-
Cambridge University Press is collaborating with JSTOR to
digitize, preserve and extend access to International Journal of
Middle East Studies.
http://www.jstor.org
Early Ottoman Administration in the Wilderness: Some Limits on
Comparison Author(s): Joel Shinder Source: International Journal of
Middle East Studies, Vol. 9, No. 4 (Nov., 1978), pp.
497-517Published by: Cambridge University PressStable URL:
http://www.jstor.org/stable/162076Accessed: 01-08-2015 09:20
UTC
Your use of the JSTOR archive indicates your acceptance of the
Terms & Conditions of Use, available at
http://www.jstor.org/page/ info/about/policies/terms.jsp
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 [email protected].
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Int. J. Middle East Stud. 9 (1978), 497-517 Printed in Great
Britain 497
Joel Shinder
EARLY OTTOMAN ADMINISTRATION IN THE WILDERNESS: SOME LIMITS ON
COMPARISON
A synthesis of Ottoman administrative history has yet to be
written, and it is unlikely that one will appear in the near
future. The task is enormous, and more glamorous subjects continue
to receive priority. Even now the field of adminis- trative history
exists largely as an ancillary to the study of Ottoman diplomatic
instruments or as a foundation for the study of the modernization
of traditional society. In the latter case it has fallen under the
spell of institutional history, where three theses and at least one
'antithesis' scurry in their murine way across the tiles of the
Ottoman edifice. Despite the fact that a developed literature is
lacking, specialists in other disciplines have used the Ottoman
example for broad comparative studies of bureaucratic empires.1
Their premature attempts have perpetuated the notion already
endemic in Islamist circles that what we know of Islamic government
is all there is to know and need be known. Several themes, however,
have dominated the study of governing institutions in the Middle
East with a force that has surely impeded progress and fresh
thought. Our first results from the four theses referred to earlier
are terribly outdated at worst or in need of modification at best.
New source materials in the Ottoman archives and new readings of
older materials long subject to scholarly scrutiny call for a
reexamina- tion of those leading themes and the theses they inspire
before any attempt at synthesis and comparison is made.
The first twentieth-century thesis on Ottoman institutions was
advanced by Alfred Howe Lybyer in his World War I classic, The
Government of the Ottoman Empire in the Time of Suleiman the
Magnificent, reprinted in New York (1966) from the 1913 Cambridge,
Massachusetts, edition of Harvard University Press. Lybyer based
his research on Western materials, from which he concluded that the
Empire may be understood in terms of two institutions: the Ruling
Institution and the Moslem Institution. The first consisted of the
personal slaves of the sultans, slaves who with few exceptions were
the conscripted sons of Christian parents. In contrast, the Moslem
Institution was peopled entirely by
AUTHOR'S NOTE: I wish to acknowledge the support of the National
Endowment for the Humanities and of the State University of New
York Research Foundation in the preparation of this essay. I would
also like to thank Harvard University's Center for Middle Eastern
Studies for the facilities made available to me.
1 S. N. Eisenstadt, The Political Systems of Empires (London,
1963), is an example of the way in which the Ottoman case has been
used for comparison and model-building.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
498 Joel Shinder
freeborn Muslims. Members of the first wielded military and
administrative power, while those of the second were responsible
for the maintenance of the Islamic faith (theology, law,
philosophy, and other religious sciences). Whereas the first group
is European, Christian, and 'white,' the second is Asiatic, Muslim,
and 'Turkish.' When the strict separation of these staffs broke
down late in Siileyman's reign, the Empire's fate was sealed.2
Late in the thirties Paul Wittek, a most scrupulous Ottomanist,
presented his famous ghazi thesis in The Rise of the Ottoman Empire
and in a Revue des Jtudes Islamiques articles, 'De la defaite
d'Ankara a la prise de Constantinople,' both appearing in 1938.
During the fourteenth and well into the fifteenth centuries,
according to Wittek, the chief tension in the Empire was that
between the independent and 'antinomian' frontier warriors of the
faith or ghazis on the one hand and the orthodox purveyors of a
High Islamic, hinterland tradition, the ulema, on the other. The
sultan's authority could be maintained only by balancing these two
elements, whence the devsirme or child levy, at once a compromise
with the ghazi ideal of forced conversion (or death) and a
constraint on ulema pretensions. Ulema success would threaten the
raison d'etre of the state, the Holy War of the ghazis, and bring
on decline.
During the sixties Professor Stanford J. Shaw, of the University
of California, Los Angeles, then of Harvard, prepared an
unpublished study of the Ottoman Empire in two parts: 'The
Formative Years' and 'Decline and Reform' (available at Harvard
University's Center for Middle Eastern Studies, Reading Room). Shaw
observed a tension between the Turkish aristocracy, scions of
prominent Turkoman tribal families, and the devsirme class of slave
converts together with their freeborn Muslim sons. With the defeat
of the Turkish aristocracy, the devsirme class split into rival
factions temporarily allied with other groups such as the ulema and
harem cliques. The sultan could no longer depend on the
countervailing force of the aristocracy to check the excesses of
the victorious if fragmented devsirme class. Although the
aristocracy of the late sixteenth and seventeenth centuries
produced reformist tracts designed to restore to them the powers
usurped by the devsirme, the sultan was either too weak or too
dependent on devsirme military muscle to accept these proposals as
the bases for policy. The aristocracy's military contribution,
chiefly cavalry, was simply outmoded. Artillery and infantry were
needed, and these were the devsirme's forte. Shaw's view,
therefore, is a correction to the Lybyer thesis, and it has found a
degree of support recently from the thorough studies of Halil
Inalcik.3
2 Lybyer, Government, pp. 36-37, 50. His thesis was developed
and refined by H. A. R. Gibb and Harold Bowen in Islamic Society
and the West, Vol. I, Islamic Society in the Eighteenth Century,
Part I (Oxford, I950). I am indebted to Ms. Judith-Ann Corrente of
Harvard University for her comments in a seminar paper, 'Approaches
to Ottoman Institutions: An Historiographical Essay,' which I found
helpful in this analysis.
3 Professor Halil Inalcik has an extensive list of publications,
the most relevant of which are the following: 'Ottoman Methods of
Conquest,' Studia Islamica, 2 (1954), 103-I30; 'The Emergence of
the Ottomans,' in The Cambridge History of Islam, Vol. I
(Cambridge, I970), pp. 263-295; The Ottoman Empire in the Classical
Age 1300-1600 (New York,
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 499
Against these three essentially bipolar theses stands an
antithesis, if such it is, developed variously by Lewis V. Thomas,
Norman Itzkowitz, and Halil Inalcik.4 In contrast with the theses,
which are political in emphasis, the anti- thesis is a cultural or
social model of the traditional, High Islamic Near Eastern state.
The model describes a corporate society of Men of the Sword, Men of
the Pen (both making up the Ottoman ruling or askeri class), Men of
Negotiation, and Men of Husbandry (both making up the Ottoman
subject class of reaya). The Men of the Pen in the Ottoman system
are then subdivided into the ilmiye and kalemiye, the careers of
religious-legal knowledge (the ulema) and of the bureaucracy (the
kiittab). Ottoman absolutism was founded on Persian traditions of
statecraft modified by Islamic law, solidified by the Turkic
tradition of dynastic succession, which replaced the Islamic theory
of election, and effected by the reign of justice within a circle
of equity. This circle has eight propositions: a state requires a
sovereign authority to enforce rational and Holy Law; to have
authority a sovereign must exercise power; to have power and
control one needs a large army; to have an army one needs wealth;
to have wealth from taxes one needs a prosperous people; to have a
prosperous subject population one must have just laws justly
enforced; to have laws enforced one needs a state; to have a state
one needs a sovereign authority. Justice is fundamentally the
maintenance of corporate order - keeping the four classes of men
and their subdivisions in place. This is done through the Holy Law
of Islam supplemented with if not complemented by the rational and
customary law of the sultanate, kanun. But the key fiscal and
administrative unit for the implementation of order is the timar or
quasi-feudal regime, which is seen as part of the continuing
Persian legacy, through the Abbasids and Seljuks, to the Ottomans.
