This article was downloaded by: [109.121.11.15]On: 22 December 2011, At: 08:49Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Al-MasaqPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/calm20

Surveying the Aspect of the MedievalWest Anatolian TownJason T. Roche

Available online: 14 Dec 2010

To cite this article: Jason T. Roche (2010): Surveying the Aspect of the Medieval West AnatolianTown, Al-Masaq, 22:3, 249-257

To link to this article: http://dx.doi.org/10.1080/09503110.2010.522384

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae, and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand, or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.

Al-Masaq, Vol. 22, No. 3, December 2010

Surveying the Aspect of the MedievalWest Anatolian Town

JASON T. ROCHE

ABSTRACT The aim of this short article is modest: it means to fill a lacuna in scholarly

output by offering a concise and accessible survey of the physical structure of the typical

west Anatolian town in the High Middle Ages. Attempts to locate such a study meet with

disappointment. If one wishes to look through the eyes of medieval travellers in Anatolia,

whether they be merchants, pilgrims or soldiers, and discover what type of construction they

witnessed when approaching and entering a typical town, one is compelled to trawl through

a great number of specialist articles and monographs dealing with specific archaeological

sites or particular narrow periods of history. This laborious exercise will be made somewhat

redundant by a brief synthesis of the appropriate evidence which historians and

archaeologists have addressed and compiled since the late 1950s when attempting to

reconstruct the development of the Byzantine city. The article traces the slow development

of the typical Anatolian urban form and aspect from the late fourth century, through the

mid-seventh to mid-eighth centuries, and then through to a period of urban recovery until

the latter part of the twelfth century. The choice of periods separated by some 800 years is

not arbitrary: the physical character (and function) of the typical town began to change in

the late fourth century, and the form it obtained during the seventh and eighth centuries

continued to be the one retained (with inconsequential variations to the general pattern)

during the intermediate periods of Byzantine recovery

Keywords: Anatolia – towns; Byzantine empire – towns; Towns – in Byzantium;

Nicaea/Iznik, Bursa, Turkey

The aim of this short article is modest: it means to fill a lacuna in scholarly output

by offering a concise and accessible survey of the physical structure of the typical

west Anatolian town in the High Middle Ages. Attempts to locate such a study meet

with disappointment. If one wishes to look through the eyes of medieval travellers

in Anatolia, whether they be merchants, pilgrims or soldiers, and discover what type

of construction they witnessed when approaching and entering a typical town, one

is compelled to trawl through a great number of specialist articles and monographs

dealing with specific archaeological sites or particular narrow periods of history.

This laborious exercise will be made somewhat redundant by a brief synthesis of

the appropriate evidence which historians and archaeologists have addressed and

Correspondence: Jason T. Roche, Room A-308, Department of History, Faculty of Arts and Sciences,

Faith University, Buyukcekmece 345000, Istanbul, Turkey. E-mail: jason.t.roche@gmail.com

ISSN 0950–3110 print/ISSN 1473–348X online/10/030249-9 � 2010 Society for the Medieval Mediterranean

DOI: 10.1080/09503110.2010.522384

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011

compiled since the late 1950s when attempting to reconstruct the development

of the Byzantine city.1

We will first consider the contemporaneous documentary evidence which

scholars investigated for valuable insights into the form of the archetypal west

Anatolian town. Did contemporary terminology employed in narrative works

provide historians with accurate information, or were there, for example, official

state documents which proved more useful? After reflecting upon the value of a

philological approach to the aims of this article, we will assess the value of

archaeological data. Archaeologists have until recently tended to overlook the

medieval period, choosing instead to concentrate on unearthing the ruins of

antiquity. Clive Foss’s seminal work has done much to address this neglect.

Archaeological surveys in Lycia and Pamphylia, for example, have discovered

remains of medieval civic and ecclesiastical buildings. Others have uncovered

evidence of the proliferation of domiciles at famous ancient towns such as Ephesos

(modern Selcuk) and Sardis (near modern Salihi). However, Foss’s work on

Byzantine fortifications has had the greatest impact. Reliable narrative sources

provided him with the year of construction for named strongpoints. Following

surface surveys at such sites, he was then able to attribute types of mortar, varieties

of masonry and methods of construction, defence and decoration to a particular

limited period. The previously unknown construction chronology of other

fortifications could then be ascertained via analogy and historical probability.

