Journal of Interdisciplinary History, XLVII:1 (Summer, 2016), 1–25.

Faisal H. Husain
Changes in the Euphrates River: Ecology and
Politics in a Rural Ottoman Periphery, 1687–1702
The Euphrates River mobilizes water and sediment into a dynamic
and complex flow powered by the energy of the sun and gravity of
the earth. It delicately changes through space and time and adjusts
to prevailing environmental conditions. Spatially, runoff and sedi-
ment production predominate in the headwater regions of the
Taurus Mountains and Anatolian highlands before the river turns
into a sediment sink and loses much of its water through dissipation
and evaporation in the Iraqi alluvium. Temporally, the river dis-
plays markedly different characteristics in the summer and spring,
with contrasting discharge rates and water-surface elevations.
Between 1687 and 1702, these timeless, ever-recurring re-
gional and seasonal changes were eclipsed by intense ecological
disturbances that transformed the Euphrates’ hydraulic architec-
ture. A dramatic rupture in the river’s flow occurred when a large
segment of it, approximately 100 miles in length, escaped its estab-
lished channel and gushed into a new one. The abrupt relocation
of the river, a process called avulsion by geologists and hydrolo-
gists, profoundly altered the ecology and politics of Iraq and im-
periled the stability of the Ottoman Empire in the east, threatening
traditional centers of power and permitting otherwise lesser tribes
to enjoy a temporary ascendance. Thousands of lives were lost
during the intervening years, and numerous settlements were
abandoned and left to ruin.
This article documents the metamorphosis of the Euphrates
River in the late seventeenth century. Beginning in 1687, a
prolonged meteorological anomaly and an ill-fated irrigation pro-
ject divided the middle Euphrates in Iraq into two capricious
branches (Figure 1). For a constellation of reasons, the Ottoman
central and provincial administrations were incapable of resolving

Faisal H. Husain is a doctoral candidate, Dept. of History, Georgetown University. He is the

author of “In the Bellies of the Marshes: Water and Power in the Countryside of Ottoman
Baghdad,” Environmental History, XIX (2014), 638–664.
The author thanks J. R. McNeill for his support.
© 2016 by the Massachusetts Institute of Technology and The Journal of Interdisciplinary
History, Inc., doi:10.1162/JINH_a_00939
2 | FAI S A L H . H U S A I N

Fig.1 The Tigris and Euphrates Rivers

NOTE Map details are approximate.

the environmental crisis until the river had completely abandoned

its original bed in 1700. The channel pattern that emerged there-
after withstood the assault of an engineering expedition tasked with
undoing the avulsion in 1701/2, thus facilitating the fall of the
Ottoman Rumahiyya fort southwest of Baghdad and the ascen-
dancy of the Khazaʿil tribe.
Much like the Nile and the Danube, the Tigris and Euphrates
Rivers had an inestimable significance for the Ottoman Empire.
These two arteries pulsed with the circulatory movement of nat-
ural resources, goods, and people. The flow of water within this
capillary system made irrigation agriculture in vast tracts of arid
C H A N G E S IN TH E E U P H R A T E S R I V E R | 3
land possible and nurtured the daily lives of tens of thousands
of productive, tax-paying peasants. Moreover, water made both
rivers a low-cost and navigable transportation network that knitted
together various Ottoman provinces in Anatolia, Syria, and Iraq
and connected the Ottoman heartland with the world of the
Indian Ocean.
Basra in the sixteenth century was the most important hub
downstream, serving as a clearinghouse for commodities arriving
from India, Iran, Iraq, Anatolia, Syria, and Arabia—textiles, spices,
wool, dates, rice, slaves, and concubines. From Basra, a sizeable
proportion of the commercial and human traffic made its way
northward via riverboats as far as Baghdad on the Tigris and Hilla
on the Euphrates, despite the difficulty of upstream navigation.
Ottoman agents exploited this bustling movement by maintaining
customhouses at several points along the rivers to collect tolls on
behalf of the provincial administration. From Baghdad, Hilla, and
other nearby shipyards, camel caravans further distributed the riv-
erine cargo toward Aleppo, Iskenderun, Tripoli, and the far cor-
ners of the Ottoman Empire.1
More convenient and significant was downstream transporta-
tion. From upstream provinces—such as Birecik, Raqqa, Diyarbakır,
and Mosul—officials managed the riverine transport of heavy artil-
leries, wheat, barley, and other provisions for the needs of the
Ottoman armies and forts on the eastern frontiers. Armaments and
rations aside, soldiers, rowers, carpenters, and other skilled laborers
descended downstream to buttress Ottoman military efforts. As a
result, the Tigris and Euphrates became a lifeline for Ottoman de-
fenses during conflicts with the Safavids, the Portuguese, and the
tribal groups that regularly challenged Ottoman hegemony in the
east from the sixteenth century onward. Flowing downstream as
well under the auspices of the Ottoman administration were cru-
cial natural resources that Iraq’s alluvial plain lacked, notably, the
roofing-grade timber and metals necessary for the construction and
the reconstruction of dams, forts, bridges, and the holy shrines in
Najaf and Karbala. This intricate web of interests, tightly knit with

1 Cengiz Orhonlu and Turgut Işıksal, “Osmanlı Devrinde Nehir Nakliyatı Hakkında
Araştırmalar: Dicle ve Fırat Nehirlerdine Nakliyat,” Tarih Dergisi, XIII (1962–1963), 79–
102; Suraiya Faroqhi, “Crisis and Change, 1590–1699,” in Halil İnalcık and Donald Quataert
(eds.), An Economic and Social History of the Ottoman Empire, 1600–1914 (New York, 1997), II,
483–487.
4 | FAI S A L H . H U S A I N

water, would unravel once the Euphrates unexpectedly altered its

pattern.2
The environmental shift under analysis appears in the most
comprehensive histories of Ottoman Iraq as a sudden, arbitrary
rupture that took place in 1700, unconnected with any preceding
crisis that had engulfed the region. That dating is based on chron-
icles that recorded only the climax of a long-term process. The
discussion herein, which combines tree-ring analysis (dendrochro-
nology) with untapped archival sources, argues that the Euphrates
started to re-position its course at an earlier date, in part because of an
exceptionally dry period in central Anatolia in 1687 and 1688. This
earlier dating has major historiographical implications. It uncovers
distinct causes that lay behind the river’s tumultuous reconfiguration
and connects the calamitous events that afflicted Iraq between 1687
and 1702, once thought to be random, into a coherent whole. As
Ball remarked, “Flowing water is not simply an unstructured chaos
but contains persistent forms that can be recognized, recorded,
analyzed—forms, moreover, that are of great beauty, of value to
the artist as well as the scientist,” not to mention historians of
Ottoman Iraq.3
The study of morphological change in the Tigris and Euphrates
has been largely the domain of archaeologists and earth scientists.
Given that a significant portion of the literature is theoretical, this
article aims to bring an empirical, holistic approach to the study of
avulsion, concerned not only with causal factors but also with
society’s response and adaptation to it. Archaeologists and scientists
have deftly outlined the skeleton of avulsion in Mesopotamia, to
which historians can add the sorely needed flesh and blood. The
eyewitness accounts utilized herein offer a unique opportunity to

