Journal of Interdisciplinary History, XLVIII:3 (Inverno, 2018), 385–392.

Ehud R. Toldedano
Expectations and Realities in the Study of
Enslavement in Muslim-Majority Societies

Slavery, Agriculture, and Malaria in the Arabian Peninsula. By Benjamin
Reilly (Athens, Ohio University Press, 2015) 216 pag. $75.00 cloth $28.95 paper

For the past two decades, but more intensely during the last few
years, the body of knowledge dedicated to understanding how and
why modern-era Middle Eastern and North African societies en-
slaved humans has grown and deepened signiﬁcantly. The study of
the ways and means by which women and men from various geo-
graphical regions around the Mediterranean and the Indian
Ocean’s western shores were enslaved has been enriched and diver-
siﬁed to a degree unknown in the earlier cycles of interest regard-
ing enslavement. As the focus in the scholarship gradually shifts to
a more global view of servile labor and non-Atlantic forms of bond-
age, top-level research is continuously being shown on screens and
placed on library shelves across the world of learning. The fourth
volume of the Cambridge World History of Slavery, just published, Rif-
ﬂects the growing attention to, and coverage of, slavery in human
history from the Paciﬁc rim, via the Indian Ocean, to the Atlantic.1
The inclusion of other areas and the move toward a truly
global understanding of enslavement owes a great deal to the efforts
of scholars who research non-Atlantic societies, including Muslim-
majority ones. In recent years, the study of the history of enslave-
ment in those parts of the globe has achieved a volume, a dynamic,
and a maturity that make it one of the leading, cutting-edge sub-
ﬁelds in Middle Eastern and North African studies. Thanks to the
respectable list of monographs and articles published within the

Ehud R. Toledano is the Director of the Program in Ottoman and Turkish Studies, Tel Aviv
Università. He is the author of As If Silent and Absent: Bonds of Enslavement in the Islamic Middle
East (Nuovo paradiso, 2007); Slavery and Abolition in the Ottoman Middle East (Seattle, 1998).

© 2017 by the Massachusetts Institute of Technology and The Journal of Interdisciplinary
History, Inc., doi:10.1162/JINH_a_01163

1 David Eltis, Stanley L. Engerman, Seymour Drescher, and David Richardson (eds.),
Cambridge World History of Slavery (New York, 2017), IV.

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu

/
j
io

N
H
UN
R
T
io
C
e
–
P
D

F
/

4
8
3
3
8
5
1
7
0
2
1
0
5

/
j
io

N
H
_
UN
_
0
1
1
6
3
P
D

B
sì
G
tu
e
S
T

o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3

386

| E H UD R . T O L E DA NO

past decade or so in this subﬁeld, we also have the means to create a
better understanding of neglected phenomena in Atlantic slavery,
such as the nature of the relationships formed in household bond-
age, not just in gang servile labor on large estates and cash-crop
plantations.2 Non-Atlantic societies, especially in the Indian Ocean
world and Muslim-ruled areas, have had speciﬁc, often variegated
forms of slavery, developing diverse notions of abolition and gen-
erating diverse post-emancipation histories.

Reilly’s book has to be seen as part of that trend. It offers an
excellent account of agricultural enslavement in the region that
hosts the modern states of Saudi Arabia, Yemen, Oman, the United
Arab Emirates (UAE), Kuwait, and Bahrain, thereby ﬁlling a lacuna
in the expanding study of comparative bondage. Reilly seeks to
describe and explain the “system of slave and servile agricultural
labor, employing mainly sub-Saharan Africans, that prevailed in
the traditional Arabian Peninsula” (1). He adds that his study “will
argue that, as a result of the interaction between economic, cultural,
and environmental factors,” a hybrid character emerged that
combined well-known Middle Eastern elements of the practice—
mainly domestic, military-administrative, and other nonagricultural
iterations—with elements that were “strikingly similar to slave systems
in the Atlantic world of the seventeenth through the nineteenth
centuries” (2). His efforts center around the argument that, contrary
to the prevailing view, agricultural enslavement in the Arab world
deserves recognition for its pervasiveness and signiﬁcance.

