Ayalon-SystemPaymentMamluk-1957

The System of Payment in Mamluk Military Society
Author(s): David Ayalon

Source: Journal of the Economic and Social History of the Orient , Aug., 1957, Vol. 1,
No. 1 (Aug., 1957), pp. 37-65
Published by: Brill
Stable URL: https://www.jstor.org/stable/3596038
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Brill is collaborating with JSTOR to digitize, preserve and extend access to Journal of the
Economic and Social History of the Orient
This content downloaded from

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
All use subject to https://about.jstor.org/terms
37
THE SYSTEM OF PAYMENT IN

MAMLUK MILITARY SOCIETY
BY
DAVID AYALON
(The Hebrew University, Jerusalem)
TABLE OF CONTENTS
Preface
Introductory Notes
I. Mamluk Army Units, Military Schools, etc.
2. The Ratio between the Dirham and the Dinar
The List of Payments to the Mamluk
Pay in the Bahri Period
Pay in the Circassian Period
(a) TheJadmak'ya
(b) The Nafaqa
The Nafaqa of the Amirs
Pay to the Mamluks of the Amirs
(c) The Dress (Kiswa)
(d) The Meat (Lahm)
(e) The Sheep for Sacrifice (Adbiya, rarely Dahayd)
(f) The Fodder ('Aliq)
(g) The Horses and Camels
The Distribution of Horses
I. to the Amirs
2. to the Mamluks
Plots for Growing Clover as Horse Food
The Camels
The Prices of the Horses and Camels
Total Amounts of Payment
i. Sums Paid to a Whole Expeditionary Force
2. Total Amounts of Other Payments
The Offices Responsible for the Army's Pay
The Office of the Vizir (Difwdn al-Wizdra)
The Office of Ustdddr (Diwdn al-Ustdddrzya)
The Office of Ndzir al-Khdss
The Army Expenditure and the State of the Treasury
Confiscation and Extortions of Money to Cover the Expenses of the Campaigns
Appendix A: The Expenses of sultan Q~ytbay during the Early Years of his
Reign
Appendix B: Data on Mamluk Currency

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
38 D. AYALON
Preface1)
No great religion in human history has ever been so intimately con-
nected with and so much dependent upon military might as Islam has
been from its very inception. Yet our knowledge of muslim armies and
military societies, especially in the Middle Ages, is extremely meager.
This is partly due to the fact that the study of the subject has been greatly
neglected by the orientalists; but, in my opinion, this is not the main
reason. The main blame for such a gloomy state of affairs should be laid
at the door of the Muslim sources themselves, for though these sources
deal so much and so often with wars, campaigns and battles they say very
little about the structure, organization and functioning of the armies
involved. In this respect the sources of the Mamluk sultanate (i 2 0 o-1517)
constitute an outstanding exception, for they contain information on the
Mamluk army and military society which far surpasses in richness,
variety and accuracy anything of the same kind which can be found in
other Muslim Mediaeval sources.2) This fact is of extremely great signi-
ficance, for the Mamluk army was not a force of secondary importance
in comparison with other Muslim armies. In its heyday it was the
strongest army in Islam and one of the strongest in the world.3) Even
i) This paper deals with the various kinds of pay which the mamluk received,
except his income from his feudal fief. The paper is a chapter of a work, origin-
ally written in Hebrew, on the Mamluk military society and army. The com-
plete work has not yet been published. Parts of it were published either in book
form or as articles.
2) There is no doubt that a much better and more reliable picture can be
drawn of the army and of the military society of the Ottoman Empire than of
those of the Mamluk sultanate, but this is true only of the Ottoman Empire from
the sixteenth century onwards. Our knowledge of that empire in the Middle Ages
is much inferior to our knowledge of the Mamluk sultanate. As for the technical
military treatises as a source for the study of the history and organization of the
Muslim armies of the Middle Ages, they can be safely used only after exhaustive
scrutiny and with the greatest caution. Some of the limitations and defects of
those military treatises were pointed out in my work "Gunpowder and Firearms
in the Mamluk Kingdom", London, 1956, pp. XI-XIII, XV-XVII.
3) Ibn Khaldtin, who spent many years in the Mamluk sultanate, states in un-
equivocal language that the center of power in the Muslim world passed from
the Abbasid caliphate to Egypt owing to the Mamluks. In a passage which
throws a most clear light on the Mamluk system of servitude and which deserves
special attention and study (Kitadb al-'Ibar, V, p. 371, 11. 4-27) he pays a glowing

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
PAYMENT IN MAMLUK MILITARY SOCIETY 39
at the time of its decline few Muslim armies could compare with i
very doubtful whether the Ottoman army, which was in the zenith
power in the early sixteenth century, could have defeated the dec
Mamluk army at all in 1516-17 without the extensive use of the
revolutionary weapon, firearms, a weapon which the Mamluks cou
adequately employ for socio-psychological reasons.1) Nor is this a
Mamluk army, like a great and vital section of the Ottoman arm
longed to a special institution which developed only within the bo
ries of Islamic civilization and without which the history of Islam
have been very different, viz., the institution of the so-called "s
armies which lasted for no less than a thousand years from the
ninth to the early nineteenth centuries. Thus the study of the Ma
in Egypt and Syria from the middle of the thirteenth to the beginnin
the sixteenth centuries inevitably has far wider implications tha
period or the region with which it is directly concerned.
What was said in the preceding lines about the Mamluks of Egyp
Syria in general applies to their system of payment as well. It may we
that the following account of Mamluk pay will not look very impr
in comparison with what is known about the pay system of Euro
armies of the same period, where the scholars studying the subjec
in many cases use archive materials, a kind of source which is so
for the Muslim countries in Mediaeval time.2) As far as the worl
Islam is concerned, however, there is little likelihood of its being p
tribute to the Mamluks and declares roundly that it was they who saved
when it was on the verge of extinction.
i) See "Gunpowder and Firearms, etc.", pp. 46ff and pp. io8-i I.
2) Most of the surviving archive material of Mediaeval Islam has come
to us through the contemporary historians who copied state documents
works from time to time. Even in this respect the Mamluk sultanate se
have been luckier than most other Muslim Mediaeval states or countri
few Muslim historical works can compare in the richness of their archive m
to Masdlik al-Absdr of Ibn Fadl Allah al-'Umari and to Subh al-A'sh
Qalqashandi. The original Mamluk documents found in the convent of St
rine's in the Sinai peninsula and part of which are now being prepared
publication by the Hebrew University bear striking testimony to the me
care with which al-Qalqashandi copied the documents contained in his b
to the accuracy of his information and comments on the same.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
40 D. AYALON
to furnish a description of
which could compare with
One of the great merits o
the bulk of these can be
for most of the data conc
isolated statements about the amounts due to the mamluk for each of the
various items of his salary. Where such general statements occur, they
have to be corroborated by strong additional evidence, not usually
available, which is a serious disadvantage.1) The great advantage of the
Mamluk data lies in the fact that they are given in connection with
actual cases of pay. The stereotyped kind of information is the following
one: The mamluk was entitled to a certain sum of money, but the autho-
rities, mainly owing to the lack of funds, paid him a smaller sum. Both
sums are mentioned by the sources. This kind of information repeats it-
self over and over again, and the figures always tally. So much for the
pay of the individual mamluk. The same applies to the figures quoted
by the Mamluk sources for the sums paid to the whole army, to a whole
expeditionary force or to a considerable section of the army. The accura-
cy of these figures can be established by means of multiplying the pay
of the individual mamluk by the number of the soldiers involved in
each case. As I have tried to show elsewhere, the Mamluk sources usu-
ally furnish very accurate figures on the numerical strength of the Mam-
luk armed forces.2)
A wealth of information is supplied by Mamluk sources on the curren-
cy of the realm and on the ratio between the different monetary units.
The reliability of this information can be established by the same method
I) The numerous general statements and definitions contained in the Mam-

luk encyclopedias must be carefully checked with the aid of the day-to-day in-
formation furnished by the chronicles. In most cases they have to be consider-
ably modified in the light of this information.
2) D. Ayalon, "Studies on the Structure of the Mamluk Army", BSOAS,
1953, pp. 222-228; 1954, pp. 70-73. The figures on the size of the Mamluk
military expeditions are also accurate as I have attempted to prove in an un-
published chapter of my work on the Mamluks called "The Army on the
March". Nor is that all: there is, in my opinion, complete accord between the
figures on the size of the Mamluk army, the size of its military expeditions and
the sums paid to the army (see below the sections: "Sums Paid to a Whole Ex-
peditionary Force" and "Total Amounts of other Payments").

