Fleet K. The Cambridge History of Turkey: Volume 1, Byzantium to Turkey, 1071-1453

Подождите немного. Документ загружается.

machiel kiel

town.

118

To the extensive foundations of

Ishak Bey in

Usk

up, S¸ihabuddin Pas¸a

in Filibe and Sultan Murad in Edirne and Selanik, all created in the 1430sand

early 1440s, should be added four mosques and mescids, three zaviyes, two cara-

vansarys, a hamam and two bridges, all of Gazi Turhan Bey, in various places

in Thessaly, but none now extant (the vakﬁye of these foundations was written

in 1446), as well as the monumental domed mescid of S¸arabdar Hasan Bey in

the castle of Vize from 1444 and the Yes¸ilce Cami, also known as

Imaret of

Mezid Bey, in Edirne, which was ﬁnished two years earlier.

119

This whole group of buildings marks the summit of a highly creative period,

in which Ottoman architecture deﬁnitively found its own code of aesthetics.

The year 1444 ended with the ‘Varna crisis’ when the huge crusader army under

King Wladislaw Jagiello and the redoubtable Jano

s Hunyadi put the whole

of Ottoman northern Bulgaria to the torch before being ﬁnally annihilated.

Another crisis took place in 1448 when the second battle of Kosovo ﬁnally

removed the threat of invasion from the west. These years were not very

productive for architecture. Only shortly before his death (1451) did Sultan

Murad again show conﬁdence in the future by starting a new palace on the

banks of the Meric¸ river outside Edirne. It was to be ﬁnished by his son and

successor, Mehmed II, with whom a new period begins.

The Ottoman architecture of the Balkans around 1450 shows that by then

the land had been wholly integrated into Ottoman culture. The plans for the

buildings and their whole conception came from north-western Anatolia as

an imported art, but the masonry often shows local inﬂuences.

120

The vast

118 For a plan and elevation of the building which has now disappeared see L. de Beyli

L’habitationbyzantine (Grenoble and Paris,1902), p. 71. In later local literature the opinion

that the great and speciﬁcally Ottoman work was of Byzantine origin was repeated

until recently. See also the excellent new study by Paschalis Androudis, ‘Historical

and Archaeological Evidence on the Great Karavan-Saray (B

uk Kervansarayı) of

Thessaloniki’, Makedonik

a 35 (2005–6), 63–97. For the Bey Hamamı see M. Kiel, ‘Notes

on Some Turkish monuments in Thessaloniki’, Balkan Studies 11 (1970), pp. 123–48,

reprinted and updated in Kiel, Studies. Good photographs and plans can also be found

in Anna Zombou-Asimi, ‘Bey Hamam, Thessaloniki, Greece’, in Secular Architecture in

the Balkans, ed. Slobodan

Cur

c and Evangelia Hadjitryphonos (Thessalonike, 1997),

pp. 318–21. Anna Zombou was the leader of the restoration of the great hamam. See

also Ayverdi, Osmanlı Mim

ısinde, ii,pp.536–7 (with the incorrect reading of the date

as 840, failing to recognise the s¸am

an (8) before the arba‘

ın).

119 Both the two latter buildings have long been in an incredible state of decay. The zaviye-

mosque of Mezid Bey was recently drastically ‘restored’, losing most of its original

features. For this building see Ayverdi, Osmanlı Mim

ısinde, ii,pp.397–400,

and Kuran,

Mosque,pp.126–8. For its vakıf property see G

okbilgin, Pas¸a Liv

ası,pp.243–4.

120 For local inﬂuences see Robert Ousterhout, ‘Ethnic Identity and Cultural Appropria-

tion in Early Ottoman Architecture’, Muqarnas 12 (1995), 48–62. See also the excellent

overview of Slobodan

Cur

c, ‘Architecture in the Age of Insecurity’, in

Cur

cand

Hadjitryphonos (eds.), Secular Architecture in the Balkans,pp.19–51.