2
Some of the better known works on this list are Toledano, As If Silent and Absent: Bonds of
Enslavement in the Islamic Middle East (Nuovo paradiso, 2007); Hakan Erdem, Slavery in the Ottoman
Empire and Its Demise 1800–1909 (New York, 1996); Eve Troutt Powell, Tell This in My
Memory: Stories of Enslavement from Egypt, Sudan, and the Ottoman Empire (Stanford, 2013);
Madeline Zilﬁ, Women and Slavery in the Late Ottoman Empire: The Design of Difference (New York,
2013); Chouki El-Hamel, Black Morocco: A History of Slavery, Race, and Islam (New York, 2014);
Muhammad Ennaji, Slavery, the State, and Islam (New York, 2013); Ismael M. Montana, IL
Abolition of Slavery in Ottoman Tunisia (Gainesville, 2013); Behnaz Mirzai, A History of Slavery
and Emancipation in Iran, 1800–1929 (Austin, 2017); Terrence Walz and Kenneth Cuno (eds.),
Race and Slavery in the Middle East: Histories of Trans-Saharan Africans in 19th-Century Egypt, Sudan,
and the Ottoman Mediterranean (New York, 2010); Alaine Hutson, “‘His Original Name Is . . . ’:
Remapping the Slave Experience in Saudi Arabia,” in Sabine Damir-Gielsdorf et al. (eds.),
Bonded Labour: Global and Comparative Perspectives (18th–21st Century) (Bielefeld, 2016), 133–162;
Michael LaRue, “The Frontiers of Enslavement: Bagirimi and the Trans-Saharan Slave Routes,"
in Paul E. Lovejoy (ed.), Slavery on the Frontiers of Islam (Princeton, 2004), 31–54; Amal N. Ghazal,
“Debating Slavery and Abolition in the Arab Middle East,” in Mirzai, Montana, and Lovejoy
(eds.), Islam, Slavery and Diaspora (Trenton, 2009), 139–153; Matthew S. Hopper, Slaves of One
Master: Globalization and Slavery in Arabia in the Age of Empire (Nuovo paradiso, 2015).

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu

/
j
io

N
H
UN
R
T
io
C
e
–
P
D

F
/

4
8
3
3
8
5
1
7
0
2
1
0
5

/
j
io

N
H
_
UN
_
0
1
1
6
3
P
D

B
sì
G
tu
e
S
T

o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3

E NS L A V E M E NT IN M US L I M – M A J O R I T Y S O C IE T I E S

| 387

Another major feature of Reilly’s contribution is the use of a
truly interdisciplinary approach to historical explanation. He pro-
vides a convincing account that links the employment of enslaved
sub-Saharan Africans to agricultural work in the mosquito-infested,
malaria-infecting moist lowlands and wadi bottoms, where palm-
shaded, date-growing oases provided the main cultivable areas in
the largely arid deserts of Arabia. Unlike the Bedouin populations
and the enslaved Ethiopians who largely performed domestic and
nonagricultural tasks, Sudanese and other Africans possessed a resis-
tance, either inherited genetically or acquired through immuni-
zation, to the various types of malaria prevalent in those oases.
Reilly argues that this feature of agricultural bondage in Arabia also
“explains the predominance of sub-Saharan African agricultural
labor” in the U.S. South and the Caribbean (115–119). During much
of the nineteenth and early twentieth centuries, the abundance of
enslaved Africans from those regions, purchased at reduced prices,
“combined to make sub-Saharan Africans the preferred type of slave
for Arabian agricultural.” To establish that point, he uses a variety of
evidence from DNA and epidemiology involving malaria-causing
mosquitoes.

Among other interesting and innovative methods, Reilly uses
data about well depths in cultivated areas to assess the risk of
malaria in agricultural settlements and the corollary presence of rel-
atively immune Africans in communities where shallow wells were
more common (119–121). The wealth of sources consulted in this
book is impressive, ranging from journals about genetics and geog-
raphy, Arabic chronicles, and travel accounts in both European and
Middle Eastern languages. Although some of the methods are clearly
circumstantial in nature, requiring some “backing and ﬁlling” and a
fair measure of speculation, the weight of the accumulated evidence
amply carries the burden of proof. Reilly has indeed demonstrated
that “malaria has played an important role in the history of the
Arabian Peninsula, most notably in mediating the relationship be-
tween traditional Arabian agriculture and African servile labor”
(121–122). He maintains that the presence of agricultural enslave-
ment in Arabia also owes much to two additional factors—“the tech-
nologically unsophisticated, labor-intensive methods utilized . . . ,
and the widespread availability of African slaves” (127).