Through Foss’s work, the construction date of a large number of previously obscure

walls and towers in a great part of western Anatolia has been established. More

recently, the work of archaeologists such as Chris Lightfoot have built on Foss’s

work whilst breaking new ground through excavations and other techniques.

Excavations in progress at Amorion (near modern Emirdag), for example, are

discovering vital information on the town’s history, such as its patterns of

settlement throughout the late Roman and Byzantine Ages.

This article will trace the slow development of the typical Anatolian urban form

and aspect from the late fourth century, through the mid-seventh to mid-eighth

centuries, and through to a period of urban recovery until the latter part of the

twelfth century. The choice of periods separated by some 800 years is not arbitrary:

the physical character (and function) of the typical town began to change in the late

fourth century, and the form it obtained during the seventh and eighth centuries

continued to be the one retained (with inconsequential variations to the general

pattern) during the intermediate periods of Byzantine recovery.

Greek hagiographies, literary works and historians of the Byzantine age

concerned themselves primarily with the court, church, army and the frontier.

They therefore provide little in the way of detailed specific information on any

particular fortress, town or city they may mention. Such a place may be called a

kastron or polis. A kastron was technically a castle; a polis was a city, equivalent to the

Latin urbs. But we cannot glean from this vocabulary if the place in question was

a castle or a flourishing city: the official and unofficial kastron and polis civic and

military forms and functions, and contemporaneous traditional assumptions

concerning their physical character and civic and military responsibilities, blurred

and changed over time reflecting an inconsistent lexis. Islamic literary geographical

1 This survey of primary and secondary sources does not profess to be inclusive. The aim is to offer a

synthesis of representative studies and evidence.

250 Jason T. Roche

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011

descriptions can be uniform and they tend to refer to Asia Minor during the

Byzantine period as a country of fortresses rather than cities. This is important

because, to Arab observers at least, most ‘‘cities’’ appeared to be little more than

castles, regardless of the terms used in Greek narrative sources.2

Because of such problems and contradictions in the narratives, historians

working over three decades ago turned to official documents such as the Notitiae

Episcopatuum when attempting to reconstruct Byzantine urban history. The Notitiae

Episcopatuum are lists of metropolitanates, bishoprics and archbishoprics

which were composed for purposes of protocol and used in courts and synods.

The council of Chalcedon in 451 decreed that every city or polis would be the

seat of a bishop, and consequently the concept of a polis became associated with a

bishop’s presence. As the number of sees on the lists remained near constant,

accordingly, it was argued, so must have the number of cities. It is now accepted

that the terminology in the lists represents an idiosyncrasy of church language

and does not actually tell us anything about the size and nature of a settlement.

When archaeological evidence is used in conjunction with written sources, one

frequently notices that the seat of a bishop, that is, a supposed city/polis identified in

the Notitiae Episcopatuum, was little more than rubble, and often at best, a hill-top

fortress.3

A purely philological approach to discovering the form of medieval centres of

habitation is clearly flawed. We must turn to archaeology for our answers. Ankyra

(modern Ankara), for example, was a theme capital and the literary record reflects

its commercial and military importance during the Byzantine age. Foss has revealed

that it contracted around the mid-seventh century to a ‘‘fortress-town’’ occupying

a site not much larger than its small citadel at 350�150 m. The recent series

of archaeological excavations at Amorion suggest that whilst its antique/late Roman

site was extensive (known as the ‘‘lower city’’), the main seventh/eighth-century site

contracted to the ‘‘upper city’’ and was comparable in size, form and function to

Ankyra. In spite of this, it is clear from the textual record that Amorion remained

a place of considerable commercial and military importance throughout the

Middle Ages.4

There is relatively little archaeological knowledge about most of the other

places within western Anatolia’s interior, but the transformation and contraction of

Ankyra’s and Amorion’s urban forms during this period were paralleled further

west. The studies cited here, which utilise both written and archaeological

evidence, indicate there are many more examples of formerly major urban centres

which underwent a similar transformation and contraction. The centralisation

of state power during the course of the later fourth, fifth and sixth centuries was

2 John Haldon, ‘‘The idea of the town in the Byzantine Empire’’, in The Idea and Ideal of the Town Between

Late Antiquity and the Early Middle Ages, ed. Gian Petro Brogiolo, Bryan Ward-Perkins (Leiden: Brill,

