennicae, 1994. 551–62.
---. Rice and Millet: An Archaeological Case Study of a Sri
Lankan Transbasin Irrigation System. Goteborg: Goteborg
University, 1996.
Porembski, S. and W. Barthlott, ed. Inselbergs: Biotic
Diversity of Isolated Rock Outcrops in Tropical and
Temperate Regions. New York: Springer, 2000.
Scott, J. C. Seeing Like a State: How Certain Schemes to
Improve the Human Condition Have Failed. New Haven,
CT: Yale University Press, 1998.
Seneviratne, S. The Archaeology of the Megalithic – Black
and Red Ware Complex in Sri Lanka. Ancient Ceylon 5
(1984): 237–307.
Sharma, R. K. and T. K. Sharma. Textbook of Irrigation
Engineering. Vol. I. Irrigation and Drainage. New Delhi:
Oxford, 1990.
Shaw, J. and J. Sutcliffe. Ancient Irrigation Works in the
Sanchi Area: An Archaeological and Hydrological Inves-
tigation. South Asian Studies 17 (2001): 55–75.
Thomas, M. F. Geomorphology in the Tropics: A Study of
Weathering and Denudation in Low Latitudes. Chichester:
John Willey & Sons, 1994.
Water Management in Peru
C. R. O RTLOFF
The Chimu Empire of ancient South America in the
time period between 900–1480 CE dominated the
north Peruvian coast from the Santa to the Lambeyeque
Valleys west of the Andean Cordilera Negra mountain
range. This region, in terms of present-day geographi-
cal locations, extended just north of the Peruvian
capital city of Lima to the Ecuadorian border and
eastward from the Pacific Ocean coast to the eastern
slopes of the Andes. From the central administrative
center at Chan Chan in the Moche Valley, successive
generations of Chimu rulers exercised political and
economic control of adjacent valleys through adminis-
trative centers charged with overseeing and maximiz-
ing agricultural production and development. Within
the territorial domain of the empire, many of the
westward-running rivers leading runoff water from
highland Andean rainfall collection zones to fertile
coastal fluvial valleys were intercepted and redistrib-
uted through extensive canal distribution systems to
irrigate agricultural fields. Many valleys even today
still contain well preserved canal networks from this era
as land under cultivation by the Chimu far exceeded
present-day cultivation areas by perhaps as much as
50 per cent. While only a few of the ancient valley
irrigation systems have been extensively explored,
mapped and analyzed to any extent (Eling 1986; Kosok
1965; Ortloff et al. 1982, 1985; Ortloff 1988, 1993),
some novel canal hydraulic control features have been
recently discovered that warrant analysis as they
provide a window into the level of hydraulic science
existing within the Chimu empire of ancient South
America.
Since most major valleys under Chimu control
showed evidence of massive state-sponsored, hydraulic
canal infrastructures to support irrigation agriculture, it
 Water management in Peru 2215

follows that a hydraulic science was co-developed to quebrada-crossing aqueducts are shown in Figs. 1
provide tailored flow rate canals to supply field systems and 2; the Hoya Hondada aqueduct is the largest and
distant from river inlets. As considerations of soil and final aqueduct in the downstream sequence and is
crop types, crop water demands, alternate field system shown in Figs. 3–5.
watering strategies, valley topography and defense
from drought and large rainfall runoff events influence
canal design and placement, an accompanying hydrau-
lic science with flexibility to design and modify canals
according to these considerations is expected in the
archaeological record. Most probably, cumulative
observations of water flow phenomena over time
served to provide a database for empirical design
principles used for canal layout and design. Canal
volumetric flow rate relies on inlet geometry, bed slope,
canal cross-section, wall roughness and subcritical,
critical, and supercritical flow characterization. Allied
technical disciplines related to route layout, surveying,
water delivery sequencing and water routing through
multiple canal branches are additional key technologies
vital to understanding Chimu hydraulics practice. Water Management in Peru. Fig. 1 View of one of
While investigation of all aspects of applied Chimu the three destroyed aqueducts breaching a deep,
hydraulic science would give a complete picture of erosionally-incised quebrada upstream from the Hoya
their technology base, the present investigation is Hondada Aqueduct. Photo by C. R. Ortloff.
focused on but one facet of the Chimu engineering
repertoire: hydraulic control systems.
This entry details an investigation of a recently
discovered canal hydraulic feature of the Chimu
Talambo-Farfán Canal located in the Jequetepeque
Valley of northern Peru. While the remains of the canal
system are of late Chimu origin due to association with
the Chimu site of Farfán, some upstream versions of
canal segments may be associated with earlier valley
occupation by Gallinazo and late period Moche occu-
pants. For the present analysis, however, only the last
Chimu phase of canal construction is considered.

Water Management in Peru. Fig. 2 Another view of

Analysis Results a further destroyed aqueduct in the upstream sequence
The Talambo-Farfán Canal originates far upstream in from the Hoya Hondada Aqueduct. Photo by C. R. Ortloff.
the valley neck of the Jequetepeque River and passes
late hillside Chimu occupation zones in the Talambo
region through a series of aqueducts and deep canal
cuts through upvalley low hills before finally emerging
onto the agricultural plains east of Farfán. The canal
was further extended to provide water to the extensive
agricultural field zones directly south of the Chimu
mountain redoubt of Farfán Sur. In the canal extension
region, deep quebradas (canyons) were formed over
time from successive El Niño rainfall runoff events W
sculpting deep erosion channels into the soft soil
deposits that formed natural obstacles to canal
extension. In order to bridge the multiple quebradas,
a series of three large earth-fill aqueduct structures
and many small aqueducts were constructed to extend Water Management in Peru. Fig. 3 View of the low
the canal to the vast Pampa de Faclo bordering the Hoya Hondada aqueduct across the wide quebrada.
site of Pacatnamú. Two of these (largely destroyed) Photo by C. R. Ortloff.
2216 Water management in Peru
Water Management in Peru. Fig. 4 Alternate view of
the Hoya Hondada aqueduct. Photo by C. R. Ortloff.
Water Management in Peru. Fig. 5 View of the
cross-section of the Hoya Hondada aqueduct at the destroyed
south end; height of the channel approximately 5 m from
ground level. Photo by C. R. Ortloff.
Due to the much larger depth and width of the
Hoya Hondada quebrada as compared to upstream
quebradas, aqueduct design changes from upstream
high fill height designs breaching narrow and deep
gorges to maintain slope to a long, low height aqueduct
deep within the quebrada with a steep 40 degree
approach chute from the southernmost upstream bluff.
After the canal passes over this low aqueduct, it
continues to the downstream northern sidewall of the
Hoya Hondada quebrada to field systems located on the
southern boundary of Pampa de Faclo approximately
8 km south of the ancient Moche-Chimu religious
center at Pacatnamú. The intent of the canal extension
through difficult terrain north and west of Farfán Sur
was apparently to provide water to settlements and field
systems in the south part of the pampa east of
Pacatmanú. While further canal extension to the city
limits of Pacatnamú may have been the ultimate intent
of the canal builders, traces of a continuous connection
path to the many canal fragments on the Pampa de
Faclo are yet to be discovered.
Water Management in Peru. Fig. 6 Demonstration of a
hydraulic jump in a hydraulic test channel induced by a
supercritical flow interacting with a plate obstacle; a similar
hydraulic phenomena is induced at the channel intersection
point of the steep chute and low slope Hoya Hondada
aqueduct channel. Photo by C. R. Ortloff.
The canal and aqueduct design within the Hoya
Hondada quebrada area contains many novel hydraulic
features indicative of the state of Chimu hydraulic
knowledge. As opposed to using a massive earth-fill
aqueduct that would span the 150 m width and 20 m
height to connect opposite sides of the quebrada with
a constant slope canal, a 5 m high aqueduct is
constructed low in the quebrada with a 25 m long
steep chute joining the upstream high elevation part of
the canal to the downstream low aqueduct. This
alternate design eliminates the possibility of a massive
high dam structure trapping a large height of quebrada
rainfall runoff behind it that would constitute a dam
breakage hazard to downstream occupation areas. The
low dam design, should breakthrough occur, would be
easily repairable due to its low volume of fill material
and contain relatively less impounded water than the
alternate design. While some Chimu aqueducts span-
ning quebradas contained culverts or boulder bases to
let water pass underneath to alleviate water damming
during heavy rainfall events, use of a low aqueduct
height design reduced backwater hydrostatic pressure
effects that lowered destructive transverse forces that
could lead to breakthrough failure.
Continuing, with no upstream hydraulic controls, the
effect of a high velocity, high volumetric flow rate
down the steep 40-degree downward-sloped chute to
the low slope aqueduct would be to create a massive
hydraulic jump at the angle-change junction. Fig. 6
shows a typical hydraulic jump created by an obstacle
placed in a high velocity, supercritical flow – in this
case, the hydraulic jump is typical of that resulting from
the effect of the rapid flow down the steep chute – low
slope aqueduct junction. The severity and height of a
large hydraulic jump is sufficient to destroy the
aqueduct by turbulent erosion’s acting on the unlined

aqueduct structure unless a hydraulic control is in place
upstream of the aqueduct to dissipate stream energy,
lower stream velocity and/or remove excess water from
the canal by a side weir – which is another mode of
stream energy reduction. The height of the hydraulic
jump also would cause sidewall overflow from the
aqueduct leading to precious irrigation water loss for
downstream field systems. By lowering the channel
water velocity and/or volumetric flow rate, the height
of the downstream hydraulic jump can be reduced to a
level that does not imperil the integrity of the aqueduct
structure or cause sidewall overflow water loss
problems. This then constitutes the hydraulic design
problem faced by Chimu engineers.
The channel upstream of the Hoya Hondada
aqueduct shows that an energy dissipation hydraulic
structure had been installed to influence flow leading
on to the aqueduct. Figs. 7 and 8 show a ground view
of this hydraulic control structure in its unexcavated
state. The hydraulic structure consists of two pairs
of opposing boulders with a 70 cm separation dis-
tance between boulders and with a 13.2 m down-
stream separation distance between boulder pairs. The
Water Management in Peru. Fig. 7 Unexcavated view of
the hydraulic control structure upstream of the chute and the
Hoya Hondada aqueduct; structure consists of opposing
boulder pairs in the main channel separated by a 13.2 m
distance. Photo by C. R. Ortloff.
Water management in Peru 2217
channel containing this structure also had variations
in width and sidewall angle. A FLOW-3D computer
model (FLOW-3D 2004) incorporating field measured
channel geometry is shown in Fig. 9. In the distance
between the boulder pairs, a side overflow weir that
activates when water height exceeds a given height is in
place that led to a channel that guided overflow water
into the quebrada downstream of the Hoya Hondada
aqueduct. The hydraulic function of the control
structure is next described by use of numerical solution
of the governing equations of fluid flow; results of the
solutions indicate the free surface shape and internal
velocity within the channel and provide the basis to
understand the hydraulic functioning of the control
section.
The first upstream choke (defined as the open
separation zone between boulder pairs) apparently
forms a hydraulic jump at incoming stream velocities
past about 7 m/s while the downstream choke apparently
forms a hydraulic jump at inlet velocities about 3 to 4
m/s. The second downstream choke is therefore con-
trolling as it actives first at a lower limiting velocity.
For all cases, an increase in water velocity occurs in the
zone downstream of the first upstream choke due to
Water Management in Peru. Fig. 8 Ground view of the
hydraulic control structure upstream of the chute and
Hoya Hondada aqueduct. Photo by C. R. Ortloff.
2218 Water management in Peru
Water Management in Peru. Fig. 9 FLOW-3D computer
model of the excavated hydraulic control section upstream
of the chute and Hoya Hondada aqueduct from measured
field data. Note the presence of the side overflow weir.
Photo by C. R. Ortloff.
Water Management in Peru. Fig. 10 Solution results for
water free surface geometry for a channel flow rate exceeding
3 m3/sec; note the spillage over the side weir from the action
of the downstream choke limiting the transmitted flow rate on
to the Hoya Hondada Aqueduct. Photo by C. R. Ortloff.
water gravitational acceleration on a hydraulically steep
slope influenced by expansive channel cross-sectional
geometry changes. These velocity changes account for
the different hydraulic behavior of each choke. Since
flow into the second choke is supercritical over the range
of subcritical input flow rates, the second choke forms a
hydraulic jump and the flow rate is limited by critical
conditions at the throat. As the choked flow rate of the
downstream choke is less than that of the upstream
choke, the flow rate difference results in a water height
Water Management in Peru. Fig. 11 Solution results for a
channel flow rate less than 3 m3/sec; note that flow is fully
contained within the channel and the side overflow weir is not
activated. Photo by C. R. Ortloff.
Water Management in Peru. Fig. 12 Plot of flow rate
(m3/sec) into the control structure vs. flow rate exiting: the
difference is the flow exiting over the side weir. Note that
transmitted flow rate is limited past 3 m3/sec due to the action
of the choke system and overflow weir activation. Photo by
C. R. Ortloff.
change between chokes that activates water overflow
from the side weir. The overflow flow rate is equal to
the volumetric flow rate difference between up- and
downstream chokes. Fig. 10 shows side weir over-
spillage typical of inlet flows exceeding about 7 m3/s;
Fig. 11 shows conditions obtaining when the inlet
volumetric flow rate is less than 3 m3/s and the side weir
is inactive. Fig. 12 shows that the flow rate passed on to
the Hoya Hondada aqueduct is automatically altered
past 3 m3/sec inlet flow due to activation of the overflow
weir. The upper curve represents a linear condition
present when no control section is present and the input

flow rate is equal to the output flow rate in the channel.
The lower curve indicates that when the control section
is in place, the output flow exiting from the control
section is reduced by the side weir overflow to a limiting
value about equal to 3 m3/s. The limiting flow rate of
3 m3/s then ultimately limits the height of the
downstream hydraulic jump before the aqueduct to
acceptable design values to control turbulence genera-
tion and contain the flow in the aqueduct channel
without sidewall overflow.
The double choke system described is augmented
with yet a further downstream energy dissipation
system on the steep sloped chute. The Fig. 13 contour
map, which shows the channel depression where the
control section exists, contains a leadoff bifurcation
channel that subtracts a fraction of the water down-
stream of the hydraulic control structure and leads it
around a small hill into a low slope channel. This flow
rejoins the main channel flow at a downstream junction
point. The lower subcritical velocity stream from the
lower slope channel acting on the main supercritical
stream at the channel junction creates a further energy
dissipating oblique hydraulic jump in the channel. The
hydraulic jump converts velocity kinetic energy into
random turbulent and potential, height change energy.
This has the effect of subtracting further energy from
the stream approaching the chute junction zone. Note
that no energy gain exists as subtracted flow from
the main channel is added back into the downstream
flow to maintain the constant flow rate. The net effect
of the two energy reduction controls – the first from
subtracting energetic water by means of the overflow
Water Management in Peru. Fig. 13 View of a minor
hydraulic control system found extensively in the Pampa
Faclo field system. The presence of a dual, opposing stone
choke downstream of a steep chute creates a hydraulic jump
ahead of the choke. This reduces the velocity of water
entering a field system to promote gradual water absorption
into the soil over time. The inlet from the main canal is
tailored only to let a certain amount of water into the chute to
make the system work as planned, indicating careful design
of this irrigation system control. Photo by C. R. Ortloff.
Water management in Peru 2219
weir, the second from hydraulic jump energy dissipa-
tion – is to limit the channel flow rate and also reduce
entry velocity onto the Hoya Hondada aqueduct to
limit hydraulic jump erosive and overflow damage.
Although a large part of the steep slope channel is no
longer extant, there is indication that stones placed in
the chute bed provided yet a further energy dissipation
mechanism to reduce stream velocity. The totality of
these hydraulic controls then serve to limit erosion
damage to the Hoya Hondada aqueduct – particularly
under El Niño conditions. For example, as the coastal
area is frequently subject to massive El Niño rainfall
runoff from the nearby Cerro Faclo mountain range,
large canal flow rates arising from water washing into
the canal could easily produce excessive canal flow
conditions beyond the design flow rate. An excessive
flow rate could destroy the Hoya Hondada aqueduct
by creating a massive hydraulic jump at the channel-
aqueduct junction. By diverting water over the weir
into a channel leading to the downstream side of the
aqueduct together with use of additional energy dissipa-
tion controls, excess water and energy is removed from
the canal to provide a protective hydraulic feature for
the aqueduct.
As the canal can transport up to 3 m3/s before weir
overflow activation, this flow rate represents a
sustainable maximum flow value to a location 35 km
from the river inlet. This flow rate may be available
to the Pampa Faclo fields and settlements, despite
evaporation and seepage losses, by blocking all other
canal branches from the main canal to temporarily
direct all flow to the Pampa Faclo. As the needed water
volume delivered through each canal branch to a field
sector was a known quantity to sustain crops in that
sector according to their importance, yield, and water
requirements, the assignment of water volumes there-
fore was metered for maximum agricultural productiv-
ity effect. As settlements on the outlying southern
section of Pampa Faclo were somewhat remote from
the Farfán center and field systems discovered to date
rather limited in size, it may be surmised that water
resources reserved for this sector were proportional to
its potential expansion and importance. The fact that
water could be directed to this secondary location
perhaps indicates that further expansion of the area’s
agricultural potential was anticipated but still in the
construction phase. The major effort to construct
multiple elaborate aqueduct structures to bring water
to this zone through difficult topography perhaps W
indicated that a valuable agricultural resource in the
Pampa Faclo land area overrode labor-to-build con-
siderations – perhaps due to population increase
pressures.
The transition to Chimu from Moche eras in the
seventh century CE sees a vast expansion of popula-
tion, cities, administrative centers, agricultural and
2220 Water management in Petra
settlement zones as well as a development of hydraulic
technology that permits intra- and intervalley canal
network development to exploit available agricultural
zones. The success of the agricultural program,
supported by a knowledge base of technical achieve-
ments, undoubtedly underlies this expansion. The
present study adds an example of an application of
hydraulic technology that played a role to protect
critical aqueduct structures from flood damage to
ensure field system survival through time. Other
examples of steep-slope channel constrictions formed
by opposing stones set with a narrow opening are
numerous within the Farfán field system area (Fig. 13)
indicate an understanding of creative use of hydraulics
knowledge to enhance the efficient use of field systems.
Basically channel constrictions of this type create a
hydraulic jump ahead of a constriction with a high
height, low velocity flow leaving the throat that flows
slowly into downstream distributive channels within
the field system to regulate water seepage rates into the
soil. The channel size, slope and inlet configuration
supplying water to maintain a stable hydraulic jump
without overflow must therefore be intelligently
constructed to allow just a sufficient flow rate for that
purpose. This design capability appears to have been
well understood by Chimu engineers as demonstrated
by many canal design examples.
Examples of hydraulic controls in the form of flow
rate limiting and velocity-reducing hydraulic structures
point to a yet little explored creative aspect of Chimu
irrigation agricultural practice and the Chimu hydraulic
science knowledge base. While examples illustrating
applications of Chimu hydraulics engineering are
somewhat limited by scant exploration and analysis
to date, undoubtedly more remains to be discovered as
focus is given to exploration and analysis of hydraulic
features. By computer analysis of such systems, aspects
of Chimu hydraulic science will be revealed and point
to a new evaluation of the contribution of indigenous
South American cultures to the hydraulic sciences.
References
Eling, H. PreHispanic Irrigation Sources and Systems in
the Jequetepeque Valley, Northern Peru. Eds. M. Turpin,
H. Eling and R. Matos. Andean Archaeology. UCLA
Archaeology Publications, 1986: 130–49.
FLOW-3D. V.8.4 Software-Flow Science, Inc., Santa Fe,
New Mexico, 2004.
Kosok, P. Life, Land and Water in Ancient Peru. New York:
Long Island University Press, 1965.
Ortloff, C. R. Chimu Hydraulics Technology and Statecraft
on the North Coast of Peru. Economic Aspects of Water
Management in the Prehispanic New World. Eds. V.
Scarborough and B. Isaac. Greenwich, Connecticut: JAI
Press, 1993. 327–87.
---. Canal Builders of Ancient Peru. Scientific American
256.12 (1988): 100–7.
Ortloff, C. R., Moseley, M. E. and R. Feldman. Hydraulic
Engineering Aspects of the Chimu Chicama- Moche
Intervalley Canal. American Antiquity 48 (1982): 572–99.
---. Hydraulic Engineering and Historical Aspects of the
PreColumbian Canal Systems of the Moche Valley, Peru.
Journal of Field Archaeology 12.1 (1985): 77–87.
Water Management in Petra
C. R. O RT L O F F
Many scholars have studied the political history of
Petra (Taylor 2001; Guzzo and Schneider 2002; Glueck
1959, 1965; Hammond 1973; Levy 1999; Auge and
Denzer 2000; Bowersock 1983; Bourbon 1999;
Markoe 2003). Despite the fact that control of water
is essential to an understanding of life in the desert,
there has been little scholarship on hydraulic engineer-
ing at this site. Figure 1 shows details of the supply and
distribution system leading water to the urban core of
Petra. Numbered locations denote major buildings,
temples, and site features listed in the Appendix. Shown
are major dams (d-), minor dams (d), cisterns (c), water
distribution tanks (T), and springs (S). The grid system
(A,B,C;1,2,3) serves to define an area coordinate system
composed of 1.0 km2 grid boxes to locate various
features mentioned in the text.
The urban core of Petra lies in a valley surrounded
by high mountainous terrain. Seasonal rainfall runoff
passes into the valley through many canyon streambeds
(wadis) and drains out through Wadi Siyagh (A;2). While
water storage is a partial key to the city’s survival, a
number of springs internal and external to the city (Fig. 1:
Ain Mousa, Ain Umm Sar’ab, Ain Braq, Ain Dibdiba,
Ain Ammon, al Beidha, Ain Bebdbeh) provided water
channeled and/or piped to the urban center.
The main Petra water supply originated from the Ain
Mousa spring about 7.0 km east of the town of Wadi
Mousa (Fig. 1, D;1) combined with waters of the minor
Ain Umm Sar’ab spring; this supply still serves the
associated tourist complex (2), (3) located outside of
the Siq (C;1) entrance (10). The Siq is a 2 km long,
narrow passage through high mountains bordering the
eastern part of the city core area. Early phases of water
supply utilized Ain Mousa spring water channeled
through the Siq (dashed line, 29, B;2) to the urban core
of the city as far as Q’asr al Bint (29) with final
drainage into the Wadi Siyagh (A;2). Due to dam and
flood bypass tunnel construction at the Siq entrance
and infilling and paving of the Siq floor both in ancient
and modern times to reduce flooding, the channel now
lies under the current pavement surface attributed to
Nabataean construction under Aretas IV and later
 Water management in Petra 2221

Water Management in Petra. Fig. 1 Petra site features and water distribution systems.

Roman paving efforts. Recent excavations in front of

the Treasury (11, C;1) have revealed remnants of this
early open channel. While this channel provided water
to early, low population phases of the city, the later
concentration of urban settlement areas north and south
of Wadi Mousa (B;2) represented a transition towards
full city status with a cosmopolitan society involved in
trade and commerce. With demands to increase water
supply to spreading urban settlement areas resulting
from population increase and a desire to match the city’s
prosperity from trade with an appropriate elevation in
symbols of success, extensive use of pipelines followed
to bring larger amounts of water to areas not reachable
Water Management in Petra. Fig. 2 The Zurraba (al
by the low elevation, open channel system. Pipeline
Birka) reservoir.
systems, however, introduced new hydraulic design
complexities that involved knowledge of ways to
maintain stable piping flow whose maximum transport record, as well as insights into the management strategy
flow rate matched (or exceeded) spring flow rate input. of these assets, indicate the level of technical achieve-
Flows in poorly designed pipeline systems are capable ment of Nabataean engineers.
of a number of transient, self-destructive hydraulic One example of later phase technological advances
instabilities (e.g., water hammer, pressure surges, is a reservoir at Zurraba (1, D;2) (Figs. 1 and 2). It was
transient wave structure, flow intermittency, or internal constructed to store and transmit water along the Wadi
oscillatory hydraulic jumps). Thus analysis of Petra’s Shab Qais (D;2) around the northern flank of the Jebel
piping systems provides insight into the available el Khubtha mountain (C;2) in an elevated channel W
technical knowledge base applied to problem solution. (D;3, 40) containing piping (Fig. 3) that continued over
Pipeline routing involved constant angle contour path royal tombs (22–24) to supply a typical large basin at
surveying through rugged, mountainous terrain as well its terminus (Fig. 4). Descending channels from this
as choices of hydraulic technical parameters (slope, basin to cisterns at the base of the mountain added
diameter, internal wall roughness, sinuosity, and supply water to that collected from rainfall runoff for urban
head) that govern piping carrying capacity. Choices of housing needs, celebratory rituals at nearby tomb
these parameters, as extracted from the archaeological complexes and for piping elements conducting water
2222 Water management in Petra
Water Management in Petra. Fig. 3 Elevated water
channel/piping system on west face of el Khubtha Mountain.
Water Management in Petra. Fig. 4 Elevated settling
basin typical of the el Khubtha channel/piping system.
further on to city center locations. While runoff capture
is one probable sourcing of the Zurraba reservoir,
connection to local spring sources, including Ain
Mousa, remains probable for additional charging.
Although reservoir water could be used to supplement
the Siq open channel flow, later city phases involved
shifting Ain Mousa water supplies to a Siq piping
system after the open channel was abandoned. In this
case, rainfall runoff and spring charging still enabled
the Zurraba reservoir to supplement Ain Mousa Siq
pipeline flows when required. The Jebel el Khubtha
pipeline appears to be the main outflow path for
reservoir water along Wadi Mataha (B;2, C;2) and
surplus water, after cistern topping, was most certainly
directed to the main city fountain (Nymphaeum 42,
B;2) through a pipeline (as some pipeline fragments in
the area suggest). From a systems point of view, the
reservoir served principally to maintain cistern levels
by on-demand water release while the Ain Mousa
spring provided the continuous spring supply source to
the Nymphaeum through piping supported in a channel
through the Siq (Fig. 5a, b; Ortloff 2005). The ability to
provide an “on-demand” water supply from this backup
Water Management in Petra. Fig. 5 (a) Piping channel
elements on the north side of the Siq. (b) Channel trough on
the north side of the Siq for piping placement.
source would prove most useful to large caravans
entering the city that would place a sudden demand on
water supply capability.
Pipeline Carrying Capacity Considerations: The
Zurraba–Jebel el Khubtha System
While a spring produces a given volumetric flow rate, a
limitation on how much can be transported by pipeline
relates to pipeline technical characteristics (diameter,
internal roughness, slope and supply head). Piping
design considerations require the spring output flow
rate to match (or be less than) the theoretical carrying
capacity of the pipeline. Technical examination of
Nabataean pipeline designs then yields insights into
solutions to increase flow rate throughput. If the long
Jebel el Khubtha piping system were to function in full-
flow mode typical of a very low-flow rate, its flow rate
would be somewhat less than that derived from an open
channel, near critical flow mode due to internal wall
frictional resistance; larger diameter piping would then
be required to match the spring flow rate. For a steeper
slope design, gravitational acceleration causes flows to
become supercritical and tend to an internal free surface
normal depth. Rapid supercritical flows, however,

may be subject to intermittent zones of subcritical full
flow induced by internal piping wall roughness and
curvature resistance effects as well as transient
hydraulic jumps that create pulsations in delivery flow
rate at the piping exit. Such effects can lead to
destructive tensile forces that weaken mortared piping
joints. The best piping design to produce a stable
volumetric flow rate is therefore a partially full flow at
near, but below, critical conditions that empties water
gently into a terminal reservoir. Selection of this piping
design would then be a measure of knowledge of
hydraulic principles required to achieve a steady, high
flow rate to the terminal elevated basin placed far left in
C;2 and would explain the high elevation positioning of
this pipeline (to maintain a low slope) around Jebel el
Khubtha. The Nabataean design, given its slope and
piping diameter, closely matches the near critical flow
rate (Morris and Wiggert 1972) and provides for the
largest possible flow rate from the Zurraba reservoir to
meet on-demand large flow rate requirements. Addi-
tional benefit from the Nabataean design resides in
the presence of partial, open channel flow in the
piping to greatly reduce the leakage rate compared to
a pressurized system. Since particles settle in the
reservoir, no particle transport occurs to clog piping;
this is particularly important as access to the high
elevation piping (25 m above the ground) on the near
vertical Jebel el Khubtha mountain face would limit the
possibility of cleaning procedures. The combination of
all these features designed into the Jebel el Khubtha
pipeline indicates that much thought went into the best
placement and design of this system to achieve the
multiple goals that ensured not only system longevity
but also rapid, on-demand water delivery capability
with minimum leakage.
From an exit pipeline leaving the basin to a ground
level cistern, additional piping led to the Nymphaeum
fountain to complete the Jebel el Khubtha circuit from
the Zurraba reservoir. As the Nymphaeum was a major
water supply to the urban core and market areas, much
effort and innovation was employed to guarantee its
year-round functioning from the Siq piping system
(Ortloff 2005) supplemented, on-demand, by the long
Jebel el Khubtha pipeline from the Zurraba reservoir.
Supplemental Water Supply Systems Supply
Redundancy
A number of cisterns and dams on Jebel el Khubtha
(Akasheh 2003) (C;2) captured and stored rainfall
runoff. Some of the upper level cisterns appear to have
channels leading to ground level cisterns that led to
urban housing or field areas to the west of Jebel el
Khubtha to supplement the water supply from the
Zurraba system. As previously mentioned, the Siq floor
open channel was abandoned in late Nabataean phases
Water management in Petra 2223
and replaced by a Siq north wall pipeline system
(Fig. 5a, b) that extended to the area across from the
theater district (B;1) and ended at the Nymphaeum.
Thus at least two separate supply lines (Siq and Jebel el
Khubtha) led to the Nymphaeum to ensure supply
redundancy. The construction for the Siq pipeline
system is generally attributed to Malichus II or his
predecessors, Aretas IV or Obodas III, in the first
century BCE or early in the first century CE (Guzzo
and Schneider 2002). Since water demands south of
Wadi Mousa (transecting urban Petra) are high due to
the nearby marketplace, theater, temple, and housing
districts and significant water resources are available
from the north side piping systems, a pipeline transfer
connection from the Nymphaeum to this area is logical
for development of this area. While a bridge or piping
from the north side of the Jebel el Khubtha system
in the El Hubtar Necropolis area (20, B;2) across the
Wadi Mousa may have existed in the vicinity of the
theater (19, B;1) to carry water (at the same level) to
the south side, traces are lost due to extensive erosion
flood damage. In addition to water delivered by these
means, the theater water source was supplemented
from large, upper level reservoirs in the Wadi Farasa
area, and pipelines originating from Ain Braq and Ain
Ammon sources (Fig. 1) again indicating built-in
supply redundancy from multiple sources. Some of
the larger reservoirs appear to function in connection
with a spring supply system and are situated to collect
rainwater runoff; reservoir usage, therefore, is mainly
to provide water for occasional peak requirements.
Surface cisterns appear to be opportunistically placed
to collect rainwater runoff; other than seasonal rain
recharge, the numerous, widely scattered catchments
appear to serve local community needs for supplemen-
tal supplies of lower quality water when piped water is
not readily accessible.
Traces of a south side piping system (Fig. 6) are
found in front of the theater. Two parallel pipelines
W
Water Management in Petra. Fig. 6 Dual pipelines
continuing past the theater to supply tanks (T) above the
Cardo.
2224 Water management in Petra

Water Management in Petra. Fig. 7 Great Temple on the south side of Wadi Mousa.

continue past the theater along the ridge (B;2) above

the commercial district along the Roman Cardo (25),
Hadrian’s Gate (43), upper and lower marketplaces and
the Paradeisos water garden (Bedal 2004) to locations
above the Great Temple (28, Figs. 1 and 7; Joukowsky
2001, 2003) where they form part of the water supply to
structures located in (B;2). The two separate pipelines
may indicate branch lines to separate destinations or a
later elevation change that continues piping to Q’asr al
Bint through the Great Temple to supply the Sacrificial
Altar area, although no excavations exist to connect the
multiplicity of subterranean canals below the altar to a
specific water source. Hadrian’s Gate (43) separates the
secular commercial district from the western sacred
temple district containing the Great Temple, Temple
of the Winged Lions (26), and Q’asr al Bint. The
Paradeisos water garden complex west of the gate
consisted of an open house structure situated on a
platform island within a large water filled basin; bridge
structures connected the island to outer precincts, and
greenery added to the city’s elegance as indicated by
reconstructions reported by Bedal (2004). The basin
walls contain overflow channels as well as supply
piping that may emanate from both the Nymphaeum Water Management in Petra. Fig. 8 Fragment of one
piping extension into this area and water from a south of the elevated water collection basins (T) above the
side spring supply system. Distributed along this piping Cardo area.
system, a number of elevated basins (T, Fig. 1) lined
with hydraulic plaster (Fig. 8) served as receiving core region contains the marketplace area, water
basins; earth-fill mound structures extended from these requirements were high; consequently, additional
tanks to the lower Cardo area and served as pipeline supplies are channeled to this area by means of an
support structures. As the basins are elevated 20 m on a underground channel (B;1; B;2) from the combined
bluff above the Cardo, sufficient head existed to flows from Ain Braq and Ain Ammon. Some as yet
provide pressurized water for fountains in the market unexcavated branch of this system running through
area below as well as for the Great Temple (and high elevation channels may be part of the system that
possibly Q’asr al Bint). Because the south side urban provided water to piping located in front of the theater.

Water from these springs may be supplemented by
elevated cistern storage water from the Jebel Attuf
area (B;1) in one of the many high places (12, 13, B;1) of
the city. Water to the Lion Fountain (14, B;1) and al
Hamman pool area in the vicinity of elite tombs (16–18;
B;1) came from this supply line that continued on to the
Great Temple area and clearly indicated that a continuous
spring supply is part of the system due to the presence of
the Lion Fountain. A large elevated cistern located on a
plateau above the Tomb of the Roman Soldier (16)
(Browning 1982) also contributes rainfall runoff water
supplies into this system. Details of the Wadi Farasa
water system in this area (B;1) have been investigated
(Schmid 2000) and indicate the existence of large
reservoirs and piping systems that not only serve local
usage, but also have sufficient capacity to transfer water
further west to the Great Temple area. Numerous
channels, pipelines, and multiple cisterns within, and
leading from the Great Temple, indicate that water
supplies within the temple were abundant (Joukowsky
1999, 2001, 2003). Water export lines to the marketplace
area and the Q’asr al Bint region from the Great Temple
served as part of the water system.
Water Supply System Management Operations
The evolution of the water system to incorporate piping
networks transformed the site to meet the demands of a
large urban population estimated to reach 30,000
(Guzzo and Schneider 2002). The water system
incorporated both intermittent, on-demand supplies
piped from large reservoirs or drawn from cisterns and
continuous supply piping systems from remote springs
to provide daily requirements of city inhabitants. No
water could be wasted. As a consequence, transfer
piping from north side systems (Jebel el Khubtha and
Siq pipelines to the Nymphaeum) provided water that
could be transferred to south side downhill locations for
further usage or storage before final discharge into the
low elevation Wadi Siyagh.
Water storage through use of major, on-site dams
presented yet a further aspect of Petra’s water system.
For example, on the north side of Wadi Mousa,
numerous high status structures exist in the B;2
quadrant [Temple of the Winged Lions, Royal Palace
(41), North Defense Wall and Fortress (35), Conway
Tower (54)] and are logically associated with some
water supply system. A dam (d) at Wadi Turkamaniya
(B;2) may have trapped and stored sufficient runoff
to provide water to the lower reaches of the Temple
of the Winged Lions although no excavation data is
available. Excavations reveal that lower portions of
both the Temple of the Winged Lions and the Great
Temple spanned the Wadi Mousa stream by means of
bridging. It appears then that supply redundancy
derived from pipelines from different spring/reservoir
Water management in Petra 2225
sources crossing from one to another part of the city is an
aspect of the design approach. This design philosophy
ensures that water supply to any area may be composed
from different sources depending upon variations in
individual spring flow rates and reservoir/cistern storage
amounts and implies that management oversight was
in place to monitor and control the system network.
While cisterns are well dispersed through the urban
settlement area, a main underground channel starting
from Ain Bebdbeh north of (D;3) and running toward
the convergence of Wadi Mataha and Wadi al Nassara
(B;2), runs into the lower reaches of the north side
below the temple areas. To illustrate the Nabataean
mindset to utilize all water resources, a further element
based upon on-site dams constituted yet further
complexity to the water management picture. Local
histories mention the existence of large dams – one on
the Wadi Mataha (Taylor 2001), the other on the Wadi
al Nassara (Fig. 1). Destroyed remains are found to
verify that these dams provided water storage from
rainfall runoff within urban Petra. Judging from
Nabataean placement of the Wadi Mataha dam (d),
piping to the nearby Nymphaeum must have been an
additional third backup water source to the fountain.
Since the Ain Mousa spring could also serve to place
water behind the Wadi Mataha dam through the Wadi
Shab Quais pipeline branch from Zurraba (in addition
to seasonal rainfall runoff storage behind the dam from
diversion of the Wadi Mousa stream through the bypass
tunnel, 8 in Fig. 1), the water level behind the dam
could be maintained to provide backup water to the
Nymphaeum throughout the year. The Nymphaeum
could then be supplied by contributions from stored
runoff water behind the Wadi Mataha dam, a canal or
pipeline from Ain Bebdbeh, the pipeline along the
western face of Jebel el Khubtha feeding ground level
cisterns and pipelines, and the north side Siq pipeline.
This degree of redundancy indicates that planning for
water supply variations was a consideration addressed
by a complex design that could tap into various
pipeline-water storage resources depending upon
available supplies.
Flood Control, Groundwater Recharge, and the
Great Temple Water Subsystem
Floodwater drainage during the rainy season was a
major concern. Since heavy rainfall and flooding
characterize the Petra area, measures to divert Wadi W
Mousa floodwater from the Siq by means of a bypass
tunnel (Fig. 1, 8 at C;1), a low dam at the Siq entrance
and elevation of the Siq floor near the entrance
provided flood control. While this strategy had proven
effective in deflecting small flood events, continuous
deliberate infilling and accumulating flood deposits in
the Siq helped protect against floodwater incursion.
2226 Water management in Petra

Water Management in Petra. Fig. 9 FLOW-3D calculations of a rectilinear model of the Siq piping for entry full flow
input flow rates of 1.0 (a), 2.0 (b), 5.0 (c), 10.0 (e) ft/s (0.305, 0.610, 1.52, 3.05 m/s) velocity indicating internal flow
development for the observed wall roughness distribution. Fig. 9d shows results for smooth interior piping at 1.52 m/s initial,
full flow inlet velocity.

While large flooding events had negative consequences,
there were also ways to utilize the sudden water bounty:
storage dams across the numerous wadis intersecting
the urban core served to reduce floodwater entry into the
city while seepage from the impoundments provided
water table recharge suitable for well extraction during
protracted drought. Thus a fraction of the seepage from
dam storage, canals and pipelines ultimately can be
recaptured and used as an ultimate groundwater defense
against drought on a citywide basis.
The same idea could also be used on a more
localized basis for elite structures. Within the Great
Temple, an elaborate south boundary wall drainage
channel system collected infiltrated rainfall seepage
and directed it to a nearby underground cistern with
50 m3 capacity located within the eastern, upper part of
the temple structure. A channel connected to the upper
part of this cistern conducted overflow water to lower
level structures before exiting to Q’asr al Bint and Wadi
Siyagh in order to regulate its maximum capacity.
Large channels located in an upper room north of the
Theatron (Joukowsky 2003) of the temple most likely
indicate the terminus of a subterranean channel from
the Ain Braq, Ain Ammon system (B;1, B;2) with
water transfer access to the cistern. Some additional
water sources may have been available from springs in
Wadis Kharareb and Ma’Aisert (Fig. 1) although piping
connections await further excavations. Channel water,
supplemented by cistern water to meet peak demands,
was then distributed to subsidiary open cisterns located
in east and west sides of the temple and then through
subterranean channels under the lower temenos (a

Water Management in Petra. Fig. 10 Typical rippled wall
pattern within the interior of some Nabataean piping
elements.
sacred enclosure around a temple or holy site) platform
to lower level rooms near the temple entrance stairway.
Thus the cistern functioned as a reservoir adding stored
runoff and seepage water to the channel-delivered
base supply when required – much in the same way
that previously described reservoir–pipeline systems
worked to meet occasional peak demand requirements.
Because such a system is contained within the temple
itself, its position of importance as a major canal
terminus and water distribution node is clear from the
complexity of hidden channels, cisterns, and piping
thus far discovered. Perhaps the internal water system
of the temple, capable always of providing ample water
supplies for rituals, had special significance to
demonstrate the premier role of religion in the lives
of the Nabataeans; only later under Roman rule are
these supplies used for more utilitarian Cardo market-
place purposes indicating Roman predilection to
practical concerns.
While the north side piping of the Siq provided the
main potable water supply, the south side channel
system was probably meant for animal watering and
may have been supplied by a channel from Ain Braq
and supplemented from a cistern atop the bluffs with a
drop hole to this channel (C0, Fig. 1).
Per Capita Water Availability
For estimates of total water volumetric flow rate into
the city, conservatively assuming about a third of the
north side supply rate for the south side due to the less
robust south side springs, and assuming a combined
flow rate between the Siq and Wadi Shab Qais reservoir
release lines to be about 40 m3 h−1, then the city could
receive at minimum 50 m3 h−1 from these sources.
While additional sources [Ain Braq (0.8 m3 h−1), Ain
Dabdabah (2.5 m3 h−1), Ain Ammon, and Ain Siyagh
(<1.0 m3 h−1)] add to this estimate, released storage
Water management in Petra 2227
water and additional springs would add yet further
capacity to arrive at a yet higher flow rate estimate –
perhaps to 100 m3 h−1 total if less conservative leakage
estimates are used. For about 30,000 inhabitants, then
at least 0.04 m3 day−1 (40 l day−1) would be available
on a per capita basis, which is minimal to maintain
hygienic standards. Considering public use of foun-
tains, watering troughs, baths, water gardens, and
workshops, per capita water availability is well within the
Rome urban water usage rate of 0.6 m3 day−1 per person
(Butterfield 1964). Considering hot climate survival
water intake per person of about 0.003 m3 day−1, then
surplus water beyond human consumption/survival rates
is available by these estimates. Estimates for water
storage from dams and cisterns on Jebel el Khubtha
alone (Akasheh 2003) is 0.36 m3 per person; for a
personal consumption rate of 0.003 m3 day−1, then
theoretically, considering evaporation losses, about a
2 month emergency supply was available through Jebel
el Khubtha cisterns if spring flow rates declined
precipitously.
A similar calculation shows that Zurraba reser-
voir contained about 3 weeks emergency supply and
additional cisterns and dams on site, particularly those
of the Wasi Farasa system (Schmid 2000) add yet
further reserves totaling several month’s supply. While
per capita water supply for Petra is somewhat lower
than Rome city standards, still the amounts supplied to
the city on a continuous basis are more than sufficient
to maintain quality living standards.
As the city underwent major construction in the
period 50 BCE–100 CE during the reigns of Obodas II
(III) and Aretas IV, Roman/Greek technologies most
probably influenced water system design. The avail-
ability of advanced Roman surveying techniques used
for aqueduct system design (Lewis 2001) would be
particularly useful in constructing long distance lines of
prescribed slope to and within Petra to maintain stable
open channel flows in piping to match spring flow
rates. While the design capability to achieve such
balances still remains elusive due to fragmentary
knowledge of ancient hydraulics practices, some hints
of water flow rate measurement capability exist (Cohen
and Drabkin 1966; Ortloff and Crouch 1998).
In essence, the Nabataeans utilized all possible
above-and-below groundwater supply and storage
methodologies simultaneously. While water storage in
contemporary Hellenic cities also emphasized cistern
water storage for household use, the Petra systems W
advanced this technology to citywide systems with
elaborate dam and cistern systems that served both
water storage and flood control purposes. Water storage
in groundwater aquifers was also practiced by multiple
dam system storage; this allowed for the possibility of
constructing wells as backup systems should all other
supply systems decline due to long-term drought
effects.
2228 Water management in Petra

Appendix: Site Features

1. Zurraba reservoir (al Birka) 31. Quarry
2. Forum Rest House 32. Lion triclinium
3. Park Entrance 33. El Dier
4. Hospital 34. 468 monument
5. Dijn monuments 35. North City Wall
6. Obelisk Tomb and Bab el Siq 36. Turkamaniya tomb
Triclimium
7. Entrance elevated arch 37. Armor tomb
8. Flood bypass tunnel and dam 38. Little Siq
9. Eagle monument 39. Aqueduct
10. Siq 40. Al Wu’aira Crusader Castle
11. Treasury (El Khasneh) 41. Byzantine tower
12. High place 42. Nymphaeum
13. Dual obelisks 43. Paradeisos, Market, Hadrian’s Gate
14. Lion monument 44. Wadi Mataha dam
15. Garden tomb 45. Bridge abutment
16. Roman soldier tomb 46. Wadi Thughra tombs
17. Renaissance tomb 47. Royal tombs
18. Broken pediment tomb 48. Jebel el Khubtha high place
19. Theater 49. El Hubtar Necropolis
20. Uneishu tomb 50. Block tombs
21. Royal tombs 51. Royal tombs
22. Sextius florentinus tomb 52. Obelisk tomb, snake monument
23. Carmine façade 53. Columbarium tomb
24. House of dorotheus 54. Conway tower
25. Colonnade Street (Cardo) 55. Tomb complex
26. Temple of the winged lions 56. Convent tombs, crusader fort
27. Pharaoh’s column 57. Tomb complex
28. Great Temple 58. Pilgrim’s spring
29. Q’asar al Bint 59. Jebel Ma’Aiserat high place
30. Museum 60. Snake monument

A water supply system consisting of dams, cisterns,
channels, pipeline networks, and groundwater storage
exploited multiple spring supply systems and rainfall
runoff collection. Examination of two different pipeline
system designs with different slopes and delivery
requirements (one on-demand, the other continuous
flow) indicates that technology is in place to provide
designs that minimize leakage, maximize flow rate,
minimize particle transport, and eliminate transient
flow instabilities that cause system vibration and
prevent steady flow from developing. Further ad-
vances, as related to the Siq pipeline, relate to water
purification by four easily cleanable settling basins
(Bellwald 2004) whose positioning eliminates a
complex flow stability problem (Ortloff 2005). The
totality of solution options confirms that a hydraulic
design methodology is applied with skill to solve
complex hydraulic engineering problems.
While it is traditional to look for earlier Greek and
Roman technical advances (Ortloff and Crouch 2001;
Cohen and Drabkin 1966) that improve the Nabataean
system, few are found, indicating that Roman engineers
perhaps viewed the Nabataean system as optimum. In
this case, it is likely that the water management techni-
ques observed by the Romans served to increase their
library of water conservation and management techni-
ques as applied to desert cities and outposts. While
details related to technical transfer and hydraulic
engineering practices of ancient societies are still a matter
of research, it is certain that the longevity of Petra based
upon its innovative water system design constitutes a vital
chapter in the history of water management (Fig. 9).
References
Akasheh, T. Nabataean and Modern Watershed Management
Around the Siq and Wadi Mousa in Petra. Hashemite
University Report, The Hashemite University, Zarqa,
Jordan, 2003.
Auge, C. and J. Denzer. Petra, Lost City of the Ancient World.
New York: Abrams, 2000.
Bedal, A. The Petra Pool-Complex: A Hellenistic Paradeisos
in the Nabataean Capital. Piscataway, NJ: Gorgias Press,
2004.
Bellwald, U. Personal Communication, 2004.
Bourbon, F. Petra: Art, History and Itineraries in the
Nabataean Capital. Vercelli: White Star, 1999.
Bowersock, G. Roman Arabia. Cambridge, MA: Harvard
University Press, 1983.

Butterfield, R. Ancient Rome. New York: Odyssey Press,
1964.
Cohen, M. and I. Drabkin. A Source Book in Greek Science.
Cambridge, MA: Harvard University Press, 1966.
Glueck, N. Rivers in the Desert. New York: Straus and
Cudahy Publishers, 1959.
---. Deities and Dolphins. New York: Strauss and Cudahy
Publishers, 1965.
Guzzo, M. and E. E. Schneider. Petra. Chicago: University of
Chicago Press, 2002.
Hammond, P. The Nabataeans – Their History, Culture and
Archaeology. Gothenburg: Åström, 1973.
Joukowsky, M. S. Petra Great Temple. Brown University
Excavations 1993–1997. Vol. 1. Providence: Brown
University Publications, 1999.
---. Petra Great Temple. American Journal of Archaeology
106 (2001): 435–58.
---. Petra – The Great Temple Excavation, 2002
Field Campaign. Providence: Brown University Report,
2003.
Levy, U. The Lost Civilization of Petra. Cambridge, MA:
Floris Books, 1999.
Lewis, M. J. T. Surveying Instruments of Greece and Rome.
Cambridge: Cambridge University Press, 2001.
Markoe, G. Petra Rediscovered. New York: Harry N.
Abrams, Inc. Publishers, 2003.
Morris, H. and J. Wiggert. Applied Hydraulics in Engineer-
ing. New York: The Ronald Press, 1972.
Ortloff, C. R. The Water Supply and Distribution System
of the Nabataean City of Petra (Jordan), 300 BCE–AD
300. Cambridge Archaeological Journal 15:1 (2005):
93–109.
Ortloff, C. R. and D. Crouch, Hydraulic Analysis of a
Self-Cleaning Drainage Outlet at the Hellenistic City of
Priene. Journal of Archaeological Science 25 (1998):
1211–20.
---. The Urban Water Supply and Distribution System of
the Ionian City of Ephesos in the Roman Imperial
Period. Journal of Archaeological Science 28 (2001):
843–60.
Schmid, S. The International Wadi al-Farasa Project.
Association for the Understanding of Ancient Cultures
Report 2002, Basel, Switzerland, 2000.
Taylor, J. Petra and the Lost Kingdom of the Nabataeans.
Cambridge, MA: Harvard University Press, 2001.
Water Procurement in the Prehistoric
Caribbean
P E TER D REWETT
When Ceramic Age peoples arrived in the Caribbean,
water was readily available on some volcanic islands
like Dominica but virtually absent on most of the coral
islands like Barbados. Initially this would have been a
major constraint to settlement and survival. In years
of regular rainfall, rainwater could be collected from
rock pools, caught in foliage of plants and collected
Water procurement in the prehistoric Caribbean 2229
in pottery vessels or even upturned conch shells. When
the rains failed, however, drought would have led
to serious problems on islands without the ability to
maintain surface water supplies in the form of rivers
and streams. On at least three islands in the southern
Caribbean, Barbados, Mustique and Carriacou, Amer-
indians developed a specific alternative way of guaran-
teeing a water supply. It may be assumed that the
technique was used on other islands although archaeo-
logical evidence has not yet been located.
On coral rock islands (Machel 1999), where the
bedrock is highly permeable, salt seawater may
infiltrate under all of the island or at least around the
coastal fringe. Rainwater falling onto the island will
seep down until it reaches denser salt water. It will then
flow towards the sea on this denser water. At the head
of beaches it will be quite shallow before it mixes
with seawater (Schultz 1995). Somehow Amerindians
recognised this source of sub-surface fresh water and
tapped into it. Although similar to sinking wells
elsewhere in the world, the process of tapping into this
water required great skill; otherwise salt and fresh
water would be mixed and the water source spoiled.
Excavations at the Amerindian settlement at
Heywoods (Port St Charles) on the west coast of
Barbados revealed two methods used to tap into this
shallow supply of fresh water (Drewett 2000). The main
Ceramic Age settlement at Heywoods was occupied
from about AD 600 to AD 1200 (Late Saladoid –
Troumassoid). Several round houses and burial areas
have been excavated together with wood- and pot-lined
water holes (Drewett 2000).
Two wood-lined water holes were excavated. These
consisted of wood planking some 1 m in length sunk
into the soft sand and held in place by large coral rocks.
Both were roughly square in shape (Fig. 1).
W
Water Procurement in the Prehistoric Caribbean.
Fig. 1 Wood-lined water hole at Heywoods, Barbados. ca.
AD 800–1000 (source by author, photo by Abel Drewett).
2230 Water procurement in the prehistoric Caribbean
The wood used was false mastic (Mastichodendron
sp.) and lignum vitae (Guaiacum sp.). One was dated
to ca. AD 780–1020 (B-147314) and the other to ca.
AD 790–1030 (B-147313). In both wells calabash
(Crescentia cujete) remains were recovered suggesting
these were used to scoop up the fresh water (Fig. 2).
More common, and possibly more successful, were
the pot-lined water holes. A shallow hole was dug into
the underlying wet sand and a large cooking/storage pot
was sunk into the wet sand. These pots had been pre-
prepared with their bases carefully removed (Fig. 3).
The pot was carefully pressed into the wet sand
while more sand was scooped out of the interior.
Additional pots were then stacked on top, acting as a
cofferdam to stop wet sand slumping back in. The pots
were held in place with coral rock boulders and/or
wooded wedges. After a couple of hours any sediment
in the water would settle with the pressure of the
underlying water pushing the fresh water up into the
pot-lined hole (Figs. 4 and 5).
It is uncertain whether these pot-lined water holes
were left open or had a removable cover. None were
Water Procurement in the Prehistoric Caribbean.
Fig. 2 Calabash water scoops from wood-lined water hole at
Heywoods, Barbados. Fragments on left ca. AD 800–1000.
Modern example on right (source and photo by author).
Water Procurement in the Prehistoric Caribbean.
Fig. 3 Large cooking/storage pot with base removed
found in pot-lined water hole at Heywoods, Barbados. ca.
AD 1000–1200 (source by author, photo by Abel Drewett).
found covered although several had large sherds of
pottery in the water holes, which could have been used
as lids. Flat cassava griddles found elsewhere on the
site would have also made suitable lids, as would
organic materials like large leaves or planks of wood. A
major problem was probably local soiling of the water
by domestic rubbish or even human burials, which
were also within the domestic area. This might explain
why holes had to be regularly replaced. Twenty three
were systematically excavated during the main phase of
salvage excavations at Heywoods during 1998/1999,
but many more were noted during building work for the
Port St Charles marina constructed on the site.
Some of the pot-lined water holes had a highly
decorated pot rest resting on the top of the stack. These
are like ceramic collars in which large unstable pots
Water Procurement in the Prehistoric Caribbean.
Fig. 4 Pot-lined water hole at Heywoods, Barbados. ca.
AD 1000–1200. Plan view. (source by author, photo by
Abel Drewett).
Water Procurement in the Prehistoric Caribbean.
Fig. 5 Pot-lined water hole at Heywoods, Barbados. ca.
AD 1000–1200. Section view (source by author, photo by
Abel Drewett).

were set to make them stable on uneven ground. One
found was 32 cm in diameter and 17 cm high. It was
red-slipped with white on red paint and grooved
decoration in a concave-ended rectangle and vertical
line pattern. This was among the finest pottery found on
the site, so it would suggest great importance of the
mouth of the hole. Tapping into this underlying source
of water may have involved great perceived risk
involving careful ritual, for Amerindians believed that
a river of the dead ran under their houses and
settlements. Ritual specialists were no doubt involved
in the placing and digging of these pot-lined water
holes and perhaps hallucinogenic drugs, like cohoba,
were taken prior to digging rituals (Drewett 2002).
How widespread the use of pot-lined wells was in the
prehistoric Caribbean remains uncertain particularly
because of limited archaeological excavation in the
southern Caribbean. To date they have certainly been
recorded on three sites on Barbados: Heywoods, Spring
Garden and Maxwell. They have also been recorded on
Mustique and Carriacou. Many further examples may
have been lost by coastal erosion or await discovery.
References
Drewett, P. L. Prehistoric Settlements in the Caribbean:
Fieldwork in Barbados, Tortola and the Cayman Islands.
London: Archetype Publications, 2000.
---. Amerindian Stories: An Archaeology of Early Barbados.
Bridgetown: Barbados Museum and Historical Society,
2002.
Machel, H. G. Geology of Barbados. Bridgetown: Barbados
Museum and Historical Society, 1999.
Schultz, C. The Carriacou Hypothesis: Bottomless Stacked
Pots, A Study in Amerindian Fresh Water Procurement.
Proceedings of the XVI International Congress for
Caribbean Archaeology. Basse Terre: International Asso-
ciation for Caribbean Archaeology, 1995. 217–28.
Water Systems in Bronze
and Iron Age Israel
AVRAHAM FAUST
The Near East is located on the edge of the desert. Part
of the region is arid\semi-arid and, as droughts are a
frequent phenomenon, water is a scarce resource even
in the Mediterranean climate zone. Water was therefore
an important factor determining the situation of set-
tlements in antiquity. Initially, many settlements were
located near water sources, mainly springs (rivers are a
rare phenomenon in Israel). However, since security
was also a major consideration, larger settlements
gradually tended to be situated on top of hills. In this
Water systems in Bronze and Iron Age Israel 2231
conflict of interests, security had, in most cases, the
upper hand, and alternative means of obtaining
drinking water had to be secured.
Wells
A well is an artificial shaft sunk into the ground all the
way to the water table. One of the earliest known wells
was uncovered at the submerged Neolithic village of
‘Atlit Yam. Later wells are known from Beersheba,
Lachish, Tel Zeror, etc., and it seems as if the practice
became widespread from the Iron Age onward (and
mainly from the Classical periods). Notably, however,
wells are typical of lower regions and plains. It is more
difficult to dig wells in mountainous areas because of
both the depth of the water table and the hardness of the
rocks. Wells were, therefore, only a partial solution, and
in most cases people used different methods (but see
below).
Cisterns
Cisterns were cut out of rock in order to collect and
store rainwater. The first known cisterns were appar-
ently dug into the soft limestone of the Coastal Plain in
the late Chalcolithic or early Early Bronze Age at
Mesar. The earliest cisterns were not plastered, but due
to the softness of the stone could still hold the water.
The first plastered cisterns were discovered in several
Middle Bronze Age sites (e.g., Hazor and Gezer) (see
Extra 1). Their use, however, was still sporadic.
Plastered cisterns became widespread only during Iron
Age I, and mainly in Iron Age II. The use of cisterns
became even more widespread in later periods. While
cisterns became common, and at times dozens of
cisterns were unearthed in a single site, it appears that
they were still dug at the family level. Communities, let
alone cities/states, had to develop a better solution to
the danger posed by the scarcity of water.
Large-Scale Water-Works
Large settlements, if not located on a large spring or a
river, needed a guaranteed water source for the summer,
and all the more so for periods of siege. After all, the
fate of a conquered city in the ancient Near East was
horrific, and any effort was done to secure the city and
to enable it to withstand siege.
The earliest large-scale water works, evidently
beyond the scope of the household, are dated to the W
Early Bronze Age – the time period when the first cities
emerged in the southern Levant. These early water
systems are basically large pools, usually located in the
lowest part of the settlements, into which rainwater was
collected. Such systems were found at ‘Arad, ‘Ai and
probably also at Tel Qashish. ‘Arad, located on the edge
of the desert, is a unique example. The city was built
2232 Water systems in Bronze and Iron Age Israel
not on top of a hill, but rather on a basin, probably in
order to ease the collection of water. A slightly different
type of water system was located outside the city of
Jawwa, in Transjordan. Here, dams were built on a dry
wadi, absorbing floods, and securing water for the dry
season. Such large-scale works were able to support a
large population.
Still, in the climatic conditions of the Land of Israel, in
which droughts are frequent, rains cannot be regarded
as a secure source of water. In cases of failing rain,
people would have to go to the nearest spring (usually
just below the city). In times of siege, however, which
began to become a major consideration as time pro-
gressed, such an enterprise was not feasible. Cities had
to rely on a more permanent source of water. They deve-
loped more sophisticated methods to guarantee access
to permanent sources of water, i.e., to springs (which
were usually located in the valleys below the cities).
Waterworks technology seems to have advanced
substantially in the first half of the second millennium
BCE. The notable water system of this period (Middle
Bronze Age) is that of Jerusalem (see Extra 2). Here, an
attempt was made not to collect rainwater, but to secure
access to the water of the nearest spring – the Gihon –
which was located below the city at the Kidron valley.
From within the city, a large tunnel was dug in the hard
limestone. The tunnel went outside the city wall until it
opened into a large pool that was located not far from
the spring and was protected by fortifications. Water
flew from the spring to the pool by way of two tunnels.
Due to level differences, the engineers had to raise
the level of the water by more than 2 m, in order to
let them enter the first tunnel. This was done by the use
of dams.
The real “breakthrough” in water systems, however,
was accomplished during the Iron Age, and mainly
during Iron Age II. A relatively large number of Iron
Age water systems were unearthed over the years,
and these can be divided into the following types:
1. Covered stairways (galleries). This is the simplest
form of Iron Age water work. A shallow tunnel
was dug on the tell’s1 slope, from beneath the city
wall, all the way down to the spring. The tunnel
was then covered. Thus, in times of siege, city
inhabitants could sneak from the city to the spring
through the covered stairway, without being seen
by the enemy. Such systems were discovered in
Megiddo and Tell es-Saiyidyeh. Obviously, these
systems had their limitations. The spring was
known to the enemy, and the gallery could easily
be discovered. Once discovered, it could not have
been used anymore.
1 2
A tell is an artificial mound formed by the accumulated
remains of ancient settlements.
2a. Tunnels. Many Iron Age water systems were
dug under the ground, from within the city to the
spring. Initially, a large shaft was dug through
the earlier layers of the tell. Since these were
pretty loose, the shaft was supported by massive
retainer walls that prevented their collapse. After
reaching bedrock, a smaller tunnel was dug in
the rock, at an angle that lowered it toward
the spring. Such systems were discovered for
example in Gibeon and Megiddo (early phase of
the tunnel).
2b. A modification of this type of water system is
also known. In some cases, a tunnel brought the
water from the spring toward the bottom of the
shaft (i.e., toward the city). Such systems are
known, for example, at Megiddo (a later phase);
the Warren Shaft system in Jerusalem (see note:
Jerusalem water system during the Iron Age) is a
variation of this type.
While in many cases the enemy who besieged the
city could still enjoy the water supplied by the spring,
such systems provided (1) a secure approach to the
water by the defenders; (2) a better chance of con-
cealing the system from the enemy. Still, such systems
were still susceptible to discovery, and the enemy
could harm the water source itself (e.g., poison it).
Furthermore, the enemy could even use the tunnels as
a mean to approach the city in surprise.
3. “Tunnels” dug to the water table. This type is very
similar in its form to the previous one. A wide
vertical shaft was dug through the tell’s layers to
bedrock, and this was followed by a narrow
diagonal tunnel that was dug in the rock. This
diagonal tunnel, however, did not reach the spring;
it descended until it reached the water table. While
very similar in form to type b, this is simply a large
and complex well. Such systems were not only
concealed from the enemy, but were also complete-
ly secured. Some were discovered, for example, at
Hazor and Gezer.
4. The last type of large scale Iron Age water works
is that of large underground reservoirs, dug into
bedrock (inside the cities). Rainwater was collect-
ed through an extensive system of plastered
tunnels (e.g., at Beth Shemesh), or from flood-
waters in nearby wadies2 (Beersheba). Smaller
reservoirs are also known in this period, e.g., in the
fort at ‘Arad. While not as reliable as permanent
water sources, it appears as if such large reservoirs
could absorb enough water underground, to
overcome the limitations discussed above (regard-
ing the dependence on rain water).
A wadi is a valley, ravine or channel that is dry except in the
rainy season.