2 Salih Özbaran, “XVI. Yüzyılda Basra Körfezi Sahillerinde Osmanlılar: Basra Beylerbeyliğinin
Kuruluşu,” Tarih Dergisi, XXV (1971), 51–72; Jon E. Mandaville, “The Ottoman Province of
al Hasa in the Sixteenth and Seventeenth Centuries,” Journal of the American Oriental Society,
XC (1970), 486–513; Abdurrahman Sağırlı, “Cezayir-i Irak-ı Arab veya Şattü’l-Arab’ın Fethi—
Ulyanoğlu Seferi—1565–1571,” Tarih Dergisi, XLI (2005), 43–94; Selçuk Dursun, “Forest and the
State: History of Forestry and Forest Administration in the Ottoman Empire,” unpub. Ph.D. diss.
(Sabancı University, 2007), 50–51; Orhonlu and Işıksal, “Dicle ve Fırat”; Faroqhi, “Crisis and
Change,” 483–487.
3 Philip Ball, Nature’s Patterns: A Tapestry in Three Parts. II. Flow (New York, 2009), 11–12.
For the history of Ottoman Iraq, see ʿAbbas al-ʿAzzawi, Tarikh al-ʿIraq bayna Ihtilalayn (Beirut,
2004), V, 178–182; S. Helmsley Longrigg, Four Centuries of Modern Iraq (Oxford, 1925), 121–
122; Clément Huart, Histoire de Bagdad dans les temps modernes (Paris, 1901), 139–142.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 5
move the analysis of channel evolution from the inanimate, all-
seeing GIS map to the living experience of anxious urban scribes,
weary construction workers, and valiant tribal sheikhs situated in
the middle of the changing alluvial environment.4

DROUGHT The dread of famine gripped Baghdad on Tuesday,

December 7, 1688. Starving families from Mosul and the Kurdish
regions flocked to the city begging for aid. Epidemics infected the
refugee population, many of which failed to find shelter in the in-
creasingly overcrowded provincial capital despite the best efforts of
its notables. Rumors and panic gave way to anarchy. A group of
Janissary soldiers (elite infantrymen) approached and murdered an
attendant at the shrine of Abu Hanifa (d. 767) for allegedly engag-
ing in a monopoly that aggravated currency inflation. The prices
of wheat, barley, meat, dates, and raisins skyrocketed in a year re-
membered by the people of Mosul as that of the Great Inflation.5
Famine, displacement, and inflated prices came in the wake of
a dwindling flow in the Tigris and Euphrates. Change in the rivers’
water levels occurred during a sharply defined and distinctive weather
anomaly, registered in the rings of Anatolian trees and the pages of
Ottoman state records and chronicles. As Silahdar Fındıklılı Mehmet
Ağa (d. c. 1726) wrote, protracted wars and drought in the “Muslim
lands” resulted in a great famine and inflation that became extremely
intense after 1687. Reports penned by Ottoman judges and officials
in Diyarbakır and Akşehir, and as far west in Anatolia as Eskişehir
and Seferihisar, conveyed to Istanbul the subjects’ tribulations. A

4 For notable scientific works about avulsion in Mesopotamia, see Carrie Hritz and
Tony J. Wilkinson, “Using Shuttle Radar Topography to Map Ancient Water Channels
in Mesopotamia,” Antiquity, LXXX (2006), 415–424; Vanessa Mary An Heyvaert and Cecile
Baeteman, “A Middle to Late Holocene Avulsion History of the Euphrates River: A Case Study
from Tell ed-Dēr, Iraq, Lower Mesopotamia,” Quaternary Science Reviews, XXVII (2008), 2401–
2410; Steven W. Cole and Hermann Gasche, “Second- and First-Millennium BC Rivers in
Northern Babylonia,” in Hermann Gasche and Michel Tanret (eds.), Changing Watercourses in
Babylonia: Towards a Reconstruction of the Ancient Environment in Lower Mesopotamia (Ghent,
1998), 1–64; Galina S. Morozova, “A Review of Holocene Avulsion of the Tigris and Euphrates
Rivers and Possible Effects on the Evolution of Civilizations in Lower Mesopotamia,” Geoarch-
aeology, XX (2005), 401–423; McGuire Gibson, “Population Shift and the Rise of Mesopotamian
Civilisation,” in Colin Renfrew (ed.), The Exploration of Culture Change: Models in Prehistory
(London, 1973), 447–463.
5 Yasin al-ʿUmrari, “Al-Durr al-Maknun fi al-Maʾathir al-Madiya min al-Qurun,”
MS Arabe 4949, 568, Bibliothèque nationale de France, Paris. See also Dina Rizk Khoury, State
and Provincial Society in the Ottoman Empire: Mosul, 1540–1834 (New York, 1997), 35.
6 | FAI S A L H . H U S A I N

dendroclimatic study corroborates Silahdar’s account and the offi-

cial reports, revealing that Sivas near the catchment area for the
twin rivers had experienced drought in 1687 and 1688. The anom-
alous precipitation was a signal of the climatic regime of the Late
Maunder Minimum (1675–1715), a period characterized by con-
siderably low sunspot numbers, colder temperatures, and highly
variable climatic conditions tending toward extremes, especially
in the continents of the Northern Hemisphere. As precipitation
declined in Anatolia, drought struck in Crimea, central Russia,
and India; sea ice accumulated on the coast of Iceland; and hail-
storms and gales smashed fields and vineyards in Switzerland. Firm
connections between these extreme weather events are elusive, but
they conformed to the prevailing, unstable climatic norm.6
Baghdadi chroniclers established the necessary correlation be-
tween the Sivas precipitation reconstruction and the Tigris–Euphrates
flow in 1687/8. In the words of Nazmizade (d. 1723), a provincial
official, “By the wisdom of the One who cannot be called to account

6 Silahdar Fındıklılı Mehmet Ağa, Silahdar Tarihi (Istanbul, 1928), II, 243. For tree-ring
study, see Rosanne D‘Arrigo and Heidi M. Cullen, “A 350-Year (AD 1628–1980) Tree Ring
Record of Turkish Precipitation: Linkages to Tigris-Euphrates Streamflow and the NAO,”
Dendrochronologia, XIX (2001), 169–177; for Diyarbakır, Başbakanlık Osmanlı Arşivi, Istanbul
(hereinafter BOA), Maliyeden Müdevver Defterleri (hereinafter MAD.D) 3871, 63; BOA, Bab-ı
Defteri Başmuhasebe Diyarbakır Hazinesi 14/54; for Akşehir, BOA, Ali Emiri, İkinci Ahmed,
2/115; for Eskişehir, BOA, İbnülemin, Sıhhiye, 1/76; for Seferihisar, BOA, İbnülemin, Dahiliye,
9/880; for Russia and Crimea, Andrei O. Selivanov, “Global Climate Changes and Humidity
Variations over East Europe and Asia by Historical Data,” in Michel Desbois and Françoise
Désalmand (eds.), Global Precipitations and Climate Change (New York, 1994), 88; Yevgeny P.
Borisenkov, “Climatic and Other Natural Extremes in the European Territory of Russia in
the Late Maunder Minimum (1675–1715),” in Burkhard Frenzel, Christian Pfister, and Birgit
Gläser (eds.), Climatic Trends and Anomalies in Europe 1675–1715 (New York, 1994), 88; for
India, Brian M. Fagan, Floods, Famines, and Emperors: El Niño and the Fate of Civilizations
(New York, 2009), 8–9; for Iceland, Astrid E. J. Ogilvie, “Documentary Records of Climate
from Iceland during the Late Maunder Minimum Period A.D. 1675 to 1715 with Reference
to the Isotopic Record from Greenland,” in Frenzel, Pfister, and Gläser (eds.), Climatic Trends,
17–18; for Switzerland, Pfister, “Switzerland: The Time of Icy Winters and Chilly Springs,”
ibid., 218–219; for the Maunder Minimum, John A. Eddy, “Solar History and Human Af-
fairs,” Human Ecology, XXII (1994), 23–35; Pfister, “Spatial Patterns of Climatic Change in
Europe A.D. 1675 to 1715,” in Frenzel, idem, and Gläser (eds.), Climatic Trends, 287–316; Sam
White, The Climate of Rebellion in the Early Modern Ottoman Empire (New York, 2011), 133,
215–222; John F. Richards, The Unending Frontier: An Environmental History of the Early Modern
World (Berkeley, 2003), 66–67; John L. Brooke, Climate Change and the Course of Global His-
tory: A Rough Journey (New York, 2014), 175–176; Geoffrey Parker, Global Crisis: War, Cli-
mate Change and Catastrophe in the Seventeenth Century (New Haven, 2013), 13–17; for more
information about the study of climate in history, see the pioneering special issue, “History
and Climate,” Journal of Interdisciplinary History, X (1980), 583–861.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 7
for anything He does, dread and fear of inflation overcame the hearts
of people when the dearth of quenching rainfall brought the Tigris
and Euphrates to a halt.” Ghurabzade (d. 1691) added that the rivers
flooded their banks. These terse yet invaluable observations are cru-
cial to comprehend the events that unfolded during the following
months. Nazmizade’s remark that the rivers came to a halt alludes
to a crippling effect on their status as agents of transport. Rainfall
shortage diminished stream power, the means by which rivers over-
come friction, transport sediment, and perform the basic geomorphic
work that creates their channels. A feebler stream intensified sedi-
ment accumulation within the channels, raised the riverbeds, and
prompted a spillover, as Ghurabzade noted.7
The Euphrates, which is more typical of the hydraulically vul-
nerable rivers of the arid zone than is the Tigris, was at a greater
risk under drought conditions. No significant tributaries join it be-
tween the Khabur River in Syria and Shatt al-Arab, a distance of
some 750 miles. The feather-like catchment basin of the Tigris, in
contrast, receives tributary contributions, draining the Zagros
Mountains from its eastern bank along most of its course, notably
from the Greater Zab, Lesser Zab, and Diyala. Furthermore, the
Euphrates’ gentler slope makes it more sluggish than the Tigris,
which has carved a deeper bed at a significantly lower altitude.
The Euphrates is also longer than the Tigris; in fact, it is the lon-
gest river in western Asia. The mountain basin near Erzerum from
which it debouches is more distant from lower Iraq than is the
source of the Tigris. Mount Karaca, a shield volcano in southeast
Anatolia, deflects the Euphrates westward and further extends its
length, forcing it to traverse several hundred miles through the
parching heat of the Syrian Desert. Therefore, whereas the Tigris,
like the majority of river systems in humid environments, receives
a push from tributaries that preserve the vigor of its flow through
most of its course, the Euphrates, longer in channel, gentler in gra-
dient, and without hydrological support in Iraq, enters the alluvial