1999), pp. 1–23.3 George Ostrogorsky, ‘‘Byzantine cities in the Middle Ages’’, Dumbarton Oaks Papers, XIII (1959):

47–66; Wolfram Brandes, ‘‘Byzantine cities in the seventh and eighth centuries – different sources,

different histories?’’, in The Idea and Ideal of the Town, ed. Brogiolo and Ward-Perkins, 25–57.4 Clive Foss, ‘‘Late antique and Byzantine Ankara’’, Dumbarton Oak Papers, XXXI (1977): 27–87; Chris

Lightfoot, ‘‘The public and domestic architecture of a thematic capital: the archaeological evidence of

Amorion’’, in Byzantine Asia Minor (6th–12th cent.), ed. Stelios Lampakis (Athens: Institute of Byzantine

Research, 1998), pp. 303–320; Klaus Belke, Galatien und Lykaonien, Tabula Imperii Byzantini,

Osterreichische Akademie der Wissenschaften Philosophisch-Historische Klasse, 4 (Vienna, 1984),

pp. 122–125.

Surveying the Aspect of the Medieval West Anatolian Town 251

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011

generally followed by urban and rural devastation caused by constant warfare

during the period of the Persian wars in the first decades of the seventh century and

the Arab conquests of the mid-seventh to the mid-eighth centuries. The subsequent

economic and demographic decline was often exacerbated by forced migration and

natural disaster. In consequence, most of the places which in Classical Antiquity

were recognisably urban centres first began to lose their classical structures, and

then to contract to become medieval fortresses with essentially a military existence,

if they did not disappear altogether. This process of contraction and disappearance

was often repeated in the countryside.5

There is little evidence of major building activity other than the construction

and repair of religious structures, or (re)fortification as a defence against Arab

incursions. Foss states that the reconstructed defences at Sardis are a good example

of the building work typifying this period. The remaining wall which surrounded

the acropolis is now inaccessible without a long, uphill hike, although its defensive

function is evident to see from the plain of the ancient city. Archaeology has

revealed that new fortresses were also built during this period. Examples include

Malagina which was situated in a verdant region of the same name near the lower

Sangarios River (approximately modern Geyve to modern Osmaneli) and a simple

fortified refuge near modern Kutahya. Any evidence of urban expansion at this time

is exiguous, and it appears that contemporary Islamic geographers’ descriptions

were accurate. Byzantine Anatolia generally consisted of cities and towns which

appeared to be little more than fortresses.6

The rate of transformation of centres of habitation from the urban poleis

of antiquity to the Byzantine kastra should not be overstated and neither should the

abandonment of sites. The initial deterioration of classical structures was gradual.

Most places had already begun to transform and exhibit archaeological and

topographical features of the middle Byzantine kastron as early as the fourth and

fifth centuries, but particularly in the sixth. The extensive excavations at Amorion

and other sites indicate that whilst the main area of occupation contracted to a

fortress-citadel during the seventh and eighth centuries, discrete settlements

resembling villages continued to be inhabited within the old Roman walls. It is also

argued that one result of the Arab incursions was a growth in importance

of fortified towns such as Gangra (modern Cankiri), which provided refuge for the

5 James Russell, ‘‘Transformations in early Byzantine urban life: the contribution and limitations of

archaeological evidence’’, The Seventeenth International Byzantine Congress. Major Papers (Washington,

DC: Aristide D. Caratzas, 1986), pp. 135–154; Wolfram Brandes, Die Stade Kleinasiens im 7. und 8.

Jahrundert (Berlin: Berliner byzantinistische Arbeiten 56, 1989); Wolfgang Liebeschuetz, ‘‘The end of

the ancient city’’, in The City in Late Antiquity, ed. John Rich (London: Routledge, 1992), pp. 1–49; John

Haldon, Byzantium in the Seventh Century: The Transformation of a Culture, 2nd edn (Cambridge:

Cambridge University Press, 1997), pp. 92–124, 459–461; Leslie Brubaker and John Haldon,

Byzantium in the Iconoclast Era (c.680–850) (Aldershot: Ashgate, 2001), pp. 146–156; Chris Wickham,

Framing the Early Middle Ages: Europe and the Mediterranean 400–800 (Oxford: Oxford University Press,