The galleries (type 1) were the simplest type, and
indeed they were used earlier in the Iron Age. It appears
as if type 2 was developed to overcome the limitation of
the galleries. Type 3 was seemingly the most advanced
system. We do not know if these systems were developed
on the basis of the knowledge that there was a water table
“down there” (the people of the time had to have had this
knowledge, as they were familiar with wells), or whether
hewers of a type 2 system accidentally reached the water
table, and this discovery gave rise to the development of
this type (structurally, type 3 is very similar to type 2).
This reconstruction is of course speculative, and we need
more detailed chronological information in order to
confirm or refute it.
Various water systems are also mentioned briefly in
the period’s written sources, e.g., the Bible (e.g., II
Kings 20: 20; Isaiah 22: 9–11; II Chr. 32: 30), and
ancient inscriptions (e.g., the Mesha inscription).
The most sophisticated Iron Age water system is
Hezekiah’s tunnel, which was dug in Jerusalem during
the late eighth century BCE. A 533 m. long underground
tunnel brought the water from the Gihon spring to a
lower point that was located in the other side of the city,
in order to collect the water in a pool in a secure place.
In later periods, and especially from the Hasmonean
and Roman period onward, larger water systems, e.g.,
ten km of aqueducts were built, but this is beyond the
scope of this entry.
Extra 1: The Open Cisterns of the Negev in the
Tenth Century BCE
During the tenth century BCE, the period of the biblical United
Monarchy, dozens of settlements were built in the Negev desert of
southern Israel. The exact nature of these sites is debated: some view
them as royal forts, others believe that only the initiative was royal,
while some consider that the entire phenomenon resulted from the
settlement of the desert’s nomads. Notably, these sites were located
away from the few springs that exist in the Negev, and their main
source of water was large open cisterns that were discovered at
practically every site. These large cisterns were probably covered
with skins in order to minimize evaporation. It is also likely that when
the rains failed, the inhabitants of many of the sites could go back to
the springs, located several kilometres away, and bring drinking water
from them.
Extra 2: Jerusalem’s Water Systems During the
Iron Age
The Gihon spring of Jerusalem is one of the major springs in the
central hill country. It is likely that it was one of the causes for the
establishment of the city on the low hill, later known as the City of
David. Notably, the spring was used by the inhabitants of the city
from earliest times. The most ancient water systems are two tunnels,
probably dated to the Early or Middle Bronze Age. During the
Middle Bronze Age an impressive water system was dug in
Jerusalem (above in the text). It is during the Iron Age, however,
that the most famous water-works were constructed. As many issues
are still debated, the following is only a brief summary that is
intended to give a succinct overview (and it cannot deal with all the
Water systems in Bronze and Iron Age Israel 2233
debated details). The earlier Iron Age system, known as the Warren
Shaft (after Captain C. Warren who rediscovered it in the nineteenth
century), enabled free access to water through a system of underground
tunnels and a shaft. In the past, many scholars believed that it is through
this shaft that David conquered Jerusalem (II Samuel 5: 6–9). It appears,
however, that this system was constructed only in the eighth century
BCE. At this time, improvements were made in the existing system
of tunnels (that were cut in the Middle Bronze Age). The works
included lowering the bottom of an existing tunnel that led from within
the city wall, outward (henceforth, the upper tunnel), thus, exposing
a natural karstic shaft that was not known before. Once the shaft
was discovered, its significance was apparently observed, previous
works ceased, and it became the central part of a new system. A lower
tunnel, connecting the spring and the bottom of the shaft, was dug,
enabling the water to flow from the former to the latter. Thus, people
could walk in the upper tunnel, from within the city walls to the top of
the shaft, and draw water. The level of the water at the bottom of the
shaft was increased through the use of dams, making water drawing
easier. Notably, this system was not very convenient, and was probably
intended to be used only at times of siege. The Warren Shaft system,
however, was not in use for long, and it went out of use in the late
eighth century BCE, when Hezekiah’s tunnel was constructed.
Hezekiah’s tunnel involved the digging of an underground tunnel
that led from the spring to a pool at the other side of the City of David
ridge, within the city walls. Hence, all the water from the spring was
secured. The last stage of the enterprise involved the meeting of two
groups of hewers who worked from both sides of the tunnel
simultaneously; this is commemorated in the famous Siloam
Inscription3. Hezekiah’s tunnel is an impressive engineering enter-
prise. Even today, some of the accomplishments still raise questions as
to how they were achieved. Among the intriguing questions are, for
example, how the workers had enough air during the work, how were
the two groups able to meet, and how were they able to maintain such a
shallow sloping angle.
References
Aharoni, Y. The Archaeology of the Land of Israel.
Jerusalem: Shiqmona, 1978.
---. The Land of the Bible: Historical Geography. Philadel-
phia: Westminster Press, 1979.
Amiran, R. Early Arad I. Jerusalem: IES, 1978.
Amit, D., J. Patrich, and Y. Hirschfeld. The Aqueducts of
Israel. Portsmouth: David Brown Book Company, 2002.
Bagg, A. M. Dealing with Water Rights in the Ancient Near
East. Cura Aquarum in Israel. Ed. C. Ohlig, Y. Peleg, T.
Tsuk. Siegburg: DWhG, 2002. 223–32.
Barkai, G. The Iron Age II–III. The Archaeology of Ancient
Israel. Ed. A. Ben-Tor. New Haven: Yale University Press,
1992. 302–73.
3
The inscription was discovered in 1880 by a boy who was
bathing in the waters of the Gihon Spring, and was studied by
Conrad Schick, one of the first explorers of Jerusalem.
Engraved in the rock, the inscription describes the meeting of W
the two groups of hewers who had begun digging from
opposite ends of the tunnel. “The tunneling was completed…
While the hewers wielded the ax, each man toward his
fellow… there was heard a man’s voice calling to his fellow…
the hewers hacked each toward the other, ax against ax, and
the water flowed from the spring to the pool, a distance of
1,200 cubits…” The inscription is now in the Istanbul
Museum.
2234 Water systems in Bronze and Iron Age Israel
Baumgarten, Y. Y. The Use of the Water-Pulley in Iron Age II
Water Systems. Cura Aquarum in Israel. Ed. C. Ohlig,
Y. Peleg, T. Tsuk. Siegburg: DWhG, 2002. 233–8.
Ben Tor, A. The Archaeology of Ancient Israel. New Haven:
Yale University Press, 1992.
Bunimovitz, S. and Z. Lederman. Beth Shemesh: Culture
Conflict on Judah’s Border. Biblical Archaeology Review
23/1 (1997): 42–9.
Cole, D. How Water Tunnel Worked. Biblical Archaeology
Review 6\2 (1980): 8–29.
Eph’al, I. Siege and its Ancient Near Eastern Manifestation.
Jerusalem: Magnes Press, 1996 (Hebrew).
Faust, A. A Note on Hezekiah’s Tunnel and the Siloam
Inscription. Journal for the Study of the Old Testament 90
(2000): 3–11.
---. Warren Shaft, Yes It Was Used to Draw Water. Biblical
Archaeology Review 29/5 (2003a): 70–6.
---. The Warren Shaft System in Light of the New
Discoveries. New Studies on Jerusalem: The Ninth Volume.
Ed. E. Baruch, U. Leibner, A. Faust. Ramat Gan: Bar-Ilan
University, 2003b. 7–19. (Hebrew), 26* (English Abstract).
---. The Israelite Society in the Period of the Monarchy: An
Archaeological Perspective. Jerusalem: Yad Ben Zvi,
2005a (Hebrew).
---. The Israelite Village: Cultural Conservatism and
Technological Innovation. Tel Aviv 32 (2005b): 204–19.
Frumkin, A. The Hydrogeology of Israel and the Problem of
Water Supply in Antiquity. The Aqueducts of Israel.
Journal of Roman Archaeology Supplementary Series
number Forty-Six. Ed. D. Amit, J. Patrich, and
Y. Hirschfeld. Portsmouth, The Society for the promotion
of Roman studies, 2002. 21–4.
Frumkin, A., A. Shimron, and J. Rosenbaum. Radiometric
Dating of the Siloam Tunnel, Jerusalem. Nature 425
(2003): 169–71.
Gibson, S. Agricultural Terraces and Settlement Expansion in
the Highlands of Early Iron Age Palestine: Is There Any
Correlation Between the Two. Studies in the Archaeology
of the Iron Age in Israel and Jordan. Ed. A. Mazar.
Sheffield: Sheffield Academic, 2001. 113–46.
Gill, D. Subterranean Waterworks of Biblical Jerusalem:
Adaptation of a Karst System. Science 254 (1991):
1467–71.
---. How They Met? Biblical Archaeology Review 20\4
(1994): 21–33.
Gvirtzman, H. Israel Water Resources: Chapters in Hydrolo-
gy and Environmental Sciences. Jeruslalem: Yad Ben Zvi,
2002 (Hebrew).
Haiman, M. Water Sources and the Iron Age II Settlement
Pattern in the Negev Desert. Cura Aquarum in Israel.
Ed. C. Ohlig, Y. Peleg, T. Tsuk. Siegburg: DWhG, 2002.
23–31.
Helms, S. W. Jawa: Lost City of the Black Desert. Ithaca,
New York: Cornell University Press, 1981.
Herr, L. G. The Iron Age II Period: Emerging Nations.
Biblical Archaeologist 60/3 (1997): 114–83.
Herzog, Z. The Archaeology of the City: Urban Planning in
Ancient Israel and its Social Implication. Tel-Aviv: Tel
Aviv University, 1997.
---. Water Supply at Tel Beersheba in the 1st Millennium
BCE. Cura Aquarum in Israel. Ed. C. Ohlig, Y. Peleg,
T. Tsuk. Siegburg: DWhG, 2002. 15–22.
Kempinski, A. and R. Reich eds. The Architecture of Ancient
Israel, Jerusalem: Israel Exploration Society, 1992.
Kern, P. B. Ancient Siege Warfare. Bloomington: Indiana
University Press, 1999.
King, P. and L. W. Stager. Life in Biblical Israel. Louisville:
Westminster John Knox Press, 2001.
Lamon, R. S. The Megiddo Water System. Chicago: Chicago
University, 1935.
Mazar, A. The Archaeology of the Land of the Bible, 10,000–
586 BCE. New York: Doubleday, 1990. 478–85.
---. Jerusalem’s Water Supply in the First Temple Period. The
History of Jerusalem: The Biblical Period. Ed. S. Ahituv,
A. Mazar. Jerusalem: Yad Ben Zvi, 2000. 195–232
(Hebrew).
Meiron, E. A New Look at Jerusalem’s Water System
During the Middle Bronze 2 Period. Cura Aquarum in
Israel. Ed. C. Ohlig, Y. Peleg, T. Tsuk. Siegburg: DWhG,
2002. 7–13.
Ohlig, C., Y. Peleg, and T. Tsuk Cura Aquarum in Israel.
Siegburg: DWhG, 2002.
Olsen, J. P. Water Works. Anchor Bible Dictionary. Vol. 6.
New York: Doubleday, 1993. 883–93.
Porat, Y. Hydraulic Plaster in Aqueducts as a Chronological
Indicator. The Aqueducts of Israel. Ed. D. Amit, J. Patrich,
and Y. Hirschfeld. Portsmouth: David Brown Book
Company, 2002, 25–36.
Pritchard, J. B. The Water System of Gibeon. Philadelphia:
Pennsylvania University Press, 1961.
Reich, R. and E. Shukrun. Light at the End of the Tunnel.
Biblical Archaeology Review 25/1 (1999): 22–33, 72.
---. System of Rock-Cut Tunnels Near Gihon in Jerusalem
Reconsidered. Revue Biblique 107 (2000): 5–17.72.
---. Channel II in the City of David, Jerusalem: Some of Its
Technical Features and their Chronology. Cura Aquarum
in Israel. Ed. C. Ohlig, Y. Peleg, T. Tsuk. Siegburg:
DWhG, 2002a. 1–6.
---. Reconsidering the Karstic Theory as an Explanation for
the Cutting of Hezekiah’s Tunnel in Jerusalem. Bulletin of
the American Schools of Oriental Research 325 (2002b):
75–80.
Rosen, B. and Z. Greenberg. Water Sanitation in Pre-Modern
Jerusalem. Cura Aquarum in Israel. Ed. C. Ohlig, Y. Peleg,
T. Tsuk. Siegburg: DWhG, 2002. 285–93.
Seger, J. D. Water Tunnels. Oxford Encyclopedia of
Archaeology in the Near East. Ed. E. Meyers. Vol. 5.
New York: Oxford University Press, 1997. 331–3.
Shanks, H. I Climbed the Warrens Shaft (But Joab Never
Did). Biblical Archaeology Review 25/6 (1999): 33–4.
Shemesh, A. O. Water Sources in the Bible Propertorship,
Drawing Rights and Allotment – Part 1. Beit Mikra 50/1
(2004): 55–73 (Hebrew), 93 (English abstract).
---. Water Sources in the Bible Propertorship, Drawing Rights
and Allotment – Part 2. Beit Mikra 50/2(2005): 99–108
(Hebrew), 200 (English abstract).
Shiloh, Y. Jerusalem Water Supply Under Siege: The
Rediscovery of Warrens Shaft. Biblical Archaeology
Review 7\4 (1981): 24–39.
----. Underground Water Systems in the Land of Israel in
the Iron Age. The Architecture of Ancient Israel. Ed.
A. Kempinski and R. Reich. Jerusalem: Israel Exploration
Society, 1992. 275–93.
Shimron, A. E., A. Frumkin, J. Rosenbaum, and Y. Porath.
The City of David Waterworks: A Geological and
Engineering Overview. New Studies on Jerusalem:
Proceedings of the Fourth Conference. Ed. E. Baruch.
Ramat Gan: Bar Ilan University, 1998. III–XVI.

Stern, E. ed. The New Encyclopedia of Archaeological
Excavations in the Holy Land. Jerusalem: Israel Explora-
tion Society, 1993.
Tsuk, T. Pools. Oxford Encyclopedia of Archaeology in the
Near East. Ed. E. Meyers. Vol. 4. New York: Oxford
University Press, 1997a. 350–51.
Tsuk, T. Cisterns. Oxford Encyclopedia of Archaeology in the
Near East. Ed. E. Meyers. Vol. 2. New York: Oxford
University Press, 1997b. 12–3.
Vincent, L. H. Underground Jerusalem. London: Horace
Cox, 1911.
Wilkinson, J. Ancient Jerusalem: Its Water Supply and
Population. Palestine Exploration Quarterly 106 (1974):
33–51.
Wilson, W. and C. Warren. The Recovery of Jerusalem.
London: R. Bentley, 1872.
Zorn, J. R. Tell en-Nasbeh: A Re-evaluation of the
Architecture and Stratigraphy of the Early Bronze Age;
Iron Age and Later Periods. Ph.D. Dissertation, University
of California, Berkeley, 1993.
---. Estimating the Population Size of Ancient Settlements:
Methods, Problems, Solutions, and a Case Study. Bulletin
of the American Schools of Oriental Research 295 (1994):
31–48.
Weaving Cotton in Ethiopia and Nubia
M ICHAEL G ERVERS
Archaeological material finds provide invaluable
information about the nature and source of textiles
and textile fibres in Nubia in the early Christian era. In
the half century since Griffith and Crowfoot first
identified cotton among the textile fragments excavated
at Karanog and Meroe, much more cotton fabric has
been retrieved from the rescue excavations of early
habitation sites along the Nile subsequently inundated
by Lake Nasser; i.e., between Aswan in Upper Egypt
and Wadi Halfa in northern Sudan. The oldest
fragments identified as cotton have been dated to the
Roman period and may be as early as the first century
AD. The great majority are attributable to the late
Meroitic period, ca. 200–ca. 330 AD, at which time
cotton appears to have been the most commonly
available fabric. The following period, from ca. 330 to
ca. 550 AD, sees a swift and remarkable change in the
predominant textile fibre in the region. The use of
cotton plummets and is replaced by animal fibre,
largely camel’s wool.
The quantity of cotton found in excavations of the
Roman and late Meroitic periods has led many to
conclude that the fibre reached Lower Nubia from the
south, where it is thought to have been grown, since
this area lies in the rain belt and could have provided
the environment necessary for growth. However, no
specific sites for cotton cultivation in Nubia and the
Weaving cotton in Ethiopia and Nubia 2235
Sudan at this period have been identified and in view
of the chronology of the archaeological evidence,
and of what is known of the political and economic
situation and agricultural production of the period,
there is good reason to suppose that vegetable fibres for
textiles were not then grown in Nubia or even in the
Sudan. Rather, the Nubians, lacking any such indige-
nous source of raw material, were obliged to import
whatever they required either as cloth or raw fibre.
These goods included linen, which, throughout much
of the Middle Ages under the provisions of the baqt
agreement, was one of the principal items received
by the Nubians in exchange for slaves. The baqt was
first formulated in the seventh century, by which time,
in the absence of adequate supplies of cotton, the
Nubians seem to have sought imported linen as an
alternative. The cotton would have been gossypium
arboreum, and its ultimate source would have been
the west coast of India. How it reached Nubia is
uncertain. It could have come via the eastern desert trade
routes from such ports as Suakin on the Red Sea, or via
the kingdom of Axum, which by 300 AD. enjoyed a
thriving trade with India. The costs and difficulties of
overland transport, however, make it more likely that
cotton reached Nubia through the Roman world to the
north via the Nile. Since the archaeological cotton
remains from Nubia were not made from locally grown
fibres, there can be little question of cotton reaching
Ethiopia from the Sudan. One early reference which
is universally accepted speaks of the direct importation
to the Red Sea Axumite port of Adulis of cotton from
India in the first century AD. The first indication we
have of the production of fine cotton cloth in Abyssinia
occurs at the end of the thirteenth century in the
commentary of Polo: “Good cotton and buckram cloths
are woven here”. If his source for the supply is accurate,
one can reasonably conclude that by that time cotton
cultivation and weaving was well established in the
highlands.
Textiles, especially imported, have always been an
expensive commodity. Their distribution through the
ranks of a society depends directly on the geographical
diffusion of the raw material, the quantities produced
and the availability of mechanisms to spin and weave it
into cloth. An Egyptian source for cotton grown in
India would explain why a blue and white cotton
tapestry decorated with Egyptian symbols was among
the Meroitic finds from Qasr Ibrim. It would also
explain why the cotton, if imported raw from India and W
locally spun, would appear different in texture from
contemporaneous cotton cloth of Indian origin discov-
ered at Palmyra. Cotton cloth woven in the humid
atmosphere of India could be much finer than that
produced in the arid climate of Egypt or Upper Nubia.
The latter, if woven on the warp-weighted loom, as is
supposed, would have to be much thicker and coarser
2236 Wei Boyang
to prevent it from breaking under the stress of the
weights. Consequently, it would have appeared softer
than the material from Palmyra and more like wool.
When cotton became less available from the fourth
century and was replaced by wool, the warp-weighted
loom would have been all the more appropriate for
weaving it. Weaving in Ethiopia today is carried out
almost entirely on a pit-treadle loom, whose treadles
are suspended in a hole in the ground. The weaver sits
at the edge of the pit with his legs inside so that his feet
can operate the treadles. The origin of this loom in
Ethiopia is uncertain, although numerous writers have
noted how closely it parallels the Hindu loom, and
some suggest that the technology is the result of direct
importation from the Indian subcontinent. Endrei
believes this loom originated in India for the purpose
of cotton weaving. The specific characteristics of the
Abyssinian loom, according to Boser, are its sturdy
frame, the width of the comb and consequently of the
cloth produced, and the depth of the pit. Boser goes on
to say that regardless of how this loom got to
Abyssinia, it did not spread thence to other parts of
Africa. It did not spread, until recent times at least, even
to southern and western Ethiopia. A close relative is the
warp-weighted pit loom which is found frequently in
Persia, Syria and Egypt where it was, and is, used
particularly for weaving woollen, and in Egypt at any
rate for weaving linen thread. It does not seem to be
appropriate for weaving with a cotton warp as the strain
placed by the weights on the thread would cause it to
break. This restriction would explain the absence of the
warp-weighted loom in Ethiopia, where cotton is by far
the principal fibre used for weaving today. The
technical association of the Ethiopian loom with cotton
would also argue in favour of a common development,
if not introduction, of cotton growing and weaving to
the Abyssinian highlands.
References
Bergman, Ingrid, Hans-Åke Nordström, and Torgny Säve-
Söderbergh. Late Nubian Textiles. The Scandinavian Joint
Expedition to Sudanese Nubia. Vol. 8. Lund, 1975.
Boser-Sarivaxévanis, Renée. Les tissus de l’Afrique
Occidentale. Basle: Birkhäuser, 1972.
Casson, Lionel. The Periplus Maris Erythraei. Princeton,
New Jersey: Princeton University Press, 1989.
Endrei, Walter. Der Trittwebstuhl in frühmittelalterlichen
Europa. Acta Historica 8 (1961): 107–36.
Gervers, Michael. Cotton and Cotton Weaving in Meroitic
Nubia and Medieval Ethiopia. Textile History 21 (1990):
13–30.
Griffith, F. L. L. and G. M. Crowfoot. On the Early Use of
Cotton in the Nile Valley. The Journal of Egyptian
Archaeology 20 (1934): 5–12.
Lamm, Carl Johan. Cotton in Mediaeval Textiles of the Near
East. Paris: P. Geuthner, 1937.
Lombard, Maurice. Les textiles dans le monde musulman du
7e au 12e siècle. Paris: Mouton, 1978.
Maureen, Mazzaoui. The Italian Cotton Industry in the
Later Middle Ages, 1100–1600. Cambridge: Cambridge
University Press, 1981.
Nicholson, G. Edward. The Production, History, Uses and
Relationships of Cotton (Gossypium spp.) in Ethiopia.
Economic Botany 14 (1960): 3–36.
Pfister, R. L’introduction du coton en Égypte musulmane.
Revue des Arts Asiatiques 11 (1937): 167–72.
Polo, Marco. Trans. Ronald Latham. The Travels of Marco
Polo. Harmondsworth, Middlesex: Penguin Books, 1982
(1958).
Thurman, Christa C. Mayer and Bruce Williams. Ancient
Textiles from Nubia: Meroitic, X-Group and Christian
fabrics from Ballana and Qustul. Chicago: Art Institute of
Chicago, 1979.
Watson, Andrew M. Agricultural Innovation in the Early
IslamicWorld.Cambridge: CambridgeUniversityPress,1983.
Wei Boyang
FABRIZIO P REGADIO
Although many Chinese works relate the name of Wei
Boyang to the origins of alchemy, nothing is known
about him from a historical point of view, and his figure
may be entirely legendary. Some sources, which place
him in the second century, relate that he came from
Shangyu (in modern Zhejiang) and transmitted his
teaching to Xu Congshi, who in turn handed it down to
Chunyu (alternative spelling: Shunyu) Shutong. The
text deemed to embody the gist of this transmission
is the Zhouyi cantong qi (Agreement of the Three
According to the Book of Changes), a scripture in verses
that is attributed to Wei Boyang. The commentaries
usually interpret the “Three” as Heaven, Earth, and Man,
or as Daoism, cosmology, and alchemy. The Book of
Changes is the Zhouyi (or I Ching, also spelled Yijing),
the renowned divination manual, which also includes
a classical exposition of cosmology in its section com-
monly known in the West as “Great Treatise.”
The genesis of the Cantong qi is as obscure as its
putative author. Current research tends to consider that
an originally Han text on cosmology may have been
elaborated into an exposition of alchemical doctrines,
perhaps at different dates until the seventh century. The
very few references to a Cantong qi before that time
seem to refer to the original cosmological treatise,
whose circulation may conceivably have been restricted
due to its association with an unofficial and proscribed
body of writings propounding esoteric interpretations of
the Confucian classics.

Since the late Tang, however, the Cantong qi began
almost abruptly to exert a far greater influence in the
history of Chinese alchemy than that of any other
text. Hidden in the highly allusive language, and the
thick layers of symbols and images that characterize
this scripture, is an exposition of the doctrine that
inspired a variety of commentaries (about 30 of which
are extant) and other texts, written between the eighth
and the nineteenth centuries, in both Daoist and Neo-
Confucian traditions. One of the most well-known
exegeses was written by the Neo-Confucian philoso-
pher Zhu Xi, who commended the high literary quality
of the text.
The Cantong qi would hardly be intelligible without
recourse to the related literature. Although it was
occasionally read as a treatise on cosmology, most
commentaries agree that it articulates the theories found
in the tradition of the Book of Changes, applying them
to the alchemical work. The gua (trigrams and
hexagrams) of the Book of Changes are used both
to construct a cosmological model and, at the same
time, to represent phases of the process performed
by the alchemist. The various possible levels of
reading of the text possibly mean that it was used
within both the main traditions of alchemy in China,
i.e., waidan or “external alchemy” and neidan or
“internal alchemy.”
See also: ▶Alchemy in China, ▶Divination
References
Fukui Kōjun. A Study of Chou-i Ts’an-t’ung-ch’i. Acta
Asiatica 27 (1974): 19–32.
Ho Peng Yoke. The System of the Book of Changes and
Chinese Science. Japanese Studies in the History of
Science 11 (1972): 23–39.
Needham, Joseph, Ho Peng-Yoke, and Lu Gwei-Djen.
Science and Civilisation in China, Vol. 5: Chemistry and
Chemical Technology, Part 3: Spagyrical Discovery and
Invention: Historical Survey, from Cinnabar Elixirs to
Synthetic Insulin. Cambridge: Cambridge University Press,
1976. 50–75.
Partington, J. R. Ancient Chinese Treatise on Alchemy.
Nature 136 (1935): 287–8.
Pregadio, Fabrizio. Alchemical Doctrines and the Textual
History of the Zhouyi cantong qi. Cahiers d’Extrème-Asie
8 (1995).
The Representation of Time in the Zhouyi cantong qi.
Cahiers d’Extreme-Asie 8 (1995): 155–73.
Usaburō, Imai, ed. Shūeki sandōkei bunshō tsū shingi
kōhon (Collated text of the Zhouyi cantong qi fenzhang
tong zhenyi). Tōkyō Kyōiku Daigaku Bungakubu kiyō 57
(=Kokubungaku kanbungaku ronsō) 11 (1966): 1–89.
Lu-ch’iang and Wu Tenney L. Davis. An Ancient Chinese
Treatise on Alchemy Entitled Ts’an T’ung Ch’i. Isis 18
(1932): 210–89.
Weights and measures in Africa: Akan gold weights 2237
Weights and Measures in Africa: Akan
Gold Weights
G EORGES N IANGORAN -B OUAH
The subject of gold weights is complex and multidi-
mensional. It can be understood only when placed in
the context of its original cultural environment, which
was linked intimately to gold for its physical substance
and to the package (dja) in which it came for its
sociocultural identity.
The Akan country, on the Gulf of Guinea in West
Africa, was and is an area of gold deposits. This metal
is both feared and worshipped. Well before the first
contacts with Europeans in the fifteenth century, the
Akan people used gold dust as a medium of exchange.
However, the concept of “gold weights” came from
Western traders (Dutch, Portuguese, English, and
French).
The first question to ask is: are the figures really
weights? In describing these figures, the Baolé use
several different terms: Dja-yôbwê, sika-yôbwê, shin-
dra-yôbwê, nsangan-yôbwê, ngwa-yôbwê. Let us look
at the meaning of each of them.
. Dja-yôbwê (dja stone). Here the term refers to the
contents; it designates what is contained in the
package or dja. These original elements would be in
stone and would be concerned with different realms
of human knowledge deemed worthy of interest.
. Sika-yôbwê (stone of gold) money. In this second
case, the term describes a price or a monetary total.
These elements would deal with economics, calcu-
lation, and mathematics.
. Nsangan-yôbwê (stone of a fine). This term
describes the price of a fine, a tax, or a tribute, and
is concerned with elements of economics and
finance.
. Ahindr-yôbwê (proverb stone). This term evokes the
notion of speech, thought, and discourse, the fields
of literature and philosophy.
. Ngwa-yôbwê (game stone). This term refers to the
figures comprised of graphic signs. The exercise of
reading or deciphering the signs was considered a
kind of game, involving elements of imitation and
intellectual training.
As we see, the term “weights” appears nowhere in W
these descriptions.
The dja or sanaa was originally a package made of
animal skin or thick cloth in which the Akan placed
their figurines and certain accessories. It seems to have
been a kind of encyclopedia written in miniature
figures. The package and its contents symbolized the
2238 Weights and measures in Africa: Akan gold weights

economic power of the living king and the spirits of the artisans used the lost wax method to produce the
dead sovereigns. The act of taking possession of the dja weights.
signified for a new ruler that he was assuming the The Akan used a system of computing weight
power to raise taxes, impose fines, and take measures to consisting of 11 units. It began at dama and ended with
increase the state treasury. For his subjects, knowing bèna. It was possible to multiply bèna by infinity and
that the ruler was in possession of the dja of the state the values went from single to double or were
meant that they judged him capable of administer-
ing the financial and economic heritage of the country.
Many different objects can be found in a dja, but we
will mention here only the essential, original ones.
The equipment is a group of accessories used to
manipulate the gold power. They consist of (1) balance
scales, used to determine the weight value of different
quantities of gold dust used as a medium of exchange;
(2) spoons to put the gold dust on the plates of the
scales; (3) boxes of many different kinds, their covers
often decorated with graphic designs. They are used to
hold gold dust already weighed or about to be weighed;
(4) winnowing baskets used to separate and rid the gold
powder of any impurities; and (5) sieves to separate the
different grains of gold dust.
The weights provide knowledge of the weight and
monetary value of the quantity of gold powder placed
in the plate of the scale. There are three kinds of
weights – figurative weights, weights with graphic
designs, and geometric weights (Figs. 1, 2 and 3). Gold
weights were (usually) made of an original alloy whose
composition was similar to that of bronze and brass.
However, there are also weights made of silver, copper,
and solid gold. Weights and Measures in Africa: Akan Gold Weights.
The weights were made by the Tounfouê, an artisans’ Fig. 2 Weights with Graphic Designs. Photographs by the
group, different from blacksmiths and jewelers. These author. Used with his permission.

Weights and Measures in Africa: Akan Gold Weights. Weights and Measures in Africa: Akan Gold Weights.
Fig. 1 Weights with Figurative Elements. Fig. 3 Weights with Geometric Elements.
 Weights and measures: Animal-shaped weights of Burma
multiplied by two. There were three series of weights –
small, medium, and large. They could be added and
multiplied.
The small weights series consisted of ten monetary
units and was used for all sorts of small transactions:
ba = unit = 0.148 g
ba (gnon) = ba × 2
ba (nsan) = ba × 3
ba (nan) = ba × 4
ba (nou) = ba × 5
ba (nzien) = ba × 6
ba (nzo) = ba × 7
ba (motchué) = ba × 8
ba (brou) = ba × 10
The medium weights series consisted of 7 units. The
computation is done from simple to double, and each
unit has multiples and submultiples.
Assan = 4 m.v.
Gbangbandia = 4 m.v.
Tya = 5 m.v.
Anui = 5 m.v.
Gua = 5 m.v.
Anan = 5 m.v.
Tyasue = 5 m.v.
Total = 33 m.v.
These 7 units comprise 33 monetary values. The
smallest value is météba which equals 12 ba or 1.77 g.
The largest value is the ta, which equals 348 ba or 51 g
of gold.
In practice, the system worked as follows. For
example, the gua, the fifth unit, comprised the
following five monetary values:
Météba = 12 ba = 1.77 g of gold
Adjratchui = 24 ba = 3.55 g of gold
Tra = 48 ba = 7.54 g of gold
These are all sub-multiples of gua.
Gua = 96 ba = 14.20 g of gold (Unit of this series)
Guagnan = 192 ba = 28.40 g of gold (Multiple of
gua)
The large weight and monetary values series had
only 3 units. They were:
Banda = 384 ba = 56.80 g of gold
Banna = 432 ba = 67.44 g of gold
Pereguan = 478 ba = 71.92 g of gold.
In considering the weights and numeric representa-
tion, we will consider here only those weights with
graphic signs which correspond with calculation and
mathematics.
Concerning the signs and marks, the anthropologist
François H. Abel has written: “A. Amélékia, a well
known man named Diénélou confirmed for me that
the Ancients knew how to read from the weights…
In the village of Lomo-north, in the region of Toumodi,
the village chief knew that the signs on the weights had
meaning.”
Savary, the Director of the Department of Black
Africa in the Museum of Ethnography in Geneva has
2239
written: “Each weight is the product of two signs
written on it… Reading it is sometimes simple, but
often difficult. This is because some Black Africans
had a different concept for numeric figuration and for
the representation of the product of two numbers.
[Also] zero did not exist…”. In the system, figures and
numbers are represented by vertical and horizontal
lines, such as marks and arrows similar to those still
seen in charcoal in the houses of African villages.
Commenting in 1605 on the Akan system of
accounting and calculating, the Dutch explorer and
historian Pieter de Marées made this remark: “The
Negroes have weights of copper and tin which they
have cast themselves, and, although they do not divide
in the same way we do, it comes out the same, and the
accounting is always correct.”
For the people of the Akan civilization, the gold
weights contained in the dja constitute a fundamental
cultural text, comparable to the Christian Bible, the
Islamic Qu˒rān, or the Hindu Veda. It is within this
sacred package that they had consigned, in idiogram-
matic letters and signs, their knowledge and values to
be passed on to posterity. In this way, their descendants
would not have to reinvent that which had been known
by their ancestors and which constituted the foundation
of their civilization.
References
Abel, H. Les Poids à Peser l’Or en Côte d’Ivoire. Bulletin de
l’Institut Français d’Afrique Noire 16B (1952): 23–95.
de Marées, P. Description et Récit Historial du Riche
Royaume d’Or de Guinée. Amsterdam: C. Claeffon, 1605.
Garrard, T. F. Akan Weights and Gold Trade. London:
Longmans Group, 1980.
Gluck, J. Die Goldgewichte von Oberguinea. Heidelberg:
Carl Winter’s Universitat’s Buchhandlung, 1937.
Niangoran-Bouah, G. L’Univers Akan des Poids à Peser l’Or.
3 Vols. Abidian: NEA–MLB, 1984, 1985, and 1987.
Savary, C. Poids à Peser l’Or du Musée d’Ethnographie de
Genève. Bulletin Annual du Musée d’Ethnographie de
Genève, no. II, 1969.
Zeller, R. Die Gold Gewichte von Asante (Westaftica).
Leipzig: Baessler Archiv (Beiheft III), 1912.
Weights and Measures:
Animal-shaped Weights of Burma
W
D ONALD G EAR , J OAN G EAR
The royal animal-shaped weights of the Burmese
Empires appear to be unique not only as a weight
system, but also as one of the most important artifact
series, especially in a country with but few durable
artifacts. Their shape motifs are mythically leonine
2240 Weights and measures: Animal-shaped weights of Burma
(lion-like), elephantine, anserine (goose-like), and gal-
linaceous (poultry-like). They symbolize Buddhist and
pre-Buddhist beliefs, and the Burmese institution of
monarchy and its continuity, dynastic, and imperial
changes.
For the purpose of this account the Burmese empires
comprise those regions of the Southeast Asian
peninsula now known as Myanmar (Burma) and
northern Thailand (Siam).
Buddhist art in India gave rise to the Burmese choices
of animal motifs to use on the weights. The anserine
weight, in a tenth century Indian style, apparently was
chosen first during a period of Theravāda revival.
However, the motifs were ancient even then, originating
before the Burmese entered Burma. Shamanistic and
animistic influences, both of the post-1000 BCE steppes
and China, are also evident. Even more obvious is the
use of Chinese models and execution for most of the
style groups. Moreover, the decision to use Buddhistic
symbolic animals may also have been influenced by the
enduring association of Indian and Chinese Mahāyāna
Buddhists with commerce.
The mass units and scales reached Burma from India
before the twelfth century AD, but India itself obtained
them from Achaemenid Persia, which in turn, obtained
them from Assyria and Babylonia. However, with the
growth of Chinese trade in Southeast Asia from the
thirteenth century onward and the adoption in Burma of
Chinese mass standards, the Indian mass system was
displaced. (A mass unit is defined here as that which
was used in multiples to form weights mainly used for
heavy trade goods and in fractions mainly for bullion or
small amounts of other valuable items, the multiples
and fractions together forming the mass scale).
A Burmese term often given to the animal-shaped
weights is shway arlay (gold weights) which may have
been derived from the expression sri arlay (king’s
weights). However, in the East the word gold has
customarily been associated with royalty. Also, until
the late eighteenth century the weights of Burma were
mainly used on behalf of the king, who enjoyed
monopolies in trade and was the chief import/export
broker.
Of the three main weight shapes, the leonine and
anserine were used throughout the empires, while those
of elephant shapes were used only in north Siam. The
first two shapes were still in use in 1970 along the more
remote routes that pass from northeast Burma into
Yunnan, north Siam, north Laos, and the Thai cantons
of Vietnam.
Most commonly, the diameters and heights of the
weights do not exceed about 70 × 120 mm. In shape
they occur in three leonine- and seven bird-weight style
groups which can be placed in stylistic sequence. A
sequence in the elephant-shaped weight styles has not
been identified with certainty (Fig. 1).
Weights and Measures: Animal-shaped Weights of
Burma. Fig. 1 Weights of different shapes. Photograph by
the author.
The leonine shape, called to (taw = “royal” in
Burmese) is a stylized combination of the parts of four
animals. The lion of west and south Asia supplied the
model for the open-mouth, bearded head and probably
for the crouching torso. The horns were either bull-like
or were adapted from the antlered muntjac deer of
eastern and southern Asia. The tail of two of the styled
groups is usually lion-like, but it is occasionally horse-
like. With one exception, all the tails are characterized
by the artificially raised tail base of the ancient
Yunnanese horse. The legs and feet were modeled
upon those of the elephant. All these animal represen-
tations are present on the abacus of the Asokan pillar at
Sarnath.
The standing bird shapes of six of the bird style
groups are in the form of a stylized Chinese mandarin
drake (aix galericulata) having a knobbed crest and
known as a hintha (Pali). The model for the squatting/
brooding, gallinaceous style group has not been
identified.
The elephant shapes are naturalistic. In Siam, the
weights of elephant shape, known as chang in Thai, are
difficult to separate from those figurines of similar shape
intended for other purposes. The animal representations
stand on usually pyramid-shaped bases which may be
rectangular, octagonal, circular, or hexagonal.
Frequently, impressed signs 4–19 mm in size are
found, usually on the front or right side of the base. In
shape they may be script-like circular, square, bird- or
feline-like or in patterns normally of 4, 5, 6, 8, and 9
rays, diverging from a central point. These also can be
placed in sequence.
The weights are made of a metal alloy, the essential
components being copper (50–80%), lead (3–30%)
and tin (0–20%). In some of the nineteenth century
weights zinc may amount to as much as 35%. The
alloy, formally known as ganza (Tamil kamsa), was
imported either in the form of broken vessels from
Canton or as the coins known as cash (Sanskrit,
karsha) from Yunnan.
The kyat was the mass unit of the to and hintha
weights and varied with time from about 14 g in the
fifteenth century to over 16 g in the eighteenth century.
 Weights and measures: Animal-shaped weights of Burma
The 20 kyat (320 g) weights constitute about 50% of all
weights, while weights with a mass of about 250 kyats
(3,750 g) amount to about 2%. About one-half of the
elephant weights weigh less than 45 g and 2% have a
mass greater than 130 g. The average mass unit of the
upper part of the gallinaceous mass scale is about
11.2 g and that of the lower part is about 13.9 g. The
average mass unit of the elephant-shaped weights is
about 12.7 g (Fig. 2).
The essentially decimal mass scale of the to and
hintha weights, based on the mass unit of one kyat is
1/8 (or 1/10); 1/4 (or 1/5); 1/2; 1; 2; 5; 10; 20; 50; 100;
250. The elephant-shaped weights occur on the
following scale: 1/8; 1/4; 1/2; 1; 2; 5; 10; 20; 40; ?;
60. The gallinaceous scale, at about 37 g, separates into
an upper binary scale (1; 2; 4; 8; (?); 32 and what may
be a lower trinary scale, 1/9; 1/3; 1. Bengalese and
Chinese mass scales of the period are similar in part.
Ninety-nine percent of the to and hintha weights fall
within ± 12% of the mean mass of a particular style
group while about 51% of them fall within ± 2%. These
variations lie in the making of the weights not in ad-
ditions, subtractions, oxidation, nor wear. The accuracy
of the elephant weights (and figurines) is less.
During the nineteenth century the weights were
made at a village near the capital under the supervision
of the Chief Minister. These standard weights were
stored in the treasury and issued to officials in other
towns. It was a criminal offense to use weights other
than those made “at the palace”. Special sets of weights
were kept for the purposes of comparison and the
settling of disputes. Though only two nation-wide
“standardizations” have been recorded, the earliest in
the eleventh century, nevertheless it was the required
duty of each new monarch, upon ascending the throne,
to verify the weights.
Copies of animal-shaped weights, usually somewhat
crude, are still made today in eastern Burma (Shan
States) using techniques similar to those of thirteenth
century AD and second century BCE Burma and
Weights and Measures: Animal-shaped Weights of Burma. Fig. 2 Weights of different sizes. Photograph by the author.
2241
Yunnan. The weights were made by the lost-wax
technique, using clay to surround the wax model and
metal molds to shape the wax model. The high-melting
point wax used was produced in Yunnan by an insect.
Among the goods imported from India which may
have required weighing were opium, indigo, and
mercury. From Yunnan came gold leaf, silver ignots,
copper-alloy, cash, salt, white insect wax, cinnabar, and
tea. Burma exported gems, including pearls and coral,
costly medicines, musk camphor, gums, resins, waxes,
ivory, rhinoceros horn, and rare woods. Considerable
trade would have been done with other Southeast Asian
states, e.g., tin from Malaya and Laos, arsenic, lead,
and silver from the Shan States, zinc from Laos.
Seeds and a Chinese type of equal-arm balance were
used for the weighing of small highly valuable articles
and the steelyard or datchin for large masses. For
moderately valuable articles, e.g., foreign silver coins
and ingots, the animal-shaped weights would have
been used.
Until after the middle of the eighteenth century
ordinary village commerce was conducted by barter,
counting, purchase with cowries, and by measurement
of volume or length. Weights were rarely necessary.
However, from the last quarter of the eighteenth
century onward, it was the ordinary Burman, rather
than the king’s officers, who made use of the weights,
mainly to weigh currency ingots. This change was
caused by the Burmese wars of the time. These resulted
in the cessation of the supplies of copper alloy and
cowries, led to the Burmese capture of the large lead
and silver mines of the Shan States and so to the
copying of the Chinese practice of using chopped lead
as a low value currency. As a result of the increased
need for weights to weigh the abundance of lead and
silver ignots, there was a flood of copies from the
formerly Burmese north Siam region, now occupied by
south Siam (Ayuthya). This new domination also led to
the replacement or adjustment of the animal-shaped
weights to agree with those of south Siam.
2242 Weights and measures in China
Each of the weight shapes together with its base, and
each of the parts of the weights, was intended to convey
a particular meaning. The meanings had to be
understood by illiterate speakers of many different
languages. Though they were intended to be under-
stood primarily by animists and Buddhists, they would
also have been understood by Confucianists, Daoists,
and others. Characteristic of animist/Buddhist South-
east Asia (though originating much further away in
space and time) was the belief in the earthly, semidivine
king. This is what the to weight was intended to
symbolize. It combined the physical characteristics of a
potential Buddha (bodhisattva) and a universal mon-
arch (cakravartin). These, in turn conveyed the idea of
secular power both to animist and Buddhist, being
associated with earth, fertility, and healing, with
legitimacy of rule and righteousness of conquest (a
prerogative of a universal monarch). New styles of to
weights were issued only at the times of the “righteous
conquests” of a Burmese king, usually at the beginning
of a dynasty with the accompanying empire-building.
The hintha weight symbolizes, among other things,
the heavenly perfection and purity of the Buddhist
faith, especially that of the Theravāda belief character-
istic of Burma and Siam.
The association of feline and anserine representa-
tions for symbolic purposes has an origin more remote
than Buddhism. The lion–duck association is present
on some Asokan pillars. Lion-shaped weights were in
use in Assyria about 1500 BCE and duck-shaped
weights in Sumeria and Babylonia.
Concerning the elephant weights the kings of Burma
and Siam (Ayuthya) were the incarnations of Indra, the
chief of the gods, who was ancient before Hinduism,
and who rode on an elephant. Thus, one reason (among
several) for the choice of the elephant was to symbolize
these kings.
The most generally useful of the techniques for
establishing the chronological sequence of the weights
were the style and sign sequences; the relation of the to
weight styles, through their symbolism, to the dates of
the dynastic changes and territorial expansions; the
sequence of increase in the average unit mass of each
style group; the dates of the unit masses obtained from
the weighings recorded by many European traders from
AD 1515, and their relation to the average unit masses
of the style groups.
References
Annandale, N. Weighing Apparatus from the Southern Shan
States. Memoirs of the Asiatic Society of Bengal 5 (1917):
195–205.
Braun, Rolfe and Ilse Braun. Opium Weights. London: Braun,
1983.
Decourdemanche, J. Traite des Monnaies, Mesures et Poids
Anciens et Modernes de l’Inde et de la Chine. Paris: Institut
Ethnographique International de Paris, 1913.
Fraser-Lu, Sylvia. Burmese Opium Weights. Arts of Asia 1
(1982): 73–81.
Gear, Donald and Joan Gear. Earth to Heaven: The Royal
Animal-Shaped Weights of the Burmese Empires. London:
Twinstar, 1992.
Mollat, Helmut. Die Standardformen der Tiergewichte
Birmas. Baessler Archiv N. F. 32 (1984): 405–40.
Temple, Richard. Currency and Coinage Among the
Burmese. Indian Antiquary. 1897, 4 articles; 1898, 9
articles; 1919, 15 articles; 1927; 1 article; 1928, 5 articles.
Weights and Measures in China
H ANS U LRICH V OGEL
Metrology means the “science of weights and mea-
sures.” But when referring to premodern periods the
term “knowledge of weights and measures” is more
appropriate. In a more concrete sense, metrology is
the art of calculation with number, weight, and measure
units in the economic and fiscal domains as well as
in science. “Metrosophy” may be defined as “number
speculation within cosmological philosophemes,” but
from a more inclusive perspective the relationships
between magical, religious, and political thought on the
one hand and metrology on the other also have to be
taken into account. In both metrology and metrosophy,
numbers are of great importance. In metrology, they are
used for defining and counting measure and weight
units, while in metrosophy they serve as basic stuff for
the creation of systems of number symbolisms, magical
numbers, and correlative numerologies.
For the study of metrosophy and metrology, two
different types of sources are basically available. First,
we may mention sources of the tradition type, like the
metrosophical and metrological sections in chapters of
the dynastic histories. Although these sections also
contain information of a metrological nature, it is clear
that they are principally metrosophical in character. The
second important types of sources are concrete remains,
especially the relatively great number of real ancient
weights and measures which either have been excavated
at archaeological sites or which survived by having been
handed down from generation to generation. While the
sources of the tradition type are closely related with
metrosophical thought and number systems, concrete
remains inform us almost exclusively of metrological
aspects and realities. The tension between these two
clusters of information and data is of particular
importance and is rewarding to be investigated.

Chapter 21A in the Hanshu, written in the first
century AD, contains one of the first surviving texts
that gives evidence of a system integrating metrology
with such diverse fields as numerology, musical pitch,
astronomical phenomena, astronomy, cosmology,
historiography, and political, ethical, and moral
thought. The textual arrangement of this chapter makes
clear that metrology belonged to the traditional Chinese
science of mathematical harmonics (lü or lülü).
Chapter 21A of the Hanshu states that harmonics
consist of five categories: number, musical pitch, length
measures, capacity measures, and weights and balances.
All five had their origin in the huangzhong pitch-pipe,
which had been created and standardized by the
legendary Yellow Emperor. Obviously, the antiquity of
these conceptual and technical devices was an important
criterion for their purported eternal validity. Within the
five categories a certain hierarchy is expressed, ranking
numbers above, followed, in descending order, by pitch,
length, capacity, and weight measures. Numbers are
obviously of primary importance, because they provided
the basic counting units for the other four categories.
Metrology was grounded on a numerological system
combining the numbers of the Five Phases (Wuxing)
and other numerological concepts with the numbers
derived from the changes of yin and yang as they are
described in the Yijing (I Ching Book of Changes) and
its commentaries. Moreover, weights and measures were
conceived as an integrated system. With grains of
medium-sized black millet, the length and capacity of
the huangzhong pitch-pipe were measured and thus the
basic length and capacity measure units derived.
The weight of the grain filling of the pitch-pipe also
served as the basis for fixing the weight units (Fig. 1).
Weights and Measures in China. Fig. 1 Handling the
Wang Mang proto sliding callipers. From Liu Dong-rui, “The
Earliest Slide Ruler in the world – the Bronze Slide ruler of
the Xin Dynasty (AD 923). Bulletin of the Museum of
Chinese History 1 (1979): 96 (in Chinese).
Weights and measures in China 2243
Concrete remains of ancient Chinese weights and
measures are of great importance for supplementing the
information contained in the largely normative written
sources. A spectacular example are the proto sliding
callipers of the Wang Mang period (AD 9–23), of
which apparently three pairs survived.
The intaglio inscription on the movable part says:
“Manufactured at the first day of the first month of
the first year of the Shijian guo reign-period.” This
corresponds to March 15, AD 9, the first day of Wang
Mang’s interregnum. The front side of the fixed part
of the sliding callipers is divided into 5 in. (cun), that
of the movable part into 5 in. In addition, the 4 in. of the
fixed part are further subdivided into ten parts ( fen)
each. With the Wang Mang proto sliding callipers outer
diameters of less than 4 in. could be measured. The
most conspicuous difference between modern sliding
callipers and the Wang Mang piece is that the latter
does not have the differential scale of the vernier.
No less spectacular is the great number of other
Chinese metrological remains. For instance, more than
240 ancient length measures are known, which can
serve as a basis for ascertaining the “true” length of a
length measure unit of a given period. This is no easy
task as may be shown in the case of the length measure
unit chi of the Eastern Han period. Modern measuring
has shown that the actual lengths of 85 concrete chi
measures of this period vary from 20.49 cm to
24.73 cm. Obviously, state efforts at unification and
standardization of weights and measures were of
limited success and collided with diverging interests
having their origin in different regional, local, political,
economic, fiscal, and social conditions. Small wonder
that in the Suishu, the dynastic history of the Sui
dynasty written in the early seventh century, 15
different chi measures are listed.
The development of the types of weighing appara-
tuses can be divided into three stages. From the fifth
century BCE to the second century AD balances with
equal arms prevailed. In the period from the third to the
ninth centuries the steelyard with unequal arms appeared
and was used together with the ancient balances. Finally,
the third stage from the tenth to the early twentieth
centuries was characterized by the emergence of the
highly accurate deng steelyard and the introduction of a
decimal system of weighing units. An important
promoter in the introduction and improvement of the
deng steelyard was the eunuch Liu Chenggui (950–
1013). Liu was in charge of the court treasury and fiscal W
affairs. The purpose of the introduction of the deng
steelyard was to reduce disputes in the weighing of
precious metals, which arose between different officials
involved in fiscal affairs.
See also: ▶Acoustics, ▶Five Phases, ▶Yinyang
2244 Weights and measures in Egypt
References
Cheng, Panji. Exploration on Containers for Medicinal
Measurement. Chung-hua i shih tsa chih (Peking, China:
1980) 30.2 (2000): 109–13.
Ferguson, John C. Chinese Foot Measure. Monumenta Serica
6 (1941): 357–82.
Guo, Zhengzhong. The Deng Steelyards of the Song Dynasty
(960–1279): In Commemoration of the One Thousandth
Anniversary of their Manufacture by Liu Chenggui. Une
activité universelle: peser et mesurer à travers les âges.
Ed. Jean-Claude Hocquet. Caen: Editions du Lys, 1994.
297–306.
Iwata, Shigeo. The Changes in Linear Measures in China
and Japan. Acta Metrologiae Historicae: Travaux du IIIe
Congrès International de la Métrologie Historique. Linz,
7–9 Oct. 1983. Linz: Trauner Verlag, 1985. 117–37.
Jun, Wenren and James M. Hargett. The Measures Li and
Mou During the Song, Liao, and Jin Dynasties. Bulletin of
Sung-Yuan Studies 21 (1989): 8–30.
Loewe, Michael. The Measurement of Grain during the Han
Dynasty. Toung Bao 49 (1961): 64–95.
Qiu, Guangming. Zhonguo lidai duliangheng kao (Investiga-
tions into the Length, Capacity, and Weight Measures of
China Through the Ages). Beijing: Kexue chubanshe,
1992.
Renn, Jürgen and Matthias Schemmel. Waagen Und Wissen
in China: Bericht Einer Forschungsreise. Berlin: Max-
Planck-Institut f ür Wissenschaftsgeschichte, 2000.
Vogel, Hans Ulrich. Metrology and Metrosophy in Premo-
dern China: A Brief Outline of the State of the Field. Une
activité universelle: Peser et mesurer à travers les âges.
Ed. Jean-Claude Hocquet. Caen: Editions du Lys, 1994.
315–32.
Vogel, Hans Ulrich. Aspects of Metrosophy and Metrology
During the Han Period. Extrème-Orient, Extrème-Occident
16 (1994): 135–52.
Vogel, Hans Ulrich. Zur Frage der Genauigkeit antiker
Längenmaβe und deren interkulturelle Zusammenhänge
im Lichte chinesischer metrologischer Sachüberreste.
Rainer S. Elkar, u.a.: Vom rechten Maβ der Dinge: Beiträge
zur Wirtschafts- und Sozialgeschichte. Festschrift für
Harald Witthöft zum 65. Geburtstag. Scripta Mercaturae
Verlag, St. Katharinen, 1996.
Vogel, Hans Ulrich, Guo Zhengzhong, and Qiu Guangming.
Bibliography of Works on Historical Metrology Relating to
China (in preparation).
Weights and Measures in Egypt
R UTH H ENDRICKS W ILLARD
The gradual change from a nomadic to a settled
existence, which began about 8000 BCE, marked
humankind's determination to shape the environment
to their requirements. Needing units of measure for
both building and agriculture, people chose the most
readily available and useable references – simple parts
of the human body in various positions. The need for
uniform measurements was recognized, and as early as
3000 BCE official reference standards of length,
volume, and weight were being maintained in temples
and royal palaces in Egypt, Mesopotamia, and around
the eastern seaboard of the Mediterranean.
The standards developed in these ancient cultures
moved westward, mostly as a result of trade, to the
Greek and Roman empires, thence to Gaul and Britain
via the Roman conquest. Egypt, by virtue of its
geography a great trading nation and of its climate a
preserver of archaeological treasures, has provided an
incomparable record of this metrological heritage.
Measures of Length
Throughout the ancient Middle East, the basic unit of
linear measure was the cubit, which was defined as the
length of the forearm from the point of the elbow to
extended fingertips. In Mesopotamia, the cubit was
generally subdivided into 2 ft each of 3 or 4 palms, a
palm having 4 digits. In Egypt, however, the most
ubiquitous standard was the royal cubit. It was
subdivided into 7 palms (28 digits), these divisions
having significant mystical relationship to the four
7-day phases of the 28-day lunar month by which time
was reckoned. The basic sub-unit was the digit (finger
width): 4 digits = a palm; 5 digits = a hand; 12 digits
(outstretched forefinger tip to little finger tip) = a small
span (1/3 cubit); 14 digits (outstretched thumb tip to
little finger tip) = 1 large span (1/2 cubit). Measurements
smaller than a cubit were expressed in digits (up to 10),
in palms and digits, or rarely, as 1/2 or 1/3 cubit.
Surveys of the Great Pyramid confirm its construction
according to the royal cubit of 20.62 in (52.4 cm). The
skill of the Egyptian workmen is shown by the fact that
the sides of its base vary no more than 0.05% from the
mean length of 440 cubits (9069.45 in or 230,364 m).
The Egyptians also employed a short cubit of 6 palms
(24 digits) = 17.68 in (44.9 cm) for general purposes
including calibration of the Nilometers (masonry
stairwells leading down to the river at points where the
priests recorded the rise of the annual floods).
Egyptian measuring scales are square rods with one
face half-beveled to form a reading surface. The
fractions of the digit are shown by dividing the first
into halves, the second into thirds, and so on down to
sixteenths. The oldest extant cubit rods date from the
18th Dynasty, 1550–1307 BCE; none are longer than
the double cubit, 41.2 in. Most of those used by the
workmen are made of wood with deeply cut notches of
division. The stone cubit rods are generally ceremonial,
belonging to the temples and crowded with inscriptions
referring to the religious aspects of the digits as
connected with the gods and with the signs of the
nomes. On these, the calibrations are often carelessly
marked.
Land was measured by a linear unit called khet or
khet u nu (reel of cord) 100 royal cubits in length.