190

Ottoman expansion into the Balkans

expanses of the Balkans were a new land for Islam, where all the institutions

it needed had to be imported since they did not exist in the local architectural

repertoire (mosques, medreses, hans, hamams, zaviyes, t

urbes and so on). It

might be suggested that this factor greatly helped to push the development

of Ottoman architecture forward and gave it more space for experimenting

than in the more traditional Anatolia. By 1450 the south-eastern Balkans had

become fully integrated in Ottoman culture, becoming a part of the Turkish

world. The census and taxation registers of the early period make this clear

for the ethnic and religious situation. The preserved monuments of Ottoman

architecture make the intensity of this integration visible to us still today.

191

Ottoman warfare, 1300–1453

al fodor

The formative period, c.1300–1370s

TheearlyOttomanmilitary organisation was a peculiar amalgam of Turkoman

nomadic, Seljuk-Ilkhanid and Byzantine elements, but it was the nomadic

tradition which predominated among the various constituent parts in the

early period of the formation of the state. As in other Eurasian nomadic

societies, it maywell have been very difﬁcult to distinguish between society and

military among the Turkomans living in the Bithynian marches. All followers

of the Ottomans capable of ﬁghting could and did participate in raiding or

in defence if need arose. The predatory raids were called akın in Turkish,

and those taking part in them were akıncı (later, as the Islamic character of

the state strengthened, the term gazi, that is, warrior of the faith, came to

be preferred).

Yet, apart from the undifferentiated bands of warriors, we see

from the outset a relatively small but well-organised and well-trained force

which gathered around the ruler and rendered services to him in war and

peace.

This type of ‘military retinue’ was not an Ottoman invention. It was a

universal institution which had existed in the east and the west since time

immemorial. In the empires of the Inner Asian Turks and Uighurs the retinue

of the kagan (ruler) was called buyruk, while the Karahanids and the Mon-

gols knew it under the names koldas¸ and n

or, respectively.

The same body

1 Muharrem Ergin, Dede Korkut Kitabı, vol. I: Giris¸ – Metin – Faksimile; vol. II: Indeks, Gramer

(Ankara, 1958–63), ii,p.8; Ahmedi, ‘Ah

med

ı’s History of the Ottoman Dynasty’, tr. Kemal

Silay, Journal of Turkish Studies16 (1992), 137,147;cf.P

al Fodor, ‘Ahmed

ı’s D

asit

an asa Source

of Early Ottoman History’, ActaOrientalia Academiae ScientiarumHungaricae 38, 1–2 (1984),

52–3;Doukas,Decline and Fall of Byzantium to the Ottoman Turks by Doukas, ed. and tr.

Harry J. Magoulias (Detroit, 1975), pp. 133–4. According to Ruhi (Halil Erdo

gan Cengiz

and Yas¸ar Y

ucel, ‘R

ıT

ıhi’, Belgeler 14, 18 (1989–92), 376), Ertu

grul was appointed

akıncı bas¸ı by Sultan Alaeddin.

2 Gy

orgy Gy

orffy, ‘Die Rolle des buyruq in der altt

urk

ischen Gesellschaft’, Acta Orien-

talia Academiae Scientiarum Hungaricae 11 (1960), 169–79; Mario Grignaschi, ‘Les guerri

ers

192

Ottoman warfare, 1300–1453

of warriors appeared among various western and eastern European ‘barbar-

ian’ peoples, many of whom had founded their own states by the end of the

ﬁrst millennium (Celtic vassus, German Trost or Gesinde, Russian dru

zina and

Scandinavian hird, for example).

The military retinue was a group of armed,

mainly free men (the majority of them foreigners), who served on a volun-

tary basis and were attached personally to the leader. They were his closest

companions, friends and servants; they commanded the troops in wars, while

a select group of them served as his bodyguard. Their livelihood was secured

by their master, predominantly from the booty acquired during incursions

and wars. The strength of these retinues ranged from a few dozen to 3,000

men. When the founders of the new states began to transform their personal

might into territorial power, they relied heavily on their military retinues, del-

egating them to and settling them on the territories they controlled. In this

Gefolgschaft-type of state, it is the military retinue to which the origins of the

formal institutions of power can be traced back.

The counterpart of this institution among the O

guz/Turkomans, the prede-

cessors of the Ottomans, was known as yoldas¸ and yi

git.

In addition, the chiefs

and their most renowned companions willingly assumed the title alp or alp-

eren (hero, brave man), an epithet of the distinguished nomadic adventurers-

warriors from the time of the ancient Turks.