2005), pp. 626–635. Mark Whittow has recently questioned the consensus of opinion expressed in this

paragraph. See ‘‘The Middle Byzantine economy (600–1204)’’, in The Cambridge History of The

Byzantine Empire c.500–1492, ed. Jonathan Shepard (Cambridge: Cambridge University Press, 2008),

pp. 465–492 (here 478–486).6 Clive Foss, Byzantine and Turkish Sardis (Cambridge, MA: Harvard University Press, 1976), pp. 57–

59; Clive Foss, ‘‘Byzantine Malagina and the Lower Sangarius’’, Anatolian Studies, XL (1990): 161–183;

Clive Foss, Survey of Medieval Castles of Anatolia I: Kutahya, British Institute of Archaeology at Ankara,

Monograph 7 (Oxford: Oxford University Press, 1985), pp. 95–98.

252 Jason T. Roche

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011

surrounding population. Nevertheless, the archaeological record appears to agree

with John Haldon’s suggestion that by the ninth century generally small but distinct

communities continued to consider themselves as ‘‘citizens’’ of the polis within

whose walls their communities lived, and that the kastron, which retained the name

of the ancient polis, provided a refuge in the case of attack.7

The anonymous Persian author of the geographical treatise Hadud al-’Alam

summarised the general situation in the tenth century. ‘‘In the days of old’’,

he writes, ‘‘cities were numerous in Rum, but now they have become few. Most of

the districts are prosperous and pleasant and have an extremely strong fortress

on account of the raids which the fighters of the faith direct upon them. To each

village appertains a castle where in times of flight they may take shelter’’. Such

strongholds often retained a bishop’s presence and continued to act as focal points

for markets and fairs. Indeed, the Byzantine economy began to flourish in the ninth

century and continued to do so throughout the tenth and eleventh centuries. There

is clear evidence of an economic revival and of construction work in Sardis,

Amorion, Ankyra, and Attaleia (modern Antalya). But Foss has shown that the

essential military nature of centres of habitation, firmly established during the

seventh and eighth centuries, generally continued as the form of settlement when

they were newly built and refortified during the period of demographic and

economic growth under the Macedonian dynasty (867–1025).8

The Turkish incursions into Anatolia, however, particularly those in the wake

of the Battle of Mantzikert (1071) when the greater part of the country fell into

Turkish hands, coupled with the resultant Byzantine internal anarchy and civil

strife, understandably appear to have provoked a hiatus in urban development.

Foss maintains that the break in development at Sardis, as evinced by the

archaeological record, is a microcosm of the rest of western Anatolia. Successive

Komnenoi (1081–1185) recovered parts of Anatolia, and concerned themselves

with the founding and rebuilding of defensive structures and the re-colonisation

of those areas which they had returned to Byzantine hegemony. The literary

record shows that following his accession to the imperial throne in 1081, Alexios I

Komnenos attempted to stabilise the geopolitical situation near his capital by

constructing fortresses such as Kibotos (modern Hersek) in Bithynia. After the

Turks had been pushed back from the western coastal plains and river valleys and

on to the central Anatolian plateau in the wake of the First Crusade in 1097/8,

Alexios began the reconstruction and repopulation of Aegean coastal towns,

7 John Haldon, Warfare, State and Society in the Byzantine World, 565–1204 (London: Routledge, 1999),

pp. 249–250. Chris Lightfoot has produced a series of reports on the Amorion excavations;

see, for example, Chris Lightfoot, ‘‘Die Byzantinische Stadt Amorium: Grabungsergebnisse der Jahre

1988 bis 2008’’, in Byzanz - das Romerreich in Mittelalter, Teil 2,1 Schauplatze, ed. F. Daim and J.

Drauschfe (Mainz: Monographien des Romisch–Germanischen Zentralmuseams 84/2, 1), pp. 293–307.