Measuring cords were made of palm-fibre or flax-fibre
and knotted at 5-cubit intervals. For long distances the
foot (12 digits) and the pace (double step) were used;
2 paces measured heel to toe = the extended arms
(4 cubits), and 100 paces = a stade. This distance later
used by the Greeks in the Olympic foot races known as
a stadium.
Measures of Area
The basic unit for reckoning areas of fields and nomes
[a name given to regional divisions in Egypt] was the
setat, the square of the khet, containing, therefore,
10,000 square cubits. Such an area was called a cubit of
land, but was visualized as 100 parallel strips of land
each 1 × 100 cubits. Ten setats were called a thousand
of land. In devising ways to re-establish field
boundaries after the annual flooding of the Nile, the
Egyptians discovered the rudiments of geometry. A
new unit formed from the diagonal of a square with
sides equal to the royal cubit was found to be divisible
by 40 digits of 0.73 in, twice the length of the remen
(forearm from elbow to clenched knuckles) and
consequently named the double remen. Half of the
area of 100 × 100 cubits was also called remen in land
measure. By having two standards, one the diagonal on
the square of the other, it was possible to denote areas
in squares equal to one-half or double the area of
others.
Measures of Time
Through studies based primarily on the movements of
the heavenly bodies in relation to the regular flooding
of the Nile the Egyptians established two calendars.
The priests followed an exact year of 365 1/4 days,
while the civil year was subdivided into 3 seasons
of four months each having 30 days – inundation,
cultivation, and harvest – plus 5 days-upon-the-year.
Measures of Volume
In establishing standard units of capacity for barter,
taxation, and medication convenient reference was
made to human factors such as a mouthful (tablespoon-
ful), handful, cupful (two handfuls) and sack- (hide-)
ful for easy carriage. The principal Egyptian standards
from small to large were the ro, hin, hekat, and khar.
Medications were prescribed in terms of the ro
(mouthful, tablespoonful) = 1/2 fl oz (14.5 cc). By the
usual Egyptian method of doubling this standard
produces 2 ro = 1 handful (fl oz, jigger); 4 ro = 2
handfuls ( jack, jackpot); 8 ro = 4 handfuls (gill); 16
ro = 2 hands cuppedful (cup); 32 ro = jugful (pint, hin);
64 ro = pitcherful (quart) and, by extension, the old
English measures including the gallon, peck, bushel,
barrel, hogshead, and tun.
Weights and measures in Egypt 2245
The hin for liquids was subdivided dimidially down
to 1/32 = 1 ro. There was also 1/3 hin known as the
khay, thirds being a peculiarity of Egyptian metrology
occurring also in the subdivision of the cubit and qedet
standards. There were two principal standards for larger
volumes. Multiples of the hin were decimal: 10 hin =
1 hekat (4.7 L). The hekat was the official corn
measure: hekat = 10 hin; 30 hekats = 1 cu cubit. The
khar, orginally a hide or sackful, was a special
standard = 20 hekats (94 L). Subdivisions of the hekat
in the dimidial series 1/2 to 1/64 (=5 ro), an amount
frequently prescribed for internal medications) were
always written in symbols now known as the Horus-eye
notation. The Rhind Mathematical Papyrus, attributed
to the 12th Dynasty (1991–1783 BCE), provides the
following conversion table: 9,600 ro = 300 hin = 30
hekat = 3/2 khar = 1 cubic cubit (37 U.S. gallons).
The early use of additional capacity standards is
documented by wall paintings in a Mastaba tomb of the
3rd Dynasty, ca. 2630 BCE. Two series of cylindrical
capacity measures, one of copper, for measuring wine
and oil, and one of hooped wood staves for grain, have
been shown to conform to a mixed series in binary
progression based on the standards of both the
Egyptian hin and the Syro-Phoenician kotyle (21.4 cu
in or 350 cc). Other standards of volume used in Egypt
include the Syro-Babylonian log (33.1 cu in or 542 cc),
the Attic kotyle (17.2 cu in or 282 cc), and the Persian
kapetis (74.9 cu in or 1227 cc).
Measures of Weight
A search for the origins of weighing reveals that this
intuitive use of the lever evolved independently in at
least three ancient civilizations – Mesopotamia, the
Indus Valley, and Egypt – and that all three were
routinely employing balance and weights by the middle
of the third millennium BCE. The weights, made in
durable stone, have survived in sufficient numbers to
permit study of the ancient weight standards. However,
very few balances, which were usually constructed of
wood in the earliest periods, exist today.
It is possible, nonetheless, to trace not only the
evolution of weighing instruments but also the uses to
which they were put in dynastic Egypt during the
period 2700–650 BCE. As a means of enhancing the
quality of the afterlife of their leaders, Egyptian artisans
carved or painted scenes from their earthly life in tombs
and temples. Because of a mild, dry climate, many of W
those monuments survived into modern times. While it
is still possible for visitors to experience the excitement
of discovering the weighing scenes in situ, not all of
the tombs and temples have survived the two centuries
of study and tourism to which they have been
subjected. Colors have faded; walls and in fact entire
structures have disintegrated or been destroyed or
closed. We turn, therefore to the remarkable work of the
2246 Weights and measures in Egypt

Egyptologists of the nineteenth and early twentieth The Evolution of the Equal Arm Balance
centuries, a period of intensive recording and copying All of the balances used in Egypt before the Roman
of Egyptian inscriptions, reliefs, and paintings. Hence, occupation in 30 BCE were equilateral, doubtless
it becomes necessary to consider not only possible inspired by the yokes with which men balanced heavy
errors in the original, but also the errors resulting from loads on their shoulders. During the Old Kingdom
the deterioration of the works being copied. (2686–2181 BCE) most balance beams were flat and
For information regarding the practice of weighing, rectangular, pierced vertically for support at the
we are indebted primarily to Sir W. M. Flinders Petrie. fulcrum and for the hangers, which were single cords
He was the first Egyptologist to identify, collect, and hooked at the end to hold baskets or bags of gold dust.
study the ancient weights for the insights they could With but one exception, shown in (Fig. 1), they were
provide into the cultural and trade relations among the made of perishable wood until about 1550 BCE.
various nations of the Near East. His numerous Hence, the only records of their design and use come
publications include chapters on the metrological from reliefs, paintings, and inscriptions.
findings of each site excavated. Most large museums Significant innovations during the Old Kingdom
have specimens of ancient Egyptian weights. include specialized tongs to hold metal vessels,
The practice of weighing began with the discovery counterweights of a predetermined mass, tongues,
of metalworking. The earliest weights are of a small and plummets to help in sighting the vertical. Nearly
order, suggesting that at first only small quantities all of the 5th and 6th Dynasties weighing scenes come
needed to be weighed. Egyptians developed two from the tombs of royal officials, and depict the
indigenous weight standards and as an international
trading center employed at least six more. The Egyptian
national standard was based on the qedet, which varied
over time from 138 to 152 gr giving a median value
of 144 gr (9.33 gm). It was multiplied decimally:
10 qedet = 1 deben; 10 deben = 1 sep, as first recorded
in a 3rd Dynasty tomb painting. The unit was
sometimes divided into thirds as evidenced by a series
of graves from the 1st Dynasty (2920–2770 BCE). The
qedet is the basis of nearly all statements of weight
from the 18th Dynasty onward. The qedet and deben
were also known as the kite and uten when applied to
pieces of copper used as a medium of exchange.
Less well known but more influential in world affairs
was the beqa, which was associated with the weighing
of gold and silver from pre-Dynastic times and used in
Egypt for nearly 4,000 years. Although the standard is
seldom mentioned in Egyptian inscriptions, numerous
beqa weights have survived. These are sometimes
marked with the hieroglyph for gold (nub) perhaps
recalling its origin in the Nubian gold fields. As in most
Middle Eastern standards, the basic unit was called a
shekel; 50 beqa shekels = 1 mina. The beqa shekel was
multiplied decimally to 7,000 shekels and subdivided
dimidially down to 1/16. Over time, the several distinct
values merged into a single standard of 192 gr (12.4 g),
which was adopted by the Greeks around 700 BCE as Weights and Measures in Egypt. Fig. 1 This is the earliest
the standard of Aegina. From it evolved two of the known Egyptian balance. It was skillfully cut from pinkish
Roman pounds, the silver denarius and the gold aureous brown limestone, a material commonly used for making
coinage. In the seventh century AD it was adopted by the vessels in prehistoric times, but seldom later. The three
Arabic Empire for bulk gold and it ultimately became suspension holes were drilled with a primitive bow drill and
flint point. The beam is 3.35 in long. Its width varies from 1.6
the basis for the English troy weight standard.
in to 2.0 in and its depth from 1.7 in to 2.0 in. The diameter of
Other standards in use in Egypt included the the central hole is .03 in and that of both end holes is .07 in.
Palestinian peyem, 120 gr (7.78 g); Mesopotamian As to sensitivity, a change of 1 in 500 can be seen by the
daric, 129 gr (8.36 g); Attic stater, 135 gr (3.75 g); change of level. A plaster model of this balance was tested at
Persian khoirine, 178 gr (11.53 g), and Phoenician sela, the Science Museum in London. Photo at the Petrie Museum,
218 gr (14.13 g). University College, London by O. W. Willard in 1993.
 Weights and measures in Egypt 2247

weighing of metal in the course of manufacturing. (1550–1070 BCE). While the equal arm balance of
There is no evidence of buying selling, or exchanging the Old and Middle Kingdoms had been modified in
goods by weight in this early period. ways that made using it more convenient, there was still
The decline of the Old Kingdom was followed by the problem of accuracy. The end suspension cords
the First Intermediate Period, an epoch of chaos, civil could wander in holes drilled vertically through the
war, hunger, violence, and invasion. No evidence of beam and those holes, even if exactly equidistant from
weighing has been found from this period, which the fulcrum initially, might not remain so after the
encompassed the 7th to 10th Dynasties and lasted from wearing away inherent in use. Either early in the New
2134 to 2040 BCE. Eventually Theban warlords gained Kingdom (or possibly late in the Second Intermediate)
supremacy and established the Middle Kingdom with a radically different scale beam made its appearance.
its capital first at Thebes and later at Itjtawy, a newly Constructed of either wood or bronze, it was round
built city near Memphis. in section. Suspension cords entered at the beam end
through longitudinal bores and emerged through
The Middle Kingdom intersecting radially drilled holes, being knotted on
top of the beam. Any inequality in the distance of the
Weighing scenes are found at Beni Hasan, the most
beam ends from the fulcrum could be adjusted by
noteworthy necropolis of the Middle Kingdom, where
simply filing the longer end, restoring the balance to
39 tombs are cut out of the rock cliffs at Beni Hasan, on
accuracy. This principle was later improved upon in the
the east bank of the Nile. The tomb paintings here
large standing balances with flared beam ends that
depict balances with an improved suspension system.
allowed the four strings to diverge from the lowest
The beam is suspended from a bracket near the top
point on the flange, making for even greater accuracy.
of the upright and can swing more freely because of a
While the great majority of representations from this
ring added to the top of the tongue. As in the fifth and
period depict large, stately standing balances, the extant
sixth Dynasties, the plummet appears to hang from
specimens are all small and portable. The beams, which
a point near the bottom of the tongue suggesting that
may be of bronze or wood, are invariably round in
the tongue and beam are rigidly attached and move
section. The Cairo Museum displays several. A hand-
together. The baskets are now suspended by multiple
held balance at the Science Museum in London was
cords, depicted as two but presumed to be four,
excavated at Amarna and dates from the reign of
since that is also the case in the New Kingdom, from
Akhenaten, 18th Dynasty, c. 1331 BCE. The length of
which we have extant pans with four holes. One
the round, tapering wooden beam is 30 cm and the
representation shows workers weighing two sets of
diameter of the copper pans is 7.4 cm; the sensitivity is
objects against one another. Another representation
plus or minus 2 gr per hundred in each pan. Bronze
shows the introduction of shallow, concave pans made
animal weights characteristic of the 18th Dynasty are
possible by the multiple suspension cords. Weight
shown with this balance. Two previously unpublished
boxes are shown below each standing balance.
specimens were found at the University of California's
A totally new aspect of weighing appears in one 12th
Phoebe Apperson Hearst Museum of Anthropology.
Dynasty relief. Here we see a jeweler who has left the
Both were excavated at Naga ed-Deir during the
workplace to show his finished product to the nomarch
1899–1904 Hearst Expedition led by G.A. Reisner.
(local ruler). Reading between the lines, we note that he
While it is impossible to date either one with certainty,
has suspended the balance from a full-length stand, but
both have the general characteristics of New Kingdom
by using an extension cord, he has brought the scale
balances (see Fig. 2).
and its contents down to a level he can reach while
Specimen 6-13624, from a cemetery dating from the
kneeling before his ruler.
late Old Kingdom through the Middle Kingdom, was
During the Second Intermediate Period, from 1640
found near the entrance of tomb 434 with the
to 1532 BCE, the political situation was unstable, with
excavator's notation ghadim (fill). The wood is acacia,
several dynasties ruling in various areas simultaneously.
While there was undoubtedly turmoil during the
takeover, Egyptian life was enriched by the introduc-
tion of horses, wheeled chariots, hump backed cattle,
and new tools, weapons and musical instruments. W
Innovations relevant to metrology included the working
of bronze and weights cast in the shapes of animals.

Weights and Measures in Egypt. Fig. 2 Two New

The New Kingdom 1550–1070 Kingdom-type balance beams. Photo by O. W. Willard by
Both the equipment and the philosophy of weighing permission of the Phoebe Apperson Hearst Museum of
underwent major changes during the New Kingdom Anthropology, University of California.
2248 Weights and measures in Egypt
sub-species not identified. The round, tapering beam is
50.3 cm long with a diameter of 2.7 cm at the center,
drilled in the New Kingdom manner for central and end
suspensions. Unlike the Amarna balance above, the
beam has a definite arch in the middle. There is the
imprint of a narrow saddle or support of some kind at
the fulcrum. The ends are knobbed, with a decorative
channel around each. On each side of the suspension
holes, there is a decorative double annulet with a
channel between. No pans were found.
Specimen 6-2702 is believed to have come from a
cemetery dating from early dynastic to Roman times.
The wood is acacia, sub-species not identified; the
round beam is 31.2 cm long and 2.0 cm in diameter at
the center. There is a bulge in the center of the beam
and the ends are slightly flared, foretelling the more
pronounced flanges of the large ceremonial balance.
There is also an unexplained diametric hole about
1.0 cm from the fulcrum hole. No pans were found.
The most magnificent weighing scenes are those
representing the brilliant 18th Dynasty, which ruled
Egypt from 1550–1307 BCE. The ceremonial bal-
ances, sometimes 10 ft tall, are often adorned with the
head of a god as a finial. The beam hangs from a
bracket that may be in the shape of the feather
representing Maat, the goddess of truth. The bronze
tongue, which has evolved into a pointer, moves with
the beam. In many examples, the operator holds the
plummet in his hand as if to inspect it for flaws before
beginning the weighing procedure. Shallow concave
pans, usually of metal, are suspended by multiple cords
thought to be four in number although sometimes
depicted as two or three. The conventional stone
weights, geometric in form, have been replaced with
handsome animal shapes in cast bronze.
Ambiguities in the drawings coupled with the lack of
physical specimens have fostered intense interest in the
central suspension and indicator system. The explana-
tion most generally accepted was offered by Petrie in
1888:
The beam was suspended by a loop or ring from a
bracket projecting from the stand; this bracket is
shown in side view though at right angles to the
beam just as the Egyptians drew a full eye in a side
face. Then below the beam, a long tongue was
attached, not above the beam, as with us. To test
the level of the beam, a plummet hung down from
the tongue, and it was this plummet which was
observed to see if the tongue was vertical and the
beam horizontal. The weigher is often shown
steadying this plummet with his hand, as it would
be set swinging by the motions of the beam. Such
is the whole system, which is so simple that it
seems strange that any mistake could be made
about it to say nothing about the mechanical
absurdity of the explanation which has been
current for so long.
The “mechanical absurdity” theory to which Petrie
referred had been offered by Wilkinson in 1837.
Referring specifically to an illustration showing the
balance being used by the Qabbaneh, or Public
Weighers, he wrote,
The beam passed through a ring suspended from a
horizontal rod immediately above and parallel to
it, and when equally balanced, the ring, which was
large enough to allow the beam to play freely,
showed when the scales were equally poised. It
had the additional effect of preventing the beam
tilting when the goods were taken out of one, and
the weights allowed to remain in the other. To the
lower part of the ring a small plummet was fixed
and this being touched by hand and found to hang
freely indicated, without the necessity of looking
at the beam, that the weight as just.
Though acclaimed for his understanding of the
Egyptian artistic canon and for the accuracy of his
copies, Wilkinson overlooked the obvious fact that the
vertical plane of the bracket, shown in profile, would
have been perpendicular to that of the beam. The
single-ring hanging mechanism as shown could not
possibly have worked. It would appear that either the
original drawing was erroneous, or deterioration in the
ancient surface caused Wilkinson to miscopy it.
Recent research has centered in the nature and
function of the triangular shape bisected by the plumb
line that project below the beam. Spiegler postulates an
arrangement of three strings fastened at the beam and
equipped at the tip with a plummet. As the beam is
tipped, a slack string allows the plummet to swing.
Jenemann and Robens note that such a mechanism
would work only with large deflections that could be
seen more easily by visual observation. Furthermore, in
colored drawings the triangle is clearly opaque,
indicting a solid object. They then suggest that the flat
side of the triangular plane, while shown frontally in
the drawings, is actually perpendicular to the beam,
while the plummet hangs independently in front of the
tongue, an arrangement considered to be more sensitive
than one in which the tongue pierces the beam
longitudinally.
Skinner states, “For larger wooden or bronze beams
a bronze tongue was driven vertically through the
center of the beam with a ring on top for suspension
from a bracket. The tongue protruding below the beam
acted as a swinging index, which was sighted against a
freely-suspended plum-bob to denote when the beam
was horizontal and in balance.” In the absence of New
Kingdom specimens, perhaps conclusions can be
drawn from a small bronze goldsmith's balance dating

to the Greco-Egyptian period, 304–30 BCE. The beam
is suspended from a ring in the bracket intersecting a
ring in the top of the tongue or pointer, which pierces
the beam and extends below it. The plummet hangs
from the bracket, behind the tongue. The original is in
the Cairo Museum and its replica, a plaster cast, is in
the Science Museum, London.
Expanding Roles for the Balance: Trade,
Tribute, Offerings
With Egypt's transformation to an international power,
new roles evolved for the balance. Along with the
customary depictions of jewelers and other metal
workers there are scenes of the balance used in weighing
tribute to the pharaoh and offerings to the national god
Amun and, for the first time, in domestic and foreign
trade. Short gold or silver cylinders about five inches in
diameter served as a sort of currency. Variable in mass,
these must have been weighed in each transaction.
Medicine however, continued to be measured rather
than weighed.
The earliest known depiction of the balance in com-
merce is found in a remarkable painting from the tomb
of Kenamun, Mayor of Thebes and Superintendent of
the Granaries of Amun under Amenophis III, 1391–
1353 BCE. Syrian ships were anchored on the Nile off
Thebes, while on shore two male shopkeepers engage
in private trading with the sailors. Since the commod-
ities – sandals, apparel, and various objects – are too
large to weigh on the hand balances shown, the sellers
are obviously weighing out the amount of gold needed
to make the purchases.
Among the various public officials were the
Qabbaneh, or public weighers, who erected their
balances in the market place while a notary stood by
to record the details. They were required by law to adjust
the sale of each commodity with the strictest regard to
justice, without favoring either the buyer or the seller.
The penalty for falsification of weights or records was
the loss of both hands, inflicted upon the seller, the
weigher, and the notary alike. The Qabbaneh were still
functioning in early twentieth century Thebes.
Hatshepsut, who ruled as Egypt's only female
Pharaoh (1473–1458 BCE) built at Deir el Bahari a
magnificent temple with extensive terraced gardens
facing the Nile. She dispatched an expedition to the
fabled land of Punt in search of treasures to be offered
to the sun god Amun, to whom the temple was
dedicated, and recorded the presentation in a monu-
mental relief on the south face of the second terrace. In
the weighing scene, a massive balance is piled high
with cylindrical gold ingots being weighed against
some loaf-shaped weights of the Middle Kingdom and
some bronze animal-shaped weights characteristic of
the 18th Dynasty. In the background are trays holding
Weights and measures in Egypt 2249
additional loaf shaped weights and quantities of
electrum (a naturally-occurring alloy of gold and
silver) in bars and rings. The inscriptions read,
The balances accurate and true of Thoth, which
the King of Upper and Lower Egypt (Hatshepsut)
made for her father Amun, lord of Thebes, in order
to weigh the silver, gold, lapis lazuli, malachite,
and every splendid costly stone, for the sake of the
life, prosperity, and health of her majesty.
Weighing the gold and electrum, the impost of
the southern countries, for Amun-Re, lord of
Thebes. Recording in writing, reckoning the
numbers, summing up in millions, hundred of
thousands, tens of thousands, thousands, and
hundreds.
Thutmose III, successor to Hathepsut, so extended
the Egyptian holdings in Asia and northern Ethiopia
that he has been referred to as the Napoleon of Egypt.
During this period, the coffers of the state were greatly
enriched by both tribute from and trade with the nations
they had subdued. A weighing scene in the tomb of his
Vizier, Rekhmire, is inscribed in part, “This great heap
of electron, which is measured by the heket, making
36,692 deben.
The Papyrus Harris, now in the British Museum,
represents the testimony of Rameses III, Pharaoh
(1194–1163 BCE) as he nears the end of his life and
affords an understanding of both the physical nature
and the significant role of the ceremonial balance in the
New Kingdom. Describing his stewardship as the god's
representative, he states as part of the prayer and recital
of the king's benefactions,
I made for thee splendid balances of electrum, the
like which had not been made since the time of the
god. Thoth sat upon it as guardian of the balances,
being a great and august ape of gold in beaten
work. Thou weighest before thee, O my father Re,
when thou measurest of gold and silver by the
hundred thousands, brought as tribute before thee
from their coffers, and given to thy august treasury
in the house of Atum. I founded for it (namely the
balance; evidently offerings were made to it) daily
divine offerings in order to supply its altar at early
morning.
In another section, Rameses describes the king's
gifts to Re including (in part), “Fine mountain gold and
gold for the balances 1,278 deben, 9 2/3 kedet1. Crude W
silver for the balances and silver in vessels 1,891
deben, 1/2 kedet. Black copper for the balances 67
deben, three kedet.
1
The gold amounts to some 311 3/4 pounds troy, and the
silver to nearly 461 pounds troy.
2250 Weights and measures in Egypt

Weights and Measures in Egypt. Fig. 3 The Judgment Scene: Ani's heart being weighed in the balance (the first three of six
panels). Left to right, top: The company of the gods, who hear Ani's testimony. Lower register, left to right: Ani and his wife
Thuthu entering the Hall of Judgment; Rehenet and Meskhnet, the Goddesses of birth; (small figures. L 2 r) Ani's soul, Ani's
Embryo, Ani's luck or destiny; Anubis testing the tongue of the balance; Thoth recording the result of the weighing, The
Devourer of the Unjustified; Not shown here: Horus, the falcon-headed god, introducing Ani into the presence of Osiris; Ani
kneeling before Osiris, Osiris enthroned within a shrine. Behind him are Isis and Nepthys, and before him upon a lotus stand,
the children of Horus (Budge 1969). The original Papyrus of Ani is in (the British Museum, No. 10470 Sheet 3 and 4).

Weighing the Soul
Souls too were weighed. The Egyptian concept of right
conduct as a prerequisite for admission to paradise
grew out of the cult of the god Osiris, which with its
promise of immortality began to flourish during the
First Intermediate Period. But it was first represented
pictorially in the New Kingdom, when an elaborately
illustrated roll of papyrus known as the Book of the
Dead (or, more properly, The Book of Coming Out by
Day) was placed in the tomb. Along with spells and
instructions for entering into the kingdom of the gods,
each Book of the Dead included a weighing scene in
which the heart of the deceased, often represented by
the hieroglyph for heart, was weighed against the
feather of truth, symbol of the goddess Maat. The
deceased was required to make a “negative confession”
denying the commission of 42 specific sins such as
giving short measure or falsifying the scales of the
balance. The jackal-headed god Anubis tested the
plummet of the scales, while the ibis-headed god Thoth
prepared to record the results and a beast called the
Devourer of Souls stood by to perform his duty, should
the applicant's heart be found wanting (see Fig. 3).
This allegorical concept, first presented graphically
by the Egyptians, persisted in the Christian world
through the Middle Ages as demonstrated by surviving
legends, stained glass windows, and paintings in which
Anubis is replaced by St. Michael and the feather
replaced by a toad. In 1901, Dr. Duncan McDougal of
Haverhill, Massachusetts conducted the first of several
experiments to determine whether the soul has weight.
He found that in addition to the gradual weight loss
related to the loss of body fluids as death approached, at
the exact moment of death there were sudden weight
losses of 3/8 oz, 1/2 oz, and 3/4 oz. He attributed this
weight loss to the departure of the soul. In his book,
Weighing the Soul, Fisher relates other scientific
research dating from Leonardo da Vinci in 1515 to
American and European physicists in modern times. As
recently as 1972, according to Agence France Press, a
Swedish doctor set out to determine the mass of the
human soul by placing the deathbeds of seriously ill
patients on extremely sensitive scales. As each died and
the soul left the body, the needle dropped 21 g, he
reported.
The reliefs and paintings with which the Egyptians
adorned tombs and temples celebrate the enduring
values of their culture over a period of more than 3,000
years. The 18th Dynasty weighing scenes with their
monumental balances and graceful animal-shaped
weights are generally considered the apex of the genre.
After this period of artistic and religious ferment Egypt
returned to the old familiar ways, recasting their
weights in geometric shapes. And as time went on,
the illustrations grew more and more fanciful, as if the
artists preparing them were not familiar with the actual
instruments. While these enigmatic drawings, at once
lively and serene, do not reveal all the technicalities of
the Egyptian balance, they do provide a window into an
ancient culture whose achievements in many fields are
still reflected in modern life.
Whether constructed of wood or of bronze, the
improved Egyptian balance became the standard
weighing instrument of the Near East until Roman
times. Glassmaking, another technical innovation of
the 18th Dynasty, was applied to metrology with the
production of a few weights during the Roman and
Byzantine periods. The Arabs, after their conquest of
the country in AD 640, continued the tradition, issuing
a fine series of reference weights for their coinage.

These are circular discs of transparent glass in various
colors molded with a thick rim surrounding a central
depression and civil inscriptions in Arabic. The Arabic
standard of metrology continued to be dominant
through the nineteenth century with standards varying
city by city and by commodity. The metric system was
made permissible in Egypt in 1873.
See also: ▶Calendars in Egypt, ▶Pyramids
References
Acta Metrilogie IV; Vle Congres International de Metrologie
Historique; Cahiers de Metrologie. Jean Claude Hocquet,
editeur; Tomes 11–12, 1993–1994. Editions Diffusion Du
Lys BP 1653 104020 Caen Cedex.
Bell, Barbara. The Oldest Records of the Nile Floods.
Geographical Journal 136 (1970): 569–73.
Berriman, A. E. Historical Metrology. New York: E. P.
Dutton, 1953. New York: Greenwood Press, 1969.
Breasted, J. A History of Egypt. New York: Scribner, 1905.
Bi, P. H. Weights and Measures. Chambers Encyclopedia,
Oxford: Oxford University Press, 1967.
Budge, E. Wallis. The Gods of the Egyptians. New York:
Dover Publications, 1969.
Davies, N. D. G. and R. Faulkner. The Rock Tombs of Der el
Gabrawi 1,2. London: Egypt Exploration Fund, 1902.
Davies, N. and R. Faulkner. A Syrian Trading Venture to
Egypt. Journal of Egyptian Archaeology 33 (1947): 40–6.
Ducros, H. Étude sur les Balances Égyptienne. Annales du
Service des Antiquités de l'Égypte 7 (1908): 32–53.
---. Deuxième Étude sur les Balances Égyptienne 10 (1910):
240–53.
---. Troisième Étude sur les Balances Égyptienne 11 (1911):
251–56.
Edwards, I. E. S. The Pyramids of Egypt. Hammondsworth,
Middlesex: Penguin Books, 1947. (Reprinted 1977).
Fisher. Len. Weighing the Soul. New York: Arcade
Publishing, 2004.
Glanville, S. R. K. Weights and Balances in Ancient Egypt.
Proceedings of the Royal Institution of Great Britain 29
(1937): 10–40.
Griffith, F. L. Notes on Egyptian Weights and Measures.
Proceedings of the Society of Biblical Archaeology 14
(1892): 403–40; 15 (1893): 301–15.
Helck, Wolfgang H. Masse und Gewichte, Lexikon der
Ägyptologie. 6 vols. Wiesbaden: Harrassowitz, 1972–1984.
Hoare, F. R. Eight Decisive Books of Antiquity. New York:
Dorset Press, 2004.
Jenneman H. R. and E. Robens. Indicator System and
Suspension of the Old Egyptian Scales. Thermochimica
Acta, 152 (1989): 249.
Leake, Chauncey D. The Old Egyptian Medical Papyri.
Lawrence: University of Kansas Press, 1952.
Lucas, A. Ancient Egyptian Materials and Industries. 4th ed.
London: Edward Arnold, 1926. (Reprinted 1962).
Martin-Pardey, Erva. Waage. Lexikon der Ägyptologie. 6 vols.
Wiesbaden, 1972–1984.
Petrie, W. M. F. Glass Stamps and Weights; Ancient Weights
and Measures. 2 vols. Egypt: The British School of
Archaeology, 1926. Reprinted Warminster, England: Aris
& Phillips, 1974.
Weights and measures of the Hebrews 2251
Skinner, F. G. Weights and Measures. London: Her Majesty's
Stationery Office, 1967.
Spiegler, O. Die Bestimmung der Gleichgewichteslage der
Ägyptischen Balkenwaagen, Bulletin of the Society of
Historical Metrology 1, No. 2. 59–63.
Vleming, Sven V. Masse und Gewichte. Lexikon der
Ägyptologie. 6 vols. Wiesbaden, 1972–1984.
Wilkinson, Manners and Customs of the Ancient Egyptians
II. London: John Murray, 1837.
Weights and Measures of the Hebrews
L IONEL H OLLAND
Weights
The earliest Biblical reference to a unit of weight is in
Genesis 23:16: “Abraham came to an agreement with
him and weighed out… four hundred shekels (of silver)
of the standard recognized by merchants.” Abraham
was a new immigrant in Canaan. The shekel units
which he used to weigh out the silver for Ephron the
Hittite may have been those of his native Mesopotamia,
with which he would have been most familiar.
The Mesopotamian shekel weighed about 8.3–8.5 g.
Sixty shekels = 1 maneh (Greek: Mna, mana. English
translations of the Bible often render maneh as “pound”)
of 497–508 g. Sixty maneh = 1 kikar (Greek: Talanton;
English: talent) of 29.8–30.5 kg. This system, which was
originally Sumerian, in the fourth or third millennium
BCE, was used with little change for thousands of years
by the peoples of Mesopotamia; the terms SH-Q-L,
M-N-H, and K-K-R are Semitic in origin and were first
used by the Akkadians. Systems similar in structure,
though varying in detail and in the absolute mass of the
units, were used by most of the peoples of the Eastern
Mediterranean, well into Roman Imperial times.
The term shekel is, in short, a generic one, and in this
context could apply to any one of a number of units
(including the Mesopotamian) in use in Canaan in the
Bronze Age. Since the price of 400 shekels for the Cave
of Machpelah was quoted by Ephron himself (Genesis
23:14), these may very well have been shekels of the
Syrian standard (also called Ugaritic), weighing about
9.4 g. Weights on this standard were used throughout
Canaan and northern Syria, and are found at Bronze Age
sites in the Aegean sea, as far north as Troy. The standard
also reached Egypt, where its unit became the qedet. W
Three thousand Syrian shekels made one talent.
Genesis 24:22 says: “…the man took a gold nose-
ring weighing half a shekel, and two bracelets for her
wrists weighing ten shekels…” The Hebrew word
rendered in English as “half a shekel” is beka’. The
word means “half,” and appears on certain Iron Age
weights described below.
2252 Weights and measures of the Hebrews
The next mention of units of weight in the Bible is in
the Book of Exodus. Several hundred years have
passed; the Children of Israel, now a numerous people,
have received the Law at Mt. Sinai. This includes
instructions for the payment of taxes and the perfor-
mance of religious ritual, which call for clearly defined
units of measurement. The first specifically Hebrew unit
of weight is shekel ha-kodesh, which the King James
version translates as “shekel of the sanctuary” and the
New English Bible as “shekel by the sacred standard,”
“twenty gerahs to the shekel” (Exodus 30:13). This unit
is referred to repeatedly (Exodus 30:24, 38:24–26,
Leviticus 5:15, 27:25, etc.). Reference is now also made
to talents of silver and gold: the talent weighed 3,000
shekels (Exodus 38:25, 26). There are no surviving
artifacts of this period from which the absolute mass of
these units might be inferred.
It should be stressed that the shekel ha-kodesh and its
fractions were used by the Hebrews for religious
purposes, but not necessarily in commerce. In the ancient
world, as in medieval Europe, there were many weight
systems in simultaneous use in international trade.
Weights representing all the major systems of the ancient
Near East are found at all important archeological sites in
Israel, and were obviously in use there. These systems
are often linked by networks of interrelationships, with
multiples of different units having a common mass and
being used interchangeably.
The oldest extant weights bearing Hebrew inscrip-
tions date from the period of the Kingdom of Judah
(tenth to sixth centuries BCE). They are normally of
limestone, dome-shaped with a flat bottom; a few
bronze specimens exist. There are several types, all
rare. The total number of known specimens of all
groups is a few dozen. They are:
. Necef. Average mass about 9.8 g. The necef is not
mentioned in the Bible. The name means “half ”: it
may have originally been half of some more ancient
unit. Scholars today identify the necef with the
Canaanite/Syrian/Aegean unit already described. It
is also equivalent to 1 Egyptian qedet and to four-
fifths of the royal shekel of Judah (see below).
. Beka’. This word also means “half ” (Genesis 24:22,
see above). The mean mass of this group is about 6 g.
The beka’ is one-half of the “shekel of the king” first
mentioned in II Samuel 14:26.
. Pym (Petrie: peyem). Mean mass about 7.8 g. The
word appears once in the Bible (I Samuel 13:21):
“The charge was two-thirds of a shekel (Hebrew:
pym) for (sharpening) ploughshares and mattocks,
and one-third (of a shekel) for sharpening the axes
and setting the goads.” The pym is two-thirds of the
royal shekel.
. Shekel. Weights of this group are inscribed, not with
a word, but a sign γ. Its meaning has been variously
interpreted as being an official adjuster’s mark, a
symbol of the bag (tsror) in which lump silver was
often carried, and a stylized representation of the
royal scarab emblem of the kings of Judah. Sir
Flinders Petrie read it as a monogram composed of
the Greek letters chi λ and omicron ο, and on this
basis, erroneously attributed these weights to a
(supposed) system which he called Khoirine. The
mean mass of the shekel, calculated from all known
marked specimens, is about 11.4 g.
The shekel sign is accompanied by a mark indicating
the denomination: 1, 2, 4, 8, 16, 24. Since numbers of
shekels are given in the Bible in multiples of 50, this
binary multiplication seems illogical; it is explained by
the fact that this shekel is equal in mass to one and a
quarter Egyptian qedets. Hence, 4 shekels = 5 qedet,
8 shekels = 10 qedet, etc. The denomination marks on
the shekel weights are, in fact (apart from simple
strokes to indicate 1 and 2), Egyptian hieratic decimal
numerals, the sign for 4 shekels meaning in Egyptian 5,
that for 8 meaning 10, and so on. These weights thus
could be conveniently used interchangeably to weigh
either in shekels or qedets. In fact, it can be seen that all
the units here described are closely related. The
Canaanite/Syrian/Aegean maneh of about 940 g can
be divided into either 80 royal shekels, 100 necef (or
qedet), 120 pym, or 160 beka’, providing simple
arithmetical relationships between the systems preva-
lent in different regions.
The name “royal shekel” is confirmed by the
existence of a two-shekel weight inscribed in Hebrew
“L-M-L-KH” (of the king). A series of fractional
weights also exists, both for the shekel and for the
necef, showing that both were divided into 20 units
(gerah). The denominations of these are also indicated
by Egyptian hieratic numerals.
Scholars attribute most of these weights to some time
near the end of the seventh century BCE, leading some
of them to suggest that the religious reforms of Josiah
(II Kings 22, 23; II Chronicles 34, 35) were accompanied
by a reform of weights and measures. There is no direct
documentary evidence of this. In fact, biblical references
are less than enlightening. Ezekiel, foreseeing the
rebuilding of the Temple, orders that the maneh shall
consist of “twenty, and twenty-five, and fifteen (i.e., 60)
shekels.” This is so inconsistent with other data that the
translators of the New English Bible assume a
misspelling in the Hebrew text, and render essrim as
“ten” (Ezekiel 45:10), giving 50 shekels to the maneh
as before. After the Temple was rebuilt, the people of
Israel pledged to pay an annual tax of one-third shekel,
not one-half as in Moses’ time (Nehemiah 10:32).
Nothing more is heard of the shekel ha-kodesh; it is
last mentioned in Numbers 18:16. Yet the correct
weight of the Temple tax was a subject of deep concern
to the Jewish religious authorities; it is discussed at
length in the Mishnah, the textbook of Jewish laws. By

the first century BCE it had been decided that all
religious taxes should be paid in “shekels of the sacred
standard, of the maneh of Tyre.” The Tyrian shekel at
that time weighed a little over 14 g and was famous for
its purity of metal and uniformity of weight. No other
coin would do. This is the reason why Jesus found
so many money-changers in the Temple precinct
(Matthew 21:12): Jewish pilgrims arriving from foreign
parts and wishing to pay their dues needed Tyrian
shekels in order to do so.
From AD 66 to 70, when Jerusalem was besieged by
the Romans, and Tyrian coin was no longer obtainable,
the shekels struck throughout the siege by the
embattled Jews (probably for religious use) were of
the Tyrian standard.
The use of Hebrew weights and measures for secular
commerce almost certainly ended with the fall of the
Kingdom of Judah and the destruction of the Temple by
Nebuchadnezzar in 586 BCE. There is no evidence
from the successive periods of Persian, Hellenistic, and
Roman domination of the Land of Israel (or from times
of relative independence, under the Hasmonean and
Herodian dynasties) of the use of any specially Jewish
units. Rabbinical discussions of weights and measures
usually center on attempts to define Biblical units in
terms of contemporary non-Jewish systems, in order to
ensure the proper performance of the Mitzvoth
(commandments).
Measures of Length and Area
The Hebrew unit of linear measure was the ammah or
cubit – the length of a man’s forearm, from the elbow to
the tip of the middle finger. Deuteronomy 3:11 actually
uses the expression le-ammat ish – “by the forearm of
an adult male.” The measurement of length by units
based on parts of the human body is universal among
ancient cultures.
The Bible mentions no unit greater than a cubit. Kaneh
ha-midah, the surveyor’s measuring rod (or reed), was
six cubits in length (Ezekiel 40:5). The measurements of
the inner temple are given by Ezekiel as 500 × 500 kaneh
(Ezekiel 42:16–20): the Authorized Version translates
these as “reeds”, the New English Bible as “cubits”.
Surveyor’s measurements of up to 25,000 cubits are on
record (Ezekiel 45:1). Long distances are expressed in
terms of a day’s journey, derekh yom (Genesis 30:35;
Numbers 11:31; II Kings 3:9).
The Hebrews used two cubits, one a hand’s breadth
(tefah or tofah) longer than the other (Ezekiel 40:5,
43:13). The longer cubit was used in the rebuilding of the
Second Temple (Ezekiel 40–48), and also, presumably,
in building the Temple of Solomon, since the dimensions
quoted for the Holy of Holies and the Sanctuary are the
same in both cases (20 × 20 cubits and 20 × 40 cubits,
respectively: I Kings 6:14–28, Ezekiel 41:1–4).
Weights and measures of the Hebrews 2253
The use of two cubits suggests an association with
Egyptian linear measure. The greater (“Royal”) Egyptian
cubit was of seven hands, each hand of four fingers (also
four spans of seven fingers each). Its length of about
52.5 cm is attested by many surviving measuring rods
and by records of the dimensions of surviving monu-
ments. The lesser Egyptian cubit was of six hands.
The Biblical fractional units of length are the span
(zeret – Exodus 28:16, 39:9; I Samuel 17:4), the hand
(tefah or tofah – Exodus 25:25, 37:12; I Kings 7:26; II
Chronicles 4:5), and the finger (etzba – Jeremiah 52:21).
The only direct clue to the absolute value of the Hebrew
cubit is the Siloam inscription (ca. 700 BCE) from which
Conder estimates a value of 17.1 in. (43.4 cm); he warns,
however, that this estimate may be in error by 10%
or more.
The Mishnah (second to third century AD) states that
the greater cubit was of six hands, the lesser of five; the
hand is of four fingers. Six hands of Egyptian
dimensions would measure 45 cm. Most rabbinical
estimates, however, are in the range 58–66 cm.
There are no specially named units of area in the
Bible. Ezekiel, discussing the layout of rebuilt
Jerusalem, speaks of “twenty-five thousand cubits
square (revi’it)” (Ezekiel 48:20).
Measures of Cubic Capacity
…your bushel (Heb. ephah) and your gallon (Heb.
bath) shall be honest. There shall be one standard
for each, taking each as the tenth of a homer, and
the homer shall have its fixed standard (Ezekiel
45:11).
Solid and liquid measure are thus given a common
basis. The volumetric measures mentioned in the Bible
are Ephah (Leviticus 19:36; Deuteronomy 25:14
(“measure”); Judges 6:19; Ezekiel 45:24); S’ah, seah
(Genesis 18:6; I Samuel 25:18; I Kings 18:32 (always
“measure”)); Kab, cab (II Kings 6:25); Omer (Exodus
17:36 – “An omer is one-tenth of an ephah”); these are
always associated with solid measure. Letekh (Hosea
3:2 – AV: “a half homer of barley”; NEB: “a measure of
wine”) is mentioned only once. Homer – (Ezekiel
45:11, see above) and Kor, cor (I Kings 4:22, 5:11) are
associated with both solid and liquid measure. Bath
(I Kings 7:26, 38; I Chronicles 2:10) and Log (Leviticus
14:10, 12, 15, 21, 24) are associated with liquid measure.
The relations between any of these measures are only
mentioned once or twice. The Mishnah gives a table W
relating some of them: 1 ephah = 3 se’ah; 1 se’ah = 6 kab;
1 kab = 4 log; 1 log = 6 eggs. The water displaced by 1 egg
is about 55 ml (there are differing rabbinical views on
the precise quantity), which would give the log a volume
of 330 ml and the ephah and bath each about 24 l.
Attempts to establish mathematical relationships
between ancient units of length, mass, and volume
2254 Weights and measures in the Indus valley
were popular a century ago, but today are not taken
seriously. Reliance on post-Biblical authors such as
Josephus Flavius is also risky: Josephus lived more than
a thousand years after King Solomon, and is more likely
to have been concerned with giving his (mainly non-
Jewish) reading public a notion of orders of magnitude
than with achieving a maximum of mensural precision.
Hebrew weights and measures will always remain a
subject fraught with uncertainty, which will be dispelled
only when archeologists are able to recover the actual
instruments of measurement.
References
Barkay, G. A Group of Iron-Age Scale Weights. Israel
Exploration Journal 28.4 (1978): 33–4, 209–17.
Conder, C. R. Hebrew Weights and Measures. The Palestine
Exploration Fund Quarterly Statement, London (1902):
175–95.
De Sgani, Leah. Shitot ha-Mishkal be-Eretz Israel (Weight
Systems in the Land of Israel). Perakim be-Toldot Ha-Mis-
khar Be-Eretz Israel (Chapters in the History of Commerce in
the Land of Israel). Ed. B. Keidar, T. Dothan, and S. Safrai.
Jerusalem: Ben-Zvi Memorial, 1990 (in Hebrew).
Kletter, Raz. The Inscribed Weights of the Kingdom of Judah.
Tel Aviv Journal of the Institute of Archaeology of Tel Aviv
University 18 (1991): 121–63.
Petrie, W. M. F. Ancient Weights and Measures. London:
Department of Egyptology, University College, 1926.
Scott, R. B. Y. The Shekel Sign on Stone Weights. BASOR
153 (1959): 32–5.
---. Shekel-Fraction Marking on Hebrew Weights. BASOR
173 (1964): 53–64.
---. The Scale-Weights from Ophel, 1963–64. Palestine
Exploration Quarterly 96 (1965): 23–4, 128–39.
---. The N-S-F Weights from Judah. Bulletin of the American
Society for Oriental Research 200 (1970): 62–6.
Weights and Measures
in the Indus Valley
S HIGEO I WATA
The golden era of the Indus civilization in ancient India
extended from 2300 to 1750 BCE. This vast civiliza-
tion had significant uniformity and standardization in
its material culture, as reflected in its town planning,
building construction, pottery, metallurgy, and system
of weights and measures.
The prosperity of the Indus cities depended to a large
extent on trade. Many raw materials were brought by
land and sea routes from within and outside the Indus
valley. Inland trade must have extended beyond the
Baluchistan to Afghanistan route and the Iranian high-
lands on the one hand, and to the Punjab and Aravalli
hills on the other. Overseas trade covered the Makran and
Persian Gulf ports on the west and the Gujarat and
Konkan ports, if not those of the Malabar coast in the
south. The writing system has not yet been deciphered,
and the names of their measuring units are still unknown.
Length
A unit of length, which corresponds to a fathom, was
revealed in the process of analyzing the city plans of
Lothal, Surkotada, Kalibangan, Dolavira, Harappa, and
Mohenjo-daro. The mean value with standard deviation
of the units used in these cities was 168 ± 1.1 cm.
The minimum division of graduation found in the
segment of an ivory-made linear measure excavated in
Lothal was 1.79 mm (that corresponds to 1/940 of a
fathom), while that of the fragment of a shell-made one
from Mohenjo-daro was 6.72 mm (1/250 of a fathom),
and that of bronze-made one from Harapa was 9.33 mm
(1/180 of a fathom).
Area and Volume
The use of area measures should have been quite
common. There is, however, no piece of evidence that
could be connected with an area measure. In the
excavations carried out at various sites of the Indus
civilization a variety of pots made of clay, and sometimes
of metal, have been discovered. No systematic determi-
nations of the volumes of pottery seem, however, to have
been made.
Mass
The oldest known weight in the Indus measuring
system was excavated from Dashli Tepe, south
Turkmenia in Russia. This era dates back to the fifth
millennium BCE. The other three weights belong to the
fourth millennium BCE and were discovered in
northern Iran. These weights belong to pre-Indus
civilization. In the third millennium BCE the Indus
measuring system was further developed in the ancient
regions of Iran and Afghanistan.
A total of 558 weights were excavated from Mohenjo-
daro, Harappa, and Chanhu-daro, not including defec-
tive weights. They did not find statistically significant
differences between weights that were excavated from
five different layers, each measuring about 1.5 m in
depth. This was evidence that strong control existed for
at least a 500-year period. The 13.7-g weight seems to
be one of the units used in the Indus valley. The
notation was based on the binary and decimal systems.
Eighty-three percent of the weights which were
excavated from the above three cities were cubic, and
68% were made of chert.
Balance pans were made of copper, bronze, and
ceramics. A bronze beam was found with the two pans
in Mohenjo-daro. The fulcrum was the cord-pivot type
(Fig. 1).

Weights and Measures in the Indus Valley. Fig. 1 Balance
pans (photograph by the Maninichi Newspapers, 1961; used
with permission).
Weights and Measures in the Indus Valley. Fig. 2 Weights
(photograph by the Mainichi Newspapers, 1961; used with
permission).
The measuring system used in the Indus valley was
different from the Mesopotamian and Egyptian measur-
ing systems, but the sensitivity of precision balances
used in these regions is assumed to have been
comparable. The weights excavated from Taxila (sixth
century BCE to seventh century AD) descend from the
system of weights used in the Indus civilization (Fig. 2).
References
Bhardwaj, H. C. Aspects of Ancient Indian Technology. Delhi:
Motilal Banarsidass, 1979.
Hori, Akira. A Consideration of the Ancient Near Eastern
Systems of Weight. Orient 22 (1986): 16–36.
Iwata, Shigeo. On the Standard Deviation of the Weights of
Indus Civilization. Bulletin of the Society for Near Eastern
Studies in Japan 27.2 (1974): 13–26.
---. Development of Sensitivity of the Precision Balances.
Travaux du 1er Congrès International de la Métrologie
Historique. Zagreb: Jugoslavenska academija znanosti i
umjet-nosti, Historijski Zavod, 1975. 1–25 + Fig. 1.
---. History of Weighing Scales. Journal of Japan Society for
Design Engineering 38.9 (2003): 438–51.
Mainkar, V. B. Metrology in the Indus Civilization. Frontiers
of the Indus Civilization. Ed. B. B. Lal and S. P. Gupta.
New Delhi: Books & Books, 1984. 141–51.
Rao, S. R. Lothal and the Indus Civilization. London: Asia
Publishing House, 1973.
Weights and measures in Islam 2255
Weights and Measures in Islam
U LRICH R EBSTOCK
In the sphere of Islamic influence the Quranic
injunction “to give full measure and to weigh with
the right scales” (Qur’ān 17:35), led - in the long run -
to systems of measuring that were subjected to the
authority and control of politics and law. But only a few
measuring standards were substituted by new Islamic
prescriptions. The most lasting novelty set off through
the Quranic revelation affected the measure of time. It
assumed a dual character. The natural solar year gave
way to the ritual lunar year (29–30 days of the month),
except in some fields of public administration and
astronomical science. Within the sphere of the metro-
logical systems of measures and weights, however, the
Qur’ān remains vague. Among the terms most often
mentioned, kayl or mikyāl (Sura 12:12 and passim,
‘measure of capacity’), mīzān and mithqāl (Sura 6:152,
21:47 and passim, ‘weight’), range first. Others, like
qint. ār (3:75, ‘hundredweight’), darāhim (3:75, pl. of
dirham), dīnār (3:75) and h.abba min hardal (21:47,
˘
‘grain of mustard’), are used in a metaphorical sense. A
more systematic elaboration of metrical definitions and
ratios, mainly for juridical purposes, was effectuated in
the ‘traditions of the Prophet’ (h.adīth, pl. ah.ādīth). The
characteristic aspects of the genesis of Islamic law – a
cumulative development until the fourth and tenth
century, geographically restricted proliferation of the
various law schools (madhāhib), lack of normative
authority – did, however, not allow for the introduction
of uniform and ubiquitously accepted Islamic metric
systems. Thus, the weights and measures which were
used in Arabia and outside it, in the lands conquered by
the Muslims, co-existed side by side, replaced each
other, sometimes only by function, or by name, or
intermingled. The striking feature of the metric systems
that were in use in the early, medieval and pre-modern
Islamic countries was their diversity. Neither in time,
nor in space, could standard values develop that were
accepted beyond their regional borders and their rulers’
period. Although some names of units of measure,
e.g. dhirā٬, dirham or rat. l, were widely diffused
throughout the Islamic world, their absolute values and
use in practice differed considerably. Notwithstanding
this bewildering array of measures, a few basic terms
that have survived until modern times stand for the
W
cultural continuity between the Ancient Orient and the
Islamic world: rat. l (Greek litron, ‘litre’), irdabb (Greek
artabe, Persian ‘measure of capacity’), mat. ar (Greek
metretes, ‘measure of 40 L’), k.ist. (Greek xestes, Latin
sextarius, ‘jug’), k.int. ār (Latin centenarius, ‘hundred-
weight’), k.af īz (Persian ‘measuring cup’), k.īrāt. (Greek
2256 Weights and measures in Islam

keratios, ‘carat’), istār (Greek stater, ‘weight of gold
coin’), dirham (Greek drachme), dīnār (Latin denari-
us), kayl (Aramaic measure of capacity), mann/mannā
(Babylonian unit of weight), kurr (Babylonian measure
of capacity).
The development of systems of measure was
influenced by the interplay between cultural tradition
and the order of authorities. Until the tenth century,
the spread of Islam brought about an intercontinental
economic and cultural sphere which amalgamated mea-
suring standards of Egyptian, Arabic, Greek, Roman-
Byzantine, Mesopotamian and Persian origin. This
resulted in a multiplicity of regional and func-
tional systems of measurement, which were constantly
modified by power politics, institutional reforms or,
simply, by the rulers’ autocratic order. Quite often, the
advent of new dynasties brought about the introduction
of new metrological standards, mainly in the field of
the basic weight units of currency of gold (mithqāl) and
silver (dirham), and the exchange rate of gold dīnār
and silver dirham, which were fixed at a very early state
by the canonical texts of the Qur’ān and the prophetic
tradition (h.adīth). Contrary to the more or less stable
weight rate (10 weight dirhams equal 7 weight
mithqāl), which everyday use polished into the handier
ratio of 3:2, the prescribed rate of value (10 silver
dirhams equal 1 gold dīnār) incessantly deteriorated
over the course of time. Ratios of 12:1, 16 4/5:1, 20:1,

Alphabetic list of abbreviations used

ar aruzza, pl. aruzzāt grain of rice

ash ashl, pl. ashwāl, ushūl part of a ‘rope’
az azla/azāla, pl. azāla, azālāt unit of capacity measure
٬as ٬ashīr, pl. ٬ashīrāt, ٬ushrān, a٬shur tenth
ba bat. t. a leather bottle
bā bāb, pl. abwāb portion
bā٬ bā٬, pl. abwā٬ arms’ span
da dānik., pl. dawānik., dawānīk. 1/6 of a dirham/dīnār
dh dhirā٬, pl. dhirā٬āt, adhru٬ ell
dī dīnār, pl. danānīr Gold dīnār
dir dirham, pl. darāhim silver-dirham
dj djarīb, pl. djirbān, adjriba unit of square measure
djo djou barleycorn (Persian)
fa fals, pl. fulūs, aflus small coin
fad faddān, pl. fadādīn yoke of oxen
far farq unit of capacity measure
fā fātil small unit of weight
ghi ghirāra, pl. gharā’ir sack
h.a h.abba, pl. h.ubūb, h.abbāt seed
h.ab h.abl, pl. h.ibāl, h.ubūl rope, thread
h.i h.iml, pl. ah.māl camel-load
ir irdabb, pl. arādibb unit of capacity measure
is istār, pl. asātir unit of weight measure (Greek)
is. is.ba٬, pl. as.ābi٬ width of middle finger
ka kaff, pl. kaffāt hand
kā kāra load carried on the back
kay kayl/kayla, pl. akyāl, akāyil unit of weight measure
kayl kayladja, pl. kayladjāt, kayālidj unit of capacity measure
ku kurr, pl. akrār unit of capacity measure
k.a k.afīz, pl. ak.fiza, k.ifzān unit of capacity weight
k.ab k.ab(a)d.a, pl. k.abad.āt width of fist
k.ad k.adah., pl. ak.dāh. unit of capacity measure
k.al k.alam, pl. ak.lām ‘strip’
k.ām k.āma build, fathom
k.as. k.as.aba, pl. k.as.abāt pole, rod
k.in k.int. ār, pl. k.anāt. īr ‘hundredweight’
k.is k.ist. , pl. ak.sāt. ‘portion’
k.ī k.īrāt. , pl. k.arārīt. carat
k.u k.ulla, pl. k.ulal, k.ilāl jug
ma mann/mannā, pl. amnān, amunnā’ unit of weight measure
mak makkūk, pl. makākīk unit of capacity measure
mar marzbān unit of capacity measure (Persian)
 Weights and measures in Islam 2257

Alphabetic list of abbreviations used (Continued)

mat. mat. ar, pl. amt. ār unit of liquid
mi mithk.āl, pl. mathāk.īl unit of weight measure
mish mishk.ā٬ drinking-vessel
mu mudd, pl. amdād unit of capacity measure
mud mudy, pl. amdā’ unit of capacity measure
na nak.īr ‘small spot’ (Arabic)
nu nügi, pl. nügiler unit of weight measure (Turkish)
ok. ok.k.a, pl. ok.k.alar unit of weight measure (Turkish)
pe peymāne bowl
r rat. l, pl. art. āl ‘litre’
ru rub٬, pl. arbā٬ fourth
sha sha٬īr, pl. sha٬īrāt grain of barley
shi shibr, pl. ashbār span of hand
si silsila, pl. salāsil chain
su sunbul, pl. sanābil ear of grain
s.a s.ā٬, pl. as.wu٬, as.wā٬ unit of capacity of weight
th thumn, pl. athmān eighth
ti tillīs unit of capacity measure
t. a t. assūdj, pl. t. asāsīdj unit of weight measure
٬ush ٬ushr, pl. a٬shār tenth
ūk. ūk.īya, pl. awk.iyā, ūk.īyāt unit of weight measure
wa waiba unit of (dry) capacity measure
was wask., pl. awsāk. (camel’s) load
za zabīl basket made of palm-leaves

30:1 and even 50:1 are recorded. Nevertheless, modern
Islamic jurists insist on referring to the canonical rates
(10:7; 10:1) when, for example, fixing the minimum
income (nis.āb) for the obligatory alms payment (zakāt)
by Muslims of a determinate portion of their lawful
property. The canonical ratio of the value of gold and
silver (10:1) reflected the historical situation in the
Mediterranean region and the Middle East after the
Roman period (12:1). During the Il-Khānid period
(thirteenth to fourteenth century), silver from Central
Asia was massively imported to the West, the price of
silver sank again, for a short period, to the Roman
value. Gold, in contrast, remained remarkably stable
over the millennium. Under the Persian king Darius the
Great (522–486), mutton cost the same as in Anatolia
in 1340: the equivalent of 1.9355 g pure gold.
The overall cultural diversity of the Islamic world
corresponds to the diversity of the metric systems,
which came into use between the Atlantic and the
Indian sub-continent. Three geographical units can be
differentiated: The Islamic Arab West, from Andalusia
to Iraq, Persia and the adjacent areas under Persian
influence, and India. The following comments omit the
metric systems of India (see EI 2 VI, pp. 121a–122a,
s.v. makāyīl, and VII, pp. 138a–140b, s.v. misāh.a) and
concentrate on the development in the Arab West, in
consideration of the situation in Persia. Emphasis is laid
on the early Islamic and medieval period. The absolute
equivalents in modern metric values stem back to
archaeological evidence or observations of European
travellers.
Measures of Length
Along with the basic unit of length, the dhirā˓, several
other units were used, some of them only for particular
purposes (construction, geometry, etc.). In theory, i.e.
without considering their actual common occurrence or
precise values, these units could be arranged to the
following equation:
1 ash = 1 si = 10 bā = 10 k.as. = 15 bā٬
(or k.ām) = 60 dh (= Persian gaz) = 360
k.ab = 1,440 is. = 3,600 fa = 8,640 sha.
The ‘black ell’ (al-dhirā٬ al-sawdā’), being ca.
54.04 cm, is said to refer to the length of the ell (from
the elbow to the tip of the middle finger) of a slave
of the Caliph al-Mans.ūr (r. 754–775) or the Caliph
al-Ma’mūn (r. 813–833). Another etymology links
the measure to the unit by which the ‘Nilometer’ of the
island of al-Rawd. a was operated. There are almost
thirty variants of the ell, some varying 30-fold from
the original. By the eleventh century, at least 11 differ-
ent types of dhirā’ can be differentiated:
. 1 dhirā٬ sawdā’ = 1 + 1/7 + 2/3 ·1/7 dh al-yad (of the W
hand)
= 1 + 1/8 +1/9 dh al-h.adīd (iron ell)
. 1 dh fid.d.īya (silver) = 1 – 1/7 dh al-sawdā’
. 1 dh yūsuf īya (of Abū Yūsuf, d. 798) = 1− 2/3 ·1/7
dh al-sawdā’
. 1 dh hāshimīya (of the Banū Hāshim) = 1 + 1/8 +
1/10 dh al-sawdā’
2258 Weights and measures in Islam
. 1 dh bilālīya (of Bilāl b. Abī Burda, d. 739) = 1 dh
al-sawdā’ + 2 + 2/3·1/7 is.
. 1 dh fid.d.īya (al-misāh.a) = 7 or 8 dh al-yad
. 1 dh ٬umarīya (of ٬Umar b. ٬Abdal٬azīz, d. 720) =
1 + 1/2 dh al-yad
. 1 dh mīzānīya (surveyor’s ell) = 3 dh al-yad
. 1 dh mābahrāmī = 1 + 1/2 dh al-h.adīd
In addition to these different norms of the dhirā˓,
a multitude of ells was used depending on the
profession involved: carpenters, cloth-makers, con-
structors etc. Moreover, the ells used in different cities
under the same name differed: the medieval cloth-ell of
Damascus (ca. 63.035 cm), for example, was 1/12
longer than the cloth-ell of Cairo (58.187 cm).
Measures of Area
The calculation of the surface of (straight) areas operated
with the conventional measures of length. The basic units,
however, were the k.af īz and the djarīb, two specific
measures of surface area. Originally and throughout the
Islamic period, both units also served as measures of
capacity. One djarīb was conceived of as representing the
surface area of agricultural land which could be sown
with the amount of seed one djarīb contained.
Based on the ratio of the length units (1 ash = 10
bā = 60 dh = 360 k.ab = 1440 is.), the following ratio of
units of surface area measurement can be generated:
1 ash = 60 dh · 60 dh = 3,600 dh 2 = 1 dj
and:
1 dj = 10 k.a = 10· 360 dh 2
= [in Persia] 60 ka = 600 ٬as = 600 · 6 dh2
= 100 ٬as = 100 · 36 dh 2
This djarīb was called the ‘small’ djarīb, being 100
square k.abad.a (or k.as.aba; the units being often
exchangeable) which renders: 100 · (399 cm · 399 cm)
2
= 1,592 m2. The ‘big’ djarīb had 5,837 1/3 m2, i.e. 3 2/3
‘small’ djarīb, and corresponded roughly to the
predominantly Egyptian faddān which was calculated
ashl bāb dhirā٬
ashl 1 dj 1 k.ab 5/3 ash
bāb 1 ash 1 ash = 6 dh 2
dhirā٬ 1/36 ash = 1 dh 2
k.abad.a
azla3 dhirā٬3 k.afīz3
azla3 1 100 6,000
dhirā٬3
as 400 square k.as.aba, i.e. 6,368 m2. During the
nineteenth century, the faddān was reduced to
4,200.833 m2.
If multiplied with one another these units render the
matrix (see above).
There is substantial evidence that the professional
surveyors during the Abbasid period used a specific
system of calculation. They divided the biggest unit,
the azla, into 100 dh mīzānīya which corresponded to
48 is. ٬umarīya (see above), hence:
1 az = 100 dh2 = 100 · 122 k.ab2 = 100 · 122 · 42 is.2
In the Turkish lands of the Ottoman Empire (Minor
Asia, Iraq, Syria and Palestine) the dönüm (turn), Arabic
dūnam, was – until recent times – the standard measure of
area. Originally measuring 939 m2, it has been adjusted in
colonial times to 1,000 m2 (in Iraq to 2,500 m2).
Measures of Capacity
Most of the confusion about the system of the Islamic
measures of capacity, both in primary medieval and
in modern secondary texts, dates back to the Oriental
practice to measure grain, pulse, and some liquids in
capacity, but not in weight. The Arabic term mīzān does
not clearly differentiate between the two. The transition
from volume to weight needs the related quantity of the
litre of water: the volume of approx. 75–77 kg of wheat
and 60–72 kg of barley correspond to the volume of 100
kg/L of water.
From this economic and agricultural use of measures
of capacity the proper mathematical and technical
calculation of volumes must be set apart. This calculation
is built on the calculation of the surface area multiplied by
the third dimension. The names used for the cubic units of
measure do not change. Related to the dhirā٬ mīzānīya
and based on the ratio 1 az = 100 dh 3 = 100 ku, the
following values are produced: (see matrix below)
Most of the units of measures of capacity are regarded
as units of weights too. It is therefore impossible
to separate the two systems properly. Depending on
k.abad.a is.ba٬
1/6 + 1/9 ash 1/24 + 1/36 ash = 2 1/2 dh 2
1/36 ash = 1 dh 2 1/144 is. = ¼ dh 2
1/216 ash = 1/6 dh 2 1/864 ash = 1/24 dh 2
2
1/1,296 ash = 1/144 dh 1/5,841 ash = 1/576 dh 2
k.abad.a3 is.ba٬3
172,800 = 102 · (12 k.ab)3 11,059,200 = 102 · (48 is.)3
1,728 110,592

the material measured, additionally different types of
the same unit, e.g., a ‘honey-fark.’ or a ‘barley-irdabb’,
were used. The absolute values of these types differed
considerably in different regions and periods. In order to
allow a comparative overview, units that are related to
each other by practical use are grouped together. Minor
local variations and temporal changes are ignored.
Few of these units have a canonical background:
1 was = 60 s.ā٬ = 240 mu = ca. 252 L. Far bigger than
this mudd (ca. 1.05 L) of Medina were the mudd of
Egypt and Iraq (2.5 L), of Syria (3.67 L), of the Maghreb
(4.32 L), and that of Jerusalem (100 L). The prophetic s.ā٬
was exactly 4.2125 L. Being the quantitative lower limit
(nis.āb) for the liability for the zakāt (alms) taxes, the
measure of 5 wask. of dates, for example, was equated in
value with 5 ūk.īya (= 200 dir = 529.9 g), 20 dī (or
mithk.āl, = 84.7 g, see later), 5 dhawd (camels), the nis.āb
of cotton (5 was = 1,600 dir ٬irāk.ī à 130 g), or 50 kay.
Therefore, the values given for one wask. greatly differ. In
the time of Hārūn al-Rashīd (around 800), a short-lived
wask. (1 was = 2 1/2 prophetic was) was introduced.
Towards the end of the seventh century, the k.af īz
(usually corresponding to 1/10 dj = 1/60 kurr) was
used instead of this prophetic s.ā٬ in Iraq. Another
specific k.af īz of capacity is recorded from Iraq around
990: 1 k.a = 1 k.affa (basket) = 1/2 zabīl (basket made of
palm leaves).
In Egypt grain, but in particular wheat, was measured
by irdabb: 1 ir = 6 wa = 24 ru = 48 k.ad = 90 ma = 96 k.ad
(small) = ca. 90 L. Different values extant for the irdabb
(between 72.3 kg, modern 182 L) may be explained also
by the difference of volume between, for example,
wheat, barley and lentils (100:80:104).
Egyptian flour was measured in tillīs: 1 ti = 3/2 wa =
3 ba = 15 ma = 24 k.ad = 22.5 L. There, the waiba of
rice (1 wa = 8 k.ad = 24 r kabīr), as observed around
1665, contained only 12.5 l. Three centuries before in
Tunis, it was equal to 12 prophetic mudd (ca. 12.6 l).
In Medina and Iraq, honey but also wheat was
measured in fark. : 1 fa = 3 s.a = 36 r bagdādī = 19 L. In
Egypt and Syria, the mudy – not to be confused with
the mudd - replaced the k.ist. when not oil but food was
measured. It is sometimes called ‘the Syrian djarīb’,
sometimes equated with the k.af īz. The practice in Syria,
however: 1 k.a = 8 mak = 12 s.a, the indication: 1 mud =
15 mak = 22 1/2 s.a does not confirm this. In Palestine, a
square mudy was known (1 mud 2 = 1 h.ab · 1 h.ab).
Olive oil was merchandised in mat. ar (1 mat. =
2 k.u = ca. 17 kg) in the Maghreb, in k.ulla (1 k.u = 12
th = 27 r = 13.6 kg) in Andalusia. In Egypt, the thumn
corresponded to 1/8 k.ad (today 0.29 L), in Qayrawān to
6 prophetic mu = 6.32 L. Oil and other liquids were
also measured in k.ist. : In Iraq, the ‘small’ k.ist. (1 k.i = 3
r = 1.22 L) was half of the ‘great’ k.ist. , in Egypt it was
half of a s.ā٬: 211 L; elsewhere the k.ist. is given as:
1 mat. = 4 k.i = 21 1/3 r djarwī (see below) = 192 ūk.
Weights and measures in Islam 2259
(capacity) = 256 ūk. (weight). In Andalusia, wine and
vinegar were sold in rub٬ (1 ru = 1/4 k.ad = 18 r = 216
ūk. = 1,728 mi = 8.16 L, in Persia the peymāne (bowl,
8.3 L) was in use for this purpose. In Iraq, wine, but
also oil and honey, were measured by makkūk or
mishk.ā٬ (drinking-vessel): 1 mak = 48 th à 50 dir = 64
mish à 37 1/2 dir = 7.5 L.
Another widespread unit of capacity was the
ghirāra, mainly used for grain: 1 ghi = 3 ir mis.rī =
12 kay = 14 mak = 72 mu dimashk.ī = 73 1/2 mu
mis.rī = 265 L. In Egypt, the kayla = 8 k.ad was 7.5 L
(modern 16.5 L).
This kayla is not identical with the kayladja,
presumably an originally Persian unit of capacity
measure: 1 kayl = ½ s.a = 1/3 mak = 3/14 ghi = 1/6 k.a
wheat = 1/5 k.a barley = ca. 1 7/8 ma = 2.5 L (or 2 L in
East Iran).
The most basic of all grain measures, especially in
the Islamic East, was the old Babylonian kurr.
1 ku = 30 kā = 60 k.a = 480 mak = 600
‘ush/‘as = 1,440 kayl = 5,769 ru = 7,200
r = 11,520 th = 2,925 kg (wheat).
Smaller than this ‘big’ kurr of Baghdad was the kurr of
Wāsit. and Bas.ra (1 ku = 60 k.a = 480 mak = 1,440 kayl
à 600 dir of wheat = 2,700 kg); a ‘reformed’ kurr even
amounted only to: 1 ku = 60 k.a à 25 r baghdādī =
609.375 kg (wheat). Moreover, depending on the kind
of grain measured, different akrār were used: In
fourteenth century Baghdad, the kurr of wheat weighed
2,925 kg, that of barley 2,437.5 kg, and that of rice
3.656,25 kg. The common sub-units of the kurr, the
k.af īz, makkūk, kayladja, and thumn differed respec-
tively, sometimes not only proportionally. Thus, in
twelfth century Aleppo, a quite different makkūk
existed: 1 mak = 19 sun = 28.5 r à 684 dir à 3.125
g = 60.92 kg. About the same time, the k.af īz of H.amāh
was 7/8 k.a of that of Shayzar. In Aleppo, 4 mak made
one marzbān (1 mar = 1/4 mak = 19/4 sun = 57/8 r =
4,873 1/2 dir = 15.23 kg).
Towards the end of the tenth century, the mathemati-
cian al-Būzdjānī compared the new ‘reformed’ (Arabic
mu٬addal) djarīb - this djarīb was not measured with 10
but with 2 1/2 k.a only – which was introduced after 978
by his Lord, the Būyid ٬Ad.ud al-Dawla, with four
different common types of the kurr. His systematic
treatment of the issue will throw some light on the
complex variety of the units used and their specific
relation when being transformed from one into another W
(see matrix below).
Besides simsim (sesame), h.int. a (wheat), djahkan-
dam (mixture of 1/2 h.int. a + 1/2 sha٬īr), and sha٬īr
(barley), a fifth category is formed to include all kinds
of grain and dry goods that do not belong to one of the
aforementioned categories: nuts, like almonds, pista-
chios and hazelnuts, dried pears, plums etc. From the
2260 Weights and measures in Islam