With the ascendancy of the

Chinggisid states, there was a clear tendency to replace the term yoldas¸ by the

word n

or/n

oker, its Mongolian equivalent. In view of this we can regard it as

domestiques dans la f

eodalit

e turque’, in VI. T

urk Tarih Kongresi Ankara 20–26 Ekim 1961 .

Kongreye Sunulan Bildiriler (Ankara, 1967), pp. 210–11; B. Y. Vladimirtsov, Mo

golların

Ic¸tima

Tes¸kil

atı. Mo

gol G

oc¸ebe Feodalizmi,trans.Abd

ulkadir

Inan (Ankara, 1944), pp. 133–46.

3 Walter Schlesinger, ‘Herrschaft und Gefolgschaft in der germanisch-deutschen Verfas-

sungsgeschichte’, Historische Zeitschrift 176, 2 (1953), 225–75; Reinhard Wenskus, Stammes-

bildung und Verfassung. Das Werden der fr

uhmittelalterlichen gentes (Cologne and Graz,

1961), pp. 346–74; Otto H

otsch, ‘Adel und Lehnwesen in Rußland und Polen und ihr

Verh

altnis zur deutschen Entwicklung’, Historische Zeitschrift 108 (1912), 542–70; Franti

sek

Graus, ‘Die Entstehung der mittelalterlichen Staates in Mitteleuropa’, Historica 10 (1965),

38–53; Eric Hoffmann, ‘Knut der Heilige und die Wende der d

anische

Geschichte im 11.

Jahrhundert’, Historische Zeitschrift 218 (1974), 531, 550–3.

4 Faruk S

umer, O

guzlar (T

urkmenler), Tarihleri – Boy Tes¸kil

atı – Destanları (Ankara, 1972),

pp. 402–3; cf. Ergin, Dede Korkut Kitabı, ii,pp.338–9;Yas¸ar Y

ucel, Kadı Burhaneddin Ahmed

ve Devleti (1344–1398) (Ankara, 1970), p. 69;S¸inasi Tekin, ‘XIV. Y

uzyılda Yazılmıs¸Gazi-

lik Tarikası “G

azili

gin Yolları” Adlı Bir Eski Anadolu T

urkc¸esi Metni ve Gaz

a/Cih

Kavramları Hakkında’, Journal of Turkish Studies 13 (1989), 141 n. 9; Enveri, Le Dest

d’Um

ur Pacha (D

ust

urn

ame-i Enver

ı), ed. Ir

ene M

elikof

f-Sayar (Paris, 1954), pp. 76, 78.

5 M. Fuad K

opr

u, ‘Alp’, in

Isl

am Ansiklopedisi [hereafter

IA], vol. I (Istanbul, 1941), pp. 379–

84;M.FuadK

opr

u, Osmanlı

Imparatorlu

gunun Kurulus¸u, 2nd edn (Ankara, 1972), pp. 146–

53;S¸inasi Tekin, ‘T

urk D

unyasında Gaz

a ve Cih

ad Kavramları

Uzerine D

us¸

unceler. II.

ı Teriminin Anadolu ile Akdeniz B

olgesinde

Itib

arını Yeniden Kazanması’, Tarih ve

Toplum 19, 110 (1993), 77–8.

193

al fodor

natural that the early followers of Ertu

grul and Osman are repeatedly referred

to by such terms in the earliest Ottoman chronicles.

Osman’s entourage seems to have been heterogeneous, consisting of Turks

(his closest, permanent followers and warriors from neighbouring lands who

foughton a temporary basis), Byzantines, Catalans and – in small but increasing

number – slaves.

His retinue may have numbered a few hundred men at

most. His allies and semi-independent companions also had their own retinues

and sometimes acted on their own account. Apparently, they formed several

smaller ‘marches’ within the larger Bithynian frontier.

When they engaged in

large-scale military adventures or faced enemy attacks, they joined their forces

including the tribal warriors. This happened in 1301–2 at Bapheus and in 1329

at Pelekanon when possibly the entire force which could be mobilised from

the Ottoman territory fought against the Byzantines.