7 Murtaza Nazmizade, Gülşen-i Hulefa (Istanbul, 1730), 106b; Ahmad Ghurabzade, “ʿUyun
Akhbar al-Aʿyan bi-man Mada fi Salif al-ʿAsr wa-l-Zaman,” MS Add 23309, 271b–272a,
British Library, London. For the relationship between stream power and sedimentation in
the Mesopotamian context, see Wilkinson and Hritz, “Physical Geography, Environmental
Change and the Role of Water,” in Wilkinson, Gibson, and Magnus Widell (eds.), Models of
Mesopotamian Landscapes: How Small-Scale Processes Contributed to the Growth of Early Civilizations
(New York, 2013), 20.
8 | FAI S A L H . H U S A I N

plain a much weaker river. Through seepage, evaporation, and ir-

rigation, it suffers a considerable diminution in volume, velocity, and
power downstream—vulnerable to the resistance of the channel bed
sediments to its movement—and loses definition in southern Iraq as
the ratio of sediment to flow surges.8
As a result, precipitation decline in 1687/8 disproportionately
affected the Euphrates, weakened its already feeble stream, and ag-
gravated its vulnerability to sedimentation and riverbed erosion. It
also brought the river closer to what geologists term the threshold
of critical power—the precarious point at which the power avail-
able for a stream to transport its sediment load equals the power
needed to accomplish it. In other words, prolonged drought
further destabilized an inherently unstable river downstream and
increased the risk of clogging, overbank spillage, and channel mi-
gration. In such perilous situations, rivers can adjust and find a way
to provide the velocity needed to transport the load supplied from
upstream. In this case, help came unexpectedly from a man named
Sheikh Dhiyab.9

SHEIKH DHIYAB Like foraging moose and dam-building beavers,

humans perturb specific components of their environments, often
inadvertently and with unintended consequences. Around the
turn of 1689, Sheikh Dhiyab attempted to direct irrigation waters
down the levee slopes of the Euphrates north of Rumahiyya
through a controlled levee break. Pressured by a hydraulically un-
balanced stream since 1687, his project ultimately burst out of con-
trol and ushered in a new channel configuration.10
Dhiyab was an ordinary sheikh engaged in an ordinary agrarian
pursuit. Yet, the unintended consequence of his action earned him
ill repute in Istanbul and eternal condemnation in the form of one
sentence in the official history of the Ottoman Empire compiled at
the time: “One of the Arab sheikhs issued and channeled water by
creating a fissure in the [Euphrates] River in order to irrigate his
gardens,” wrote Raşid Efendi (d. 1735), a court historian, before

8 Andrew S. Goudie, Arid and Semi-arid Geomorphology (New York, 2013), 204–245; M. G.
Ionides, The Régime of the Rivers, Euphrates and Tigris (London, 1937).
9 William B. Bull, “Threshold of Critical Power in Streams,” Geological Society of America
Bulletin, XC (1979), 453–464.
10 For the indirect consequences of moose herbivory and beaver engineering, see Robert J.
Naiman, “Animal Influences on Ecosystem Dynamics,” BioScience, XXVIII (1988), 750–762.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 9
recounting the ramifications. The nameless sheikh’s fateful action
sheds light on his predicament and that of other farmers in southern
Iraq.11
A number of environmental complications stood in the way
of Sheikh Dhiyab’s basic aim to water his gardens. Canal excava-
tion was essential due to the region’s meager annual rainfall, less
than the 200 mm minimum required for dry farming. The low gra-
dients of Iraq’s flat alluvial plain, which Evilya Çelebi (d. c. 1684)
likened to the Kipchak steppe in the Ukraine and Tavernier (d.
1689) to the terrain in Holland, posed an intricate challenge to
the canal excavator. Conducting water from the Euphrates re-
quired a channel that sloped at a steeper gradient than the landscape
to avoid sedimentation but not so steep as to invite accelerated ero-
sion. Sheikh Dhiyab was technically ill-equipped for the task and
had to walk a tightrope between the extremes of sedimentation and
erosion to solve this engineering dilemma.12
Fortunately, the Euphrates lent a hand. After passing Ramadi,
it flowed, as it does still, several meters above plain level in an
elevated bed of its own making. Instead of cutting deeply, it trans-
formed into a sediment-sorting machine, depositing coarser and
heavier materials adjacent to the banks and lighter ones into distant
basins, the distance traveled for each sediment class being inversely
related to grain size.
The resultant river levees have historically bestowed local in-
habitants with numerous advantages. The most compelling one, at
least for Sheikh Dhiyab, was a lateral gradient steeper than the
vexingly flat plain itself, along which he could direct irrigation
waters down the levee slope toward his gardens. The Euphrates
had endowed Sheikh Dhiyab with natural levees that he could
break to lay the basis for a flexible and cost-effective irrigation
module easily excavated, resistant to siltation, and manageable
by the kin group to which he belonged.13

11 Raşid Efendi, Tarih-i Raşid (Istanbul, 1865), II, 525.

12 Evliya Çelebi, Seyahatname (Istanbul, 1896), IV, 415; Jean-Baptiste Tavernier, Les six voy-
ages de Jean-Baptiste Tavernier, ecuyer Baron d’Aubonne, qu’il a fait en Turquie, en Perse, et aux
Indes, Pendant l’espace de quarante ans, & par toutes les routes que l’on peut tenir (Paris, 1676), I, 217.
13 Piet Buringh, Soils and Soil Conditions in Iraq (Baghdad, 1960), 144–148; Wilkinson,
Louise Rayne, and Jaafar Jotheri, “Hydraulic Landscapes in Mesopotamia: The Role of Hu-
man Niche Construction,” Water History, VII (2015), 397–418.
10 | FAIS AL H. HUSAI N

