The Mongols, the Turks and the Muslim Polity
Author(s): Bernard Lewis
Source: Transactions of the Royal Historical Society, Vol. 18 (1968), pp. 49-68
Published by: Cambridge University Press on behalf of the Royal Historical Society
Stable URL: http://www.jstor.org/stable/3678955
Accessed: 11-08-2016 03:52 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
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.
Cambridge University Press, Royal Historical Society are collaborating with JSTOR to digitize,
preserve and extend access to Transactions of the Royal Historical Society
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
THE MONGOLS, THE TURKS AND
THE MUSLIM POLITY
By Professor Bernard Lewis, Ph.D., F.B.A
READ II MARCH 1967
TEN years ago a well-known Swiss writer o
affairs published an article on patriotism a
Tamong the Arabs. Discussing the attitude
to the past, and their tendency to substitute fanci
for serious history, he quotes 'a high Syrian go
as saying, 'in deadly earnest': 'If the Mongols h
libraries of Baghdad in the i3th century, we A
had so much science, that we would long since
atomic bomb. The plundering of Baghdad
centuries.'1
This is of course an extreme, even a grotesque formulation, but
the thesis which it embodies is not confined to, and was not
invented by, romantic nationalist historians. Deriving ultimately
from the testimony of contemporary sufferers, it was developed
by European orientalists, who saw in the Mongol invasions the
final catastrophe which overwhelmed and ended the great Muslim
civilization of the Middle Ages. As the barbarians had destroyed
the Roman Empire, it was thought, so the Mongols destroyed
the Caliphate-except that the destruction was more terrible
and more permanent, and the new masters, unlike the Germanic
barbarians in Europe, could neither learn from others, nor themselves
create anything new. This judgment of the Mongols,
sometimes extended to include the Turkish invaders who had
preceded them out of the steppe, was generally accepted among
European scholars, and was gratefully, if sometimes surreptitiously,
borrowed by romantic and apologetic historians in Middle
1 Arnold Hottinger, 'Patriotismus und Nationalismus bei den Arabern',
Neue Ziiricher Zeitung, 12 May 1957. On modern Muslim views of the
Mongol invasions see further W. Cantwell Smith, Islam in Modern History
(Princeton, N.J., 1957), pp. 32 ff., 164 ff.; G. E. von Grunebaum, Modern
Islam: the Search for Cultural Identity (Berkeley and Los Angeles, 1962),
PP. 44 ff., I85, 213, 255-6.
49
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
50 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
Eastern countries as an explanation both of the ending of
golden age, and of their recent backwardness. It was expr
with characteristic force by the famous English orientalist
Granville Browne, who saw in the Mongol invasion 'a
trophe which... changed the face of the world, set in m
forces which are still effective, and inflicted more suffering o
human race than any other event in the world's history of
records are preserved to us."
To Browne, writing in Cambridge in the early years of
century, it may well have seemed that the Mongol conque
a calamity of unparalleled magnitude, and that a civilizati
stricken could never fully recover. But for the less in
historians of a less tranquil age, the horrors of the past as
a milder aspect. The great Russian orientalist V. V. Ba
writing in Moscow in 1917, was able to achieve a more tol
view of Mongol destructiveness, and a more robust assess
of the recuperative powers of their victims. 'It would be a mis
however, to consider that cultural life could only continue i
localities which had escaped the inroads of the Mongol tro
It is true that a cultured land had been conquered by a wild
still believing in the efficacy of human sacrifice. When a town
taken, except for the artisans who were needed by the conq
the inhabitants were sometimes subjected to total mas
People, who had survived these horrible experiences, natu
thought that the country will [sic] not arise again for an
thousand years. Influenced by the opinion of writers conte
ary to that epoch, European scholars have believed th
Mongols dealt a heavier and more devastating blow to the c
life of Asia and Eastern Europe than, for example, was dealt
cultural life of Southern Europe by the Great Migrat
Peoples. In reality, the results of the Mongol invasion we
annihilating than is supposed. ... Besides a not numerous m
contingent the Mongol Khans brought with them their cu
councillors [sic] who helped them to establish their rule a
1 E. G. Browne, A Literary History of Persia from Firdawsi t
(London, 1906), pp. 426-27; cf. idem., A History of Persian Literatur
Tartar Domination (Cambridge, 1920), pp. 14-15. Like most other W
writers, Browne bases his account of the Mongols largely on Ba
d'Ohsson's Histoire des Mongols, Ist ed. 1824, 2nd considerably am
ed., The Hague and Amsterdam, 1834-35.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
MONGOLS, TURKS AND THE MUSLIM POLITY 5I
apply to the new country that harmonious and well-constructed
governmental and military organization which had been elaborated
at the time of Chenghiz Khan himself.'"
Since then, a more intimate experience of catastrophe on the
one hand, and a deeper knowledge of Islamic history on the other,
have confirmed some-though not all-of Bartold's insights.
In our own time we have seen, in the heart of Europe, rulers and
armies compared with whom the Mongol Khans and the Tatar
hordes appear almost as angels of mercy-and we have seen the
swift recovery of the lands they ravaged. Not all scholars would
now fully accept Bartold's views on the benevolent and progressive
character of Mongol rule. They were well received in
Mongol, Tatar and Turkish circles; others however have suggested
some revisions, and among Russian scholars in particular
there has in recent years been a sharp reaction against them, and
against what is called the 'racialist-nationalist idealization of the
Turco-Tatar nomads' by pan-Turkist writers. Professor I. P.
