The Silk Roads:

an ICOMOS Thematic Study

by Tim Williams
on behalf of ICOMOS
2014
The Silk Roads
An ICOMOS Thematic Study

by Tim Williams
on behalf of ICOMOS
2014
International Council of Monuments and Sites
11 rue du Séminaire de Conflans
94220 Charenton-le-Pont
FRANCE

ISBN 978-2-918086-12-3

© ICOMOS
All rights reserved
Contents
STATES PARTIES COVERED BY THIS STUDY ......................................................................... X
ACKNOWLEDGEMENTS ..................................................................................................... XI
1 CONTEXT FOR THIS THEMATIC STUDY ........................................................................ 1
1.1 The purpose of the study ......................................................................................................... 1
1.2 Background to this study ......................................................................................................... 2
1.2.1 Global Strategy ................................................................................................................................ 2
1.2.2 Cultural routes ................................................................................................................................. 2
1.2.3 Serial transnational World Heritage nominations of the Silk Roads .................................................. 3
1.2.4 Ittingen expert meeting 2010 ........................................................................................................... 3
2 THE SILK ROADS: BACKGROUND, DEFINITIONS AND SCOPE ........................................ 6
2.1 The Silk Roads .......................................................................................................................... 6
2.2 Silk Roads’ misconceptions: neither silk nor roads .................................................................. 7
2.3 The nature of exchange: high, medium and low value goods .................................................. 9
2.4 Geographical scope ................................................................................................................ 11
2.5 Chronological scope ............................................................................................................... 14
2.6 Geography, climate & vegetation .......................................................................................... 15
2.6.1 Ecological zones ............................................................................................................................. 15
2.6.2 Topography .................................................................................................................................... 16
2.6.3 Hydrology ...................................................................................................................................... 16

3 ISSUES ...................................................................................................................... 19
3.1 Scale of the overall study ....................................................................................................... 19
3.2 Scale of existing mapping ....................................................................................................... 19
3.3 Sites included - and omitted .................................................................................................. 19
3.4 The quality of information on sites and their dating .............................................................. 20
3.5 Site locations .......................................................................................................................... 20
3.6 Place names: ancient & modern ............................................................................................ 21
4 PREPARATION OF A GEOGRAPHICAL INFORMATION SYSTEM ................................... 22
4.1 Introduction ........................................................................................................................... 22
4.2 Data sources .......................................................................................................................... 22
4.2.1 Published works on the Silk Roads ................................................................................................. 22
4.2.2 Historical accounts ......................................................................................................................... 22
4.2.3 State Party Tentative Lists .............................................................................................................. 23
4.2.4 Published map data ....................................................................................................................... 24
4.2.5 Digital data sources ........................................................................................................................ 24
4.2.6 Climate data................................................................................................................................... 25
4.2.7 Ecological and hydrographic data ................................................................................................... 25
4.2.8 Topographic & physical geography ................................................................................................. 26

5 CONCEPTUAL APPROACH TO MAPPING: NODES, ROUTE SEGMENTS & CORRIDORS . 27

5.1 Mapping the routes ............................................................................................................... 27
5.2 Definitions: nodes, route segments & corridors .................................................................... 27
5.2.1 Node .............................................................................................................................................. 27
5.2.2 Segment......................................................................................................................................... 29
5.2.3 Corridor ......................................................................................................................................... 29
5.2.4 Adapting the corridor buffer .......................................................................................................... 30
5.2.5 Conclusions .................................................................................................................................... 30
5.3 Digitising routes ..................................................................................................................... 30
6 ANALYSIS ................................................................................................................. 31