1. Ratio of kurr-Variants

Types of kurr mu٬addal kāmil fālidj hāshimī sulaymānī

mu٬addal 1 2 2 1/2 3 3 1/2 + 1/4

kāmil 1/4 1 1 1/4 1 1/2 1 1/2 + 1/4 +1 1/8
fālidj 2/5 4/5 1 1 1/5 1 1/2
hāshimī 1/3 2/3 5/6 1 1 1/4
sulaymānī 1/6 + 1/10 1/3 + 1/5 2/3 4/5 1
[ratio] 60 30 24 20 16

2. Djarīb per kurr

Types of kurr mu٬addal kāmil fālidj hāshimī sulaymānī

djarīb/kurr 24 12 9 3/5 8 6 2/5

fraction 1/3 · 1/8 1/2 · 1/6 1/2 · 1/6 + 1/6 · 1/8 1/8 1/8 + 1/4 · 1/8

3. Ratios Between sub-units (as indicated by Al-Būzdjānī)

makkūk ٬ushr kayladja rub٬ rat.l thumn

kurr 480 600 1,440 5,760 7,200 11,520

makkūk 1 1/4 12 15 24
٬ushr 1/4 + 1/6
kayladja 5 8
h.ubūb 4 5 60
rub٬ 1/12 + 1/48 1 1/4
thumn 1/20 + 1/60 1/2 + 1/8

4. Ratio of Capacity Between Different Kinds of Grain a) dirham al-fid.d.a (silver dirham)
Calibration of the silver (dirham) and gold (dīnār)
simsim h.int.a djahkandam sha٬īr coins was done with the help of glass weights. The
simsim 1 2 2 2/3 4 earliest preserved exemplars date back to the second half
h.int. a 1/2 1 1 1/3 2 of the eighth century. The dirham weight defined the
djahkandam 1/4 + 1/8 1/2 + 1/4 1 1 1/2 weight of the dirham coin, the mithk.āl weight the weight
sha٬īr 1/4 1/2 2/3 1 of the dīnār. The most precise glass weights of the mithk.āl
[ratio] 8 4 3 2 have an average weight of 4.233 g (max. tolerance 1/3
mg). Archaeological finds affirm both the weight of the
dirham in accordance to the canonical ratio of dirham:
matrixes 1–4, exactly 280 possible combinations result mithk.āl (= 10:7): 2.97 g, as to the ‘rounded’ ratio (= 3:2):
by which the transfer of one given quantity (and value) 2.82 g. An exceptional mithk.āl weight was in use in Egypt
of one commodity into another can be calculated. under the Ayyubid dynasty and in the Maghreb under the
Example: If 24 ku of oats (h.urt. umān = category of Almohad dynasty (4.722 g).
sha٬īr) should be transferred into kurr sulaymānī of The mithk.āl gold and the dirham silver were divided
pepper grass (h.abb al-rishād = category of h.int. a), then into k.īrāt. and h.abba.
the rule of seven is required, in short: mi gold = 20 k.ī = 60 h.a;
1
24 ku kāmil sha٬īr – 1/6 · 24 = 22 dir silver = 12 k.ī = 48 h.a (Iraq)
1
ku sulaymānī h.int. a + 30 k.a. mi gold = 24 k.ī = 60 h.a;
1
dir silver = 16 k.ī = 60 h. (Arabia, Egypt, Syria)
1
Measures of Weight Hence, the values (see matrix below).
The entire Islamic system of weights is based on the In addition to these general systematic differences a
dirham and the rat. l. The rat. l is the most common variety of deviating systems from different regions,
smallest unit, or reference, of weight. The weight of the authors and periods are recorded (tenth to thirteenth
dirham is used for two different purposes. The two century; indicated as I–V), that integrate sub-units
values differ correspondingly: like the dānik., t. assūdj, ٬ashīr, fals, and aruzza (which

gold k.īrāt.
Iraq 0.212 g
Egypt etc. 0.176 g
dānik. k.īrāt. t.assūdj
I. Dīnār 12 1/2 20 24
Dirham 12 24
II. Dīnār 6 20 baghdādī
24 bas.rī
Dirham 6 48 ˘baghdādī/bas.rī
III. Dīnār [12] [20] 24
IV. Dīnār 24/20 576
elsewhere corresponds to 25 h.a hardal, grains of
˘ defines the ha
mustard, i.e. ca. 0.0186 g); one author
.
hardal as 1/70 of a h.abba (which is sometimes replaced
˘by ‘kamha’, grain of wheat), 60 of which make one
. .
silver dirham, i.e. 1 h.a hardal = 0.0007 g. From the
vague comments of sources,˘ it must be assumed that
most of these different systems were in use as weight
measures too (see matrix below).
According to the actual ratio of value between gold
and silver currency, the moneychangers had to take
several factors into consideration when transferring
amounts of money from one currency into the other.
This could result in thirteenth century Egypt, for
example, when 16 4/5 dir were equivalent to 1 dī, in the
following calculations:
1 dī = 1,440 h.a fid.d.a [10 · 60/7· 16 4/5 = 1,440]; and
1 h.a gold = 1/5 + 2/25 k.ī = 16 4/5 h.a silver; or
1 fa = 2 2/5 h.a silver = 1/7 h.a gold [2 2/5: 1/7 = 16 4/5].
b) dirham al-kayl (weight dirham)
In contrast to the homogeneous evidence of the weight
of the ‘silver dirham’ the extant values of the weight of
the ‘weight dirham’ deviate considerably from one
another. They range from 3.086 g to 3.148 g. When not
indicated otherwise, the following comments will be
based on the established average standard value of 1
dir = 3,125 g with which the ‘canonical’ (ratio 10: 7)
mithk.āl of 4.464 g is corresponding. From textual
evidence some of which are included in the matrix
above, different regional values of the dirham/mithk.āl
weight can be deduced:
Egypt 3.125/4.68 g; Syria (Aleppo twelfth century)
3.14/4.427 g, (Aleppo nineteenth century)
1 dir = 3.167 g, Damascus 3.086/4.62 g; Anatolia
(Ottoman period) 3.086/4.81 g; Iraq 3.125/4.46 g;
Iran (fourteenth century) 1 mī = 4.3 g, (sixteenth
century) [3.26]/4.639 g; Maghreb 3.3/4.722 g;
East Africa (sixteenth century) 1 mī = 4.41 g.
Weights and measures in Islam 2261
gold h.abba silver k.īrāt. silver h.abba
0.0706 g 0.247 g 0.062 g
0.0706 g 0.186 g 0.0495 g
h.abba ٬ashīr fals aruzza
60
48 60
60 baghdādī 60
72 hurās./shāmī
60 96
36 hurās./shāmī
60 ˘ 240
[72] 600/7 600
With the exception of Persia, where the mann domi-
nated the system of weight measures, the rat. l became
the most common and widespread unit of weight
measure in the Islamic world, comparable in size and
function to the European ‘pound’ (Pfund, livre, libra,
Italian loan word ‘rotolo’). The rat. l was measured in
dirham. Depending on what was measured, and where
and when, the rat.l could take different numbers of dirham
(values between 96 and 1.040 are recorded) of different
dirham weights (standard value: 1 dir = 3.125 g).
If integrated into the early Meccan system:
1 r = 2 ma [à 130 dir] = 12 ūk. = 480 dir = 1/100 k.in,
the mithk.āl weights produce the following (fictitious)
relation:
1 mi = 20 k.ī = 60 [or 100] h.a = 10/7 dir = 1/336 r
(for Iraq; 1 k.ī = 0.223 g)
1 mi = 24 k.ī = 96 h.a = 3/2 dir = 1/320 r (for Mecca,
Egypt etc.; 1 k.ī = 0.195 g).
From archaeological (glass weights) and textual
evidence, several hundred rat. l weights are known. The
following list enumerates (in order of size, with ‘[…]’
values developed) some of the standard rat. l weights
repeatedly recorded (see matrix below).
Besides the rat. l, the mann was an important unit of
weight everywhere in the Islamic world, in particular
in the Persian East, where it weighed between 260 dir
(= 816.5 g) and 2,080 dir (= 6,656 g). A similar variety
of mann weights was used In Asia Minor (twelfth
century onwards). Until the fifteenth century it was
used instead of one half of a rat. l à 130 dirham. Then a
W
‘big’ mann (ca. 3 to 3.25 kg), and a ‘middle’ mann of
1.920 g came into use. During the Safawid period
(sixteenth century), a ‘super’ mann, later called the
‘royal’ mann (between 5.7 and 6 kg), was introduced.
The Ottomans used the ok.k.a (1 ok. = 2 nu = 400 dir à
3.207 g = 1.2828 kg) instead of the rat. l. Its stability
was proverbial: Okka her yerde dört yüz dirhem gelir
(… to be no different from anybody else).
2262 Weights and measures in Islam

Egypt (Abbasid period) 96 dir 300 g

Rūmī (Asia Minor) I 102 6/7 dir 321.43 g
Umayyad period [110 dir] 340 g
Rūmī (Asia Minor) II 120 dir 375 g
Iraq (medieval) 128 4/7 dir 401.79 g
Abbasid period (Egypt, 130 dir 406.25 g
Baghdad)
Maghreb 130 dir 406.25 g
Maghreb 137 1/7 dir 428.57 g
Umayyad period (Egypt) [140 dir] 437.5 g
Maghreb (Fāt.imid period) 140 dir 437.5 g
Egypt (later Abbasid period) 144 dir 450 g
Fulfulī 150 dir 468.75 g
Maghreb (Ibn Bat.t.ūt.a) 150 dir 468.75 g
‘big’ Egypt (Abbasid period) 160 dir 500 g
Maghreb (Ibn Bat.t.ūt.a) 180 dir 562.5 g
Laithī 200 dir 625 g
Djarwī 312 dir 975 g
Turkestan (fourteenth centu- 330 dir 1,031.25 g
ry)
Fes/Marrakesh (fourteenth 336 dir 1,050 g
century; = 16 ūk.)
Aleppo (twelfth and thir- 480 dir 1,500 g
teenth century)
Syria/Palestine (fourteenth 592 1/2 dir 1,851.56 g
century)
H.ims. (twelfth century) 684 dir [sic] 2,137.5 g
Aleppo (after thirteenth cen- [724 dir] 2,273 g
tury)
Jerusalem (medieval) 800 dir 2,500 g
H.ims. (Syria, medieval) 864 dir 2,700 g
Constantinople (eighteenth 876 dir 2,800 kg
century)
Jerusalem (nineteenth cen- 900 dir 2,812.5 g
tury)
Iran (Shīrāz, Fārs; in mann) 1,040 dir 3,250 g

The biggest unit - besides the rather colloquial
h.iml, camel-load (1 h.i = ca. 250 kg) -, was the k.int. ār,
the hundredweight (= 100 r). Depending on the type
of rat. l it was based on, the k.int. ār weights differed.
In medieval Egypt, different k.int. ār weights were
common: fulfulī (pepper) = 100 r à 144 dir = 45 kg;
laithī = 100 r à 200 dir = 62 kg; djarwī = 100 r à 312
dir = 96.7 kg; mannī = 100 r à 260 dir = 81.25 kg;
‘big’ = 24 ru = 240 r à 160 dir = 38,600 dir = 120 kg.
In a treatise composed by a customs officer in the
thirteenth century, additional k.int. ār names, but no
values, for specific goods are mentioned. While the
k.int. ār of Syria (Aleppo, H.ims., H.amāh) was always
equivalent to 100 local rat. l, it was taken for 100 mann
in late medieval Iraq. In Iran (fifteenth century)
and Asia Minor (Ottoman period) 1 k.int. ār weighed
ca. 57 kg.
The smaller weight unit of istār (1 is = 4 1/2 mi =
6 3/7 dir = 20.07 g), only known from Egypt, was used
there to weigh silk: 1 s-k.-t. [?] = 3 ru = 90 man = 180 is.
The Quranic ‘h.abba min hardal ‘(the ‘grain of mustard’,
see above), being 1/70 h.a of˘ 1/60 dir each (= ca. 0.0007 g),
seems to have remained the smallest unit of weight in use
in the Islamic world. If calculated properly, the fictitious
nak.īr (1djo = 6 [h.a] hardal = 72 fa = 432 fāl = 2,592
na = 1/96 mi = 0.045˘ g) would correspond to ca. 5 ng.
al-Kurashı̄
The˙ research of the history of weights and measures
and their use in the Islamic world is based on a variety
of sources. Unfortunately, no particular literary type of
text developed that could claim to be called ‘profes-
sional’. The information available is scattered over
texts on law, social and economic history, administra-
tion and geography. They generally lack a systematic
character, i.e. ignore comparative and proportionate
references. The most recent endeavor to collect all
information available in the historical sources was
undertaken by Mah.mūá Fākhūrī and S.alāh. al-Dīn
knawwām in: Madjmū˓at wah.dāt al-qiyās al-˓arabīya
 Weights and measures in Islam 2263

Aleppo: Weights
1 dir = 60 h.a 1 dī = (22 + 1/2) k.ī = 90 h.a
1 k.ī = 4 h.a = 2/45 dī 1 dī = 3/2 dī (Iraq)
1 h.a/dī = (6/7 + 2/21) h.a/dī (Egypt) 1 h.a = [(1 + 3/25) h.a (Syria)]
1 k.ī = (1 + 1/4) k.ī (Iraq) 1 k.ī = (1 + 1/8) k.ī (Iraq)
1 r = 7,560 k.ī

Ant.āk.iya: Weights
1 r = [16/17 r sulaymānī]
1 r = [4/5 r z.āhirī]
1 r = 384 dir = [12 ūk.] = [17 1/7 mi] = (268 4/5) mi
1 ūk. = 32 dir (22 2/5) mi
1 r = (3/5 + 1/25) r (Syria) = (2 + 1/2 + 1/20 + 1/100) r fulfulī = 4/5 r z.āhirī = (2/3 + 1/4 + 1/100) r h.aythamī

Ardabīl: Weights
1 r = [9/5 r sulaymānī] 1 r = 1.080 r = [12 ūk.] = 756 mi
1 ūk. = 90 dir 1 mi = 63 dir
1 r = 1 4/5 r sulaymānī = 8 2/5 r (Iraq)

Asyūt. (Egypt): Weights

1 r = [5/3 r sulaymānī] = r (T.ah.āwī, ٬Akkā) 1 r = 720 dir = 1/5 k.is
1 r = 1,000 dir = 700 mi = [12 ūk.] 1 ūk. = 83 1/3 dir = 58 1/3 mi
1 r = 1 2/3 r (Syria) = 31/3 r djarwī = 6 2/3 r fulfulī = (7 + 2/3 +
1/9) r (Iraq)

Baghdad: Measures of capacity

1 r = [3/16 s.a (H.idjāz) = [2 ma] = [1/4 mu (Damascus)] = 3/14 r (Syria)

Bardha٬a (Azarbaydjan): Weights

1 r = [7/5 r sulaymānī] 1 r = 840 dir = [14 ūk.]
1 ūk. = 588 mi 1 ūk./mi = 49 dir
1 r = 1 2/3 r sulaymānī
= (6 + 1/3 + 1/5) r sulaymānī

Damascus: Measures of capacity

1 mu = 1 4/7 mu (H.idjāz) = 4 r (Baghdad)

Weights
1r = 12 ūk. = 600 dir = 1 r sulaymānī 1 ūk. = 50 dir
1 r = 420 mi = 3,600 da = 14,400 k.ī = 36,000 h.a 1 k.ī = 1 3/8 da (Iraq)
1 ūk. = 35 mi = 300 da 1,200 k.ī = 3,000 h.a 1 k.ī = [15/16 k.ī (Aleppo)]
1 h.a = (6/7 + 1/28) h.a (Aleppo)
1 ūk. (small, silk) = 10 dir = 1/50 r = 1/20 is

Diyār Bakr (N-Syria)

1 dir = 60 h.a 1 dī = 22 1/2 k.ī = 90 h.a
1 k.ī = 4 h.a = 2/45 dī 1 h.a = 1/4 k.ī = 1/90 dī W
Diyār Mud. ar (N-Syria)
1 dir = 60 h.a 1 dī = 22 1/2 k.ī = 90 h.a
1 k.ī = 4 h.a = 2/45 dī 1 h.a = 1/4 k.ī = 1/90 dī
2264 Weights and measures in Islam

Djarwī: Weights
1 r = [6/7 ma (Syria, general)]
1 r = [1/2 r sulaymānī ]
1 r = 300 dir = 1/2 r (Syria) = (1/5 + 1/10) r (T.ah.āwī)] = 1 1/6
ma = 210 mi
= 2 1/3 r (Iraq) = [3/10 r (Asyūt.ī)]

Djazīra: Weights
1 dir = 60 h.a 1 dī = 22 1/2 k.ī = 2/45 dī
1 k.ī = 4 h.a = 2/45 dī 1 h.a = 1/4 k.ī = 1/90 dī

Egypt: Measures of area

1 fad = 100 dh · 100 dh = 20 k.ab

Measures of capacity
1 ku = 1 k.in fulfulī = 1/4 k.in (Syria)

Weights
1 dir = 60 h.a 1 da = 6 h.a
1 dir = (1/2 + 1/5) dī = 16 4/5 k.ī 1 k.ī = 3 4/7 h.a
1 dī = 10/7 dir = 24 k.ī = 85 5/7 h.a/dir 1 da/dī = 8 4/7 h.a/dir
1 k.ī = (1/7 + 1/14) da [sic] 1 da = 2 2/5 k.ī
1 dī = 600/7 h.a 1 k.ī = 25/7 h.a = (1/24 + 1/42) dir
1 h.a = (1/5 + 2/25) k.ī 1 h.a/dī = (1/100 + 1/600) dī
1 h.a/dī = [(1 + 1/21) h.a/dī (Aleppo)] 1 k.ī = [15/16 k.ī (Aleppo)]
1 k.ī = [21/25 k.ī (Iraq)] 1 k.ī/dir = [3/8 da (Iraq)]
1 r = 1 k.in fulfulī = 1/4 k.in (Syria)

Filast.īn (Palestine, incl. Tiberias): Measures of length and area

1 h.ab = 40 dh 1 mudy = 1 h.ab · 1 h.ab

Weights
1 ra = 420 mi = 3,600 da = 14,400 k.ī = 36,000 h.a
1 ūk. = 35 mi = 300 da = 1,200 k.ī = 3,000 h.a

Fulfulī: Measures of capacity

1 ra = [7/32 s.a (H.idjāz)] = [7/8 prophetic mu]

Weights
1 r = [7/12 r (Syria, general)]
1 r = 2 r djarwī
1 r = [1/4 r sulaymānī]
1k.in = 1 r (Egypt) =1/4 k.in (Syria)
1 r/ūk. = 12 1/2 dir = (8 1/2 + 1/4) mi = 1/4 r (Syria) =(1/3 + 1/4)
ma
1 r = [6/7 r (Iraq)] = [3/8 r h.aythamī = [5/16 r z.āhirī] =[3/2 r
(Asyūt.ī)]
= [3/2 r (T.ah.āwī)] =[25/64 r (Ant.ākiya]
1 r = 150 dir = 105 mi

Ghaylānī, see Yemen

H.aithamī: Weights
1 r = 400 dir = [12 ūk.] 1 ūk. = 33 1/3 dir
1 r = 2/3 r (Syria) = 2 r laythī = 2 2/3 r fulfulī = 3 1/9 r
(Iraq) = 1 5/9 ma (Iraq) = 1 5/6 r z.āhirī
= 1 11/150 ra (Ant.ākiya)
 Weights and measures in Islam 2265

H.idjāz: Measures of capacity

1 was = 60 s.a = 240 mu 1 mu = 1 3/4 mu (Damascus)
1 s.a = 5 1/3 r (Baghdad) = 1 1/7 r (Syria) = 4 4/7 r fulfulī = 2 2/
3 ma = 8 r (Abū H.anīfa: Baghdad)

Weights
1 dī = 24 k.ī 1 r = 3/601 r sulaymānī

Iraq: Measures of length and area

1 dj = 60 dh · 60 dh = [3,600 dh 2] = 10 k.a = 100 ٬as
1 k.as. = 6 dh = 1 bā = 1/10 ash 1 dj = 1 ash · 1 ash
1 ٬as = 1 bā · 1 bā 1 mud = 30 k.al 2 = 1687 1/2 dh 2

1 k.a = 7 1/2 dh = 1/4 (or 1/3) si

Measures of capacity
1 ku = 60 k.a 1 mu = 3/4 prophetic mu

Weights
1 dī = 6 da = 60 h.a = 20 k.ī 1 k.ī = 3 h.a
1 dir = 48 t. a = 48 h.a = 6 da 1 da/dir = 8 h.a/dir
1 dir = (1/2 + 1/5) dī 1 h.a/dir = 1/48 dir
1 t. a = 4/5 h.a (Syria) 1 dī = 2/3 dī (Aleppo)
1 k.ī = [4/5 k.ī (Aleppo)] 1 k.ī = 1 1/7 + 1/21 k.ī (Damascus, Egypt)
1 da = 2 2/3 k.ī/dir (Dam., Egypt) 1 da = 2 1/3 k.ī
1 k.ī = 8/9 k.ī (Aleppo, weight) 1 r = [3/14 r sulaymānī]
1 ūk. = [1/1,200 k.in (Syria)] 1 r = 1/2 r (Syria, general)
1 r = 128 4/7 dir = 90 mi 1 r = [12 ūk.]
1 ūk. = 10 5/7 dir = 7 1/2 mi
1 r = 1,800 k.ī = 6,171 3/7 h.a/dī = 5,400 h.a/dir = 3/14 r (Syria)
= 6/7 r fulfulī = 1/2 ma = [9/14 r laythī]
1 r = [9/28 r h.aythamī] = [9/35 r (Asyūt.ī)] = [5/42 r (Ardabīl)]

Laythī: Weights
1 r = [1/3 r sulaymānī] 1 r = [7/9 ma (Syria, general)]
1 r = 100 dir = [6 3/33 ūk.] 1 r = [1/2 r h.aythamī]
1 ūk. = 16 1/2 dir = 140 mi = 11 2/3 mi sulaymānī
1 r = 1 1/4 r fulfulī = 6/7 ma = 1 5/9 r (Iraq)

Makāyīl al-nabīy (prophetic measures)

1 mu = 1 1/3 r (Iraq) = 171 3/7 dir = 3 3/7 ūk. (Syria) = 1 1/7 r fulfulī = 120 mi

Sulaymānī: Weights
1 r = 1 r (Damascus) = 200 1/3 r (H.idjāz) = 4 r fulfulī = 4 1/3 r (Iraq) = 3 r laythī = 2 r djarwī = 2 4/19 r (Ghaylānī)
= 1 1/2 r h.aythamī = 1 1/4 r z.āhirī = 5/4 r (Bardha٬ī) = 5/9 r (Ardabīl) = 3/5 r (Asyūt.ī) = 1 1/16 r (Ant.ākiya)
1 r = [2 2/15 ma (Syria, general)] = [7/3 ma (Syria, general)] = [5/7 r (Bardha٬ī)] = [5/9 r (Ardabīl)]

Syria: Measures of capacity

1 ghi = 3 k.a (Iraq) = 12 ru = 72 mu 1 ūk. = [7/24 prophetic mu]
1 r = [7/8 s.a (H.idjāz)] = [4 2/3 r (Baghdad)]
W
Weights
1 dir = 60 h.a 1 h.a = 6 da
1 da = 10 h.a 1 dir = (1/2 + 1/5) dī = 16 4/5 k.ī/dī
1 k.ī = 3 4/7 h.a 1 dī = 1 3/7 dir = 24 k.ī = 85 5/7 h.a
1 da/dī = 8 4/7 h.a/dī 1 k.ī = (1/7 + 1/14) h.a
1 da = 1 4/5 k.ī 1 dī = 600/7 h.a
1 h.a = (1/5 + 2/25) k.ī 1 h.a = 4/5 t. a (Iraq)
2266 Weights and measures in Islam

1 h.a/dī = (1/100 + 1/600) dī 1 h.a = (6/7 + 1/28) h.a (Aleppo)

1 k.ī = 25/7 h.a = (1/28 + 1/42) dir = 1/24 dī

Weights (Syria, specific)

1 r = 420 mi = 3,600 da = 14,400 k.ī = 36,000 h.a
1 ūk. = 35 mi = 300 da = 1,200 k.ī = 3,000 h.a
1 k.in = [4 k.in fulfulī] = [4 r (Egypt)] = 42,000 mi = 60,000 dir
1 r = [4 r/ūk. fulfulī] = [2 r djarwī] = [14/3 r (Iraq)]
1 r = [3/2 r h.aythamī] = [3/5 r (Asyūt.ī)] = [16/25 r (Ant.ākiya)]

Weights (Syria, unspecified)

1 k.in = 100 r = 233 1/3 ma = 466 2/3 r [sic] = 1,200 ūk. (Iraq)
1 ma = 260 dir ≈ 257 1/7 dir = 180 mi = 2 r (Iraq) = 1 5/7 r fulfulī = 13/30 r sulaymānī
1 ma = 1 2/7 r laythī = 6/7 r djarwī = 3/7 r sulaymānī
1 r/ūk. = 5 1/7 ūk. = 1/4 ma = 3 3/4 mi

T.ah.āwī: Weights
1 r = [3/10 r djarwī ] 1 r = 1 r (Asyūt.ī)

Yemen (Ghaylānī): Weights

1 r = [19/42 r sulaymānī ] 1 ūk. = (15 1/2 + 1/3) mi sulaymānī
1 r = [271 2/3 + 2/7] dir sulaymānī = 190 mi sulaymānī

Z.āhirī (Fāt.imid): Weights

1 r = 480 dir = [12 ūk.] = 336 mi 1 ūk. = 40 dir = 28 mi
1 r = 4/5 r (Syria) = 1 1/5 r h.aythamī = 3 1/5 r fulfulī = 1 1/4 r
(Ant.ākiya)
= (3 2/3 + 2/30) r (Iraq)

Ells (adhru٬, unspecified):

dhirā٬ al-yad
1 dh = 2 shi (cloth) = 18 is. (medium) = 24 is. (without thumb) = [3/4 dh (k.āsimī)]

dhirā٬ k.āsimī (= hāshimī)

1 dh = 1 1/3 dh (yad) = 24 is. = 6 k.ab 1 k.ab = 1/6 dh
1 Is. = 1/4 k.ab = 1/24 dh

dhirā٬ hāshimī (mālikī)

1 dh = 8 k.ab = [3/5 dh (mābahrāmī)]

dhirā٬ mābahrāmī (sūd)

1 dh = 1 2/3 dh (hāshimī) = 60 fa = 1/3 (or 1/6) k.ab
1 dj = 36 dh · 36 dh

al-islāmīya, Beirut: Maktabat Lubnān Nāshirūn, 2002.
Only occasionally, external sources, records of both
European and Oriental travellers, allow the fixation of
absolute values. With regard to their geographical and
chronological diversity, the reconstruction of entire
systems of measurement and of their relation to each
other has just begun.
During the tenth century, mathematicians became
aware of the complexity of the metric systems in use.
Their particular perspective on the issue differed from
what their legal and other colleagues had been noting
down hitherto. They not only tried to present a systemized
enumeration of all units related, but also endeavoured,
for pedagogical reasons, to systematize the usage of
measuring units in popular treatises. They sometimes
even expressed efforts to standardise and facilitate the
conversions customs, market and tax officers had to
enact. Thus, stimulated by theoretical manuals, mathe-
matically standardised methods of measuring and of
converting quantities from one system into another
became popularised. To some extent, these devices even
had retroactive effects on the practice of Islamic laws.
Among these manuals, the “Book on the Basis of
Arithmetic and the Division of Inheritances” (Kitāb
 Weights and measures in Japan 2267

al-Tadhkira bi-us.ūl al-h.isab wa l-farā’id. ), written

by the Damascene mathematician ٬Alī b. al-Khid. r
al-٬Uthmānī al-urashī (1030–1067 AD) contains the
most coherent information on the metric systems that
were in use in the Islamic Middle East up to the lifetime
of the author. The following list of units of weight,
currency, capacity and length, together with the
localities they refer to arranged in alphabetical order,
is extracted and, if indicated by ‘[…]’, concluded from
different chapters of this Kitāb al-Tadhkira.

See also: ▶Nilometer

References
Cahen, C. Quelques Problèmes Économiques et Fiscaux de
l’Iraq Buyide D’après un Traité Mathématique. Annales de
L’Institut d’Études Orientales 10 (1952): 326–363.
Ehrenkreutz, A. S. The Kurr System in Medieval Iraq.
Journal of the Economic and Social History of the Orient
5 (1962): 305–314.
Encyclopaedia of Islam, New edn. (= EI 2), Vol. I-XI, Leiden:
E.J. Brill 1986ff. Entries: misāh.a, mīzān, s.ā٬, zakāt.
Hinz, W. Islamische Masse und Gewichte. Umgerechnet ins
Metrische System. Leiden: E.J. Brill, 1955.
Hinz, W. Islamische Währungen des 11. bis 19. Jahrhunderts.
Umgerechnet in Gold. Ein Beitrag zur islamischen
Wirtschaftsgeschichte. Wiesbaden: Otto Harrassowitz, 1991.
Ibn Manthū Lisān al-٬arab, I-VI [15 books in 6 vols.]. Beirut:
Dār S.ādir, n.d.
Miles, G. C. Early Arabic Glass Weights and Stamps. New
York: American Numismatic Society, 1948.
al-Qurašī, ٬Alī Ibn al-Hid. r at-Tadkira bi-us.ūl al-h.isāb wa
l-farā’id.. (Buch über˘ die Grundlagen der Arithmetik und
der Erbteilung). übersetzt, kommentiert und in Faksimile
herausgegeben von Ulrich Rebstock. Frankfurt am Main:
Institute for the History of Arabic-Islamic Science, 2001.
[Islamic Mathematics and Astronomy, Vol. 107].
Rebstock, U. Rechnen im Islamischen Orient. Darmstadt:
Wissenschaftliche Buchgesellschaft, 1992.

Weights and Measures in Japan. Fig. 1 Linear Measure.

Weights and Measures in Japan Drawing by The Ministry of Finance. Doryoko Shurui Hyo
(The Classification Table of Weights and Measures) Genbei
Kinokuniya, 1875. Chos 1, 4–6, 9–11.
S HIGEO I WATA

Japan used the 17.3 cm-long linear measure unit that was
common to all the regions in East Asia for a period of
25,000 years. Under the influence of China, the
Japanese measuring unit gradually lengthened during
the period from the end of the eleventh century BCE to
the middle of the third century BCE. Then, the length of
the Japanese measure shaku was stabilized at 23 cm and
remained unchanged until the end of the second century
AD. Various civil disturbances in China had the effect of
lengthening the linear measure substantially to 29 cm
until the middle of the seventh century, and no more
significant variation has since been observed (Fig. 1).
Length
Under the Chinese Tang dynasty, a law was enacted
mandating the use of two methods based on large and
small linear measures. The large scale was 1.2 times as
long as the small scale, which we refer to as the ancient W
linear measure. The small scale was used for music,
astronomy, and ceremonial items.
Japan introduced this Chinese measuring system in
701. The large scale later became known as kanejaku,
which refers to an L-shaped ruler used by architects.
The small scale gradually dropped out of use. Linear
measurement tools were mainly made of wood, though
2268 Weights and measures in Japan

Weights and Measures in Japan. Fig. 2 Land Surveying. From Tokugawa Bakufu Kenchi Yoryaku (Essential Summary of
the Government of Prefectures, Tokugawa Shogunate). Tokyo: Kashiwa Shobo, 1915.

ivory was also used in ancient times. The largest unit of

length in ancient times was the ri, which was 5–70 cho
(a cho was 109.1 m). The ri was set to 36 cho in the
seventeenth century. One shaku was one-tenth of a jo.
Metal linear measurement tools began appearing in the
medieval era.
The average value of one shaku was 29.6 cm in the
seventh century, 29.7 cm in the eighth century, 29.8 cm
in the ninth century, 29.9 cm in the tenth century,
30.1 cm in the twelfth century, 30.2 cm in the fourteenth
century, and 30.3 cm in the eighteenth century. The rate
of elongation was slower than in China.
Several linear measures for sewing appeared in the
medieval era. The length of one shaku ranged from 1.15
to 1.27 times that of a kanejaku. These linear measures
for sewing were neatly arranged into two kinds of scales. Weights and Measures in Japan. Fig. 3 Measures.
Drawing by The Ministry of Finance. Doryoko Shurui Hyo
One scale, 1.20 times as long as a kanejaku, was called
(The Classification Table of Weights and Measures) Genbei
gofukujaku, while the other was 1.25 times as long as the Kinokuniya, 1875. Chos 1, 4–6, 9–11.
kanejaku and called a kujirajaku. The gofukujaku gra-
dually dropped out of use in the first half of the
eighteenth century and was finally abolished in 1875. bu, which was also referred to as a tsubo. These names
appeared in the medieval era. Land surveyors made
Area their measurements using flax string measures and
The earliest area unit we know of was the shiro, which wood measuring poles (Fig. 2, 3).
was in use before the sixth century. One shiro was 30
shaku square. In 646, new units were defined: the bu, Volume
which was six shaku square; the tan, equal to 360 bu; The names for most units of volume also came from
and the cho, equal to ten tan. One tan has been equal to China. The basic unit was the sho. One sho averaged
300 bu since the end of the sixteenth century. For land 750 cm3 in the eighth century, 800 cm3 in the ninth
surveying purposes, one bu was defined as six shaku century, 970 cm3 in the eleventh century, 1,400 cm3 in
five sun square, rather than six shaku square. At the end the fifteenth century, and 1,650 cm3 in the sixteenth
of the sixteenth century, a bu was defined as six shaku century. Since the seventeenth century, the average
three sun square. Despite this, the bu was commonly value has remained constant at 1,804 cm3. During the
considered to be six shaku square, and was finally medieval era, different measures of volume were used
defined as such in 1891. A se was ten times as large as a for goods, taxes, and financial transactions (Fig. 4).

Weights and Measures in Japan. Fig. 4 Steelyard.
Drawing by The Ministry of Finance. Doryoko Shurui Hyo
(The Classification Table of Weights and Measures) Genbei
Kinokuniya, 1875. Chos 1, 4–6, 9–11.
The Tokugawa-Shogunate (1603–1867) established
the East and West Measure Guilds in 1655, and exerted
strong control over the new standard. Use of sho con-
tinued until 1958, when the shaku-kan system was
abolished.
Rectangular measures were made of wood. The
brims of many dry measures were protected by bamboo
or iron laminae. Most of the large dry measures also
featured a narrow iron bar positioned diagonally across
the top. Round wooden sticks were used to level off the
tops of measures. Liquid measures had lacquered
surfaces to prevent the contents from leaking.
Standard sizes for measures have been one to, seven
sho, five sho, one sho, five go, two go five shaku, and
one go since 1655, and one to, five sho, one sho, five go,
two go five shaku, one go, and five shaku since 1875.
Mass
Between the eighth century BCE and the fifth century
AD, the Chinese expressed mass in terms of the liang,
which was equal to 14 g. In this measuring system,
one jin was equal to 16 liang, and one liang to 24 zhu.
The civil disturbances of the third and seventh centuries
also had their effect on the liang, which tripled in mass.
Under the Chinese Tang dynasty, a law stipulating two
measuring scales was enacted, which fixed the mass of
the new large liang at three times that of the mass of the
older liang, which became used only for medicine.
Japan adopted this Chinese measuring system as well,
changing the names of jin, liang, and zhu to the kin, ryo,
and shu, but retaining the same characters. The absolute
mass of the ryo in the large scale decreased year by year,
while that in the small scale became rather unclear. The
average mass of one ryo was 42 g in the eighth century,
40 g in the tenth century, 39 g in the twelfth century, and
37.47 g at the end of the sixteenth century. Under the
Weights and measures in Japan 2269
Weights and Measures in Japan. Fig. 5 Balance. Drawing
by Shigeo Iwata (1981).
strong control of the Tokugawa-Shogunate, the average
mass of the ryo remained almost unchanged until the end
of the nineteenth century. The kan and monme appeared
in the medieval era. One kan was equal to 100 ryo, and
one monme equal to one-tenth of a ryo. These units
gradually became standard.
Chinese use of the steelyard dates back to at least
several centuries BCE. The steelyard was also used
in Japan, where the Tokugawa-Shogunate established
the East and West Steelyard Guilds in 1653. Steelyard
beams were made of white oak, persimmon wood, ivory,
horn, and bone. The beam had from one fulcrum to eight
fulcra, which were made using string, cord, or rope.
A copper pan or bronze hook was installed on the end
of the beam, and the bob-weight was made of bronze or
iron. Steelyards were used for most commercial transac-
tions (Fig. 5).
Balance makers did not belong to a guild, so
standards for making balances were never formulated.
The most famous balance makers were members of the
Nakabori family, who lived mainly in the western
counties of Japan. The balance beam was suspended by
its pointer from the center of a box-mounted rectangu-
lar wooden stand. A central pin was used for the main
fulcrum, and during the final stages of weighing, W
bearing friction was overcome by tapping on the pin
with a small wooden hammer. Balances were mainly
used in the manufacture of gold, silver, and copper
coins, and in money-changing transactions.
Beginning in the seventeenth century, users of the
balance had to be supplied with weights made by the
Goto family. Two sets of weights were available: one
consisted of weights of 20, 10, 5, 4, 3, 2, 1 ryo, 5, 4, 3,
2270 Weights and measures in Japan

2, 1 monme, and 5, 4, 3, 2, 1 fun, and the other was
identical to the first but for the addition of 50 and 30
ryo weights. Weights were made of bronze, and
engraved on the surface with the inspector’s mark
and maker’s crest, leaves and flowers of the Japanese
paulownia. The Goto family belonged to the Guild of
Goldsmiths, and were inspectors authorized by the
feudal government. They examined user’s weights
periodically by comparing them with standard weights.
Transition to the Metric System
In 1886, Japan signed the Treaty of the Meter, and
promulgated the Law of Weights and Measures. As a
result, the shaku-kan system and metric system began
to be used jointly in 1891. The conversion coefficients
between these two measuring systems were also fixed
at the same time (Table 1). The law went into formal
effect in 1893. In 1903, the Central Inspection Institute
of Weights and Measures was established in Tokyo.
Japan also adopted the units of the foot-pound system
as legal in 1909.
A Measurement Law was passed by the Diet in 1951,
and went into effect on March 1, 1952. It mandated that
measuring units be based on the metric system. The
shaku-kan system and foot-pound system were abol-
ished in 1958.
A new Measurement Law was promulgated on May
20, 1993, and went into effect on November 1, 1993.
The law made a sweeping revision of the regulations by
adopting the Système Internationale d’Unités. The
units of the other measuring systems on commercial
transactions and certifications were abolished by the
end of the twentieth century.

Weights and Measures in Japan. Table 1 Conversion tables of Japanese weights and measures (1891–1958)

Length
ri cho ken shaku sun bu rin mo Metric system

1 36 2,160 12,960 129,600 1,296,000 12,960,000 129,600,000 3.927 km

1 60 360 3,600 36,000 360,000 3,600,000 109.1 m
1 6 60 600 6,000 60,000 1.818 m
1 10 100 1,000 10,000 30.30 cm
1 10 100 1,000 3.030 cm
1 10 100 0.303 cm
1 10 0.0303 cm
1 0.00303 cm

Area
cho tan se bu (tsubo) go shaku Metric system

1 10 100 3,000 30,000 300,000 9917 m2

1 10 300 3,000 30,000 991.7 m2
1 30 300 3,000 99.17 m2
1 10 100 3.306 m2
1 10 0.331 m2
1 0.0331 m2
Volume
koku to sho go shaku Metric system

1 10 100 1,000 10,000 180,400 cm3

1 10 100 1,000 18,040 cm3
1 10 100 1804 cm3
1 10 180.4 cm3
1 18.04 cm3
Mass
kan ryo monme fun rin mo Metric system

1 100 1,000 10,000 100,000 1,000,000 3.750 kg

1 10 100 1,000 10,000 37.50 g
1 10 100 1,000 3.750 g
1 10 100 0.375 g
1 10 0.0375 g
1 0.00375 g

References
Iwata, Shigeo. Changes in Mass Standards in Modern Japan.
Bulletin of the Society of Historical Metrology, Japan 1.1
(1979): 5–9. Table 1–7+Fig. 1.
---. Japanese Scales and Weights. Equilibrium. Spring (1981):
319–26.
---. Changes in Linear Measures in China and Japan. In Acta
Metrologiae Historicae I. Ed. Gustav Otruba. Linz: Rudolf
Trauner Verlag, 1985. 117–37, 488–9.
---. Changes in the Chinese Standard Mass Unit. Acta
Metrologiae Historicae II. Ed. Harald Witthöft and
Cornelius Neutsch. Linz: Rudolf Trauner Verlag, 1989.
117–29.
---. Changes in the Japanese Weights and Measures Standards
under the Influence of China. The 15th Symposium
on Historical Metrology. Japan, November 6, 1992. 1–17.
---. The Length Standards in China, Korea, and Japan, (1) 300
BC–1700 AD. Bulletin of the Society of Historical
Metrology, Japan 17.1 (1995): 53–65.
---. The Length Standards in China, Korea, and Japan, (2)
5000–300 BC. Bulletin of the Society of Historical
Metrology, Japan 17.1 (1995): 53–65.
Weights and Measures
in Mesoamerica
J EAN -C LAUDE H OCQUET
Mexico is said to have used an incredible variety of
weights and measures changing according to the
region and the period. Different uses of the weights and
measures increased the complexity of the situation. The
task of the historian consists of restoring the ancient
systems and understanding their use by posing the
questions: who was using the weights and measures and
for what purpose? The more powerful of the two parties in
the exchange would impose upon the other modes of
measuring and counting to his own advantage.
Indian Measures
In 1520, Hernan Cortes had the opportunity to see the
market of Tlalecolco and the transactions which went
on there. He wrote, “Everything is sold by count and
measure, and up till now I haven’t seen anything sold by
weight.” Antonio de Herrera y Tordessilas reported that
the Indians “have measures for everything” and that the
judges controlled the weights and measures and punished
the cheaters. But his information has not been confirmed.
However, the level of development attained by the
Mesoamerican people before the Conquest makes it very
probable that they had an ensemble of measures and
standards used in the apportionment of land, commercial
transactions, levying of taxes, construction, pharmacy,
etc., and that these measures constituted a unified and
coherent system.
Weights and measures in Mesoamerica 2271
After the Conquest, the Spanish imposed Andalusian
and Castilian measures such as brazas, codos, and
pal-mos, that the Indians adopted easily, as, like their
own, they were anthropometric. The result of this
adoption was that Mesoamerican symbols of measure
were transcribed in European terms on the pictograms
of indigenous manuscripts written in the Nahuatl
language. Bars represented digits, little black circles
were for 20s, and other signs represented hands or
feet. Thus, for example, the Indians gave the name
tlamamalli to the load (carga) that a man (porter,
tameme) could carry on his shoulders during a day’s
work. In the novohispano system, this load had the
name tameme, by a simple movement of language and
meaning. It was equivalent to a Spanish half-fanega or
to two arrobas. This indigenous load varied in weight,
in the function of the weighed merchandise, and with
the strength of the carrier; for cocoa it was equivalent to
three xiquipilli of 8,000 grains each.
Among the measures of capacity, of great diversity,
the smallest was the centlachipinilli, which literally
means “a small something” (una gota de algo) and
which was itself a multiple of centlachipiniltontli
(una gotilla de algo). There were also small measures
used in pharmacy, such as spoonsful (cemixcolli,
cenxumatli, etc.). Testal designated the quantity of
cornmeal necessary to make a tortilla for one person.
To measure dry materials, in particular for grains, they
were often able to use measures equivalent to those
imported from Spain: the acalli, for example, of which
we do not know the weight, was part of an arithmetic
system with the cencaauhacalli equal to a half-fanega
and with the cuauha-caltontili equal to a celemin. Such
equivalent approaches make it possible to conclude that
the conquerors tried to impose their own system while at
the same time adopting the native terms.
Measures of length were used for the preparation of
clothing, especially for ponchos, made of bands of
different widths. Most measures of length were land
measures, based on proportions of the human body:
cemizteltl, cemmapilli, and cemmatl, literally a finger, a
palm, and a span. This “span” designated the height of a
man, from foot to raised hand, but it also measured the
diagonal, from the left foot to the raised right hand, and
it stood for three Spanish varas (yards or cubits). Such
a dimension proves that the span was a geometric
measure and not an anthropometric one; it was obtained
by attaching an anthropometric measure such as a digit,
measured by its length, to a coefficient. European W
measures of length did not have the same way of
varying, in which a “foot” could vary between 28 and
60 cm. However, for land measures, certain spans were
smaller, and the vara was therefore equal to two spans.
In Tula, there was a division of land in order to reform
the tax law, and “each Indian was given one hundred
varas in length and twenty in width. Each one of these
varas made two spans; this is the measure which the
2272 Weights and measures in Mesoamerica
Indians use” (Castillo Farreras 1972). It is possible that
this text makes reference to an Indian vara and a
Spanish braza. Standards of measure and length existed,
and, to measure land, poles or cords were used, just as
European surveyors did.
Road measures also existed, which were used by
travelers, soldiers, and merchants. These were the
cennecehuilli and the cennetlalolli, which Spanish
authors tended to translate as “leagues.”
Weights and Measures of New Spain
In the sources of the eighteenth century the range of
different names involved – carga, fanega, and arroba –
hides the numerical relations which link multiples and
submultiples. The carga is made up of two fanegas,
composed in turn of six to eight arrobas. In general the
historian goes from the largest to the smallest, whereas
in practice, the peasant, the muleteer, or the trader
proceeds in the reverse order. In the same way that the
metric quintal is made up of grams and kilograms
between which the factor 10 governs all relations, in the
eighteenth century, 2 and 6 (equals two multiplied by
three) governed the relations of the carga with its
submultiples. The decimal metric system is a scientific
system, mathematic in character, invariant, exterior to
humans. In contrast, previous systems were based on
people or on observations concerning them, or on their
work and capacity to complete certain tasks with the aid
of animals. That is, if a mule must climb steep paths
with its load, the muleteer compensates for this by
lowering the weight of the load by one or two arrobas.
At times, the sources mention cargas pequeñas in
contrast to cargas regulares de mula.
As long as freighting relied chiefly on pack trains
and carts, the weight of the loads was determined
by distances to the customer, the availability of
pasture en route, as well as by the nature of roads
and tracks. In the case of steep mountain paths,
broken country or long distances, the carga, the
traditional load of a mule, could be lowered in
weight. So a mule load could comprise one and a
half or two fanegas. In 1832 the Governor of
Oaxaca stated that because of the poor roads the
carga could never consist of more than eight
arroads (Ewald 1985).
Generally the custom of mentioning the number of
arrobas making up a fanega or a carga does not elimi-
nate the uncertainties which surround all quantitative
data. In 1751, the carga of salt in Tehuantepec consisted
of 14 arrobas. Another eighteenth-century source stated
that moist Tehuantepec salt was 16 arrobas, while drier
salt was computed at 12–14 arrobas.
The dampness or dryness of goods being transported
markedly affects their weight. The presence of water
increases the weight of light goods such as grains, but
it lessens the weight of heavier ones, in particular
salt. In Tehuantepec, the variation in weight often
reached a quarter. Other variations were also noticeable
between the different regions: “In eighteenth century
Yucatan, one fanega of old salt weighed approximately
9 arrobas, one fanega of new salt 10–11 arrobas”
(Ewald 1985). The measure known as carga in the
valley of Mexico was often mistakenly referred to as
fanega in Yucatan.
But there is another trap: Ralph Roys, for example,
noted that a fanega “is variously defined as 1.6 bushels,
and a load of 100 pounds of grain,” which is an
example of recent efforts to align traditional measure-
ments with those of the national market governed by
the decimal metric system, whilst at the same time
conserving the ancient names. But “this statement has
led a number of scholars to assume that a fanega
weighs 100 pounds, which is not necessarily the case.
The fanega, like the bushel, is a measure of capacity,
not weight. Hence, the weight of a fanega will vary
enormously depending on the goods being measured
by this standard. A fanega of corn may weigh close to
100 pounds, but salt is much heavier. An extensive
survey of the literature on weights and measures used
in the Maya area – from Yucatan to El Salvador sug-
gests that a fanega of salt commonly weighed about
115 kg” (Andrews 1983). It is important not to confuse
measures of capacity, which are volumes, and units of
weight which measure mass; weighing the contents of
a volumetric measure can only lead to considerable
distortions, since each body has a specific weight by
which it distinguishes itself from all others.
The foreigners who traveled or worked in nineteenth-
century Mexico reported that a fanega of salt weighed
approximately 70 kg and a carga 140 kg. J. Buschmann
stated that the carga consisted of 138 kg. In 1912–
1913, however, the business records of Salinas, Mexico
calculated the carga of Colima salt at 161 kg. In 1916–
1917, the carga was again lowered to the more cus-
tomary 140 kg. The nineteenth-century sources gave
300 pounds as the equivalent of a carga. In summary,
here are some of the results found in the historical
literature on the subject.
In Mexico, on the central plateau (Tehuantepec salt):
carga 1
fanega 2 1
arroba, dry salt 14 7 1
arroba, moist salt 16 8 1
pounds, dry salt 300 150 21.4
kg 138 69
In Yucatan, the substitution of the fanega for the carga
does not modify the system.

β 1
arroba, old salt 9 1
arroba, new 11
salt
pounds 300 33.33 1 Peruvians chewed lime and lipta with coca leaves as
kg 138 0.460
The historian will recognize the mule load of 300
pounds and the 16 ounce pound, such familiar territory
that he would think he was in Castile in the time of the
Catholic Kings.
References
Alonso, M. M. Medidas Indígenas de Longitud. Mexico:
CIESAS, 1984.
Andrews, Anthony P. Maya Salt Production and Trade.
Tucson: University of Arizona Press, 1983.
Castillo Farreras, V. Unidades Nahuas de Medida. Estudios
de Cultura Nahuatl. Vol. 10. Mexico City: Instituto de
Investigaciones Historicas, Universidad Nacional Autonoma
de Mexico, 1972. 195–223.
Ewald, Ursula. The Mexican Salt Industry. 1560–1980.
A Study in Change. Stuttgart: G. Fischer, 1985.
Garcia Acosta, Virginia. Los Precios del Trigo en la Historia
Colonial de Mexico. Mexico DF: Centro de Investigaciones
y Estudios Superiores en Antropologia Social, 1988.
Gual Camarena, Miguel. El Primer Manual Hispanico de
Mercaderia (SigloXIV). Barcelona: Consejo Superior de
Investigaciones Cientificas, 1981.
Hocquet, Jean Claude. Anciens Systèmes de Poids et Mesures
en Occident. London: Variorum Reprints, 1992.
---. La Métrologie Historique. Paris: Presses universitaires de
France, 1995.
Roys, Ralph L. The Political Geography of the Yucatan
Maya. Washington, District of Columbia: Carnegie Insti-
tution of Washington, 1957.
Villena, Leonardo. Weights and Measures in Islamic Spain.
Acta Metrologiae Historicae (1985): 298–303.
Weights and Measures in Peru
S HIGEO I WATA
Ancient Peru covered most of the Andes, where the
Peruvian culture lasted for more than 10,000 years.
Because most of the region was in the highlands, more
than 3,000 m high, their metrology was slightly different
from other civilizations. Also, they had no form of
writing until 1532. For calculation and for recording
metrological units decimally, they used quipus, spatial
arrays of multicolored knotted cords. The names of the
units and the values differed significantly depending on
the period, the people, and the region.
Weights and measures in Peru 2273
Length
Peruvians used linear measurements that were related
to human measurement patterns. Examples are listed in
Table 1.
a stimulant. They invented a measure called cocada,
which measures how far one can walk while under the
drug’s influence. This distance is 3 km on level ground
or 2 km on an uphill slope when carrying a 45 kg load
for 35–40 min in Pataz county.
A twelfth-century pre-Inca site in Puruchuco, located
about 6 km from Lima, was recently restored. The
linear measurements of each part of the structure
showed that the average stature of men from head to toe
was 1.701 m, while the height from the middle fingertip
to one’s toe was 2.106 m. They used a geometrical
design of proportions based on the golden section
(1:1.618) between the sides of the rectangles.
Area
The Aymara called the area where they could raise
one or two heads of cattle the callapa. Tupu was the
Incan name for an area of cultivated land or pasture
where a married couple could live. The Inca called
the area, which was equal to 625 m2 huiri. This was
the area a man could cultivate in 1 day. The Cuzco
called the area equal to 282 m2 of sweet potatoes or
corn papacancha.
Volume
The Peruvians used a single hand-cupped gowpen to
measure small portions of grain. They also used gourds,
dried pumpkins, pots, and jars to measure things. The
names of the units and the values are shown in Table 2.
The Quechua called a broad crate filled with coca or
red pepper a runcu. One-half of a runcu was called
checta runcu, one-fourth was called cutmu, and one-
eighth was called sillcu.
Mass
Because fewer weights have been discovered in the
Peruvian region than in other civilizations, it has been
difficult to estimate the mass standard accurately. We
assume that the average mass of unit was 23.1 g. The
notation was based on the decimal system. The
maximum unit was assumed to be 23.1 kg, which is
about 80% of the maximum units from other civilizations,
27–30 kg in Egypt, Mesopotamia, Indus, and China. W
This is probably because of the low concentration of
oxygen in the atmosphere in the Andean highlands.
Thirty-nine percent of the weights were made of
stone, 32% of iron, 24% of lead, and 5% were made
of other nonferrous metals. One-third of the weights
were globe-shaped; the others were conical, cylindrical,
or in spindle and other irregular forms.
2274 Weights and measures in Peru

Weights and Measures in Peru. Table 1 Examples of Peruvian linear measurements

Quechua Aymara Inca Ratio Metric system

Handbreadth ttkhlli 0.051 8.5 cm

Link yaku vicu 0.107 18 cm
Span capa chia 0.125 21 cm
Cubit cuchuch tupu 0.25 42 cm
Yard sikya 0.50 84 cm
Pace chillque 0.89 150 cm
Fathom rikra 1.00 168 cm
loca 102 168 m
ecca 103 1.68 km
cocoda 1.19–1.79 × 103 2–3 km
League tupu tupu 3.33 × 103 5.6 km
yapu 5.00 × 103 8.4 km
chuta (sayhua) 104 16.8 km
guamanin 105 168 km

Weights and Measures in Peru. Table 2 Hand-cupped

“gowpen” measure of the Peruvians

Quechua Aymara Ratio Metric

system
(cm3)

Gowpen poktoy iuu 0.043 300

(double
hand-
cupped)
laqui 1.00 7,000
pokcha hullu 2.00 14,000

Weights and Measures in Peru. Fig. 2 Movable load scale

(adapted by the author from P. Rivet. La Balance Romaine
au Pérou. L’Anthropologie 33 (1923): 535).

All types of scales, namely balance, steelyard, bismar,

Weights and Measures in Peru. Fig. 1 Bismar (adapted by
the author from M. Uhle. La Balance Romaine au Pérou.
Journal de la Société des Américanistes de Paris 17
(1925): 335).
and movable load, were discovered in the northwest
region of South America. The fulcrum was the cord-pivot
type. Balance beams were made of wood, horn, and
bone with humans, monkeys, birds, and geometrical
patterns engraved on them. The two items supporting
the load and weights were mostly nets, and partly pans
of metals. The balance might have been invented by the
ninth century (Figs. 1 and 2).