Petrushevsky has formally declared that Bartold's evaluation of
the consequences of the Mongol invasions for the economic development
of Iran and neighbouring countries 'cannot be accepted
by Soviet historiography'.2 But even Professor Petrushevsky
1 V. V. Bartold, Mussulmnan Culture, translated from the Russian by
Shahid Suhrawardy (Calcutta, 1934), pp. 11o-x2; cf. the very much better
Turkish translation edited by M. Fuad K6priilii, Islam medeniyeti tarihi,
2nd ed. (Ankara, 1963), p. 62. Bartold's views on the Mongol invasions and
their effects are developed in many of his writings. In attempting a more
positive assessment of the Mongols, he was to some extent anticipated by Sir
Henry Howorth (History of the Mongols, London, 1876-88) and, still more,
by L6on Cahun (Introduction a l'histoire de l'Asie, Paris, 1896). These works
were, however, written without reference to oriental sources, and are of no
scholarly significance. Cahun's book, written with some skill and much
enthusiasm, became a source of inspiration for Turkish and pan-Turkish
nationalist theories.
2 I. P. Petrushevsky, Zemledelie i agrarniye otnosheniya v Irane xiii-xiv
vekov (Moscow-Leningrad, 1960), p. 36; Persian translation by Karim
Kashivarz, KashavarJ va mundsabdt-i art! dar Iran 'aahd-i Moghlzl, i (Tehran,
1344 solar), p. 48. This statement is obviously prescriptive, not descriptive,
and, like other such decisions recorded on behalf of Soviet historiography,
may not be determined exclusively by the findings of historians and the
evidence of the sources. A clue may be found in hostile allusions, without
citation of authors or titles, to 'pan-Turkists', i.e. those who ascribe a common
identity and purpose to the Turkic peoples inside and outside the
U.S.S.R. Bartold is declared innocent of complicity in such villainy. His
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
52 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
affirms that the processes of development, interrupted by the
catastrophe of the 13th century, were resumed and completed in
the 14th, and concedes that Bartold had some reason to react
against the one-sided presentation of the Mongols as destructive
savages. Most scholars would now agree that the harmful
effects of the Mongol conquests were not as great, as lasting,
or even as extensive, as was once thought.
The reconsideration of the impact of the Mongols on the Islamic
world has been concerned with three periods, before, during and
after their irruption, and with three questions; what did they
destroy, what did they achieve, and what did they leave behind
them? The traditional answer to the first of these questions is
that they destroyed the Caliphate, and with it the great ArabicIslamic
civilization that had flourished under its aegis. 'Islam',
says a contemporary Syrian historian, 'has never suffered a greater
and more decisive disaster than this',' and other historians, of
that and later ages, have shared this opinion. The destruction of
the Caliphate, still, even in its decay, the legal centre of Islam
and the symbol of its unity, and the establishment of a heathen
domination in the Islamic heartlands, were indeed a bitter blow
to Muslims, and it is not surprising that their anguish has echoed
through the centuries. But the real significance of this act of
destruction has been much exaggerated. The golden age of classical
Islamic civilization had long since ended, and the Mongols
conquered a society that was far advanced in decay. The Caliphs
had lost most of their effective power, and by abolishing the
Baghdad Caliphate the Mongols did little more than lay the ghost
of something that was already dead. Even some modern nationalist
historians, their perceptions sharpened by more recent reverses,
have begun to appreciate this. 'Some of us still believe', says
Professor Constantine Zurayk, 'that the attacks of the Turks and
the Mongols are what destroyed the Abbasid Caliphate and Arab
errors are attributed to his lack of Marxist discipline, not to sinister panTurkist
motives. Cf. Professor Petrushevsky's introduction to the new
edition of Bartold's collected works (Sochineniya, i (Moscow, 1963), especially
pp. 32-3)x
Ibn Wasil, Mufarry al-kuruib, MS. Paris, Arabe 1703, fol. I26 b, cit.
D. Ayalon, 'Studies on the transfer of the 'Abbasid Caliphate from Baghdad
to Cairo', Arabica, vii (I96o), p. 59.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
MONGOLS, TURKS AND THE MUSLIM POLITY 53
power in general. But ... the fact is that the Arabs had be
defeated internally before the Mongols defeated them ....'a
In another respect, too, the effects of the Mongol conquest
have been exaggerated-in the extent and consequences of t
material damage done by them. Certainly, the damage was gre
The immediate blows of the Mongols, though no doubt trivial
modern standards, were terrible and overwhelming. Even
Persian historian Juvayni, a servant and admirer of the Khan
who sees in them the instruments of God's purpose, tells how
they destroyed whole cities and massacred or deported th
inhabitants. Their ravages were not confined to the cities;
many areas the extirpation of the military aristocracy, the de
or flight of the peasantry, left vast lands untenanted, uncultivate
and unclaimed, often permanently abandoned to nomadic her
men. Recent studies have shown that the damage done by the
Mongols to the economy of Persia was not limited to the actu
destruction during the campaigns of conquest. The ill effects
depopulation and the neglect of irrigation works were aggrava
by harsh and extortionate policies, which degraded and im
poverished the peasants, and set back the development of agr
culture and of the rising feudal society of the immediately pr
ceding period.2
Yet these effects, however terrible, were limited both in ext
and duration. Egypt, which by this time had become and
ever since remained the chief centre of Arab Islam, was neve
conquered by the Mongols, and was thus only indirectly affec
by their coming. Syria suffered only raids, and after the defe
of the Mongols by the Mamluk army of Egypt at 'Ayn Jfliit
Palestine in i26o, was incorporated in the Egyptian Sultanate a
protected by Mamluk power from Mongol attack. Arab Afric
was never invaded; Turkish Anatolia was long dominated by t
Mongol state in Persia, but suffered little direct interference,
survived to cradle the last and greatest of the Islamic Empire
1 Constantine K. Zurayk, The Meaning of the Disaster, trans. R.
Winder (Beirut, 1956), p. 48; cit. G. E. von Grunebaum, Modern Isla
p. 255.