At that time the Ottoman army consisted largely of mounted warriors

with some supplementary infantry. The poet As¸ık Pas¸a, who clearly regarded

details of a warrior’s weapons as important, portrayed the alp of his time

in his Garibname (1330) as a soldier having a good horse, carrying a keen-

edged sword (kılıc¸), bow (yay), arrows (ok) and lance (s

ung

u), and bearing

an impenetrable and awe-inspiring coat of mail (ton).

As the conduct of the

two battles mentioned above clearly shows, early Ottoman warfare was based

largely on the ancient steppe tradition. Its principal elements included the

avoidance of close combat, placing the mobile mounted troops in ambush, the

feigned retreat with incessant shooting in order to disrupt the enemy’s battle

array, and repeated sudden attacks until the enemy became totally confused

and ﬂed.

The greatest challenge was the conquest of the Byzantine fortresses

6 As¸ıkpas¸azade, Tev

ıh-i

Al-i Osman,inOsmanlı Tarihleri, vol. I, ed. N. Ciftc¸io

glu,

(Istanbul, 1949), pp. 107, 112 (n

or), pp. 98, 100, 105–6, 108 (yoldas¸); Mehmed Nes¸ri,

Kitab-ı Cih

an-n

a, ed. Faik Res¸it Unat and Mehmed A. K

oymen, 2 vols. (Ankara, 1949,

1957), i,pp.120–1, 146–7; cf. Halil

Inalcık, ‘Osmanlı Tarihine Toplu Bir Bakıs¸’, in Osmanlı,

vol. I, ed. G

uler Eren, Kemal C¸ic¸ek and Cem O

guz (Ankara, 1999), pp. 51–5. At the same

time both alp and yoldas¸ had a religious-mystical connotation meaning spiritual devo-

tion (‘companionship’) to God and Suﬁ leaders; cf. Mine Mengi, ‘Garib-namede Alplık

Gelene

giyle

Ilgili Bilgiler’, Belleten 48, 191–2 (1984–5), 488–95;

Elvan C¸ elebi, Men

akıbu’l-

Kudsiyye F

ıMen

asıbi’l-

Unsiyye (Baba

Ilyas-ı Horas

ıveS

alesinin Menkabev

ı Tarihi), ed.

Ismail E. Er

unsal and A. Yas¸arOcak(Ankara,1995), for example, pp. 25/285, 58/653,

59/658, 82/936, 156/1822.

7 Cemal Kafadar, Between Two Worlds: the Construction of the Ottoman State (Berkeley, Los

Angeles and London, 1995), pp. 122–38; Heath W. Lowry, The Nature of the Early Ottoman

State (Albany, 2003), pp. 55–94, 114–43.

8 As¸ıkpas¸azade, Tev

ıh,pp.99–100, 109;N

es¸ri, Kitab-ı Cih

an-n

a, i,pp.128–9.

9 Mengi, ‘Garib-namede’, 483–7.

10 Johannes Kantakouzenos, Geschichte, ed. Georgios Fatouros and Tilman Krischer, 2

vols. (Stuttgart, 1986), ii,pp.21–35; cf. N. Jorga, Geschichte des osmanischen Reiches nach

194

Ottoman warfare, 1300–1453

and fortiﬁed towns. In the earlier phase the Ottomans blockaded a town or

fortress for years or even decades, setting up stockades and watch towers, and

occupying and, when necessary, devastating the surrounding countryside.

It was during these prolonged blockades that villages or other sources of

income began to be granted in order to provide for the subsistence of the

retainers who were constantly under arms.

These assignments, which were

allocated for the ‘nourishing’ or ‘feeding’ of the warriors in the strict sense,

were called timars, a Persian term meaning care, forethought or solicitude,

and the holder was called a timar eri. In this period the timar was an undiffer-

entiated, general form of remuneration given to the grantee to enable him

to make a living. This early stage is reﬂected in the ﬁfteenth-century written

sources where the concept of timar-holding is most often designated by the

term ‘to eat a timar’ (timar yemek).

It is worth noting that the armed ret-

inue was similarly called ‘fed’, ‘nourished men’ (threptos anthropos) among the

Danubian Bulgars, and in the Kievan Rus state the princely retainers, who

were nominated governors, were assigned portions of princely revenues for

‘nourishment’ (kormlenye).