Weights and Measures in Peru. Fig. 3 Balance (from
the Museo Amano).
Miquel de Estete, who attended Francisco Pizarro,
found a steelyard on the coast of Ecuador in 1531. The
steelyard was graduated from the middle of the beam to
the end, and had a bob-weight suspended from the arm.
The steelyard was used to weigh gold and silver. The
bismar was discovered at the market in Tarma, 170 km
northeast of Lima. The total length was 27.8 cm, and the
capacity was assumed to be 924 g. The movable load
scale, called wipe in the actual place, was discovered in
Huarochiri, 90 km to the east of Lima. The scale was made
of wood, and was used to weigh coca leaves (Fig. 3).
References
Iwata, Shigeo. Ancient Peruvian Mass Standard and Scales.
Bulletin of the Society of Historical Metrology, Japan 7.1
(1985): 23–33.
Nordenskiöld, Erland. Emploi de la Balance Romaine en
Amérique du Sud avant la Conquête. Journal de la Société
des Américanistes de Paris nouvelle série 13 (1921): 169–71.
Rivet, P. La Balance Romaine au Pérou. l’Anthropologie 33
(1923): 535–8.
Rostworowski de Diez Canseco, Maria. Pesos y Medidas en el
Perú Pre-Hispánico. Lima: Imprenta Minerva, 1960. 1–27.
---. Mediciones y Computos en el Antiguo Peru. Cuadernos
Prehispanicos 6 (1978): 21–48.
Uhle, Max. La Balance Romaine au Pérou. Journal de la Société
des Américanistes de Paris nouvelle série 17 (1925): 335–8.
Wakeham Dasso, Roberto S. Puruchuco, Investigacion Arqui-
tectonica. Lima: Universidad Nacional De Ingenieria, 1976.
1–25 + 6.
Western dominance 2275
Western Dominance
M ICHAEL A DAS
Scientific curiosity was a major motive behind Western
overseas expansion from the fifteenth century onward
and technological innovations, particularly in ship-
building, navigational instruments, and firearms, made
that expansion possible. But early European explorers
and conquistadors did not rely heavily on evidence of
scientific or technological achievements as gauges of
the worth of the peoples they encountered or as explana-
tions for their growing dominance in the Americas and
maritime Africa and Asia. In encounters with the great
centers of civilization in Africa and Asia, European
superiority in these endeavors was highly selective,
marginal, or in many areas nonexistent. In fact, tra-
velers to China and the Indian subcontinent in the early
centuries of expansion were as likely to dwell on the
technological deficiencies of the West, when compared
to these great civilizations, as to boast of European
advantages. As in India, China, and Japan, the
Europeans were able to make little headway into the
heartlands of the Islamic world in this era. That standoff
and the fact that they had borrowed so heavily from the
scientific learning and technology of Muslim cultures
in the centuries of Europe’s emergence as a global
force, rendered it unlikely that material standards
would supplant the long contested religious differences
that the Europeans had employed to set themselves off
from and above the followers of Islam.
In Africa, disease and geographical barriers and the
power of coastal kingdoms prevented the Europeans from
translating their technological edge into significant
conquests. Failure to move into the African interior also
meant that the Europeans had only the vaguest notions
about African epistemologies or understandings of the
natural world, which were usually dismissed as super-
stition or fetishism. In sharp contrast to their experience
in Africa, European invaders encountered few disease
barriers in the Americas. In fact, diseases from smallpox
to the measles became powerful allies of the Spanish
conquistadors in their assaults on the heavily populated
and highly advanced civilizations of Mesoamerica and
the Andean highlands. The long isolation of the
Amerindian peoples from the Afroeurasian people and
cultures left them highly vulnerable to both the microbes
borne and the Iron-Age technology wielded by the
W
European invaders. Nonetheless, the Spanish tended
to attribute their startling successes in battle against
seemingly overwhelming numbers of Aztec or Inca
adversaries and the rapid conquests that followed to
supernatural forces and to the superiority of their
2276 Western dominance
militant brand of Christianity over the “heathen” faiths
of the indigenous inhabitants of the “New” World.
Lacking in-depth knowledge of the epistemologies
and scientific learning of most of the peoples they
encountered in the early centuries of overseas expan-
sion and often enjoying only very selective (but at
times critical) technological advantages over them,
European explorers, missionaries, and Crown or
Company officials were unlikely to rely on material
standards to judge the level of development attained by
other cultures or to compare overseas civilizations to
Europe itself. Until at least the end of the seventeenth
century, religious beliefs, or the Europeans’ certitude
that they possessed vastly superior understandings of
the transcendent world, predominated as the gauge by
which other cultures and peoples were assessed and
ranked. Additional cultural variables, such as the
position and treatment of women, were frequently
cited as evidence of advancement or savagery; and
physical features, especially skin color, were some-
times emphasized in attempts to distinguish and rank
the peoples encountered overseas.
Scientific and technological gauges of past attainments
and present abilities remained peripheral to most
evaluations of the peoples and cultures that the Europeans
encountered as they expanded across the globe. None the
less, signs of material advancement – the existence of
large cities, sophisticated techniques of fortress construc-
tion, or evidence of complex scientific instruments –
were often noted and even cited to support arguments
regarding the level of development achieved by different
peoples. In two areas in particular, in the perception and
measurement of time and in perspectives on space
associated with artistic and mathematical advances of the
Italian Renaissance, European overseas observers began
to see a clear divide between the West and all other
civilizations and cultures. In addition, as early as the
sixteenth century, European commentators began to rank
African cultures beneath those of Asia and the Americas,
not so much because of skin color or other physical
features, as has often been argued, but due to what was
perceived as a markedly lower level of material culture in
Africa than that found by European travelers and traders
in India, China, or central Mexico.
Although ideologies justifying overseas expansion
and the domination of non-European peoples from the
fourteenth to the early eighteenth century were rooted
in religious belief and were generally culture – rather
than racially – oriented, material accomplishment,
including the assumed capacity for invention and
scientific thinking, was increasingly associated with
racist defenses of the enslavement of Africans.
Defenders of the slave trade sought to counter the
abolitionists’ objections with often lurid descriptions of
the alleged savagery of African life and the debased
level of African material culture. Along with skin color
and other physical differences, racist writers, such as
Samuel Estwick and Dominique Lamiral, emphasized
material backwardness and ignorance of the workings
of the natural world as proof of the subhuman nature of
Africans that justified their subjugation as slaves.
By the last decades of the eighteenth century, this
rather broad association between racial ideology,
material culture, and the defense of slavery was refined
and enhanced by the rise of racist theories allegedly
grounded in scientific experimentation and reasoning.
Physicians and ethnologists devised a variety of
measurements – from skull size and shape to genital
configurations – in efforts to prove that there were
innate physical, mental, and moral differences between
human racial groups. The fact that the measurements
reflected a priori assumptions and were based on small
and suspect samples, and that even the racial categories
themselves were hotly contested, did not prevent
“scientific” racism from winning widespread support
from European scientists, social commentators, and
politicians throughout the nineteenth century. The
tenets of scientific racism were popularized among
the middle and working classes by practitioners of
phrenology, whose booths could be found at county
fairs and on the promenades of seaside resort towns,
and by the pulp press, where allegedly scientific proofs
of European racial superiority were linked to social
evolutionist arguments for imperialist expansion.
By the first decades of the nineteenth century,
scientific and technological gauges of human achieve-
ment and worth were clearly in the ascendant. Earlier
measures of the level of development achieved by
different cultures continued to be cited. Religious
belief, for example, remained of paramount importance
to missionaries active in overseas lands. But even
missionaries increasingly linked Europe’s advances in
the sciences and invention to the rhetoric of Christian
proselytization. Ignoring past and contemporary ten-
sions between science and religion in Europe itself,
prominent missionaries, such as the Abbé Boilat and
David Livingstone, argued that Christian culture had
been particularly receptive to scientific investigation
and technological innovation, and that conversion to
Christianity would promote the scientific and material
development of colonized peoples. The growing numbers
of ethnologists and professionally trained anthropolo-
gists, who found in the colonies relatively safe and fertile
environments for their research, also privileged nonsci-
entific or non-technological standards, such as modes of
political organization or gender relations, in assessing the
level of development attained by African, Asian,
Amerindian, and Pacific Island peoples. But evidence
of material culture was increasingly linked to societal
advance, and indigenous “superstitions,” or at best
magical beliefs, contrasted with the scientific mindset
that was seen as typical of the educated West.
A number of factors account for the dominance of
material standards, particularly those linked to science

and technological innovation, in nineteenth-century
ideologies of European global hegemony. Most
critically, the transformations wrought by industrializa-
tion from the middle of the eighteenth century in
England and somewhat later in Belgium, Germany,
France, and Italy made the gap in scientific and
technological capacity and material development
between western Europe and non-Western societies
increasingly apparent to European and non-European
observers alike. Maxim guns, steamboats, and railway
lines carried elements of Europe’s industrialization to
colonized areas, and champions of imperialist expan-
sion reasoned that these wonders could not help but
impress subjugated peoples with the unprecedented
degree to which European societies had advanced over
their own. Not only did European superiority in science
and technology seem obvious, but also it could be
empirically tested in ways that claims of higher
religious understanding or moral probity could not.
Europeans had vastly more firepower, could produce
incomparably greater quantities of goods much more
rapidly, and could move both these products and
themselves about the globe with much greater speed
and comfort than any other people, including the once
highly touted Chinese. In an age when what were held
to be scientific proofs were authoritative, attainments
that could be measured statistically were viewed as the
most reliable gauges of human ability and social
development. Mechanical principles and mathematical
propositions could be tested; cast iron or steel bridge
spans could be compared for size and strength with the
stone or wooden structures of non-Western societies;
and human skulls could be quantified in seemingly
infinite ways to assess the highly variable mental
capacity of the “races of man.”
The preeminence gained in the nineteenth century by
scientific and technological standards of human worth
and ability not only bolstered proponents of theories of
European racial supremacy, but it also proved vital to
various formulations of the civilizing mission ideology
that both inspired and rationalized European imperialist
expansion from the early 1800s to 1914. Chauvinistic
politicians in the metropoles and imperial proconsuls in
the colonies increasingly stressed the importance of the
diffusion of Western science and technology to what
they viewed as the benighted peoples and backward
lands that had come under European control. Propo-
nents of the civilizing mission confidently predicted
that the world would be remade in the image of
industrializing Europe. Given Europe’s material ad-
vancement, it was seen as appropriate that Europe and
North America serve as the sources of capital, both
machine and financial; of entrepreneurial, scientific,
and managerial expertise; and of manufactured goods
for the rest of the globe. In this view, the non-Western
world, including both areas that had been formally
colonized and those that had come under the informal
Western dominance 2277
sway of the Great Powers, were best suited to provide
abundant and cheap land, labor, and raw materials that
were required to fuel the industrial economies of
Europe and North America.
According to the “improvers” or nonracist advocates of
the civilizing mission, the spread of Western education
among colonized peoples – emphasizing the inculcation
of at least rudimentary Western scientific learning and
technological skills – would provide the critical means
by which the material level of non-Western societies
would gradually be raised. Though they approved of
the diffusion of essential Western technology to overseas
areas under the paternalist supervision of European
colonizers, racist apologists for imperialism had little
faith in the ability of subjugated peoples to master
the sciences or engineering of the West. Thus, they
envisioned the period of European “tutelage” extending
for centuries, if not indefinitely, into the future.
The non-Western peoples who were the targets of the
European colonial enterprise were very often awed and
overwhelmed by their initial encounters with the
science and technology of the industrializing West.
Whether they were indigenous leaders resisting the
growing encroachments of European forces or scribes
and merchants who allied themselves with the invaders,
the colonized could not help but be impressed with the
clear and increasing advantages in power that the
Europeans gained through their superior capacity to tap
the resources of the natural world, to produce material
goods, and to devise more deadly weapons. As surveys
taken as late as the post-World War II era, such as those
which form the basis for G. Jahoda’s White Man, the
science and technology of the colonizers gave them an
aura of magical power among the colonized masses
in many areas. Though Western-educated Africans,
Asians, or Polynesians were likely to scoff at such
expressions of popular admiration, most came to
accept that Western science and technology were not
only on the whole superior to their own but essential
for the future “development” of societies they hoped
someday to rule. Therefore, nationalist critiques of
imperial domination often deplored the fact that
colonialism had severely constricted the flow of
science and technology from the West to dominated
areas, and demanded that technical education and
scientific facilities for indigenous peoples be expanded
and improved.
By the last decades of the nineteenth century,
however, a number of influential African, Asian, and W
Caribbean thinkers were mounting cogent challenges
both to notions of European racial superiority based on
evidence of scientific and technological achievement
and to the advisability of the wholesale transformation
of non-Western cultures and societies along Western,
industrial lines. Much of this resistance to the
hegemonic ideologies of the Western colonizers
focused on efforts to reassert and revitalize indigenous
2278 Western dominance
epistemologies, modes of social organization, and
approaches to the natural world. Thinkers such as
Vivekananda and Aurobindo Ghosh contrasted an Indian
spiritualism with the deadening abstractions of Western
materialism. African writers, such as the Caribbean-born
Edward Blyden, deplored the devastating impact of the
Atlantic slave trade on African cultures and celebrated the
Africans’ strong sense of community, reverence for and
care of the elderly, and sophisticated artistic creations.
Ironically, these defenses of colonized cultures were
buttressed by contemporary European anthropological
studies, usually carried out under the auspices of
colonial administrations; by the intense interest in
“Oriental” religions fashionable among European
intellectuals in the decades before World War I; and
by the “discovery” of the abstract power of African
masks and other forms of “primitive” artistic expres-
sion by Picasso, Derain, Matisse, and other avant-garde
artists in the early 1900s. Inadvertently, however, the
works of these first generations of Indian and African
critics of European hegemonic ideologies often vali-
dated the very materialistic standards they sought to
contest. For example, Indian thinkers, particularly
Vivekananda, repeatedly stressed the scientific accom-
plishments of India’s ancient civilizations, while
African and West Indian writers, most notably Anetor
Firmin, claimed Egypt, with its impressive engineering
and architectural feats, as a civilization that black
Africans had done much to build.
With the coming of the First World War, non-
Western critics of what had been characterized as the
excessively rationalistic, impersonal, and materialistic
West found numerous, and highly vocal, European
intellectual allies. The horrific trench slaughter on the
Western Front and the multitude of ways that Western
scientific knowledge and experimentation were har-
nessed to the war effort as a whole raised profound
doubts for noted thinkers, such as Paul Valery, Sigmund
Freud, and Georges Duhamel, about the long-assumed
progressive nature of Western science and technology.
After such a savage and suicidal war, the tenets of the
civilizing mission rang hollow, and “scientific” racist
thought came under increasing assault in both western
Europe and the United States. Collaboration with
indigenous elites was increasingly stressed in the
governance of the colonies, and a rhetoric of science
and technology as agents of development through
cooperation with indigenous peoples permeated colo-
nial policy making.
European doubts about the directions taken by the
industrial West and its global hegemony gave new
impetus to African, Asian, and Caribbean critiques of
European global hegemony. René Maran’s Prix
Goncourt-winning novel, Batouala, mocked the pre-
tensions of racial superiority held by European
colonizers, and idealized village life in French West
Africa. The poets of the négritude movement, most
powerfully L. S. Senghor and Aimé Césaire, mourned
the suffering and destruction wrought by European
science and technology in Africa and the lands of the
slave diaspora, and inverted racist epithets by exulting
blackness, intuition, affinity for the natural world, and
indifference to inventiveness. Aurobindo Ghosh and
Rabindranath Tagore viewed the war as the fulfillment
of earlier Indian prophecies of a coming cataclysm in
the aggressive and materialist West, and proof of
importance of India’s spiritual mission in the modern
age. Mohandas Gandhi also cited the senseless violence
and colossal destructiveness of the war in support of his
sweeping assaults on industrial society. In the decades
after the war, he sought to formulate for India (and
implicitly for other colonized areas) a community-
centered, low tech, and conservationist alternative to
industrialized society as it had developed in the West.
In this same period, Gandhi also worked out a strategy
for confrontational but nonviolent protest that repeat-
edly proved an effective antidote to the advanced
technologies of repression employed by Western
overlords in the decades of decolonization from the
1920s onward.
Though battered and under assault, the scientific and
technological underpinnings of ideologies of Western
dominance survived the crisis of two global wars and
the powerful critiques of Gandhi and the négritude
writers largely due to the emergence of the United
States as the global power from the 1920s onward.
Entering the First World War late and just when new
technologies had restored a war of motion and decision,
Americans continued their long-standing infatuation
with science and technology after the conflict. Faith in
the essentially progressive nature and beneficence of
science and technology informed the development
(later called modernization) theory that came to
dominate both American thinking on colonial issues
and that of the European and Japanese imperialist rivals
of the United States. After World War II, America’s
chief rival, the Soviet Union, also championed a
rhetoric of development that privileged science and
large-scale industrialization. With modernization theory
(in a number of capitalist and socialist versions) in the
ascendant, non-Western alternatives to social develop-
ment and economic well-being, such as that formulated
by Mohandas Gandhi, were marginalized or openly
spurned by the Western-educated elites that governed the
new states that emerged from the collapsing colonial
empires. Though alternative approaches have gained
significant support in some of these new nations, most
notably India and Tanzania, international agencies and
Western and non-Western planners continue to rely
upon modernization schemes, based overwhelmingly
on Western precedents, to solve the problems of poverty
and growing wealth differentials and the demographic
and environmental dilemmas that have been building on
a global basis for centuries.

See also: ▶Magic and Science, ▶Colonialism and
Science, ▶Science as a Western Phenomenon, ▶East
and West, ▶Technology and Culture
References
Adas, Michael. Machines as the Measure of Men: Science,
Technology and Ideologies of Western Dominance. Ithaca,
New York: Cornell University Press, 1989.
Cohen, William B. The French Encounter with Africans:
White Response to Blacks, 1530–1880. Bloomington:
University of Indiana Press, 1980.
Curtin, Philip. The Image of Africa: British Ideas and Action,
1780–1850. Madison: University of Wisconsin Press, 1964.
Gould, Stephen J. The Mismeasure of Man. New York:
Norton, 1981.
Irele, Abiola. Négritude or Black Cultural Nationalism. The
Journal of Modern African Studies 3.3 (1965a): 321–48.
---. Négritude–Literature and Ideology. The Journal of
Modern African Studies 3.4 (1965b): 499–526.
Jahoda, G. Whiteman: A Study of the Attitudes of the Africans
to Europeans in Ghana Before Independence. Oxford:
Oxford University Press, 1961.
Jordan, Winthrop D. White over Black: American Attitudes
Toward the Negro, 1550–1812. Chapel Hill: University of
North Carolina Press, 1968.
Leclerc, Gérard. Anthropologie et Colonialisme: Essai sur
l’Histoire de l’Africanisme. Paris: Fayard, 1972.
Nandy, Ashis. Science, Hegemony & Violence: A Requiem for
Modernity.Delhi: Oxford University Press, 1988.
Stephan, Nancy. The Idea of Race in Science: Great Britain
1800–1960. Hamden, Connecticut: Greenwood, 1982.
Windpower
R ICHARD L. H ILLS
There is little doubt that the first practical use of the
wind as an energy source other than as the motive
power for sailing ships occurred in the East. Those of
Persia were probably the first but precisely when is
uncertain. According to a story of al-T.abarī, writing
around AD 850, and later writers, the second orthodox
Caliph, Umar ibn al-Khat.t.āb, was murdered in AD 644
by a captured Persian technician, Abū Lu’lu’a, who
claimed to be able to construct mills driven by the
power of the wind and was bitter about the taxes he had
to pay. This early date cannot be confirmed but Arabic
geographers of the tenth century all confirm the
existence of windmills in the region of Seistan in north
eastern Iran. For example, al-Mas˓ūdī around AD 950
wrote of the wind driving mills and raising water from
streams.
Nothing has survived to show how such mills
pumped water but later drawings of one type of
windmill suggest that it could have been derived from
Windpower 2279
horizontal corn-grinding watermills. In the lower part
of such watermills, the water fell down a chute and
struck blades placed radially around a vertical shaft.
The top of the shaft passed into a higher room and
carried the upper millstone and so rotated it above the
bedstone which was set on a floor in the middle of the
mill. When developed into a windmill, the wind rotor
was situated in the bottom of the mill. The building was
constructed with four “loop holes” to direct the wind on
to the blades from whichever quarter the wind might
be coming. There might be eight or ten blades, and
the wind was directed by the loop holes onto one side of
the rotor. In this way the sails were pushed round on that
side away from the direction of the wind while those
advancing into the wind on the other side were shielded
by the walls around the loop holes. The grinding stones
were placed in an upper room with the upper stone
turned by an extension of the rotor shaft in the
“underdrift” manner. None of these mills has survived.
The type which may still be in use today had the
grinding stones situated in a room below the rotor with
the sails above. This necessitated a change to the
“overdrift” method of driving the stones where the
spindle passed down through the bedstone to a bearing
underneath, which had to support the weight of both the
upper stone and the sails. It was possible to disconnect
the driving shaft at the place where it was connected to
the upper stone at the “rynd” to allow for the stones
being separated for dressing. The layout for grinding
was similar to other corn mills with a hopper mounted
on the wall from which the grain was fed into the central
hole of the upper stone through a chute or shoe while
the flour was passed out around the circumference of
the stones.
The advantage of this second layout was that the
rotors with the sails could be greatly enlarged. Mills
near Seistan and on the borders of Iran and Afghanistan
might have rotors approximately 5 m (16.4 ft) high by
3 m (9.8 ft) in diameter. There could be six or eight sails
which had wooden framing that was interlaced with
straw or covered with wooden boards. The upper parts
of these mills were built with one wall that directed the
wind on to one side of the rotor while another wall
shielded the other half. Sometimes matting screens
were erected to help channel the winds to the sails.
There were no brakes, but more screens might be
placed across the slots between the walls to regulate the
wind reaching the sails; to secure the mill when out of
use, the rotor and upper millstone were lowered to rest W
on the bedstone. A wind speed of 22.40 m s−1 (50 mph)
was needed to drive these mills at only 30 rpm. Such a
mill, working with intermittent wind for about four
months of the year, would grind enough flour for about
15 families. Seistan was known as the land of the winds
and between mid-June to mid-October the wind
regularly blew from the north for a period of 120 days.
2280 Windpower
The mills were built in line to face these winds and the
famous example at Neh had a long row of 75 mills.
Because these horizontal mills needed such strong
and regular winds to power them, they did not spread in
this form much beyond the borders of Iran and
Afghanistan. They were invented earlier than the
Western vertical type but it is doubtful whether there
is any connection between the two. Those in the West,
which have been described as the “full admission, axial
flow type,” first appeared around AD 1150 possibly
either in the southeast of England, the northwest of
France, or Flanders. However these Persian mills were
probably the source of the later Chinese mills and
possibly the Tibetan wind-powered prayer wheels.
The wind-powered Tibetan prayer wheels were not
surrounded by elaborate buildings to guide the wind
onto the rotors. The most common form had a vertical
axle with horizontal spokes at its top, on the ends of
which were fixed sails shaped like cups so that they
caught the wind on one side of the rotor, but were
smoothed or curved to present less resistance on the
other. The wind would be caught in the concave part of
the cup, or a curved sheet, during half the circle to turn
the rotor, while in the other half, the convex or
streamlined side would be advancing into the wind and
so present less resistance. Power output was minimal
but was sufficient to turn the prayer wheels which was
the only use to which they were applied.
Prayer cylinders designed for automatic repetition of
the Buddhist mantra are unlikely to have been produced
before the reign of K’ri-srong-Ide-brtsan in AD 755 to
797 when Buddhism conquered Tibet. It is unlikely that
such cylinders were turned by the wind at this period,
although no doubt there were prayer flags fluttering in
the wind from around this time. Early in the twelfth
century, a new fashion for mechanical piety swept China,
but again it seems doubtful whether this included wind-
powered prayer wheels, which must therefore be placed
later, and certainly therefore after the Persian mills.
More is known about the adoption of the windmill in
China. Once again early dates for this have been
suggested, but these are doubtful. In about AD 1230,
Yelu Chu Zai was captured by Jinghi Khan and became
his minister. He was an extremely good scholar,
administrator and mathematician. An accurate descrip-
tion of the Persian windmill has been discovered in his
memoirs, with a comment on how good it would be if
the Chinese used it. A Chinese book of the seventeenth
century, the Zhu Qi Tu Shu, describes the windmill as if
it were a European invention, which could be a mistake
for Persia. In China, these mills were used for raising
water and evolved into an entirely different form from
the ones in Persia or Tibet.
Chinese horizontal windmills are still used today
along the eastern sea coast north of the Yangtze and in
the region of Thangku and Taku near Tientsin to
operate chain pumps through gearing for raising salt
water for salt pans or fresh water for irrigation. Once
again they have no elaborate structures for directing the
wind onto the rotors but are closely linked to the way
the sails of junks operate. Canvas-covered sails are
mounted at the ends of radial arms, each with its own
mast, in such a way that they can be spread to catch the
wind on the side of the rotor turning away from the
wind and be feathered to present the least resistance on
the side turning against the wind. The sails can pivot on
these masts, and, as on a junk, the sail extends to the
front of the mast. The longer, or driving, side of the sail
is tied to the rotor framework by a piece of rope of such
a length that, when the sail is turning with the wind, it is
held into the wind, but can rotate out of the wind when
advancing against the wind. The angle of the sail can be
set by the length of rope to a position in which the wind
can do useful work on it for more than 180°.
These mills must have been the most efficient of the
horizontal types developed before the twentieth
century, but all horizontal windmills suffer from the
same problem, that only a small part of the wind rotor
can be used to its maximum efficiency at any time. The
theoretical maximum power coefficient for a simple
horizontal windmill is only one third, but in practice it
will be much less. These are the reasons why the
horizontal windmill has not been developed further and
few examples remain at work today.
In the eastern Mediterranean, at some period during
the Middle Ages, vertical windmills appeared with a
different type of sails from those normally used in the
West. The horizontal wind shaft was extended in front
of the mill like a ship’s bowsprit so that ropes from it
could help to stay six or eight radiating spars. The sails
were triangular pieces of canvas like the sails of a
modern yacht and worked in the same way. The leading
edge of each sail was attached to the spar, round which
it could be wrapped to reef it in strong winds. The free
corner was secured by a rope in much the same way as
a ship’s boom. These sails were much lighter than those
of a conventional western windmill and have found a
new application in water pumping mills in some
developing countries today.
References
Harverson, Michael. Persian Windmills. The Hague: Interna-
tional Molinological Society, 1991.
Hills, Richard Leslie. Power from Wind. A History of
Windmill Technology. Cambridge: Cambridge University
Press, 1994.
Needham, Joseph. Science and Civilisation in China. Vol. 1,
Introductory Orientations, and Vol. IV, Physics and
Physical Technology, Part II, Mechanical Engineering.
Cambridge: Cambridge University Press, 1954 and 1965.
White, Lynn. Medieval Technology and Social Change.
Oxford: Clarendon Press, 1962.

Wine in Anatolia
A LEXIA S MITH
The quest for the earliest evidence of viniculture has led
many scholars to Transcaucasia and Anatolia, the latter
of which spans the Asian portion of modern-day
Turkey and derives its name from the Greek for “Land
of the Rising Sun”. It is not known precisely when
viniculture (the cultivation of grapes for wine) or
viticulture (cultivation of the vine) as were first practiced,
but both were well established in Southwest Asia by
the beginning of the third millennium BCE. Several
lines of evidence have provided insight into early wine
production, including textual accounts, archaeological
data such as drinking vessels, chemical analyses of
residues within vessels, and ancient plant (archae-
obotanical) data, as well as modern biogeographical
observations.
The modern distribution of wild grapes is useful for
demarcating the area within which grape domestication
could have occurred in the past. Today, wild grape
(Vitis vinifera L. subsp. sylvestris [C.C. Gmelin]
Berger), is found as a forest climber along the coastal
regions of central and southern Europe, areas surround-
ing the Black Sea, and the southern reaches of the
Caspian Sea, with isolated patches growing in
Turkmenistan and Tadzhikistan. Anatolia clearly falls
within this natural range.
Archaeobotanical remains provide more detailed
information. (See Agriculture: Ancient Methods of
Agriculture by Alexia Smith in this volume for a more
detailed description of archaeobotanical remains and
their recovery.) In general, wild berries are smaller and
more acidic than their domesticated counterparts.
While entire grape berries can be found, archaeobota-
nists most typically encounter charred pips which tend
to become shorter and rounder following domestica-
tion. In practice, however, distinguishing between wild
and domesticated grape pips is challenging, since
considerable overlap in their shape and size exists.
Consequently, determining the precise point in time
when grapes became domesticated is difficult. What is
clear from the archaeobotanical record is that wild
grapes have been collected for millennia to be eaten
fresh or dried. Fermentation of grapes into wine
requires fruit with a minimum sugar content of 10%
and the presence of the principal wine yeast, Saccha-
romyces cerevisiae, which naturally occurs on grape
skins. Experiments with wine making may have taken
place as early as the Neolithic period (ca. 10,500–5500
BCE), well before viticulture developed as a
specialized venture. Some of the earliest grape finds
come from Öküzini Cave in southwest Anatolia, and
Wine in Anatolia 2281
date to the Late Palaeolithic (16,000–10,500 BCE).
Grapes also appear consistently during the Neolithic
and Chalcolithic (5500–3000 BCE) periods at sites
such as Can Hasan III, Çayönü, and Korucutepe. From
the Chalcolithic to the Early Bronze Age (ca. 3000–
2000 BCE), a rapid increase in the presence of grapes at
sites such as Kurban Höyük in southern Anatolia
reflects the growth of viniculture.
Changes in the range of Early Bronze Age pottery
vessels are also linked with an intensification or
expansion of viniculture. At this time chalices and
other forms associated with drinking such as the depas
amphikypellon (a tall, narrow, two-handled tankard)
began to increase in number. Other drinking vessels
associated with wine include rhyta (sing. rhyton),
which were fashioned out of metal or ceramics in the
form of an animal’s head. Storage vessels include
amphorae1 and large pithoi2. At sites such as Boğazköy
and Alişar Höyük, containers fashioned after a cluster
of grapes have been found. The association of vessels
with wine can be affirmed using infrared spectroscopy
through the detection of tartaric acid, a compound that
occurs naturally in grapes and forms the primary acid
within wine. Tartaric acid was detected in a large
ceramic tub found in Early Bronze Age levels at Titriş
Höyük in southeastern Anatolia, linking it with
activities such as grape pressing.
Early texts written in Assyrian dating to the Old
Assyrian Colony Age (ca. 2000–1750 BCE,
corresponding with the Middle Bronze Age) and the
Hittite period (ca. 1600–1200 BCE, Late Bronze Age)
have been synthesized by Ronald Gorny and provide
additional insights into early wine production. Texts
from Cappadocian trading colonies refer to the months
of September and October as qitip karānim, the
“harvesting of grapes” season. Grape harvesting
festivals are also mentioned in Hittite texts (EZEN
GIŠ
GEŠTIN túh-šu-u-wa-aš). In Hittite, grapes or
grapevines were called wiyana (which features in the
etymology of the word “wine”) and are represented in
texts with the Sumerogram GIŠGEŠTIN. The official
position of “GAL.GEŠTIN” is mentioned in the Palace
Chronicle, an Old Hittite text that documents the court
during the reigns of Hattušili I and Muršili I. The early
responsibilities of GAL.GEšTIN included supervision
of vineyards and their products, which underscores the
importance of vineyards as an elite enterprise. Since the
majority of the texts document political, economic, and
W
1
An amphora is a jar, usually made of clay, with a narrow
neck and two handles, used by ancient Greeks and Romans
for holding oil or wine.
2
Pithoi were large ceramic storage jars, mainly used for
storing agricultural produce such as olive oil, wine, olives,
raisins or grain. Such large jars must have been made in
several sections and joined together before firing.
2282 Wine in Anatolia
religious matters, bias in the records exists, and little is
known about grape production or wine consumption
among the general populace. It seems fair to link early
viniculture with the elite, however. Since grape vines
require a number of years to become established and
yield fruits, only people of means could make such a
commitment outside of a garden setting. Alexander
Joffe even argues that production of alcoholic beverages
may be linked to the rise of the state by allowing
emerging elites to increase their level of power through
surplus production and control over craft production.
A number of Hittite laws pertaining to viniculture
bear testimony to well established practices and likely
reflect concerns of earlier time periods. The laws
clearly outline relative worth of various commodities
and assign appropriate penalties for theft or damages to
property. Damage to a vineyard is covered by Law 113,
whereby the guilty party must take the damaged grapes
and provide grapes from their own vines. Penalties for
theft are discussed in Law 101 and combine monetary
compensation with corporal punishment involving a
spear. Over time corporal punishment was stricken
from the law and the fine increased from one shekel per
vine to six shekels for a free individual and three
shekels for a slave. One Hittite shekel is roughly
equivalent to 12.5g of silver and could purchase
2 parīsu of wine, 3 parīsu of emmer wheat, 6 parīsu
of barley, or 160 shekels of copper. Each parīsu is
believed to correspond with 30 liters.
While wine was undoubtedly valued as a pleasing
beverage, its earliest uses were restricted to Hittite
rituals, festivals, and religious ceremonies, which in
general were conducted to gain favor of the gods with
respect to the particular issue or problem at hand.
Offerings of wine to the gods are mentioned in several
Hittite prayers, including the Prayer to Tešub of
Kummani, the Prayer to the Sungoddess of the
Underworld, and the Prayer of Muwatalli to the
Stormgod Pihašašši. A number of myths, such as
the Ašertu Myth˘ and the Myth of the Serpent Illuyanka,
discuss the inebriating power of wine and the potential
dangers of becoming intoxicated. The cleansing nature
of wine is stressed in several medical rituals, and
similarly in rituals intended to purify land and prepare
ground for the founding of a new palace. Soldiers also
used wine as a metaphor for blood during an oath
pledging ritual intended to secure their loyalty.
The association of wine with ritual and religious
beliefs perpetuated for some time. During the Lydian
period (687–546 BCE) in western Anatolia, Bacchus
was celebrated as the god of wine. During Byzantine
times in the early part of the first millennium A.D.,
wine was believed to be blessed by the sky-god;
vineyards and stored wine afforded protection from evil
spirits. Wine production decreased significantly fol-
lowing the conversion of Anatolia to Islam, although
wine has continually been drunk. Today wine produc-
tion is increasing again and grapes are widely
cultivated in many parts of Turkey, including the
Mediterranean coast and Central Anatolia.
References
Algaze, G., K. Ataman, M. Ingraham, et al. The Chicago
Euphrates Archaeological Project 1980–1984: An Interim
Report. Anatolica 13 (1986): 37–148.
Davis, P. H. ed. Flora of Turkey. Edinburgh: Edinburgh UP,
1967. 521–522.
Goetze, A. Laws from Mesopotamia and Asia Minor: The
Hittite Laws. Ancient Near Eastern Texts Relating to the
Old Testament. Ed. J. B. Pritchard. Princeton: Princeton
UP, 1969. 188–97.
Gorny, Ronald L. Viticulture in Ancient Anatolia. The
Origins and Ancient History of Wine. Eds. P. E. McGovern,
S. J. Fleming, and Solomon H. Katz. Philadelphia, PA: The
University of Pennsylvania Museum of Archaeology and
Anthropology, 2000. 133–74.
Hoffner, Harry A., Jr. Alimenta Hethaeorum. Food Produc-
tion in Hittite Asia Minor. American Oriental Series 55.
New Haven, CT: American Oriental Society, 1974.
Jacquat, Christiane and Danièle Martinoli. Vitis vinifera L.:
Wild or Cultivated? Study of the Grape Pips found at Petra,
Jordan: 150 B.C.–A.D. 40. Vegetation History and
Archaeobotany 8 (1999): 25–30.
Joffe, Alexander H. Alcohol and Social Complexity in
Ancient Western Asia. Current Anthropology 39 (1998):
297–322.
McGovern, Patrick E. Ancient Wine. The Search for the
Origins of Viticulture. Princeton and Oxford: Princeton UP,
2003.
Martinoli, Danièle. Les macroreste botaniques de la Grotte
d’Öküzini. La Grotte d’Öküzini: évolution du Paléolithique
final du sud-ouest de l’Anatolie. Ed. I. Yalçinkaya, M. Otte,
J. Kozłowski J, O. Bar-Yosef. Liege: Eraul. 2002. 91–4.
Powell, Marvin A. Wine and the Vine in Ancient Mesopo-
tamia: The Cuneiform Evidence. The Origins and Ancient
History of Wine. Ed. P. E. McGovern, S. J. Fleming,
Solomon H. Katz. Philadelphia, PA: The U of Pennsylvania
Museum of Archaeology and Anthropology, 2000. 97–122.
Smith, H. and Glynis Jones. Experiments on the Effects of
Charring on Cultivated Grape Seeds. Journal of Archaeo-
logical Science 17 (1990): 317–27.
Stager, L. E. The First Fruits of Civilization. Palestine in the
Bronze and Iron Ages. Papers in Honor of Olga Tufnell.
Ed. J. N. Tubb. London: Institute of Archaeology, 1985.
172–188.
Stanislawsky, Dan. Dionysus Westward: Early Religion and
the Economic Geography of Wine. Geographical Review
65 (1975): 427–44.
Zohary, Daniel. The Domestication of the Grapevine Vitis
vinifera L. in the Near East. The Origins and Ancient
History of Wine. Ed. P. E. McGovern, S. J. Fleming,
and Solomon H. Katz. Philadelphia, PA: The U of
Pennsylvania Museum of Archaeology and Anthropology,
2000. 23–30.
Zohary, Daniel and Maria Hopf. Domestication of Plants in
the Old World. The Origin and Spread of Cultivated Plants
in West Asia, Europe, and the Nile Valley. Oxford: Oxford
UP. 2000.
 Winemaking and viticulture in the ancient near East and Egypt
Winemaking and Viticulture
in the Ancient Near East and Egypt
V IRGINIA B ADLER
In the ancient Near East and Egypt the development of
winemaking and viticulture has a discontinuous record
due to the nature of both historical and archaeological
evidence. Historical records include texts and pictorial
evidence illustrating grapes, grape vines, wine produc-
tion, and wine consumption. Tracing winemaking and
viticulture to its more distant origins in prehistory is
now possible through careful archaeological investiga-
tions and scientific and chemical analyses. A signifi-
cant step toward understanding how scant prehistoric
evidence could contribute to reconstructing the pres-
ence and manufacture of such a specific, delicate and
liquid substance as wine emerged with the observation
of stains on the interiors of certain pottery jars and the
subsequent chemical analysis and identification of
key components of these residues. Badler et al. (1990)
pioneered chemical residue analysis (the identification
of tartaric acid as a marker for grapes and wine)
coupled with archaeological evidence to establish new
data points in ancient winemaking history. Guasch-
Jané et al. (2004) have further differentiated red from
white wine by the presence of malvindin-glucoside and
its unique by-product, syringic acid.
Prehistoric Wine and Ancient Viticulture
Because of extensive cultivation, crossbreeding, seed
dispersion, and deliberate transplantation, it is difficult
to determine by grape genetics alone the geographic
origins of winemaking and viticulture. Probably they
are located in the general area of the wild Eurasian
grapevine (Vitis vinifera sylvestris) that presently grows
from Spain to Lebanon, inland along the Danube and
Rhine Rivers, around the Black Sea and southern
Caspian Sea, at the headwaters of the Tigris and
Euphrates Rivers, and farther east in Central Asia
(McGovern 2003: 7). Winemaking preceded viticulture,
as grape cultivation is not necessary to make wine – it
can be made from the wild grapes that were plentiful in
certain areas of the Near East. The development of
winemaking could have followed a similar path as that
of breadmaking and the cultivation of grains. The
motivation for both grape and grain cultivation would
have been to provide convenient local sources and a
stable supply of raw materials for the end products.
Viticulture, the purposeful cultivation of grape vines,
would have also provided the opportunity to exercise
more control over both grape quality and quantity. If
there were no wild grapes in the area, a taste for wine
2283
imported from other regions would motivate the
transplantation of grape vines to secure a local supply
of the beverage.
If burnt or carbonized, grape pips (seeds) can be
preserved for many centuries. However, it is difficult to
determine whether these ancient grape pips are from
wild or cultivated grapes, due to variations in the pips
themselves and the shape changes that occur when they
are carbonized. Carbonized grape pips have been found
in many areas of the Near East and Egypt, dating to
3000 BCE or earlier: Susa and Hajji Firuz Tepe in
Iran; Hama, Mureybit, Abu Hureyra, Aswad, Jawa,
Tell esh-Shuna, Ohalo, Jericho, “En Besor and Bab
edh-Dhra” in the Syro-Palestinian area; and Buto,
Minshat Abu Omar, Tell Ibrahim Awad and Abydos
in Egypt (McGovern 2003: 2, Map 2). Many of these
areas (Egypt, most of the Syro-Palestinian region, and
southern Iran) are now too arid to support the wild
grape.
Neolithic Wine ca. 5400–5000 BCE: Hajji Firuz
Tepe, Iran (▶http://www.museum.upenn.edu/
new/exhibits/online_exhibits/wine/wineneolithic.
html)
Viticulture was likely first established in the Neolithic
period (from about 8000 to 4000 BCE), when for the
first time in human prehistory there were numerous
sedentary villages established in the Near East and
Egypt supported by domesticated plants (such as grains)
and animals (such as sheep and goats). Fired ceramic
vessels first made their appearance ca. 6000 BCE, and it
is within these vessels that we can find chemical
evidence for the earliest wine.
Hajji Firuz Tepe is located southwest of Lake Urmia
in northern Iran, in the Zagros Mountains at an elevation
of over 1,200 m above sea level, along the eastern
periphery of the wild grapevine’s modern distribution.
Jars with yellowish and reddish interior residues have
been analyzed, and both residues are that of resinated
wine. The resin was from the terebinth tree (Pistacia
atlantica) that grows throughout the Middle East and
would have been added as a preservative (McGovern
et al. 1996).
Late Chalcolithic Wine ca. 3500–2900 BCE: Godin
Tepe, Iran
The chemical analysis of residues in jars from Godin W
Tepe established that the technique could be used to
determine the existence of wine in prehistoric contexts
(Badler et al. 1990; Badler 1995). Wine was found in a
distinctive jar type (piriform, with rope decoration in
an inverted U shape on opposite sides of the vessel)
from two different contexts dating from 3500–3100
BCE and 3100–2900 BCE. The complete vessel from
2284 Winemaking and viticulture in the ancient near East and Egypt
the later period has a bung hole a short distance from
the base drilled after firing, presumably to decant the
beverage.
The homeland of viticulture may be Transcaucasia (the
region of modern Georgia, Armenia, and Azerbaiijan
located between the Black Sea and the Caspian Sea).
An early sixth-millennium BCE Shulaveris-Gora jar
was shown to have contained resinated wine, making it
the earliest known wine jar (McGovern 2003: 75). At
Godin Tepe there is an intriguing link between
winemaking and the appearance of the first Transcau-
casian type pottery. The earliest sherd with wine
residue dates from 3500–3100 BCE, when there are
significant numbers of Late Uruk type artifacts
suggesting contact with sites in southern Iraq (such as
Nippur and Warka/Uruk) and Iran (Susa). However, the
earliest evidence for wine production (a funnel, heavy
lid and empty wine jars, as well as a bin perhaps used
for treading grapes) date from 3100–2900 BCE, when
there is a marked increase in the number of Transcau-
casian type sherds. In fact, the funnel itself is similar to
the Transcaucasian sherds in its method of manufacture
(handmade and fired in a reduction atmosphere), and
the heavy lid is a Transcaucasian pottery type.
Late Uruk Wine ca. 3500–3100 BCE – Susa, Iran
and Warka/Uruk and Tello/Girsu, Iraq
The residues of ancient resinated wine have been
detected in Late Uruk type jars (a droop spouted jar, a
piriform jar, and a stone vessel with vertical rope
decoration) from Susa in southern Iran (McGovern
et al. 1997), a droop spouted jar from Warka/Uruk
(Badler et al. 1996), and a droop spouted jar from Tello/
Girsu in southern Iraq (McGovern et al. 1997). Since
grapes will not grow in the hot dry climate of southern
Iraq (there are no grape remains from this area), the
wine was most likely an import from further north, and
perhaps an important commodity in the Late Uruk trade
network (Algaze 1995).
The Early History of Wine in Syro-Palestine
and Egypt
Whereas the Near East looked north to Transcaucasia
for the introduction of wine and winemaking, archaeo-
logical and clay pottery jar composition evidence
clearly suggests that the origins of Egyptian wine
and winemaking were to its east from the Levant:
modern Israel, Palestine, and Jordan (McGovern 2003:
85–103). Hundreds of jars containing chemically
attested wine and wine yeasts were found in tomb U-j
of the first Scorpion king (Dynasty 0) dated by
radiocarbon determination to ca. 3150 BCE. The
closest stylistic parallels to these jars are from Early
Bronze I sites in the Levant, later confirmed by
additional scientific analyses. Although the wild
grapevine does not grow there today, it flourished in
the region in ancient times, beginning in the Epi-
Paleolithic period nearly 20,000 years ago (McGovern
2003: 212). By at least 3500 BCE, the domesticated
grapevine (Vitis vinifera vinifera) had taken hold in the
region. Very little archaeological or chemical evidence
for winemaking has been found so far, although small
spouted Iron Age vats could have been used to tread
the grapes.
Although there was a royal winemaking industry
in the Nile Delta region from as early as Dynasty 3
(ca. 2700 BCE), the wild grape never grew in ancient
Egypt. It seems most likely that both the grape vines
and the technology for making wine were imported
from the Levant. The earliest “wine labels” in the world
belong to Dynasties 1 and 2 of the Early Dynastic
period (ca. 3100 – 2700 BCE), and show an already
well developed viticulture tradition (McGovern 2003:
85ff.).
Once local viticulture and winemaking were estab-
lished in the Nile Delta, it became a source for royal
wine. When there were disruptions, however, such as
the Hyksos invasion (ca. 1670–1550 BCE), wine was
again imported in large quantities from the Levant and,
in particular, Southern Palestine. Exceptionally large
quantities of wine were imported into the Hyksos
capital of Tell el-Dab’a/Avaris located in the east Delta
region of Egypt. The Egyptian winemaking industry
rebounded (as evidenced by increasing quantities of
native Egyptian wine jars) beginning in the earliest
Middle Bronze IIA phases (ca. 1950 BCE) at Avaris.
There is evidence that the Egyptian Delta supplied the
New Kingdom towns and cities at that time. Avaris also
has the only excavated winemaking installation in Egypt,
predating Hellenistic times (McGovern 2003: 119).
Detailed tomb paintings provide additional evidence
for the Egyptian wine industry, from picking and
crushing the grapes to filling the jars (the tomb of
Nakht at Thebes [1550–1295 BCE]), the addition of
additives and stoppering (the tomb of Khaemweset, son
of Ramesses II [1279–1213 BCE], at Thebes), and a
sequence of grape picking, crushing, pressing, and
wine storage (tomb of Intef, Royal Herald under
Thutmosis III [1479–1425 BCE], also at Thebes).
Chemical analysis of the residues in the wine jars from
the tomb of Tutankhamun (d.1327 BCE) established
that this wine was made from red grapes (Guasch-Jané
et al. 2004).
Wine in the Ancient Near East
There are chemically identified ancient resinated wine
residues in a small jar from Warka/Uruk dating to the
late Jemdet Nasr/early Early Dynastic period (begin-
ning of the third millennium BCE; McGovern et al.
 Winemaking and viticulture in the ancient near East and Egypt
1997). In the early third millennium BCE, there are
grape pips and even grapevine wood identified from the
site of Tepe Malyan (ancient Anshan) in the Shiraz
region of the southern Zagros mountains (McGovern
2003: 165). Elamite seals depict drinking scenes (most
likely of wine) under grape arbors (McGovern 2003:
165–166). Texts from the city of Ur mention a jar full
of geštin (the Sumerian word for wine). At Tello/Girsu,
geštin was stored in rooms, and geštin was imported
from the eastern (presumably the southern Zagros)
mountains. Texts mention grapevines growing within
walled gardens in and around the region of the city of
Tello/Girsu by the end of the third millennium BCE
(McGovern 2003: 150). The law code of the Babylo-
nian king Hammurapi, dating to ca. 1790 BCE,
mentions wine shops. There are detailed accounts of
the wine trade (with cities further north) in the archives
of the Syrian city of Mari, located along the Middle
Euphrates, and dated to ca. 1750 BCE when the city
was destroyed by King Hammurapi, and in the archives
of the eighteenth century BCE site of Tell al-Rimah/
Karana’a in upper Mesopotamia (McGovern 2003:
169–173).
In the first millennium BCE, the Assyrian rulers
not only consumed large quantities of wine, but also
undertook an ambitious program of planting new
vineyards in their territories. The “Assyrian Doomsday
Book” recorded tens of thousands of vines planted in
the upper Balikh River to the east of Carchemish, in the
eastern part of their territory at Yaluna, probably
northeast of Nineveh, and in the land of Zamua in
the district of Sulaymanyah. The Assyrians produced
wine described as sweet, good, aged, or strong, and in
both red and white varieties. Additionally, some wines
were spiked with other ingredients (McGovern 2003:
190–191). The Nimrud wine lists contain detailed
accounts of the distribution of wine rations. In a wall
relief of the palace at Nineveh, Assurbanipal and his
wife are shown beneath a grapevine bower drinking
what is probably wine, similar to the Elamite scene
mentioned above. Chemical analysis of residues from a
bronze drinking cup from Nineveh, a bronze goblet and
pottery “beer jug” with built-in strainer spout, from Uruk
(all from seventh to sixth century BCE tombs) tested
positive for resinated wine (McGovern 2003: 201).
References
Algaze, G. Fourth Millennium B. C. Trade in Greater
Mesopotamia: Did it Include Wine? The Origins and
Ancient History of Wine. Ed. P. E. McGovern, S. J.
Fleming, and S. H. Katz. Luxembourg: Gordon and
Breach, 1995.
Amerine, M. A., et al. The Technology of Wine Making.
4th ed. Westport, CT: Avi Publishing, 1980.
Badler, V. R. The Archaeological Evidence for Winemaking,
Distribution and Consumption at Proto-Historic Godin
2285
Tepe, Iran. The Origins and Ancient History of Wine. Ed.
P. E. McGovern, S. J. Fleming, and S. H. Katz.
Luxembourg: Gordon and Breach, 1995. 45–56.
Badler, V. R., P. E. McGovern, and R. Michel. Drink and Be
Merry! Infrared Spectroscopy and Ancient Near Eastern
Wine. Organic Contents of Ancient Vessels: Materials
Analysis and Archaeological Investigation. Ed. W. R. Biers
and P. E. McGovern. Philadelphia: MASCA, The Universi-
ty Museum of Archaeology and Anthropology, University
of Pennsylvania, 1990. 25–36.
Badler, V. R., P. E. McGovern, and D. L. Glusker. Chemical
Evidence for a Wine Residue from Warka (Uruk) Inside a
Late Uruk Period Spouted Jar. Baghdader Mitteilungen 27
(1996): 39–43.
Barnett, R. D. Assurbanipal’s Feast. Eretz-Israel 8 (1985):
1*–6* .
Biers, W. R. and P. E. McGovern eds. Organic Contents
of Ancient Vessels: Materials Analysis and Archaeological
Investigation. Research Papers in Science and Archaeol-
ogy 7. Philadelphia: Museum Applied Science Center for
Archaeology (MASCA), University of Pennsylvania
Museum, 1990.
Broshi, M. Wine in Ancient Palestine: Introductory Notes.
Israel Museum Journal 3 (1984): 21–40.
Dayagi-Mendels, M. Drink and Be Merry: Wine and Beer in
Ancient Times. Jerusalem: Israel Museum, 1999.
Farkaš, J. Technology and Biochemistry of Wine. 2 vols.
New York: Gordon and Breach, 1988.
Guasch-Jané, M. R., et al. Liquid Chromatography with Mass
Spectrometry in Tandem Mode Applied for the Identifica-
tion of Wine Markers in Residues from Ancient Egyptian
Vessels. Analytical Chemistry 76 (2004): 1672–7.
Johnson, H. Vintage: The Story of Wine. New York: Simon
and Schuster, 1989.
Kinnier Wilson, J. V. The Nimrud Wine Lists. London: British
School of Archaeology in Iraq, 1972.
Kislev, M. E., D. Nadel, and I. Carmi. Epipalaeolithic (19,000
BP) Cereal and Fruit Diet at Ohalo II, Sea of Galilee, Israel.
Review of Palaeobotany and Palynology 73 (1992): 161–66.
Kitchen, K. A. The Vintages of the Ramesseum. Studies in
Pharaonic Religion and Society in Honour of J. Gwyn
Griffith. Ed. A. B. Lloyd. London: Egypt Exploration
Society, Occasional Publications, No. 8. 1992. 115–23.
Lesko, L. H. King Tut’s Wine Cellar. Berkeley, CA: B.C.
Scribe, 1977.
Lutz, H. F. Viticulture and Brewing in the Ancient Orient.
Leipzig: J. C. Hinrichs, 1922.
McGovern, P. E. Wine of Egypt’s Golden Age: An
Archaeochemical Perspective. Journal of Egyptian Ar-
chaeology 83 (1997): 69–108.
---. Wine for Eternity. Archaeology 5.4 (1998): 28–34.
---. The Foreign Relations of the “Hyksos”: A Neutron
Activation Study of the Middle Bronze Pottery from Tell el-
Dab’a (Ancient Avaris). British Archaeological Reports
International Series. Vol. 888. Oxford: Archaeopress, 2000.
---. Ancient Wine, The Search for the Origins of Viniculture.
Princeton, NJ: Princeton University Press, 2003. W
McGovern, P. E., S. J. Fleming, and S. H. Katz eds. The
Origins and Ancient History of Wine. Luxembourg:
Gordon and Breach, 1995.
McGovern, P. E., et al. The Beginnings of Winemaking and
Viniculture in the Ancient Near East and Egypt. Expedition
39.1 (1997): 3–21.
---. Neolithic Resinated Wine. Nature 381 (1996): 480–81.
2286 Woodcarving in Malay architecture

Michel, R. H., P. E. McGovern, and V. R. Badler. The First Rashidin 1999; Nik Hassan and Nik Abdul Rahman
Wine and Beer: Chemical Detection of Ancient Fermented 1998). In historical and geographical contexts, Malay
Beverages. Analytical Chemistry 65 (1993): 408A–13A. timber architecture and carving in peninsular Malaysia
Milano, L., ed. Drinking in Ancient Societies: History and
Culture of Drinks in the Ancient Near East. Padua: Sargon, stretches from the north in the province of Pattani
1994. in Southern Thailand to the south in the state of Johor.
Oren, E. D., ed. The Hyksos: New Historical and Archaeo- Similar floral, geometric, fauna, and calligraphic
logical Perspectives. University Museum Monograph 96. motifs appear on nonarchitectural components such as
Philadelphia: University of Pennsylvania Museum, 1997. boats, weapons, furniture, grave-markers (Fig. 1),
Phillips, R. A Short History of Wine. London: Penguin, 2001. house utensils, and musical instruments. The most
Seltman, C. Wine in the Ancient World. London: Routledge &
elaborate carving on a boat is the makara carved in
Kegan Paul, 1957.
Stager, L. E. The First Fruits of Civilization. Palestine in the relief and perforated styles depicting mostly flora and
Bronze and Iron Ages: Papers in Honour of Olga Tufnel. Ed. sometimes figurine motifs. The hilts of kris (Fig. 2),
J. N. Tubb. London: Institute of Archaeology, 1985. 172–88. badek (small dagger), and spears are some of the
Unwin, T. Wine and the Vine: An Historical Geography of weapons that are intricately carved on timber such as
Viticulture and the Wine Trade. London: Routledge, 1991. kemuning (Murraya paniculatum), leban (Vitex pub-
Walsh, C. E. The Fruit of the Vine: Viticulture in Ancient escens), sena (Pterocarpus indicus), or nibong (On-
Israel. Harvard Semitic Monographs 60. Winona Lake,
IN: Eisenbrauns, 2000. cosperma tigillarium). The craftsmen also showed their
Zohary, D. and M. Hopf. Domestication of Plants in the Old ideas and skills on more than 15 types of furniture and
World: The Origin and Spread of Cultivated Plants in West house utensils.
Asia, Europe, and the Nile Valley. 3rd ed. Oxford: Oxford In carving Malay craftsmen demonstrated high skills
University Press, 2000. of art changing abstract ideas into physical beauty. This
Zohary, D. and P. Spiegel-Roy. Beginning of Fruit Growing developed through a long apprenticeship whereby
in the Old World. Science 187 (1975): 319–27.
the skills and knowledge of woodcarving were passed
on. By imitating a carved masterpiece, a woodcarver
gradually modified the motifs and produced his own
design onto the timber piece. For example, a motif of
Woodcarving in Malay Architecture leaves and buds of ketumbit or getamguri (two shrubs) are
gradually modified by the carvers to demonstrate their
own stylish identity called air-tangan, but they keep the
I SMAIL S AID fundamental motif form (Nik Rashiddin 1999). This
pattern becomes the trademark, both for the craver and
Meaning and Significance
Malay vernacular architecture is characterized by
timber as the main building material, pitched roofs
and woodcarving on some of the building components.
As early as the fourteenth century, woodcarving was a
significant craft in the vernacular architecture practiced
by Malay craftsmen in Peninsular Malaysia and
Southern Thailand (Ismail and Ahmad 2001). Palaces,
houses, mosques, gateways, tombs, and wakafs (pavi-
lions) are adorned with more than 20 types of carved
components (Ismail 2002). The craft is an abstract
manifestation of craftsmen’s ideas into architectural
timber components, symbolizing artistic expression
into three architectural components: structural, elemen-
tal, and decorative (Ismail 2002). The craftsmen or
woodcarvers depict five types of motifs, in order of
importance: floral, geometric, calligraphy, fauna, and
cosmic. These motifs are laid in six geometrical or floral
shapes including oblong, square, circle, octagonal flora,
semicircle and triangle. In definition, woodcarving
is an art of partially removing wood from a board,
a block or a plank following specific motifs, patterns,
and orders. The art originated from two fronts: the
kingdom of Langkasuka from the north, and the Woodcarving in Malay Architecture. Fig. 1 A grave
Srivijaya empire from the south of the peninsula (Nik marker (Nik Rashiddin 2003).