2 See Ann K. S. Lambton, Landlord and Peasant in Persia (London
1953), p. 77 ff., Petrushevsky, op. cit., and, on the Mongol Empire in gener
J. J. Saunders, 'Le nomade comme bitisseur d'empire: conquete arabe
conquete mongole', Diogine, no. 52 (i965), pp. 85-109, where other rec
literature is cited.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
54 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
that of the Ottomans. Persia, indeed, was hard hit-but even
here by no means the whole country was affected. In South
Persia, the local dynasties submitted voluntarily to the Mongols,
and their cities, not looted by the invaders, continued to flourish.
Even in those parts of Persia which were actually overrun and
devastated, there was some recovery, and before long some Persian
cities were again centres of industry, trade and culture.
Only in one country did the Mongol conquest leave permanent
injuries-in Iraq, once the metropolitan province of the Caliphate.
Here, as elsewhere, the immediate effect of the invasion was the
breakdown of civil government; in Iraq this also meant the
decline of the elaborate irrigation work on which the prosperity,
even the life of the country depended. But whereas in Persia
there was a partial recovery once the new regime was firmly in
control, in Iraq there was hardly any.' The Mongol I1-Khans of
S. W. Asia, like the Seljuks before them, made Persia, not Iraq,
the centre of their power; Tabriz, their residence, grew into a great
and wealthy city. Even before the Mongol conquests, Iraq had
lost much of her importance; the coming of the Mongols, the
destruction of the Caliphate, and the emergence of new centres,
finally ended it. The Mongols conquered Persia and Iraq, but
failed to conquer Syria and Egypt; these, under the Mamluk
Sultans, formed the base of the most important Muslim military
power of the day, and the most dangerous adversary of the I1Khans.
Iraq now became an outlying frontier-province, abandoned
to the destructive inroads of the Bedouin, who moved into
the breaches made by the Mongols and, unlike them, did not pass
on, but stayed. The valley of the Tigris and Euphrates was cut off
from the Mediterranean lands by the Mongol-Mamluk conflict;
it was overshadowed by the rise of the new Persian centre to
which it was subordinated, and outflanked by the flourishing
Turkish states in Anatolia, now under Mongol suzerainty. Iraq
could no longer serve as a channel for east-west trade, which now
passed through two other, competing routes-the Mongol
northern route, through Anatolia and Persia, and the Mamluk
southern route, through Egypt and the Red Sea. Bereft of the
1 Even in Iraq, however, the extent of the economic damage done by the
Mongols has been exaggerated. See the important study by Dr Ja'far H.
Khesbak, 'Ahwal al-'Iraq al-iqtisadiyya fi 'ahd al-Ilkhlninyin al-Mughil',
Majallat Kulliyyat al-,Jddb (Baghdad), (1961), pp. 1-56.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
MONGOLS, TURKS AND THE MUSLIM POLITY ?
Caliphate and ruled by a Mongol governor, Baghdad could no
longer be the centre and the rallying point of Islam. This role
passed to Cairo and later to Constantinople, leaving the fallen
city of the Caliphs to centuries of stagnation and neglect.
The dethronement of Iraq and the partial devastation of
Persia are the significant exceptions in the general picture of
gradual recovery and renewed activity in the Muslim Middle
East. Clearly, such a revival could not have been accomplished
under the heel of destructive and unteachable savages, such as the
Mongols of the conventional image. The opposite extreme is
expressed, in a lyrical passage, by the Polish Altaist Wladyslaw
Kotwicz,
Dans leur empire, [he says] les Mongols firent rigner l'ordre
et le droit, organiserent une administration uniforme, entreprirent
l'oeuvre de reconstruction des pays en ruines, de
relkvement de l'industrie et du commerce, d6veloppbrent des
rapports culturels avec les territoires les plus recules de l'ancien
monde.
Leur autorit6 6nergique fit effectivement rigner, sur la plus
grande surface de cet ancien monde, une vraie Pax mongolica.'
Even allowing for the natural affection of a Mongolist for the
Mongols, and for the revisions imposed by more recent research,
there is a considerable element of truth in this picture. Once the
conquests, with their attendant horrors, were completed, the
Mongols were quick to appreciate the advantages of peace and
order, and the pax mongolica became a reality in their vast
dominions.