Foss, Byzantine and Turkish Sardis, 55–62; Clive Foss, ‘‘Archaeology and the ‘Twenty Cities’ of

Byzantine Asia’’, American Journal of Archaeology, LXXXI (1977): 469–486; Clive Foss, ‘‘The cities of

Pamphylia in the Byzantine age’’, in Cities, Fortresses and Villages of Byzantine Asia Minor, Clive Foss

(Aldershot: Ashgate, 1996), pp. 4, 1–62 (here 24–46).8 Vladimir Minorsky, ed. and trans., Hadud al-‘Alam: The Regions of the World (London: Gibb Memorial

Trust, 1937), p. 157; Foss, ‘‘Late antique and Byzantine Ankara’’, 27–87; Clive Foss and David

Winfield, Byzantine Fortifications, An Introduction (Pretoria: University of South Africa, 1986), pp. 142–

144; Clive Foss, ‘‘Cities of Pamphylia’’, 1–62; Foss, ‘‘Archaeology’’, 469–486; Foss, Byzantine and

Turkish Sardis, 70, 74–75; Lightfoot ‘‘Die Byzantinische Stadt Amorium’’; Whittow, ‘‘Middle Byzantine

economy’’, 473–476.

Surveying the Aspect of the Medieval West Anatolian Town 253

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011

including the fortress town of Adramyttion (modern Edremit) which the Turks had

razed to the ground. Anna Komnene informs us that the fortresses of Korykos

(modern Kizkalesi) and Seleukeia (modern Silifke) in Cilicia were rebuilt and

refortified as early as 1099. Foss has shown that fortresses were also rebuilt or

expanded by the Komnenoi at Telmessos, Xanthos, Patara, Myra and Limyra on

the Lycian coast. The year after his accession in 1119, John II Komnenos retook

and refortified Phrygian Laodikeia (near modern Denizli). Narrative sources

corroborated by archaeological evidence demonstrate that he also built the

fortresses of Lopadion (modern Uluabat near Karacabey) in 1130 and

Akhyraous (near modern Pamukca) around 1140. This evidence also confirms

that John’s successor, Manuel I Komnenos, carried out a significant amount of

construction work. For example, he rebuilt the walls of Attaleia in Pamphylia. A

fortress was built and garrisoned around 1145 at Malagina in Bithynia, although

there is evidence of a defensive construction from the seventh century. The fort of

Pithekas (modern Kuplu?) in Bithynia was rebuilt and garrisoned around the same

time. Also in Bithynia (on the coast), Manuel constructed a fort named Pylai on a

much older site. Other examples include the fortified refuges he built near the

Lydian fortified towns of Pergamon (modern Bergama), Adramyttion and Chliara

(near modern Soma), which themselves received new walls between 1162 and

1173. Manuel also rebuilt the fortresses of Dorylaion (near modern Eskisehir) and

Choma, also known as Soublaion (modern Homa), in 1175/6.9

Construction work was not restricted to fortifications during this later period,

and some scholars have argued that the textual record suggests larger urban centres

existed in the eleventh and twelfth centuries than in the preceding centuries.

We know that ecclesiastical sources, contemporary histories and occasional

9 Jason T. Roche, ‘‘In the wake of Mantzikert: The First Crusade and the Alexian reconquest of Western

Anatolia’’, History: The Journal of the Historical Association 94 (2009): 135–153; for Sardis, see Foss,

Byzantine and Turkish Sardis, 66–76; for Kibotos, see Anna Komnene, The Alexiad of Anna Comnena,

trans. E. Sewter (Middlesex: Penguin, 1969), pp. 203–204; for Adramyttion, see Anna Komnene,

Alexiad, 436–437; for Korykos and Seleukeia, see Anna Komnene, Alexiad, 363; for the Lycian coast,

see Clive Foss, ‘‘Lycia in history’’, in Foss, Cities, Fortresses and Villages, 1, 1–37; for Laodikeia, Lopadion

and Akhyraous, see Anna Komnene, Alexiad, 453; John Kinnamos, Deeds of John and Manuel Comnenus,

trans. Charles Brand (New York: Colombia University Press, 1976), pp. 14–15, 38; Niketas Choniates,

O City of Byzantium, Annals of Niketas Choniates, trans. Harry Magoulias (Detroit, IL: Wayne State

University Press, 1984), p. 9; William Ramsey, The Historical Geography of Asia Minor, Royal

Geographical Society, Supplementary Papers 4 (London, 1890), pp. 160, 201–202; Clive Foss, ‘‘The

defenses of Asia Minor against the Turks’’, Greek Orthodox Theological Review, XXVII (1982): 145–205