Reilly’s book occasions a discussion about certain general
issues in this subﬁeld of enslavement studies—that is, the ﬁeld not

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu

/
j
io

N
H
UN
R
T
io
C
e
–
P
D

F
/

4
8
3
3
8
5
1
7
0
2
1
0
5

/
j
io

N
H
_
UN
_
0
1
1
6
3
P
D

B
sì
G
tu
e
S
T

o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3

388

| E H UD R . T O L E DA NO

concerned exclusively with Atlantic slavery. Given the constantly
growing volume of high-quality scholarship that has emerged in this
area within the last two decades, the time has come to abandon the
cliché that it is understudied. The relatively scant attention devoted
to research about agricultural bondage in the Middle East and North
Africa (MENA) and the complete lack of such coverage in Arabia are
reason enough for the focus that Reilly offers; he need not have
extended his complaint about the lack of scholarship to the entire
region or even to the “Islamic world,” as he frequently does in this
lavoro.

One question that arises is, What should be the appropriate
unit of study for an examination of slavery in this part of the
mondo? Should it be the Ottoman Empire, which had an articu-
lated economic system and a coordinated bureaucracy that gov-
erned most of the MENA territories; O, as Reilly seems to believe,
the Arabic-speaking world (other variants in the book being “the
Arab world” and “the Arabic world”), which is a cultural notion;
or the totality of Muslim-majority societies across Asia and Africa
(Reilly’s “Islamic world”), a religious-civilizational approach? A
this reviewer, framing enslavement within the Ottoman imperial
context seems to offer the most coherent sociocultural, historical
basis for analysis and interpretation. In that light, Reilly’s work is
primarily a local-history study of enslavement in Arabia, particu-
larly its agricultural component, rather than an attempt to address
a broader “Arab world,” a context that is fraught with ambiguity
regardless of thematics.

Another problem is to determine how much research about
enslavement in the MENA regions, especially within the Ottoman
and Qajar empires and the Arab successor states, is sufﬁcient to
make a general case. Are we to expect an output similar to, Dire,
the treatment of enslavement in Atlantic societies? Should we
not think of the MENA regions as more akin to African or Asian
societies, such as on the Indian subcontinent or Asia Paciﬁc rim,
than to Atlantic ones? What will convince scholars who venture
into this subﬁeld not to begin every study with a banal, and pat-
ently false, statement about the “lamentable paucity” of coverage?
The amount of research in enslavement studies, as in other ﬁelds
of social and cultural history, depends on the pervasiveness of
enslavement within a given society, and the existence of a con-
stituency that sees such history as its heritage and identity base.

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu

/
j
io

N
H
UN
R
T
io
C
e
–
P
D

F
/

4
8
3
3
8
5
1
7
0
2
1
0
5

/
j
io

N
H
_
UN
_
0
1
1
6
3
P
D

B
sì
G
tu
e
S
T

o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3

E NS L A V E M E NT IN M US L I M – M A J O R I T Y S O C IE T I E S

| 389

The sheer magnitude of enslavement in the Americas and the
concomitant demand for knowledge about it among descendent
communities—such as the African-American and the Brazilian
ones—are the factors that determined the level of scholarship. Two
leading historians of Brazilian enslavement state that “Brazilian
historians and economists are doing more studies on their insti-
tution of slavery than is now occurring in the United States, nonostante
the imbalance in the size of the historical profession in the two coun-
tries.”3 The sheer size of Brazil’s enslaved population, the duration of
the practice, and the quest of descendants for historical research
into their heritage have created this large-scale ﬂow of research.
Hébrard writes that not until the 1970s did “the history of slavery
[in Brazil become] a central focus of intellectual debate, including
heated disputes over politics and memory. Once this had begun,
nothing could stop the rush of research or the sheer intensity of
argument that still characterizes this extremely rich area of Brazilian
academia.”4 Unfortunately, much of this literature has remained in
Portuguese.

All available research, including Reilly’s current book (seem-
ingly contrary to the thrust of his argument), reveals that enslave-
ment in the MENA regions was, by and large, of a different nature
and volume from that in the Atlantic world. Consequently, due to
nation-building pressures in the successor states, descendent com-
munities of enslaved Africans and Circassians remain more inter-
ested in negotiating issues of identity politics than in exploring their
own history. The attempt to integrate as equal citizens in the Arab
countries of the MENA regions, as in Turkey and Iran, takes prece-
dence over any desire to emphasize cultural or political differences,
let alone to press demands for multicultural inclusion or to seek
reparations for past injustices. Given the nature of that research
ambiente, the investigations of MENA bondage during the past
decade and a half is hardly inadequate, despite the potential and
desire for more to come. Hence, to borrow a phrase from Reilly’s
testo, we “should put to rest once and for all the often-cited adage
that” enslavement in the Middle East is sorely understudied. Realistic

3 Herbert S. Klein and Francisco Vidal Luna, Slavery in Brazil (New York, 2010), ix.
4
Translating the Americas, 1 (2013), 49.