Some beneficial effects of Mongol rule in Persia are discernible
almost immediately. Once firmly established, the Khans brought
a measure of security and stability. In contrast to barbarian
Europe, there was no permanent reversion from a money economy
to barter, from an urban to a rural way of life. The merchants
raised their heads again, and the I1-Khans, for their own good
reasons, gave them every encouragement. Their interest was
more than that of the greedy savage who has learnt to tend instead
of killing his dairy cattle-though even to learn that, in
so short a time, would have been no small achievement. The
1 Wladyslaw Kotwicz, 'Les Mongols, promoteurs de l'id6e de paix
universelle au d6but du xiiie siecle', Rocjnik Orientalistyc~ny (Cracow),
xvi (1950), p. 429.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
56 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
Il-Khans gave active help to what they regarded as useful sc
such as medicine, astronomy and mathematics; after thei
version to Islam at the end of the I3th century, they ext
their patronage to Islamic learning, and by the 14th c
Muslim Khans were raising magnificent edifices for I
worship and scholarship. In one respect the Mongol con
actually brought some advantage to the lands of Islam-th
the broadening of Muslim horizons. The Mongol world wa
limited to the familiar Muslim lands of the Middle East and Ce
Asia. It included southern Russia and, most important of a
Far East, with which Muslim Western Asia was now unite
the first time in a single imperial system. In this way Per
opened to Chinese influence, notably in art and technolog
Mongols also exposed the Muslims to other contacts, as Eur
seized the opportunity offered by their presence to explo
land and sea routes through Persia to China and Indi
benefits of these journeys, it may be noted in passing, are
apparent in Europe than in the Middle East. A good exam
the wide outlook and interests of the Mongol era in Persia
Jdmi' al-tavdrikh, the Assembly of Histories, by the P
historian Rashid al-Din (1247-13 I8). Rashid al-Din was a J
convert to Islam, a physician, scholar and minister, w
entrusted by the I1-Khans with the task of preparing a un
history. He assembled a team of collaborators and inform
including two Chinese scholars, a Buddhist hermit from Ka
a Mongol specialist on tribal tradition, and a Frankish mo
well as a number of Persian scholars, and with their help
posed a vast history of the world, from Ireland to China.
attempting a universal history, going beyond the conf
their own civilization and its accepted precursors, Rashid
and his colleagues anticipated European historical scholars
half a millennium.
The Mongols, then, though they ravaged some of the lan
Islam and abolished the Baghdad Caliphate, did not de
Islamic civilization, which was far advanced in the decline
they came and which, in new forms, rose again after their com
But their advent marked the turning-point in a process of
which, in the course of time, transformed the whole patte
society and government in the Middle East. The Mongols
relatively few in number; their direct rule in the Middle E
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
MONGOLS, TURKS AND THE MUSLIM POLITY 57
limited to the northern tier, and to a brief period. They bequeath
neither a language nor a religion to the lands they conquered
and whether their dominion was 'historically progressive'
'historically reactionary', as Russian scholars of successive gen
ations have argued, its effects in either direction were exhaus
within a century. Yet the historical instincts of those who, fr
contemporaries onwards, saw in the Mongol conquests the en
of one era and the beginning of another, were fundamentall
sound; their error was the common one of telescoping a long
complex evolution into a single dramatic event. The great chan
in medieval Islam cannot be understood only in terms of
brief episode of Mongol conquest and domination; it must
seen against a broader background, involving a longer per
than the reign of the Khans, and the movement of more numerou
peoples than the Mongol tribesmen of Jenghiz and his heirs.
Professor Zurayk, it will be recalled, links the Turks with t
Mongols as the invaders of the collapsing Arab Caliphate.
association is not new. 'It is a remarkable thing', says a I3
century Damascene chronicler, discussing the defeat of t
Mongols by the Mamluks at 'Ayn Jdlfit, 'that the Tatars wer
defeated and destroyed by men of their own kind, who w
Turks." Rashid al-Din also links the two together. The Assem
of Histories begins, as one would expect, with 'the present mas
of the world'. Volume I is in two parts, the first dealing with
steppe peoples in general, the second with Jenghiz Khan and
successors. The first, concerned with the divisions, genealogi
and legends of the tribes of the steppe, includes Turks as well
Mongols, and in time became a source-book for Turkish hero
and historiographic myths. Even a Turkish tribal origin-myt
as Professor Hatto has remarked, is 'fused with a wishful trave
of the saga of the more dazzling Mongols ... at the poetic lev
of myth and folk-tale.'2
The Turks and the Mongols were ethnically, culturally a
linguistically distinct; yet they had much in common. Both ca
x Abu Shama, Tarajim rzjdl al-qarnayn al-sddis wa'l-sabi', ed. Muhamm
al-Kawthari, Cairo, 1947, p. 208.
2 A. T. Hatto, 'lHamdsa iv', in Encyclopaedia oflslam, revised ed., iii, I
The whole problem of Turkish-Mongol relationships is discussed in
important article by Professor Ibrahim Kafesoilu, 'Ttirk tarihinde Mogo
ve Cengiz meselesi', Tarih Dergisi, v (1953), pp. 105-36.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
58 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
into the Middle East from the steppe-lands of Central and N. E.
Asia, where they shared a common way of life and were subject
to similar influences. These affinities brought the Mongols closer
to the Turks than to any of the other peoples they had conquered.
Jenghiz Khan himself made use of Uygur Turkish advisers, and
ordered the adoption of the Uygur script for the Mongol language.
The Mongols, few in number, leaned heavily on Turkish support
in both war and government. In time, the Mongols in the Islamic
lands were merged into the mass of their Muslim subjects, and
even lost their language, adopting various forms of Turkish in
its place. The very name Tatar, once that of a section of the
Mongols, has for a long time been applied to the Turkish-speaking
Muslim inhabitants of the territories that were once ruled by
the Mongol Khans of the Golden Horde.
The great migrations of the steppe peoples into the Middle
East began in the ioth century, when the Turkish tribes of
Central Asia crossed the Jaxartes and began their march of conquest
westwards. They ended in the period after the death
of Timur or Tamerlane, the last of the great Turkish worldconquerors,
in 1405. Even then, the trickle of Central Asian
tribes continued for a while, until it was stopped by the double
barrier of Safavid and Ottoman power on the plateaux of Iran
and Anatolia.