Page | iv
 6.1 Routes & corridors ................................................................................................................. 31
6.2 WHS and TL distribution ........................................................................................................ 32
6.2.1 Existing WHS .................................................................................................................................. 32
6.2.2 Tentative list sites .......................................................................................................................... 33
6.3 Silk Roads site categories ....................................................................................................... 34
6.3.1 Definitions ..................................................................................................................................... 34
6.3.2 Number of sites and their distribution ........................................................................................... 34
6.3.3 Category 1 Infrastructure ............................................................................................................... 36
6.3.4 Category 2 Production ................................................................................................................... 47
6.3.5 Category 3 Outcomes ..................................................................................................................... 48
6.3.6 Conclusions: Silk Roads Categories ................................................................................................. 52
6.4 Empire systems ...................................................................................................................... 53
6.5 Nomads .................................................................................................................................. 54
6.6 Buried archaeological sites .................................................................................................... 54
7 THE WAY FORWARD ................................................................................................. 56
7.1 A single property? .................................................................................................................. 56
7.2 Nomination strategy .............................................................................................................. 56
7.2.1 Introduction ................................................................................................................................... 56
7.2.2 Themes, individual sites or corridors? ............................................................................................ 57
7.2.3 Selecting specific corridors ............................................................................................................. 60
7.2.4 Site selection within chosen corridors ............................................................................................ 61
7.2.5 Compilation of nomination dossiers & an overall Silk Roads framework......................................... 63
7.2.6 Harmonizing and developing Tentative Lists .................................................................................. 63
7.3 Additional recommendations ................................................................................................ 63
7.3.1 Further research ............................................................................................................................ 63
7.3.2 Documentation and national inventories ....................................................................................... 64
7.3.3 Capacity building ............................................................................................................................ 64
7.3.4 Developing the role of expert groups ............................................................................................. 64
7.3.5 Digital data dissemination .............................................................................................................. 65
7.3.6 International funding and support mobilized by UNESCO ............................................................... 67
ANNEXES ......................................................................................................................... 68
1 DRAFT CONCEPT STATEMENT FOR THE SILK ROADS.................................................. 69
2 TENTATIVE LIST SITES FROM THE STUDY AREA ......................................................... 71
3 SELECTED CORRIDORS .............................................................................................. 76
3.1 Introduction ........................................................................................................................... 76
3.2 Selected corridors .................................................................................................................. 76
4 DATABASE & GIS .................................................................................................... 103
4.1 Introduction ......................................................................................................................... 103
4.2 Silk Roads Database ............................................................................................................. 103
4.2.1 The Sites table.............................................................................................................................. 103
4.2.2 Other tables ................................................................................................................................. 104
4.3 Silk Roads GIS ....................................................................................................................... 104
4.3.1 Introduction ................................................................................................................................. 104
4.3.2 Data drawn from the database or directly digitised ...................................................................... 106
4.3.3 Scanned and digitised data .......................................................................................................... 108
4.3.4 Digital archaeological & historical data ........................................................................................ 110
4.3.5 Physical data: ArcGIS Online ......................................................................................................... 120
4.3.6 Physical & environmental data: downloadable............................................................................. 126
4.3.7 Software ...................................................................................................................................... 138

5 BIBLIOGRAPHIC SURVEY ......................................................................................... 142

Page | v
REFERENCES CITED ......................................................................................................... 143

Page | vi
List of figures

Figure 1. The vast geographical extent of the terrestrial Silk Roads, showing major routes (in red) and
other significant routes (orange). ......................................................................................................... 6
Figure 2. The broad study area (in green), covering some 25 million km 2. ......................................... 11
Figure 3. The study area (green) with major (red) and other significant (orange) routes. .................. 12
Figure 4. Chang’an (China) to Antioch (Turkey): 6,478 km ‘as the crow flies’ (Google Earth image
Landsat © Orion-ME 2014 © Google 2014 [accessed April 2014]). .................................................... 13
Figure 5. Chang’an in China to Antioch in Turkey, by perhaps the most direct route taking into
account geographic obstacles: closer to 7,250 km (Google Earth image Landsat © Orion-ME 2014 ©
Google 2014 [accessed April 2014]). ................................................................................................... 14
Figure 6. The diversity of ecological and topographic zones that the Silk Roads passed through. ..... 16
Figure 8. The major desert areas along the Silk Roads. Even here, there is a complexity of desert
forms, from the high plateau deserts surrounded by chains of high mountains, such as the Gobi, the
Djungarian, and the Taklamakan, as opposed to the lowland dunal deserts such as the Karakum. ... 17
Figure 9. A portion of Central Asian routes against a backdrop of the AS cropland data. .................. 26
Figure 10. Selecting nodes and segments. The principal sites between Khotan (left) and Lop
Nor/Loulan (right). The green lines are the segments identified in the OWTRAD dataset, and simply
link nodes with straight lines. In red, a more complex picture, with more settlements and a route
digitised to reflect local topography (rivers, oases, etc) (from the Historical Atlas of Eurasia). .......... 29
Figure 11. Adjusting the boundary of specific corridors. In (a) all the principal sites lie within a
corridor defined by a generic buffer 30 km to either side of the main segment defined between two
nodes (yellow). In (b) some sites (in red) lie outside the corridor. In (c) the buffer is redrawn to take
into account the local topography/ecology – which constricts the landscape available in some places,
and broadens it out in others. ............................................................................................................ 30
Figure 12. Distribution of current Tentative List sites (yellow triangles) across the study area. ......... 34
Figure 13. Stopping places as mapped from the OWTRAD data, showing something of the scale of
the sites in the central area of the Silk Roads. .................................................................................... 36
Figure 14. Distribution of caravanserai (orange circles: based on OWTRAD data) in Syria, with the
concentration of known sites between Damascus and Palmyra, and the paucity of sites in the eastern
Syrian desert. ...................................................................................................................................... 38
Figure 15. Major linear water systems (in blue) and some of the major drainage systems (in green,
such as the Tigris/Euphrates and the Ob), against the Silk Roads (red and orange). .......................... 41
Figure 16. Qanāts crossing a now desertified landscape in Central Asia (Google Earth)..................... 46
Figure 17. A sample of the Buddhist monasteries from the OWTRAD data sets, displayed in Google
Earth. See Annex 4.3.4.3 for details. ................................................................................................... 50
Figure 18. The wide fertile river system of the Ganges enabled sites to develop over a broad area and
there were multiple routes across this landscape. At present two corridors have been defined,
encompassing the main urban developments and monuments, but a single wider corridor probably
would be more appropriate. ............................................................................................................... 62
Figure 19. Mountain passes and steep-sided valleys would require a spatially much tighter definition
of the corridor, which might, in places, only extend a few hundred metres from a very discreet route.
Here the Karakoram highway (corridor 14) and through the high Karakoram (corridor 43) have been
deliberately left un-buffered at this stage. ......................................................................................... 62
Figure 20. Some selected corridors (or parts of corridors) highlighted in purple (details in next
figures). ............................................................................................................................................... 76
Figure 21. Sample corridors in the west.............................................................................................. 77
Figure 22. Sample corridors in the central area. ................................................................................. 78
Figure 23. Sample corridors in the east. ............................................................................................. 79
Figure 24. Example of sites in the database shown on the GIS. Key: Green = Existing World Heritage
Sites; Yellow = Tentative List Sites; White = Other sites of interest. ................................................. 107