Here the key to Ottoman studies, then, is not the polarities of the
political models. Ottoman society was too complex. Diversification
and stratification, shifting alliances and alignments within and
between occupational classes are what the sociocultural model
emphasizes. Rather than tension between two groups with a central
authority seeking balance, this model suggests the genesis of a
ruling elite drawing membership from and controlling admission to
all service areas. It is clearly more sophisticated than its
competitors. All four, however, are ruled by themes that pervade
Islamic studies in the twentieth century.
This point may be elucidated through the simple graphic device
of the chart, whose object is to afford the examiner an opportunity
to discern the meanings of symbols A and B. What is the underlying
structure of the chart? Or, put
1973). Inalcik, unlike Shaw, prefers to see the 'Turkish
Aristocracy' as the march lords or uc begs and limits the tension
between this group and the devsirme party to the period ca.
I362-ca. 600oo. He also works toward a synthesis by incorporating
the 'Muslim Aristocracy' or chief ulema families into his scheme.
The effect is to bring Wittek and Lybyer (as corrected by Shaw)
together.
4Thomas, A Study of Naima (New York, 1972); Itzkowitz,
'Eighteenth Century Ottoman Realities,' Studia Islamica, 26 (I962),
73-94, and Ottoman Empire and Islamic Tradition (New York, 1972),
and, with Max Mote, Mubadele: An Ottoman-Russian Exchange of
Ambassadors (Chicago, 1970); Inalcik in the works cited in n. 3
above, but especially in Classical Age.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
t_rl O O
.
ca
FOUR THESES IN OTTOMAN INSTITUTIONAL HISTORY
Origin of thesis I. Lybyer 2. Wittek
3. Shaw/inalcik
A (State) Ruling Institution
Ghazi (to Devsirme) Devsirme Party Devsirme Kapl Kullarla or
Central Government
Intermediate Status B (Society) Moslem Institution
Ulema
Turkish Aristocracy Ulema/Muslim Aristocracy March Lords
4. Inalcik/Itzkowitz/Thomas Askeri (Rulers) Men of the Pen
Bureaucracy Religion/Law
Men of the Sword Kapi Kullar Timariots Auxiliaries
Miners Guards (e.g., bridges) Road crews Falcon breeders Sheep
breeders, marketers
Reaya (Subjects) Men of Negotiation Merchants Artisans
Men of Husbandry Farmers Pastoralists
a Persons in the immediate service of the sultan, whether in the
palace itself or in the provinces, including the devsirme; either
salaried, supported by estates, or in the combination of these
two.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 501
another way, what do the symbols 'A' and 'B' stand for, and what
is their thematic foundation? (See chart on facing page.)
One theme in Umayyad and Abbasid history (A.D. 661-750 and
750-I258, respectively) is the struggle between advocates of
centralized and those of decentralized government. A second popular
theme has been the antagonism of different racial groups in
government itself, if not throughout the society. Arabs, Persians,
Kurds, Berbers, Turks, and Circassians figure prominently in the
application of this theme. Yet a third is the rivalry of the
military and the bureaucracy in such states as the Abbasid, Great
Seljuk, Mamluk, and Ottoman. Where these themes have not masked a
stark Western ethnocentrism, they have often served as handmaidens
to ideologies past and present, East and West: nationalism,
secularism, Islamic reformism, or Arabism. More specific themes of
the same formulation exist for the Ottoman case: on origins -
Byzantine/ European vs. Muslim/Asiatic; on ruling elites - gallant
frontiersman vs. effete hinterlander; on power loci - Palace vs.
Porte; on policy - reaction/Islamic Orthodoxy vs.
reform/Westernization; on broad cultural influences - Turkish
romanticism vs. Islamic ecumenicism. It is obvious that these
themes as well have ideological content or potential. The chief
problem raised by the examples, however, is not the advancement of
ideology. Their bipolar formulation violates the principle in logic
of the excluded middle. This requires that any ambiguous statement
be either true or false. For example, it is incorrect to ask
whether the Ottoman government was in the hands of free Muslim
Turks or of converted non-Turkish slaves and to expect that one is
true, the other false. The correct question is whether or not the
Ottoman government, once clearly defined, was in the hands of free
Muslim Turks, and so forth. Disregard of the principle creates
artificial situations of tension between alternative choices which
an historical society may or may not have faced. It produces loaded
or complex questions of the nature: 'How did you spend the money
you stole'?
Although the preceding examples do not immediately degenerate
into fallacies, a principal theme in Islamic history does - a theme
that, as is demonstrated, is the underlying structure of the
mystery chart above. This theme is the disruptive tension between
the ideals of the Holy Law of Islam, the shari'a (feriat in Ottoman
usage), and the practical needs of sovereign states. Inquiries
based on the theme conclude from comparisons of Islamic political
theory and isolated examples of actual practice that Islamic
governments have consistently failed to be 'Islamic' and that
Islamic society has failed to be 'political.' This succinct version
borders on equivocation, perhaps inevitably in view of the heavily
theoretical bias of the research supporting the theme. The research
closes in on legal theory and executive practice, demanding
absolute purity of the first, corruption of the second:5
Even the most harmonious co-operation of jurisprudents and
executive officialdom could not have prevented the gap between the
ideal and the actual, the normative and
5 Gustave E. von Grunebaum, Medieval Islam: A Study in Cultural
Orientation, 2d ed. (Chicago, 1956), p. 143.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
502 Joel Shinder
the practical, the precedent of sacred law and the makeshift
decision of the executive order, from widening until it became
unbridgeable. The pious condemned the ruler's deviations from the
established norm of the Prophet's days, and in fear for their souls
they evaded his call when he summoned them to take office.
Thus the government of Allah and the government of the sultan
grew apart. Social and political life was lived on two planes, on
one of which happenings would be spiritually valid but actually
unreal, while on the other no validity could ever be aspired to.
The law of God failed because it neglected the factor of change to
which Allah had subjected his creatures. [Legal theory] had,
unwittingly perhaps, relinquished that grandiose dream of a social
body operating perpetually under the immutable law which God had
revealed in the fulness of time.
There is an opposition here between government or state and
society, and it is suggested that this is a uniquely Islamic
phenomenon despite historical evidence to the contrary. Because the
legists of early Islam wrestled with the problem of theory and
practice for ages bereft of the Prophet's guidance, it is argued,
all subsequent states and societies, in being Islamic, suffered the
same difficulty. This is the fallacy of division, where the
properties of the whole are considered true also for the parts
(assuming, of course, the validity of the properties of the whole
as described). The fallacy of composition, where the properties of
the parts are considered true also for the whole, is introduced
when certain key Islamic states, like the Abbasid, are considered
models for the entire Islamic community or umma.
When the full theme is drafted into use for the discussion of
Ottoman state and society, the fallacy of division applies. One
consequence of such reasoning is the judgment that practitioners of
Ottoman government, as of any Islamic government, uniformly
displayed a total lack of administrative morality. This comes from
the uncritical reading of a spate of 'government' manuals which can
rival The Prince for the challenge they pose the literary analyst.
An eleventh- century manual urges, 'Commit no forgery for a trivial
object, but [reserve it] for the day when it will be of real
service to you and the benefits substantial.'6 Noting evidence of
corruption in Ottoman administrative ranks, a thinker of this ilk -
steeped, to be sure, in Oriental lore - would find his general
conclusion exonerated. The abundance of Ottoman manuals which urge
probity would not impress him. A comparative perspective, in other
words, has thrown the Ottoman child into the wilderness where, left
to the devices of nature, he would thrive on the milk of two gray
wolves, the beasts of State and Society or, respectively, symbols A
and B in the chart above (p. 500). In terms of the theme, it should
be remembered, 'society' is chiefly the ulema and other popular
leadership outside the framework of the government proper.