The scheme did not go as planned, however. All of their

advantages notwithstanding, levee breaks, whether initiated by
human or natural agency, represent points of weakness in the levee
bank that could expand to the detriment of the river. The weak
point provided by Sheikh Dhiyab became a node of avulsion, from
which the Euphrates branched, producing an offshoot that joined
the trunk channel (already under stress since 1687, due to diminished
discharge) in conveying all of the water and sediment delivered to
the river. Sheikh Dhiyab initiated the levee crevasse sometime after
October 1688. By September 1689, Ottoman officials were report-
ing that it was growing at an exponential rate, flooding villages on its
way and depriving those on the main course of sufficient water.
Thus, in this first, partial phase of avulsion, the Euphrates bifurcated
below Hilla into two low-energy, erratic branches—the parent
channel and the Dhiyab Canal—before reuniting above ʿArja. The
onset of this new river regime brought an ecological quandary that
strained the Ottoman provincial administration.14

PROVINCIAL CRISES Under stable conditions and careful manage-

ment, governments at the time could often turn a partial avulsion
into an opportunity to expand irrigation networks, transportation
routes, and settlement centers by maintaining sufficient and con-
trolled flow in the parent and new channels. The Baghdad admin-
istration during the late seventeenth century, nevertheless, found
itself entrapped in a vicious cycle difficult to break. Epidemiolog-
ical, political, and financial crises impaired its ability to mount an
effective response to the environmental shift, which amplified those
crises.
The dwindling water supply of the Tigris and Euphrates after
1687 created a panic that propelled more and more people into
Iraq’s urban centers, thereby maximizing the virulence of the pla-
gue when it hit the region. Raging in northern and western Persia
as early as 1684, the plague surfaced in Baghdad in March 1690 to
claim 100,000 lives. Several months later, it reached Basra, where it
killed 500 daily, and hit Baghdad again in early 1691. The damage
and loss of life in Baghdad were of apocalyptic proportions.

14 Nazmizade, Gülşen, 120b; BOA, Bab-ı Defteri Başmuhasebe Bağdat Hazinesi (hereinafter
D.BŞM.BGH) 1/ 27, 29. For nodes of avulsion, see Wilkinson, Archaeological Landscapes of the Near
East (Tucson, 2003), 82–85.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 11
“Friends and loved ones eschewed each other, the dues of fellow-
ship were ignored, and everyone was preoccupied with his own
life and careless about the conditions of others,” wrote Nazmizade.
In his eyes, the scene was reminiscent of the Qurʾan’s portrayal of
doomsday: “The Day man flees from his own brother, his mother,
his father, his wife, and his children” (80:35–37).15
“Plague wields a power that is disproportionate to the deaths
that it causes,” Wilson remarks. In late seventeenth-century Iraq, the
high mortality that it had caused in Baghdad and Basra compounded
the damage that the Euphrates had done, creating an opening for
mobile pastoral groups to prey on the afflicted urban populations.
The extraordinary career of Sheikh Maniʿ, son of Maghamis and
leader of the Muntafiq tribe, epitomized the rural–urban change in
power dynamics. In 1691, he marched audaciously toward Basra
with an armed force of 2,000 to 3,000 men. Standing in his way
was an Ottoman contingent of no more than 500 men headed by
Basra’s governor Ahmed Pasha. Terrified and outnumbered, 400
of the Ottoman troops deserted; the remaining 100 stood loyally
with the pasha and died at his side on the battlefield. Another
humiliating defeat in 1693 ended the mission of Baghdad’s gover-
nor to re-assert control over southern Iraq. The threat of Sheikh
Maniʿ and other emboldened tribal leaders hung over Baghdad
until the end of the seventeenth century.16
After the Euphrates veered off course, plague broke out, and
the countryside revolted, Baghdad plunged into a financial melt-
down, its treasury “totally broken” (kulliyet ile maksūr). The bud-
gets of 1689 and 1690 sounded the alarm bells. Losses totaled
68,310 and 95,542 guruş (a large Ottoman silver coin), respectively,
more than triple the budget surplus that the provincial treasury
recorded in 1670. As late as 1702, officials at the Baghdad treasury
were still dealing with a loss of 134,413 guruş that they attributed
to the damage made by the relocation of the Euphrates.17

15 Nazmizade, Gülşen, 113a, 114a; Ghurabzade, “Akhbar,” 272b.

16 Nazmizade, Gülşen, 114a–115a; Raşid Efendi, Tarih-i Raşid, II, 180–181; Defterdar Sarı
Mehmed Paşa (ed. Abdülkadir Özcan), Zübdet-i Vekayiat: Tahlil ve Metin, 1066–1116/1656–1704
(Ankara, 1995), 417, 454–455; Mary Elizabeth Wilson, “The Power of Plague,” Epidemiology,
VI (1995), 459.
17 BOA, Mühimme Defterleri (hereinafter MD) 104, no. 204, 50; BOA, D.BŞM.BGH 1/17, 27;
BOA, D.BŞM.BGH.D 16735; BOA, MAD.D 9891, 14–17.
12 | FAIS AL H. HUSAI N

The myriad forces of disease, politics, and finance conspired to

sustain the perilous, segmented path of the Euphrates and allow it to
evolve unmolested. Depopulation and financial distress handicapped
the provincial administration’s ability to mobilize sufficient labor
and resources to control the shifting riverbed, creating a political
vacuum around Baghdad that rural forces exploited. What was ini-
tially an engineering complication (levee failure) gradually became a
major military threat that called for an armed force large enough to
defeat an emboldened rural enemy.

IMPERIAL CRISES The imperial council complained to Baghdad’s

governor in late 1694, “From all sides, the cursed infidels are at
present daringly assaulting the Muslim lands.” The Ottoman Em-
pire, mired in an epic struggle against armies of the Holy League
(1683–1699), suffered crushing defeats and enormous fiscal strain.
The debilitating demands of an exacting and prolonged war se-
verely hampered the attempts of Ottoman policymakers to devise
a proper response to the Iraqi crisis. In late June 1692, the imperial
council ordered Baghdad’s governor to dam the Dhiyab Canal and
use state money (māl-i mīrī) to reconsolidate the Euphrates. Experts
in water control from the Rumahiyya region who had witnessed
the development of the Dhiyab Canal appeared before the gover-
nor in Baghdad to offer their assessment of the project’s feasibility
and potential costs. The provincial administration concluded that
as long as the Dhiyab Canal did not completely drain the main
river, it could focus on other urgent goals requiring less time, ef-
fort, and resources. Hence, provincial officials ignored Istanbul’s
order for the time being.18
Willingly or otherwise, the imperial council followed suit and
shelved any major hydraulic project, focusing instead on uprisings
in the countryside. Two daring tribal leaders, Sheikh Maniʿ in the
south and Bebe Süleyman in Shahrizor to the north, had blatantly
undermined what the authorities regarded as the very purpose of
the Ottoman state—safeguarding the welfare of the subjects. Rural
rebellion became even more urgent when it spiraled out of con-
trol around the Safavid border, threatening the peace with Persia

18 BOA, MD 105, no. 394, 94; BOA, MD 104, no. 201, 49; Nazmizade, Gülşen, 120b; Caroline
Finkel, Osman’s Dream: The Story of the Ottoman Empire, 1300–1923 (New York, 2007), 289–
321; Virginia Aksan, Ottoman Wars, 1700–1870: An Empire Besieged (New York, 2007), 18–36.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 13
established by the Zuhab Treaty in 1639. Ottoman decision
makers unequivocally affirmed their interest in maintaining friendly
relations with the Safavids and avoiding provocations, at least as
long as the main Ottoman army was embroiled on the western
front.19
The embattled Porte (the Ottoman seat of power) adopted
various measures to contain the fallout caused by rural turmoil—
among them, partial tax exemptions and resettlement plans for the
disadvantaged, diplomatic exchanges with Safavid authorities, and
constant admonishments to provincial officials to conduct them-
selves with clemency and justice toward the subjects. The imperial
council’s strategy regarding Bebe Süleyman and Sheikh Maniʿ
oscillated between the politics of appeasement and outright war-
fare. Although the two rebels had the blood of Ottoman officials
on their hands, the Porte sent Sheikh Maniʿ a conciliatory letter
(istimāletnāme) in 1693, granted him additional landholdings, and
appointed Bebe Süleyman as bey of the Bebe Kurds around
1695. In more acrimonious moments, the council called for Bebe
Süleyman’s head and supported two half-hearted military expedi-
tions against Sheikh Maniʿ in 1695 and 1698, both of which ended
in failure.20
Thus, the Porte, like Baghdad, was trapped in a profoundly
difficult situation during the late seventeenth century. Lacking
the capacity to mend the disorderly flow of the Euphrates, it de-
cided to give priority to immediate political action over long-term
hydraulic management. As a result, from its inception, the Euphrates’
partial avulsion threatened to push the system toward another dire
condition too costly to reverse. That grim prospect materialized by
1700, forcing the Porte to change its political calculations.