Indications of the fact that the ﬁrst timar-holders

were chosen from among the members of the leaders’ retinues can also be

detected in the ﬁfteenth-century timar registers. A grant was most often made

on the basis of what was referred to as yoldas¸lık which alluded to participation

in a campaign and the display of bravery, although originally – as was seen

above – it only denoted companionship and the performing of (excellent)

service as a retainer of a military leader.

The origins of the timar system are still being disputed. Some scholars

argue that it derives from the Seljuk ikta, others from the Byzantine pronoia.

den Quellen dargstellt. Erster Band (bis 1451), vol. I (Gotha, 1908), pp. 165–7;RudiPaul

Lindner, Nomads and Ottomans in Medieval Anatolia (Bloomington, 1983), pp. 25–6, 30–2.

11 Colin Imber, The Ottoman Empire, 1300–1650: the Structure of Power (Handmills and New

York, 2002), pp. 254–5.

12 As¸ıkpas¸azade, Tev

ıh,p.105;Nes¸ri, Kit

ab-ı Cih

an-n

a, i,pp.122–3, 158–9.Thereis

documentary evidence showing that this practice existed as early as under Orhan; see

Nicoar

a Beldiceanu, Le timar dans l’

etat ottoman (d

ebutXIVe–d

ebutXVIe si

ecle) (Wiesbaden,

1980), p. 25.

13 Cf. e.g. Halil

Inalcık (ed.), Hicri 835 Tarihli Suret-i Defter-i Sancak-i Arvanid (Ankara, 1987),

passim.

14 V. B e

vliev, Die protobulgarischen Inschriften (Berlin, 1963), pp. 282–3; Igor Iakovlevich

Froianov, Kiievskaia Rus. Ocherki socialno-politicheskoi istorii (Leningrad, 1980), pp. 88–9;

Uwe Holbach, Der russische F

urstenhof vor 16. Jahrhundert (Stuttgart, 1985), p. 100.

15 The notion of yoldas¸lık is ﬁrst attested to in a document of 1391;Lowry,The Nature,p.62.

In the early timar registers its synonyms are erlik and s¸irmedlik.

16 Gyula K

aldy-Nagy, ‘The First Centuries of the Ottoman Military Organization’, Acta

Orientalia Academiae Scientiarum Hungaricae 31, 2 (1977), 147; Mark C. Bartusis, The Late

Byzantine Army: Arms and Society, 1204–1453 (Philadelphia, 1992), pp. 184–5; Imber, The

Structure of Power,pp.194–6.

195

al fodor

Whichever might prove true, the ﬁrst timars were apparently hereditary, as

were the appanages and estates (m

ulk, vakıf ) assigned to family members, allies

and military commanders.

The principle and practice of heritability were not

alien either to ikta of the Mongolian period, or to pronoia of the Palaeologan

era.

The size of the timars must have varied widely for, if credence can be

given to the later chroniclers, prominent leaders could obtain entire former

principalities as timars.

The ruler and the major ﬁgures in the state had continuously to make up

for their lost retainers who had settled on their timars. At the same time,

the protracted sieges of Byzantine cities called for the creation of an efﬁcient

infantry. This dual challenge was met by Orhan, probably around 1330, when

he set up an army of foot soldiers called yaya (literally ‘footman, infantry’)

recruited from the people in his own territory. Owing to the lack of con-

temporary sources, we have to fall back on ﬁfteenth–sixteenth-century tradi-

tion concerning these troops. The earliest material maintains that the yayas

were the ruler’s own private (has) soldiers or armed escorts who replaced the

yoldas¸es and wore white caps (ak b

ork) to differentiate them from the retain-

ers of other lords and the tribal warriors who all wore red caps.

Although

another, later, stratum of the tradition claims that the yayas were recruited

from among the tax-paying peasants (reaya), who served on a voluntary basis

in return for exemption from various taxes and were given minimal pay (one

or two akc¸e a day) during warfare, it is almost certain that this scheme reﬂects

the realities of the ﬁfteenth century projected with some distortions into

the past.

At any rate, the transformation must have been rapid and radical

because the geographer al-‘Umari – undoubtedly exaggerating – reported in

around 1331 that Orhan had 25,000 or 40,000 mounted troops and innumerable

infantry.