In the establishment of Turkish power and the spread of
Turkish customs over the lands of Islam, two periods are particularly
significant. One was that of the Seljuk Great Sultans, who
ruled for about a century from the conquest of Baghdad in Ioy55
to the death of Sultan Sanjar in I157. The other was that of the
Mongol conquests of the 13th century and the period of Mongol
supremacy and influence that followed them.
The Seljuks had entered Islam as condottieri, and had served
various Muslim rulers, including the Ghaznavids, before they
carved out an independent state of their own. They were devout
and earnest Muslims, and, as a great Russian historian has remarked,
'it is quite natural that the first Saljuqids... were better
Muslims than [the Ghaznavids] Mahmid and Mas'id, just as
Saint Vladimir was a better Christian than the Byzantine Emperors.'x
They were also free Turks, with their roots in Central
1 W. Barthold, Turkestan down to the Mongol invasion (London, 1928),
p. 305.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
MONGOLS, TURKS AND THE MUSLIM POLITY 59
Asia, and with memories both of the older Turkish kingdoms and
of the tribal traditions of the Oghuz. We can point to many
Turkish elements among the titles, ranks and emblems of the
Seljuk Sultanate; we can also see the first phases of a profound
transformation of Islamic state and society, part of which must
surely be attributed to the incursion of the steppe peoples.
The transformation is completed with the second and greater
of the steppe Empires-that of the Mongols. Their rule, though
of brief duration, was of great significance, for it was at this time
that the main characteristics of the post-Mongol phase of Islamic
government were formed. The first Mongol rulers of Persia were
pagans-the first to rule over an important Islamic territory since
the beginnings of Islam. Their system of government was
avowedly non-Islamic-based on the so-called Yasa of Jenghiz
Khan. This seems to have been a codification of Mongol rules
and customs; it was held to be binding on the Khans themselves
as well as on their subjects, both Mongols and others. Even after
the conversion of the I1-Khans of Persia to Islam, the Mongol
code remained effective, and Mongol practices were only gradually
and partially modified under the influence of Muslim administrative
and legal traditions.
The Mongol influence was of course strongest in those areas
where the rule of the Mongol Khans persisted-in Central Asia,
in Persia and in the territories of the Golden Horde in Russia. It
was, however, by no means limited to these areas. The SyroEgyptian
Empire of the Mamluks, though it escaped Mongol
conquest, was profoundly influenced by the Mongol example and
by the Mongol deserters and refugees who migrated to Egypt.
During the i3th and 14th centuries the Mongols enjoyed the
immense prestige of victory and conquest; they were in consequence
imitated in warfare, even in dress-as Europe was imitated
in the i9th and 2oth centuries. The Mamluk emir of I3thcentury
Egypt wore his Tatar coat and hat in much the same way
and for much the same reasons as his modern equivalents wear
fitted tunics and peaked caps. Both are alien to Islam-but both
were the symbols of real power.
Far stronger than in Egypt was the Mongol influence in
Turkish Anatolia. After conquering Persia, the Mongol horsemen
had swept on to Mesopotamia and Anatolia, where they had dealt
the Seljuk Sultanate of Rfim a blow from which it never recovered.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
60 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
After dragging out an attenuated existence for some fifty years,
it finally disappeared at the beginning of the 14th century. Most
of eastern and central Anatolia became subject to the I1-Khans of
Persia and was ruled either by Mongol governors or by Turkish
vassals. Even after the decline of I1-Khan power had permitted
the development of local autonomies, the administrative and financial
system which the Il-Khans had impressed on the country
continued to function. It was still working under the Anatolian
princes and survived to exercise a formative influence on the
institutions of the Ottoman state.
After the death of the Il-Khan Abfi Sa'id in 1336, the Mongol
dominions in the Middle East broke up, and a number of smaller
states, ruled by Mongol or Turkish dynasties, appeared in Persia,
Mesopotamia and Anatolia. Those of Persia were of short duration.
Farther east, Timur had succeeded in making himself ruler
of the Mongol successor state in Central Asia. In 1380, already
master of Transoxania, he invaded Persia and in the next seven
years overran the whole of it. He twice defeated the Khan of the
Golden Horde, raided India, annexed Iraq and then overran Syria
and exacted homage from the Mamluk Sultan. In 1394 and 1400
he invaded Anatolia, and in 1402 inflicted a crushing defeat on the
Ottomans at the battle of Ankara, capturing the Ottoman Sultan
Bayezid. He died in 1405 while preparing an invasion of
China.
Timur was a Turk and a Muslim-but he was proud to relate
himself to the Mongol Imperial house by marrying a princess of
the line of Jenghiz Khan. He led mixed Mongol and Turkish
armies, in which the former were the dominant element but the
latter the great majority. His career has been variously represented
as a reaction of Islam against the Shamanism of the Mongol
Khans and as the last convulsion of the Altaic invasion. Unlike
the Khans of the earlier conquests Timur was, or claimed to be, a
pious Muslim, and amid the enormous destruction he wrought he
was careful to show deference to the places and personnel of the
Islamic faith. But his system of government was still in the Mongol
tradition, and despite the noticeable Islamizing tendency, his
codes of law are true to the spirit of the Yasa of Jenghiz Khan.
With his death, the great movement of the steppe peoples that
had begun in the Ioth century seems to have come to an endthough
the infiltration of tribes continued and, what was more
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
MONGOLS, TURKS AND THE MUSLIM POLITY 6I
important, the penetration of nomads already in the Middle East
into the structure of urban life and civilization.