DALTABAN MUSTAFA PASHA In August 1701, an Ottoman equerry

arrived in Baghdad carrying an imperial edict to governor Daltaban
Mustafa Pasha, upon whom Sultan Mustafa II (r. 1695–1703) hung
high hopes. Four months earlier, while performing the Friday
prayer at the Selimiye Mosque in Edirne, Mustafa II received news

19 BOA, MD 104, no. 218, no. 225, no. 473–474, no. 686–688, no. 691–692, 53–54, 106,
161–162; BOA, MD 105, no. 395, 95; BOA, MD 111, no. 532–537, no. 1195–1196, 161–162, 348.
20 BOA, MD 104, no. 235, 57; BOA, MD 105, no. 393–395, 93–95; BOA, MD 106, no. 532, no. 575,
no. 1281, 161, 165, 324; BOA, MD 111, no. 224, no. 1196, 67–68, 348; Raşid Efendi, Tarih-i Raşid,
II, 225–226, 355–356; Nazmizade, Gülşen, 115a–117a; Defterdar, Zübdet, 675.
14 | FAIS AL H. HUSAI N

that Daltaban Mustafa Pasha, described as “a vigorous and tyran-

nical Serbian, illiterate but thrusting,” had just re-established
Ottoman control in lower Iraq and re-conquered Qurna and Basra.
The emissary read the imperial edict, praising the pasha’s loyal
service and relaying to him his new task—to dam the Dhiyab Canal
and restore the Euphrates to its former channel.21
The long-awaited order arrived at a time when the map of
the Ottoman world was being re-drawn. The signing of the
Karlowitz Treaty in 1699 ended hostilities between the Ottoman
Empire and the Holy League and diminished Ottoman influence
in eastern and central Europe. In the east, Mustafa II and Shah Sultan
Husayn of the Safavid Empire resolved border violations stemming
from the activities of Bebe Süleyman and Sheikh Maniʿ diplomati-
cally, despite deep mistrust on both sides. More important than the
geopolitical scene was the emergence of a new channel configura-
tion in the Euphrates. Ottoman authorities had not kept track of
intensified sedimentation occurring since 1687. By 1700, when
war in the west was over and the trunk channel had become com-
pletely “filled and shut” (memlū ve munsed) and left dry and sandy,
chroniclers in Istanbul and Baghdad were moved to issue grim
accounts of the situation. The sight of sand hills (kum depeleri) stand-
ing in the middle of the derelict conduit later mesmerized even the
Ottoman engineers. The Euphrates had entered its second, full avul-
sion phase, diverting its flow entirely to the Dhiyab Canal, which
suddenly expanded and, in the words of Raşid Efendi, “came to pos-
sess the strength and vigor of a large river.”22
The central authorities could not tolerate this development.
The new Euphrates played havoc with agrarian and commercial sec-
tors of the economy important to the Ottomans while rendering
most of its ecosystem services to rural outcasts and rebels. Farmers
had their fields either flooded and submerged or completely stripped
of their water supplies, depending on their location; merchants had

21 Silahdar (ed. İsmet Parmaksızoğlu), Nusretname (Istanbul, 1966), II, 49, 95–115; “1113’te
Basra Çevresinde Arap Eşkiyasına Karşı Yapılan Askeri,” MS 2062/4, 75a–91a, Esad Efendi,
Süleymaniye Library, Istanbul; Nazmizade, Gülşen, 121a; Longrigg, Four Centuries of Modern
Iraq, 95.
22 Raşid Efendi, Tarih-i Raşid, II, 509–518, 525; Defterdar, Zübdet, 706–713, 720–722;
Abdülkadir Özcan (ed.), Anonim Osmanlı Tarihi, 1099–1116/1688–1704 (Ankara, 2000), 154–
160; Al-ʿAzzawi, Tarikh al-ʿIraq, V, 163–165, 168–177; Silahdar, Nusretname, II, 95, 108; “Basra
Çevresinde,” 75b, 86b; Nazmizade, Gülşen, 120a–b.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 15
the movement of their vessels and caravans along the rivers inter-
rupted. Meanwhile, mobile pastoralists, loathed by the administra-
tion for their alleged “natural disposition to depravity and
wretchedness,” acquired an auspicious opportunity to seize power.
The Euphrates’ new course refueled the process of wetland forma-
tion, creating new marshes and engorging old ones. Soon, the water-
logged, anaerobic sites turned into places of refuge and escape for rural
groups to re-assert their autonomy and challenge state authority.23
The ending of the long war in the west and the settlement of
border disputes in the east finally permitted the rehabilitation of
the Euphrates to become a top priority. The Ottoman emissary
to Baghdad informed Daltaban Mustafa Pasha that the Porte was
prepared to offer all of the financial support and labor required to
achieve the Herculean task. Rural uprisings made the initiative all
the more urgent. Given a lull in major military engagements, con-
fidence was high that the Empire could finally crush the despised
pastoral enemies within its borders. “When the degree of igno-
rance displayed by the barefoot and naked wandering Arabs be-
comes evident,” a contemporary wrote, “nothing is easier than
having their ears pulled.”24

AN ENGINEERING FEAT The Porte brought into play what White

termed an “imperial ecology,” tapping into vast reserves of natural
resources and workers throughout the Empire to remake the
Euphrates. The provisioning of wood, which took precedence, fell
on the backs of the governors of Maraş and Raqqa. They dutifully
sent woodcutters to several mountain zones in Maraş and Malatya,
where they cut 42,200 logs of different kinds and sizes and trans-
ported them via animals to various points along the Euphrates’
shores. The rafts that picked up the logs from each spot met at
the Birecik shipyard in the upper Euphrates. Lighter wood,
12,000 palm and mulberry trees, came from Hilla in Iraq. The
rough fibers surrounding the bases of palm fronds produced thou-
sands of ropes and baskets. Aleppo’s tax collector was in charge of