Page | vii
Figure 25. Example of digitised routes. Red = main routes; yellow = more minor routes. ................ 107
Figure 26. Example of corridors: main routes buffered (red hatching extending 30 km either side of
the route). ......................................................................................................................................... 108
Figure 27. A geo-rectified scan of The archaeological map of Iraq, overlain with digitised site
locations. .......................................................................................................................................... 109
Figure 28. Geo-rectified scan of map from Siroux 1949. ................................................................... 109
Figure 29. Example of Siroux digitised caravanserai locations (named) and routes. ........................ 110
Figure 30. The multiplicity of information available in the Google Earth version of the Historical Atlas
of Eurasia. ......................................................................................................................................... 111
Figure 31. Example of the Historical Atlas of Eurasia ‘Silk Roads routes’ as a GIS shapefile. ............ 111
Figure 32. Example of the Historical Atlas of Eurasia ‘Eurasian Empires’ as a GIS shapefile: any of the
individual empires can be turned on or off. ...................................................................................... 112
Figure 33. Coverage of Innermost Asia maps (from the Digital Silk Roads project) in Google Earth. 113
Figure 34. Detail of Innermost Asia maps (from the Digital Silk Roads project) in Google Earth ...... 113
Figure 35. Coverage of Serinda maps (from the Digital Silk Roads project) in Google Earth. ............ 114
Figure 36. Detail of Serinda maps (from the Digital Silk Roads project) in Google Earth. ................. 114
Figure 37. Example of OWTRAD nodes connected by straight lines – in this case from Whitfield 1999.
.......................................................................................................................................................... 115
Figure 38. Example of OWTRAD monastic data in Google Earth. Legend - White circles: Clusters and
groups of monasteries; Yellow: Theravada monasteries; Green: Sammitiya, Sarvastivada &
Mahasanghika monasteries; Red: Mahayana monasteries; Magenta: Vajrayana monasteries; Blue:
Indeterminate Buddhist tradition. .................................................................................................... 116
Figure 39. Example of OWTRAD wiki monastic entry. ....................................................................... 117
Figure 40. Example of UNESCO World Heritage Sites in Google Earth – with direct link to UNESCO
World Heritage Centre information. ................................................................................................. 119
Figure 41. Example of Warwick Ball’s (1982) gazetteer of sites in Afghanistan, as seen in ArcGIS. .. 120
Figure 42. World Physical Map. ........................................................................................................ 121
Figure 43. World_Reference-Overlay on top of World_Physical _Map (see above). ........................ 121
Figure 44. World Boundaries and Places on top of World_Physical _Map (see above). ................... 122
Figure 45. World shaded relief. ......................................................................................................... 123
Figure 46. Example of World Topo Map. .......................................................................................... 124
Figure 47. DeLorme World basemap. ............................................................................................... 125
Figure 48. World_SRTM, with gradual colour change representing relief. ....................................... 126
Figure 49. World_SRTM2 has a simplified colour scheme that represents the data in five categories.
.......................................................................................................................................................... 127
Figure 50. AS cropland ...................................................................................................................... 128
Figure 51. AS pasture. ....................................................................................................................... 128
Figure 52. Köppen-Geiger climate classification (see key below). .................................................... 130
Figure 53. Example of 1:10m Physical Vectors from Natural Earth: showing mountain ranges, deserts,
steppe, major valley systems, etc, on the back drop of Natural Earth II. .......................................... 132
Figure 54. Example of desert backdrop. ........................................................................................... 132
Figure 55. Example of World Elevation Contours (with no backdrop). Individual contours lines can be
displayed or re-coloured. .................................................................................................................. 133
Figure 56. Example of World Linear Water layer, with intermittent and perennial streams. ........... 134
Figure 57. Example of simple World Drainage System map. ............................................................. 135
Figure 58. World Drainage Systems: can map and chart basin areas, discharge levels, sedimentation
load, and distance. ............................................................................................................................ 135
Figure 59. World Drainage Systems: example of charting basin area, sedimentation and discharge.
.......................................................................................................................................................... 136
Figure 60. Example of World Gazetteer. Labels and symbols can be rescaled. ................................. 137
Figure 61. Example of World Cites. ................................................................................................... 138