17 Cf. As¸ıkpas¸azade, Tev

ıh,p.104.

18 Alessio Bombaci, ‘The Army of the Saljuqs of R

um’, Annali Istituto Oriente di Napoli 38,

28 (1978), 365; Bartusis, The Late Byzantine Army,pp.179–82.

19 Nes¸ri, Kitab-ı Cih

an-n

a, i,pp.166–7; Feridun M. Emecen,

Ilk Osmanlılar ve Batı Anadolu

Beylikler D

unyası (Istanbul, 2003), p. 94.

20 Oruc¸, Die fr

uhosmanischen Jahrb

ucher des Urudsch, ed. Franz Babinger (Hanover, 1925),

pp. 15–16, 89–90;As¸ıkpas¸azade, Tev

ıh,pp.117–18;Nes¸ri, Kitab-ı Cih

an-n

a, i,pp.152–7.

21 K

emal Pas¸a-o

glu S¸ems

uddin Ahmed

Ibn-i Kemal, Tev

ıh-i

Al-i Osman, vol. ii. Defter, ed.

S¸erafettin Turan (Ankara, 1983), pp. 49–51; Saadeddin, T

u’t-tev

arih, vol. I. (Istanbul,

1279/1862–3), p. 40; cf. Muzaffer Arıkan, ‘Yaya ve M

usellemlerde Toprak Tasarrufu’, in

Atat

urk Konferansları

viii

. 1975–1976 (Ankara, 1983), pp. 179–80; Halime Do

gru, Osmanlı

Imparatorlu

gu’nda Yaya–M

usellem–Taycı Teskil

atı(XV.veXVI.Y

uzyılda Sultan

u Sanca

gı)

(Istanbul, 1990), pp. 10–11.

22 Al-‘Umari, Al-‘Umari’s Bericht

uber Anatolien in seinem Werke Mas

alik al-abs

ar f

ımam

alik al-

ams

ar, vol. I: Text, ed. Franz Taeschner (Leipzig, 1929), pp. 21, 41; cf. Ibn Battuta, Voyages

196

Ottoman warfare, 1300–1453

Although there are hardly any reliable data available for the ﬁrst half of

the fourteenth century, a comparison between the thirteenth-century Seljuk

and ﬁfteenth-century Ottoman organisations suggests that the early Ottoman

military administration and territorial division followed the Seljuk model.

One of the key ﬁgures of the Seljuk military organisation during the Mon-

gol domination was the subas¸ı (its Arabic and Persian synonyms being zaim,

s¸ihna, ser-les¸ker, server), who was consistently also called bey (amir in Arabic),

the ofﬁce he ﬁlled being called zeamet, emaret, serveri or ser-les¸keri.Thesubas¸ı

bey or zaim governed a territorial unit called a vilayet, or a larger city, and

was the commander (zaim

u’l-c

uyus¸) of the ikta-holder cavalry (sipahi) and the

fortress garrisons (mustahfız) under his authority.

Comparing the informa-

tion provided by a document of 1348 with data concerning subas¸ıs in ofﬁcial

registers from the ﬁrst half of the ﬁfteenth century and the early chronicles,

we ﬁnd that the Ottoman territorial division was the immediate successor to

the Seljuk system.

Subas¸ı beys(zaim

u’l-c

uyus¸) remunerated by large timars

were appointed to lead the vilayets or other territorial units of similar size (il).

It was their duty to command the timar eri cavalry (also called sipahis after

the Persian model) in military operations, while they also played a role in civil

administration. S

uleyman, son of Orhan, who appeared without any title in

d’Ibn Batoutah, vol. II: Texte, ed. and tr. C. Defr

emery and R. Sanguinetti (Paris, 1877),

pp. 321–2: ‘Ce sultan est le plus puissant des rois Turkomans, le plus riche en tr

esors, en

villes et en soldats.’