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
which was adopted here in order to establish the reliabi

formation on the army's payl). Any serious attempt at th
economy and finances of the Mamluk sultanate should,
as its basis and starting-point the currency and pay system
society, for in these two domains we tread on firm ground.
Additional reasons for giving procedence to the army
society over any other section of the population in the st
nomy of Egypt and Syria under the Mamluks are the fo
whole life and economy of the realm were geared to th
Mamluk upper class; b) Our knowledge of the ruling mili
incomparably better in almost every respect than our kn
other class.
From the above it should not be concluded that after

a most detailed and trustworthy picture of Mamluk cur
(including the revenues from the fiefs) one can expect
harvest in the study of the economic problems of the rea
work has still to be done in many other fields of Maml
order to render such study successful, we should not att
at one blow; only limited targets should be set at first,
because so few Islamists are equipped with the requisite e
ledge. In the present paper I have deliberately refrained
cending the strict limits of the subject and from dealin
plications of my conclusions on the general economy
state.
Introductory Notes
Before starting with the study of the main subject of the prese
paper, it is essential to do two things. First, to explain briefly some of
terms - especially those pertaining to the army units and to the mil
schools - which recur most frequently in this paper, but which
described in detail in other chapters of my work on the Mamluk mil
society. Second, to give a short sketch in a chronological order of
changing ratio between the two principal coins of the realm: the dir
and the dinar.2)
I) On the Mamluk currency see below: "Introductory Notes".

2) The references given in the following notes on Mamluk currency are
a tiny part of the mass of data available on the subject. See also Appendix

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
42 D. AYALON
i. Mailuk Army Units, M

The Mamluk forces sta
parts:
I. The Royal Mamluks (al-mamidlik as-sultdnmya). These were of two
categories.
(a) The mamluks of the ruling sultan (mushtarawdt, ajldb or iulbdn).
(b) Mamluks who passed into the service of the ruling sultan from the
service of other masters (mustakhdamin). These were divided into two
groups.
I. Mamluks who passed into the service of the reigning sultan from
that of former sultans (mamidlik as-saldtin al-mutaqaddima, qardenfs, or
qardniFsa) ;
2. Mamluks who passed into the service of the reigning sultan from
that of the amirs, because of the death or dismissal of their masters
(sayfzya).
II. The Amirs' Mamluks (mamidlik al-umard', ajnidal-umard').
III. The Troops of the Halqa (ajndd al-.halqa), a corps of free, i.e. non-
mamluk, cavalry. There was within the .halqa a special unit composed of
the sons of the amirs and the mamluks, called awlddan-nds.
The Royal Mamluks constituted the backbone and the elite of the
military society. They were given first-rate training, and in the Bahri period
(648/1250-784/1382) numbered not less than io,ooo troops; in the
Circassian period (784/1382-922/15 6) they never exceeded this figure,
and usually fell far short of it. In the major engagements fought by the
Mamluk sultanate, the brunt of the fighting was borne by the Royal
Mamluks, who formed the main force in all the important military ex-
peditions. In the words of al-Qalqashandi: "The Royal Mamluks are the
most important and the most honoured part of the army, the closest to
the sultan, and the owners of the largest feudal estates, and it is from
among them that the amirs of various ranks are appointed".') The history
of the Mamluk sultanate is actually first and foremost the history of the
Royal Mamluks. No other unit could compare with this corps either in
military efficiency or in political power. What greatly increased their
power and their influence on the affairs of the state was the fact that they
were not spread out as garrison troops in scattered parts of the realm,
I) Daw' as-Subh, p. 243, 11. I1-12.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
but were all concentrated in the capital, with a large section

quartered in barracks in the Cairo citadel, which served both
of the sultan and seat of the government.
The Royal Mamluks did not, however, constitute a hom
unit. Their division into separate bodies, or groups or faction
sing. td'ifa) was based solely on the allegiance of these bodies t
sultans. The Mamluk system of servitude instilled in the
feeling of profound loyalty toward his master and liberator
the one hand and toward his fellows in servitude and liberation (
sh/ya) on the other. This twofold loyalty was one of the pr
around which the entire military and social system of Mamlu
revolved. The sultan and his mamluks formed a tightly-knit
whose members were united by strong bonds of solidarity.
the Mamluk sultanate was wealthy and powerful, and so long
was well-trained, well-disciplined and ruled with an iron
feeling of fellowship or khushddshzya constituted a positive
esprit de corps, tending to foster moderate competition among t
bodies, stimulated the troops and prevented stagnation. With
of the sultanate under the Circassians, however, when discip
weak and the selfish impulses of the various bodies burst fo
unbridled manner, this feeling of solidarity turned against th
became one of the main factors which sapped its strength. T
bodies of the Royal Mamluks indulged in politics and in a st
power. In this struggle the mamluks of the ruling sultan (aj
tarawdit) were bound to win, because they had the full suppo
master the sultan; because they were more compact than the r
forming a homogeneous td'ifa; because they were more num
because they were younger than the members of the other tawd'
a very short time after the beginning of Circassian rule their
over the other bodies of the Royal Mamluks was firmly establ
were the real rulers of the realm and even the sultan himself bec
captive. The khIssak~ya, i.e. the bodyguard or select retinue w
nished all the pages and most of the amirs of the sultanate, w
picked from among the mushtarawdt.
Second in importance were the mamluks of the former sultans
In contrast to the mushtarawdt, who formed a single and hom
.td'ifa, the qard~ns did not, naturally enough, constitute a un

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
44 D. AYALON
Their factions were se

between them was their
juniors and lorded it ov
forget how the mamluk
from the status of musht
same grudge against tho
of conflict among the v
considerable, especially
these factions was very
reigns. The existence of
was skillfully exploited
reduce the influence and
tans.
The sayfitya were Royal Mamluks who had formerly been mam
amirs. Each of these had served under a different master before he was
thrown together with the others into the same td'ifa. No wonder, there-
fore, that the sayfzya were indifferent and strange to one another, to the
mushtaraawdt, to the qardn4 and to the sultan. They lacked the essential
unifying bond of solidarity with the khushddsh on the one hand and of
loyalty to the ustddh on the other. Their allegiance to any particular
sultan was loose and vacillating. They were extremely weak as a political
unit and seldom acted independently to improve their position. They
usually made common cause with the qardunf in their struggle for a
better status.
The absolute supremacy of the mushtarawdt over the qardngS and the
sayfzya found its expression in numerous ways, of which the most con-
spicuous were the participation in battle and the pay. The sultan tried
to spare the lives of his own mamluks as best he could and to send the
members of the other two groups into the thick of the fighting. He also
discriminated the aqrdnJs and s~ayflya by often giving them lower pay than
to his mzusbtarawdt, though legally they were entitled to the same pay.
The mamluks of the amirs were socially far inferior to the Royal Mam-
luks. They constituted no serious political factor in the Mamluk military
society; rebellions and mutinies of great proportions were extremely
rare among them. In addition, they were necessarily less well trained
than the Royal Mamluks, for they did not have access to the first-rate
military schools in which the latter grew up and studied. As we shall see

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
below the political weakness and the insignificance of the

luks is accurately reflected in the treatment they received
masters regarding their pay and in their lukewarm and resig
to this treatment.
The halqa constituted the most important non-mamluk ele
sultanate's army. It seems to have been created by Salih
Ayyubid. Under the Ayyubid and early Mamluk sultans it
extremely high position both socially and militarily and was
to the Royal Mamluks. With the steady rise of the Mamluk elem
military society the .halqa was bound to suffer heavily. Th

which fell upon it was the land re-allotments following the
veys (rawks) of the end of the 7th/I3th and the beginning of t
centuries and which resulted in drastic curtailments of the .

fiefs. The decline of that unit which continued throughou
teenth century was greatly accelerated in the fifteenth. Wit
there was an 6lite unit called awldd an-mds. These were the s
luks and Mamluk amirs, born as free men and as Muslims in
sultanate and bearing Arab names. As such they could not
into the Mamluk upper class and were therefore relegated t