Recognizing the official status of the ulema in the Ottoman state,
the Inalcik-Itzkowitz-Thomas proposition is more strictly
functional in its askeri-reaya definition.
The Ottoman state should be fully characterized, however, as
primarily Islamic, Turkish, dynastic, monocratic, and agrarian. To
the extent that it was
6 Reuben Levy, trans., A Mirror for Princes: The Qdbus Ndma by
Kai Kd'uis Ibn Iskander, Prince of Gurgan (London, I95I), p.
209.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 503
Islamic, it enjoyed only some elements of a civilization to
which it fell heir, and it contributed much to that civilization.
The same may be said for its Turkic legacy. In neither case is the
legacy full. The dynasty, the monocracy, and the economy differ
vastly from earlier and later forms, Turkish or Islamic. Similarly,
Ottoman administration differs from earlier and later varieties in
accordance with development under unique conditions, particularly
those introduced with the expansion into Europe and, only
subsequently, into the Islamic heartlands of the Middle East,
thence into North Africa. These distinctions, unfortunately, have
not been preserved. Identical terms for administrative offices and
methods as between the Ottomans and other Islamic states have been
taken for coincidence and continuity in institutions, regardless of
chronological and spatial gaps. To the shifting of personnel across
frontiers has been added a lengthy baggage train containing the
paraphernalia of a paradigmatic and reified 'traditional Near
Eastern State and Society.' The important conclusion is that it is
indispensable to study Ottoman administration for itself and in its
own context. The comparative study of Islamic government is more
than a welcome endeavor, provided that the comparative approach
yields results other than mirror images of a self- fulfilled
prophecy. The remainder of this essay, accordingly, is concerned
with what the past bequeathed to Ottoman administration (to the
extent that current research allows) and how the bequest has been
presented in historical writings. Ideally, this discussion would be
followed by an in-depth examination of Ottoman administration from
the reign of Sultan Mehmed II (145I-148I) to the first third of the
eighteenth century, by which time major changes in the imperial
administration had largely run their course. The materials used
here, however, are not drawn from the early period merely to
introduce a continuous survey. Even a cursory review is not
possible in the present state of the field. The chief intention is
rather to discuss notions which have affected the study of Ottoman
administration in any period. One such notion, the Turkish ghazi
thesis, has its roots in the first Ottoman chronicles.
An early source hostile to bureaucratic, centralized government
introduces the history of Ottoman central administration with the
following story. One market day in about the year A.D. 1300 a man
from the neighboring principality of Germiyan came to the court of
Osman Beg, founder of the Ottoman dynasty and first of the great
ten leaders of that house. The visitor offered to purchase the tax
farm on market tolls. Osman ingenuously asked, 'What is a market
toll?' The Germiyanid explained what market tolls were and, noting
the just prince's ire, advanced his claim by demonstrating the
universality of the practice in all sovereign states since the
beginning of time. Yielding to the stranger, who was supported by
the local kadi and military commander, Osman promulgated
regulations known as kanun - executive law - to govern not only
market taxes but also timars, the estates whose usufruct went to
the support of warriors and other servants of the dynasty.7
The tendentiousness of this story is more significant than the
content. On one 7
'A?ikpa?azade, Tevdrih-i Al-i Osman, ed. 'Ali (Istanbul,
I332/II915), pp. I9-20. The
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
504 Joel Shinder
side is the simple, pure leader of warriors for the faith, whose
sole concern is and should be advancing the frontiers of Islam at
the cost of the Christian infidels. On the other side stand the
external forces of the status quo, princes like those of Germiyan
who, together with the Islamic religious establishment (the kadi)
and the entrenched household military commander, prefer stable,
regulatory government to the material sacrifices and disorder of
jihad. Outsiders, it is claimed, sullied the high aims of the
Ottoman house, which is identified with the ghazi movement. The
tsar is good, but he is misled by his advisors! The principal
interest of the earliest chronicles as they have come down to us,
then, is the ghazi's exploits against the unbelievers.8
Sophisticated government, whose Anatolian purveyors were the
religious-legal scholars of the ulema class, is considered the
chief hindrance to the ghazi effort. The simplistic, highly
syncretistic and, to some extent, mystical Islam with Shi'i
heterodox tendencies to which the ghazis independently adhered
would not tolerate the extensively developed and institutionalized
system of siyisa shar'iya. This was a theoretical symbiosis of
politics and Islamic Holy Law which regulated not only the status
and financial obligations of lands and peoples absorbed into the
Abode of Islam by conquest or surrender, but also the prerogatives
and responsibilities of those in authority.9 According to this
doctrine, the two classes of emirs or secular princes and the ulema
are in authority. The first keeps order, and the second serves as
the heir and guardian of the Prophet's path. It is what the ghazis
had presumably rejected in leaving the heartlands of the Islamic
Middle East to open new frontiers in the company of clansmen and
confederates. (Never mind the compulsion to move ever farther
westward with the Mongols breathing hard not far behind.)
Interloping strangers from the heartlands with their manners,
customs, and institutions were not wanted even for the sake of
legitimizing and consolidating the recent conquests. Sword,
compound bow, and pony were thought sufficient to that end. More
important, the ghazis justifiably felt that the existence of the
timar regime and kanun legislation at this early date is a moot
point. 'A?ikpasazade is henceforth referred to as Apz., preceded by
the editor's last name.
8 Friedrich Griese, ed., Die altosmanische Chronik (Tevarikh-i
Al-i Osman) des 'Aszk- pasazdde (Leipzig, I929), in addition to the
'All and Atsiz editions; Oruc, Karamani Mehmed, $ukrullah, Tursun
Beg, Ahmedi, and Ne*ri in several editions (but do see the
Qift9io/lu, N. Atsiz collection, Osmanlh Tarihleri, Vol. I,
Istanbul, I949) - the earliest chronicles we know - possibly share
a common prototype.
9 Refer to the discussion of Ibn Taymiya's docrine of siyasa
shar'iya in E. I. J. Rosen- thal, Political Thought in Medieval
Islam: An Introductory Outline (Cambridge, 1958), pp. 58, 6o. It
should be noted, however, that even in the small states which were
successors to the Abbasids the doctrine was often ignored in
practice, its author having little if any following. Ira M. Lapidus
has demonstrated that in many Syrian and Mesopotamian towns and
cities new elites emerged during periods of upheaval, common from
the ninth century. Military regimes which controlled former Abbasid
provinces were incapable of 'reordering' local situations. The
urban ulema stepped into the vacuum and, through intermarriage with
merchant, administrative, and landowning families, forged a new
elite defined by religious qualification. This development,
however, should not be superimposed on the scene in western Asia
Minor from the thirteenth century. See 'The Evolution of Muslim
Urban Society,' Comparative Studies in Society and History, 25
(1973), 21I-50, esp. pp. 39-4I.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 505
strangers would divert the resources of the new conquests - that
is to say, booty - from the hands of the actual victors and put
them to use for the benefit of others through an organized and
regular system of taxation.