23 Osman Turan, T

urkiye Selc¸ukluları Hakkında Resm

ı Vesikalar. Metin, Terc

ume, Aras¸tırmalar

(Ankara, 1988), documents nos. 4–11, 16–17, 20–1; cf. Claude Cahen, La Turquie pr

ottomane (Istanbul and Paris, 1988), pp. 197–201. The institution continued in the Aydın

beylik too: Eﬂaki, Les saints des derviches tourneurs, ed. and tr. Cl. Huart, 2 vols. (Paris,

1918), ii,pp.391–2; cf. Himmet Akın, Aydıno

gulları Tarihi Hakkında Bir Aras¸tırma (Ankara,

1968), pp. 15, 30; Enveri, Le Dest

an d’Um

ur Pacha,pp.50, 64, 116. For the traces of its survival

in the state of Kadı Burhaneddin, see Y

ucel, Kadı Burhaneddin Ahmed,pp.44, 65–6, 102.

24 T

he document has been published several times (cf. the references in I. Beldiceanu-

Steinherr, Recherches sur les actes du r

egnes des sultans Osman, Orkhan et Murad I (Munich,

1967), p. 106), e.g. by Gli

sa Elezovi

c, Turski spomenici, vol. I, Parts 1–2 (Belgrade, 1940–

52), part 1,pp.1110–12,part2,pp.218–20; the timar registers used are:

Inalcık, Hicri 835

Tarihli, passim and p. xxiii;Hazim

Sabanovi

c, Kraji

ste Isa-bega Ishakovi

ca. Zbirni katastarski

popis iz 1455. godine (Sarajevo, 1964); Nikolai Todorov and Boris Nedkov, Fontes Turcici

Historiae Bulgaricae, series XV–XVI, vol. XIII/II (Soﬁa, 1966), pp. 11–41, 119–231; Hamid

Had

zibegi

c, Adem Hand

c and E

sref Kova

cevi

c, Oblast Brankovi

ca. Op

sir

ni katarski

popis is 1455 godine, 2 vols. (Sarajevo, 1972); Du

sanka Bojani

c-Luka

c, Vidin i vidinskijat

sand

zak prez 15–16 vek (Soﬁa, 1975), facs. pp. 1–111; Melek Delilbas¸ı and Muzaffer Arıkan,

Hicr

ı 859 Tarihli S

uret-i Defter-i Sancak-ı Tırhala, 2 vols. (Ankara, 2001); from among the

narrative sources see particularly Nes¸ri, Kitab-ı Cih

an-n

a, i,pp.224–7, 242–3, 262–3,

276–7; for some further occurrences of the term in early documents: Lowry, The Nature,

pp. 88–9 (1365–85); N. Beldiceanu, Le timar,pp.26 (1397–8); Paul Wittek, ‘Zu einigen

uhosmanischen Urkunden (I–VII)’, Wiener Zeitschrift f

ur die Kunde des Morgenlandes 54

(1958), 240 (1412)a

nd55 (1959), 125–6, 129 (1402: s¸ihne).

197

al fodor

a document of 1324 as a witness,

was described as zaim

u’l-c

uyus¸ ve’l-asakir

and had also gained the title of pas¸a in the source of 1348 cited above. All

this appears to conﬁrm that later chroniclers mirrored the reality of the age

faithfully, ascribing the titles bey, subas¸ı and pas¸a repeatedly to the noted yaya

commanders and marcher lords (for instance, Evrenos) of the second half of

the fourteenth century.

During the second third of the century, however, a new military-

administrative unit was taking shape as a new organisational level above the

vilayetsorsubas¸ılıks: the sancak (district or sub-province). Originally, the word

meant ﬂag or banner, and it designated an army unit or company without

territorial connotation. Some scholars argue that the prototype of the territo-

rial sancak is to be found in Orhan’s grant of Bursa and its vicinity to his son

Murad as ‘the ruler’s district’ (bey sanca

gı) after the capture of

Iznik (1331).

Two factors may have played important roles in the subsequent multiplica-

tion of sancaks. First, there was a dynastic drive to curb the subas¸ıs with local,

familial or tribal bonds by means of new military leaders more strongly depen-

dent on the rulers. Second, probably as a result of military considerations and

for reasons of consolidation, there was a tendency to incorporate the newly

conquered Turkoman principalities into the Ottoman system by keeping their

original boundaries, and the larger sancak must have been better suited to this

purpose.