(here 159–161); for Attaleia, see Foss, ‘‘Cities of Pamphylia’’, 50; for Malagina, which is also

occasionally referred to in the sources as Metabole, see John Kinnamos, Deeds, 37; Niketas Choniates,

O City of Byzantium, 31; Foss, ‘‘Byzantine Malagina’’, 61–83; for Pithekas, see John Kinnamos, Deeds,

38; Klaus Belke and Norbert Mersich, Phrygien und Pisidien, Tabula Imperii Byzantini, Osterreichische

Akademie der Wissenschaften Philosophisch-Historische Klasse, 7 (Vienna, 1990), pp. 141–142; for

Pylai, see John Kinnamos, Deeds, 56; Ramsey, Historical Geography, 187; for the refuges in Lydia,

see Niketas Choniates, O City of Byzantium, 85; Foss, Byzantine Fortifications, 147; Foss, ‘‘Defenses’’,

166–170; for Dorylaion, see John Kinnamos, Deeds, 220–222; Niketas Choniates, O City of Byzantium,

99–100; Bar Hebraeus, Political History of the World, part 1, trans. Ernest Budge (Oxford: Oxford

University Press, 1932), p. 306; Michael the Syrian, Chronicle, ed. and trans. Jean-Baptiste Chabot,

Chronique de Michel le Syrien, Patriarche Jacobite d’Antioche (1166–1199), 3 (Paris, 1905; repr. Brussels,

1963), p. 369; for Choma, see John Kinnamos, Deeds, 223; Niketas Choniates, O City of Byzantium, 100;

Bar Hebraeus, Political History, 306; Michael the Syrian, Chronicle, 369; for Hierapolis, see Celal S� imsek,

‘‘Ikinci sezon Hierapolis Roma hamami (Muze Binasi) Kazi Calismalari’’, Muze Kurtarma Kazilari

Semineri 5: 243–263.

254 Jason T. Roche

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011

mentions in hagiographies actually tell us little about the physical size of a place.

There is, however, some archaeological evidence that Ankyra expanded beyond

its citadel’s walls in the final centuries of Byzantine rule. Foss has shown

that considerable building activity took place at Ephesos on the Aegean coast in

the twelfth and thirteenth centuries both within and outside its fortified walls.

The archaeological record shows that Sardis thrived after the Arab invasions and

before Mantzikert, and distinct settlements flourished again amongst the remains

of antiquity at the end of the twelfth century. Foss has revealed that in Lycia at

Lebissos, a cathedral was rebuilt, other parts of the town were occupied, and

settlements around Myra on the coast received new chapels and walls. Such

developments are indicative of burgeoning urban populations and economies, and

even twelfth-century Hierapolis (modern Pammukale), situated amongst transhu-

mant Turkmen tribes and close to the Seljuk strong holds on the southwestern edge

of the Anatolian plateau, witnessed small-scale industrial activity within the safety

of its ancient walls. Invariably, however, such economic activity and construction

work was on a much smaller scale compared to that conducted in antiquity.10

Newly built and rebuilt civic, ecclesiastical and military structures rarely

compared in size with their late Roman equivalents. The newly built and

reoccupied fortified towns, for example, were larger than the many simple

fortresses which peppered the countryside, but not always a great deal larger.

The ancient Bithynian town of Nikomedia (modern Izmit) is a good illustration

of this. Described by Ibn Khurradadhbih around 845 as ‘‘now ruined’’, its location

on a major route to Constantinople nevertheless ensured its continued importance

and the Byzantines soon recovered the town after Mantzikert. Odo of Deuil

described Nikomedia in the winter of 1147 as set amongst thorns and brambles and

still ‘‘with lofty ruins testifying her ancient glory’’. The populated part of the city

had contracted from the coastal plain (with its ‘‘lofty ruins’’) to its fortress-citadel

during the seventh and eighth centuries. The fortress-city occupied an area of only

about 1000 m�500 m on the hilltop overlooking the vast expanse of the ancient

city, even following rebuilding work in the twelfth century.11

The Aegean coastal cities which were brought back within the imperial domain

at the very end of the eleventh century gave the following impression to Odo

of Deuil: ‘‘we came upon many ruined cities, and others which the Greeks had

built up from the ancient level above the sea, fortifying them with walls and

towers’’. One of these ruined cities was Pergamon. Manuel Komnenos rebuilt its

defensive walls between 1162 and 1173 to surround the ancient acropolis and

enclose the settlements built on the slope of the citadel. This construction work can

be viewed as typically representative of the greater extent of development

10 Alexander Kazhdan and Ann Wharton Epstein, Change in Byzantine Culture in the Eleventh and Twelfth

Centuries (Berkeley, CA: University of California Press, 1985), pp. 36–39; Foss, ‘‘Late antique and