Unqualified acceptance of this interpretation is risky. In his
story of the market toll, 'Aslkpasazade (1393-1481) left his own
imprint on his recension of the prototypical chronicle. His
disapproval of Sultan Mehmed II's administra- tive policies is
clear. Where sultans through Murad II (1421-1451) merely led or
ordered campaigns resulting in conquests, Mehmed II went beyond
conquest and had scribes busily registering and confiscating the
booty.10 Bayezid I (1389-I403) had also committed this 'error' of
rapid centralization, but Timur had put an end to his excesses
after the Battle of Ankara (1402) by restoring local autonomy to
Asia Minor dynasts. The evil of bureaucracy itself is attributed to
external (Germiyanid) influences on the pristine form of Ottoman
govern- ment. Those who perpetuated the evil, the ulema, also come
under the author's fire. The Candarli family of viziers is singled
out for having introduced corrup- tion into the world.1 Acting
under the influence of their breed of officialdom, sultans
themselves began to accumulate wealth, that sure sign of corruption
resulting in the fall of states and rulers, the destruction of the
soldiery, and chaos. Bayezid I and his personal agents had suffered
these very consequences. It is not only the influence of the ulema,
then, which is responsible for disruption. Part of the blame is
also placed on the use of personal agents or slaves, kapt
kullarz.12
'Aslkpasazade's argument is poorly developed. The ghazis were as
much refugee tribal fragments from the east fleeing Mongol advances
and seeking new pastures as they were earnest warriors for Islam
occupied more with the future of their souls. The late
fifteenth-century politics of the chronicler must not be confused
with early thirteenth-century realities. The attempt to use the
early chronicles to discern administrative developments, therefore,
could not make much progress. The information is fragmentary and
somewhat unreliable. Another source was relied on to fill in the
lacunae. This was the body of precedents or traditions coming to
the early Ottoman principality from the Islamic heartlands through
the mediation of the Seljuk state of Rum and its later overlord,
the Ilhanid Empire.13 The Ottoman's institutional debt to the
10 See Giese, Apz., pp. I75-I77. The importance of the
juxtaposition offeth etmek (to conquer) and kaleme almak (to
record) was pointed out to me by Professor Rudi Lindner of Tufts
University.
11 Atsiz, Apz., pp. 1i8, I39. The view is shared by other works
of the genre, such as 'Ahmedi' in Nihat Sami Banarli, XIV. aszr
Anadolu fairlerinden Ahmedi'nin Osmanlz tari- hi: Ddsztan-z
tevdrih-i miilik-i dl-i Osman ve CemSid ve Hursid mesnevisi
(Istanbul, 1939), p. 74. $ukrullah's Behcet in Atsiz, Osmanli
Tarihleri, p. 57, and Banarh's 'Ahmedi,' p. 83, relate how Bayezid
I had to restrain the kadis from oppressing the people.
12 This general historical judgment of trial by ordeal is made
much of in 'All, Apz., pp. I97-198. History in these chronicles is
fundamentally moralistic, teleological, and polemical.
13 Supplementing the early chroniclers with tradition (Turkish,
Islamic, Middle Eastern) is standard fare in twentieth-century
Turkish historiography, represented in
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
506 Joel Shinder
Seljuk-dominated border principalities, to the Seljuks of Rum,
to the Great Seljuks of Iran-Mesopotamia and, later, to their
successors, the Ilhanids, is posited. This assumption bears
examination under three headings: (i) composi- tion and training of
bureaucratic cadres; (2) Seljuk and Ilhanid precedents; and (3)
forms of the tradition of the vizierate.
There is some evidence that the Ottoman chancery scribes early
in the fifteenth century relied on earlier collections of letters
and other forms of guidebooks from non-Ottoman lands in their task
of creating an Ottoman methodology for recruiting, training, and
entitling the membership of the scribal cadres. Their sources were
of three kinds: insa or miinseat (examples of correspondence, both
official and unofficial); 'mirrors for princes' (books of counsel
by fathers for sons, or by viziers for princes or sultans); and
adab (collections of anecdotes, homilies, and excerpts from
classics in prose and poetry). The first genre related directly to
the functions of the early administration, the second to the
general rules which governors should follow both to maintain power
and to rule justly. The third was concerned with 'discipline,' the
achievement of cultural refine- ment and wisdom which supporters of
the state were to cultivate not only for the prestige of the state
itself but also for the very identity of the bureaucratic elite in
a corporate society.
The oldest known Ottoman insa is the Teressiil of the poet
Ahmed-i Da'i, who died after 1421. Based on Arabic and Persian
texts, the Teressiil consists of two parts, the standing form for
such works. Part One contains advice to scribes, and the rules
prescribed here are termed edeb, Turkish for adab. Part Two is the
practical section, where models of letters and forms of address are
presented to exemplify the theoretical considerations of the
introduction. The usage of the term edeb is an interesting
indication of the kind of appreciation early Ottomans with their
pragmatic bent had for classical Islamic genres. The author's
personal career is also interesting in the light of the market toll
story of 'Asikpasazade, inasmuch as he first served under the
Germiyan prince Ya'kub Beg II (I387-1390, 1402-1429) before
entering the service of the Ottoman Emir Siileyman (1405) and
Sultan Murad II. Perhaps this is 'the man from Germiyan' displaced
one century!14
The Teressiil's importance in the formation of an Ottoman
chancery, however, is not easily assessed. Bj6rkman15 and Tekinl6
detect parallels and continuity, three generations of scholarship:
Zeki Velidi Togan, Umumi Tiirk Tarihine Girif, Vol. I, En Eski
Devirlerden I6. Asra Kadar (Istanbul, I946); I. H. UzuncarSlih,
Osmanli Devleti Teskildtzna Medhal (Istanbul, 1941); and in the
works of Halil Inalclk as cited.
14 Ismail Hikmet Ertaylan published a facsimile of the
Terressiil in Ahmed-i Dd'Z, Hayatz ve Eserleri (Istanbul, 1952),
pp. 325-328. Also see W. Bjirkmann, 'Die Anfiinge der tiirkischen
Briefsammlungen,' Orientalia Suecana, 5 (Uppsala, I956), 22-23, on
insa forms. A possible source for the Terressiil was published by
Adnan Sadik Erzi, ed., Selfukiler Devrine did Insa Eserleri . . .
(Ankara, I963).
'5 Bjorkman, 'Anfinge,' p. 29. 16 6inasi Tekin, ed.,
Mendhicii'l-InSa: The Earliest Ottoman Chancery Manual by Yahyd
bin Mehmed el-Kdtib from the i5th Century, in Sources of
Oriental Languages and Litera- tures: Turkic Sources, Vol. II
(Roxbury, Mass., I971), p. II.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 507
respectively, in inSa from Sultan Mehmed I (I403-I421) through
Sultan Mehmed II. Tekin goes further, concluding from the 1412
Bursa copy of the famous Persian Sa'adetndme (written around I307
by 'Alaeddin-i Tebrizi) of the Ilhanid period that 'the Ottomans
are copying the administration and institutions of the Selcuks
(sic) and Ilhanids.'17 Barkan is more cautious in this respect but
still subscribes to the view that models and parallels constitute
grounds for concluding the identity of institutions and practices
without regard to region and period or nature of absorption into
the Ottoman realm.l8
However difficult the problems posed by the use of insa
literature among the early Ottomans, the influence of the 'mirrors'
genre together with that of adab is even harder to establish with
firm evidence. Nizam al-Mulk's famous Siyaset- name and the
writings of Nasir al-Din Tisi were probably read. A translation of
Kai Ka'fis's Qabfs Ndma from Persian into Turkish was certainly
made for Murad II in his first reign (1421-1443), and Turkish
manuscripts are more numerous than those in the original Persian,19
but the degree to which these tracts on Oriental governance
actually shaped the Ottoman effort in the con- struction of an
administration cannot be determined. The ideals expressed in such
works may well have been shared by early Ottoman scribes, however,
even if their own conditions were quite different. A look at this
ideal is necessary, therefore, provided that it is ranked in the
category of theory or aspiration. It would be misleading to view
the ideal as a product or reflection of realities.
The ideal is founded in the golden age of Abbasid rule (ca.
750-900). Under the Abbasids 'administrative appointments were
likely to be made from candidates belonging to a rather narrow
circle of families. These families were in possession of the
secrets of government technique, they were familiar with empire
conditions, and they had the necessary connections.' Offices were
not hereditary, but professions like the military and the
bureaucracy required special knowledge and abilities, so they were
largely exclusive. Where clerks tended to be free men or mawali
(non-Arab clients) and Persian, soldiers tended to be freedmen and
slaves of Turkish or other non-Arab and non-Persian descent.20
If the Ottoman scribal corps of the early period aspired to an
equal degree of exclusivity, they may well have sought the cultural
identity and etiquette which accompanied the Abbasid civil service
corps. That cultural form has been ascribed as a Bildungsideal,
polite education or, in Arabic, adab. The original connotation of
this term was 'the discipline of the mind and its training ...