Of the great changes that can be discerned in Islamic government
and society during and after the invasions of the steppe
peoples, how much can be attributed to the influence of the
invaders? The question is by no means easy to answer. The Turks,
after their conversion to Islam, had surrendered themselves to
their new religion almost completely. Partly because of the
simple intensity of the faith as they encountered it on the frontiers
of Islam and heathendom, partly perhaps because their conversion
to Islam at once involved them in Holy War against their own
unconverted kinsmen beyond the borders, the Muslim Turks
sank their national identity in Islam as the Arabs and Persians
had never done.
Yet something of the Turkish past survived. The Turkish
language, brought from Central Asia by the first migrants and
invaders, lived on and emerged triumphant in a new Muslim
dress. Turkish rulers, even in lands of old Islamic traditions, used
titles and symbols of authority that go back to pre-Islamic
Turkish antiquity. Even in the Ottoman Empire, the symbols of
the bow and the arrow and the horsetail remained to commemorate
the mounted archers from the steppe that had first crossed the
rivers from Central Asia into the lands of Islam; the Altaic titles
of Khan and Beg were used or conferred by a sovereign whose
roots of power led back to the Mongol Khans as well as to the
Sultans of Islam.
The persistence of these old Turkish titles and emblems, long
after the Islamization of the Turks, symbolizes the survival, at a
deeper level, of habits, practices, and beliefs inherited from an
earlier age. The identification and evaluation of these survivals
is however a task of no small difficulty. The evolution of Islamic
and Persian notions and practices of government is well documented
and has been fairly well studied. Those of the Turks,
however, are still little known and have formed the subject of
some dubious theorizing.
The attempt has been made by some historians to explain the
whole structure of Ottoman administration in the Imperial age by
reference to the nomadic herdsmen who invaded Anatolia in the
i ith century-tather as at one time a school of historians in the
West tried to trace the British parliamentary system to the alleged
TRANS. 5TH S. - VOL. 18 - E
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
62 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
practices of primitive Germanic tribes. In avoiding fanciful and
exaggerated hypotheses on steppe origins, we should not however
fall into the opposite error of underrating them. Both th
Turkish and the Mongol dynasties that ruled over Islam during
the formative period between the i ith and I4th centuries were o
steppe origin, and even when they had been long assimilated in
the cities and river valleys of the Middle East, new waves of
nomadic invaders from the steppe were still breaking into the
lands of Islam and seeping into the apparatus of government
When the Mongol victories had brought a new aristocracy and
new law from the steppe, the Turks rediscovered their pride in
their ancestors and their ancestral way of life, and sought more
self-consciously after the emblems and prerogatives of a specifically
Turkish sovereignty.
The cultural and political baggage of the steppe peoples when
they entered the world of Islam was not limited to their own
native inheritance. They had for long been in contact with other
sedentary civilizations-for example, with the ancient, littleknown
but highly important Iranian cultures of Central Asia,
the influence of which can be traced through pre-historic Iranian
borrowings in the Turkic languages. Easier to observe, and more
relevant to our present inquiry, is the influence of China-clearly
visible on both the pre-Islamic Turks and the Mongols. It is from
Chinese sources that we first hear of the Turks, as a tributary
people among the barbarians beyond the N.W. frontier of the
Chinese Empire. The earliest Turkish records-the 8th-century
Orkhon inscriptions-reveal profound Chinese influence, and in
a sense express a kind of Turkish national revolt after a long
period of subjugation to China.
Several of the later Turkish tribes and peoples which entered
the Islamic world were still strongly affected by Chinese civilization.
Still more so were the Mongols and their kin. The first
important group of these to become known to the Muslims were
the Kara Khitay, who appeared on the North Eastern frontiers o
the Empire of the Great Seljuks. Of Mongol or Tunguz stock
they had conquered Northern China in the ioth century and
founded a Chinese dynasty. The name Cathay commemorate
their period of rule. In the early x2th century they were driven
out of China by another related people, and began to move westwards.
Towards the middle of the x2th century they conquered
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
MONGOLS, TURKS AND THE MUSLIM POLITY 63
Transoxania from the Karakhanids and set up a vast Empire
stretching from the Oxus to the Yenissei and the border of
China. The Seljuk Sultan Sanjar, trying in vain to stem their
advance, suffered a humiliating defeat at the battle of the Katvan
steppe in 1141.
This little-known engagement must rank among the decisive
battles of Asian history. In Persia, it accelerated the decline of
Seljuk power and the break-up of the Seljuk Great Sultanate into
a number of small states. In Central Asia it confirmed the domination,
over what was now old Muslim territory, of a dynasty of
far eastern origin, with Chinese Imperial experience. Their
language of government, we are told, was Chinese, and they introduced
many elements of the Chinese administrative and fiscal
system.
With the great Mongol conquests, Muslim South West Asia
passed under the control of a people of East Asian origin, dominated
since the childhood of their race by the vast majesty of
China. Jenghiz Khan himself leaned heavily on Chinese precedent
and advice; in his first expedition in 1219 across the Jaxartes into
the lands of Islam, he was accompanied by his Chinese counsellor,
Ye-lu Ch'u ts'ai, a high Chinese official and, incidentally a descendant
of the former Kara Khitay ruling house. By the time when
Jenghiz Khan's grandson, Hiilegii Khan, advanced across the
Oxus, on a new campaign of westward conquest, the Mongols
had conquered China itself-and the subjugated lands of Islam
were incorporated in an Empire that had its capital in Peking.