The awldd an-ntis and the .halqa declined together and g
dividing line between them was obliterated. Eventually the
fell into disuse, and one could find under the name of awlid
by side with the sons of mamluks, Cairo merchants, artisans
of nouveau-riche elements who paid their way into the regimen
improve their social standing. Though they were still called
did little to justify that name. Both the .halqa and the awldd
deprived of their horses and ceased to take part in militar
No wonder, therefore, that these two units were the most
element in the whole military society of the Mamluk sultan
our information on their pay is much scantier than the inf
the pay of the other elements of which that society was compo
We would emphasize in this connection that the data on th
various sections of the civilian population of the sultanate a
far and less reliable than those on the pay of the military. It th
not serve as a basis for any study of real value.
Brief notice will be made now of the Mamluk officers
called amirs (sing. amir, pl. umard') and were divided into three

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
46 D. AYALON
(a) The highest rank was that

entitled to keep in his service
soldiers of the .halqa in the field
of the state were selected from
officially fixed at 24, but whic
command of a thousand .halqa s
titious even in the period when t
yet the most common name of
(amir alf).
(b) The second highest rank was
or amir .tablkhdna). Each amir o

in his service 40 horsemen, wh
even 80.
(c) The third rank was that of an Amir of Ten (amtr 'ashara). All such
amirs were entitled to keep in their service between io to 20 horsemen.
The number of the Amirs of Forty and Amirs of Ten was not fixed
At the closing decades of Mamluk rule we find the number of the first
group fluctuating between Io and 75 and that of the second group
varying between I8J and 225.
A few words must be said in the present context about the military
schools or barracks of the Cairo citadel. These were called tibdq (sing.
tabaqa) and were exclusively dedicated to the training of the Royal
Mamluks. The Mamluk boys brought from abroad spent several years
in these institutions until they reached maturity and became fully-fledged
soldiers. On finishing the school they received their certificate of manu
mission together with their horse and equipment. After having left the
school the mamluks did not sever their connections with it. They con-
tinued to belong to the old barracks where they had received their recruit
training. This connection with the old barracks was expressed in many
ways, one of them being the order of the pay parade. The mamluk
received their pay according to their tibdq, the total number of which was
about twelve. The most important and famous among these tibdq were
tabaqat ar-Rafraf, tabaqat at-Tiziya, tabaqat az-Zimim and tabaqat al
Ashrafiya. A very considerable part of the barracks' personnel was com-
posed of eunuchs who served as a barrier between the young recruits
and the adult mamluks. The head of the whole school (muqaddam al
mamdlik as-sultAinya), his deputy (nd'ib muqaddam al-mamdlik as-sultddnya

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
and the heads of each of the twelve barracks (muqaddamii at-t

all eunuchs.1)
2. The Ratio between the Dirham and the Dinar
During the Bahri period and the first two to three decades of the Cir-
cassian period there existed a more or less fixed ratio between the dirham
(the silver coin) and the dinar (the gold coin). When changes did occur
they usually were slight. The dinar was ordinarily equivalent to 20
dirhams. Sometimes its value rose to 25 or 28 dirhams.2) But since the
reign of the Circassian sultan Faraj (80I41398-808/1405; 809/1406-815/
1412) the dirham became more and more debased as a result of the
growing practice of mixing it with inferior metals. The dinar, on the
other hand, was almost always made of gold3) both under the Circassians
and in the period which preceded their rule4). Thus the difference in the
respective values of the two coins steadily increased. Mamluk sources
say that the debasement of the dirham started in 806/1403, i.e. in exactly
the same year which, according to the same sources, witnessed the be-
ginning of the general economic decline of the realm.5) In actual fact
the debasement of the dirham started somewhat earlier as did the econo-
mic decline of the Mamluk sultanate.
In 805/I402 the dinar was equivalent to 60 -65 dirhams ;6) in 807/1404

to 100oo dirhams;7) in 815/1412 the dirham declined at an accelerated
pace;8) during the period 818/I415-856/1452 the value of the dinar
I) The above description is based on "Studies on the Structure of the Mam-
luk Army", BSOAS, London, 1953, pp. 203-228, 448-476; 1954, PP. 57-90;
and on "L'Esclavage du Mamelouk", Jerusalem, 195 I.
2) Sub , III, p. 442; p. 444. Ta'rikh Bayrit, p. 112. Nujam, V, pp. 73-74; PP.
351-352; P. 560, 11. -6; VI, p. 7; P. 71; p. 272. Zetterst~en, p. 224, 11. I1-14.
I.usn al-Muadara,
3) For II, see:
an exception p. 213, 11. 24-26.
Ibn Iyis, V, pp. 86-87.
4) This does not mean that the purchasing power of the dinar was the same
under the Bahris as under the Circassians. In the present state of our knowledge,
however, there is no possibility whatsoever of establishing even approximately
the changing value of the dinar throughout Mamluk rule.
s) For a description of the economic decline of the Mamluk sultanate see
"Gunpowder and Firearms, etc.", pp. o103-107.
6) Nujzm, VI, p. Io 6, 11. -6.
7) Nujbm, VI, p. I i .
8) Sub , III, p. 442.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
48 D. AYALON
changed between 230 and 285

the dinar was equivalent to
cessful attempts were made t
dinar);2) in 862/I458 the valu
the dinar.3) From that date o
fix the ratio between the d
evidence, however, that th
continued without break to th
The List of Payments to the Ma

The mamluk received the fo
(a) the jdmakiya - the mo
(b) the nafaqa - mainly o
the accession of a new sultan to the throne
(c) the kiswa - a yearly or a half-yearly pay to cover the mam-
luk's expenses on his dress
(d) the la.hm - the daily meat ration
1)Nujzjm, VI, p. 27;p. 3 5 6;p. 5 37;p. 5 96; pp. 667-668; VII, p. 11;p. I I;
p. 224. HIawddith, p. I1; p. 68; p. 76; p. 99.
2) H.awadith,
Nujpm, VII, p. 473. p.
Ibn208;
Iygs,p.II,225;
p. 57.p. 231; p. 271; pp. 294-297.
3) Nujam, VII, p. 496. Ibn Iyis, II, p. 6i.
4) The difficulty lies in the fact that the sources no longer mention the ratio
between the dirham and the dinar. They either mention the ratio between the
nisffidda dirham and thefls (pl.ful/s) or quote the value of a ra.tl of/jul/s(Nujam,
VII, p. 496. HIawddith, p. 332; p. 445; P. 503; P. 51o; p. 530o. Ibn Iyis (KM),
III, p. IS; p. Ioz02; p. 117; p. 231; p. 385; IV, p. 251; p. 295; p. 327; p. 338; V,
pp. 87-88). I could not fix for certain the exact relation of the nisffldda to the
fils or to the dinar during the greatest part of the period of the existence 6f the
first-named coin. On the fils and the nisffldla, see E. Strauss, "Prix et Salaires a
l'Epoque Mamlouke", REI, 1949, PP. 52-53, where the changing value of the
dirham in relation to the dinar is also discussed.
5) For an illuminating description of the adulteration of the currency under

the Mamluks see: Ibn IyZis, V, pp. 87-88. During the greatest part of Circassian
rule the sultanate was swamped by European gold coins with images (ad-dandnfr
al-ifrantjya al-mushakhkhasa) which were of much higher quality than the local
coins. According to Ibn Taghribirdi these coins swept not only Egypt and
Syria but also the Yemen, the IIij iz, the Ottoman state (bildd ar-Rim) and Iraq
and Iran (bi/ddash-sharq) (Nujtim, VI, p. 596).