Jean M. Hébrard, “Slavery in Brazil: Brazilian Scholars in the Key Interpretive Debates,"

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu

/
j
io

N
H
UN
R
T
io
C
e
–
P
D

F
/

4
8
3
3
8
5
1
7
0
2
1
0
5

/
j
io

N
H
_
UN
_
0
1
1
6
3
P
D

B
sì
G
tu
e
S
T

o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3

390

| E H UD R . T O L E DA NO

expectations proclaim that it is not; the pace and quality of what is
in the pipeline are adequate testimony.

A related issue concerns the weight of agricultural enslave-
ment within the larger bag of bondage forms practiced in the MENA
regions and in other Muslim-majority societies. The “often-cited
adage” to which Reilly objects is that “slavery in the Arab world
was overwhelmingly consumptive rather than productive in char-
acter” (153–154). We have come a long way since the days when
domestic labor was not given its proper economic value; the dis-
tinction between “consumptive” and “productive” forms of servile
labor is now all but extinct. But even after eliminating any differ-
ence between these two forms, the research done thus far clearly
shows that agricultural bondage was not the predominant mode
of human exploitation in the MENA regions. Except for such known
cases of slave labor in Egypt’s cotton cultivation during the 1860s, IL
estates bordering the Sahara, Oman’s date plantations in Zanzibar,
and now, thanks to Reilly’s work, the Arabian oases, agriculture in
the modern Middle East was largely free and, to a signiﬁcant extent,
small-hold.

The large majority of enslaved persons in the Ottoman and
Qajar empires and their successor states were Africans and women
who toiled in urban households. Reilly’s book does not change
that picture despite the important light that it casts upon the
few pockets of agricultural enslavement on the margins of those
imperial domains. Numbers matter, but Reilly recognizes that the
data about the volume of the trafﬁc are unreliable, open to inter-
pretation and speculation; in most periods and for most peripheral
regions, they cannot really support far-reaching conclusions. None-
theless, he constantly prefers higher estimates, sometimes even the
totally imagined ﬁgures provided by travelers (75–79). The numer-
ical evidence in my own work largely agrees with the estimates of
such trusted scholars as Austen, Lovejoy, E, more recently, Wright
that the annual number of Africans coerced into the Ottoman
Empire during much of the nineteenth century fell between
16,000 A 18,000 women and men.5

5 Ralph A. Austin, “The 19th Century Islamic Slave Trade from East Africa (Swahili and
Red Sea Coasts): A Tentative Census,” in “The Economics of the Indian Ocean Slave Trade
in the Nineteenth Century,” Special Issue of Slavery and Abolition, IX (1988), 21–44; idem,
“The Mediterranean Islamic Slave Trade out of Africa: A Tentative Census,” ibid., XIII
(1992), 214–248. See also Thomas M. Rick’s thorough consideration of numbers in “Slaves

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu

/
j
io

N
H
UN
R
T
io
C
e
–
P
D

F
/

4
8
3
3
8
5
1
7
0
2
1
0
5

/
j
io

N
H
_
UN
_
0
1
1
6
3
P
D

B
sì
G
tu
e
S
T

o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3

E NS L A V E M E NT IN M US L I M – M A J O R I T Y S O C IE T I E S

| 391

Only a small fraction of those persons arrived in Arabia, and a
much smaller number of them reached the arable oases that Reilly
describes for agricultural servile work—a ﬁgure in the low thou-
sands for men. The history of these enslaved men is certainly wor-
thy of Reilly’s impressive investigation. Tuttavia, the scale of their
enslavement is not comparable to that of Atlantic, African, Indian,
or Chinese agricultural enslavement; nor are the modes of exploi-
tation and labor organization on cash-crop plantations. As already
noted, certain similarities to Atlantic practices might be detectable
on Egyptian cotton ﬁelds of the 1860s, Zanzibar’s date plantations,
and Reilly’s Arabian oases but not enough to shift the balance
dramatically.