Byzantine Ankara’’, 27–87; Clive Foss, Ephesus after Antiquity: A Late Antique, Byzantine and Turkish City

(Cambridge: Cambridge University Press, 1979), pp. 116–117; Foss, Byzantine and Turkish Sardis,

70–76; Foss, ‘‘Lycia’’, 1–37.11 Ibn Khurradadhbih, Kitab al-Masalik wa‘l-Mamalik, ed. Michael de Goeje, Bibliotheca geographorum

arabicorum, 6 (Leiden, 1889; repr. Baghdad, 1967), pp. 106, 113; ‘‘ruinis sublimibus antiquam sui gloriam

. . . probat’’, Odo of Deuil, De Profectione Ludovici VII in Orientem, ed. and trans. Virginia Berry (New

York: Columbia University Press, 1948), p. 88; Clive Foss, Survey of Medieval Castles of Anatolia II:

Nicomedia, British Institute of Archaeology at Ankara, Monograph 21 (Oxford: Oxford University Press,

1996), pp. 1–41. Compare with the fortress of Lopadion with walls at 475 m�150 m: Foss, Byzantine

Fortifications, 145–466.

Surveying the Aspect of the Medieval West Anatolian Town 255

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011

undertaken at any one existing site during this later period. Pergamon’s form can be

taken as the archetype which Byzantine towns in the twelfth century had continued

to acquire since the eighth century even with the growth and recovery accomplished

under the Macedonian and Comnenian dynasties. Indeed, Pergamon’s develop-

ment was entirely typical. It had retracted to a heavily walled fortress on the

acropolis by the ninth century, and then it slowly recovered to occupy the slope of

the hill. Manuel’s walls enclosed individual settlements of crudely built houses

centred on a small number of churches by the end of the twelfth century. There is

some archaeological evidence of limited industrial activity, and parts of the

expanded town were set amidst the ruins of the monumental buildings of antiquity.

Such sites, together with numerous smaller garrison forts and outposts generally

situated on rocky outcrops and prominences typified the Anatolian provincial

countryside well into the Seljuk period and beyond.12

Archaeology then, occasionally supporting, but often acting as a corrective to the

written sources, demonstrates that the typical Byzantine town in the twelfth century

had slowly expanded with demographic and economic growth beyond the limits to

which it had contracted during the seventh and eighth centuries. It is nevertheless

clear that such settlements did not resemble cities in any modern sense of the word.

They began to lose this form and function by the late fourth century, and by the

ninth, the typical urban centre of habitation no longer consisted of the monumental

public buildings and permanent houses of the once flourishing classical city situated

in a plain. The inhabitants of such centres in the middle Byzantine period generally

continued to reside well within the remaining walls of the classical polis, but in

villages situated within the circumference of the walls. Perhaps separated by

cultivated land and largely self-sufficient, the villages were often situated on the hill

topped by the citadel of antiquity which afforded them protection – a phenomenon

which was to continue until the end of our period. Foss sums up the typical urban

development: ‘‘The medieval city developed with different considerations, in which

security was paramount . . . The transformation took place in the Dark Ages

[mid-seventh to mid-eighth centuries]: succeeding centuries brought no major

change’’. The evidence for Lycia, Pamphylia, and indeed most of Asia Minor

‘‘reveals the same picture of late antique prosperity, Dark Age catastrophe, and very

limited Byzantine recovery’’. This deduction is repeated in the works cited in this

article.13

The ancient fortified town of Nikaia (modern Iznik) may have been an important

exception to the established pattern of contraction and expansion mainly because

it was regarded as Constantinople’s Bithynian bulwark against invasion from the

central plateau. Although very little relevant archaeological work has been

completed inside its great circuit of walls, studies of the impressive ancient

fortifications, which dwarf the modern village in height and circumference, indicate

that many repairs and improvements were made during the eighth and ninth

centuries. Repairs to the existing fortifications were also carried out after the

earthquake of 1065 and the crusaders’ siege of 1097. The archaeological record

12 ‘‘multas urbes destructas invenimus et alias quas ab antiqua latitudine supra mare Graeci restruxerant,

munientes eas muris et turribus’’, Odo of Deuil, De Profectione, 106; Foss, ‘‘Archaeology’’, 469–486;