23 BOA, Bab-ı Defteri Başmuhasebe Musul Hazinesi 3/118; BOA, D.BŞM.BGH 1/17; Silahdar,
Nusretname, II, 95–97; “Basra Çevresinde,” 75b–76a.
24 “Basra Çevresinde,” 56b. For analysis of Baghdadis’ views of their hinterland and deep
mistrust of rural tribal groups, see Khoury, “Violence and Spatial Politics between the Local
and Imperial: Baghdad, 1778–1810,” in Gyan Prakash and Kevin M. Kruse (eds.), The Spaces of
the Modern City: Imaginaries, Politics, and Everyday Life (Princeton, 2008), 186–187.
16 | FAIS AL H. HUSAI N

providing oakum, hawser, marline, and iron nails. When all of the
resources collected from Maraş, Malatya, and Aleppo assembled in
the Birecik shipyard—together with cauldrons for cooking pitch,
carpentry tools, forges, spades, pickaxes, sacks, bags, cannons, ex-
plosive bombs, and mallets—Ottoman authorities loaded the sup-
plies into open-topped boats and shipped them downstream to
Iraq.25
The manpower assembled under the command of Baghdad’s
governor was no less impressive. Steersmen, rowers, caulkers, and
blacksmiths from the areas around Birecik descended upon Iraq via
the Euphrates with cargo. Infantrymen escorted the boats as they
traveled. Governors and pashas of the eastern provinces and dis-
tricts of Kütahya, Diyarbakır, Shahrizor, Mosul, and Köy Sancak
joined their forces with Baghdad’s Janissaries. From Istanbul came
the deputy commander of the Janissaries (kul kethüdası) with a reg-
iment, gunners, and armorers. Artisans, including workers in brass
and carpenters, and a military band took part in the campaign.
After four months of travel, preparation, and ceremony, this
motley crew departed from its meeting point in Baghdad in early
December 1701 and crossed to the western Euphrates bank
through Hilla, where it unloaded provisions from the boats.26
After violently crushing a rebel force, the expedition was
ready to embark on the grand project, which it sought to accom-
plish in two major steps—(1) dredging and re-digging the sedi-
ment-chocked channel and (2) damming and closing the Dhiyab
Canal that had absorbed all the flow by 1700. As soldiers and
workers pitched their tents, Daltaban Mustafa Pasha surveyed
the defunct channel with the high-ranking officers, blacksmiths,
and carpenters. The plan was to dig a channel approximately
4 km long, 90 m wide, and 15 m deep. The followers of each
military leader, as well as locals from Hilla, Hasaka, Karbala, and
Najaf, had specific areas to dig. The project began on Decem-
ber 22, 1701.27

25 BOA, MAD.D 966, 192–215; BOA, MAD.D 3595, 6, 24, 72–74, 101–102; BOA, MD 111,
no. 2261, no. 2349, 637, 659; Silahdar, Nusretname, II, 99; “Basra Çevresinde,” 78a; White,
Climate of Rebellion, 15–51.
26 BOA, MAD.D 966, 192–215; Silahdar, Nusretname, II, 101–104; “Basra Çevresinde,” 80a–
82b; Nazmizade, Gülşen, 121a–b.
27 Silahdar, Nusretname, II, 105; “Basra Çevresinde,” 83b–84a; Nazmizade, Gülşen, 121b.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 17
The Porte had drafted traders and craftsmen to accompany the
expedition and supply it with provisions. Bakers, grocers, vegetable
sellers, butchers, drapers, and silk manufacturers established a mar-
ketplace in the campsite and attracted customers with “beautiful
melodies.” The authorities neglected neither hygiene nor morale;
they established several baths and coffee shops and recruited confec-
tioners and perfume makers. Ships carrying food from Baghdad,
Hasaka, and Hilla on a daily basis sustained the camp. “As famously
said,” a chronicler wrote, “even bird milk was found.” Known by
other historians as the military market (ordu pazarı), this organiza-
tional method for supporting large campaigns, according to Suraiya
Faroqhi, perfectly suited the circumstances that the Ottoman Em-
pire had faced since the late sixteenth century—an increasingly hos-
tile political environment, financial crises, and ecological pressures.28
About 4,000 workers were employed on the project, carrying
soil on their backs in sacks and bags. Transported earth formed hills
on the sides of the river. A military band played with gusto to lift
everyone’s spirits. In the afternoon of February 7, 1702, after forty-
eight days of digging, when water started to burst into the channel
under reconstruction, the barrier between the Euphrates and the
channel was removed to allow the river to run its new course. In
celebration of their accomplishment, officers and workers read
Qurʾanic chapters, made sacrificial offerings, and raised their hands
in prayer. Meanwhile, 200 naked men braved the gushing waters
with rafts to continue clearing the new channel. More than 20,000
people on both sides watched the spectacle with awe, crying,
“God is great! There is no god but God!”29
Following the reclamation of the old riverbed, blacksmiths,
carpenters, and weavers, with the assistance of twelve ships from
the imperial fleet and a few thousand oxen, focused their efforts
on damming the Dhiyab Canal to complete the diversion. Nature,
however, refused to cooperate. The Euphrates became unruly in
spring, the season when melting snow brings high floods in the
river, and swallowed the thousands of soil-filled mats and baskets
that workers had rolled into the water to establish the damming

28 Silahdar, Nusretname, II, 104–105; “Basra Çevresinde,” 83b; Suraiya Faroqhi, The
Ottoman Empire and the World around It (London, 2007), 108; Rhoads Murphey, Ottoman
Warfare, 1500–1700 (New Brunswick, N.J., 1999), 90–93.
29 Silahdar, Nusretname, II, 105–106; “Basra Çevresinde,” 84a–85a; Nazmizade, Gülşen,
122a; Faroqhi, “Crisis and Change,” 483.
18 | FAIS AL H. HUSAI N

Table 1 Tax Collections from the Rumahiyya Fort

YEAR c. 1538 1544 1577 1670
AMOUNT 74,386.5 akçe 257,412 akçe 242,520 akçe 6,300 guruş
SOURCE 1538: Başbakanlık Osmanlı Arşivi, Istanbul (hereinafter BOA), Tapu Tahrir Defterleri
(hereinafter TT.D) 1028, 121. 1544: Tapu ve Kadastro Genel Müdürlüğü, Kuyud-ı Kadime
Arşivi, Ankara (hereinafter TKG.KK), TT.D 228, 25b–26a. 1577: TKG.KK, TT.D 29, 309b–314b.
1670: BOA, Bab-ı Defteri Başmuhasebe Bağdat Hazinesi 1/17.

structure. In late March, the river inundated the Ottoman camp

and carried away remaining timbers. Due to a timber shortage, ris-
ing water levels, and the campsite’s putrefaction, workers decided
to abandon the project on March 30, 1702. After appointing
guards to oversee the partially built dam, they met with Daltaban
Mustafa Pasha to compose a letter to the Porte describing their
ordeal and justifying their decision, and dispatched the remaining
provisions back to Baghdad.30

RUMAHIYYA Within a twenty-five year period, feedback loops

between natural and anthropogenic forces gave violent birth to
a new hydraulic order in Iraq’s Middle Euphrates region that
produced long-term winners and losers. The obliteration of the
Rumahiyya fort was a major case in point. The fort emerged dur-
ing the Ilkhanid period in Iraq (1258–1336) as a riverine transit hub
and apparently a center for spear production. Under Ottoman rule
in the sixteenth century, it bustled with a dyeing workshop, a
slaughterhouse, a market, a press (for making juice or oil), a spin-
ning mill, a tannery, and even a gaming house (qumār hane). The
town’s governor sometimes ran Baghdad’s affairs when its gover-
nor left on a military campaign. Tax-collection figures indicate
that the town received a major boost after the Ottoman conquest
of Baghdad in 1534, flourishing throughout the sixteenth and
seventeenth centuries despite periodic Ottoman–Safavid confron-
tations in Iraq (Table 1).31
When compared with other market towns in the Baghdad
province c. 1577, Rumahiyya appears to have been nothing special

30 Silahdar, Nusretname, II, 107–112; “Basra Çevresinde,” 85a–88b; Nazmizade, Gülşen,

121b–122b.
31 Muhammad Rida al-Shabibi, “Al-Rumahiyya,” Lughat al-ʿArab, I (1913), 461–465; Tapu
ve Kadastro Genel Müdürlüğü, Kuyud-ı Kadime Arşivi, Ankara (hereinafter TKG.KK), Tapu
Tahrir Defterleri (hereinafter TT.D) 29, 313b–314a; Nazmizade, Gülşen, 62b–63a.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 19
Table 2 Major Market Towns in the Baghdad Province, 1577
TOWN TAX COLLECTED (AKÇE) TOWN TAXABLE POPULATION