The rise of the classical Ottoman army, 1370s–1453

In the second half of the century, especially in the 1360s–70s, the Ottoman

military underwent fundamental changes. These, and the following, decades

witnessed the emergence and consolidation of the troops and structures that

determined Ottoman warfare until the middle or end of the sixteenth century,

and in some regards even much longer. In short, this was when the classi-

cal Ottoman war organisation emerged. We will discuss each of the main

components in turn.

25 Lowry, The Nature,p.76.

26 Nes¸ri, Kitab-ı Cih

an-n

a, i,pp.246–7, 256–7, 262–3; Tunca Kortantamer, Leben und

Weltbild des altosmanischen Dichters Ah

med

ı unter besonderer Ber

ucksichtigung seines Diwans

(Freiburg im Breisgau, 1973), pp. 141–2;cf.

Inalcık, Hicri 835 Tarihli,p.xxiv.

27 Nes¸ri, Kitab-ı Cih

an-n

a, i,pp.162–3;KemalPas¸a-o

glu, Tev

arih-i

Al-i Osman, ii,pp.48;

Emecen,

Ilk Osmanlılar,p.93.

28 Emecen,

Ilk Osmanlılar,pp.95–9.

198

Ottoman warfare, 1300–1453

The timar-holding sipahis

By the last two decades of the fourteenth century, the timar-holding sipahi

had become the backbone of the provincial army and of the whole Ottoman

military organisation. The sipahi (also referred to as timar eri, timar-hor, ehl-i

timar, s

uvari) was ﬁrst and foremost a mounted soldier who had to perform

military service in exchange for his prebend granted by the state (the grant was

valid as long as he met his obligations). The timariots constituted the largest

single unit among the various central and provincial troops. This was also due

to the fact that the ofﬁcers of other troops, of the peasant militias, auxiliary and

paramilitary troops, as well as a considerable portion of the court mercenaries

and garrison troops, were also incorporated into the timar system. There were

many civil ofﬁce-holders, too, who received timars in lieu of salary and many of

them were also obligated to go to war, in the same way as the sipahis proper.

By the ﬁfteenth century, therefore, the timar-holders were recruited from

the most heterogeneous sources of manpower. In the ﬁrst half of the century

therewasstill a signiﬁcant, but continuouslydecreasingnumber of old retainers

described in the sources as ‘ancient’ (kadimi ) or ‘of sipahi origin’ (sipahi asıllı)

among the timar-holders.

They inherited the service and often the estate, or

a part of it, over generations; sometimes even women or underage children

were entitled to the estate provided that there was a relative who took care

of them and someone who carried arms for them. The Muslim sipahis of the

conquered Anatolian principalities and the Christians of Byzantine-Slav origin

also belonged to this category; they had entered into Ottoman service in order

to keep at least parts of their privileges and old ikta, pronoia or bas¸tina estates.

The importance of the latter is evidenced by the following ﬁgures: 18 per cent

of all timars in Arvanid (Albania) in 1431–2 and 17 per cent of the timarsin

the district of Tirhala in Greece in 1454–5 were in the possession of Christian

sipahis.

The central power had long been determined to increase the timar-holders’

dependency and to complement or replace them with elements more closely

tied to it. The ﬁrst aim was ensured by the method of mass deportation:

29 Vera P. Mutafchieva, Agrarian Relations in the Ottoman Empire in the 15th and 16th Centuries

(Boulder, 1988), pp. 1–60; N. Beldiceanu, Le timar,pp.38–46.

30 On the term kadimi, see Nicoar

a Beldiceanu and P. S. N

asturel, ‘La Thessalie entre

1454/55 et 1506’, Byzantion 53, 1 (1983), 110–11.

31 Halil

Inalcık, ‘1431 Tarihli Timar Defterine G

ore Fatih Devrinden

Once Timar Sistemi’,

in IV. T

urk Tarih Kongresi Ankara, 1948 (Ankara, 1952), p. 137; Delilbas¸ı and Arıkan, Hicr

859Tarihli,p.xli; for a general survey of the subject, see Halil

Inalcık, Fatih Devri

Uzerinde

Tetkikler ve Vesikalar,vol.I(Ankara,1987), pp. 137–84; Halil

Inalcık, ‘Ottoman Methods

of Conquest’, Studia Islamica 2 (1954), 104–29.

199