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
(e) the aad.hiya - sheep distributed to the mamluks on

of the Feast of Sacrifices
(f) the 'aliq - the fodder ration (distributed twice a week)

(g) al-khayl wal-jimdl - horses and camels (distributed irregularly)
Pay in the Ba.hri Period

Our information on the various kinds of pay in the Mamluk sultanate,
extremely rich though it is, suffers from certain defects, the most im-
portant of which is that it is not evenly distributed; for the bulk of this
information belongs to the later Mamluk period and only a small fraction
of it pertains to early Mamluk rule. The reason for this unevenness is not
far to seek: during the early part of its existence the Mamluk sultanate
was rich and prosperous and its pay machine worked smoothly, thus
attracting little attention. In later years, however, when the sultanate be-
came impoverished and its economy declined, it became increasingly
difficult to pay the army in full and on time. We owe it to the constant
struggle between the army and the treasury over the salaries that such a
wealth of data is available on the Mamluk system of payments in the
Circassian period.
Our information on pay in the Bahri period is not only scanty but also
stereotyped for it is limited to almost a single kind of pay, viz., the
nafaqal.)
Under the Bahri Mamluks the nafaqa to the various ranks of the army
was as follows: Amir of a Thousand received between 1,ooo000 and 3,000
dinars (very important amirs received up to 10o,ooo000 dinars each in ex-
ceptional cases); Amir of Forty - between 400 and 1,ooo000 dinars; Amir
of Ten - no information is furnished on the nafaqa of this category in
the sources at my disposal; muqaddam al-.halqa - 1,ooo000 dirhams (about 5o
dinars); Royal Mamluk - 0oo dirhams (about z5 dinars)2). When Sultan
an-Nisir Muhammad b. Qalhiin paid, after his defeat by Ilkhin Ghizin
in 699/I299, the sum of 5o dinars to every Mamluk private soldier, it
was regarded as something extravagant and "unheard of" (anfaqa fi al-
I) For a detailed description of the nafaqa see below the section dealing with
"Pay in the Circassion Period".
2) Sulik, I, p. 449; p. 682. 11. 20-22; p. 724; II, p. 499, 11. 7-9. Nuj]m (c),
VIII, pp. 259,1. i8-260, 1. 5; IX, p. 146, 11. I1-I4. Zetterst6en, p. z128; p. 228.
Nujim, V, pp. 14, 60.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
Jo D. AYALON
jaysh nafaqa md sumi'a bi-mithlihd qa.t.t kdinayju't al-juzndya khamsbna dindran)

and this in spite of the fact that the Mamluk army lost most of its equip-
ment and ammunition in the battle against the Mongols of Persia.1)
From the data presented it may be concluded that the size of the amirs'
nafaqa in the Bahri period was not very different from that of the nafaqa
of the amirs in the Circassian period, though its limits were not as fixed
and rigid. The nafaqa of the simple mamluks, on the other hand, was
much lower in the Bahri period. Near the end of Bahri rule the nafaqa of
the simple Mamluk rose considerably and in the transition period
between the Bahris and the Circassians it reached its highest peak.2)
In this connection it would be desirable to make a comparison betwen
the pay of the Mamluk and Ayyubid armies. Unfortunately, such a com-
parison is not feasible in the present state of our knowledge.3) I would
like, however, to draw attention to a passage in the chronicle of the late
Mamluk historian Ibn Taghribirdi (died 874/1469) which might have
been copied from an earlier source. According to that historian the bud-
get of the Mamluks, and especially their expenses on the army, were
much higher than those of the Ayyubids. The army of the late Ayyubid
sultan al-Kimil Muhammad (615/I2i8-635/i238) and his son as-Salih
Najm ad-Din Ayyib (637/1240-647/I249) numbered Io,ooo men. They
wore modest uniforms, their equipment was scarce and their pay barely
sufficient. The Mamluk sultan al-Malik az-Zihir Baybars (65 0o/1260-
676/1277) raised the number of his army to 40,000 and spent on it huge
sums of money.4) Only additional data will establish the accuracy of this
isolated statement of Ibn Taghribirdi's.
Pay in the Circassian Period

(a) The Jdmakya
The jdmakya, the monthly pay, was the most regular and frequent pay
of the army.5) Its payment was normally announced a few days before
I) Duwalal-Isldm, II, p. 15 8, 11. I3-I8.

2) See below: "Pay in the Circassian Period".
3) My impression from the reading of Ayyubid sources, which was by no
means exhaustive, is that their data on the pay to the army are far less numerous
and far less reliable than those of the Mamluk sources.
4) Nuj]m (c), VII, pp. 197-198.

5) The term jdmak[ya is quite frequent in Ayyubid sources. Its use before the
Ayyubid period has not been followed up by the present writer.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
PAYMENT IN MAMLUK MILITARY SOCIETY 1I
the fixed date.') Its cancellation was made known by means of the
"torch-bearers" (mashd'il~ya)2). Its distribution usually started at the begin-
ning of the second half of the Muslim month, probably on the first
day of ceremonial procession (mawkib)3) in that part of the month. The
I5th, I6th or I7th of the month were the most common days for the
commencement of the jdmakya pay4), though instances are known of
payment being made somewhat earlier or later.5) The distribution of the
idmakya lasted ordinarily until the zoth or 21st of the month, but occa-
sionally it was stretched up to the 27th.6) The old custom was to finish its
payment in three consecutive days of ceremonial procession. In 85 8/1454
the mamluks of the ruling sultan (iuzlbdn) stoned the Majordomo (ustdddr)
for stretching its payment over five mawkib days instead of three.7) In
861/1457 the mamluks demanded again that the payment of the jdmakya
should be completed during three mawkib days only.8) But this demand
was never met. During the last 70 to 80 years of Mamluk independent
rule thejdmakya was distributed in four mawkib days.9)
Both the jdmakya and the nafaqa were paid in a special parade ('ard)
which took place in the Royal Courtyard (al-.hawsh as-sulztdtf) in the Cairo
i) Nuj~ m, VII, p. 425, 11. 6-7; p. 691, 11. 4-5; p. 826, 11. 12-13; p. 829, 11.
14-15 ; P. 830, 1. ii.
2) Ibn Iyis, V, p. 430, 11. I5, x8, I9. The mashd'iltya served mainly as exeuc-
tioners, leaders of punished persons under disgraceful conditions, town criers,
etc. See Van Berchem, Corpus Inscriptionum Arabicarum, II, Syrie du Sud, J~rusa-
lem, HIaram, p. 320, note 7. Al-Maqrizi, Histoire des Sultans Mamluks (ed. Qua-
trembre), vol. I, part II, p. 4, note 5.
3) On the mawkib see L. A. Mayer, Mamuk Costume, Genbve, 1952, p. 78, and
note 3, where the difference between mawkib and khidma is also explained.
4) I.Iawddith, p. 698, 11. 6-7. Ibn Iyis (KM), III, p. 391, 1. 20; IV, p. 18, 11.
18-19, 2i; p. 237, 11. 18-20; p. 291, 11. 3-6; p. 307, 11. 8-ii; p. 312, 11. I3-i4; P.
326, 11. 4-6; p. 330, 11. 4-6; p. 350, 11. 20-23; p. 360, 11. 2-3; p. 382, 11. 1I-16;
pp. 386, 1. 22-387, 1. 7; P. 416, 1. 16; p. 430, 11. 20-21; p. 480, 11. 9-10; V, p. 19,
11. I6-Is8; p. 78,11. 5-8.
5) Hawddith, p. 134, 1. 23; p. 194.
6) -Iawddith, p. 514, 11. 9-14. Ibn IyIs, V, p. 45, 11. 12-I5. See also ibid., IV,
p. 471,11. 4-5 ; P. 483,1. 11i.
7) Nujam,
8) I.awadith,
VII, p.p.474,11.
201, 11.
6-I 3;I0o-12.
P. 477,11. 5- II.
9) Hlawddith, p. 486, 11. 6-io. See also the references in the following notes.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
52 D. AYALON
citadel.1) When Sultan

in 909/I 5o03 he som
always took place in t
schools (muqaddam al
Chief of the Corps o
markets (mu.htasib)
cials;3) but al-Ashra
time after his becom
during the pay of th
of Mamluk rule, pay
During the parade t
In the absence of the
deputized for him. T
together with the hig
he who read the nam
official parades.5) Bo
to the tabaqas. Each
mamluk, on hearing
advance towards the
I) HIawddith, p. 68i,
696, 11. 8-9. Ibn Iyas (
the sultan transferred h
bench called mastaba (Ib
2) Ibn Iyis, IV, p. 144,1
in the Mamluk Kingdo
place in the mayddns in
during the greatest par
whereas in ra's
3) On the thenawba,
Bah.riseeperiod next"Studies
D. Ayalon, to nothing
on the is knownofabout
Structure the pay-parades.
the Mamluk
Army", BSOAS, 1954, pp. 60-6i. See also the references in note i .
4) Ibn Iyas (KM), III, p. zz22, 11. 21-24; pp. 323,1. 21-324,1. Io.
5) Nujqm, VII, p. 218, 11. io-zo20. Hawrddith, p. I13, 11. 7-23. Ibn Iyis, IV, p.
291,11. 3-6; p. 307.11. 8-u ; p. 3 3,11. 20-23; p. 413, 11. 6-7; P. 416,11. u6-u9;V, p.
19, 1l. I6-i8; p. 78. 11. -8. Kdtib al-mamdlik or kitib al-mamdlik as-sulftdnya
(his office was called kitadbat al-mamilik) was in numerous cases of Coptic origin
Often he was a Q.di (Nuidm, VII, p. 449, 11. Io-I I; p. 713, 11. I4-I 5 ;P. 765,1. 18.
Hawdidith, p. 193, 1). I-3; p. 220zzo, 11. 3-4, 7-9; P. 363, 1. I; p. 672, 1. 17. IbnIyis
(KM), III, p. 143, 1. 5; p. 240, ll. 2z-1; p. 359, 1. I9;p. 388, 1. I ; IV, p. 99, 11.
i6-17; p. xx8, 11. 3-6; p. 2zI, 1. 7;p. 321, 11. 1i-I2).