17 Ibid. 18 Omer Lutfi Barkan, XV ve XVI tncz Aszrlarda Osmanlz
Imparatorlugunda Zirai
Ekonominin Hukuk ve Mali Esaslarz, Vol. I, Kanunlar (Istanbul,
1943), pp. lxxi-lxxii. On the problems involved in the use of insa
model documents for historical purposes, see Irene
Beldiceanu-Streinherr, Recherches sur les actes des regnes des
sultans Osman, Orkhan et Murad I (Monachii, Romania, i967),
passimn. This is a critical analysis of several early texts.
19 Levy, Mirror, p. xxi. 20 von Grunebaum, Medieval Islam, p.
213.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
508 Joel Shinder
characterized by combining the demand for information of a
certain kind with that for compliance with a code of behavior.'
Although the religious judge and theologian were trained strictly
in the sciences of tradition, canon law, and scholastic theology,
the perfect kdtib or scribe had to be competent in grammar, law,
theology, literature, astronomy, mathematics, medicine, lexicology,
political theory, and the science of administration. The ability to
recite the Koran and Traditions by heart was also necessary. In
fact such encyclopedic knowledge was really narrowed to grammar,
belles lettres, and history on the theoretical plane, and
calligraphy and style on the practical level. Both recruitment and
training, therefore, suggested preference for the all-rounder who
could pass freely through the bureaucracy and assume functions as
diverse as chancery and finance with equal facility.21
From the Abbasid ideal to Ottoman practice a yawning gap of
almost five hundred years exists. Many have tried to bridge the
empirical gap through studies on Seljuk and Ilhanid administrative
organs and methods. First the Great Seljuks in Iran and
Mesopotamia, then the Seljuks of Rum in Asia Minor proper and,
finally, the Ilhanids with supremacy in both regions are held to
have mediated the Abbasid-Ottoman exchange with but a few
Mongol-Turkic variations. The full Abbasid apparatus, therefore, is
seen to have been as readily available to the Ottomans as the
published traditions of the Prophet were to Muslims of that
age.
Under the first two Ottoman princes, Osman and Orhan (together,
ca. 1281-1326), the structure of the Ottoman state and its
administrative methods were probably very similar to those of the
other Anatolian principalities at best. Little more than this
statement of probability can be hazarded. Not even the very
officials of the early Ottoman state are clear. For example, an
important finance officer for the Seljuks of Rum was the mustawfi.
His office and duties may be a precedent for the Ottoman finance
officer, the defterdar, or for the commissioner of the cadaster,
the defter emini. The inception of neither Ottoman office, however,
is known accurately. Likewise, just as the Seljuk royal council
(divan-i hdss) may be the model for and the functional equivalent
of the Ottoman council (divan-i hiimdyun), the exact functions and
ranks of member officials concerned with military, judicial,
financial, and other affairs could and probably did differ
considerably.22 Because the Seljuks and the Ottomans are quite
alien to the modern reader, it is easy to accept functional
equivalency based on titulature and vice versa. The faith, the
culture, the language, and the civilization seem to be the same in
each case. This argument, however, would have the Congress of the
United States of America and the Parliament of the United Kingdom
functionally equivalent entities in respects going far beyond
the
21 Ibid., pp. 213, 250, 253-255, and Walther Hinz, 'Das
Rechnungswesen oriental- ische Reichsfinanzimter im Mittelalter,'
Der Islam, 29 (I950), I-2.
22 Although many parallels may be sought, or analogues found,
fundamental differences obtain between various Seljuk offices (in
functions, hierarchical position, and power) and the later Ottoman
forms. This is clear in the work of Osman Turan, Tiirkiye
Selfuklularz Hakkznda Resmi Vesikalar: Metin, Terciime ve
Arastzrmalar (Ankara, I958), pp. I-62.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 509
responsibility of legislation and the relationship of these
bodies to 'the people.' And in this case, after all, there is a
direct historical link between the two nations. Historical
differences do in fact obtain to alter institutions even in the
proximate case of the Seljuks of Rum and the Ottomans. The case for
the Ottoman timar's deriving from the Byzantine pronoia rather than
from the Seljuk iqtf' is cogent enough in itself to call for
significant changes in administra- tion.23 It is hardly proper,
then, to fill in early Ottoman unknowns with Seljuk vaguely
knowns.
Although the case for the Seljuks is weak, that for the Ilhanids
from the reign of Shazan (I294-I304) and his vizier Rashid al-Din
as the true precepts of the Ottomans in the field of government is
more coherent. Ilhanid and the later Timurid administrative
handbooks are found in the libraries of Istanbul. The use of
Persian phrases in official records and the adoption of the Persian
solar calendar for the fisc, as well as the Ottoman use of siyakat
cipher script for financial registers, could stem from Ilhanid or
Ilhanid-Seljuk practice.24 Some malefactor, after all, had to
create that cipher which has blinded more than one Ottomanist. The
wording of Ilhanid berats, which certified various state
liabilities to individual parties, is similar to that of Ottoman
documents of the same name.25 From evidence of this nature, the
doyen of Ottoman history to the imperial age, Halil Inalclk,
asserts that Sultan Bayezid I introduced the full 'Turkish-Islamic'
system of central government as developed in Persia under the
Mongol Ilhans.26
This system included provincial land and population surveys,
fiscal methods, a central treasury, and a bureaucracy. Dominant in
the system was Bayezid's extension of his personal ghulam-kapzkulu
or slave corps. One religious judge turned territorial prince
actually called Bayezid 'son of a Mongol, totally lacking knowledge
and grace,'27 so great did he sense Mongol influence at the Ottoman
court. The similarity of Ilhanid and Ottoman administration is
clearest in the forms of official registers maintained. Under the
Ilhanids seven basic types of registers have been described.28
These include the following: a journal for daily transactions and
appointments; a general state-of-the-treasury register; a register
of ordinary payments; a register of working capital;
revenue-expenditure books for individual cities and provinces; a
general register of the realm's total revenues; and books of
regulations on taxation and other matters. Three of the types (the
first, third, and fourth) bear the same names used in the Ottoman
system, if for different purposes (ruizname, tevcihat, and
tahvil7t). The Ilhanid extraordinary tax known as 'avdrid and the
tax levied in kind to provide
23 Claude Cahen, Pre-Ottoman Turkey (New York, 1968), pp.
I82-183. 24 Hinz, Rechnungswesen, pp. 4-6, I3. 25 Ibid., pp. 20-22,
and I. P. Petruchevsky, 'The Socio-Economic Conditions of Iran
under the Il-Khans,' in The Cambridge History of Iran, Vol. V,
The Saljuq and Mongol Periods (Cambridge, I968), p. 494.
26 Inalcik, 'Emergence,' p. 280. 27 Togan, Umumi Turk, p. 33I.
Togan (pp. 329-330), is in agreement on the synthetic
'Turkish-Islamic' system of central government. 28 Hinz,
Rechnungswesen, pp. 114-134.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
5 1 Joel Shinder
food and fodder for the army in a given district, the 'ulffa
tax, have exact Ottoman counterparts in the same meaning.29
Finally, the Ottoman practice of having secular kanun/yasak
regulations side by side with shari'a prescriptions is also an
Ilhanid practice.30 Moreover, Ilhanid land tenure fell into four
general categories, each of which has its Ottoman analogue: state
lands; private royal domains; lands of the religious institutions
and pious endowments; private lands (mulk). This categorization,
however, preceded the Ilhanids as much as it did the Ottomans in
accordance with custom and Holy Law. Besides, the tenacity of
landed classes is a well-known historical phenomenon, and this
certainly affected classification of lands.