Woodcarving in Malay Architecture. Fig. 2 A kris hilt
(Nik Rashiddin 2003).
for his architecture or craft. Differences in motifs, layouts
and perforations occur from one region to another. For
example, craftsmen in the northeastern states of Kelantan
and Terengganu depicted different floral motifs from the
central states of Negeri Sembilan and Melaka. The motifs
of woodcarving signify the identity of the carvers in a
region. In spite of the differences, Malay craftsmen
observed three common aspects: (a) applying four
principal design forms, (b) deducing motifs from home
landscape elements, particularly flora and fauna, and (c)
manifesting natural beauty as a devotional act to the
Creator and as a gift to his fellow men (Mohd Taib 1997).
The degree of complexity in carving varies from
one component to another; intricate, or complex ones
include door leaves and wall panels, and simple
carvings include gable board and fascia board. From
the carving characteristics including motifs and types
of perforation and incision, one could differentiate the
architecture of one state from another. For example, a
large gable board distinguishes a Terengganu house,
whereas a large-latticed gable portrays a Perak house.
As one gets closer to the carved components, the
distinction is further distinguished by the composition of
the motifs. The elevation of the Terengganu house is
adorned with perforated wall and ventilation panels of
varying sizes carved in floral motifs: leaves, tendrils,
and flowers of local plants such as ketumbit, getamguri,
keraknasi, jari buaya, and bakawali (Ismail 2000;
Ismail and Ahmad 2001). The panels of Perak houses are
carved in different floral motifs including sunflowers
and ketola, sometimes mixed with cosmic motifs. These
are regionally specific motifs found in Malay woodcar-
vings in Peninsular Malaysia. However, similarities
occur in depicting geometrical motifs on veranda railings
Woodcarving in Malay architecture 2287
or tall window and fascia boards. Likewise, similar
Arabic calligraphy motifs are depicted on door and
window ventilation panels.
Architectural Components
Carved building components can be categorized into
three types: structural, elemental and ornamental. The
structural components include pemeleh (bargeboard),
tunjuk langit (king post), cross beams, stringers of
stairs and braces. The pemeleh is the most prominent
component in palaces, aristocratic houses and mosques
in the state of Terengganu and in Kelantan. These
differentiate the vernacular architecture of this region
from other states in peninsular Malaysia. For example,
a palace known as tele house, built by Sultan Zainal
Abidin in 1868 in the city of Kuala Terengganu,
Terengganu, has two pairs of pemelehs, one at the rumah
ibu (main building) and the other at the rumah tengah
(subsidiary building). Likewise, Langgar Mosque in the
city of Kota Bharu, Kelantan, also has two pairs of elegant
pemelehs with a simple motif of barking deer on its
three-tiered roof (Marsiti 1976). This exterior component
handsomely extends the profile of the building toward the
sky, making the mosque a landmark in the settlement.
The next important structural carved component in
a house or a mosque is the tunjuk langit (kingpost).
It supports the roof by holding the ridge beam and
rafters. The Kampong Laut mosque in Kota Bharu has
eight carved tunjuk langits (Zahar 1989). This timber
mosque is perhaps the oldest timber mosque, perhaps
170 years old, in Peninsular Malaysia (Ismail 2001).
The tunjuk langit distributes the roof load to the tie
beams. The lower end usually is attached with a carved
component called buah buton. This is an ornamental
piece which covers the tenon and mortise joint of the
tunjuk langit to the tie beam. In contrast, the Telok
Manok mosque in Southern Thailand has three tunjuk
langits, joining the three-tiered roof to the tie beam one
above another (Ahmad 1998) (see Fig. 3). They are
beautifully carved with lotus blossom motifs and their
ends are equipped with buah butons using similar
motifs. The tunjuk langits and buah butons greatly
enhance the beauty of the interior. Apart from the
tunjuk langit as a structural member distributing the
roof load to the tie beams, braces, or brackets are also
structural components supporting the tie beam load to
the post. In the Kampong Laut mosque, braces are
elaborately carved in shallow relief with lotus blossoms W
as centerpieces framed by leaves of a different plant in
an intricate arabesque.
There are at least five carved elemental components
in Malay houses and mosques. These are essential
components that make up the entrances and fenestra-
tions. They are the ventilation panels of doors or
windows, door or window leaves, walls, railings and
2288 Woodcarving in Malay architecture

Woodcarving in Malay Architecture. Fig. 3 Elevation of the Telok Manok mosque in Southern Thailand illustrating the
carved components (KALAM, Universiti Teknologi Malaysia 1998).

gates. Ventilation panels characterize the architecture

of a region. For example, the Tengku Long palace
in Kuala Terengganu has 15 tall windows at its rumah
ibu, the main house, with perforated ventilation panels
(Fig. 4a,b). Each panel was designed with different
floral motifs in symmetrical and asymmetrical layouts.
They are part of the fenestration of the building,
allowing air and light in. These perforated boards are
placed on top of doors, windows, or walls allowing
circulation of air and light into the building. More
elaborate motifs, a combination of Quranic calligraphy
bordered by floral motifs, are carved for interior door
ventilation panels as can be seen in Fig. 5a.
Some ventilation panels are carved in simpler forms,
as in the Telok Manok mosque (Fig. 6). A floral motif
of a weed called ketumbit is abundantly carved on
the ventilation panels which are located around the
building and are placed above the walls, windows and
doors. The carvings transformed tender leaves of
ketumbit into an arabesque form both in relief and
perforated styles. The perforations not only allow air to
flow into the building but also permit light to illuminate
the interior. This is often dark because the interior is
covered by the pyramidal roof and also because of the dark
hue of the timber walls and the underside of the roof. The
late afternoon sun casts intricate shadows on the praying
space when light passes through the carved panels. This
phenomenon adds beauty to the interior setting of the
mosque. At night, the scene would be reversed when
light from lamps passes through the perforations casting
Woodcarving in Malay Architecture. Fig. 4 (a) A tall
a silhouette of the carved panels. The silhouettes of window at Tengku Long palace, an example of Terengganu.
arabesque and geometric forms can be clearly seen from (b) A ventilation panel with arabesque floral motif on top of a
outside especially in the absence of light from other window.
surrounding buildings. Therefore the perforated panels
add to the characterization of the mosque architecture. of liturgical furniture from which the iman delivers his
Inside a timber mosque, the mimbar (pulpit) demands sermon (Yacub Zaki 1978). The mimbar of Kampong
the most appreciation from the congregation. It is a piece Laut mosque is profusely decorated with perforated

carvings of floral motifs. The most prominent panel is
the front carved board attached to the canopy carved
in gunungan, one of four principal forms. It resembles
the form of the headdress worn by Malay women
in the marriage ceremony. The side and back walls
of the mimbar are fully furnished with perforated
carved panels as part of the structural component of the
four posts.
The gate is also an elemental component for
vernacular Malay architecture; it demarcates the access
to a palace or an aristocratic house. It is installed as part
of the wall or fence in single or double-leaf doors. The
leaf is generally carved in simple floral motifs with
straight railings, as in the gate of Kota Lama Duyong,
an aristocratic house built in 1920 in Kuala Terengganu
(Mohd Hanif 1997). This house has five types of gate
incorporated with a concrete wall surrounding the
building. Big and double-leaf gates are meant for
honored guests located at the front yard of the building
Woodcarving in Malay Architecture. Fig. 5 (a) A
ventilation panel on top of an interior door of Tengku Long
palace. (b) A panel with Quranic calligraphy and arabesque
floral motifs.
Woodcarving in Malay Architecture. Fig. 6 Floral motif on ventilation panels at the door and wall of Telok Manok mosque
(KALAM, Universiti Teknologi Malaysia 1997).
Woodcarving in Malay architecture 2289
compound (Fig. 7a). Smaller single or double-leaf gates
are located near to the kitchen and deck leading to the
rear of the house and are used by the residents and
neighbors (Fig. 7b).
The third type of carved components found in Malay
vernacular houses and mosques are ornaments. Som
and buah buton are ornaments that are carved and
attached to the structural members of the building. Som
is the decorative component located at the ridge end
of the pyramidal roof (Fig. 8). Each roof tier has
four soms. Langgar mosque has the largest and most
elaborate soms on its first and second roof tiers. A pair
of pemelehs accentuates the third tier roof. From a
considerable distance the soms and pemeleh accen-
tuate the profile of the mosque, giving it a distinctive
vernacular style. Telok Manok mosque has the most
soms; each ridge end and minaret roof is decorated with
floral motif soms. Another ornament that gives character
to the timber mosque is the buah buton – a decorative
component attached to end of tunjuk langit jointing the
tie beam or rafter to the tie beam. The attachment hides
the tenon and mortise joint and is not a structural part
of the building. Telok Manok Mosque has the most
buah butons – 18 at the tie beams and three at the end
of the tunjuk langits. All the carvings are in lotus
blossom motif, but each differs in form, resembling the
natural form of the flower that shrinks in the morning
and gradually opens during the day (Fig. 9). The
differences in form resulted from the interpretations of
several craftsmen.
Types of Motif
Malay craftsmen applied five types of motif: floral,
geometric, calligraphic, faunal, and cosmic onto the
carved components. Flowers dominated Malay wood-
carving; the carvers used the fruit, stems, tendrils,
leaves, and of course the flower. Fruit of pomegranates,
lotus flowers, sunflowers, ketumbit, ketola and baka-
wali, leaves of getamguri, and stems and tendrils of
ipomea are depicted in a variety of abstract forms. The
pomegranate is chosen as a motif because of its
interesting shapes and bright orange flowers. The
lotus is selected because of its auspicious status and
sacredness. Furthermore, weeds such as ketumbit with
bright yellow flowers and getamguri with wavy foliage
are recognized by Malay craftsmen in Terengganu and
2290 Woodcarving in Malay architecture
Woodcarving in Malay Architecture. Fig. 7 (a) A gate at
Kota Lama Duyong for residents and neighbors. (b) A larger
gate at Kota Lama Duyong for honored guests.
Kelantan; they translated their beauty into the tangible
art (Fig. 10). Craftsmen in Perak and Negeri Sembilan
favored bright yellow flowers and the intertwining
character of ketola and large, bright sunflowers on their
door leaves and ventilation panels. Malay craftsmen
observed the beauty of their surroundings and trans-
ferred their intangible value into a physical product that
could be appreciated by others.
Fauna are rare in Malay woodcarving because Islam
prohibits depicting figurine motifs. However, a few
craftsmen still carved fauna motifs such as: (a) a pair of
fighting cockerels on a window ventilation panel of a
Negeri Sembilan house, (b) a group of ducks waddling
Woodcarving in Malay Architecture. Fig. 8 Image of
soms at Masjid Telok Manok (KALAM, Universiti Teknologi
Malaysia 1997).
Woodcarving in Malay Architecture. Fig. 9 Lotus
blossom motif on buah butons at Masjid Telok Manok
(KALAM, Universiti Teknologi Malaysia 1998).
Woodcarving in Malay Architecture. Fig. 10 A ventila-
tion panel of Kota Lama Duyong in Kuala Terengganu carved
in floral motifs, getamguri and sulur kacang (KALAM,
Universiti Teknologi Malaysia 1997).
in a row on a stringer of a Malacca house, and (c) a lizard
head on a corner of a fascia board of a Perak house
(Abdul Halim 1986). These motifs are generally carved
in abstract forms, hiding the real figure of the fauna.
Calligraphy is also applied as a motif at wall and
ventilation panels on Kelantan, Terengganu and Negeri
Sembilan houses. It is either carved in relief, perforated
or a combination of both. The Malay craftsmen put
Quranic verses into wood panels written in several
Arabic styles. An excellent example is a wall panel at a
house in Kampung Pulau Panjang, Kelantan where a
Quranic verse is carved in perforated form and laid
symmetrically on an axis (Fig. 11).

Woodcarving in Malay Architecture. Fig. 11 A wall panel carved in traditional calligraphy motif at a house in Kampung
Pulau Panjang, Kelantan.
This carving demonstrated the craftsman’s devotion
to Islam and adoration of the Quran. The legacy of
calligraphy in Malay letter writing has contributed
woodcarving, which has selected it as one of its motifs
(Syed Ahmad 1992).
In addition to arabesque motifs, Malay craftsmen
also apply geometric motifs on door leaves, ventila-
tion panels, wall panels, railings, and partitions. The
configuration can be a series of diagonals repeatedly
copied throughout the component. Swastikas and stars
are more complex motifs that dominate the carving
configurations, for example, on a ventilation panel of a
wall. Repetition of similar motifs creates a sense of
beauty and contrast against adjacent foliate or calli-
graphic motifs. The motif of the swastika, which
originated from Chinese woodcarving, can be seen in
ventilation panels of Kota Lama Duyong (Mohd Hanif
1997). Geometrical motifs are easier to carve than the
other four types and thus repetitive components such as
railings are done by apprentices or sometimes wives of
the master craftsmen (Nik Rashiddin 1999).
The motif of the cosmos is seen the least in Malay
woodcarving. The depiction of this motif in Malay carving
illustrates the influence of Chinese craftsmanship and
the pluralism of cultural values in vernacular architecture.
A ventilation panel at Mohd Ali Kulup Mat Yassin’s
house in Perak is an example where leaf and flower are
combined with a cloud.
Principal Forms and Shapes of Carved
Components
In terms of iconography, Malay woodcarving depicts
four principal forms including stupa, makara, lotus and
gunungan (Fig. 12).
The stupa form can be found in house components
such as buah buton or on the newel posts of stairs or
gates. These are decorative components that enliven the
interior of the building. Buah buton is attached to the
end of the kingpost hiding the tenon and mortise joint
to a tie beam of the roof structure. It is the only
volumetric component that becomes the jewel of
carving, beautifying the interior of the house. The
gunungan is a silhouette of a mountain or tree of life, a
symbol of status. The gateway of a house compound is
carved in gunungan, but other components including
Woodcarving in Malay architecture 2291
Woodcarving in Malay Architecture. Fig. 12 Principal
forms of Malay woodcarving.
the ventilation panel of the door or window and door
leaf also used this form. The makara reflects the center
of the cosmological imagination for a mythological sea
monster in Pattani and Kelantan. Gable-end boards on
Terengganu or Kelantan houses depicted the form of
makara that enhanced the façade of the building. This
component differentiates the form of the house in
Terengganu and Kelantan over houses in Perak and
Negeri Sembilan. The lotus is a symbol of purity, and
its form is applied to a variety of house components,
including door leaves, fascia boards, ventilation panels
and buah butons. Apart from house components, the
form is carved at the foot of tombstones and bases of
kris hilts (see Farish and Khoo 2003).
The carved components are designed in six geometri-
cal or floral shapes including oblongs, squares, circles,
octagonal flora, semicircles, and triangles (Fig. 13).
Squares and oblongs are common layouts for many
components, particularly door leaves and ventilation
and wall panels. They are easier to carve than flowers
or circles since their outlines are straight. Ventilation
panels on doors, windows, and walls of Terengganu
and Kelantan houses are dominated by these perforated
components in floral motifs. The layout and size of
the components are proportionate with the elevation
of the building and volume of its interior space. The
house builder and craftsmen carved sufficient amounts
of fenestration to allow cross ventilation to give thermal
comfort to residents without depending on mechanical W
means (Ismail and Ahmad 2001).
Occasionally, a floral octagonal or circle layout is
carved for special components such as the base of a
ceiling lamp. Such carving can be found in aristocratic
houses such as Kota Lama Duyong. The intricacy of
the shape and carving denotes the skillfulness of the
Malay craftsman and wealth of the house owner.
2292 Woodcarving in Malay architecture
Woodcarving in Malay Architecture. Fig. 13 Shapes of carved components in the Malay houses.
Woodcarving in Malay Architecture. Fig. 14 A
perforated, arabesque wall panel carved in a symmetrical
layout with a vertical axis.
Apart from the shapes, the carvings are generally
symmetrical on at least one axis. The axis orientates the
carving pattern, repeating the pattern from left to right
or from the top to the bottom of a component. Fig. 14
illustrates a perforated carving of a wall panel of Wan
Embong house in Kuala Terengganu. Most ventilation
panels, wall panels, door and window leaves, and flat
railings are done in symmetrical formats with floral
motifs.
Factors Influencing Timber Selection
Generally, Malay craftsmen would apply three factors
in selecting appropriate timber for their carvings.
The factors include availability of timber, its phy-
sical characteristics and durability, and the craftsmen’s
spiritual beliefs toward the timber species. The hierarchy
of selection depends on the type of carved components.
For example, the making of house components such as
door leaves or ventilation panels is clearly determined
by the availability of timber which could easily be
obtained in large volume. On the other hand, the Malay
craftsmen chose only kemuning or kenaung for making
kris hilts and sheaths because of their fine interlocked
grains and deep yellow sapwoods overlapping with
dark brown heartwood. In addition, these timbers are
regarded as possessing good spirit that must be respected
and that will accompany the weapon (Faris and Khoo
2003; Ismail and Ahmad 2001).
Availability of Timber
Tropical rain forests of Peninsular Malaysia produce a
variety of quality, durable timber grown in the low-
lying undulating land and hills of the Main Range
(Appanah 1993). Heavy hardwood species such as
chengal (Balanocrpus heimii), balau (Shorea spp.),
and merbau (Intsia palembanica) are the favorites.
They are durable species that resist attacks from
termites, powder-post beetles and fungi, which would
harm their structural properties including flexibility,
stiffness, and hardness. These are suitable for making
large building components such as door leaves, barge-
boards, kingposts, wall, and ventilation panels as well
as small ones including railings and buah butons. The
same timber species are used to make structural
building components such as posts, beams, and floor-
boards as well as the carved components. It is common
to find an entire house made from chengal, balau, or
merbau in the states of Terengganu, Kelantan, Perak,
and Pahang in the peninsula. For example, a house at
Kampong Bolok in Pahang built in the 1920s, was
constructed wholly from merbau.
The timbers are air-dried under a shed or under the
houses, allowing prevailing winds and the sun to dry
them while protecting them from the rain. No preserva-
tive treatment is applied to the timber since the resins of
most dipterocarps are able to resist the powder-post
beetles, termites, and fungi attack (Farmers 1987). The
drying (seasoning) process may take months or even
years for some timber species particularly kemuning,
kenaung, and sena.
Medium hardwoods from the forests such as medang
(Litsea grandis), kundang hutan (Bouea macrophylla),
and keladan (Drybalanops oblongifolia) are sometimes
used for carving door leaves, furniture such as beds and
cabinets, and musical instruments such as the kenong
and kompang. These timbers are more prone to attacks
from powder-post beetles and fungi and thus they are
placed where they will not come into contact with
moisture.

Craftsmen would optimize timber choice by harvest-
ing readily available fruit trees grown in house com-
pounds and orchards. They would cut large branches or
sometimes the trunk of mature trees including ciku
(Achras zapota), jackfruit (Artocarpus heterophyllus),
rambai (Baccaurea bracteata), belimbing (Averhorra
belimbii), bacang (Mangifera foetida), and kundang
(Bouea macrophylla). Since these timber pieces are
relatively small, the craftsmen would carve them into
household tools and utensils and musical instruments
in floral and geometrical relief motifs. The tools and
utensils include coconut graters, ladles, food containers,
biscuit moulds, and rehal (a cradle in which to place the
Quran during readings). The practice of consuming
timber from cultivated trees suggests that the Malay
craftsmen were attentive to their environment.
Timbers for carving were also extracted from the
secondary and coastal forests where mixtures of heavy,
medium, and light hardwood species grow. Leban
(Vitex spp.), a heavy hardwood, and sena (Pterocarpus
indicus), a light hardwood, are the common species
obtained from the secondary forest for making sheaths
of badek or kris and for carving some utensils. The
craftsmen living in coastal villages would harvest
timber from penaga laut (Calophyllum inophyllum)
and kelat jambu laut (Syzygium grande) from the beach
forest. They were carved for boat paddles and grave
markers. Malay craftsmen optimized the timber found
within or adjacent to their environments.
Physical Characteristics and Durability
In woodcarving, the physical characteristics that govern
the suitability of timber are durability, color, grain
and texture, and luster. Before working on the timber,
the craftsmen select a timber piece that is free from
all defects, namely, knots, pith flecks, resin streaks,
brittlehearts, checks and splits, decay, bowing and cupping
(Nik Rashiddin 1999). Durability of hardwood is directly
related to density; high density generally suggests strong
resistance against fungi decay and boring insect attacks.
Therefore, chengal, balau, resak, and merbau are the
preferable species for most house components including
structural, elemental, and decorative types (Watson 1928).
Buildings made from these species are known to last for
more than 150 years. Their resistance is possibly due to the
silica deposits in their storage tissue (Desch 1981).
In some heavy hardwoods there is no color distinc-
tion between sapwood and heartwood, but usually the
heartwood is more deeply colored (Desch 1981). On
exposure to the air, chengal and balau are light brown to
dark red-brown and become darker as they age. On the
contrary, the merbau’s sapwood is pale yellow sharply
defined from the dark red-brown heartwood. Generally,
the final finish of the carved timber components made
from these heavy hardwoods is finished with sandpaper.
Woodcarving in Malay architecture 2293
In carving house components, the craftsmen is less
critical about the timber color but very selective when
carving weapon hilts and sheaths, furniture and musical
instruments. Only kemuning and kenaung, with bright
yellow sapwoods and dark brown heartwoods, are chosen
for the hilts of kris, badek and kerambit (a small knife).
These weapons are considered auspicious tools and used
only for ceremonial events or special occasions.
Grain and texture are two distinct characteristics of
timber. The grain refers to the direction of the fibers,
and texture applies to the relative size and the amount
of variation in the size of the cells (Desch 1981; Smith
1999). The chengal timber is easy to cut and incise
because it has straight fibers and it does not give rise to
ornamental figuring. It has a fine and even texture and
is thus suitable for almost all carvings, from large
features such as a boat’s figurehead, to a building’s
wall panels, to small objects such as rehal. The merbau
has a more interesting appearance than the chengal or
balau since it has interlocked and sometimes wavy
grain and a coarse texture with large vessels and coarse
rays (Farmers 1987). But it is more difficult to incise
since it is denser and has a higher hardness, thus it is
used for large building components such as posts and
seldom used for small crafts such as house utensils.
Lustrous woods such as merbau, kemuning, tempinis
(Sloetia sideroxylon), and kenaung have cell walls that
reflect light, particularly on quarter-sawn surfaces
(Desch 1981). The luster is a natural asset that crafts-
men exhibit in carved furniture and small crafts such as
kris hilts, knife sheaths, walking sticks, and picture
frames. The luster, however, does not last long without
a finishing coat over the timber surface. Craftsmen
apply several layers of shellac or varnish to a kris hilt
to retain the lustrous surface of the kemuning.
Spirit of Wood
In addition to the tangible characteristics of the timber,
Malay craftsmen also select timber based on its
possessing either a beneficial or malignant value (Farish
and Khoo 2003). Kemuning and kenaung are regarded
as the most auspicious species because the craftsmen
believe that they possess strong spirits that will
accompany a weapon such as a kris, badek, kerambit,
or spear. A few craftsmen in Kelantan believe that this
spirit is compatible with the iron blade. These timbers
are reserved for creating hilts and sheaths of weapons.
As work begins, a craftsman cannot be definite on what
style of kris hilt that a piece of kemuning or kenaung W
will finally become. Gradually, during the incision
process, the timber reveals its grain, texture, and luster
and only then will the craftsmen know the hilt style the
timber will become. The motifs on this hilt would be
similar to large architectural components such as leaves
of getamguri and jaribuaya, flowers of ketumbit and
2294 Woodworking in Egypt
keraknasi; these are shrubs or weeds commonly found in
Malay house gardens (Syed Ahmad 1992; Ismail and
Ahmad 2001).
Woodcarving is a significant craft in Malay vernac-
ular architectural and nonarchitectural elements. It is a
manifestation of craftsmen’s imagination using floral,
geometrical, calligraphic, faunal, and cosmic. It signifies
the identity of a regional architecture as well as a pattern
of the craftsmen from a region. In intangible perspective,
it illustrates the belief of the Malay craftsmen toward
architecture, devotion to god and their contribution to
society. To understand the significance of the craft, one
needs to investigate its physical forms in term of motifs,
principal forms and layouts. In addition, it is important to
know the characteristics of the timber species from which
the carvings are made.
References
Ahmad, Marsiti. Measured Drawing: Masjid Langgar. Kota
Bahru: KALAM, Universiti Teknologi Malaysia, Unpub-
lished, 1976.
Ali, Kamaruddin. Architecture: Unity of the Sacred and the
Profane. Islamic Civilization in the Malay World. Ed.
Mohd Taib Osman. Istanbul: Dewan Bahasa Pustaka and
IRCICA, 1997. 254–84.
Atan, Zahar. Measured Drawing: Masjid Kampung Laut,
Kota Bahru and Kelantan: KALAM, Universiti Teknologi
Malaysia, Unpublished, 1989.
Desch, H. E. Timber: Its Structure, Properties and Utilisa-
tion. 6th ed. London: The Macmillan Press, 1981.
Embong, Ahmad Zulkhairullah. Measured Drawing: Masjid
Teluk Manok. Kota Bahru: KALAM, Universiti Teknologi
Malaysia, Unpublished, 1997.
Farmers, R. H. ed. Handbook of Hardwoods. 2nd ed. London:
Her Majesty’s Stationery Office, 1987.
Fee, Chen Voon ed. The Encyclopedia of Malaysia:
Architecture. Kuala Lumpur: Archipelago Press, 1998.
Hassan, Mohd Hanif. Measured Drawing: Kota Lama
Duyong. Kuala Terengganu: KALAM, Universiti Teknologi
Malaysia, Unpublished, 1997.
Jamal, Syed Ahmad. Rupa dan Jiwa (Form and Spirit). Kuala
Lumpur: Dewan Bahasa dan Pustaka, 1992.
Lim, Jee Yuan. The Malay House. Penang: Institut Masyarakat
Pulau Pinang, 1987.
Nasir, Abdul Halim. Ukiran Kayu Melayu Tradisi (Malay
Traditional Woodcarving). Kuala Lumpur: Dewan Bahasa
dan Pustaka, 1986.
Nik, Hussein and Rashiddin Nik. Personal Communication.
Kelantan: Bachok, 1999.
---. Spirit of Wood. Kuala Lumpur: Exhibition at Badan
Warisan Malaysia, 2003.
Noor, Farish A. and Eddin Khoo. Spirit of Wood: The Art of
Malay Woodcarving. Singapore: Periplus, 2003.
Osman, Mohd Taib. Islamic Civilization in the Malay World.
Dewan Bahasa: IRCICA, 1997.
Said, Ismail. Research Report: Study of Timber Species in
Malay Woodcarving in Peninsular Malaysia. Malaysia:
Research Management Center, Universiti Teknologi
Malaysia, 2000.
---. Research Report: Visual Composition of Malay Woodcar-
ving of Traditional Houses in Peninsular Malaysia.
Malaysia: Research Management Center, Universiti
Teknologi Malaysia, 2001a.
---. Art of Woodcarving in Timber Mosques of Southern
Thailand and Peninsular Malaysia. Jurnal Teknologi B 34
(2001b): 45–56.
---. Visual Composition of Malay Woodcarvings in Vernacular
Houses of Peninsular Malaysia. Jurnal Teknologi B 37
(2002): 43–52.
Said, Ismail and Ahmad Saifuddin Abdullah. Timber Species
in Malay Woodcarving. Johor: Universiti Teknologi
Malaysia, 2001.
Shuhaimi, Nik Hassan and Nik Abdul Rahman. Early History:
The Encyclopedia of Malaysia. Vol. 4, 1998. 76–80.
Smith, J. Wyatt. Pocket Check List of Timber Trees. Rev.
K. M. Kochummen. Malayan Forest Records no. 17.
Malaysia: Forest Research Institute Malaysia, 1999.
Watson, J. G. Malayan Forest Records no. 5: Malayan Plant
Names. Singapore: FandN, 1928.
Zaki, Yacub. Allah and Eternity: Mosques, Madrasas and
Tombs. Architecture of the Islamic World. Ed. Ernst J.
Grube and George Mitchell. London: Thames and Hudson,
1978. 15–40.
Woodworking in Egypt
G EOFFREY P. K ILLEN
More is known about woodworking in Egypt than
anywhere else in the ancient world. By the New
Kingdom, Egyptian woodworking had reached its
zenith with examples of woodworking such as furniture
being highly prized and often sent as tribute to the
rulers of neighbouring countries.
Many Pre-dynastic burials in the Nile valley have the
body placed on wooden poles and covered with matting
made of plant fibre, while some burials are found in
primitive wooden boxes. By the unification of Upper
and Lower Egypt we find bed frames, Plate 1, in
common use with many examples being found in 1st
dynasty (3100–2890 BCE) tombs. The quality of these
bed frames ranged from conveniently shaped branches
Woodworking in Egypt. Plate 1 Bed frame, 1st Dynasty,
Metropolitan Museum of Art, MMA 12.187.52.
 Woodworking in Egypt 2295

that were lashed together to sophisticated examples

made from rounded poles that were jointed together
and supported on finely carved bovine shaped legs
(Petrie 1913, Pl. VIII–IX).
At the beginning of the Old Kingdom, which opens
with the 3rd Dynasty (2686–2613 BCE), we see major
advances in building and the associated trade of
carpentry. The quality of royal furniture made during
this period can be seen in those examples discovered by
the American Egyptologist George Reisner, in the 4th
Dynasty tomb of Queen Hetepheres (ca. 2600 BCE) at
Giza (Reisner 1955). When he opened the tomb, he
found that the wooden elements from which the
furniture had been made had rotted away to powder. Woodworking in Egypt. Plate 2 Cosmetic Box, 18th
However, it proved possible to reconstruct much of the Dynasty, Metropolitan Museum of Art, MMA 26.7.1438.
Queen’s furniture by studying the positions of the gold
sheaths, which had encased the furniture, and the inlays
that had fallen free and lay on the tomb’s floor.
Hetepheres’ furniture consisted of two armchairs, bed
frame, bed canopy, carrying chair and two boxes. What
Reisner and his team achieved, from a pile of unrelated
fragments of gold and faience is remarkable for it has
given us a small but superb collection of early furniture,
which rivals Tutankhamun’s, manufactured over a
thousand years later.
We see the introduction of the wooden box at the end
of the Old Kingdom. They were manufactured with
flat, gable, barrel and shrine shaped lids. Some were
very large and were designed with a pair of poles that
enabled the box to be carried by a team of porters. In
one tomb scene 14 men are carrying a box. During the
Middle Kingdom we find boxes were customised to Woodworking in Egypt. Plate 3 Table with cavetto cornice
hold cosmetics. Many were designed like crates to edge, 18th Dynasty, Metropolitan Museum of Art, MMA
hold small alabaster jars, which held perfumed oils. 14.10.5.
Other boxes have been found to contain mirrors, kohl
containers, combs and even a pair of slippers! A box
made for Sithathoriunet (ca. 1800 BCE) (Winlock stands were made from thin strips of timber braced with
1934) was decorated with gold fittings and bezels in cross and angled struts. They were fitted with a shaped
which were set with polished carnelian stones. Other collar, which held the round base of a single vase. They
elaborate boxes held cosmetics or; these were usually were covered with a gesso foundation before being
inlaid or veneered with sheets of ivory or exotic timbers painted to imitate carnelian and faience inlay. Those
bought from lands south of Egypt (Plate 2). Scribes chairs made during the Middle Kingdom had either
even had boxes in which they stored their writing short backs over which was draped a cover or cushion
implements and palettes. Their boxes were usually or they had backs of full height. Such chair backs were
painted to imitate the stringing and veneered panels curved and made from angled slats of timber. They
found on more ornate boxes. stood on slender gazelle-shaped legs. Often chairs were
Important directional changes in Middle Kingdom painted to simulate animal skin, which were painted
furniture can be seen by studying the large collection of with a technique, which resembles cow skin and was
stelae, which are preserved in the Egyptian Museum, used on an arrow quiver case, which is preserved in the W
Cairo. These Middle Kingdom stelae show that tables Egyptian Museum, Cairo.
were widely used for the display of vases or holding By the New Kingdom, the homes of officials and
water pots. Many are low with straight legs and have nobles would have been furnished with a wide range of
a single stretcher strung below the tabletop. We also wooden furniture, the most common of which would
see that Egyptian carpenters were constructing splay- have been the stools. Egyptians used a large number of
legged tables, which had cavetto cornice mouldings different types of stool. The most commonly used were
below the edge of the tabletop (Plate 3). Slender vase lattice stools that were made from thin struts of timber
2296 Wound healing in ancient Egypt
Wound Healing in Ancient Egypt1. Plate 4 Lattice Stool,
18th Dynasty, British Museum, BM EA 2476.
with angled braces supporting a double cove seat
(Plate 4). Round-legged stools appear in some of the
more important Theban tombs. The majority of legs
from these stools were hand rounded although there is a
small corpus of material, which have legs that appear to
be turned. During the New Kingdom we see carpenters
sitting on three legged stools, which allowed the stool
to rest evenly on the workshop floor. The folding stool
originated in the Middle Kingdom and was made from
two interlocking frames with a leather seat. New
Kingdom examples are more elaborate having the floor
rails and crossing spindles finished with carved goose
head terminals, which are inlaid with ivory to imitate
the eyes and neck feathers. We also see that lion legged
stools and chairs were used in the homes of high-
ranking officials.
The wooden furniture manufactured in the royal
workshops was not very different in design to that used
by the middle classes. However, they were exquisitely
embellished with gold sheet, inlaid with coloured
stones and faience or veneered with ebony and ivory.
They were also adorned with the uraeus [the sacred
serpent found on the headdresses of Egyptian rulers and
divinities, representing sovereignty] and the symbols
of kingship. Other pieces are inlaid with thousands
of slivers of coloured wood in either marquetry or
parquetry patterns. In the tomb of Yuya and Tuyu (ca.
1400 BCE), the parents of Queen Tiy and the wife of
Amenhotep III (1390–1352 BCE), a small armchair
made for Princess Sitamun was discovered. The illustri-
ous examples of furniture discovered in the Tomb of
Tutankhamun (1336–1327 BCE) show the outstanding
quality of design and construction achieved by 18th
Dynasty carpenters.
References
Killen, G. P. Furniture – Chapter 15, The Seventy Great
Inventions of the Ancient World. Ed. B. M. Fagan. London:
Thames and Hudson, 2004. 71–5.
---. Furniture and Woodworking, The Oxford Encyclopaedia
of Ancient Egypt. Ed. D. Redford. New York: Oxford
University Press, 2001. 580 –86, 516 – 19.
---. Woodworking, Ancient Egyptian Materials and Technol-
ogy. Ed. P. Nicholson and I. Shaw. Cambridge: Cambridge
University Press, 2000. 353 –71.
---. Egyptian Woodworking and Furniture. Princes Risbor-
ough, Buckinghamshire, Shire Publications, 1994.
---. Ancient Egyptian Furniture, Volume II. Warminster,
Wiltshire, Aris and Phillips, 1994.
---. Ancient Egyptian Furniture, Volume I. Warminster,
Wiltshire, Aris and Phillips, 1980.
Petrie, W. M. F. Tarkhan and Memphis V. London. British
School of Archaeology in Egypt, 1913.
Reisner, G. A. A History of the Giza Necropolis, Volume II,
The Tomb of Hetep-heres the Mother of Cheops. Cam-
bridge, Massachusetts: Harvard University Press. 1955.
Winlock, H. E. The Treasure of El Lahun, The Metropolitan
Museum of Art, New York, 1934. pp. 12–17.
Wound Healing in Ancient Egypt1
H EDVIG G YŐRY
Representations of injuries are relatively infrequent in
ancient Egypt. However, the accuracy in the depiction
of accidents in the joiner’s shop drawn on the wall
of the tomb of Ipui in Deir el-Medineh, and the
representations of the battle of Qades in Abu Simbel,
Luxor and Ramesseum temple, reveal exceptional
medical details and the consequences of specific
injuries. The exact identification of injuries on the
battle scenes is possible in two-thirds to three-fourths of
the injured (only Hittite) soldiers. Most of them were
inflicted on the chest and abdomen, or neck and head,
respectively, these being the largest targets in bow-
and-arrow warfare or in hand-to-hand fighting. For
instance, a ‘decelebrate rigidity’ is shown in a Hittite
soldier struck by an arrow on the top of his head, or a
haematoma (accumulation of blood) is shown deep in
the abdominal wall caused by an arrow entering another
Hittite soldier’s right lower-chest or abdomen below
the rib cage. An examination of skeletons show that
arrowheads severed most of the spinal cord, resulting in
instant paralysis of the body below the neck.
1
In the text, the following abbreviations are used for two
medical papyri: Eb is Ebers papyrus and SM is Edwin Smith
papyrus.

Types of Wounds
We can differentiate several types of wounds depend-
ing on their characteristic look, origin or severity. The
same is true for ancient Egyptians. Moreover, medical
papyri, especially the Edwin Smith papyrus, are very
accurate and detailed in the description of various types
of injuries and their treatments. They tell us that the
healer was able to characterise the most important types
of injuries, impressive even by today’s standards. They
differentiated various types of swellings, wounds,
fractures and joint injuries.
The general word for an injury, where the protective
function of the skin was disturbed or hurt, i.e., a wound,
is wbn.w. Without any explanation it means the
commonest form of injury, the rupture of the soft
tissue overlying the bone and leaving a constricted
orifice. Beside this simple form, they spoke also about
a wbn.w n kft, (a gaping wound) describing an open
wound with a well-defined gash as, for instance, in
Sm 28, where the most important clinical sign for an
open oesophagus, drinking water regurgitating through
the wound, is mentioned. The axillary could also be
open (Sm 47). Another type of wbn.w could be wbn.
w sd (sg), (a wound smashing (sg) or contusion), or
wbn.w jsp (ro-r), [a wound cutting (through)]. Another
description of wound, known only from the Smith
papyrus, is wbn.w jsdb (r), [a wound piercing (to)],
which seems to be nearly synonymous with wbn.
w ˓r (n), [a wound penetrating (to). According to the
examples we have, the penetration reached a solid body
part as bone, while the piercing came through into a
void. For a simple perforation the common word was
thm. Finally, in pEbers 877c we find the wbn.w sh.m,
(a crushing wound), laceration and rupture, and in
pEbers 494 wbn.w n wbd.t (burn wound), although for
designating this last type it was enough to write wbd.t –
(burn) alone, or to give the word for the actual phase
of its development. Ancient Egyptians differentiated
various types of fractures. Beside the h.sb—simple,
open fracture—there was the closed sd fracture, where
the bone was broken in two inside the body, or
the commuted compound or impacted pšn fracture,
where the bone was broken in several places. The
above-mentioned thm and sh.m could also occur below
the skin. As a diagnostic clinical sign of fracture,
nhbhb, meaning, in all probability, ‘crepitation’ [the
˘ ˘ made by rubbing together the ends of a fractured
noise
bone], is mentioned.
Treatment
During the repair of any type of wound, the practitioner
has to face three basic factors: pain, bleeding and
infection. The ancient Egyptians realised only two of
them, since they regarded infection as part of the
recovery process. The most common complaints,
Wound healing in ancient Egypt 2297
swelling and fever indicate that wounds were often
infected.
Interestingly, they never mention anaesthesia for use
in surgical interventions although painkilling must
have been an important part of the repair procedure.
The possibility of anaesthesia was there because we
know that they collected rush, which can be used as
a narcotic, that they knew very well the effects of
alcohol, that they cultivated mandrake and poppy from
the New Kingdom on, and that the medical papyri
reveal a knowledge of hyoscyamus [henbane, a genus
of poisonous herbs] and scopolamine as well. Their use
of analgesia was thus theoretically possible, either in
the form of a type of food or concoction, and their use
as hypnotic agents might also be hypothetically
suggested. The only words, however, which might
hint at anaesthesia accompanying the Old Kingdom
circumcision scene where the use of local anaesthesia
can be deduced from a promise to make it ndm –
‘sweet, agreeable’, when the region is rubbed with a
piece of stone. Some consider that water was mixed
with vinegar over Memphite limestone, resulting in the
formation of carbon dioxide, which may have had an
analgesic effect; the process would then resemble
today’s cryo-analgesia.
Ancient practitioners were more deeply concerned
with arresting haemorrhages, since the mtw (Eb. 871),
tumours (Eb. 875) and wounds were susceptible to
bleeding.
Medical Protocol
In preparation for medical treatment in ancient Egypt a
medical protocol was first issued by describing the
symptoms, making a diagnosis and giving the form of
treatment. This process was usually followed by the
medicine man, who was usually a physician or
probably a priest of Sachmet.
Diagnosis could be divided into three categories:
(1) “an ailment, which I will treat”; (2) “an ailment
which I will contend with”; (3) “an ailment not to be
treated”. The main wound repairing methods were
bandaging with an ointment or poultice, scorching (Eb.
876, Eb. 872) and surgery. In doubtful cases, enforced
bedrest and normal diet were often recommended to the
patient until he reached the decisive moment when
the healer could give either a positive or negative
prognosis. At the end of the treatment sometimes
the verdict was also given as (a) “until he recovers”; W
(b) “until the period of his injury passes” by; (c) “until
you know that he has reached decisive point”.
Relatively fewer examples are given for incurable
ailments. Only 5 out of 48 cases in the Smith papyrus,
and a few in the tumour section of the Ebers papyrus,
as Eb. 874, may be leprosy, or Eb. 873, a possible
cavernous haemangioma, or a twisted vascular swelling
2298 Wound healing in ancient Egypt
forming knot (in this case, however, the mtw should
be straightened). The first two steps in Eb. 877 are
still unidentified, although cancer, cutaneous leprosy,
bubonic plague or neurofibromatosis has already been
suggested. In the last step in Eb. 876, the ball-shaped
extensions have tentatively been identified by Regöly-
Mérei as varicosities.
The cause of wounds can be varied. They are most
often trauma by accident or battle, blows or knocks,
scratching or scraping, bites, burns or a symptom or
stage of an illness. The preferred treatment was drug
therapy accompanied by bandages, which was supple-
mented in severe cases by small surgical interventions.
Some mechanical treatments are also known.
Surgery also took place sometimes for religious
(circumcision) or aesthetical reasons (ear or nose
restitution) and prosthetic surgery was also known,
although not all that often.
Types of bites, such as those from crocodiles (Eb.
436, H 239–40, 243), lions (H 244) and humans (Eb
432–435, H 21), were treated like any other wound,
preferably by applying fresh meat on the first day and,
subsequently, honey, wax and oil or grease as in Eb.
435 or by using other herbal remedies, occasionally
with the addition of some animal ingredients. Snake-
bites, however, were treated differently, and by
different medicine men, mainly by the snake-charmer
priest of Selket, because of the venom inflicted by the
bite into the body. Tumours were well differentiated.
Several sorts of tumours, abscesses, cysts, swellings,
and boils (h.nh.n.t, ˓o.t, sf t, ˓nw.t, šfw.t, hsd, two.w, bsj,
h.mo, sh, bnw.t, wnm-snf ) are mentioned ˘ mostly in
˘
the Ebers, Hearst and Berlin papyri, with very brief
medical instructions. These were treated by the same
methods as wounds caused by mechanical injuries.
In the Ebers papyrus there is a complete book on
wounds originating in burns: 482–509. No. 482 lists a
sequence of remedies which, it says, should be applied
consecutively. In other cases, remedies applying an
ointment or bandage are always the same.
Methods
Drug Therapy
Egyptian drug therapy can be regarded as having
evolved from a system rooted in magic that developed
from empirical observations. Sometimes a sequential
approach was used (Eb 522). In the 182 cases given in
the Smith, Ebers, Berlin, Hearst and Ramesseum III
papyri regarding wound repair, almost 130 types of
materia medica were used. All were applied externally.
More than 60% of them were herbal remedies; the
remaining types were mineral or substances originating
from animals (both in about the same degree), with
a few undetermined components. The only human
remedy was urine.
Very often (in every 6th case), fresh meat was
applied to a wound on the first day as an efficient
haemostatic and mechanical agent. It was bound on the
wound on the first day only, and was usually followed
by the application of a lint saturated with ointment
composed of grease and honey, also bound on. In
specific cases, many other ointments and poultices
were also applied (e.g., Eb 510–42). In other cases the
treatment started with these materials. Almost 80% of
the prescriptions contained oil or grease in either the
form of mrh..t or ˓d, both prepared from various plants
or animals, closing off the wound from air and
introducing some minerals or vitamins into the injured
tissues. In almost half of the prescriptions honey was
used, which can be beneficial as a wound dressing
because it inhibits the growth of micro-organisms and
is hygroscopic, thus attracting an abundant secretion of
leukocytes and antibodies. Taking into account that the
wax, which was also frequently mentioned in these
cases, must have also contained some honey, its use
was as popular among Egyptian drugs in wound
healing as oil or grease.
dor.t was often used (ca. 25%), probably because of
a disinfectant effect; this has been determined from
the analysis of its use in various ancient Egyptian
prescriptions. Unfortunately, the identification is not
fully resolved: colocynth2 and carob are suggested,
although the last one seems to be more probable.
Besides these basic medicaments, some others are also
characteristic. Typical poultice materials were beans,
peas, leeks, a sort of fibre (dbj.t), faïence powder, bsn-
clay, wš˓-mineral, jmrw – eventually gypsum or plaster,
and linen (both the usual colour and jrtj.w -a red one).
Several other materials were used for their anti-
biotic, antiseptic, astringent, rubefacient (for internal
pain), antiphlogistic (counteracting inflammation), an-
algesic, fever reducing, hygroscopic or wound sealing
effects, or because they could contribute to the forma-
tion or development of the epithelium. Most of them
are listed, but infrequently. Only a few are common:
incense, salt, acacia, sycamore, barley, millet, faeces of
various animals, Christ thorn, celery, cumin and a few
unidentified ones. Some of the materials used only
rarely might also be very effective. This can be stated
for instance for the willow (trt) having salicyl as
its analgesic, fever reducer, anti-inflammatory effects,
and being also anti-rheumatoid. It is, however, only
mentioned in three prescriptions. Another such compo-
nent is onion, listed only once, but being a good anti-
inflammatory and very effective against staphylococci,
2
The colocynth is the light spongy pulp of the fruit of the
bitter cucumber (Citrullus, or Cucumis, colocynthis), an
Asiatic plant allied to the watermelon. It comes in white balls,
is intensely bitter and a powerful cathartic. Also called bitter
apple, bitter cucumber, bitter gourd.

streptococci and some sorts of dermatomycosis [fungal
infections of the skin].
The question can be raised, however, how often a
certain prescription was used, as we can surely attest
that the medical knowledge of ancient Egyptians was
slowly but continually developing, incorporating new
methods and materials in their papyri, although they
continued to practice medicine as they had earlier
times. One of these new materials was also the willow.
Mechanical Treatment
Treatment for Joints
Dislocations (wnh) and sprains (nrw.t) were treated.
In the tomb of ˘Ipuj in Deir el-Medineh, there is a
picture of industrial accidents. Among them the use
of Kochers method for dislocated shoulders3 can be
clearly discerned. Other methods are described in the
Smith papyrus, for instance, treatment for the disloca-
tion of a jaw (Sm 25). For neurological complications
resulting from the dislocation of a cervical vertebra, the
inability to move the limbs, priapism, urinary inconti-
nence, and intestinal dilatation with gas (paralytic ileus)
are mentioned (Sm 31). In another case, a positive
Lasague sign is described, i.e. pain due to a sprain
of the vertebral column when the legs are extended
(Sm 48), which indicates a prolapsed lumbar interver-
tebral disc with nerve involvement.
Bandages
For drawing together a gaping wound they applied
strips of adhesive plaster, putting them on both sides of
the wound (Sm 10, 27). Splints could be used for
administering medicament into the mouth, as it could
be held open with them. The same supporting use is
attested at nose and ear operations. The common
application of splints was for fractures. Wounds were
usually bandaged after ointment was put on them, or
the bandage was saturated with it. The linen for
bandaging (Sm 2, 9, 10, 27, 47) varied in texture from
the finest silk-like gauze to a canvas-like coarseness.
Ancient Egyptians had special knowledge in this field,
which was explained in a specific book, as a note in the
Smith papyrus says. The skill was first taught for the
bandaging of mummies.
Treatment for Fractures
Based on many relatively good healing settings, we can
assume that ancient Egyptian bone setting was
advanced. Among the skeletal remains from ancient
Egypt, there are innumerable completely healed limbs
with good realignment of the bone, but most other bone
3
Kochers method of reduction involves bending the elbow
to 90°. The arm is slowly rotated 75° laterally, then the point
of the elbow is lifted forwards, and finally the arm is rotated
medially.
Wound healing in ancient Egypt 2299
settings often show gross shortening and misalignment.
Since fractures of the femoral bones often healed
without complication, it may indicate, on the one hand,
that the virulence and incidence of pyogenic [produc-
ing pus] pathogens was lower than they are today,
although their existence is proved by the frequency of
osteomyelitis. On the other hand, based on the skeletal
remains of pyramid workers and other Egyptians, it is
obvious that fractured limbs were splinted using an
advanced technique. Cleverly, if only one of the bones
of the lower leg or forearm was broken, the remaining
intact bone served as a splint for the broken one, and
healing was satisfactory. In instances where the damage
was so severe, amputation was an alternative course
of action.
The standard procedure for the reduction of fracture
was traction (Sm 12, 35, 36). And, indeed, many
humeri set in splint remained. The first procedure was,
after realigning the broken limb, to set it either with a
splint wrapped in bandages, or encased in a healing
poultice-cast that could be made of various (often
unidentified) materials, as, for instance, (to mention
some known substances) cow’s milk mixed with barley
(H 219) or acacia leaves bound together using gum and
water (H 223). For reducing a fracture of the clavicle
(Sm 35), the ancient Egyptians also used a modern
method, first described among Greek physicians by
Hippocrates, by stretching the patient ‘on his back with
a folded cloth between the shoulder blades’ and pulling
‘on his two shoulders until the fracture falls into
position’. Among the neurological complications, as in
the case of an impacted fracture of the cervical
vertebrae (Sm 33), aphasia, unconsciousness and
tetraplegia, are mentioned.
Surgical Processes
. Wounds were often surgically treated by suturing
with jdr stitching (Sm. 3, 10, 14, 23, 26, 28, 47). The
Smith papyrus has the first mention in history of
‘stitching’ a wound. Other wounds were not sutured,
but kept open.
. The th.n—piercing with the h.mm instrument—was
prescribed for opening the ascites [fluid in the
abdomen] in Eb 865. Regarding the only surgical
procedures for burns, Eb 504 speaks about opening a
blister, this time with a sharp thorn.
. Sandwiched between two pharmaceutical prescrip-
tions, also for a blister in Eb. 501, an incision is W
suggested, but, as usual, without any details.
. The dw-˓-‘knife-treatment’, meaning, probably, ex-
cision, was fairly popular, as the many prescriptions
show: Eb. 866 for a vascular tumour (haemangio-
ma?), Eb. 867 for a presumably subcutaneous
lipoma, Eb. 868 for a possible sebaceous swelling
with metathesis ( ˓ot nt so), Eb. 869 for an abscess,
2300 Wound healing in ancient Egypt
Eb. 870 for a swelling. The identification of the latter
has several possibilities, such as a dermoid cyst lined
with skin appendages, pilonidal sinus located in
the midline or sebaceous cyst of the scalp, as, for
instance, atheroma, Eb. 871 for an abscess in the
axillary lymph glands.
. Sometimes, the dw-˓-‘knife-treatment’ was applied
with a heated (sšmm) knife as in Eb. 872 for a
probable haemangioma.
. In some cases ‘fire’ is mentioned, treated ‘with the
so-h.mm (firebolt?–man)’s treatment. This scorching
was used in the case of the skin-like flesh protrusion
in Eb 863 or that of the (umbilical or epigastric?)
hernia in Eb 864.
. The method was applied according to the state of the
cause, as in the first two cases of Eb. 876, where the
tumour was eliminated either by excision or by
scorching fitted to its state.
In all the above cases, procedures are not described
and thus nothing definite is known. The only complex
and detailed surgical process is given in the Ebers
papyrus (Eb. 875)—with various knife and forceps
devices—and is, in all probability, the month-long
treatment for dracunculiasis (which may be the origin
of the medical emblem with snakes), and thus a very
atypical surgical invention.
Skull Repair
According to the Edwin Smith papyrus, the ancient
Egyptians recognised brain disorders and performed
skull repair. They were also the first in the history of
medicine to identify the inner parts of the cranium.
They also provided vivid descriptions of the brain and
cerebrospinal fluid. Parallel to the development of
weapons, the number of head injuries increased, thus
making the discussion of the many cranial injuries
in the Smith surgical treatise fully understandable. The
dominant factor in healing, however, seems to be the
vis mediatrix naturae, the expectant healing power of
nature. It is also manifest by the signs of treatment on
heads with cranial injuries found by Elliot Smith, Wood
Jones, Courville, and recently, Nerlich.
The ability to recognise the classic sign of a base
fracture in the skull, i.e. bleeding from the nostrils and
ears (Sm 4–5, 7–8, 17, 21–22), or the basic sign of
meningeal irritation, i.e. the inability to flex the neck to
the breast (Sm 3–5, 7), indicates a sound basic medical
understanding. And, indeed, pathological descriptions
give clues to the identification of cases, like classic
meningitis in Sm 3 or cerebral prolapse and nucleal
rigidity in Sm 6. In Sm 7, a complex case scenario is
given dealing with a fracture at the base of the skull.
First, trismus and meningeal irritation (cured by
palliating treatment with an unknown hot material,
allowing the patient to be able to open his mouth),
then infection of the wound causing tetanus is
diagnosed, again treated by palliative treatment with
mdo.t – ‘wooden chisel’ and tigernut drink (w˓h.), since
recovery was hopeless. The Smith papyrus also
discusses a contracoup case with a pulsating swelling
protruding through the cranial fracture (Sm 8), where,
besides the symptoms of the eyes looking in the
direction of the affected side, the ears and nose are
bleeding and nuclear rigidity with a hemiplegia on the
injured side is documented.
Snakebite
According to the Brooklyn Herpetological Papyrus,
snakes and the effect of their bites were accurately
identified, and their bites treated with rational medical
care. The treatments comprised the application of
bandages with salt and natron, both being anti-
agglutinant and anti-hemolytic, thus hindering the
absorption of the venom; drinking of emetic remedies
with various drugs to get rid of venom by vomiting;
fumigation as adjuvant; magical spells for general and
specific divine help and/or the dw-˓-knife-treatment
and the tštš incision in the swelling. For drug therapy
many earlier, not medically attested, substances were
used, which may be explained by the 30th Dynasty
date of the papyrus or by our missing knowledge of
empirical snakebite treatment, but eventually by both
of them.
The simple cut (dw-˓) was applied there twice, in the
case of a ‘male viper’ (Br. 31) and of a uraeus/cobra
(Br. 32); the wound is attributed to Seth and the
prognosis is positive in both cases. The action is
not specified, and, for the first case, there is nothing
else to do. In the second case, however, the knife
treatment is preceded by a remedy (‘every thing for …)
with a missing part, which is thought to be by Sauneron
‘emetics’). These two procedures resulted in vomiting,
thus saving the patient’s life. The word tštš originally
meant ‘crush’; in these cases, however, the knife
determinative finishes the hieroglyphs. Thus, here
the ‘tštš with many cuts’ means, in all probability,
the cutting of the area of the bite into pieces by a knife,
applied to the bite swelling of any type of serpent
(Br. 72a) or, specifically, to that of an unidentified
cobra(?) ‘male snake’ (Br. 81). The next step was,
however, the bandaging of the wound with salt or
natron, or, in the second case, a medicament with the
addition of incense, an unidentified liquid, and cobra
blood. The latter one, in addition to the venom, also
contains the antitoxins.
Surgical Devices
We do not know very much about medical implements.
The earliest tools come from the tomb of Qar of the 6th
Dynasty, including forty scalpels and tweezers as

well as other medical tools whose purposes are still
unknown.
For cutting the flesh they used knives of particular
shapes and sizes. Various ancient Egyptian names are
also known, such as ds, h pt, šos, h.mm(?) or swt. There
was also a peculiar fish-tailed knife, called psš-kf, used
for cutting the umbilical cord after delivery. In the
beginning the knife was a ‘disposable’ stone, most
likely made of flint or obsidian, which was still used for
cutting up dead bodies for mummification in the first
century AD (Diod. I. 91) or blades fashioned from reed
stems. However, cuttings on the body were made from
the New Kingdom on, also with a metal knife. This
innovation led to a new procedure: the blade was
heated until it was red hot before an incision was made
so that the knife both cut and sealed at the same time.
If a bone broke, bark or cartonnage cases padded with
vegetable fibre or linen for securing the injured member
by splints might be used, but Eliot Smith also mentions
splints made of reeds connected by a linen band.
For taking out pieces of bones or any other foreign
bodies, tools such as hooks and forceps were used.
There was also a pair of small leather forceps, called
hnw(h.). For discarding pus or other amorphous
formations, tweezers or spoons and spatulas or
vegetable ‘knives’ (sw = bulrush) were used. Other
plants were used in other ways; for instance, the thorn
of an acacia was used for cutting up blisters (Eb.
504 = L 52 = L56).
Wound Healing in Ancient Egypt. Fig. 1 Detail from the Smith Surgical papyrus. Source: Breasted, J. H. The Edwin Smith
Surgical Papyrus in Facsimile and Hieroglyphic Transliteration with Translation and Commentary, Chicago 1930.
Wound healing in ancient Egypt 2301
For cauterising there was the do – ‘fire drill’ and
sometimes heated metal cauteries. For closing open
wounds, various types of bandages (hoyt, sšd) or
needles for suturing by jdr stitching were employed.
Linen strips coated on one side with a gummy
substance were used, for instance, as strips of adhesive
plaster (owy). They had a sort of cotton wool made
of fibres of the plant dbj.t, which was soaked in drugs
(Sm 28) and enclosed by linen bandages (sšd). To
prevent the dislocation or slipping of the bandage, they
applied a gummy substance or varnish taken from
embalmers. A wedge covered with layers of cloth is
described in Smith 7 for use in prying open the jaw.
Medical devices seem to be varied and fairly
specialised, each used according to its function or to
the requirements of a particular stage of the operation.
Their refinement was, however, primitive, although the
technical skill existed to make complex instruments, as
we know that some cosmetic devices are astonishingly
complex and of high quality. The repertoire, naturally,
continually changed and developed. The state of our
current source material, however, does not allow us to
deduce details. At the temple of Kom Ombo, a box of
instruments for a general practitioner, carved into the
outer corridor wall and originating from the Roman
Period, contains 37 instruments, identified as metal
shears, knives, saws, suction cups, probes, a saw,
tweezers, pincers, small bags, retractors, a small scale,
lances, specula, chisel, dental or cleaning tools,
2302 Wound healing in ancient Egypt
Wound Healing in Ancient Egypt. Fig. 2 Broken bones
with splints. Source: Majno, G., The Healing hand: Man and
Wounds in the Ancient World. Cambridge 1975.
Wound Healing in Ancient Egypt. Fig. 3 Drawing with
surgical hook used during operation.
spatulas and spoons, together with some objects having
connections with magic. The ensemble presents a
good example of the last period of Egyptian medical
praxis, when Pharaonic Egyptian and Graeco-Roman
medicine were completely fused.
Mythological Background
The medical papyri point to the fact that pernicious
beings could enter and leave the organism through the
openings of the body and thus harm people. From the
Egyptian medical perspective, the wound was also an
opening. Thus wound management had to be twofold.
The first task was to repair the harm already occurred,
by closing the opening and repelling the demons who
had entered. Prevention was essential too, to assure the
defence of this gate by magical rites and material feared
by these illness-causing intruders. Seth often personi-
fied their negative power. The association of Seth with
wounds was close, whether the wound was received in
battle or as a consequence of the mythical fight between
Horus and Seth for the succession to the throne of
Osiris. According to this concept, any illness could
only be healed with religious rites.
Material in connection with a plant or an animal might
dispose some magical power and be a device through
which people could get in contact with the gods. And the
same is true for minerals. Thus any medical material
could also express the protective role of a deity itself.
This view is the most manifested in ‘magical’
components of the prescriptions. But even their choice
was based with all probability on long empirical
observations. Many of the materials are still (may be
forever) unknown for us, but most of the identified ones
have proved to be reasonably explained, and are
appropriate in terms of their effect. The explanation of
their effectiveness could only be given at that time,
however, by mythological–theological reasoning.
Various material combinations could be used for
seemingly the same type or state of illness or health,
having the same or similar effect. There were several
routes of drug administration, with a wide range of
formation for use. They were usually determined by the
choice of ingredients applied both for their physical–
chemical properties and magical assumptions. The
sycamore was, for instance, the tree of Hathor, Nut or
Isis, the Christ thorn that of Thot of Pnubis, malachite
was the mineral of Hathor, the donkey belonged to the
god Seth, excrement was disgusting even for illness-
demons, the so-wr resin bore in its name (‘great of
magic’) its magical connotation. For patients who were
wounded, bandages were also used as magical devices.
They were traps, so that invisible beings could be kept
off or removed.
Another way of making a treatment more effective
was religious–magical rite. Similar to cases in the
London papyrus (15–21, 46–48, 53–61) and also
among the Ebers papyrus prescriptions (482–500),
magic spells had to be cast, referring to Horus having
been burnt in the desert. According to the incantation of
Eb. 499, which had to be recited with mother’s milk,
Isis used ‘water in her mouth’ and ‘Nile springing
between her thighs’. Then mother’s milk, gum and ram
fur (having a high cholesterol content) were put on the
burnt wound.
The mythological background worked as a psycho-
logical support as well, which recent studies have
shown can reduce pain perioperatively and increase
patient’s satisfaction.
See also: ▶Surgery in Ancient Egypt