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
receive his pay in a purse (surra) and return to the ranks.1) When the
mamluks received their meat and other rations in money instead of in
kind the same procedure was repeated (the sources use in such cases the
following or a similar expression: wa-yastad'ihim wd.hidan ba'da wd.hidin

mithla tafriqat al-jdmakiya).2)
Usually three to four .tabaqas received their pay in one pay-day.3)
Tabaqat ar-Rafraf was always paid first. The second was, in most cases,
tabaqat at-Tgziya.4) When the treasury was empty, however, only two
tabaqas, sometimes only one tabaqa, would be paid on the same pay-day.5)
On some occasions pay and other parades were stretched over longer
periods not only because of the lack of money, but also for military
reasons: to make the enemies of the realm believe that the Mamluks had
got a very big army. When Sultan Q~nsofh al-Ghawri, on the eve of the
battle of Marj Dibiq, paid the nafaqa to his whole army in four days, he
was strongly criticized by his amirs who feared lest this hasty procedure
would lead the rulers of the Ottoman Empire and of Safawid Persia to
the conclusion that no army had been left in Egypt, which would tempt
them to attack the country.6) For the whole Mamluk period, I came
across one single case in which the distribution of pay to the army was
completed in a single day. This happened in 873/1468 when an expedi-
tionary force was hastily organized to march against the Turcoman
Shah Siwar. The pay parade lasted from morning until dusk.7)
The contemporary sources say practically nothing about how the
mamluks were dressed on pay parades. An isolated remark by Ibn Tagh-
ribirdi throws some light on this subject. In 866/1462 Sultan Khush-
qadam ordered the Royal Mamluks to wear on these occasions their
ceremonial dress (qumdsh al-mawkib), viz. the kalaftdt - hat, the tight-
sleeved coat (qabd') and the sword, "in accordance with the old custom"
I) Nujfm, VI, p. 533, 1. i6; p. 534, 1. i5; P. 685, 1. 17; PP. 687-88. Ibn al-
Furit, IX, p. 33, 11. 16-17.
2) Ibn Iyas (KM), III, p. 285, 11. i5-2i; IV, p. 371, 11. 16-17; V, p. 8, 11. 5-8;
p. 28, 1. I5.
3) HIlawddith, p. 179, 11. I2-20. Ibn Iyis, V, p. 11 3,1. 23.
4) Nujim, VI, pp. 533-34; PP. 687-88. Iawdadith, p. 514,11. 9-14.
5) Ntujzm, VII, p. 693, 11. 9-1 3; p. 831, 11. 4-5.
HIawddith, p. 514, 11. 9-14. Ibn Iyis, IV, p. 19, 11. i1-14; p. 20, 11. 17-19.
6) Ibn Iyis, V, p. 27, 11. I8-zz22.
7) I.awddith, pp. 619, 1. 23-620, 1. 2.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
54 D. AYALON
(ka-md kdnat al-'dda al-qadim

the eunuchs of the military
strictly carried out.i) This i
rather slack about their dres
In the closing decades of M
Ghawri (906/1500-922/1x I6
jdmakyi a of several months
monthly pay" (al-jdmak~ya a
to an improvement in the f
opposite. Because of the emp
give the mamluks, on the
nafaqa. So, in order to cover
pedient of advancing them
four) months.2) This expedi
he sent his expeditions agai
Ghawri who could pay a naf
hundred) to each soldier w
Portuguese.4) There were, h
jdmakya was paid in additi
shall see below.
The existence of that expedient of accelerated monthly pay, though
very harmful to the mamluk, is very beneficial to the student of the
Mamluk system of payment, for it is mainly owing to this expedient
that we can know for certain the average amount of the jdmakya.
During a considerable part of the 9th/Isth century the mamluk
received ajdmak'ya of 2,oo000 dirhams, which the sultans refused to raise
in spite of the continued debasement of the dirham. In 859/I455 the
julbdn received ajdmaktya of 2,oo000 dirhams, i.e. hardly sj dinars, for the
dinar was equivalent at that year to 370 dirhams. They demanded instead
7 dinars, in accordance with what they got several years earlier when the
I) .Hawddith,
Mamluk Costume, pp.p.13,417,11. 13-I'.
14, 16, 17, On the
18, 21-23, 27. kalaftdh and the qabd', see L. A. Mayer,
2) The devastating effects on the Mamluk and his morale of such a proce-
dure, which made him wait for his first monthly pay a very long time after his
return from the campaign, are very obvious.
3) See the references in p. 60, notes 3-4, P. 6I, note I and p. 58 note 3.
4) See the references in p. 60, notes 3-4, P. 61, note I and p. 58, note 3

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
PAYMENT IN MAMLUK MILITARY SOCIETY JJ
dinar was equivalent to 28 dirhams, but their demand w

down.') It would appear that the mamluks had to wait many
fore their monthly pay had been somewhat increased. In 893
in 922/1i5 i6 ajdmakiya of four months amounted to 8,oo000 dirh
913/1507 ajdmakiya of four months amounted to 23 dinars.3
S151 6 ajdmakiya of three months amounted to 20 dinars.4)
The jdmakJya of the members of the .halqa and of the awldd a

smaller than that of the mamluks. They usually received a m
which varied between 1,ooo000 and 2,000ooo dirhams.5) Sultan
Ghawri promised the members of his unit of arquebusiers (a.
khdmisa) to raise their jdmakiya from 150oo dirhams to 2,000ooo
According to al-Maqrizi Sultan Barqiq (784/1382-80x /328)
ted a grave mistake by his introduction of the system of ac
promotion in the army, instead of the surer system of slow
which found it expression, inter alia, in the gradual raising of th
from 3 dinars to I o.7) This, by the way, is the sole piece of
came across which indicates the existence of a gradual raise of
the private Mamluk soldier. Unfortunately no contemporary
the Bahri period corroborates the evidence of the late al-Maqr
it would mean that the jdmakvya of the mamluk in the Circa
I) Nujtm, VII, p. 474. HIawddith, p. 223.

2) Ibn Iyis (KM), III, p. 246, 11. 9-1o; V, p. 26, 1. 6.
3) Ibn Iyas, IV, p. I19, 11. 6-9.
4) IbnIySis,V, p. xI 3,1. zz22; p. 123, 1. 19.
5)Ibn
6) I.awddith,
Iyss, IV, p.p.436.
685,For11.
the7-17. Ibn
pay to Iyis (KM),
at-tabaqa III, pp.
al-khdmisa and256-257; p. 323.
to other fire-
arms units composed of awldd an-nds, see D. Ayalon, Gunpowder and Firearms in
the Mamluk Kingdom, London, I956, pp. 63-65, 74-76. The tabaqa al-khadmisa
derived its name from the fact that it received its monthly pay on a fifth pay-day
at the end of the month and not during the four regular pay-days in the middle
of the month together with the rest of the army. This was a sympton of the
Mamluks' negative attitude towards the units employing firearms. From the
size of the jdmaklya of the awldd an-nds it may be wrongly concluded that their
pay was not very much lower than that of the Royal Mamluks. It deserves to be
emphasized, therefore, that thejdmakya was the only pay which the aw/dd an-nds
received with any degree of regularity. They had practically no income from
feudal fiefs during the later stages of Circassian rule.
7) Khi/at, II, p. 214.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
J6 D. AYALON
was more or less in the middle between the minimum and the maximum
monthly pay received by the mamluk of the Bahri period.
(b) The Nafaqa