The evidence admits a strong case for Ottoman-Ilhanid
institutional ties. Nonetheless, even in that body of material much
is left to be desired. Whether particular officials like the
Ilhanid defterdar-i memalik and the Ottoman defterdar of the early
period held similar rank and fulfilled similar roles in government
is a matter for conjecture. Moreover, the general prejudice of
Orientalists for 'Islamic' governments ignores the possible
influence on the Ottomans of non- Muslim Mongol-Turkic states north
and east of the Black Sea. The kinds of registers the Ottoman kept,
however like those of the Ilhanids, give few clues apart from the
similarity itself to the method and timing of borrowing. Timing is
particularly difficult to establish owing to the formalism of the
diplomatics involved. As in the Seljuk case, therefore, a general
structure of 'Islamic' or 'traditional Near Eastern' government
cannot be assumed to presuppose functional equivalency, however
great the force of tradition and precedent. Circumstances,
policies, interests, and alignments clearly differentiate the
Ottomans from their colleagues in government throughout the Muslim
world.
These considerations suggest the tendency in earlier approaches
to advance the idea of continuity in Middle Eastern governing
traditions to the detriment of innovation or change. Some kind of
search for a stereotypical or paradigmatic 'Muslim government'
seems to be in progress, despite the prior conclusion that the
attempt in the Muslim world to create a Muslim government has
consistently failed! A look at the institution of the vizierate is
an important step in shifting the balance to the side of change.
The example is particularly crucial in that, of the many
innovations introduced by the Ottomans, the grand vizierate is out-
standing for its effects on the administrative history of the
empire.
The Arabic wazir - literally, one who carries a burden - has the
sense of aide or counselor.31 The office is neither a direct
borrowing from Sassanian practice nor a static tradition. By the
sixth century the Sassanian equivalent, the buzurg framdddr, was in
decline or non-existent, and the Umayyads had no first minister
29 Petruchevsky, 'Iran under Il-Khans,' pp. 532-533. 30 Togan,
Umumi Tirk, p. 330. The Ilhanid state was non-Muslim until
Ghazan's
conversion, and this may account for the dual legal system. The
Ottomans produced regulations under different conditions; namely,
the absorption of largely non-Muslim areas with local codes of law
into a state whose leadership was Muslim.
31 Dominique Sourdel, Le Vizirat 'Abbdside de 749 d 936 (I32 a
324 de l'Hegire), Vol. I (Damascus, 1959), pp. 50-54.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 511
of the kind the Abbasids were to develop when they shifted the
seat of govern- ment from formerly Byzantine Syria to formerly
Persian Mesopotamia.32 The first person to hold the title was Abfi
Salama, officially saluted wazir dl Muham- mad by the Kufan army in
A.D. 749. As the salutation indicates, the office was coincident
with the Abbasids' new regime of increasing orthodoxy comple-
mented by oriental influences, which was mediated by the non-Arab
secretarial classes.33
Abbasid viziers were most often recruited from the central or
provincial administration and were secretaries of recognized
competence. While court personnel rarely entered the office,
administrators of the fisc were especially prominent. In time,
recruiting was carried out almost entirely among the awlad
al-kuttdb, sons of the scribal corps, thus generating great
vizieral dynasties.34 Functionally, the Abbasid vizier headed up
the scribal corps and was responsible for their work in three major
areas: applying the injunctions of the Holy Law in fiscal matters;
resolving extra-shari'a problems; and executing the will of the
caliph according to law. The vizier, therefore, stood between the
Holy Law and the caliph's discretionary powers in those areas
outside the purview of the sacred code. His dual role was most
effective when a special relationship obtained between himself and
the caliph. In this relationship the vizier acted as 'tutor' of the
caliph at the same time that he was his servant and personal
secretary. The personal relationship between sovereign and servant,
however, was all too often jeopardized by palace intrigue and by
the attempts of some servants them- selves to become masters.
Through the authority delegated to him, the vizier could place his
own men in crucial positions and influence policy in all spheres,
including the army. On the other hand, a vizier was vulnerable in
that he did not have complete control over his own staff. Military
commanders frequently interfered in the vizier's job. The extent of
the minister's powers, therefore, ultimately depended on his
personal prestige and talents, as well as on his ability to profit
from his special relationship to his master. Vizieral instability
in the face of caliphal authority and rival groups marred the
experience of the Abbasid period even before the caliphal master
lost his power to praetorians. The vizier had no weapon to match
the brute force of the army. Even his control over the state
finances proved inadequate. The vizier, it would seem, required a
military command of his own.35
These limitations notwithstanding, for a brief period during the
late ninth and early tenth centuries the vizierate reached its
pinnacle of administrative and political power, even nominating
caliphs. From that time a number of theoretical tracts justify the
office and set down guidelines for its holders. Mawardi, for
example, distinguished two forms of vizierate, the vizierates of
delegation and of execution, and gave them a Koranic justification.
He advised the vizier to
32 Ibid., pp. 41-43. 33 Ibid., pp. 59-6I, 65-66. 34 Ibid., Vol.
II (Damascus, 9I60), pp. 565-568. 35 Ibid., pp. 6I5-656,
664-667.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
512 Joel Shinder
protect the imdm-caliph and the umma, and he cautioned the
sovereign to maintain control over the vizier and prevent the
creation of an equal and rival power in the state. No legal
framework to secure this worthy end of checks and balances was
outlined. Consequently, in practice the office depended on the
arbitrary will of the sovereign who created it.36
Under the Great Seljuk sultans, the political successors to the
Abbasids, the institution of the vizierate suffered the same
weaknesses. The military acted independently of the vizier, the
cooperation of sultanate and caliphate was seldom realized, and the
attempt to link religion and state through the madrasa system of
higher education for public servants failed. The vizier's patronage
function resulted in factionalization, and the vizier's hold over
the army through his control of the treasury disappeared when
direct support by land grants was instituted. Finally, the sultans
often went over the heads of their ministers by consulting directly
with subordinates in the bureaus.37
As an institution of government the Abbasid-Seljuk vizierate was
not a homogeneous unit ready for Ottoman adoption. From an advisory
post it became an office directing much but not all of civil
administration. Its base was the discretionary power of the
sovereign, sultan, or caliph. In its most centralized form of the
Seljuk period it still had no substantial control over the military
forces of the state, so that even the personal prestige of a
particularly able vizier only served to excite the envy and fear of
the military commanders. The Otto- mans certainly accepted the
title of vizier for their head of government and other officers of
state. They did not assume the burden of that office's full and
turbu- lent history. Neither continuity nor evolution characterizes
that history. Neither continuity nor evolution may be presumed for
the Ottoman vizierate. The themes and precedents which have
informed the study of Ottoman administra- tive history are proved
misleading or fallacious. Some of the circumstances which shaped
the early Ottoman experience in administration support this
assertion.
The structure of the Ottoman state under the first two princes,
it has been stated, was similar to that of the other Anatolian
principalities, successors to the Seljuks of Rum and dependents of
the Ilhanid empire. Territory in the princi- palities was divided
among the sons of the prince and other family members. Each member
acted at will in his own lands. The leadership was elective, and
the headman was merely primus inter pares. It is thoroughly
appropriate, then, that the Ottomans are known as such, Osmanhs,
followers of Osman and his line. Osman's reign, however, was not
very remarkable. His property at death reportedly consisted of a
province, but no gold or silver; a robe, some armor, a salt cellar,
a spoon-holder, some houseboots, several horses, some sheep, a few
oxen, and nothing more.38 The spectacular expansion of the family's
holdings
36 Ibid., pp. 715-7I7. 37 Carla L. Klausner, The Seljuk
Vezirate: A Study of Civil Administration, io55-II94 (Cambridge,
Mass., 1973), pp. 26, 40, 44-45, 88.