Far to the west, the Khan of the Golden Horde in South Russia
and the II-Khan in Persia were autonomous territorial rulers, but
they were subject to the supreme authority of the Great Khan, the
head of their family and overlord of their Empire. In time, the
Khanates of the West became independent and Islamic-but by
that time the oriental civilization of the united Mongol Empire had
profoundly affected them.
In the period following the destruction of the Caliphate, a
fundamental division becomes apparent in the Middle East,
between two great cultural zones. In the north was the zone of
Perso-Turkish civilization with its centre in the plateau of Iran,
extending westwards into Anatolia and beyond into the lands
conquered by the Ottomans in Europe, eastwards into Central
Asia and the new Muslim Empire of India. In these countries
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
64 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
Arabic survived only as the language of religion and the religious
sciences; culturally it was supplanted by Persian and Turkish,
which became the media of a new form of Islamic civilization.
To the south lay the countries where Arabic was spoken-the
derelict province of Iraq, and the new centre in Egypt, with its
Syrian and Arabian dependencies, and its African hinterland. Here,
behind the defences of a Mamluk Byzantium, the older Arabic
culture survived, and entered on its long-drawn-out Silver Age.
Persian was not known and, except in art and rather more in
architecture, the new cultural developments in the north had
little effect.
Politically, however, the Turk and the Mongol were everywhere
dominant. Mongol or Turkish dynasties ruled all the
countries from the Mediterranean to Central Asia and India, and
even the Syro-Egyptian Empire of the Mamluks was governed
and defended by a ruling class of imported slaves of Turkish
speech, mainly from the Kipchak country north of the Black
Sea.
In the I4th century the greatest of Arab historians, the Tunisian
Ibn Khaldfin, observed the almost universal supremacy of the
Turks, and saw in their coming a proof of God's continuing
concern for the welfare of Islam and the Muslims. At a time when
the Muslim Caliphate had become weak and degenerate, incapable
of resisting its enemies, God in His wisdom and benevolence had
brought new rulers and defenders, from among the great and
numerous tribes of the Turks, to revive the dying breath of
Islam and restore the unity of the Muslims. By the providential
dispensation of the Mamluk system, he affirms, they were constantly
reinforced by new importations from the steppe, who embraced
Islam with enthusiasm, yet retained their nomadic virtues
unspoilt by the corrupting influences of civilization.'
In this interpretation of events, Ibn Khaldiin is applying his
own well-known version of the myth of the noble savage. His
praise of the steppe peoples as the saviours of Muslim power is
however by no means without foundation. The military prowess
of the Turks and Mongols has never been questioned; their
1 Ibn Khaldfin, Kitdb al-'Ibar, v (Cairo, 1867), p. 371. Professor Ayalon
was the first to draw attention to this very important passage: 'The wafidiyya
in the Mamluk kingdom', Islamic Culture (1951), p. 90; cf. idem in Jewish
Observer, 23 November, 1956, p. x9.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
MONGOLS, TURKS AND THE MUSLIM POLITY 65
political contribution to the recovery, stability, and, for a while,
expansion of the Muslim world deserves more attention.
It is perhaps in the forms and functioning of government that
the great transformation wrought by the invaders from the steppe
can most clearly be seen. In the Mongol kingdom of the II-Khans,
and in the states which followed it in Persia and Anatolia, new
patterns appear that differ sharply from those of the old Caliphate.
The extent of the change may be measured against those countries
which knew Turkish but not Mongol rule-as Egypt and India,
and those that knew neither-as Morocco.
The first and most striking feature of the new era is the reinforcement
of political power. The states of the post-Mongol
era are stronger, more stable, and more enduring than those of
the past-and the states of the plateaux of Anatolia and Iran are
stronger than those of the countries less directly affected by
Turco-Mongol rule. In the six centuries before the Mongol
invasions, few states in Islam had lasted for much more than
three or four generations. The patriarchal Caliphate had perished
within forty years of the Hijra; the kingdom of the Umayyads had
lasted for less than a century; even the Abbasid Caliphs, though
they reigned in name for five centuries, wielded effective power
for little more than the first of them, and were thereafter forced
to yield it to an unending series of dynasties, some of them great
and powerful, but all of them ephemeral in the form, the extension
and the duration of their dominion. Even in the periods of their
greatness the authority they wielded, though vast, was fragile.
Institutions, regimes, realms-all were shifting and impermanent,
liable to sudden and total upheaval.
In the Turco-Mongol age all this is changed. In Egypt, the
Mamluks, recruited for the most part from the Khanate of the
Golden Horde and deeply influenced by the statecraft of their
Mongol neighbours, established a state and a government that
lasted for two and a half centuries-certainly the most stable and
powerful regime that Egypt had known since the Muslim conquest.
In Persia, lying on the main high road of invasion, things were
more difficult-but even there the heirs of Timur and the various
dynasties that followed them succeeded in maintaining the
stability and continuity of government-out of which, in time,
the territorially and administratively coherent modern kingdom
of Persia emerged. And in Anatolia, these same traditions of
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
66 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
government helped to maintain the various Turkish principalities
and the Ottoman state which eventually swallowed them all.
An important contribution of the steppe peoples to this stability
was a workable principle of dynastic succession. The juristic
doctrine of Islam was that the headship of the state was elective.
In fact the elective principle remained purely theoretical, and
Islam was ruled by a succession of dynasties, ranging from those
of the Caliphs themselves to the petty hereditary autonomies of
the provincial governors. But the elective principle remained
strong enough to prevent the establishment of any regular and
and accepted rule of succession. With the Caliphate, the fiction
of election was maintained on each accession, and beyond the
general principle that the Caliph should be chosen among the
members of the reigning family, there was no restriction of choice.