The private soldiers amongst the Royal Mamluks received the nafaqa
mainly on two occasions: a. On the accession of a new sultan to the
throne. It was usually called nafaqat al-bay'al) (far more rarely: nafaqat
as-saltana)2); b. On the eve of a military campaign (nafaqat as-safar).3)
Most of the nafaqas in the Mamluk sultanate were given on this second
occasion. The nafaqa was also sometimes given in order to appease the
army (nafaqat targdiyat al-'askar)4) or on the occasion of the first visit of
the sultan's adolescent son to the town of Cairo (nafaqat nuzizl ibn as-
sultdn).5)
As for nafaqat al-bay'a, great efforts were made to pay it in full, for the
new sultan, who did not as yet feel strong enough, wanted to be on
good terms with the army.6) It happened, however, quite frequently
that the full amount of this kind of nafaqa could be paid only to few
categories of the armed forces.7) Sultan Barsbay (825/I422-842/I438)
was the first in Mamluk history who did not pay the nafaqat al-bay'a at all.
Strange as it may seem, the mamluks did not even demand it from him,
an attitude for which Ibn Taghribirdi can find no explanation.8) Al-
AshrafQ ytb y (873/I468-901/1495) madeit a condition of his acceptance
of the office of the sultanate that he should be exempted from paying
the nafaqat al-bay'a.Y) Often the problem of the nafaqat al-bay'a became
complicated by the habit of every Circassian sultan of nominating his
i) Ibn Iyis (KM), III, p. 399, 1. II; p. 433, 1. 3; P. 462, 1. 14; IV, p. 8, 1. 19; p.
25, 11. 20-22; p. 41, 11. 12-13.
2) Nujim, VI, pp. 6-7; VII, p. 648, li. I , 17; P. 650o, in the notes. Ibn Iyss,
II, p. 25, 1. 16. Poliak, Feudalism, p. 4.
3)Ibn
4) Nujum, VI,III,p.p.33I,13,11.
Iyas (KM), 11 17-18.
i- ; IV, p.I.awddith, p.
3 1 8, 11. 2-6. 702, 1. 7. Ibn Iyis, V, p. 26, 1. 4.
5) Ibn Iyis (KM), III, p. 291, 1. I2.
6) Nujz7m, VII, p. 6, 11. 3-7; p. 37,1. I.
7) Nujam, VII, p. 428, 11. 10o-1 5. Ibn Iyis, II, p. 41, 11. I-7; p. 66, 11. 28-29;
(KM) III, p. 329, 11. I8-2I; p. 435,11. 19-2I; IV, p. 13,11. 23-24,1. 6.
8) NAujm, VI, p. 553, 11. I4-I7; p. 559, 11. 12-14.
9) Ibn lyis, II, p. 91, 11. I7-I8; (KM) III, p. 4,11. o10-Il.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
son as his successor. The son usually ruled for only a very short period,1)
yet the mamluks demanded the nafaqat al-bay'a both from the new sultan
and from the son of the old sultan.2)
With nafaqat as-safar the mamluk had to cover all his own and his ser-
vant's (ghuldm) expenses in connection with the impending military
campaign, including the repairs to his equipment and uniform.3) The
nafaqa was not distributed very frequently. Sultan al-Ashraf Khalil
(689/I290-692/1I293) received special praise because he paid the nafaqa
to the army three times during his short reign of three years.4)
The official amount of a nafaqa of any kind was i oo dinars. During the
early part of Circassian rule the mamluks did receive that sum in full ;5)
but later on the cases of paying a lower sum became increasingly fre-
quent. According to the historian Ibn Taghribirdi, al-Ashraf AynBl
(857/1i45 3-865/1460) was the first Mamluk sultan who discriminated
between the various categories of the army in the payment of the nafaqa
and this because the treasury was no longer able to pay the full sum to
the whole army as was the old custom. Thence forward the authorities
paid I00oo dinars to those soldiers only whom they were afraid of, whereas
they paid as much as they liked to those soldiers whom they considered
as harmless (ammd tafriqat al-mi'a wa-aqallfa-hddhd shay' tajaddada min al-
Ashraf Ayndl wdlidihi li-'aj al-khizdna 'an at-taswiya bayna al-jam!' wa-illd
fal-'dda al-qadima taswiyat al-kull f mi'at dindr ash-shari/ wa4-da'iJffa-baqiyat
al-'dda al-dna man khdfi~ ghd'ilatahu a'.t/zhu al-'dda al-qadima wa-man istad'afz
jdnibahu a't/ihu md arddi).6) From the reign of AynMl onwards, the pay of
i) The short reign of the sultans' sons in the Circassian period and its con-
sequences were discussed in"The Circassians in the Mamluk Kingdom",JAOS,
1949, PP. 138-139 and in "Studies on the Structure of the Mamluk Army",
BSOAS, pp. 458-59. The average duration of the reign of the pure Mamluk sul-
tans during the Circassian rule was not very long either. This added to the finan-
cial difficulties caused by the payment of the nafaqat al-bay'a.
2) Ibn Iyas, II, p. 43, 11. 24-28.
3) Nujim, VI, p. 666. Ibn Iyis, V, p. 123, 11. 20-2 1.
4) Nujim (c), VIII, p. 26, 11. 1-4. Manhal, III, fol. 64b, 11. 6-8.
5) NAujim, V, p. 528, 11. 1-2; VI, p. 55, 1. 2; p. z121, 11. 19-21; p. 228, 11. 2-6;
p. 253, 11. 9-23; pp. 480-481; p. 496, 11. 2-7. Ibn al-Furit, IX, p. 371, 11. I 1-1 5.
For additional data on the nafaqa see Ibn al-Furit, IX, p. 163, 11. 2-4. Nujlm,
V, p. 4Io, 11. I-2; VI, p. 53, 1. 6.
6) Nujzth, VII, p. 6 o, in the notes.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
S8 D. AYALON
a nafaqa of 5o, 40 or even 3

though the old custom of pa
means extinguished.l) As w
resorted to the expedient of
to make up for part of the
sometimes added 7 dinars, th
however, all these advances w
the mamluk received 130 d
exceptions which prove the r
The curtailing of payments,
practice in the Mamluk sultan
widows, the ladies (khawandd
(al-fuqahd') and the civilian o
of the pen": arbdb al-qalam
non-mamluk units of the arm
was made to ensure full pay
these efforts were successful,
to suffer. Payments were not
Even the Royal Mamluks wer
fected were the mamluks of
qaradns and third the sayfiy
the Royal Mamluks received a
another - 25, and yet anothe
received the full pay, anoth
the full sum;julbdn - o, qar

nia and sayftya - nothing. Th
I) Nujim, VI, p. I84, 11. 13-I4.

I>3, i. I9;p. z23, 11. 20-21; p. I2z
z) See the references in the next
3) Ha.wddith,
11. I5-I7; p. 623,
(KM) III, p. 246, 11. p.4-7;
11. 9-io; 264,P.11.6z5, 11.p.2-3.
z-3; IV, 119, Ibn Iyis,
11. 6-9; II, p. 94,
p. 382, 11. 10o-I; p. o107,
11. 17-20; V, p. 26, 11. 4-6; P. 113, 1. 22; p. I23, 1. I9; p. 124, 11. 4-6.
4) Nujim, VII, p. 8J2, 11. 13-16. HIawddith, p. 678, 11. 7-15; PP. 68x, 1. zz22-
682, 1. 3; p. 682, 11. 12-14. Ibn Iyis (KM), III, p. 20, 11. 22-23; p. 21, 11. I-5, 7-9;
p. 31, 11. 13-t7; P. h 7e, 11. 12-14; P. 33, 11. 3-5 ; IV, p. 2t, 11. 6-d n.
6) See the references in the preceding note.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
PAYMENT IN MAMLUK MILITARY SOCIETY J 9
luks was not confined to the nafaqa, but included other kinds of pay:
julbdn - Io, qardnSsa - 5, sayftya - 3 dinars.1)

The curtailment of the pay of the halqa and of awldd an-nds started more
or less simultaneously with the general decline of these units in the
second half of the fourteenth century. Expressions like qa.ta'a nafaqat (or
jawdmik, luz.hi~m) awldd an-nds recur most frequently in the sources.2) In
matters of such curtailment the awlddan-nds were sometimes placed in the
same category with the most underprivileged elements, like the widows,
the orphans and "the people of the pen".3)
The Nafaqa of the Amirs