References
Aufrere, S. L’univers Minéral Dans La Pensée Égyptienne.
Le Caire: IFAO, 1991.
---. L’encyclopédie Religieuse De L’univers Végétal.
Croyances Phytoreligieuses De L’égypte Ancienne I-II.
Montpellier: Orientalia Monspeliensia, 1999, 2001.
Breasted, J. H. The Edwin Smith Surgical Papyrus in Facsimile
and Hieroglyphic Transliteration with Translation and Com-
mentary. Chicago: University of Chicago Press, 1930.
Ebbell, Bendix. Die alt-ägyptische Chirurgie. Die chirur-
gischen Abschnitte der papyrus E. Smith und Papyrus
Ebers. Oslo: I kommisjon hos J. Dybwad, 1939.
Ebeid, N. I. Egyptian Medicine in the Days of the Pharaohs.
Cairo: General Egyptian Book Organization, 1999.
Estes, J. W. The Medical Skills Of Ancient Egypt, Revised
Edition. Canton, Massachusetts: Science History Publica-
tions, 1993.
Germer, R. Untersuchungen über Arzneimittelpflanzen im
alten Ägypten. Diss. Hamburg, 1979.
Griffith, F. Ll. and H. Thompson. The Demotic Magical
Papyrus of London and Leiden. London: H. Grevel & Co,
1904–1909. 1–3.
Győry, H. “Öffnen Des Sehens” Gedanken über das Rezept
Ebers 344. Göttinger Miszellen 189 (2002): 47–56.
Győry, H. Interaction of Magic and Science in Ancient
Egyptian Medicine. Zahi Hawass, ed. Egyptology at the
Dawn of the Twenty-first Century, Proceedings of the
Eighth International Congress of Egyptologists Cairo 2000,
Cairo: American University in Cairo Press, 2003. 276–83.
Hoenes, S.-E. Untersuchungen zu Wesen und Kult der Göttin
Sachmet. Bonn: Habelt, 1976.
Känel, F. Von, Les Prêtres-Ouab de Sekhmet et les
conjurateurs de Serket. Paris: Presses Universitaires De
France, 1984.
Majno, G. The Healing Hand: Man and Wounds in the
Ancient World. Cambridge: Harvard University Press, 1975.
Manniche, L. Egyptian luxuries: Fragrance, aromatherapy
and cosmetics in pharaonic times. Cairo: The American
University in Cairo Press, 1999.
Nevine El-Aref. Surprise at Saqqara, Al Ahram Weekly on-
line, 11–17 October 2001, issue no. 555. ▶http://www.
ahram.org/weekly/2001/555/tr1.htm
Nunn, J. F. Ancient Egyptian Medicine. London: British
Museum, 1996.
Regöly-Mérey, G. Surgery in Ancient Egypt. Acta Chirurgica
Academiae Scientiarum Hungaricae 15.4 (1974): 415–25.
Ruffer, M. A. Studies in the palaeopathology of Egypt.
Chicago: University of Chicago Press, 1921.
Sanchez, G. M. The injuries in the battle of Kadesh. KMT
Magazine 14/1, 2003 Spring, 58–65.
Sauneron, S. Un Traité Égyptien D’ophiologie, Papyrus Ddu
Brooklyn Museum n°47.218.48 et.85. N°11. Bibliothèque
Générale, IFAO, Le Caire 1989.
Sipos, P., et al. Special Wound Healing Methods Used in
Ancient Egypt and the Mythological Background. World
Journal of Surgery 28 (2004): 211–16.
Smith E. and W. Dawson, Egyptian Mummies. London:
G. Allen & Unwin, 1924.
Von Deiner, G. and W. Westendorf, Grundriβ Der Medizin
Der Alten Ägypter I-IX. Berlin: Akademie-Verlag,
1954–1973.
Westendorf, W. Handbuch Der Altägyptischen Medizin 1–2.
Leiden-Boston-Köln: Brill, 1999.
Writing 2303
Writing
M ARTHA J. M ACRI
Writing ranks among the most important of human
inventions. Yet it is only one of many forms of visual
communication. Others include sign language, dance,
and bodily adornment, notched tally marks, stones
signposts, as well as the more complex painting and
carving of complex narrative scenes. Some of these are,
like writing, ways of recording information, that is,
of creating a more or less permanent record available
to persons outside the immediate time or space. All
visual communication depends upon at least minimal
familiarity with the cultural conventions of those
producing it. Specifically, writing is a technique for
recording human speech. Today speech, both the
aural and the visual aspects of it, can be recorded and
reproduced electronically. However, prior to the
invention of the wax cylinder phonograph, human
language could only be recorded, however imperfectly,
in written texts.
Writing is an entirely arbitrary system in which each
symbol in a script has an agreed upon value. The
symbols may or may not have any association with a
real world object. Thus, all scripts must be learned
within the context of a specific culture, and within the
context of a specific speech community. Although the
Chinese script can be read by speakers of a number of
different languages, the conventions of the script and
the way it relates to a speaker’s own language must be
painstakingly memorized.
Writing seems to have been invented at least three
times in human history: in Mesopotamia, in China, and in
Mesoamerica. It may be that there were other indepen-
dent inventions, or it may be that all the other writing
systems have been the result of stimulus diffusion – the
spread of a good idea. An evolutionary theory of writing
sees a development from pictures to complex symbol
systems to scripts with word signs to phonetic (sound
based) writing systems.
Logographic scripts (ones in which a picture or
symbol represents a specific word) make use of
rebuses, that is, of using a picture of an object to
represent a word that sounds like the object pictured.
Such a script is always language specific. For example,
using the picture of a honey bee for the verb to be
works only in English. Logographic scripts nearly
W
always include additional elements to clarify meaning
(semantic determinatives) or to indicate the sound of
a word (phonetic complements). Ancient Egyptians
had sets of consonant signs indicating one or two or
three consonants, as well as hundreds of logographic
signs and semantic determinatives. Other examples
2304 Writing
of logographic scripts are Sumerian, Chinese, Naxi
(southern China), and Mi’kmaq (Canadian maritime
provinces).
The next step in an evolutionary theory of writing is
the addition of a significant phonetic component. In
phonetic scripts most signs refer not to words, but
simply to sounds. Of phonetic systems there are two
types: syllabaries and alphabets. A syllable is com-
posed of a vowel and any consonants that immediately
precede or follow it. In practice, in most syllabic
scripts, each sign represents a simple consonant–vowel
combination. There are, for example, individual signs
for pa, pe, pi, po, and pu instead of a single symbol for
the sound p and symbols for each vowel. Syllabic
systems often include a set of logographic signs (such
as the kanji component of Korean and Japanese), and
may have one or two symbols for a single consonant (in
Japanese there is a symbol for n, in Cherokee a symbol
for s). Although early students of writing considered
syllabaries to be merely a step toward the culmination of
the alphabet, syllabic systems have had a long history of
success. Japanese, Linear B, Akkadian, West African
scripts, and Cherokee are only a few examples.
In contrast to multiple inventions of syllabic scripts, the
alphabet seems to have been invented only once. Certainly
there are hundreds of distinctive alphabetic scripts, but
the history of each one can be traced from a single origin
in the Mediterranean region some 3,500 years ago. In
alphabetic systems, each sign represents a single sound – a
single consonant or vowel (the earliest alphabets were
largely consonantal systems). The Proto-Sinaitic, Ugaritic,
and Phoenician scripts are the immediate ancestors of
the Greek and Roman alphabets. Alphabets have an
advantage for languages with consonant clusters such
as English. The word split, for example, in a syllabary
would have to be written with four syllabic signs: si-pi-li-ti,
suggesting the presence of more vowels than the word
has. The sequence of syllabic signs could also spell spilt,
or any of a variety of nonsense words. Alphabetic systems
generally include at least a few logographic symbols,
such as the symbols for numbers.
Any script with a history – one in use longer than a
single generation, or used for more than one language
or one dialect – suffers from the effects of standardiza-
tion and carries a baggage of archaic spellings. Aunt
and ant may or may not be pronounced the same. Beau
and bow, bough and bow reflect language history, and
show clearly that English writing is not a purely
phonetic system. Standardized spellings are much easier
to read than technical transcriptions, in which linguists
using a system such as the IPA, the International
Phonetic Alphabet, try to capture the precise details of a
specific utterance.
Although no script is ever a pure example of the above
categories, each category differs in the usual number
of total signs. Counting the number of individual signs
can sometimes suggest the nature of an unknown script.
Logographic systems contain the most signs, sometimes
numbering in the thousands. Syllabic systems require
only as many signs as there are consonant + vowel
syllables in a language, usually in the range of 40–150
signs. Mixed logographic-phonetic scripts can range
from hundreds into the thousands. Alphabets have
the fewest – only enough to represent the sounds of a
particular language. Only 13 letters are needed to write
the Hawaiian language, compared with 24 symbols in
the Germanic runic script and 44 for Cyrillic. The total
number of signs in an alphabet does not necessarily
match the number of distinctive sounds in a language.
The alphabet used for English has only seven vowel
letters (a, e, i, o, u, w, and w) that are used alone or in
various combinations to represent at least 12 distinct
vowel sounds.
Another way that script types differ is in their
longevity. Two of the best known logographic scripts,
Chinese and Egyptian have had far longer life spans
than the alphabet. An identifiable Chinese script was
functional at least by 1200 BCE, and continues in use
by millions of people today. Egyptian hieroglyphic
writing ceased by the tenth century CE, but has a history
of nearly 4,000 years. Akkadian in Mesopotamia was
used for over 2,000 years. Syllabaries seem to have
briefer lives. Linear B in Mycenaean Greece was used
for about 200 years. The two syllabic components of the
Japanese script, hiragana and katagana, are slightly
over a 1,000 years old, developed by the eighth and
ninth centuries CE, respectively. The Cherokee sylla-
bary dates only from 1820. What of the alphabet?
Probably dating no earlier than the first or second
millennium BCE, its future seems promising. As a tool
of the electronic age, the alphabet is a system adaptable
to any language and is maximally efficient for keyboard
use (this may be, however, because keyboards were
invented for the alphabet).
One of the errors in an evolutionary theory of scripts
is the notion that they become increasingly abstract,
i.e., that over time scripts become more efficient,
using fewer signs and simpler sign forms. Scripts with
a heavy logographic component tend, however, to
develop in the opposite direction, collecting baggage
over time. Egyptian grew from 700 signs in Middle
Egyptian (ca. 2000 BCE) to more than 5,000 signs
by the fourth century CE. The ancient Maya script
added hundreds of logographic signs over the 700 years
of the Classic period (200–900 CE), while the core
of syllabic signs grew at a much slower rate. Chinese
characters have increased from 9,000 in the first
century CE to about 60,000 by the end of the
twentieth century, though today only about 2,400 are
in common use.
These dramatic increases happened for several
reasons. First, cultural and historical events bring

additions to a script over time. This can be seen in the
archaic spellings of many English words that reflect
pronunciations no longer used or used only in certain
dialects. Second, although a purely alphabetic script is
easier to learn and easier to write, logographic scripts,
once mastered, can be read more rapidly. The debate of
whether to teach English-speaking students to learn to
read through phonics or through word recognition
reflects the fact that even readers of alphabetic scripts
make use of both its phonetic and logographic aspects.
Third, universal literacy has not always been an ideal, and
a difficult logographic script has sometimes been seen as
a tool for limiting access to knowledge and prestige.
Finally, and probably most significantly, languages rarely
add a new sound to their sound inventory and rarely add
new syllables. Words, however, are constantly being
added. Since vocabulary is an open-ended component of
language, any logographic script, one in which symbols
represent words, is going to have to grow as new words
come into use.
Some scripts owe their invention and development
directly to contact between neighbors. One example is
the Japanese writing system. Initially Chinese char-
acters were used to approximate the Japanese language,
then syllabic features were added to provide more
specific phonetic information. Today the combination
of kanji (Chinese logographs) with the hiragana and
katakana syllabaries, make Japanese a script with one
of the world’s most arduous learning curves. Sumerian,
a largely logographic system, was adopted by the
Akkadians who like the Japanese added syllabic
features, resulting in a mixed logographic–syllabic
system that had a large syllabic component.
Some students of writing see the adoption of a
logographic script by speakers of a foreign language as
an explanation for the development of phonetic scripts.
Foreign users reinterpret logographs that carry seman-
tic and phonetic information as phonetic symbols
representing sounds without the specific meaning, or as
representing the same meaning but with the sounds it
has in the new language, or they may invent phonetic
signs to add phonetic information or grammatical
detail. Such progressions can be seen in the develop-
ment of Akkadian from Sumerian writing, or of
Japanese syllabaries, and the Korean script – both
influenced by Chinese.
The vocabulary used to characterize a script varies
from one to another. Characters, hieroglyphs, glyphs,
letters, symbols, icons, signs, and graphemes all
describe units of written language. The forms of the
signs may be described as pictographic, representa-
tional, linear, cuneiform, abstract, and geometric. The
organization of the units may be columnar, clustered,
horizontal (left to right or right to left), boustrophedon
(back and forth, literally, as the ox plows). Within the
texts themselves divisions between words, phrases,
Writing 2305
sentences, and larger units may be ignored or may be
indicated by spaces, larger symbols (e.g., upper case
letters), or a variety of punctuation conventions.
Language that is recorded in writing represents only a
small subset of language that is spoken. Nearly every
culture places specific limits on the kinds of information
presented and on the manner of presentation. Commercial
records, calendrical notations, prayers, magical spells,
laws, census records, land deeds, court records, sacred
and political histories, diaries, personal letters – these are
among the earliest written materials. Novels, commercial
advertising, and to-do lists are relatively more recent
developments. The surviving examples of some ancient
scripts are limited only to one or two genres.
Spoken language reflects an awareness of the
speaker’s audience, while written language may
ultimately reach audiences never dreamed of. Written
language tends to be more deliberate, generated at a
much slower pace than speech. Nor does it have the
meta-linguistic features that add dramatic force and
subtle nuances to a speech act. Efforts to indicate
variation in tempo, volume, and pitch (such as with
underlining, boldface, accent marks, exclamation, or
question marks) fall far short of the tools available to a
speaker. Missing, too, are facial expressions, body
posture, and hand gestures.
In contrast to recording language, there have
developed a number of systems that approach writing,
but do not fit the strictest definition of writing. Several
of these hieroglyphic scripts are or were, in fact, readable
within the culture that produced them, but with no or
with only minimal phonetic components they can have
only uncertain meaning to the uninitiated. One example is
the Mixtec writing known from eight pre-Columbian
screenfold books. In these a complex iconography is tied
to narrative images recording sequences of historical
events and genealogies. Although the meaning is often
clear, it is impossible to know exactly how the texts would
have sounded when they were read – impossible to
know precisely what words were associated with the
images. Another example is the Naxi Dongba writing of
southern China dating from the thirteenth century. This
hieroglyphic script is still used and studied today. The
meanings of the signs are complex, with minimal
phonetic information indicated. Mixtec and Naxi repre-
sent yet another challenge to the evolutionary model.
Rather than pictographic scripts always preceding true
writing, some of them were created by cultures imitating
neighboring scripts. W
Also implied in the evolutionary theory of writing is
a gradual development from picture signs to syllabic
signs. Many books on the history of writing show
charts of simple pictures for objects next to related but
more abstract forms that exist in a particular script – the
most common scripts pictured are Chinese, Akkadian,
and sometimes even the Greek alphabet. However,
2306 Writing
direct evidence of this gradual development from
pictures to a phonetic script is seldom available. An
alternate scenario is that at some point in time one
person, or a small group of persons, took an existing
symbol system, and from it chose a set of signs to
produce a phonetic system, usually a syllabary. We
know that this happened with Sequoyah, the man who
invented the Cherokee syllabary. History also tells us
that this is what happened in the fifteenth century when
King Sejong, or a committee created by him, developed
the Korean Han’gul script. Because a phonetic script is
a complex interdependent system it can only function
when a complete set of phonetic signs are in place. For
most scripts, this crucial moment of creation remains a
conjecture.
Some have asserted that writing, specifically the
alphabet, gave humans the ability to order and to
classify information. Linguistic and cognitive anthro-
pologists, however, have demonstrated repeatedly that
complex categorization also exists in societies without
writing. What then does writing contribute? Certainly
it contributes a memory beyond the life span of a single
person, knowledge that is independent of oral trans-
mission. This enables an accumulation of knowledge
across time and across cultures. Writing has given us
detailed insight into earlier societies. Several lan-
guages, such as Sumerian, Hittite, Etruscan, and
Tocharian, have no direct descendants, and are known
only through their written texts.
This essay speaks about script types and script
developments. For those who are interested in the variety
of the world’s scripts there are a number of excellent
introductory texts, some generously illustrated. For
example, see Coulmas (1989, 1999), DeFrancis (1989),
Pope (1999), Robinson (1999, 2002), Rogers (2005), and
Sampson (1990). One volume stands out for its quality
and comprehensive coverage: The World’s Writing
Systems, Daniels and Bright (1996). Also excellent are
the individual volumes of the Reading the Past series:
Bonfante (1990) Chadwick (1987), Cook (1987), Davies
(1987), Healey (1990), Houston (1989), Page (1987), and
Walker (1987). Two dramatic stories of decipherment are
Chadwick (1967) for Linear B and Coe (1999) for the
Maya script. Boone (2000) and Boone and Mignolo
(1994) offer enlightened discussions of nonphonetic
systems. The potential of the as yet undeciphered Inca
knotted khipu as a record of detailed information, perhaps
including speech can be found in Urton (2003) and
Quilter and Urton (2002).
Recent years have seen a growing number of
websites, both about scripts in general, as well as sites
devoted to a single script. Below are listed two general
websites for writing systems, both with links to many
other resources, and several sites relating to specific
scripts or regions. Searching on the web for writing
systems, or for any country, language, or script name
will yield a large number of sites. Some of these are
sponsored by academic institutions, a few are commer-
cial, and a few are sponsored by individuals. Their
quality varies, though even sites with incomplete or
nonstandard information often provide useful images.
References
Bonfante, Larissa. Etruscan. Reading the Past. Vol. 8.
Berkeley: University of California Press, 1990.
Boone, Elizabeth Hill. Stories in Red and Black: Pictorial
Histories of the Aztecs and Mixtecs. Austin: University of
Texas Press, 2000.
Boone, Elizabeth Hill and Walter D. Mignolo, eds. Writing
Without Words: Alternative Literacies in Mesoamerica and
the Andes. Durham: Duke University Press, 1994.
Chadwick, John. The Decipherment of Linear B. Cambridge:
Cambridge University Press, 1967.
---. Linear B and Related Scripts. Reading the Past. Vol. 1.
Berkeley: University of California Press, 1987.
Coe, Michael D. Breaking the Maya Code. New York:
Thames and Hudson, 1999.
Cook, B. F. Greek Inscriptions. Reading the Past. Vol. 5.
Berkeley: University of California Press, 1987.
Coulmas, Florian. The Writing Systems of the World.
Cambridge: Basil Blackwell, 1989.
---. The Blackwell Encyclopedia of Writing Systems. Malden:
Blackwell Publishers, 1999.
Daniels, Peter T. and William Bright eds. The World’s Writing
Systems. New York: Oxford University Press, 1996.
Davies, W. V. Egyptian Hieroglyphs. Reading the Past. Vol. 2.
Berkeley: University of California Press, 1987.
DeFrancis, John. Visible Speech: The Diverse Oneness of
Writing Systems. Honolulu: University of Hawaii Press, 1989.
Healey, John F. The Early Alphabet. Reading the Past. Vol. 9.
Berkeley: University of California Press, 1990.
Houston, Stephen D. Maya Glyphs. Reading the Past. Vol. 7.
Berkeley: University of California Press, 1990.
Macri, Martha J. and Matthew G. Looper. The New Catalog
of Maya Hieroglyphs: Vol. 1, The Classic Period Inscrip-
tions. Austin: University of Oklahoma Press, 2003.
Page, Raymond Ian. Runes. Reading the Past. Vol. 4.
Berkeley: University of California Press, 1987.
Pope, Maurice. The Story of Decipherment: From Egyptian
Hieroglyphs to Maya Script. New York: Thames and
Hudson, 1999.
Quilter, Jeffrey and Gary Urton eds. Narrative Threads:
Accounting and Recounting in Andean Khipu. Austin:
University of Texas Press, 2002.
Robinson, Andrew. The Story of Writing: Alphabets,
Hieroglyphs, and Pictograms. New York: Thames and
Hudson, 1995.
---. Lost Languages: The Enigma of the World’s Undeci-
phered Scripts. New York: McGraw-Hill, 2002.
Rogers, Henry. Writing Systems: A Linguistic Approach.
Malden: Blackwell, 2005.
Sampson, Geoffrey. Writing Systems: A Linguistic Introduc-
tion. Stanford: Stanford University Press, 1990.
Urton, Gary. Signs of the Inka Khipu: Binary Coding in the
Andean Knotted-String Records. Austin: University of
Texas Press, 2003.
Walker, C. B. F. Cuneiform. Reading the Past. Vol. 3.
Berkeley: University of California Press, 1990.

Websites General
Omniglot: A Guide to Writing Systems ▶http://www.
omniglot.com/index.htm.
Writing Systems of the World ▶http://logos.uoregon.edu/
explore/orthography/.
Africa African Writing Systems
▶http://www.library.cornell.edu/africana/Writing_Systems/
Welcome.html.
The Demotic Dictionary Project ▶http://oi.uchicago.edu/OI/
PROJ/DEM/Demotic.html.
Easter Island
The Rongorongo of Easter Island ▶http://www.rongorongo.
org/index.html.
Mediterranean/Mesopotamia
Cuneiform Digital Library Initiative ▶http://cdli.ucla.edu/
index.html.
Hittite ▶http://www.asor.org/HITTITE/HittiteHP.html.
The Origins and Emergence of West Semitic Alphabetic
Scripts ▶http://as3.lib.byu.edu/
imaging/negev/.
India
Languages and Scripts of India ▶http://www.cs.colostate.
edu/
malaiya/scripts.html.
Mexico and Central America
Foundation for the Advancement of Mesoamerican Studies
▶http://www.famsi.org/.
Mesoweb ▶http://www.mesoweb.com/.
Glyph Dwellers ▶http://nas.ucdavis.edu/NALC/glyphdwellers.
html.
Asia
Library of Congress: Asian Reading Room ▶http://www.loc.
gov/rr/asian/.
Selections from the Naxi Manuscript Collection ▶http://
memory.loc.gov/intldl/naxihtml/naxihome.html.
Japanese Phonic Characters ▶http://www.uweb.ucsb.edu/

christi/main.html.
Writing in China
U TA L AUER
The origin of a comprehensive writing system in China
is a matter of debate. According to an ancient Chinese
legend, writing was invented by the three mythical
emperors between the twenty-ninth and the twenty-
sixth centuries BCE. The earliest evidence for some
kind of script is marks on pottery excavated from the
Banpo site near Xi’an in Shanxi-province, dating from
the Neolithic Age. Proof for a highly developed writing
system comes from oracle bones ( jiagu wen). Datable
to around 1300 BCE, inscriptions on bovine, pig, deer
shoulder blades or on turtle shell were used for
divination. The elite of the Shang dynasty (sixteenth
Writing in China 2307
to eleventh century BCE) employed a question and
answer model, through which those in power would
ask a question, incise it into the bone, and the shaman
or priest would drill holes into it, insert a piece of
burning coal, interpret the cracks produced by the hot
coal, and then write down the answer. About 2,000
different characters have been found on Shang oracle
bones, but only half of these have so far been
deciphered. Another form of surviving document is
short texts cast on the inside of bronze vessels. There
are numerous examples of late Shang inscriptions,
typically rather short, whereas in the succeeding
Zhou period (eleventh century BCE to ca. 771 BCE)
inscriptions on bronze ritual vessels were considerably
longer. The type of script employed during the late
Zhou period was called great seal script (dazhuan).
One of the most important steps towards the
development of a comprehensive writing system was
the unification of the empire under the rule of emperor
Qin Shihuang (r. 221–206 BCE) in 221 BCE. Under
the jurisdiction of his Prime Minister Li Si, writing was
standardised into the small seal script (xiao zhuan).
This script comprised about 12,000 characters (zi).
Today, small seal script is still in use on seals and in
advertisements. Seals had been made since at least the
late Zhou period. They were used instead of a signature
and served official, commercial, personal and literary
purposes.
With the advent of the Western Han dynasty
(206 BCE–9 CE) a new type of script came into use,
clerical script (lishu). It was easier to write than the
angular small seal script and thus more suitable for the
ever-increasing amount of official, government related
texts. Based on clerical script a kind of shorthand
developed for less formal occasions like writing letters
or jotting down notes. Literally translated into English
it is called grass script (caoshu), but is also frequently
referred to as draft script, running script or fully cursive
script. To avoid confusion of technical terms, it is best
to use the unambiguous Chinese terminology. During
the Han dynasty, the potential of writing as a medium
for artistic expression was fully recognised. It was then,
that the art of calligraphy (shufa) was born. Another
less abbreviated form which emerged from clerical
script is semi-cursive script (xingshu).
During the fifth century, the last type of script was
codified, regular script or standard script (kaishu),
which is still in use today. After that, the evolution of
the Chinese writing system was complete. This had a W
culturally unifying effect this had to a high degree.
Chinese characters serve as a binding element, as an
identification factor and as bearers of an uninterrupted
tradition.
With the end of imperial rule and the founding of
the Republic of China in 1912, a debate about the
simplification of characters began. Some argued to
2308 Writing in Egypt in its literary context
abolish Chinese characters entirely and to replace them
with a romanization system in order to promote literacy.
Several reforms followed until finally in 1964 a list
of simplified characters ( jianti zi) was approved. Today
simplified characters are used in the People’s Republic
of China whereas other countries which employ Chinese
characters have maintained the complex characters
( fanti zi).
In 1979 the government of the People’s Republic of
China adopted the Hanyu Pinyin system, a phonemic
notation and transcription to Roman script. Pinyin is
very useful as a cataloguing method and for writing
Chinese on computers.
Essential for the development of the art of calligra-
phy was the material used for writing, namely the
pliable brush, absorbent paper and ink. The earliest
surviving documents written with a brush and ink
on strips of bamboo date to the third and second
centuries BCE. Already by the fifth century BCE silk
was also in use as a writing ground. Silk existed in
China since Neolithic times. Archaeological evidence
shows that paper was being made from the first or
second century BCE. Paper was much cheaper than
silk, relatively easy to make and less cumbersome than
bamboo strips. The availability of paper made possible
the steady increase in writing throughout the empire.
The earliest examples of ink are found in traces of
writing dating to the fourteenth century BCE. Chinese
ink is made from carbon, lampblack and glue which
is then dried and moulded into the form of sticks or
cakes. To obtain liquid ink, the ink stick has to be
ground by hand on an inkstone with the addition of
water. During the tenth century CE, writing equipment
(brush, paper, ink and inkstone) came to be known as
the “four treasures of the scholar’s studio” (wenfang
si bao).
Since ancient times, calligraphy belonged to the six
liberal arts (liu yi) practised by the social elite of
the country. When looking at a work of calligraphy, the
viewer can relive the process of its creation. The
writing speed, pressure, energy, any moment of
hesitation remains visible on the paper. The flexible
brush allows for an endless variation of stroke shapes.
Soon, the styles of individual masters of calligraphy
were canonized. For transmitting calligraphic styles,
various copying methods and the formation of collec-
tions were important. Before the invention of printing
in the Tang dynasty (618–906 CE), several techniques
for copying were employed. Exact copying (mu) is
basically a tracing technique. To make a free copy (lin)
the original is placed next to the sheet of paper on
which the calligrapher reproduces the work at his own
speed. Essential for a wider circulation of calligraphy
styles were ink rubbings. For this reproduction method,
a copy of the original written in ink is pasted to a stone
or wood surface and then meticulously engraved into
the surface through the copying paper. Then, a
moistened paper is placed over the engraved calligra-
phy and pushed into the incised areas with a brush.
Finally, a pad soaked in ink is dabbed over the surface.
The background of such a rubbing appears black and
the characters white.
All types of script are employed in calligraphy but
semi-cursive script and grass script are the two most
suitable for artistic expression. Among the most
notable calligraphers are Wang Xizhi (307–365) and
his son Wang Xianzhi (344–388), Yan Zhenqing
(709–782), Huaisu (725–785), Su Shi (1036–1101),
Huang Tingjian (1045–1105), Mi Fu (1051–1107),
Zhao Mengfu (1254–1322), Wen Zhengming
(1470–1559) and Dong Qichang (1555–1636). Since
the fourth century CE records and art historical treatises
on calligraphy exist. Calligraphy as the embodiment
and a highlight of Chinese culture forms an important
part of the imperial art collection.
In 1949, when the People’s Republic of China was
founded and the Guomindang defeated, the Guomin-
dang took a substantial part of the former imperial art
collection to Taiwan, where it was housed at the
National Palace Museum Taipei. The remainder of the
collection stayed at the Palace Museum Beijing. From
the beginning of the twentieth century onwards,
collectors outside Asia developed a keen interest in
calligraphy. Private collections and museums in the
USA and in Europe now own a number of representa-
tive works of Chinese calligraphy. Research at
universities on the history and the aesthetics of the art
of writing is particularly prominent at some major US
universities and in Germany.
In China, both the practise and theory of calligraphy
are taught at art academies. Exhibitions and competi-
tions are held regularly. Some contemporary Chinese
calligraphers have gained recognition outside Asia and
their works are on display in museums. Calligraphy is
an art that is very much alive. Not only is the tradi-
tion passed on but there is also experimental writing.
Young artists explore and challenge the basic princi-
ples of writing in their works. The brush might be
replaced by a neon light, ink by water, thus high-
lighting the act of writing and pointing towards the
perishable material.
Writing in Egypt in its Literary Context
W ILLIAM S. A RNETT
The evolution of Egyptian writing (hieroglyphs) is
traditionally dated to Dynasty 1 of ancient Egypt’s Proto-
Dynastic Period ca. 3100–3000 BCE. Until recently,

priority of first place in the history of writing seemed to
belong to the Sumerians of ancient Mesopotamia (now
southern Iraq). However, the recent discovery of labels in
tomb U-j at Abydos, Egypt, tomb of the Scorpion King,
have provided evidence of writing in Egypt as early as
3250–3200 BCE. The earliest Egyptian attempts at
writing consisted of potmarks, designs painted on jars,
and of motifs carved on the surfaces of cosmetic slate
palettes and other artifacts found in the graves of
Predynastic Egyptians at a number of sites (ca. 4500–
3000 BCE) and in temple deposits at Hierakonpolis.
These examples were clearly precursors to the classical
Egyptian hieroglyphs (Arnett 1982; Elkins 1999; Casson
1965: 141–148; Gardiner 1978).
The hieroglyphs worked best on stone, although they
were also written on papyrus and other materials.
Papyrus was a parchment-like paper made from the
reed plant of that name which once grew in abundance
along the banks of the Nile River. It was the most
common material used for official record keeping,
correspondence, literary works, administrative docu-
ments, etc. For these purposes the Egyptians developed
a more cursive style of writing. This occurred early
in the development of hieroglyphs. Scholars call the
earliest type of this writing hieratic (literally “priestly”
writing), although it was not used exclusively by the
priests. The scribes used lampblack and wrote with reed
pens when writing on papyrus and brushes when using
wood and other materials such as ostraca.1 Later, in the
seventh century BCE, the Egyptians developed another
even more cursive system that is known as demotic, or
popular. Generally, both of these more rapid writing
systems were written on papyrus (Figs. 1–3).
The quantity of such examples of Egyptian proto-
writing suggests that Egyptian writing may have
evolved as early as (if not earlier than) the Sumerian
cuneiform system so prominent in Mesopotamia and
much of the ancient Near East from ca. 3000–1000
BCE. It certainly proves that the Egyptians developed
their own, unique writing system and that, if they
borrowed anything from the Mesopotamians, it was the
idea of writing (Arnett 2004; Stiebing 2003: 110–112)
(Fig. 4).
The earliest extant Proto-dynastic writings consist of
numerous brief texts: funerary stelae, ivory labels and
plaques, and a few official monuments and royal
inscriptions found in situ at locations both in Egypt
proper and in the Sinai Peninsula. These early texts
1
An ostracon is a piece of pottery usually broken off from a
vase or other earthenware vessel. In archaeology, ostraca may
contain scratched in words or other forms of writing which
may give clues as to the time when the piece was used. The
word is derived from Greek ostrakon meaning a shell or a
shard of pottery used as a voting tablet. The plural of ostracon
is ostraca.
Writing in Egypt in its literary context 2309
Writing in Egypt in its Literary Context. Fig. 1 Funerary
Papyrus of Hunefer. The Sun god is shown in the form of a
great cat killing the serpent of darkness. British Museum
(EA9901.8). Photo by Jon Bodsworth. Used with his kind
permission.
reveal a hieroglyphic writing system that was already
fully developed, which led earlier scholars to believe
that the Egyptians had borrowed their writing system
from the outside world, presumably Mesopotamia. The
earlier use by the Sumerians of “symbolic tokens and
ideographic inscriptions” as means of the accounting
and recording of quantities of products and livestock
was assumed to have led to the development of
the earliest phonetic writing system in ancient Sumer.
But the labels of Tomb U-j at Abydos can be read
phonetically and may be earlier than the earliest known
Sumerian writing (Stiebing 2003: 112–113). We have
evidence of early trade and cultural exchange between
late Predynastic Egypt and southern Mesopotamia, and
it is highly likely that ideas as well as goods and
materials were being exchanged. However, the fact
remains that the two earliest systems of writing
(Sumerian cuneiform and Egyptian hieroglyphs) use
entirely different signs and methods of inscription.
While some Mesopotamian writing has survived on
stone artifacts, the majority was inscribed upon clay
tablets. In Egypt, stone stelae and walls of tombs or
temples, stone chips (or ostraca), pieces of wood, and
papyrus paper (the use of which dates back to late
Predynastic times) were the primary media upon which
the Egyptians did their writing. It is more likely that
writing evolved in both areas at about the same time
(Mark 1997; Wilkinson 2001: 150–182) (Fig. 5). W
These early texts provide only minimal details of
official political and historical events as well as the
accomplishments of the owners of the tombs in which
they were found (Lichtheim 1975: (vol. 1) 3; Pritchard
1955; Breasted 1906; Lewis 2003). It is the general
consensus that Egyptian writing was initially created
for the sole purpose of satisfying the needs of the state
2310 Writing in Egypt in its literary context
Writing in Egypt in its Literary Context. Fig. 2 Papyrus of Hunefer. The Judgment, from the papyrus of the scribe Hunefer.
19th Dynasty. Jackal-headed Anubis conducts Hunefer to the balance. The monster Ammut crouches beneath the balance so
as to swallow the heart should a life of wickedness be indicated. British Museum (EA9901). Photo by Jon Bodsworth. Used
with his kind permission.
Writing in Egypt in its Literary Context. Fig. 3
Padiamenet Papyrus. Hieratic Book of the Dead of
Padiamenet, chief baker of the domain of Amun. Dynasty 22.
British Museum (EA10063). Photo by Jon Bodsworth. Used
with his kind permission.
to record all of its administrative, political, economic,
diplomatic, and military activities and transactions
(Aldred 1998).
King Lists constitute a genre of Egyptian writing
reflecting the above – stated purposes. The earliest, the
Palermo Stone, provides the names of kings of the Late
Predynastic period through early Dynasty 5. Primarily
a list of yearly events and statistics, the main value of
this monument is in the ordering of kings by the length
of reign and the listing of their achievements as well as
recording annual Nile flood heights, a periodic census
of men and livestock, and occasional references to
public works projects and foreign trade. For instance,
there is a reference to forty shiploads of cedar wood
imported to Egypt in 1 year during the reign of Snefru
of Dynasty 4 (Gardiner 1978; St John 2003; Wilkinson
2000). Other important king lists include the Turin
Canon, the lists of royal ancestors of Sety I and Ramses
II at Abydos, the Royal List of Karnak, and the Royal
List of Sakkarah (Clayton 1994; Shaw 2000) (Figs. 6
and 7).
But, according to Lichtheim, the “first major
application” of Egyptian writing was in the omnipres-
ent Offering Lists found in the tombs of the well-to-do
Egyptians of the Old Kingdom (Figs. 8–10).
These offering lists consisted of the names of the
tomb owners, their family members, a record of
the offices and titles held by them, and an accounting
of the funerary offerings they required. By Dynasty 5,
the Offering List was shortened into a Prayer for
Offerings (Lichtheim 1975: 3; Breasted 1906: Vol. 1,
75–175) (Figs. 11–13).
As a result, wall space was freed for the development
of a descriptive narrative containing historical and
autobiographical material which the Egyptian nobles
(administrators, prophets, judges, “overseers of the
King’s land,” viziers, architects, soldiers, scribes,
priests, etc.) had written/sculpted/painted on the walls
of their tombs and/or tomb chapels. These texts and
accompanying illustrations provide us with portrayals
of the tomb owner, his family, and his servants
(Figs. 14–17).
These texts, also called “Biographies of the Nobles,”
often contain gross exaggeration and considerable
boastfulness, and were meant only for their family,
priests, and the gods they were meant to impress. Their
discovery in modern times has given historians access
to important primary source material for the recon-
struction of ancient Egyptian history. They provide
us with glimpses of the public careers and achieve-
ments of the administrative, bureaucratic, and military

Writing in Egypt in its Literary Context. Fig. 4 Jackal Images: examples which illustrate the evolution of Egyptian writing
from the earlier prototypes (proto-writing) to the historical hieroglyphs. Artwork and image by William S. Arnett. Used with his
permission.
Writing in Egypt in its Literary Context. Fig. 5 Ugaritic
Alphabet using cuneiform characters. From Ugarit (Ras
Shamrah), Syria. Aleppo Museum. Photo by William S.
Arnett. Used with his permission.
Writing in Egypt in its Literary Context. Fig. 6 Royal
Ancestors of Sety I and Ramses II (Dynasty 19). Royal
cartouches on the walls of Sety’s Temple dedicated to Osiris
at Abydos. This scene shows part of the list of previous
Egyptian Pharaohs. Photo by Jon Bodsworth. Used with his
kind permission.
Writing in Egypt in its literary context 2311
Writing in Egypt in its Literary Context. Fig. 7 Sety I and
Ramses II. Seti I and his son Ramesses II offering praise to a
list of royal ancestors. Temple of Sety I at Abydos. Photo by
Jon Bodsworth. Used with his kind permission.
personnel who served the kings throughout the history
of Dynastic Egypt. Tomb biographies also exist for
nobles and some commoners throughout the rest of
ancient Egyptian history. Lichtheim notes that, from the W
First Intermediate Period on, for many individuals, a
shortened biography would be relegated to the tomb
stela which also contained an offering prayer and an
offering scene, making the stela a “self-contained
memorial” (Breasted 1906; Lichtheim 1975: 8; 15–28).
A second literary genre, Instructions Literature, is
among the Old Kingdom’s greatest contributions to
2312 Writing in Egypt in its literary context

Writing in Egypt in its Literary Context. Fig. 8 Mastaba

(tomb) of Mereruka. The tomb of Mereruka is the largest
extant noble’s tomb at Saqqarah. Mereruka was the vizier of
Pharaoh Teti, first king of Dynasty 5 (2345–2333 BCE).
Photo by William S. Arnett. Used with his permission.

Egyptian writing. This category is also known as

Wisdom Literature, which Miriam Lichtheim calls the Writing in Egypt in its Literary Context. Fig. 9 Ka statue
“second major literary genre created in the Old of Mereruka in the Offering Chapel of his mastaba. The
family and/or funerary priests left food offerings before
Kingdom” (Lichtheim 1975: 5). They are written in
Mereruka’s statue in order to provide continual sustenance
the guise of the advice of a King, Prince or Vizier to his for him in the afterlife. Photo by William S. Arnett. Used with
son. The earliest known example is the Instructions of his permission.
Hardjedef. The sage Hardjedef (Hordjedef) was the son
of Kheops (Khufu) in Dynasty 4. He also figures as a
participant in the Three Tales of Wonder (from the
Papyrus Westcar in the Egyptian Museum of Berlin).
Prince Hardjedef claims to have written the instructions
for his son, Au-ib-re. Not too much of his Instruction
has survived, but his name is always listed among those
of the great sages of Egypt (Simpson 1973: 340–341).
Lichtheim (1975: 5–7 and 58–59) considers all three
of the earliest Instructions (Hadjedef, Dynasty 5,
and Kagemni and Ptahhotep, late Dynasty 6) to be
pseudepigraphical2 works.
Another well-known example of this genre is the
Instructions of Ptahhotep. The sage Ptahhotep, who,
supposedly, was the vizier of the Fifth Dynasty pharaoh
Izezi (2388–2356 BCE), provides maxims of good and
polite conduct and of good speech.
These works constitute a form of political propagan-
da in a literary format. They posit that order and ma’at
(justice and balance) are only possible when there is a
Writing in Egypt in its Literary Context. Fig. 10
Mereruka Wall Scene. The diminutive figure of Meri-Teti,
son of Mereruka, is seen at the feet of the Vizier Mereruka.
Mastaba of Mereruka (Dynasty 5) at Saqqarah. Photo by
2
Pseudepigrapha (from the Greek words pseudos = false and Brian Anderson. Used with his kind permission.
epigrapho = write) describes texts whose claimed authorship
is unfounded in actuality. The authenticity or value of the
work itself, which is a separate question for experienced
readers, often becomes sentimentally entangled in associa-
strong and benevolent monarch on the throne of a
tion. Yet few Hebrew scholars would insist that the Song of united Egypt. In addition, they offer advice on proper
Solomon was written by the King of Israel, or ascribe the behavior that reflects the mores of nobles of Dynasty 5
Book of Enoch to the prophet Enoch. and Dynasty 6.
 Writing in Egypt in its literary context 2313

Writing in Egypt in its Literary Context. Fig. 11 Deir

el-Bahri Offerings. In the private tombs of the nobles of the
Old Kingdom, the List of Offerings consisted of prayers and
items meant to provide sustenance to the decedent in the
afterlife. In a similar manner, the royal temples of the New
Kingdom, Kings portrayed themselves making offerings to the
gods (see Fig. 6). Here we see a wall relief from Hatshepsut’s Writing in Egypt in its Literary Context. Fig. 12 The
funerary temple that depicts her offerings to the god Amun. female Pharaoh Hatshepsut presents offerings to the God
Photo by William S. Arnett. Used with his permission. Amun. Funerary Temple of Hatshepsut at Deir el-Bahri.
Photo by William S. Arnett. Used with his permission.

Writing in Egypt in its Literary Context. Fig. 13 The Funerary Temple of Hatshepsut at Deir el-Bahri (Dynasty 18). Photo W
by William S. Arnett. Used with his permission.

A third example, The Instruction to Kagemni, most beginning of Kagemni’s Instruction is lost, and its true
likely written during Dynasty 6, alleges to be the advice date and authorship are unknown.
given to Kagemni, the vizier of Huni, the last king of The Instruction of Merikare was supposedly written
Dynasty 3 and of Snefru, the founder of Dynasty 4. The by King Khety II (ca. 2100 BCE), the father of
2314 Writing in Egypt in its literary context
Writing in Egypt in its Literary Context. Fig. 14 Stele of
Nefertiabet. Princess Nefertiabet, a daughter of the Pharaoh
Khufu (Dynasty 4), is seen here seated at her offering table in
her tomb at Giza. This wall section is in the Louvre Museum
(E22745) in Paris. Photo by Jon Bodsworth. Used with his
kind permission.
Writing in Egypt in its Literary Context. Fig. 15 Stele of
Rahotep (Dynasty 4) from Meidum. British Museum
(EA1242). Photo by Jon Bodsworth. Used with his kind
permission.
Merikare of Dynasty 10. His (Merikare’s) capital was
at Herakleopolis. However, most scholars believe that
it was actually written by scribes at the court of
Merikare. Its uniqueness lies in its new dimension as a
royal inscription in which a dying king bequeaths the
throne to his son and successor and offers him advice
on how to be a good king (Lichtheim 1975: 9–10).
The Instruction of Amenemhet I (or Amenemes I),
supposedly written by Amenemhet I to his son
Sesostris I (Senwosret I), of the Middle Kingdom
(Dynasty 12), is a more famous example of this genre.
The serenity and balance in Egyptian society as it
appeared during the Old Kingdom had been replaced
Writing in Egypt in its Literary Context. Fig. 16 The
Necropolis of Gizeh. In the shadows of the Great Pyramids,
the tombs of the princes and high-ranking nobles of Dynasty
4 were buried in tombs arranged in orderly rows so that they
would be with their king in the afterlife. Photo by William S.
Arnett. Used with his permission.
Writing in Egypt in its Literary Context. Fig. 17 The
Great Pyramids of Gizeh. Viewed from the desert to the west,
the pyramids of Khufu, Khafre, and Menkaure are ancient
Egypt’s most famous monuments. Photo by William S.
Arnett. Used with his permission.
by the political and economic turmoil and social
upheaval of the First Intermediate Period. Even though
order had been restored by the Kings of the Middle
Kingdom, our literary evidence suggests that the crown
represented a heavier burden for the monarchs of the
Middle Kingdom, and that the dangers of civil strife
and revolution were lurking just below the surface. The
apparent assassination of Amenemhet I (alluded to in
both the Instruction of Amenemhet I and The Story of
Sinuhe) suggests that things had gone awry and that
men were doing evil things. In the Instruction of
Amenemhet, the dead King (Amenemhet I) appears as a
ghost to his son Sesostris I (or Senwosret I), tells of his
assassination in the palace at the hands of his own
palace guard, and suggests that it was instigated by

someone in the harem. He advises his son to rule with a
strong hand and to trust no one in the “day of evil”
(Erman 1927: 72–74; Pritchard 1955: 418–419;
Simpson 1973: 193–197; Parkinson 1997: 203–211;
Lichtheim 1975:135–139).
The Teaching of Duaf’s Son Khety, elsewhere known
as the “Satire of the Trades,” also dates to the Middle
Kingdom and proclaims that the most exulted profes-
sion is that of the scribe. The writer belittles all other
occupations by listing all the hardships experienced by
those employed in them (Parkinson 1991: 72–76).
The Prophecies of Neferti (James 1979:102;
Lichtheim 1975: 8–9 and 139–145; Erman 1927:
110–115; Pritchard 1955: 444–446; Simpson 1973:
234–240), the Admonitions of Ipuwer, and the
Prophecy of Nefer-rohu represent a new departure in
Egyptian literature. Neferti was a lector priest (literally,
“he who carries the ritual”) at Bubastis during the reign
of Amenemhet I. This composition purports to be the
prophetic vision of someone at the court of Snefru of
Dynasty 4 who predicted the coming of the troubles of
the First Intermediate Period and of Egypt’s savior,
Amenemhet I, the first king of Dynasty 12. In other
words, this piece was probably written to justify the
seizure of the throne of Egypt by Amenemhet I. Neferti
and Ipuwer address the problem of evil and proclaim
that a strong monarchy is the only way to protect Egypt
from its dangers. The pronouncements of the sage
Ipuwer are reminiscent of those of the Biblical Prophets
who chastised the Kings of Israel for not doing their
religious and political duties.
The Admonitions of Ipuwer was probably written
in the early years of Dynasty 12 and is generally
interpreted as being either a warning to a king about the
disastrous consequences of misrule or else as a means
of contrasting the faithfulness of a good king
(presumably one of the Dynasty 12 rulers) with the
inadequacies of a bad king whose failure to tend to the
religious and political duties of the throne resulted in
the chaos and anarchy of the First Intermediate Period.
Nefer-rohu foresees the coming of Amenemhet I and
speaks of him in such familiar terms that scholars are
convinced that his prophecy was actually written during
the reign of Amenemhet (Erman 1927: 110–115;
Pritchard 1955: 444–446).
The Song of the Harper dates to the First Intermedi-
ate Period, the era of disunity and civil and social unrest
following the collapse of the Old Kingdom. It reiterates
the above-mentioned concern for the loss of order and
expresses the sense of futility in providing and
protecting the “houses of eternity” (tombs) of the kings
and nobility. In addition, the song offers an unprece-
dented philosophy of pessimism and skepticism con-
cerning the shortness of life and the uncertainties of
life after death (Lichtheim 1975: 193–197, 222–235;
James 1979: 98; Pritchard 1955: 18–22; Parkinson
Writing in Egypt in its literary context 2315
1991: 36–37; Simpson 1973: 57–74; Erman 1927:
14–29). Was this pessimism truly reflective of a new
Dawn of Conscience, as Breasted suggested (1976), or
was it simply an irreverent farce from some popular
poem or drinking song?
The Middle Kingdom (considered by later genera-
tions of Egyptians to have been the golden age of
Egyptian culture and history for its art, calligraphy and
literature) also produced a popular piece of prose,
known as the Story of Sinuhe, a work of historical
fiction. It depicts the adventures of an Egyptian
nobleman who fled into exile in Palestine, perhaps
because he feared he would be implicated in the murder
of King Amenemhet I, and only returns near the end of
his life to be buried in his homeland. The story also
presents Senwosret I (Sesostris) as a benevolent king
who restores ma’at (truth, the world order, justice, etc.)
and welcomes Sinuhe back to Egypt with open arms. It
provides the earliest literary description of Palestine
and portrays it in Bible-like fashion as a land of milk
and honey: “Abundant was its honey, many sweets
were made for me, and milk dishes of all kinds.”
Scholars are fairly certain that they know the identity
of the author of this famous story. His name was
Achthoes. The Papyrus Chester Beatty IV, in the British
Museum (#10684) avers that, “It was he who made a
book as the Instruction of King Sehetepibre.” The latter
is the name taken by Amenemhet I when he assumed
the throne. Achthoes is also known to have written the
previously mentioned Instruction book entitled the
Satire of the Trades (James 1979: 98; Parkinson 1997:
21–53; Erman 1927: 14–29; Pritchard 1955:18–22;
Simpson 1973: 57–74; Lichtheim 1975: 222–235).
The Tale of the Eloquent Peasant, like so many of
our surviving Egyptian literary texts, is also a product
of the Middle Kingdom. In fact all of the manuscripts
containing this story come from that era. It may be the
mere chance of discovery, but it could also mean that,
unlike other tales from the 12th Dynasty, this particular
example did not enjoy continued popularity during the
later New Kingdom. Social equity and the right of
commoners to expect fair treatment at the hands of the
nobility and the government are the central themes
along with the unexpected eloquence of a peasant
(Pritchard 1955: 407–410; Parkinson 1991: 64–65;
Erman 1927: 116–131; Lichtheim 1975: 169–184;
Simpson 1973: 31–49).
The Tale of the Two Brothers, a work composed in
Dynasty 19 (ca. 1210 BCE), has elements similar to the W
story of Joseph and the wife of Potiphar in Genesis 39.
But, as the noted Coptic scholar Plumley pointed out in
1958, the two stories have some common aspects, “but
it would require much greater similarity of detail
between the Egyptian and the Hebrew stories to justify
the oft-made suggestion that the Egyptian story is the
origin of the incident described in Genesis.”
2316 Writing in Egypt in its literary context
Writing in Egypt in its Literary Context. Fig. 18 Golden
Bowl of Djehuty. A solid gold bowl from the tomb of the
General Djehuty. He served the Pharaoh Tuthmosis III. His
tomb was found at Saqqara in 1824 but the contents were
distributed and the site of the tomb is now lost. Louvre
Museum. Photo by Jon Bodsworth. Used with his kind
permission.
The Capture of Joppa, found with the Story of the
Doomed Prince on the same papyrus (Harris, # 10060,
Br. Museum), is an historical romance dealing with
General Djehuty, who served Thutmosis III in his
campaigns in Syria–Palestine (ca. 1504–1450 BCE)
and depicts an event otherwise unknown in the official
Egyptian annals. Djehuty outwits the city’s defenders
by having him and several Egyptian soldiers smuggled
into Joppa in baskets (James 1979: 112; Erman 1927:
167–169; Pritchard 1955: 22–23; Simpson 1973:
81–84) (Fig. 18).
Space permits only the briefest listing of several
other categories of Egyptian writing that have survived
in great quantities on papyri and ostraca, as well as in
the form of inscriptions and paintings on the walls of
tombs and temples. The reader interested in further
study will find more scholarly information and
examples of texts in the linked sites. These additional
genres of Egyptian literature include both secular and
religious songs. In addition to the above-mentioned
Harper’s Songs, Egyptian love songs, dating to the
New Kingdom, have also survived. These have a
timeless quality and can be very romantic and erotic
(Foster 1974; Simpson 1973: 296–326).
Egyptian Poetry is found in several forms, including:
hymns to the gods and king; love poetry that is both
beautiful and passionate, and the above-mentioned
Songs of the Harper, filled with uncharacteristic
pessimism and practical questioning of time-honored
beliefs. Some of this poetry has been compared to the
Biblical Song of Songs. It has also been noted that
Writing in Egypt in its Literary Context. Fig. 19
Egyptian Harp. One of several Egyptian musical instruments
found in Egyptian tombs. New Kingdom period (British
Museum). Photo by Jon Bodswoth. Used with his kind
permission.
Writing in Egypt in its Literary Context. Fig. 20
Musicians from the tomb of Nebamon. A relief from the tomb
of Nebamon from Thebes. Musicians and dancers accompany
a feast. New Kingdom period. British Museum (EA37984).
Photo by Jon Bodsworth. Used with his kind permission.
Egyptian love lyrics “are part of the same ancient
tradition of erotic verse which we find in the cultures
of the Fertile Crescent” (Kaster 1968: 220–233;
Erman 1966: xxxi–xxxviii; Erman 1927: lviii–lxii;
Lichtheim 1975: 11–12; Simpson 1972: 169–326)
(Figs. 19 and 20).
Egyptian hymns are not only dedicated to the gods of
Egypt, but also to the kings. The most famous Egyptian
hymn is the Hymn to the Aton (Aten), which, allegedly,
inspired the wording in Psalm 104 in the Hebrew Bible.
It is the earliest known monotheistic text in history
(Pritchard 1955: 370–371; Simpson 1972: 289–295).
Egyptian letters have survived on papyrus dating as
early as the Sixth Dynasty (ca. 2250 BCE). Many of the
surviving letters were used as models in the education
 Writing in Egypt in its literary context 2317

of scribes who were trained to master the style
necessary for official, business, and interpersonal
communication.
Egyptian Religious Texts constitute by far the
greatest number of written Egyptian texts that have
survived. The most common examples of this genre
include the famous Pyramid Texts, the Coffin Texts,
The Book of the Dead, Myths and Creation stories,
Oracles and Prophecies, Hymns and Prayers, etc.
The most famous of the Egyptian Mathematical
Papyri and texts is the Rhind Mathematical Papyrus
(ca.1580 BCE). It is believed to have been composed
by a scribe named Ahmose during the reign of King
Ahmose Auserre in the Hyksos era. It is in the British
Museum.
Egyptian Medical papyri and texts are fairly
plentiful. The British Museum alone has eight
examples. Three of the more important ones include
The London Medical Papyrus (#10059) which dates to
the 18th Dynasty, The Chester Beatty Medical Papyrus
VI (#10686) and Chester Beatty X (a 19th Dynasty text
dealing with aphrodisiacs). Perhaps the most famous
medical text is the Edwin Smith Surgical Papyrus,
which is housed in the New York Academy of
Medicine. Breasted published a folio-sized edition on
this papyrus in 1932. It is a remarkable, scientific

Writing in Egypt in its Literary Context. Fig. 21 Amduat scene from the Tomb of Pharaoh Ay. West Valley of the Kings
(WV 23). The 12 baboons represent the 12 hours of the night. Photo by Brian Anderson. Used with his kind permission.

Writing in Egypt in its Literary Context. Fig. 22 Amduat scene from another wall of the tomb of Pharaoh Ay. WestValley of
the Kings (WV23). Photo by Brian Anderson. Used with his kind permission.
2318 Writing in Egypt in its literary context

approach to the practice of medicine that deals with 48
different kinds of injury cases and covers surgical
treatments for the head and thorax. Ancient Egyptians
even knew that a mother’s urine is the secret to the
determination of the sex of an unborn baby. One of the
remedies that it recommended for the treatment of
facial wrinkles was described as a lotion that had
“proved effective myriads of times.” Another very
important medical text, the Ebers Medical Papyrus
(in Leipzig) offers some insights into Egyptian surgical
techniques. There is a section on the heart and its
vessels, and one of its remedies calls for the use of
castor oil as a laxative.
The Egyptian concern with the stars and planets led
to serious study of astronomy. From the alignment of
the pyramids with the stars to the many myths
associated with the day and nighttime journeys of the
sun, they have left us both physical and textual

Writing in Egypt in its Literary Context. Fig. 23 The Goddess Nut and the Book of Gates. This painted scene (from the
ceiling of the tomb of Ramses VI (KV 9) in the Valley of the Kings, ca. 1141–1133 BCE) shows the upper half of the goddess
Nut as she prepares to swallow her “son,” the sun god Ra, at the end of his daily journey across the sky. The ceiling paintings
portray astronomical themes as well as the various stages of the night journey of the sun god through the underworld. Photo by
Brian Anderson. Used with his kind permission.