In the secular dynasties which held the real power, authority was
personal and military-and rarely survived the grandson of the
founder. Besides Islamic influences, Persian influences were also
powerful-but they came from the late, degenerate phase of the
Sasanid Empire of Persia, just before its collapse under the shock
of Arab invasion. The example it offered was of a personal absolutism,
unrestrained-and therefore unsupported-by any
entrenched rights or interests; depending on fear rather than
on loyalty. In the classical manuals of statecraft the possibility
of loyalty-by family, faith or estate-seems to be discounted
altogether, and kingship is based unashamedly on punishment
and reward.
The Turks introduced a new conception. Already in the
Orkhon inscriptions we find the notion clearly expressed of a
family singled out by God to rule over the Turks, and, more
vaguely, other peoples and lands beyond them. The same idea
reappears in an Islamic form in the correspondence of the Great
Seljuks, with their claim to an inherited and divinely sanctioned
imperial sovereignty, and again, in a pagan form, in the chancery
protocol of the Mongol Khans. For the Persians, the sovereign
was the sole autocrat; for the Turks and Mongols, sovereignty
was a family possession, and the whole family of the Khan or
Sultan had a right to share in it. In the kingdoms of the Karakhanids
and Seljuks we see the principle at work, whereby the
brothers and cousins of the sovereign are admitted to a share of
sovereignty. Under the Mongols, the whole vast Empire won by
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
MONGOLS, TURKS AND THE MUSLIM POLITY 67
the conquests was divided up into family appanages, each of
which was given to a son or grandson of Jenghiz Khan. We see it
again among the Anatolian principalities, and perhaps also in the
Seljuk and early Ottoman practice of appointing the sons of the
Sultan to provincial governorships, in which they held miniature
courts.'
In this system of family supremacy, leadership belonged to the
eldest member of the family, rather than to the son of the previous
leader. This principle of succession by seniority remained in force
in the Ottoman Empire to the very end. Though probably less
satisfactory than the rule of primogeniture, it was an immense
improvement on the anarchy of the past. In critical periods it
also assured the loyalty of the ruling family to their chief, and of
the people to their ruling family.
A ruling family, held together and sustained by strong ties of
tribal loyalty; a divine grant of authority, so sacrosanct that defaulting
members of the family were put to death by strangling
with a bow-string, to avoid the sacrilege of shedding their bloodthese
were no small advantages in setting up a regime that was
secure and accepted. But to make it permanent, in lands of ancient
culture and jaded loyalty, more was needed.
It was found. In the Turkish kingdoms there was a clarity
and cohesion in the institutional structure of state and society
that is in marked contrast with the looseness and vagueness of
classical Islamic times. The power of the state rests on and is
exercised through well-established and well-organized institutions
and social orders-army, bureaucracy, judiciary and men of religion,
with well-defined powers and functions, with regular
recruitment and hierarchic promotion. The emergence of these
new features has been variously attributed--to the steadiness and
sobriety of the new ruling groups, to changes in the system of
land tenure, to the transformation of Islamic belief and attitudes
1 Osman Turan, 'The idea of world domination among the medieval
Turks', Studia Islamica, iv (1955), pp. 80-8i; Ann K. S. Lambton, 'Quis
custodiet custodes: some reflections on the Persian theory of government',
ibid., vi (1956), p. 130. Cf Fuad Kbprillii, 'Les institutions juridique turques
au moyen-age', Belleten, ii/5-6 (1938), 41-76; idem, 'Bizans milesseselerinin
Osmanh milesseselerine te'siri hakkmda bazl miilahazalar', in Tiirk Hukuk
ve Iktisat Tarihi Mecmuast, i (193 ), 165-313; Italian translation, Alcune
osservationi intorno all' influenza delle istituwioni biyantine sulle istiturioni
ottomane (Rome, 1953).
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms
68 TRANSACTIONS OF THE ROYAL HISTORICAL SOCIETY
through the new orthodoxy, to the influence of Chinese-and
Byzantine-Imperial administration, to the introduction and
acceptance of fire-arms. All no doubt played their part-though
the determining of their relative importance is very much a
matter of argument. What is clear is that in these states, and
notably in the Ottoman Empire, land ownership and taxation,
justice and religion, government and war are better organized
and better correlated than ever before in Islam, and give to the
Turkish rulers an assurance, a competence and above all a permanence
that are new to the Islamic world.
With the consolidation of the Turkish states came an important
change in the nature of the realms over which they ruled.
Their territories were wider, their frontiers more permanent. The
constant rise and fall of petty principalities-regional or personal,
military or tribal, forming and reforming in ever different shapes
-had come to an end. After the Mongol invasions, three great
states, based on Egypt, Persia and Turkey, with more or less
stable frontiers, divided the Middle East between them. With the
Ottoman conquest of the Mamluk sultanate in 151 7 their number
was reduced to two-two great dynastic monarchies, which confronted
one another from the Caucasus to the Persian Gulf, as the
Sasanid and Byzantine Empires had done a thousand years earlier.
One of them, Persia, has survived to our own day; the disappearance
of the other has left many uncertainties that are not yet
resolved.'
School of Oriental and African Studies, London.
1 My thanks are due to Professor Ann K. S. Lambton for reading and
commenting on this paper, and to Mr G. R. Hawting for drawing my attention
to a serious omission.
This content downloaded from 87.77.121.151 on Thu, 11 Aug 2016 03:52:03 UTC
All use subject to http://about.jstor.org/terms