38 'Ali, Apz., pp. 36-37.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 513
commenced in the reign of Orhan. Ibn Battuta describes him as
the greatest of the Turkmen kings and the richest.39 Both he and
Gregory, Archbishop of Thessalonica (I347-I360), however, agree
that Orhan preferred the outdoors to the foul air and crowds of the
cities and towns he held. He never gave up his tents for the sake
of a palace, a sure sign of vigor in the mind of Ibn Khaldun,
another commentator.40
At some obscure point the Ottomans stopped being simple heads of
semi- nomadic tribes with winter and summer pasturages and became
governors of an organized frontier march. Historians for the house
developed a legitimacy argument that involves Seljuk nomination of
Osman as an independent governor. Effective nomination, of course,
would have come from the Ilhanids, but the degree of central
control over the Anatolian periphery can in fact be minimized.
Vassal or independent ruler, Osman's actions bespeak the absence of
central constraints. His military feats were not limited to the
ghazi struggle against Byzantium but included as prey long-held
Muslim lands. The Muslim Ottomans occupied the land of other Muslim
princes, who menaced each other as much as Byzantium, if not more
so.41 It would be a mistake, however, to conceive of the Ottomans
at this early date as a highly centralized state itself. The
conquest of Thrace, for example, was not accomplished by a
well-oiled, centrally directed and thoroughly coordinated military
machine matched by an effective adminis- trative apparatus.
Non-Ottoman and Ottoman leaders acted on their own to set up
political entities in Thrace. These had a Greek-speaking peasant
base, kadis, regional commanders (subasz) and army judges. Yakub,
Haci ilbeyi, and Siiley- man (an Ottoman scion) were among the more
prominent of these Thracian entrepreneurs. After Siileyman's death
no Ottoman was sent to replace him, but the conquest of Thrace
continued.42
On emerging from the black hole in their history, ca.
1364-1381,43 the Ottomans are seen to have begun their steep ascent
into the ranks of the mighty. Hidden in this lacuna, however, is
the formation of the janissary corps, the Thracian take-over, and
the development of an administration. When the murkiness clears
Murad I (I360-1389), Orhan's successor, appears as grand emir or
prince, possibly signifying his success in gaining control over the
independent agents in Thrace who felt threatened by Byzantine and
south Slav pressures. The janissary corps may well have been
initiated with this end in view.44 The final step in the Ottoman
assertion of unitary rule in their house was taken by Bayezid I
when he assumed the attribute hiimayun, imperial, which
39 H. A. R. Gibb, trans., The Travels of Ibn Battuta A.D.
1325-1354, Hakluyt Society, Series II, Vol. CXVII, Vol. II
(Cambridge, I962), p. 452.
40 G. G. Arnakis, 'Gregory Palamas among the Turks and Documents
of His Captivity as Historical Sources,' Speculum, 26 (1951),
II3.
41 Beldiceanu-Steinherr, Recherches, pp. 70-7I. 42 Ibid., pp.
45-47. 43 The lack of material for this period was first noted by
Franz Babinger, Beitrdge zur
friihgeschichte der Tiirkenherrschaft in Rumelien (14-15
Jahrhundert) (Munich, I944), p. 76.
44 Beldiceanu-Steinherr, Recherches, pp. 47-48, 241 and n.
2.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
514 Joel Shinder
his successors Musa and Mehmed I continued to use.45 While an
elevated attribute, a loyal military force and preeminency among
the princes of the region are all manifest by the end of the
fourteenth century, the growth of a central administration is
masked. Where and who are the Ottoman missi dominici?
According to the ghazi thesis of Paul Wittek, Ottoman success
was dependent on the continguity with Byzantium and the appeal to
two diametrically opposed groups, the ghazi warriors of the faith
and the representatives of high Islamic urban civilization, the
ulema. The hinterland was considered the source for the latter, the
only social group deemed capable of stabilizing the ghazi state.
The ulema, with their knowledge of Islamic principles and methods
of administra- tion, were thought to have brought to the Ottoman
capital at Bursa a pacific and tolerant Islam which guaranteed
subject peoples communal independence and a structured system of
taxation, protecting them from the depradations of booty- starved
ghazis. Artisans and merchants of the akhi fraternal organization
followed these chieftains of law and order and assured the Ottomans
a high level of productivity. The fraternal order was the first
victim of the new society. Mem- bers with secular interests entered
guild organizations (their origins unexplained), and those with
otherwordly concerns followed the dervish orders of mystics. The
ghazis were the next to suffer the jealousies of Osman's heirs, who
identified themselves with the ulema.46
On reconsideration, the Ottoman effort throughout was not so
much to harmonize tendencies as to ensure that favorable ones
prevailed. The creation of the janissary corps was a hedge against
overdependency on the independent forces of the other princes and
tribesmen. An independent and quasi-religious political, social,
and economic organization such as the akhis could not be tolerated.
But Ottoman relations with their ulema were not completely
thorough, either. Early Ottoman support in the way of pious
endowments was much greater for the popular religious fringes than
for the hinterlanders.47 The first chronicles exaggerate corruption
among the ulema in administrative posts, especially during the
reign of Bayezid I, and intimate that a blind eye was turned on
their abuses of authority. Bayezid, however, regulated the fines
and fees that officials were allowed to levy in their legal
proceedings.48 The practices of accumulating wealth, accounting for
it, and creating a treasury to hold it cannot entirely be laid at
the doorstep of the ulema alone. The initiative in structuring the
ulema into an administrative cadre came rather from the Ottoman
court itself. The Ottomans, therefore, pursued an articulated
policy which employed
45 Paul Wittek, 'Zu einigen friihosmanischen Urkunden, II,'
Wiener Zeitschriftfiir die Kunde des Morgenlandes, 54 (1958),
244.
46 Paul Wittek, 'De la defaite d'Ankara a la prise de
Constantinople,' Revue des Etudes Islamiques, 12 (1938), 5, 12-I3.
See also G. G. Arnakis, 'Futuwwa Traditions in the Ottoman Empire:
Akhis, Bektashi Dervishes, and Craftsmen,' Journal of Near Eastern
Studies, 12 (I953), 237, 240.
47 See the lists of monuments in Ekrem Hakki Ayverdi, Osmanlz
Mimarisinin Ilk Devri, Vol. I (Istanbul, I966), passim.
48 'Ali, Apz., pp. I97-I98; Giese, Apz., pp. 30, 42.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 515
all human resources at hand without any kind of ideological
orientation reflected in such terms as ghazi, hinterlander, ethical
fraternity, or mystical order. In their centralist policy the heads
of the house found ulema supporters, among others. Candarli Kara
Halil Hayruddin Pasa is the outstanding example of an
'alim-turned-administrator and commander.49 The ulema was one
important source of recruitment for administrators, but it was not
the only one. A vizier named Bayezid Pasa was originally a sipahi,
a cavalryman.50 There was also a standing Byzantine scribal class
which has to be considered along with the non- Byzantine Balkan
scribal classes, most notably the Hellenized south Slavic cadres of
the Serbian period. A contemporary observer of the early Ottomans
described the Ottoman clerks as 'iaziti' (Turkish, yazici). These
wrote in several scripts (and languages?) and included Christian
secretaries both at court and in the provinces. Their employment in
the provinces was primarily directed to the preparation of
cadastral surveys for the fisc.51 Since the Ottomans frequently
preserved features if not the bulk of pre-Ottoman Muslim and
non-Muslim law codes and customs, such scribal groups were
important additions to the state apparatus. They had an immediate,
direct influence which was not always that of 'the traditional Near
Eastern State' on the development of Ottoman adminis- tration.
After a conquest the scribes employed to write up the cadasters
were often men with local knowledge, themselves natives of the
places recorded and, therefore, recent additions to the broad
ruling class. One such scribe was a slave of an Ottoman general.