Baghdad 4,522,125 Baghdad 3,309

Mandalijin 511,463 Hilla 1,498
Bilad Ruzin 452,132 Mandalijin 1,335
Shahraban 442,902 Rumahiyya 890
Hilla 361,121 Shahraban 405
Rumahiyya 251,520 Bilad Ruzin 364
SOURCE Tapu ve Kadastro Genel Müdürlüğü, Kuyud-ı Kadime Arşivi, Ankara, Tapu Tahrir
Defterleri 29.

in terms of wealth and population (Table 2). However, its sig-

nificance extended beyond its fortified walls. As the medieval
Jerusalemite geographer al-Maqdisi (d. c. 1000) sagely observed,
“You should know that an area does not become sublime by
the number of its towns, but rather by the splendor of its rural vil-
lages. Do you not see the splendor of Nishapur and Bukhara de-
spite the dearth of their towns?” In its rural villages, Rumahiyya
was the most profitable farming region in Baghdad, densely pop-
ulated, cultivated, and grazed (Table 3). The Rumahiyya fort
functioned as the Ottoman administrative center for the far more
prosperous rural settlements and tribal groups in its vicinity, orga-
nized in five districts—Khalid, Kabsha, Malik, Zubaid Gharbi, and
Zubaid Sharqi. In total, they contributed more taxes than any
other farming region in Baghdad, followed by the plains adjoin-
ing the lower Diyala River, one of the primary tributaries of the
Tigris. The chief granary of several bygone empires, the core farm-
ing region of the Diyala River in the sixteenth century consisted of
the districts of Halis, Mahrud, Tariq-i Khurasan, and Shahraban,
located roughly between Derne and Baghdad in Figure 1.32
The Rumahiyya sub-province, where ancient Borsippa and
its hinterland once stood, was an ecological patchwork, an ecotone
between arable lands and marshlands interspersed with semi-arid
steppe and ephemeral channels. The environmental conditions
favored extensive over intensive systems of land use, evident in
the poor yields in return for inputs of labor compared with the
situation in Diyala, a more homogenous landscape with better
productivity and a more reliable irrigation agriculture. Highly

32 Muhammad al-Maqdisi, Ahsan al-Taqasim fi Maʿrifat al-Aqalim (Leiden, 1906), 228.

20 | FAIS AL H. HUSAI N

Table 3 Rumahiyya and Diyala Regions Compared, 1577

FARMING REGION RUMAHIYYA DIYALA

Tax collected 4,879,668 4,426,948

Taxable human population 12,135 4,492
71% pastoralists 100% mixed farmers
Caprid population 76,023 Sheep only 24,176 Sheep and goats
Water buffalo population 11,837 -
NOTE Percentage figures have been rounded; all figures are approximate.
SOURCES Tapu ve Kadastro Genel Müdürlüğü, Kuyud-ı Kadime Arşivi, Ankara, Tapu Tahrir
Defterleri 29, 49b–110a.

vulnerable to a seasonal flow out of phase with the agricultural cycle

and to periodic uncontrolled runoff under poor drainage conditions,
most land tracts were more amenable to stockbreeding than crop
cultivation. The complex mosaic of weeds, stubble, artificial canals,
freshwater marshes, and seasonally filled depressions created by the
Euphrates provided extensive pasturage for a considerable sheep
population, three times larger than Diyala’s, and offered natural
wallows for water buffalo (Table 3).33
Animal rearing was an efficient way to convert Rumahiyya’s
inedible and nontaxable coarse grasses, bulrush, reed shoots, and
other lacustrine vegetation into appetizing and taxable types of
food and raw material. The water buffalo in particular is renowned
for its capacity to digest low-quality roughage material otherwise
not useful as livestock feed for protein synthesis and milk pro-
duction. In the middle of the seventeenth century, Tavernier
estimated that the Iraqi female buffalo, presumably when in full
lactation, could yield as much as twenty-two pints of milk daily.
Herders found uniquely hospitable conditions for buffalo hus-
bandry in Rumahiyya’s fragmented landscape (Table 3). In return,
Ottoman officials imposed a lucrative annual tax of twenty akçe
(a small Ottoman silver coin) on every buffalo in Rumahiyya, com-
pared to one akçe on sheep and two akçe on the sale of the buffalo
hides used in various types of heavy leather manufacture. The sale of
sheep garnered a tax of two akçe.34

33 Steven W. Cole, “Marsh Formation in the Borsippa Region and the Course of the Lower
Euphrates,” Journal of Near Eastern Studies, LIII (1994), 81–109.
34 P. Mahadevan, “Distribution, Ecology and Adaptation,” in Norman M. Tulloh and
J. H. G. Holmes (eds.), Buffalo Production (Amsterdam, 1992), 1–12; TKG.KK, TT.D 29, 304b;
Tavernier, Les six voyages, 217.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 21
Nevertheless, a region given to stockbreeding and semi-
sedentary folk was fraught with environmental and political risks.
Although a successful environmental adaptation and a contribution
to biodiversity, managed mammal herbivory in Rumahiyya came
at the expense of a loss of productivity and significantly lower yields
in cereal crops. The farmers of Diyala at the northern end of the
alluvium—where arable production, the highest in value per unit
weight, ruled supreme—were able to avoid such drawbacks. More
importantly, allowing or even encouraging nomadic pastoral groups
to occupy and pasture their flocks in Rumahiyya to maximize the
land’s tax revenues bestowed political advantages to herders in their
highly contingent relationship with the ruling power. With their
wealth on the hoof, they were “both inclined and able to resist or
evade centralized government,” particularly during times of political
instability, when the chain of command inscribed in Ottoman cada-
sters and law codes eroded. In The Art of Not Being Governed, Scott
quotes a poem that finely captures pastoralists’ penchant for indepen-
dence and autarky: “Do not cultivate the vineyard; you’ll be bound /
Do not cultivate grains; you’ll be ground / Pull the camel, herd the
sheep / A day will come, you’ll be crowned.”35
An auspicious opportunity for Rumahiyya’s pastoralists to
crown a chief of their own came in 1700 when the original
Euphrates branch fell into disuse and deprived the fort of its water
supply. After the Baghdad administration suffered rapid financial
losses in the Rumahiyya fort, tribal forces assumed control of
the town in 1694. Following the complete abandonment of the
Euphrates trunk channel, Ottoman ledgers in 1702/3 record for
Rumahiyya, and its districts of Malik, Khalid, and Kabsha, an out-
standing debt of 37,782 guruş, roughly double the budget surplus
that the Baghdad treasury recorded in 1670. From 1704 onward,
the fort dropped from the ledgers all together. In 1765, a passerby
noted that the town’s dwindling population then paid tribute to

35 Ernest Gellner, “Tribalism and the State in the Middle East,” in Philip S. Khoury and
Joseph Kostiner (eds.), Tribes and State Formation in the Middle East (Berkeley, 1990), 111;
James C. Scott, The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia
(New Haven, 2009), 187. See also William Irons, “Nomadism as a Political Adaptation: The Case
of the Yomut Turkmen,” American Ethnologist, 1 (1974), 635–658; Philip Burnham, “Spatial Mo-
bility and Political Centralization in Pastoral Societies,” in L’Equipe écologie et anthropologie des
societies pastorals (ed.), Pastoral Production and Society (New York, 1979), 349–360.
22 | FAIS AL H. HUSAI N

the Khazaʿil tribe. The town’s decline became irreversible before

the end of the eighteenth century, when the French consul in
Baghdad described it as an “ancient city” that had “fallen into
ruins.” By the nineteenth century, Rumahiyya had largely disap-
peared from the historical record.36