Enslavement in non-Atlantic societies was, by and large, more
female-dominated, more domestic and less agricultural, more inte-
grative and less exclusionist, and more receptive to gradualist eman-
cipation than to one-step abolition than was its Atlantic counterpart.
The divergence in the pace of abolition—the transition from various
stages of unfreedom to various stages of freedom—has garnered very
little study thus far. In descending order, scholarship about MENA
enslavement has tended to concentrate upon African and Circassian
domestic servile labor in urban elite households, kul/harem and
gholam military-administrative bondage, enslaved menial and agri-
cultural labor, E, to a limited degree, galley slaves.

Finalmente, not surprising in a work that seeks to chart new terri-
tory within the current literature about MENA enslavement, Reilly’s
book contains several missteps. Notwithstanding the plethora of
scholarly works that Reilly consults, he has overlooked a number of
important contributions that also merit inclusion. Most conspicuous
in its absence—despite clear-cut relevance—is Timothy Mitchell’s
chapter “Can the Mosquito Speak?” in his Rule of Experts: Egypt,
Techno-Politics, Modernity (Berkeley, 2002), 19–53. Although Reilly
cites an article by Hopper with a title similar to Slaves of One Master,
he did not mention that book. Although this omission was probably

and Slave Traders in the Persian Gulf, 18th and 19th Centuries: An Assessment,” ibid., XIII
(1992), 60–70. For higher numbers and a criticism of Austen’s ﬁgures, see Lovejoy, “Commer-
cial Sectors in the Economy of the Nineteenth-Century Central Sudan: The Trans-Saharan
Trade and the Desert-Side Salt Trade,” African Economic History, XIII (1984), 87–95; idem (ed.),
Transformations in Slavery: A History of Slavery in Africa (New York, 2000), 135–159; for lower,
more recent estimates, John Wright, The Trans-Saharan Slave Trade (New York, 2007),
125–126; for Iran, Mirzai, Slavery and Emancipation in Iran, 1800–1929 (Austin, 2016).

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu

/
j
io

N
H
UN
R
T
io
C
e
–
P
D

F
/

4
8
3
3
8
5
1
7
0
2
1
0
5

/
j
io

N
H
_
UN
_
0
1
1
6
3
P
D

B
sì
G
tu
e
S
T

o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3

392

| E H UD R . T O L E DA NO

not his fault, since Reilly’s and Hopper’s books appeared in the
same year, it is signiﬁcant. Readers should be aware of it for future
reference.

Three other books about enslavement in the Arabic-speaking
world evaded Reilly’s attention. Montana’s Abolition of Slavery in
Ottoman Tunisia brieﬂy refers to agricultural enslavement in the
southern parts of Tunisia and, more importantly, contains a list of
sources for future research. Troutt Powell’s Tell This in My Memory
and Chouki El-Hamel’s Black Morocco also have something to say
about servile labor in agriculture, although neither deals with it
as a central theme. Finalmente, Walz and Cuno’s Race and Slavery in the
Middle East contains a few references to agricultural servility in the
Arabic-speaking Middle East beyond Cuno’s chapter about its pres-
ence in Egypt, which Reilly cites.

Understandably, Reilly, a newcomer to this subﬁeld, makes
addressing a lack of scholarly attention to MENA agricultural bond-
age his claim to fame. But he is slightly misleading to cite Lewis’
statement from 1990 about the “remarkable dearth of scholarly
work” on “slavery in a Middle Eastern context” given that a great
many works (including ones that Reilly cites) have emerged since
Lewis’ observation (2).6

Despite minor points of criticism, the bottom line is that Reilly
has written an excellent book that covers a lacuna of some signif-
icance in MENA enslavement studies, speciﬁcally in nineteenth- E
twentieth-century Arabia. Having deployed extensive and often
innovative and creative research, Reilly offers important insights
that can form the basis for future research into servile labor in agri-
culture in these and other regions of the Islamic and Indian Ocean
worlds. Despite its recourse to technical and scientiﬁc data to support
major arguments, this work is eminently readable, a valuable addi-
tion to the ﬁelds of Middle Eastern and enslavement studies.

6 Bernard Lewis, Race and Slavery in the Middle East: An Historical Enquiry (New York, 1990), vii.

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu

/
j
io

N
H
UN
R
T
io
C
e
–
P
D

F
/

4
8
3
3
8
5
1
7
0
2
1
0
5

/
j
io

N
H
_
UN
_
0
1
1
6
3
P
D

B
sì
G
tu
e
S
T

o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Scarica il pdf