Manfred Klinkott, Die Stadtmauer, Teil 1: Die byzantinische Befestigungsanlage von Pergamon mit ihrer

Wehr- und Baugeschichte, Altertumer von Pergamon, XVI/1 (Berlin: Walter de Gruyter, 2001); Haldon,

Warfare, 250–251.13 Foss, ‘‘Lycia’’, 30; Foss, ‘‘Cities of Pamphylia’’, 50.

256 Jason T. Roche

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011

proves that the ancient walls were maintained, reflecting Nikaia’s significant

defensive role. It is impossible to demonstrate that Nikaia’s occupied area retracted

to a citadel before slowly expanding during the periods of Byzantine recovery until

at least further archaeological excavations are undertaken in and around the

modern village. One might conclude on the existing archaeological evidence that

throughout the Byzantine Age, the ancient city of Nikaia resembled a classical polis

in form and its inhabitants continued to occupy the same expansive area they had

dwelt in since antiquity.14

It is unwise to treat Nikaia in isolation, however, even given the town’s strategic

importance. This would be to ignore 800 years of urban history and the analogous

evidence presented here. It is safe to assume that during the twelfth century, the

bulk of Nikaia’s monumental buildings were at least dilapidated, and that the city’s

inhabitants actually resided in a small number of disparate villages located well

within the circumference of the ancient fortifications. Perhaps the major difference

between Nikaia and, say, the other great town of eastern Bithynia, Nikomedia,

was that a greater number of garrison troops resided in Nikaia to help defend her

extensive walls, the town’s inhabitants and, ultimately, Constantinople. Numerous,

prosperous and thriving civil populations still existed in the late Roman period,

and were to appear again with the peace of Ottoman rule, but during the

geopolitical troubles of the High Middle Ages, populations were continually

threatened and a town’s function remained predominantly defensive despite

economic and demographic growth; facts which are reflected in the relatively low

number of inhabitants.15

This synthesis of primary and secondary source evidence has revealed that

travellers in the High Middle Ages encountered kastra that – at most – appeared to

be fortress towns actually consisting of a number of disparate villages. The urban

form and function were essentially inherited from the eighth century, regardless of

the recession and rebuilding in the decades after Mantzikert. We have seen that a

strictly philological approach to achieving the modest aims of this exercise is flawed,

and we must employ archaeological evidence in conjunction with literary sources.

Archaeological studies of Anatolia’s medieval heritage are, nevertheless, still in their

infancy, and yet the country is peppered with unstudied pre-modern remains which

are disappearing under the twin threat of modern development and natural erosion.

We may never know the size and form of twelfth-century Chalcedon, for example,

which is buried under modern Kadikoy on the Asiatic coast of the Bosphoros.

On the other hand, our knowledge of a medieval town such as Dorylaion, situated

on the northwestern edge of the Anatolian plateau 3 km outside the modern city

of Eskisehir, may be greatly enhanced by a coordinated programme of excavations

and field surveys. One looks forward to the future results from the British

Institute of Archaeology at Ankara, which is now driving archaeological research

in Turkey, and which promises to improve our understanding of the country in the

Middle Ages.

14 On Nikaia’s defences, see Foss, Byzantine Fortifications, chapter 2. Also see Clive Foss, Nicaea:

A Byzantine Capital and Its Praises (Brookline: Holy Cross Orthodox Press, 2005). A very small number

of other significant towns also retained their ancient circuits of walls, but the sites have been continually

occupied and it is impossible to trace their development. Foss, ‘‘Cities of Pamphylia’’, 49.15 Alan Harvey, Economic Expansion in the Byzantine Empire, 900–1200 (Cambridge: Cambridge

University Press, 1989), p. 199.

Surveying the Aspect of the Medieval West Anatolian Town 257

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011

Dow

nloa

ded

by [

109.

121.

11.1

5] a

t 08:

49 2

2 D

ecem

ber

2011