Another scribe rose to a field command post himself.52 Just as the
ulema did not monopolize scribal positions and traditions, neither
can they clearly be set apart as hinterlanders. In Rumeli (the
European province), especially, the ghazis under their various
independent leaders elected their own kadis to handle such
administrative questions as inheritance and taxation. Thus,
frontier Turks, not all of whom entered Europe by way of Muslim
Iran and Asia Minor, infiltrated the ranks of the incipient
'Ottoman' ulema.53
Administration in the formative period of the state was
primarily concerned with the sultan's financial claims. The basic
unit of rural exploitation was the timar, the usufruct of which
supported both civil and military servants of the Osmanli household
- the state, so to speak.54 Ottoman scribes were responsible for
developing a systematic means of registration and assignment not
only for lands falling under the timar regime but also for the
lands, forests, fisheries,
49 The importance of this family in the formative period of the
Ottoman state is out- lined in Franz Taeschner and Paul Wittek,
'Die Vezirfamilie der Candarlyzade (I4.-I5. Jah.) und ihre
Denkmiler,' Der Islam, 28 (1929), 60-115.
50 Ibid., p. 95. 51 Speros Vryonis, Jr., 'The Byzantine Legacy
and Ottoman Forms,' Dumbarton Oaks
Papers, nos. 23 and 24 (I969-I970), 275-276. 52 Halil Inalclk,
Hicri 835 Tarihli Suiret-i Defter-i Sancak-i Arvanid (Ankara,
1954), pp
xvii-xviii. 53 Giese, Apz., p. 50; Franz Babinger, ed., Die
Friihosmanischen Jahrbiicher des Urudsch.
Nach den Handschriften zu Oxford und Cambridge ... (Hanover,
1925), pp. I2-I3. 54 Irene Beldiceanu-Steinherr, 'Un Transfuge
qaramanide aupres de la Porte ottomane:
reflexions sur quelques institutions,' Journal of the Economic
and Social History of the Orient, 26 (I973), I63-I64.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
516 Joel Shinder
mines, markets, and customs posts excluded. This they did using
every reason- able source or method at hand in the newly acquired
regions. The system was not so thorough in its uniformity as in its
centralization, for the proprietary rights of the emergent state
and the usufructuary rights of individuals had to be protected. The
regulations that protected the rights of proprietor and tenants and
established the obligations of all persons holding any title to
sources of revenue came to be known as the sultan's kanun, his
executive law, a law of expediency and custom. Collections of
kanuns for particular regions prefaced the appropriate cadastral
registers. More general regulations were collected into kanunnames,
a form of codification for the use of the public, especially the
kadis responsible for applying both kanun and sharz'a in their
courts. From the language of Mehmed II's kanunname it can be
inferred that the exemplars derive from Bayezid I or Mehmed I. In
the reign of Murad I individual questions of proprietorship-tenancy
probably were not recorded in writing.55 There are references,
however, to regulations for this period. A tentative conclusion is
that Murad I was the first sultan to make pronouncements which
stipulated the varieties of land tenure and the obligations these
entailed.56 If this is indeed the case, then the inception of true
Ottoman central administration rests in Murad's reign.
Important clues to the first cadasters produced by the
administration come from the earliest surviving cadastral register
bearing the date 835/I431. It displays a primitive chronology in
the introductory section, which refers to an asil defter, the
original or source register, from the period of Mehmed I. The
inference from the phraseology is that the first register was
completed under Mehmed's father, Bayezid I, in whose reign come the
first complaints in the chronicles of bureaucratization.57 Umur
Beg, the official responsible for the composition of the register,
was the son of a foot soldier who served under Murad I and Bayezid
I. Yusuf, Umur Beg's scribe, prepared the register in tevki' script
and not in siyakat, as is customary in later registers.58 In every
other respect, however, the format is essentially that of the
imperial age. The register and chancery documents of the period are
indicative of continued development. The principles of Ottoman
accounting and the instruments for the certification of rights to
individuals are not as yet fixed. There is no heavy-handed
tradition of administration and statecraft displayed here, nor is
there a homogeneous body of hinterland civil servants perpetuating
such a tradition. That tradition, will be manufactured in the
sixteenth century by chroniclers of the ulema class rather, seeking
for themselves and their colleagues greater influence in
government.
One is nonetheless left with a final question. Was there some
cohesiveness to the emergent administrative class? If a rudimentary
but normal mode of pro-
55 Beldiceanu-Steinherr, Recherges, p. 245. 56 Babinger,
Beitrdge, p. 58. 57 Inalcik, Hicri, p. xv. Refer back to the
earlier comments on the chroniclers' views of
Bayezid I's policies. 58 Ibid., pp. xii-xiii, xvii.
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
-
Early Ottoman Administration in the Wilderness 517
gression can be suggested for the highest administrative
positions as described in the chronicles, it would look something
like: mosque professor; kadi of Bursa; kadiasker, the
judge-advocate of the army; nisancz, the head of the chancery and
keeper of the seal; vizier; and, finally, grand vizier. For the
early period as seen in the chronicles the ulema career seemed to
lead directly to the highest office in the realm below that of
sultan. Cohesiveness would seem to rest in the prestige of the
ulema class. So little is known of the early viziers, however, that
such a construction is really quite unwarranted. What made a vizier
a grand vizier, when and who the first grand vizier was, are
questions that cannot be answered. The likelihood is great that
there was no clear differentiation of roles either by training or
by function during the formative period. High-ranking commanders
loyal to the house of Osman, ulema figures who shared abilities in
politics, war, and law and, finally, local but non-Ottoman elites
contributed to the executive classes in Ottoman government. The
administrative cadres beneath the highest figures were also mixed,
including Turks and non-Turks, scribes of pre-Ottoman states
(Muslim and non-Muslim), and prominent local figures who were not
always even literate! Administration may not have meant fulltime
employment for most of the personnel involved. The rate of
expansion probably exceeded the abilities of government to govern,
which helps explain the sine- wave course of Ottoman expansion in
Europe and Asia. Moreover, local initiative takes first prize in
Ottoman history, even in the periods of greatest centralization.
One would expect the absence of a cohesive administrative corps in
the early period.
It is erroneous, therefore, to think that the government of the
Ottoman principality was the microcosm of the empire's government
or a seed of the 'traditional Near Eastern State' planted in the
fertile soil of Asia Minor by horticulturists of the ulema class;
or that government itself simply resulted from the conflict of
ghazi and ulema, Osmanli slaves and free Turkish nobility. The
question was not which elements would succeed or how far 'Ottoman
State' would withdraw from 'Ottoman Society.' It was which ruler
and of what house would prevail. Ottoman survival following the
Battle of Ankara (I402) was certainly furthered by the
administrative abilities and capabilities of the dynasty's
servants, but these functionaries were far more flexible and
imaginative than the themes and traditions which have been advanced
to classify their achieve- ments. If the Ottoman example is any
indication, the comparative study of Islamic government is on a
most inauspicious course. Might one yet cleanse these Augean
stables without detriment to generalization and the usefulness of
comparison?
MINNEAPOLIS, MINNESOTA
This content downloaded from 128.103.149.52 on Sat, 01 Aug 2015
09:20:15 UTCAll use subject to JSTOR Terms and Conditions
Article Contentsp. 497p. 498p. 499p. 500p. 501p. 502p. 503p.
504p. 505p. 506p. 507p. 508p. 509p. 510p. 511p. 512p. 513p. 514p.
515p. 516p. 517
Issue Table of ContentsInternational Journal of Middle East
Studies, Vol. 9, No. 4, Nov., 1978Urbanization, Urbanism, and the
Medina of Tunis [pp. 431 - 447]Agricultural Output and Consumption
of Basic Foods in Egypt, 1886/87-1967/68 [pp. 449 - 469]The British
Occupation of Egypt: Another View [pp. 471 - 488]First Names and
Political Change in Modern Turkey [pp. 489 - 495]Early Ottoman
Administration in the Wilderness: Some Limits on Comparison [pp.
497 - 517]An Overlooked Problem in Turkish-Russian Relations: The
1878 War Indemnity [pp. 519 - 537]The Study of 'Complex Society' in
the Middle East: A Review Essay [pp. 539 - 557]The Zentrales
Staatsarchiv of the German Democratic Republic as a Source for Late
Ottoman and Middle East History [pp. 559 - 571]Book Reviewsuntitled
[pp. 573 - 574]untitled [pp. 574 - 575]untitled [pp. 575 - 576]