Whereas the private Mamluk soldiers received their nafaqa in a special
parade, the Mamluk amirs received it at home. The stereotyped expres-
sion in this connection is: "He sent (carried, etc.) the nafaqa to the amirs
as usual" (ba'atha, arsala, .hamala an-nafaqa ild al-umard' 'ald jdrf al-'dda).4)
It happened only once in Mamluk history - on the eve of the battle of
Marj Dibiq - that the Amirs of Forty and the Amirs of Ten (but not
the Amirs of a Thousand) received the nafaqa in a pay-parade.5)
Until the days of Sultan Barsbay (825/i422-842/1438) the nafaqa was
paid to the Royal Mamluks first and only afterwards to the amirs.
When Barsbiy reversed, in 836/1433, this time-honoured order and paid
1) HIawadith, pp. 168, 1. 19-169, 1. 3; pp. 174, 1. 16-175, 1. 2; pp. 25o-251'

Ibn Iyis, II, p. III, 11. 1-7; (KM) III, p. 241, 11. 7-io; pp. 256, 1. 11-257, 1. 2;
p. 329,11. I8-21;p. 383, 11. 8-ro;p. 435, 11. 19-21; IV, pp. 13, 1. 23-14, 1. 6;
pp. 242, 1. 18-243, 1. 3; P. 359, 11. 8-1o; p. 401, 11. 14-23; P. 404,11. II-I6;p. 43I,
11. 9-14; PP. 443, 1. 21-444, 1. 3; P. 444, 11. I8-20; p. 446, 11. 2-4; V, p. I2, 11.
9-12; p. 27, 11. 6-7. See also: Nujzm, VII, p. 450,11. 7-11. Ibn Iyis, II, p. 102, 11.
24-25; p. 103, 11. 4-5; p. 230; (KM) III, p. 214, 11. 21-22; p. 231, 11. I2-16; p. 258,
11. 18-21; IV, p. 281, 11. 8-13; p. 285, 11. 17-20; pp. 424, 1. 21-425, 1. I; V, p.
6I, 11. 8-15. On the discrimination against the qardn4s and sayflya in matters of
pay, see also: "Studies on the Structure of the Mamluk Army", BSOAS,
1954, PP. 86-88.
2) Nujim, VII, p. 450. Ibn Iyis, IV, p. 22; p. 25; pp. 65-66. See also the
references p. 58, note 4.
3) See the references p. 58, note 4.
4) Ibn Iyis, I1, p. 41, 11. 22-23; p. 67, 11. 7-9; P. 93, 1. 19; p. 103, 1. 24; P.
Io6, 1. 6; (KM) III, p. 25, 11. 3-8; p. 246, 1. i; p. 264,1. 4; p. 312, 11. 19-20; IV, p.
322, 11. 15-21; p. 384, 11. I2-16.
3) Ibn Iyis, V, p. 28, 11. 14-23.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
6o D. AYALON
the amirs first, he had

stresses that the sultan'
865/1461, however, pay
was considered the norm
end of Mamluk rule, it
times the amirs who rec
nafaqa was by no means
chronic financial difficu
the better chance of recei
The amirs had to cover
and their mamluks' expe
The nafaqa of the amirs
ordinary Royal Mamluks
and curtailments.
The amounts paid to th
were as follows: The Am
than 2,oo000 dinars, wh
offices in the realm rec
'asakir received up to 5,ooo
sometimes 700 dinars. T
I) Nujzm, VI, pp. 685-68

Q.di Shuhba,
2) Nujzim, VII, fol. 99a,
p. 694, 11. 7-9.
11. Io-13.
3) Ibn Iyas, II, p. Io03, 1. 24; p. 222, 11. 6-7; (KM) III, p. 22, 11. 5-Io; IV, p.
322, 11. 15-22; p. 384,11. iz-i6. Cf. also Nujim, VII, p. 648; p. 65o, 1. 12.
4) Nujzm, V, pp. 14-1S; VI, p. 228, 11. 2-6; p. 25 3, 11. 9-23; p. 496, 11. 2-7; p.
644; pp. 685, 1. 20-686, 1. 5; VII, pp. 648, 1. 15-649, 1. 3; P. 758, 11. z-4.
Harwddith, p. 5o6, 11. 12-16; p. 698, 11. 1i-12; p. 702, 11. 3-8. IbnIlys (KM), III,
p. 7, 11. 5-I8; p. 22, 11. 5-IO; V, p. 27, 11. Io, 22-28. In 813/1410o Sultan Faraj
paid the following nafaqa to the members of the expeditionary force which
marched against Sultan (then amir) al-Mu'ayyad Shaykh: to 2 of his highest
ranking Amirs of a Thousand - 6,oo000 dinars each; to the rest of the Amirs of a
Thousand - 2,000 dinars each; to each Amir of Forty - 500oo dinars; to each
Amir of Twenty - 300 dinars; to each Amir of Ten - 200 dinars; to each
Mamluk - ioo dinars (Ta'rikhQaytbay, B.M.MS., Or. 3028, fol. 39a, 1. ii -
39b, 1. 3). When Sultan Baybars I visited Alexandria in 66I/I'263 he gave each
of his amirs a nafaqa ranging between 3,000 and 100oo dinars, according to his rank
(Ibn 'Abd
Baybars az-.Zihir,
granted Sirat (about
200,000 dirhams arz-Zio,ooo000
dhir Baybars, fol.army
dinars) to the 68a,of 11.
the 13-1). In 662/1264

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
PAYMENT IN MAMLUK MILITARY SOCIETY 6I
During two short periods, however, the amirs, and especially

al-'asdkir, got much higher nafaqas. In the years 791-792/I 389
the grim fight over the throne which interrupted the rule of Sul
qfiq and which ended in the complete victory of the Circassia
in Mamluk military society, the atdbak al-'asdkir received zoo200,
hams and i o,ooo dinars (i.e. about zo20,000ooo dinars in all). Three
Thousand received half that sum each. Each Amir of Forty
dirhams and i,ooo dinars (i.e. about 3,35 oo00 dinars). Each Mam
dirhams (i.e. about 100oo dinars). Nothing is said about the naf
Amirs of Ten.1) During part of Sultan Q ytb y's rule (873/
I495) the nafaqa of an Amir of a Thousand was 6,oo00o dinars2) an
the atdbak al-'asdkir was up to 12,000 dinars3); sometimes it ev
30,000 dinars.4)
Pay to the Mamluks of the Amirs5)

The mamluks of the amirs received their pay directly from
ters.6) According to his feudal charter (manshzr), the amir was
one-third of the income of his fief, and two-thirds were to be di
among his mamluks. The amir or his clerks were not permitt
any part of the mamluk's portion without the latter's consen
probability this apportionment was not strictly observed. In
767/1365-66 it even became necessary to enact a law requiring
to share his income equally with his troops, and our source c
Syrian coast (al-'askar as-sdhili) (ibid., fol. 8Ia, 11. 3-y). This seems
that the size of that army was quite small.
i) Ibn Q .di Shuhba, fol. 36a, 11. 10o-15. Cf. also the following sources
which quote somewhat different figures: Nluj'm, V, p. 414, 11. i- 6; p. 477,11.
4-20; p. 488, 11. I 5-I7. Ibn al-Furit, IX, p. 142, 11. 24-26.
2) Ibn Iyis, (KM), III, p. 22, 11. J-I0.
3) Hlawadith, p. 687, 11. 7-8; p. 702, 11. 3-8.
4) Ibn Iyis (KM), III, p. 7, 11. 5-18. In addition to the nafaqa the amirs
received daily rations (raawdtib) of meat, spices, bread, fodder and oil. The most
important amongst them received a dress (kiswa) as well. In RamadSin the
sultan's favourite amirs received sugar and sweets and in the feast of sacrifice
they received sheep (Sub1, IV, p. y , 11. 7-Io; p. 6, 11. 17-18. Daw' as-Subh, p.
2J8, 11. I6-I19; p. 263,11. 7-9. Khi/at, II, p. 2zi6, 11. I6-I8, 27-29).
5) Because of the scantiness of our information on pay to the mamluks of the
amirs, the data on all their kinds of pay were gathered in the present section.
6) See, for example, Nquj'm, VII, p. 694, 11. I o-13.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
62 D. AYALON
this procedure proved m

titled to assume that be
the habit of taking for h
legally entitled to a third
luk received for any item
is known, for the sourc
on this subject.
It was only on very r
disorder, that the maml
sultan and then either to
One such occasion was in
throne) when the sulta
whole pay for themselve
luk of an Amir of Forty
tan a nafaqa of 400 dirha
the same occasion to a R
1438 when a i,ooo mam
demanded a nafaqa. Such
mamluks was "unknown
ment of the contemporar
The above lines clearly
mamluks' pay whatever
they not been checked b
the pay of the mamluks
I) Khita.t, II, p. 2I6, 11.

p. 318, 11. 5-17. "Studies
I953, PP. 459-460.
IX, 2)
pp. Nuji~m,
76, 1. 22-77,V,1. p. 4I1, the
3. From 1. i.sizeIbn Q.di
of the Shuhba,
nafaqa receivedfol. 38b,
by the 11.it isi-2.
amirs Ibn al-Furit,
made clear that their mamluks got a much smaller nafaqa than that of the Royal
Mamluks. Take for example the case of the Amir of Ten, who was usually al-
lotted zoo dinars. Even if that amir had been content with half that sum, the
second half would have been distributed between those of his mamluks who
accompanied him to the field of battle and the number of whom was some-
times more than five. The same conclusion is true of the nafaqa of the mamulks
of the amirs of higher categories, though these might have received a somewhat
bigger sum than that of the mamluks of the Amirs of Ten.
3) Nujtm, VII, p. 3 9 ', 11. 8- u .