Writing in Egypt in its Literary Context. Fig. 24 The Goddess Nut and the Book of Gates (b). This scene is a continuation
of the previous figure and shows the lower half of Nut’s body where she gives birth to Ra at the beginning of each new day. The
Book of Gates illustrates the Egyptian interest in astronomy and the universe as well as a new departure in their obsession with
the afterlife and the underworld. Photo by Brian Anderson. Used with his kind permission.

evidence documenting their obsession with the uni-
verse in which they lived. The earliest evidence of
Egyptian astronomical observation takes the form of a
series of texts in the lids of coffins dating to the Ninth
Dynasty (ca. 2150 BCE) (James 1979: 124–126).
In the New Kingdom, many of the royal tombs in the
Valley of the Kings were decorated with scenes from
the Amduat (“that which is in the netherworld”)
(Figs. 21 and 22).
These are found in the form of the Book of the Day
and the Book of the Night in a burial chamber in the
tomb of Ramses VI (1144–1136 BCE). These books
deal with the day and night journeys of the sun god
Ra – in the night, as he travels through the body of
his mother, Nut, the goddess of the heavens, and in
the day, as he travels across her body, after she gives
birth to him once again in the endless cycle of birth,
death and regeneration (Silliotti 1997: 64–6); Allen
1988: 38–49; Hornung 1992: 95–113; Shaw and
Nicholson 1995: 29, 62, 79, 106, 273, 290, and 299)
(Figs. 23 and 24).
Examples of Egyptian writing have survived in fairly
great quantities but there are considerable lacunae in
our textual knowledge. These lacunae are the result of
damage to portions of extant texts as well as to the
destruction of tremendous quantities of texts recorded
on perishable materials which did not survive the
ravages of nature and time. We are indebted to the
Egyptians for their strong desire to record their history,
religious and philosophical beliefs, as well as the fruits
of their imagination and scientific, astronomical,
medical, mathematical, and biological inquiries. When
considering the great intellectual and cultural debt that
the modern world owes to this Egyptian legacy, we can
only regret that we have thus far been unable to attain
the even greater quantities of writings that remain
outside of our reach and understanding.
See also: ▶Cuneiform, ▶Nilometer, ▶Astronomy in
Egypt, ▶Pyramids in Egypt
Acknowledgments
Thanks to Brian Anderson, a former student at West
Virginia University. Photographs featured in Figs 10,
21–24 were used with his kind permission. Thanks also
to Jon Bodsworth at The Egypt Archive ▶http://www.
egyptarchive.co.uk/html/index.html. Photographs fea-
tured in Figs. 1–3, 7, 14, 15, 18–20 were used with his
kind permission.
References
Aldred, Cyril. The Egyptians. 3rd ed. London: Thames &
Hudson, 1998.
Writing in Egypt in its literary context 2319
Allen, J. P. Funerary Texts and Their Meaning. Mummies and
Magic. Ed. S. D. Auria, P. Lacovara, and C. H. Roehrig.
Boston: Museum of Fine Arts, 1988.
Arnett, William S. The Predynastic Origin of Egyptian
Hieroglyphs: Evidence for the Development of Rudimen-
tary Forms of Hieroglyphs in Upper Egypt in the Fourth
Millennium B.C. Washington, DC: University Press of
America, 1982.
---. Egypt’s Debt to Mesopotamia. Archaeology Odyssey
July/Aug. (2004): 8–9.
Breasted, James Henry. Ancient Records of Egypt. 5 vols.
Urbana: University of Illinois Press, 1906. 2001.
---. The Dawn of Conscience. New York: Scribner Book
Company, 1976.
Casson, Lionel. Ancient Egypt. New York: Time-Life Books,
1965.
Clayton, Peter A. Chronicle of the Pharaohs: The Reign-by-
Reign Record of the Rulers and Dynasties of Ancient
Egypt. London: Thames & Hudson, 1994.
Elkins, James. The Domain of Images. Ithaca: Cornell
University Press, 1999.
Erman, Adolf. The Literature of the Ancient Egyptians.
London: Methuen & Co. Ltd, 1927.
---. The Ancient Egyptians: A Sourcebook of Their Writings.
New York: Harper & Row, 1966.
Foster, John L. Love Songs of the New Kingdom. Austin:
University of Texas Press, 1974.
Gardiner, Alan H. Egyptian Grammar: Being an Introduction
to the Study of Hieroglyphs. 3rd ed. Oxford: Griffith
Institute, 1978.
Hornung, Erik. Idea into Image. Trans. E. Bredech. New
York: Timken Publishers, 1992.
James, T. G. H. An Introduction to Ancient Egypt. London:
British Museum Publications, 1979.
Kaster, Joseph. The Wisdom of Ancient Egypt: Writings from
the Time of the Pharaohs. New York: Barnes & Noble,
1968.
Lewis, Jon E. ed. The Mammoth Book of Eyewitness Ancient
Egypt. New York: Carroll & Graf Publishers, 2003.
Lichtheim, Miriam. Ancient Egyptian Literature: A Book of
Readings. 3 vols. Los Angeles: University of California
Press, 1975.
Mark, Samuel. From Egypt to Mesopotamia: A Study of
Predynastic Trade Routes. College Station, Texas: Texas
A & M University Press, 1997.
Parkinson, R. B. Voices from Ancient Egypt: An Anthology of
Middle Kingdom Writings. Norman: University of Okla-
homa Press, 1991.
---. The Tale of Sinuhe and Other Ancient Egyptian Poems
1940–1640 BCE. Oxford: Oxford University Press, 1997.
Pritchard, James B. ed. Ancient Near Eastern Texts Relating
to the Old Testament. Princeton: Princeton University
Press, 1955.
Shaw, Ian ed. The Oxford History of Ancient Egypt. New
York: Oxford University Press, 2000.
Shaw, Ian and Paul Nicholson. The Dictionary of Ancient
Egypt. New York: Harry N. Abrams, 1995. W
Simpson, William Kelly Simposon ed. The Literature of
Ancient Egypt: An Anthology of Stories, Instructions, and
Poetry. New Haven and London: Yale University Press,
1972.
Simpson, William Kelly, ed. The Literature of Ancient Egypt:
An Anthology of Stories, Instructions, and Poetry. New
Haven and London: Yale University Press, 1973.
2320 Writing in India
Stiebing, William H. Jr. Ancient Near Eastern History and
Culture. New York: Addison Wesley Longman, 2003.
St John, Michael. The Palermo Stone – An Arithmetical View.
London: The Museum Bookshop, Ltd, 2003.
Wilkinson, Toby A. H. Early Dynastic Egypt. London and
New York: Routledge, 2001.
Wilson, John A. Royal Annals of Ancient Egypt. New York:
Columbia University Press, 2000.
Writing in India
This article deals first with the evolution of writing in
India in chronological order, then with the uses of
writing.
Indus Civilization (c.2600–1900 BCE):
A Forgotten Logosyllabic Script
The Indus or Harappan Civilization flourished in and
around the Indus River in present-day Pakistan and
western India c. 2600–1900 BCE. The most extensive
urban culture of its time, it had its own script, which
perished without descendants when the Indus cities and
their urban life-style collapsed. Thousands of short
inscriptions have been preserved, mostly seal stamps
and their impressions (Fig. 1). Because the Indus script
has no obvious affinity with any other script (a debated
Writing in India. Fig. 1 A seal stamp (size 20 × 20 mm)
from Mohenjo-daro, engraved with the figure of a bison
feeding from a trough and above it, a row of Indus script
signs. The impression (in mirror image) was read from right
to left. Photo Erja Lahdenperä for the University of Helsinki,
courtesy Archaeological Survey of India.
issue) and because there are no multilingual texts, it is
exceedingly difficult to decipher.
Indus merchants sailed to the Gulf and Mesopota-
mia, where about 40 seal stamps with Indus script have
been found. Like the Mesopotamian seal cylinders,
they were used to seal containers of goods to prevent
unauthorized opening. The Mesopotamian seal inscrip-
tions contain names and titles of the seal owners, and so
do probably the Indus seal inscriptions. It is also likely
that the Indus script functioned like the Mesopotamian
cuneiform script.
Cuneiform writing started as the world’s oldest
script, Archaic Sumerian (c. 3300–2900 BCE). Initially
pictographic, its signs depicted human beings and
animals and their body parts, plants and other things.
Written on clay with wedge-shaped lines, however, the
original shapes became unrecognizable, but the cunei-
form script remained a logosyllabic writing system.
Each sign either denoted the depicted word (logos in
Greek)—an arrow-shaped sign Sumerian ti ‘arrow’—
or ‘syllabically’ its sound irrespective of its meaning —
the arrow-shaped sign could also stand for Sumerian ti
‘life’ or ti ‘rib’.
When the Indus script was created c. 2600 BCE, all
the existing scripts were of the logosyllabic type. The
Egyptian hieroglyphs and the proto-Elamite script
(used widely in the Iranian plateau) were both inspired
by the Archaic Sumerian script in the late fourth
millennium BCE. The Indus people probably got the
idea of writing from the proto-Elamites, but used their
own traditional symbols and invented new ones while
devising their own script.
In 2004 it was asserted that the Indus script is no
real writing system reproducing speech, but only a
collection of non-linguistic symbols. However, it
radically differs from such symbols widely used in
and around the Indus Valley for centuries right until the
Indus script came into being. These earlier symbols,
called ‘potter’s marks’ because they are mainly found
on ceramic vessels, usually occur one by one, whereas
the signs of the Indus script occur in long and regular
rows, with repeated sign sequences. The script emerged
when the weights and measures were standardized and
the society became more stratified and completed large
and complex building projects.
The Indus script has been “deciphered” more than a
hundred times. Most attempts assume the underlying
language to be Indo-Aryan, which however was hardly
spoken in South Asia before 2000 BCE. The horse is a
very important animal in the Rigveda, the oldest
surviving Indo-Aryan text composed by 1000 BCE, but
the horse is not found among the many animals
represented in the Indus art. On the other hand, the
Rigveda contains some Dravidian loanwords. Dravidi-
an languages such as Tamil are now mainly spoken in
South India, but one (Brahui) survives in Baluchistan

and the Indus Valley. Systematic and cross-checked
interpretations for more than twenty signs (out of
c. 400) based on the Dravidian hypothesis have been
presented and won some recognition, but generally
speaking the Indus script is still considered undeci-
phered.
Vedic Times (c. 1300–500 BCE):
Accurately Memorized Oral Literature
After the collapse of the Indus Civilization c.1900 BCE
its script was forgotten. The Indo-Aryan speakers, who
came to South Asia from the northwest during the
second millennium BCE, were illiterate, but created
sophisticated oral poetry. Over thousand hymns were
assembled into a collection called Rigveda c. 1000
BCE. Elaborate repetition techniques developed to
assist its accurate memorization have safeguarded the
Rigveda from the hazards of oral transmission.
Learning the Rigveda by heart takes five years of hard
work in early youth, and keeping it in mind requires
daily repetition. This way of memorizing Vedic texts
still goes on, and their human repositories often are
more reliable than manuscripts. The rules created for
the preservation of oral literature have been scrupu-
lously observed, and their further development led to
astonishing achievements in the study of phonetics and
grammar without the aid of writing in the first
millennium BCE.
As their standard beginning (“Thus have I heard …”)
indicates, the earliest Buddhist texts of the fourth and
third centuries BCE were likewise orally transmitted.
Old Persian Rule (c. 500–326 BCE): Writing
Re-introduced — Aramaic and Kharos.t.hı̄
Around 518 BCE, Darius the Great conquered the
northern and southern parts of the Indus Valley: they
were attached to the Persian Empire as the provinces of
Gandhāra and Hinduš for the next two centuries —
until Alexander the Great took them over and soon
thereafter died in 323 BCE. The Persian bureaucracy
used the Aramaic script all over the Empire, from
Egypt to India.
Aramaic is a variant of the Semitic consonantal
alphabet which had been created in ancient Syria much
earlier (c. 1600 BCE) on the basis of the uniconsonantal
signs of the Egyptian script (where signs could express
words that contained 1, 2 or 3 consonants). Semitic and
Hamitic (i.e., Egyptian) languages can be understood
even if the vowels are left unwritten, as they usually
are, but the consonantal alphabet is not so suitable for
writing other languages.
The Aramaic script was adjusted for writing Indo-
Aryan languages by adding markings that express
vowels to the consonantal signs; if such an additional
Writing in India 2321
sign was missing, the respective consonant was
understood to be followed by the most frequently
occurring short vowel a; consonant groups were ex-
pressed by combining consonant signs together. Signs
for consonants not represented in the Aramaic script
were also created. Why this new syllabic alphabet is
called Kharos.t. hī is a much debated question. Kharos.t.hī
greatly resembles the Aramaic script, also in being
written from right to left. It was used in the northwest of
the Indian subcontinent until c. 200 CE, and in Central
Asia until c. 600 CE. (Fig. 2).
From the Mauryas to the Guptas (c. 320
BCE—550 CE): The Brāhmı̄ script
After Alexander’s death, the Persian provinces
Gandhāra and Hinduš were quickly incorporated in his
emergent empire by Candragupta, the founder of the
Maurya dynasty (c. 325–183 BCE). Candragupta’s
grandson Aśoka (in power c. 268–232 BCE) is one of
the most famous rulers of India. Aśoka’s rock and pillar
edicts, distributed all over his large empire, are the
earliest directly preserved historical records of ancient
India. Written in local dialects they also provide precious
information about the prevailing linguistic situation.
Aśoka’s inscriptions were written in the Kharos.t.hī
script, in the Greek script (introduced by Alexander and
used by the succeeding Indo-Greeks in the Indo-Iranian
borderlands) and in the brāhmī script, which is the basis
of most writing systems used in South Asia in later
times. The Indian tradition considers brāhmī as an
invention of the creator god Brahmā, who is depicted as
holding a palm leaf book in the hand.
The brāhmī script is a syllabic alphabet functioning
like the Kharos.t.hī, but with its distinction of short and
long vowels and different initial vowels still more
perfectly adjusted to Indo-Aryan languages, and
written from left to right. It may have been commis-
sioned by Aśoka for his rock edicts around 260 BCE;
some improvements in the script appear in the pillar
edicts about two decades later (Fig. 3). In that case the
brāhmī script is based on the Kharos.t.hī and Greek
W
Writing in India. Fig. 2 Kharos.t.hī script in the
Shahbazgarhi rock edict of Emperor Aśoka (c. 260 BCE).
Read from right to left.
2322 Writing in India
Writing in India. Fig. 3 Brahmi script in the Rummindei
pillar edict of Emperor Aśoka (c. 240 BCE). Read from left to
right.
Writing in India. Fig. 4 Siddham script in the “autographed
signature of me, Śrī-Hars.a, the overlord of great kings”,
preserved in a Sanskrit inscription engraved on a copper
tablet from Banskhera, dated 628 CE.
scripts. This recent conception about the origin of the
brāhmī is however contradicted by the likewise recent
discovery of potsherds inscribed with brāhmī char-
acters in stratum J (c. 450–350 BCE) at Anurādhapura
in Sri Lanka. If the dating is correct, these finds support
the traditional derivation of the brāhmī script from the
Phoenician consonantal alphabet long before Aśoka.
Alexander’s admiral Nearchus speaks of Indians
sending each other letters written on cotton cloth, but
this statement limited to the northwest may refer to
Persian administration in Aramaic. According to other
Greek authorities quoted by Strabon, the Indians did
not use writing. The Greek ambassador Megasthenes
stationed in the Mauryan capital c. 300 BCE specifies
that the Indians had only unwritten laws.
Post-Classical Period (550 CE Onwards):
Diversified Scripts
Almost all the writing systems that are nowadays being
used to write South Asian languages go back to the
brāhmī script, which remained in use until the collapse
of the Gupta Empire c. 550 CE. The late variety of
brāhmī used in North India was c. 600 CE transformed
into two principal types of script.
The siddham or siddhamātr.kā script (Fig. 4) was in
general use in North India until c. 1000 CE, and the
Bengali (Fig. 5) and Tibetan scripts and the śāradā
Writing in India. Fig. 5 Bengali script in an autographed
poem by the Nobel laureate Rabindranath Tagore (1861–
1941).
Writing in India. Fig. 6 Devanāgarī script in a quotation
from the Sanskrit epic Mahābhārata (Otto Böhtlingk,
Indische Sprüche III, St. Petersburg 1895): “O King, a kindly
spoken word brings along many kinds of blessings, a harshly
spoken word causes misfortune.”
Writing in India. Fig. 7 Tamil script in an edition of the Old
Tamil classic Tirukkura (aphorism 93): “Virtue consists in
speaking sweet but truthful words with a smiling face and a
friendly look”.
script of Kashmir are derived from it. The siddham
script spread also to Central and East Asia, being still
used to write Sanskrit magical formulae in Japan.
The nāgarī (later called devanāgarī) script has been
used to write especially Sanskrit (Fig. 6), Hindi,
Marathi and more recently Nepalese. The gurumukhī
script of the Punjabi language and the Gujarati script
also go back to early variants of the nāgarī script. The
southern Vijayanagara empire (1336–1565) used the
nandināgarī script imported from Maharashtra.
Tamil was written in the brāhmī script from the
second century BCE. This South Indian variant of
brāhmī is the basis of the relatively simple Tamil script
used today (Fig. 7), as well as of the much more

complex grantha script used in Tamil Nadu for writing
Sanskrit, which in turn is the basis of the Malayāam
script used in Kerala. Kannada and Telugu, the two
further literary members of the Dravidian language
family in South India, are written in closely related
script derived from the Cālukya kingdom. The Sinhala
language of Sri Lanka is written in the Sinhala script
that has evolved from the Sri Lankan variant of brāhmī.
The Arabic script imported to South Asia from
Persia by the Muslims since the 8th century CE has
been used to write especially Persian — the court
language of the Moghuls — and (with important
modifications) Urdu — the Muslim variant of Hindi —
and Kashmiri, and various minor Indo-Aryan and
Iranian languages spoken in Pakistan. The Roman
alphabet was brought to South Asia by the European
colonialists who came to stay with the Portuguese in
1498, and which is used for writing not only European
languages but also Konkani spoken around Goa.
Uses of Writing
Inscriptions and Coins
Because they usually can be exactly located and also
dated, inscriptions engraved in stone are very important
sources of political and cultural history in India. There
is a continuous inscriptional tradition ever since Aśoka’s
edicts (from c. 258 BCE), but their number increases
substantially only c. 700 CE. From c. 300 CE, royal
grants of villages and the like were engraved on sealed
copper tablets. Altogether c. 100.000 inscriptions in stone
or metal are known today. Thousands of inscriptions in
Indian languages and scripts have been discovered also
outside South Asia, in Egypt, Afghanistan, Central Asia,
China and most parts of South East Asia.
Numismatic material plays a crucial role also in the
history of the dynasties of Iranian origin in the centuries
around the beginning of the Christian era. The Indo-
Greek kings are known almost exclusively from their
coins. Being bilingual, these coins moreover opened up
the ancient history of India. The text written in the Greek
script and language on the obverse side is rendered in
the Prakrit language written in the Kharos.t.hī script on
the reverse side. James Prinsep in the 1830s worked out
how to read Kharos.t.hī and applied the results to the
Kharos.t.hī inscriptions of Aśoka. These readings in turn
opened up the secrets of the forgotten brāhmī script.
Manuscripts
The book in ancient India was usually written on long
leaves of the wine palm (Borassus flabelliformis),
which had first been cut into regular rectangular form.
Wooden covers of the same size were usually put on
both sides of the bundle to protect the leaves from
breaking. The ready bunch of leaves and covers was
Writing in India 2323
then pierced in one or two or three places in the middle,
so that a rope could be inserted through the leaves to
bind them together into a bundle. The long binding
rope had some stopper at both ends which prevented it
from coming out when the book was opened and the
rope loosened: the reader places the bundle on the
ground, and then turns the leaves over from one pile to
another. When the book is closed, the leaves are
collected together, the rope is tightened, and its long
loose end is wound many times around the bundle.
“Binding” the leaves thus together has given the
ancient Indian book one its Sanskrit names, grantha.
While writing on the empty leaves the scribe would
leave some margin around the edges and the cord-
holes. This was a precaution to secure that no text
would be lost if the leaf was broken at its edges. The
leaves would also be numbered, by putting a running
number in the right or left margin of the obverse or
reverse side. The margins could also be used for
inserting in them corrections (for letters marked in the
text with symbols of cancellation or painted over),
supplementary texts (parts of text accidentally omitted
in copying but noted afterwards), glosses (explanations
of difficult words or expressions) or subtitles (indicat-
ing the contents of the text at that place) (Fig. 8).
The colophon at the end of the manuscript often
contains not only the name of the work and its author
but also the name of the scribe and possibly his
genealogy and domicile and the date of completing the
manuscript. Just as the author of a literary work, the
scribe also starts with an auspicious word and a prayer,
most frequently an obeisance to Gan.eśa, the elephant-
headed god of success, who is expected to see the task to
its conclusion and to remove any obstacles. Gan.eśa is
thought to have written down the great epic Mahābhār-
ata at the dictation of its compiler Sage Vyāsa: as the
W
Writing in India. Fig. 8 Verso side of leaf 12 of a paper
manuscript written in the devanāgarī script, dating from the
17th century. The manuscript contains the Āran.yaka Gāna of
Sāmaveda. Sāmavedic songs composed c. 1000 BCE are the
world’s oldest known music. The small numeral signs over
the lines of text are musical notation. Textual divisions are
highlighted with red colour. Photo Asko Parpola.
2324 Writing in India
divine scribe he broke one of his tusks and used it as a
stylus. This agrees with the South Indian practice of
scratching signs on the surface of the leaf by means of a
sharp iron stylus. The incised lines were afterwards
made more visible by means of the juice of plants
rubbed over the leaves. In North India, the scribe would
use ink (made of soot) and either a brush or a pen made
of reed (Fig. 8). The same methods were used if the
manuscript was illustrated with miniature pictures —
painted in several colours in North India, incised in
South India (from Orissa to Tamil Nadu and Kerala).
In northwest India, parchment may have been used
for writing during the Iranian rule, since Sanskrit
pustaka ‘manuscript’ or ‘book’ is an Iranian loanword
denoting ‘book’ and ‘hide, parchment, bark’ in the
donor language. Birch bark was used as writing
material in Kashmir. Paper started being used in North
India around 1000 years ago, but in South India much
later. Paper used to be cut in long rectangular leaves
which imitate the shape of the traditional writing
material, the palm leaves (Fig. 8). The same applies to
thin wooden plates that have likewise been used as
writing material, especially in Sri Lanka and in
communities of Indian origin in South-East Asia.
Normally manuscripts did not survive long on
account of the climate and insects, even if they were
treated with vegetable oils or kept in smoky places
avoided by insects. Palm leaves usually last maximally
some three or four hundred years. In exceptional
circumstances palm leaves have survived much longer,
however: the oldest manuscripts are from the first
centuries CE and come from the dry climate of Central
Asia. There they have been kept in pottery vessels —
exciting discoveries of Buddhist birch bark and palm
leaf manuscripts have been made in recent years in
Afghanistan and northern Pakistan. Reed baskets
(pit. aka) have provided another and more usual
implement for storing manuscripts: at one time the
entire Buddhist canon could be put into three baskets.
In ancient India, temples, monasteries, royal palaces
and even private persons had manuscript libraries,
which could contain even tens of thousands of
volumes. Today millions of Indian palm leaf or paper
manuscripts survive in public and private libraries in
and outside South Asia.
On account of their poor durability, texts to be
preserved in writing had to be copied over and over
again, which led to accumulating mistakes. Such
copying mistakes have distorted literary texts, but if a
number of manuscripts are available, their comparison
may enable reconstructing a genealogy of copying and
reconstituting the original text.
Palm leaves have been used in South India until
the early twentieth century not only for writing
literature, but also for book keeping and for legal
documents, with seal impressions attached. On the
other hand, literature — especially voluminous Vedic
texts — has been handed down also in a millennial oral
tradition until the present day.
Bibliography
Bühler, Georg. Indische Palaeographie von circa 350 a.
Chr. — circa 1300 p. Chr. Strassburg: Karl J. Trübner, 1896.
Coningham, R. A. E. et al. Passage to India? Anuradhapura
and the early use of the Brahmi Script. Cambridge
Archaeological Journal 6.1 (1996): 73–97.
Dani, A. H. Indian palaeography. Oxford: Clarendon Press,
1968.
Daniels, Peter T., and William Bright eds. The world’s
writing systems. New York: Oxford University Press,
1996.
Falk, Harry. Schrift im alten Indien: Ein Forschungsbericht
mit Anmerkungen. ScriptOralia, 56. Tübingen: Gunter
Narr Verlag, 1993.
Farmer, Steve, Richard Sproat and Michael Witzel. The
Collapse of the Indus-script Thesis: The Myth of a Literate
Harappan Civilization. Electronic Journal of Vedic Studies
11.2 (2004): 19–57.
Filliozat, Jean, and Louis Renou. L’Inde classique: Manuel
des études indiennes, II. Paris: Imprimerie Nationale, and
Hanoi: École Française d’Extrême-Orient, 1953: 665–712.
Hinüber, Oskar von. Der Beginn der Schrift und frühe
Schriftlichkeit in Indien. (Akademie der Wissenschaften
und der Literatur [zu Mainz], Abhandlungen der Geistes- und
Sozialwissenschaftlichen Klasse, Jahrgang 1989: 11).
Stuttgart: Franz Steiner Verlag, 1990.
Jettmar, Karl ed. Antiquities of northern Pakistan — Reports
and Studies, Rock inscriptions in the Indus Valley 1 Mainz:
Philipp von Zabern Verlag, 1989.
Jettmar, Karl and Volker Thewalt. Zwischen Gandhâra und
den Seidenstrassen: Felsbilder am Karakorum High-
way. Entdeckungen deutsch-pakistanischer Expeditionen
1979–1984. Mainz: Philipp von Zabern Verlag, 1985.
Lawler, Andrew. The Indus script — write or wrong? Science
306 (17 December 2004): 2026–9.
Mahadevan, Iravatham. Early Tamil Epigraphy: From the
earliest times to the sixth century A. D. (Harvard Oriental
Series, 62). Chennai: Cre-A, and Cambridge, Mass.: The
Department of Sanskrit and Indian Studies, Harvard
University, 2003.
Masica, Colin P. The Indo-Aryan Languages. Cambridge:
Cambridge University Press, 1991.
Parpola, Asko. Deciphering the Indus script. Cambridge:
Cambridge University Press, 1994.
---. Study of the Indus script. Transactions of the International
Conference of Eastern Studies 50 (2005): 28–66. Tokyo:
The Tôhô Gakkai, 2005.
Pinault, Georges-Jean. Écritures de l’Inde continentale.
Histoire de l’écriture. Ed. Anne-Marie Christin. Paris:
Flammarion, 2001 92–121.
Robinson, Andrew. The Story of Writing. London: Thames
and Hudson, 1995.
---. Lost Languages: The Enigma of the World’s Undeci-
phered Scripts. New York: McGraw-Hill, 2002.
Salomon, Richard. Indian Epigraphy: A Guide to the Study of
Inscriptions in Sanskrit, Prakrit, and the Other Indo-Aryan
Languages. New York: Oxford University Press, 1998.

---. Ancient Buddhist Scrolls From Gandhāra. The British
Library Kharos.t. hī Fragments. London: The British
Library, 1999.
---. Writing Systems of the Indo-Aryan languages. The Indo-
Aryan languages. Eds. George Cardona and Dhanesh Jain.
London and New York: Routledge, 2003. 67–103.
Shivaganesha Murthy, R. S. Introduction to manuscriptology.
Delhi: Sharada Publishing House, 1996.
Writing in Japan
C HRISTOPHER S EELEY
The modern Japanese writing system employs predom-
inantly a combination of Chinese characters (1945 are
taught in school education) and syllabic signs called
kana (close to 100 in total, in two sets, viz. the angular
katakana and the more rounded hiragana) (Henshall
2003). These are used together with an admixture of
other signs such as Arabic numerals and Roman letters.
The following sentence is an example of how the
different elements function together:
弟はすごいＳＬマニアで、１２月にまたスイス へ行くことになっています。Otōto
wa sugoi esu-eru mania de, jūnigatsu ni mata Suisu
e iku koto ni natte imasu. ‘My younger brother is a huge
fan of steam locomotives, and it has been arranged for
him to go to Switzerland again in December.’ Chinese
characters are used for the noun otōto ‘younger
brother’, the suffix -gatsu ‘month (of the year)’ and
the first part of the stem of the verb iku ‘to go’.
Hiragana are employed for elements which include
particles such as wa (topic indicator), e ‘to, towards’,
for the adjective sugoi ‘great; awful’ and the second
syllable of iku and also for the post-verbal construction
koto ni natte imasu ‘it has been arranged to ---’.
Katakana represent two European loanword items
(mania ‘fan, enthusiast’ and Suisu ‘Switzerland’), while
an acronym-type European loan (esu-eru) is repre-
sented by upper case Roman letters. Arabic numerals
are used here to represent jūni ‘twelve’ in jūnigatsu
‘December’, which can otherwise be written all in
characters, viz. 十二月 (id.).
Although still very complicated, the system in use
today represents a substantial simplification of that
which was employed in the early part of the twentieth
century. It reflects a heritage of cultural and linguistic
history in Japan that extends back to early in the first
millennium AD.
Around 400 AD the Japanese, who at that time had no
writing of their own, actively adopted Chinese writing.
At this very early period, it was the Chinese language,
not Japanese, which was represented directly in written
form. Due to the difficulties associated with using what
Writing in Japan 2325
for the Japanese was a foreign language, together with a
very large and complicated character set, initially there
was heavy reliance on Chinese and Korean scribes
(Seeley 2000: 3–15).
As time went by, the Chinese script was adapted to
represent the Japanese language, sometimes directly,
but often in a convoluted manner. The Kojiki (Record of
Ancient Matters) of 712, notable as the earliest large-
scale compilation to have survived to the present,
constitutes a useful example of such adaptation. The
Kojiki preface is in Chinese, which was appropriate
as the most formal style in a document addressed to
the sovereign; the main part of the text is in a hybrid
form which reflects Japanese influenced by Chinese
(selected by the compiler with the stated aim of ease
of reading [!]). The Japanese songs interspersed in the
text are represented by Chinese characters employed
for their sound value so as to minimise the ambiguity
of reading in material of an oral nature (Seeley 2000:
40–46). At an early period, the Japanese had the means
to represent their own language directly in writing, but
the overwhelming prestige of Chinese culture meant
that more often than not they chose a Chinese-
orientated style.
Despite the high standing of written Chinese in
Japan (and Japanese attempts at Chinese composition),
direct representation of Japanese also had its place, and
this more straightforward form of writing enjoyed a
substantial rise in popularity from around 900 AD. This
was due to a development early in the ninth century in
the context of Buddhist works in Chinese. Abbreviated
Chinese characters employed for their sound value as
phonograms – the forerunners of modern katakana
(used today for such purposes as representing loan-
words of mainly European origin) and hiragana came
to be used quite extensively as a convenient means of
annotating the text, albeit sometimes only minimally
and sporadically, to help read or ‘decode’ it as
Japanese. Abbreviation was achieved through a process
of (1) writing rapidly (cursively) and thereby omitting
strokes (e.g., 女›め me,＞波›は ha) or (2) isolating one part of
a character phonogram and taking it to represent the
whole (e.g. 保＞呆 ho, 加›カ ka). After a time, katakana-type
signs tended to be favoured for utilitarian purposes such
as annotation, while hiragana-type signs came to be
used extensively for Japanese poetry and prose. Women
were seldom given the opportunity to acquire grounding
in Chinese characters, and the convention arose
whereby writing in phonograms of the cursivised variety W
tended to be associated with women. This is reflected in
the early term for hiragana-type signs, onna-de, and
‘women’s hand/script’. This is best understood as
meaning the script typically used by women rather than
exclusively by them, for men certainly sometimes
employed kana of this type (Seeley 2000: 59–89).
2326 Writing in Japan
While some Japanese texts were written entirely in
kana, much more commonly there was at least a small
admixture of Chinese characters. In the early period,
characters were sometimes employed as a convenient
means of representing Chinese loan morphemes which
involved sound segments difficult to represent in a still
emergent system of kana signs. Orthography of this
general type, which involved a combination of characters
and kana, gradually assumed a position of increasing
importance; the role played by Chinese characters as a
guide to disambiguating what would otherwise be a
Japanese text consisting of a long string of kana signs
spaced equidistantly should not be underestimated.
In spite of the above developments, broadly spea-
king, the cultural and linguistic importance of China
continued to be very strong. Written Chinese, or rather
an approximation of it, continued to be widely per-
ceived as highly prestigious until modern times, is
being used even at the beginning of the Meiji period
(1868–1912) in some Japanese newspapers. Increas-
ingly, though, the leaders of Japanese society were
being influenced by the West. As a result of extensive
debate on language and script, a movement evolved
around the end of the nineteenth century with the aim
of bringing written language more in line with spoken
language (the Genbun itchi [Unification of Speech and
Writing] movement) (Twine 1991), and there was also
support in some quarters for reducing the number of
different Chinese characters in general use. In the year
1900 the Education Ministry (Monbushō) published a
list of about 1,200 characters for use in primary school
texts, together with a standardised set of kana signs and
a partially simplified form of kana usage. However,
there was still strong opposition from conservative
elements to reform the writing system.
A radical turning point in the Japanese writing
system occurred in the 1940s, following Japan’s defeat
in the Pacific War. In that very different environment,
simplification of the written language and script was
espoused as part of the movement to promote
democracy. Changes effected included the adoption
and promulgation in 1946 of a list of 1850 Chinese
characters approved for active general use called the
Tōyō kanjihyō (List of Interim Use Chinese Characters),
followed by several supplementary lists aimed at
restricting the number of readings associated with
individual characters and increasing the number of
officially approved simplified character shapes (e.g., 廣＞
広, 藝＞芸). A more rigorous restriction took place in the
area of given names which parents chose for newborn
children: whereas previously it was permitted to use
any Chinese characters out of a potential corpus of tens
of thousands, from late 1947 parents found themselves
restricted by law in choosing from just the 1850 in the
Interim Use List, or alternatively to use kana. This
change was seen by many as excessive, and in 1951
this restriction was eased through a supplementary list
of additional characters for use in names. 1970 onwards
a significant relaxing of guidelines and regulations
regarding the range of Chinese characters were used.
The year 1981 saw the Interim Use List replaced by a
modestly expanded new list (1945 characters) called
the Jōyō kanjihyō (List of General Use Chinese
Characters), which was given the status of a guide
(meyasu) for usage, and the name character list has
been repeatedly expanded, consisting now of close to
300 characters (Seeley 2000: 152–87).
In the Interim Use List era and before, advocates of
script reform argued that kana or Romanisation should
be adopted for Japanese because of the difficulties
associated with mechanical production of text. A
manual typewriter for Japanese was devised early in
the twentieth century, but it required a specialist
operator, and only about ten pages a day at most could
be produced. The strength of the argument for script
simplification was, however, compromised in the late
1970s with the development of a word-processor which
featured Japanese language input in Romanisation or
kana and output in conventional (character and kana)
Japanese text. The mid 1980s then saw the appearance
of portable and affordable machines which gave
consumers the ability to handle Japanese text with
output which encompassed the Chinese characters
contained in a standard corpus of about 6,300 Chinese
characters determined for universal data exchange by
the Japan Industrial Standards Institute (Gottlieb 2000;
Seeley 1994).
The popularisation of word processors and compu-
ters from the 1980s for Japanese has led to a significant
rift between the number of characters taught in the
school system and the corpus by more than three times
the size accessible via computer-generated text. This
discrepancy needs to be addressed and somehow
accommodated. Despite associated problems such as
this, what seems certain is that the position of Chinese
characters as a central feature of the Japanese writing
system is assured for the foreseeable future.
Bibliography
Gottlieb, Nanette. Word-Processing in Japan: Kanji and the
Keyboard. Richmond, Surrey: Curzon Press, 2000.
Henshall, Kenneth, Christopher Seeley, and Henk de Groot.
A Guide to Reading and Writing Japanese (3rd ed.).
Boston, Vermont, Tokyo: Tuttle Publishing, 2003.
Seeley, Christopher. The Japanese Script and Computers:
The JIS Character Codes and their Periphery. Japan Forum
6.1 (1994): 89–101.
---. A History of Writing in Japan. Hawai’i: University of
Hawai’i Press, 2000.
Twine, Nanette. Language and the Modern State: The Reform
of Written Japanese. London & New York: Routledge,
1991.
 Writing: Maya writing 2327

most of the area (Fig. 1), and smaller ones – known as

Writing: Maya Writing “affixes” – often positioned before, atop, after, or beneath
the main signs, and sometimes (in the absence of a
main sign) forming clusters of their own (Fig. 2). In his
T OM J ONES 1962 Catalog of Maya Hieroglyphs, Eric Thompson
attempted a classification of all the known glyphs of his
Some dozen or so writing systems have been identified time, assigning to “affixes” the numbers 1–370, and to
as having been in use at one time or another in ancient “main signs” the numbers 501–856, with what he called
Mesoamerica, all sharing a number of general traits that “portraits” receiving the numbers 1000a–1087. The
include (1) pictographic signs, clearly derived from study and discussion of Maya writing have benefited
things seen, used in combination with wholly abstract enormously by the preferred use of these T-numbers (as
signs, the derivations of which remain obscure; (2) a they are called) over the arbitrary nicknames and labored
reading order that generally proceeds from top to descriptions of individual glyphs that once necessarily
bottom and from left to right; (3) a vigesimal number dominated their study.
system (base 20) with a dot representing the unit, Nevertheless, epigraphic study since the publication
generally in conjunction with individualized signs for of Thompson’s Catalog has shown that this main sign/
numbers of higher orders of magnitude, though in affix distinction is an artificial one. For while it is true
certain areas, in conjunction with bars for “5” and other that certain glyph designs seem only to have been
signs for “0” to express higher numbers; and (4) a employed as main signs and that certain others were
shared, compound, cyclical calendar of 260 × 365 used exclusively as affixes, a considerable number of
(18,980) days’ duration, expressed with readily identi- glyphs are known to function in both categories with no
fiable calendrical elements such as numbers, day apparent change in their semantic or phonetic value.
names, year signs, and periods of various magnitude. This is easily seen in the variety of ways in which the
The most sophisticated writing system – and well-known bakab title is recorded in the inscriptions
certainly the most intriguing – of Mesoamerica, and (Fig. 3). Size, then, is not necessarily correlated with
that which will serve as the basis for this discussion, is significance.
the “hieroglyphic” writing of the Maya region. While not entirely true, it is generally the case that
Examples of Maya hieroglyphic writing are known both the reading order of the glyph-blocks that
from many sources: from carved monumental inscrip- comprise a text (typically laid out in a grid-like
tions in stone and wood, painted stucco-coated, screen-
fold, paper books, or “codices”, painted and carved
ceramic pots, cups, bowls, and dishes, and from other
portable artifacts of stone, shell, ceramic, and bone.
The visual style, syntax, and content of the writing was
shaped by both the nature of the medium upon which it
appeared and the message to be conveyed.
Though open to doubt for many years, there is no
question today of the Maya writing system not being Writing: Maya Writing. Fig. 1 Main signs with
true writing. It follows linguistically logical rules of T-numbers.
syntax, phonetic construction, and semantic expression,
although with the same willful departure from those rules
as that of creative writers of our own culture – even, at
times, to the point of whimsical defiance. The Maya
hieroglyphs were capable of accommodating the full
range of both the sounds and the syntactical structures of
spoken Mayan. Indeed, the relatively narrow range of
subject matter of extant Maya writing is surely more of a
reflection of its differential survival on imperishable
artifacts than of any limitations in the writing system W
itself.
The typical Maya monument inscription is composed
of a number of tightly formed glyph-blocks arranged in
vertical columns and horizontal rows, with the glyph-
blocks formed of glyphic elements of two general sorts:
relatively large ones – known as “main signs” – generally
positioned at the center of glyph-blocks and occupying Writing: Maya Writing. Fig. 2 Affixes with T-numbers.
2328 Writing: Maya writing
arrangement of columns and rows) (Fig. 4) and the
internal elements of the glyph-blocks themselves
proceed from upper-left to lower-right (Fig. 5).
Texts composed of more than one column and one
row are usually read from the upper-left corner, two
glyphs from each row, descending two columns at a
time to the bottom. The next two columns are read in
the same manner, and so forth, to the end of the text.
The reading order of uneven or exceptionally disposed
texts necessarily depart from the ideal form just
described (Fig. 4). Departure from the general principle
of upper-left to lower-right is probably more common
in the reading order of glyphic elements within the
glyph-blocks than in the organization of the latter in
the texts. This is, no doubt, due in some part to what
were apparently culturally prescribed arrangements of
certain hackneyed terms such as the AHPO superfix on
emblem-glyph titles that identify lords of specific Maya
cities (Fig. 5g) or the -NAL superfix on certain Maya
place-names (Fig. 5h), in greater part to the flexibility
of the glyphic elements in both size and shape, and in
large part to the creativity of individual scribes.
Though an argument over whether the hieroglyphic
signs were of ideographic, morphemic, or phonetic
value raged for decades in the world of Maya
scholarship, today there is general agreement that there
are glyphs of both morphemic and phonetic kinds, and
that there is at least one glyph that always functions as a
semantic determinitive. About 180 distinct glyphic
Writing: Maya Writing. Fig. 3 Three spellings of the title
bakab. All are constructed ba-ka-ba. Note the variation in
size of the elements, substitutions and the use of head
variants.
Writing: Maya Writing. Fig. 4 Reading order for the texts.
elements have so far been demonstrated to carry the
phonetic values of some 77 consonant–vowel (cv)
syllables, leaving undetected but 23 of the 100 probable
sounds employed in the ancient script. At the same
time, in addition to the signs for the 20-day names,
about 90 glyphic elements have been demonstrated to
represent consonant–vowel–consonant (cvc), or even
more complex, morphemes. Thus, words can be
recorded by using a single morphemic sign (should
there exist one that signifies the desired word), by
assembling a phonetic construction from cv-signs that
sound it out, in which case the final vowel sound is
dropped, or by combining the two, perhaps using a
cv-sign as a (silent) phonetic complement to reinforce
the pronunciation of the intended word.
In addition to the aforementioned options available
to the Maya scribe, there were four further features of
the writing system:
1. Allographs are multiple signs that were used
interchangeably to express the same phoneme. The
sound u, for example, could be expressed by any
one of at least a dozen different allographs (Fig. 6a).
2. Puns are single signs used to express homophonous
words of multiple semantic value. As an example,
the Mayan word chan means “snake”, “sky”, and
“four”. The glyphs for these three meanings could be
used interchangeably to convey the sound chan, with
Writing: Maya Writing. Fig. 5 Reading order for the
elements within a glyph-block.

Writing: Maya Writing. Fig. 6 (a) Ten allographs for
the phoneme u. (b) Three signs for chan: “sky”, “snake”,
and “four”. (c) Polyvalency of T528, as the day-name
Kawak, the phoneme ku, and with phonetic complement
ni to spell TUN-(ni). (d) Head variants for the phonemes
pa and ba.
Writing: Maya Writing. Fig. 7 Four ways of spelling balam (jaguar).
Writing: Maya writing 2329
the intended semantic value, which was sometimes
multiple, to be inferred from the context (Fig. 6b).
3. Polyvalent signs are single signs of multiple phonetic
and/or semantic value which, again, are best under-
stood from their context. Thus when the sign T528
appears framed by the day-name cartouche (serving as
a semantic determinant), it is understood to represent
the name of the day Kawak; when it appears suffixed
with T116, the sign for ni (serving as a phonetic
complement), it is understood to represent the term
tun (“stone” or “year”); and when it appears with no
affixes, it usually represents the syllable ku (Fig. 6c).
4. Head variants are main signs modified from their
generally rounded forms into the profile heads of
persons or monsters that retain their diagnostic
markings (Fig. 6d).
The range of possible sizes of the various glyphic
elements, their substitutability and variety of appear-
ance (anthropomorphic, zoomorphic, and abstract),
their use as cv syllables, cvc morphemes, and phonetic
complements, all go toward accounting for the rich
variety of visual expression that characterizes the Maya
hieroglyphic writing system. Though many words
could be written with a single glyph, those same words
could also be expressed in other ways. As an example,
the word for balam (jaguar) was variously expressed by
using the sign that represents both the word and the
concept (Fig. 7a), by using the signs for the syllables
ba, la, and ma (Fig. 7b), by using the sign for BALAM
with ma as a phonetic complement (Fig. 7c), or by
using the BALAM sign with both ba and ma
complements (Fig. 7d).
Similarly, pakal (shield) was expressed by using a
sign that portrays a shield (Fig. 8a), by the signs for the
three syllables pa, ka, and la (Fig. 8b,c), or by the
PAKAL sign with a la phonetic complement (Fig. 8d).
It will be noted that, in spite of their difference in visual
effect, both the second and third of these examples
are phonetic constructions of the syllables pa-ka-la –
the second example substituting a head variant of the
more usual pa-glyph.
2330 Writing: Maya writing

Writing: Maya Writing. Fig. 8 Four ways of spelling pakal (shield).

Writing: Maya Writing. Fig. 9 Seven ways of spelling yichnal (together with).

Despite these general principles, there were words for
which there was no single sign and the organization of
which often defied the general upper-left to lower-right
reading order. The word yichnal (together with) is a
particularly interesting example of such a construction, in
part because it always includes the previously mentioned -
NAL ending superfixed to its main sign (Fig. 9). The latter
might be either the standard sign for phonetic chi or the
“torso-and-left-arm” sign that represents the sound YICH.
The several ways of recording yichnal, with the elements
read in the usual order, were yi-NAL-chi-la (Fig. 9a),
yi-NAL-chi (Fig. 9b), yi-NAL-YICH (Fig. 9c), NAL-yi-
chi (Fig. 9d), NAL-YICH-la (Fig. 9e), NAL-YICH
(Fig. 9f ), and NAL-yi-chi-la (Fig. 9g).
Among the tasks to which a Maya scribe might be
called upon to apply his skills was the execution of
commissions to inscribe in stone the dynastic history
and military achievements of his city’s royal family.
Commissions of this sort would typically involve the
recording of a variety of genealogical relationships
such as Ah Balam yal Na Chan (Sir Jaguar is the child
of Lady Sky), where the two names are separated by a
glyph-block that defines their relationship, with the
subordinate person named first (Fig. 10a). A military
encounter might be expressed with two phrases such as
Chukah Pakal. Pakal u bak Ah Balam (Pakal was
captured. Pakal was the captive of Sir Balam), the first
being a typical verb-initial Mayan sentence followed

Writing: Maya Writing. Fig. 10 The construction of Maya sentences.
by the subject, the second defining another relation-
ship with the identical syntax as the earlier example
(Fig. 10b). Accession to political office might be
recorded as Chumwan ti ahawel Ah Balam (Sir Jaguar
was seated in the Lordship), in another typical verb-
initial Mayan sentence where a positional verb is
followed by a prepositional phrase with the subject
Writing: Maya writing 2331
W
named last (Fig. 10c). Were our scribe to turn from his
work on public monuments to the designing of a
personal book for and about Lady Sky, he might record
such sentences as Sinah u chuch Na Chan (Lady Sky
strung her warpboard) or Och-i ti te’ Na Chan (Lady
Chan wove at the post), the first a verb-initial sentence
with a transitive verb followed by a possessed object
2332 Writing: Maya writing
and the subject (Fig. 10d), and the second a verb-initial
sentence with the verb followed by a prepositional
phrase and the subject (Fig. 10e). It should be clear
from the preceding examples of inscriptions on public
monuments that Maya hieroglyphic writing is indeed a
true writing system, fully capable of recording the
language it represents.
However, for all of the marvelous achievements of
recent Maya epigraphy, much work remains. Many
glyphs are undeciphered and problems still linger
amongst glyphs for which solutions have been offered.
An understanding of the kinds of difficulties encoun-
tered by epigraphers in deciphering individual glyphs
can be gained from a close look at a fairly common
lordly title, the five basic forms of which preclude
ready demonstration of their homophony, and suggest
instead the presence of both phonetic and semantic
substitutions. Known to scholars as the “penis” title, the
three most widely distributed forms consist of a main
sign that has the appearance of a profile representation
of male genitalia infixed with the sign for sha and
suffixed alternatively with -ti or -ta affixes (Fig. 11a–f ).
Desirable as it might be to have a homophonous reading
for the three forms – and their interchangeability in
the texts is beyond doubt – the conflicting sounds of
their initiating prefixes would appear to preclude any
such solution. One form is prefixed with the sign for
yo- (Fig. 11a,b), and second with to- (Fig. 11c,d), and
Writing: Maya Writing. Fig. 11 Semantic substitutions in the “penis title.”
a third with both yo- and to- (Fig. 11e,f). Because
these substituting sounds are prefixed, the words that
they signify must begin with the sounds yo, to, and
yoto, respectively. The likelihood that the male genitalia-
main sign is intended to convey the sound AT (the pan-
Mayan word for penis) is confirmed by the interchange-
able -ti and -ta suffixes that invariably accompany it,
apparently serving as phonetic complements. If, then, the
sha infix is read before the presumed sound of the main
sign, the sound sequence should be sha-AT (ti/ta), or the
not too promising shat. However, if the previously
mentioned yo- prefix is added to this, the result is yoshat,
or yosh-at (scarred penis); if to- is added, the result is
toshat, or tosh-at (bled penis); and if yo-to is added, the
result is yotoshat, or yotosh-at (very scarred penis).
Though any such title might strike those of us with a
background in Western European culture as bizarre,
nevertheless given the wealth of evidence for the presence
of penis-perforation ceremonies in which Maya lords
engaged and the occasional portrayal of a scarred penis on
figures in Maya art, the existence of a hieroglyphic form of
a title that testified to a lord’s personal sacrifice perhaps
ought not to surprise anyone.
Understanding of the remaining two of the five
forms which this title assumes is less certain. The first
of these is a head variant that is always superfixed with
to- and postfixed with -ti or -ta (Fig. 11g,h), suggesting
the possibility that the head represents the full term

TOSHAT, or TOSH-AT, with both affixes serving as
phonetic complements. The second (of which there are
but two examples) is a phonetic construction that is in
clear contradiction to the previously discussed yosh-at
renderings: one reads yo-ta-ta, the other yo-a-ta-ta
(Fig. 11i,j). Not only is the ta sound doubled in each,
but there is also nothing present that could be construed
to represent sha, the necessary middle consonant of
yosh-at. Nor is there an easy resolution of this
difficulty; either the previously proposed yosh-at,
tosh-at, and yotosh-at readings are in error and the
arguments in support of them faulty, or the Maya scribe
that recorded the two purely phonetic constructions
(they are both at Copan) failed to include all the sounds
of the word he intended to record, or – and this is a
solution that preserves both the earlier yosh-at, tosh-
at, and yotosh-at readings and the literary reputation of
the Copan scribe – the scribe deliberately omitted the
middle (sh) consonant, as is thought to have been done
with other two-syllable words. One can only hope that
further research and further field discoveries will
someday throw stronger light on this and similar
difficulties. Nor has there ever been a time when Maya
scholars were more optimistic about the possibilities of
their ultimate decipherment of this poetically con-
ceived, strikingly beautiful, and wonderfully complex
ancient writing system.
Postscript. Since the preceding was written and
submitted to the publisher, an alternative to the reading
of the penis title presented above has been proposed by
Linda Schele based upon a recently recognized example
that appears to include a sign for the sound ho. Schele
proposes a yoh-at reading for the yo-prefixed form and
phonetic construction, a toh-at reading for the to-
prefixed form (and its head-form), but is silent regarding
the yo-to-prefixed form. This should convey some idea
of the difficulties raised in the closing paragraphs above.
See also: ▶Cuneiform
References
Coe, Michael D. Breaking the Maya Code. New York:
Thames and Hudson, 1992.
Harris, John F. and Stephen K. Stearns. Understanding Maya
Inscriptions: A Hieroglyph Handbook. Philadelphia:
University of Pennsylvania Museum, 1992.
Houston, Stephen D. Reading the Past: Maya Glyphs.
Berkeley: University of California Press, 1989.
Jones, Tom and Carolyn. Notebook for Maya Hieroglyphic
Writing Workshops. Arcata, California: U Mut Maya, 1994.
Kelley, David H. Deciphering the Maya Script. Austin:
University of Texas Press, 1976.
Marcus, Joyce. Mesoamerican Writing Systems: Propaganda,
Myth, and History in Four Ancient Civilizations. Princeton,
New Jersey: Princeton University Press, 1992.
Thompson, J. Eric S. Maya Hieroglyphic Writing: An
Introduction. Norman: University of Oklahoma Press, 1960.
Writing in the middle East 2333
Writing in the Middle East
A LAN M ILLARD
Writing and the alphabet were two of the major
inventions of the human mind. They were conceived
and developed in the Middle East between about 3500
and 1000 BCE. Other writing systems appeared in China
about 1200 BCE and in Central America during the first
millennium BCE, probably independently, although the
Chinese may have been indirectly influenced from
Babylonia.
Cuneiform Writing
Writing began over 5,000 years ago in the developing
urban culture of southern Mesopotamia known as the
Uruk IV period. At first simple pictures were scratched
on stone or clay tablets held in the hand. Some showed
a complete object, such as a fish, others part of an
object, such as the head of a cow, while others cannot
be identified. More than 1,200 signs were used in the
oldest known texts, which deal with the income and
expenditure in kind of temples. The number of signs
soon shrank to about 600 and as well as serving as
word-signs (logograms), many also came to be used
phonetically (phonograms) on the rebus principle. Thus
in English the picture of a thin man beside a king read
as ideograms could mean “The thin man is king” or
“The king is a thin man,” but taking their values
phonetically could yield the two syllables of “think-
ing,” a concept harder to express pictorially. The rebus
principle enabled the script to express any words,
including foreign terms and names. The syllabic signs
are vowels (a, e, i, u), “open” (e.g. ba, gi, du, am, ir, uz)
and “closed” (e.g. bab, gim, dur) and some longer ones
(e.g. alad, gibil) (Fig. 1). Additionally, a small number of
signs were placed before or after word-signs to indicate
their class. These determinative signs mark categories
such as “man,” “woman,” “wooden object,” “fish,”
“pot,” and plurality. By 3000 BCE the language of the
texts is evidently Sumerian, a language unrelated to any
other now known, and it is likely the script was invented
by Sumerians. The Sumerian language appears to have
many homophonous words, which suggests it may have
been tonal, and so several signs can have the same
phonetic value when transcribed into roman letters,
which scholars distinguish by accents and subscript W
numbers (thus sá “to be equal,” sa4 “to name,” sa10 “to
buy”). Simplification for speed and impressing a reed
stylus on clay produced angular strokes and resulted
in the signs rapidly losing their pictorial form, so
becoming the wedge-shaped or cuneiform script. The
range of texts expanded over the centuries from the
2334 Writing in the middle East
Writing in the Middle East. Fig. 1 Cuneiform signs: their meanings and their development. Column 1 pictorial, ca. 3000
BCE, Column 2 cuneiform ca. 2500 BCE, Column 3 ca. 1800 BCE, Column 4 ca. 650 BCE. Line 1, Meaning “hand,” Sumerian
shu, Akkadian qåtum, syllabic values shu, qat. Line 2, Meaning “mouth,” Sumerian ka, Akkadian pûm, syllabic value ka.
Line 3, Meaning “wood,” Sumerian gish, Akkadian its, syllabic values gish, its, ets.
accounts and lists of words by category of the earliest
period to embrace almost every type of written
composition. By ca. 2500 BCE archives of clay tablets
include hymns and prayers, proverbs and wise advice,
riddles, legends and myths and brief historical records.
At that time the script was adapted for writing the East
Semitic Akkadian language (ancestral to Babylonian and
Assyrian). Word signs could now be read in Akkadian
and a much wider use of phonetic values of signs became
common as words were written syllabically (e.g. the
word-sign “bull’s head,” read in Sumerian as gud, in
Akkadian as alpum, could be written al-pu-um). While
some texts were written almost wholly with syllabic
signs, especially in the early second millennium BCE (the
Old Babylonian period), scribes usually wrote with a
mixture of syllabic and word signs, preceded or followed
by the determinative signs. This system continued until
the beginning of the Christian era, spread across the Near
East, especially in the second millennium BCE, and was
adopted for writing unrelated languages such as Elamite
in Iran and Indo-European “Hittite” languages in
Anatolia. The tens of thousands of clay tablets in modern
museums attest the activity of scribes in writing down
almost every aspect of life, but the script never lost its
complexity and so writing remained a specialized craft,
usually the monopoly of the trained scribe.
Egyptian Writing
Egyptian hieroglyphic writing began almost simulta-
neously with the ancestor of the cuneiform script in
Babylonia and was possibly influenced by it. (There are
clear signs of other influences from Mesopotamia in
Egyptian culture at the end of the fourth millennium
BCE, e.g. the distinctive cylindrical seal.) The
hieroglyphs are pictures and from the first were closely
associated with narrative art, as is evident in predynastic
carvings (e.g. the Narmer Palette) and thousands of later
sculptures and objects of all kinds. They were employed
both as word-signs (ideograms) and, through the rebus
Writing in the Middle East. Fig. 2 Egyptian hieroglyphs:
mr + m + r + pyramid = “pyramid”.
principle, as sound-signs (phonograms). In the early
stages over 1,000 signs occur, a number soon diminish-
ing to 600 or 700 which were in regular use until
Hellenistic and Roman times when the number multi-
plied enormously. In their day-to-day activities, clerks,
writing rapidly on papyrus, reduced the hieroglyphic
script to a cursive form, known as hieratic, examples
occurring from the earliest dynasties onwards. A highly
cursive and abbreviated script, demotic, developed in
the first millennium BCE and continued beside the
hieroglyphs well into the Christian era. Both hieratic and
demotic were written from right to left; the hieroglyphs
could run in any direction. The hieroglyphs are, to
modern eyes, purely consonantal; the ancient readers
supplied the appropriate vowels, facilitated by the
structure of the language, as in the West Semitic
“alphabet” (see below). There are three classes of
hieroglyphs giving phonetic, apart from their ideo-
graphic, values. The 26 with a single consonant are often
called the Hieroglyphic Alphabet, but it is important to
recognize that the Egyptians did not separate them from
the other signs. Each of these seems to denote the
dominant sound of the word represented (e.g. n from nt,
“water,” dz from w’dzyt, “cobra”). About eighty signs
are biliteral and seventy triliteral. These phonograms
often followed an ideogram, in effect “spelling out” its
value, thus accompanying the sign for “pyramid,”
mr, could be the two consonant sign mr and the
“alphabetic” signs m and r (Fig. 2). In addition, a range
of determinative signs marked categories of terms,
the best known being the oval cartouche enclosing

pharaohs’ names. Problems in representing foreign
names in the early second millennium BCE led to a
special combination of signs to produce syllabic
spellings, called “group writing” or “syllabic orthogra-
phy.” Egyptian scribes normally wrote on papyrus
which decays rapidly when buried except in very dry
situations, so most of their work is lost. Enough survives
to show that, although appearing needlessly complex,
the Egyptian writing system could express as wide a
range of ideas as the Babylonian. Its complexity meant
that it remained a specialist skill, practised by, it is
estimated, less than 1% of the population.
Other Ancient Scripts
The second millennium BCE – the Middle and Late
Bronze Ages in archaeological terms – saw a pro-
liferation of writing systems. In Anatolia, the Indo-
European speaking Hittites invented a hieroglyphic
script for writing the Luwian language, about 1600
BCE. It is known from stone monuments, seals and
a few documents engraved on lead strips, but it is
likely it was written more extensively on wooden
panels, now perished. It has a basic syllabary of 70
consonant + vowel signs, a few more complex syllabic
signs and 78 logograms for commonly occurring terms
(e.g. king, country) and so is similar in concept to the
Cretan scripts (see below). The values of the syllabic
signs appear to derived acrophonically from the initial
syllable of the word each picture depicted. After the
Hittite empire ended ca. 1200 BCE, smaller kingdoms
in Turkey and north Syria continued to write with the
hieroglyphs until the seventh century BCE.
Similar to the Hittite hieroglyphs are the scripts
found in Crete, the Cretan hieroglyphs and, perhaps
descended from them, Linear A and Linear B. Only the
last has been convincingly deciphered, revealing a
simple syllabary of 54 consonant + vowel signs and
five vowel signs, with numerous logograms which
sometimes have their names written syllabically beside
them, all recording an early form of Greek. Linear B
was scratched awkwardly on clay tablets and was
probably also written on wax tablets and other
perishable materials. Apparently an offshoot of Cretan
writing is the Cypro-Minoan system current in Cyprus
in the Late Bronze Age, with rare examples found in
the Levant. Like Linear B, surviving examples are on
clay, but impressed with a stylus rather than scratched.
They seem to record notes of ownership or delivery and
longer compositions, but the script remains unread.
Later in the first millennium the Cypriot syllabary of
50 consonant + vowel signs and five vowel signs
survives on stone, pottery and coins. It is clearly a
continuation of the Cypro-Minoan signs, probably
applied to a different language, and can be read, thanks
to texts with parallel versions in Greek.
Writing in the middle East 2335
A clay disk found at Phaistos in Crete is a unique
representative of a pictographic script impressed on the
surface by 45 separate dies in various arrangements, a
precursor of printing. It may have been brought to Crete
from Turkey, but without other examples, there is little
likelihood that it can be understood.
At Byblos, north of Beirut, a dozen documents were
unearthed inscribed with signs evidently forming a
large syllabary of 114 signs, fairly clearly based upon the
Egyptian hieratic script. These “pseudo-hieroglyphs” of
the mid second millennium BCE await convincing
decipherment. There are other, isolated, specimens of
writing systems which cannot yet be read. Short texts
scratched on clay tablets about 1200 BCE at Tell Deir
‘Alla in the Jordan Valley are reminiscent of Cretan
Linear B, but clearly not the same script.
The Alphabet
Out of this medley of scripts, one arose early in the
second millennium BCE which would ultimately have
the greatest impact since the invention of writing. A
scribe working in the Levant, perhaps in Byblos, found
none of the scripts available to him, Egyptian and
cuneiform, suitable for registering his West Semitic
language, so he created a system in which each sign
stood for a major consonantal sound (phoneme). It
seems that he drew a series of simple pictures, and used
each to stand for the initial sound of its name, thus
“door” = d (the acrophonic principle), yielding between
20 and 30 signs. In the West Semitic languages (e.g.
Canaanite, Aramaic, Hebrew, Phoenician) no word
began with a vowel and it proved possible to represent
the languages adequately without signs for vowels, so
at first they were not marked at all. That continues to be
basically true for Arabic and Hebrew today. The system
was not, therefore, a true alphabet, which should have a
sign for each major sound in the language, whether
consonants or vowels, but the term can be applied to it
and its descendants for convenience. The scribe was
accustomed to the Egyptian practice of writing with ink
on papyrus and followed the Egyptian direction,
writing from right to left, as Arabic and Hebrew are
still written. The use of perishable papyrus sheets and
wooden tablets coated with wax means that most
examples of this Canaanite Linear Alphabet are lost. A
few specimens scratched on metal, stone and pottery
have been found at Middle Bronze Age sites in Canaan
and two were discovered on a rock at Wadi el-Hol on W
the desert road across the bend of the Nile north of
Thebes. The date of these two is debatable, but they
certainly belong to a very early stage in the history of
the script. More extensive examples come from the area
of turquoise mines at Serabit el-Khadem in western
Sinai. They were apparently written in the middle of the
second millennium by workmen imported from Canaan
2336 Writing in the middle East
who revered the mother-goddess, Egyptian Hathor,
whom the Egyptians entitled “lady.” Although brief or
badly damaged, so far as these inscriptions can be read,
they refer to “lady” also (West Semitic ba‘alat). The
signs in these “Proto-Sinaitic” inscriptions appear in
more developed forms scratched or painted on various
objects from Late Bronze Age Canaan, and by about
1000 BCE they are almost standardized across the
Levant, their number reduced to twenty-two, as
examples from Byblos and other sites show (Fig. 3).
Two or three centuries earlier, scribes at Ugarit
(now Ras Shamra) on the coast of Syria, trained in
Babylonian tradition, realised the Canaanite Linear
Alphabet was better suited to their West Semitic
language than Babylonian cuneiform and imitated it
by creating a cuneiform “alphabet” of twenty-seven
signs, then adding three to mark sounds necessary for
recording the completely different Hurrian language.
Scribes at Ugarit used their script to write every type of
text, letters and administrative documents, rituals,
legends and myths, on clay tablets. Their survival in
the ruins of the city illustrates the sort of documents
which it can be surmised were written also in Canaan,
but they have perished. A very few pieces of Ugaritic
writing from other sites in the Levant, as far south as
Beth Shemesh west of Jerusalem, demonstrate its
spread, but the script disappeared with the collapse of
the Late Bronze Age city states.
It was the Canaanite alphabet that survived the chaos
at the end of the Late Bronze Age to be adopted widely.
One form spread into Arabia where it became the North
and South Arabic scripts, the former being known
mainly from graffiti which shepherds and travellers left
on rocks from the sixth century BCE to the Islamic
period, the latter being attested in thousands of
inscriptions on stone in the Yemen and Hadhramaut
Writing in the Middle East. Fig. 3 Development of the
early alphabet. Column 1, Sinai mines, ca. 1500 BCE,
Column 2, Canaan, ca. 1400 BCE, Column 3, Canaan,
ca. 1200 BCE.
from the ninth century BCE until the seventh century
AD. An offshoot of the South Arabic script survives in
the Ethiopic alphabet which has augmented the letters
to mark vowels. The other form of the Canaanite
alphabet was adopted by the new West Semitic
speaking kingdoms and eventually eclipsed all other
writing systems in the region. The Israelites occupying
Canaan found the script appropriate for Hebrew (except
that it did not provide two signs to distinguish the
phonemes s and ś). They used this “Phoenician” or
“Old Hebrew” alphabet until the Persian period,
continued it for some literature and revived it for
nationalistic purposes in the Roman period. Aramean
kingdoms wrote their language with the same alphabet,
despite the lack of signs to distinguish certain sounds
(e.g. t and t). Already in the ninth century BCE
Aramaean scribes gave some characters a double value,
noting both consonants and long vowels (e.g. one sign
for w and û, one for y and î ). Under the Persian Empire
(539–331 BCE), the Aramaic language and form of
script, which had been spreading widely under
Assyrian and Babylonian rule, became the common
vehicle of administration from the Indus to the Aegean,
from the Black Sea to Egypt and the Red Sea (Fig. 4).
The Jews used it for writing Hebrew and, slightly
developed, it has been the Hebrew alphabet ever since.
Aramaic speakers in southern Turkey, in the area
around Urfa and Haran, derived from it the forms of
cursive script known as Syriac (Estrangelo and Serto)
still read among Christian churches of the Middle East.
Another form arose in the caravan city of Palmyra (first
to third centuries AD.). Most significantly, Nabataean
Arab tribes in southern Syria and Jordan began to write
with the Aramaic alphabet in the last centuries BCE and
their cursive letters, known from a few papyri surviving
dehydrated in caves near the Dead Sea and some other
examples, are clearly the origin of the Arabic alphabet
(Fig. 5). Thus the modern Arabic and modern Hebrew
alphabets have the same ancestor. However, in the
cursive Nabataean of the early Christian centuries, the
Writing in the Middle East. Fig. 4 Letters used in Hebrew
handwriting ca. 600 BCE (1), Aramaic writing on stone,
ca. 400 BCE (2) and on papyrus or leather, ca. 400 BCE (3).

Writing in the Middle East. Fig. 5 Letters in forms used in
Hebrew handwriting, 1st century (1), Nabataean inscriptions,
1st century (2), Nabataean documents 2nd century (3), early
Arabic (4).
shapes of some letters had become almost indistin-
guishable (e.g. r and z) and the Arabic language
distinguished more consonantal sounds than the
Aramaic alphabet marked, so dots were added above
or below certain letters to distinguish some and to
create new ones. In Syriac, Arabic and Hebrew, beside
some signs for consonants marking long vowels as
well, small marks were added in the early Middle Ages
above, below and occasionally within the letters to
indicate all the vowels necessary for public reading of
sacred texts such as the Gospels, the Torah and the
Qur˒ān. As Islam spread it carried the Arabic script with
it and so it came to be applied to non-Semitic
languages. For Persian additional diacritical marks
were placed on four letters to mark Iranian phonemes
and for Ottoman Turkish modifications were needed
for six. In 1928, Kemal Atatürk’s modernization
programme, the Ottoman script was replaced with
roman letters, augmented by diacritical marks to 29
characters (Daniels and Bright 1996).
The Aramaic, Hebrew and early Arabic scripts are
known today from haphazard collections of inscrip-
tions on stone and other hard surfaces and from rare
groups of texts on papyrus, mostly found in Egypt. The
reduced number of signs, 22–28, made the tasks of
writing and reading simpler than for Babylonians,
Egyptians or Hittites, and so more people may have
learnt to read and write, but widespread literacy did not
result. For most people there was no need for either
Writing in the middle East 2337
Writing in the Middle East. Fig. 6 Semitic Alphabetic
script: Column 1, Phoenician, ca. 1000 BCE, Column 2,
Moabite, ca. 830 BCE, Column 3, Greek, ca. 700 BCE.
skill, and if they had the need, they could find a
professional scribe. Nevertheless, there is some
evidence for a proportionately wider distribution of
writing in ancient Israel than in neighbouring lands
(Millard 1995).
On the coast of Lebanon scribes in the Phoenician
cities (Tyre, Sidon, Beirut, Byblos) continued to write
with the Canaanite linear alphabet. Their products are
almost entirely lost, only a few texts on stone, metal
and potsherds surviving. It was from them that the
Greeks learnt it. Greek could not be written intelligibly
without marking vowels (otherwise there was no way
to write the Greek word for “no,” ou), so signs for
Phoenician, Semitic, phonemes alien to Greek were given
new values as vowels, partly in line with the use of some
consonants to mark long vowels in Aramaic (e.g. w for û,
y for î ). That yielded the first true alphabet (Fig. 6).
See also: ▶Cuneiform
References
Daniels, Peter T. and William Bright, The World’s Writing
Systems. New York: Oxford University Press, 1996.
Oates, Joan. Early Writing Systems. World Archaeology 17.3
(Feb. 1986): 305–460.
Millard, Alan. The Knowledge of Writing in Iron Age
Palestine. Tyndale Bulletin 46.2 (1995): 207–17.