THE KHAZAʿIL “The flames of depravity and anarchy ignited, and a

shepherd in every area called for independence and proclaimed
sovereignty,” wrote a distraught Ottoman writer in the aftermath
of the avulsion of 1700. The time was ripe for Rumahiyya’s shep-
herds to deploy their herding and hunting skills, finely honed for
many decades with Ottoman acquiescence, and to exploit the po-
litical advantages of a mobile pastoral life. The most notorious
shepherd among them was Sheikh Salman—son of ʿAbbas and
head of the Khazaʿil tribe—who proclaimed his sovereign rule
in the Rumahiyya region most severely affected by the avulsion,
as well as in the nearby Hasaka, Nahr-i Shahi, and Najaf regions.
After transferring villagers to cultivate his newly conquered land,
Sheikh Salman besieged the Hilla fort, the last pillar of Ottoman
rule in Iraq’s Middle Euphrates region. The move prompted a swift
intervention by Ottoman commandos (serdengeçti) and Janissary
troops, successfully repelling the incursion but failing to dislodge
Sheikh Salman from the areas under his control, where he collected
taxes and behaved as a sovereign.37
The rise of the Khazaʿil in 1700 marked a dramatic change in
their fortunes. According to the most comprehensive accounts of
their history, the Khazaʿil, similar to other pastoral groups world-
wide, entered the historical record following a military encounter
with an agrarian, state-organized, and literate society. After Sultan
Murad IV’s re-conquest of Baghdad from the Safavids in 1638, his
new governor dispatched an expedition that brutally crushed the
tribesmen. Ottoman cadasters—especially those of 1577, largely
untapped by historians of Iraq so far—help to trace their humble
origins further back into the sixteenth century (Table 4). The

36 BOA, D.BŞM.BGH 1/17, 27, 29, 56, 59; BOA, D.BŞM.BGH.D 16735, 4; Nazmizade, Gülşen,
115b–116a; Carsten Niebuhr, Reisebeschreibung nach Arabien und andern umliegenden Ländern
(Copenhagen, 1778), II, 252-253; Jean Baptiste Louis Jacques Rousseau, Description du Pachalik
de Bagdad (Paris, 1809), 60.
37 Silahdar, Nusretname, II, 97–98; “Basra Çevresinde,” 76a–b; Nazmizade, Gülşen, 120b;
Hammud al-Saʿidi, Dirasat ʿan ʿAshaʾir al-ʿIraq: Al-Khazaʿil (Najaf, Iraq, 1974), 12–27.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 23
Table 4 The Khazaʿil Tribe, 1577
TAXABLE BUFFALO FLOCK SHEEP FLOCK TAX TOTAL
GROUP NAME INDIVIDUALS SIZE SIZE (AKÇE)
Sheikh Khalifa 22 33 35 887
Sheikh Rumh 19 20 30 882
Sheikh Hasan 7 8 10 316
Al Abu ʿAkkash 12 12 18 485
Sheikh Ghanim 14 15 25 589
Sheikh Maʿan 25 26 40 1,026
ʿAskar Raʾis 9 10 15 393
Total 108 113 173 4,578
SOURCE Tapu ve Kadastro Genel Müdürlüğü, Kuyud-ı Kadime Arşivi, Ankara, Tapu Tahrir
Defterleri (hereinafter TT.D) 29, 332b–333b. See also Başbakanlık Osmanlı Arşivi, Istanbul,
TT.D 1073, 18.

cadasters depict the Khazaʿil as a tribe of 108 herders organized

into seven groups. Even though they raised more sheep than buf-
falo, their sheep flocks (173 sheep in all) were significantly smaller
than the average size maintained by the dedicated sheep breeders
of Rumahiyya (about 521 sheep per group and village). They dis-
tinguished themselves primarily in buffalo rearing, maintaining a
herd (113 head) almost double the average size of other buffalo
breeders in the region (about 68 per group and village).38
Far from conferring prestige upon the Khazaʿil, buffalo rear-
ing more than likely made them the object of scorn. Iraqi pastoral
and agricultural tribes generally considered buffalo herding to be
beneath them, despite their recognition of its profitability. The
semi-aquatic buffalo were highly susceptible to infection from
the helminth parasites that infested the wallows of the Euphrates,
as well as to sarcoptic mange in the dry season. The anthropologist
Shakir Mustafa Salim observed that the Arab disdain for buffalo
breeding was traceable to the desert ethos of late antiquity, which
prompted the Arab tribes encroaching on Iraq since the early first
millennium to distinguish themselves from the local marsh people
by occupation. For whatever reasons, whether related to issues of
health, hygiene, culture, or a combination of those factors, the

38 TKG.KK, TT.D 29, 332b–333b; Al-Saʿidi, Al-Khazaʿil, 8–11; Max von Oppenheim, Die
Beduinen. III. Die Beduinenstämme in Nord- und Mittelarabien und Irak (Leipzig, 1952), 313.
24 | FAIS AL H. HUSAI N

Khazaʿil enjoyed little prestige among their neighbors, far behind

many of them in wealth and numbers.39
Under the astute leadership of Sheikh Salman, the Khazaʿil
seized the opportunity provided by the channel shift in the Euphrates
to transform their political standing in the region radically. A
convicted tax felon and fugitive who broke out of a Baghdad prison
around 1694, Sheikh Salman emerged in 1700 with a hefty military
force of 10,000 men armed with muskets and spears. By 1701, he had
brought most of Rumahiyya’s districts under his sway. The Khazaʿil,
the once-stigmatized buffalo herders, were now overlords. With
mobility, organizational flexibility, bold leadership, and a marshy
terrain on their side, they largely dominated settlements along the
new Euphrates channel until the early nineteenth century in defi-
ance of repeated warnings and military expeditions dispatched from
Baghdad.40

The “fatal synergy between natural and human disasters,” as Parker

puts it, in late seventeenth-century Iraq is comparable to the Ottoman
crisis in Anatolia nearly a century earlier, when the similar elements
of drought, inflation, plague, rural rebellion, and military entangle-
ment with the Habsburgs happened to coincide during the 1590s.
Nonetheless, the interaction between natural and human systems
in Iraq was unique for the prominence of fluvial processes, which
exerted enormous power in shaping and defining events from
1687 to 1702. Dwindling flow in the Tigris and Euphrates created
an atmosphere of crisis that crept over Baghdad and Basra, amplify-
ing the impact of plague in the region. Changes in the climatic,
hydraulic, and sedimentary variables governing flow in the Euphrates
triggered avulsion, abetting the military success of rebels in the coun-
tryside and contributing to the financial breakdown of the Baghdad
treasury. Interaction between the Euphrates and Ottoman epidemi-
ological, political, and financial crises was not a one-way process. The

39 Edward Ochsenschlager, Iraq’s Marsh Arabs in the Garden of Eden (Philadelphia, 2004), 27;
R. B. Griffiths, “Parasites and Parasitic Diseases,” in W. Ross Cockrill (ed.), The Husbandry and
Health of the Domestic Buffalo (Rome, 1974), 236–275; Shakir M. Salim, Marsh Dwellers of the
Euphrates Delta (London, 1962), 138–140.
40 ʿAbd al-Rahman al-Suwaidi, “Hadiqat al-Zawra fi Sirat al-Wuzaraʾ,” Add MS 18507,
20a–23b, British Library. For a comprehensive treatment of the Khazaʿilʼs history in the eigh-
teenth century and beyond, see Oppenheim, Die Beduinen, 313–333; Al-Saʿidi, Al-Khazaʿil,
12–109; Husain, “In the Bellies of the Marshes: Water and Power in the Countryside of
Ottoman Baghdad,” Environmental History, XIX (2014), 638–664.
C H A N G E S IN TH E E U P H R A T E S R I V E R | 25
depopulation, rural unrest, and bankruptcy that changes in the
Euphrates had fostered hampered effective hydraulic management
and allowed the partial avulsion of the late 1680s to become full
by 1700. When the dust had settled, literally (in the now-dry bed
of the river), a new fluvial landscape was in place. Desert and river
had changed places, Rumahiyya had turned to ruin, and the Khazaʿil
had become “kings of the Middle Euphrates.”41

41 “Reports of Administration for 1918 of Divisions and Districts of the Occupied Terri-
tories in Mesopotamia,” in Iraq Administration Reports 1914–1932 (Slough, England, 1992), 66;
Parker, Global Crisis.