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
a fight against its curtailment. Indeed, these mamluks did not cons
any political power to reckon with. Their subdued behaviour and
effacement before their masters stand in glaring contrast to the viril
offensive spirit which marks the attitude of the Royal Mamluks to
the sultan. No wonder, therefore, that the freedman of a Mamluk
had small scope for advancement and that he could never hope to
come a sultan himself.1)
When it was decided to cancel a campaign after having distribute
nafaqa, the mamluks had to return the money. This was usually
difficult procedure, for by that time most of the money would al
have been spent. The only way left was to deduct the nafaqa by i
ments from thejdmdkiya (the mamluks were exempted, however,
returning the sum they spent on buying a camel.2) The following
cident is instructive in this connection. Sultan Barsbly decided in
1433 to pay the nafaqa to the army in order to deceive Qari Yalak
Aq-Quyunlu chieftain and make him believe that the mamluk
determined to attack him. The nafaqa was distributed, however, on
the amirs, for it was considered much easier to get the money back
them than from the ordinary mamluks, as the contemporary sour
plicitly states.3)
The largest nafaqa was given to soldiers who took part in expedi
i) On the conflicts between the various units of the Royal Mamluks an

sultan, see BSOAS, I953, pp. zo6-zzz. Sultan Jaqmaq had to conceal i
youth the fact that he had been the freedman of an amir. Had this fac
discovered the only chance for him ever to join the Royal Mamluks would
been to become a sa.yft. As such he could never have been made sultan
D. Ayalon, L'Esclavage du Mamelouk, Jerusalem, 1951 , p. 2z). For isolated
of ferment amongst the mamluks of the amirs as a result of the curtailm
theirpay, seeNujtim, VII, p. 694, 11. 0o-13. Hawdidith, p. 606, 11. 12-20; p
11. 3-14; p. 666. Ibn IyIs (KM), III, p. 266, 11. 17-21. The amirs' mamluks
very often reason to complain of their masters' attitude towards them
usually they did not have the opportunity of doing so. The only sultan I
who made it a habit to hear their complaints and do them justice was Bay
wa-fi hddhibi as-sana (66z/1264) 'arada as-sultdn al-'asdkir wa-jalasa li-dhd
ku/ll kbamLis wa-ithnayn wa-lam yajri 'ard ahad hattd haqqaqa an-nazar wa-av
ishtakda
11. 2z-I4). min makhdimihi amara bi-inSdfihi minbu (Ibn 'Abd az-.Z hir, fol. 7Ia,
2) Ibn Iyus, IV, pp. 6I2, 1. 20-i22, 1. 8.
3) Nujim, VI, pp. 686-687 and especially p. 687, 11. 1-6.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
64 D. AYALON
marching to Syria and beyon

of the Mamluk sultanate thr
composed of the best units o
were far less important. The
fronts were smaller in size,
ferior units in them was much
lower. Nor is this all: even 6l
got a considerably lower nafa
part in a Syrian campaign.
Members of an expeditionar
Barsbay's rule) got 30 dinars
in the first expedition against
in the second (921/1515) -
919/1513 - 30 dinars.4)
The mamluks' contempt for
voyage was manifested in th
tions to Cyprus and Rhode
mitigated by the fact that th
as "holy wars' (jihdd) and we
828/1425 each of the gh&i<s
victorious mujdhidzn returne
In the campaign against Cyp
infrantryman or a horseman
mujdhidin... lil-fdris war-raj
'ashara dindran).7) Exactly th
866/1462.8)
The Nafaqa during the Campaign and on the Army's Return Home
The main pay in connection with a campaign was given to the mem-
I) Nujzm, VI, p. 685.

2) Ibn lyis, IV, p. 84.
3) Ibn Iyis, IV, p. 459, 11. I6-I8.
4) Ibn Iyis, IV, p. 32o, 11. 7-II.
) Nujzm, VI, p. 88, 11. 6-8.
6) Nujpm, VI, p. 594.
7) Nujzm, VII, p. 547.
8) HIawddith, p. 113, 11. 7-23; P 422. Nujum, VII, p. 218, 11. I0-20.

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00
bers of the military expedition on the eve of its departure. Usually this
was the only pay throughout the campaign. It did occur, however, that
the army received an additional nafaqa either during the campaign1) or
immediately on its return home.2) These payments were much lower
than the original nafaqa.
To be concluded
I) Ibn al-Furat, IX, p. 381,11. 15-17.
2) Nujum,
Shuhba, fol. 4ob, 11.VI,
21-24.p. 6i, 11. I2-13. Ibn al-Furlt, IX, p. 32, 11. i8-2o. Ibn Q.di

45.241.58.123 on Fri, 07 Jun 2024 12:33:15 +00:00

Ayalon-SystemPaymentMamluk-1957

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ayalon-SystemPaymentMamluk-1957

Uploaded by

Copyright:

Available Formats

The System of Payment in Mamluk Military Society

Author(s): David Ayalon

Stable URL: https://www.jstor.org/stable/3596038

This content downloaded from

THE SYSTEM OF PAYMENT IN

This content downloaded from

This content downloaded from

This content downloaded from

I) The numerous general statements and definitions contained in the Mam-

This content downloaded from

which was adopted here in order to establish the reliabi

From the above it should not be concluded that after

I) On the Mamluk currency see below: "Introductory Notes".

This content downloaded from

i. Mailuk Army Units, M

I) Daw' as-Subh, p. 243, 11. I1-12.

This content downloaded from

but were all concentrated in the capital, with a large section

This content downloaded from

Their factions were se

This content downloaded from

below the political weakness and the insignificance of the

military society the .halqa was bound to suffer heavily. Th

centuries and which resulted in drastic curtailments of the .

into the Mamluk upper class and were therefore relegated t

This content downloaded from

(a) The highest rank was that

or amir .tablkhdna). Each amir o

This content downloaded from

and the heads of each of the twelve barracks (muqaddamii at-t

2. The Ratio between the Dirham and the Dinar

In 805/I402 the dinar was equivalent to 60 -65 dirhams ;6) in 807/1404

This content downloaded from

changed between 230 and 285

The List of Payments to the Ma

(d) the la.hm - the daily meat ration

5) For an illuminating description of the adulteration of the currency under

This content downloaded from

(e) the aad.hiya - sheep distributed to the mamluks on

(f) the 'aliq - the fodder ration (distributed twice a week)

Pay in the Ba.hri Period

This content downloaded from

jaysh nafaqa md sumi'a bi-mithlihd qa.t.t kdinayju't al-juzndya khamsbna dindran)

Pay in the Circassian Period

I) Duwalal-Isldm, II, p. 15 8, 11. I3-I8.

4) Nuj]m (c), VII, pp. 197-198.

This content downloaded from

This content downloaded from

citadel.1) When Sultan

This content downloaded from

following or a similar expression: wa-yastad'ihim wd.hidan ba'da wd.hidin

7) I.awddith, pp. 619, 1. 23-620, 1. 2.

(ka-md kdnat al-'dda al-qadim

This content downloaded from

dinar was equivalent to 28 dirhams, but their demand w

The jdmakJya of the members of the .halqa and of the awldd a

I) Nujtm, VII, p. 474. HIawddith, p. 223.

This content downloaded from

monthly pay received by the mamluk of the Bahri period.

(b) The Nafaqa

This content downloaded from

This content downloaded from

a nafaqa of 5o, 40 or even 3

the full sum;julbdn - o, qar