Page | viii
Figure 62. ESRI ArcMap. .................................................................................................................... 139
Figure 63. ESRI ArcGIS Explorer: showing a detail of an area from China. ........................................ 140
Figure 64. Ayaz Tepe (Uzbekistan). On the left as seen through ArcGIS Explorer (with the World
Imagery basemap) and on the right in Google Earth. ....................................................................... 140
Figure 65. Example of viewing large areas in Google Earth, with the patchy background created by
numerous different resolutions and exposures of satellite images. ................................................. 141

Page | ix
States Parties covered by this study
Active partners in the UNESCO Serial transnational World Heritage nominations of the Silk
Roads project: Afghanistan, People's Republic of China, Republic of India, Islamic Republic of
Iran, Republic of Kazakhstan, Kyrgyz Republic, Federal Democratic Republic of Nepal,
Republic of Tajikistan, Turkmenistan, and Republic of Uzbekistan.

Countries at least partly covered by the Thematic Study and who may wish to join the
project: Republic of Armenia, Republic of Azerbaijan, People's Republic of Bangladesh,
Kingdom of Bhutan, Republic of Iraq, The State of Israel, Italy, Japan, Hashemite Kingdom of
Jordan, Republic of Korea, The Democratic People's Republic of Korea, Republic of Lebanon,
Mongolia, Islamic Republic of Pakistan, the State of Palestine, Syrian Arab Republic, and
Republic of Turkey.

Page | x
Acknowledgements
The study was supported by the UNESCO World Heritage Centre through the voluntary
contribution made by China to the World Heritage Fund and commissioned by ICOMOS
International Secretariat. Special thanks go to:

x Regina Durighello, World Heritage Programme Director, ICOMOS – International

Secretariat, Paris
x Gwenaëlle Bourdin, World Heritage Programme Senior Specialist, ICOMOS –
International Secretariat, Paris
x Susan Denyer, World Heritage Adviser, ICOMOS
x Feng Jing, Chief of Unit, Asia and the Pacific Section, UNESCO World Heritage Centre,
Paris
x Roland Lin Chih-Hung, Programme specialist, Asia and the Pacific Section, UNESCO
World Heritage Centre, Paris

Significant work on the project was undertaken by:

x Paul Wordsworth, GIS specialist and doctoral candidate, Department of Cross-
Cultural and Regional Studies, University of Copenhagen
x Dr Gaygysyz Jorayev, researcher and translator, Institute of Archaeology, UCL

The contributions of the State Parties who kindly shared Tentative Lists, bibliographic and
other data, including:
x Afghanistan, People's Republic of China, Republic of India, Islamic Republic of Iran,
Republic of Kazakhstan, Kyrgyz Republic, Federal Democratic Republic of Nepal,
Republic of Tajikistan, Turkmenistan, and Republic of Uzbekistan

Generous assistance and advice was also provided by:

x Kaori Kawakami, Associate expert, Asia and the Pacific Section, UNESCO World
Heritage Centre, Paris
x Prof Guo Zhan, Vice-president of ICOMOS International & Director of the ICOMOS
International Conservation Center-Xi’an (IICC-X), China
x Ms Lu Qiong, Deputy Director, Department for Protection of Monuments and Sites,
State Administration of Cultural Heritage
x Dr Abdisafikhan Rakmanov, Deputy Head, Department for the Protection and
Utilization of Cultural Objects, Ministry of Culture and Sports, Uzbekistan
x Dr Shakirdjan Pidaev, Director of the Fine Arts Institute of the Academy of Sciences
of Uzbekistan
x Prof Mavluda Yusupova, Chief, Section on Architecture, Fine Arts Institute, Academy
of Sciences, Uzbekistan
x Dr Server Ashirov, Chief, Archaeology Section, Principal Department for Preservation
and Utilization of the Cultural Objects, Ministry of Culture and Sports, Uzbekistan
x Prof Karl Baipakov, formerly Director of the Institute of Archaeology named after A
Margulan, Academy of Science, Kazakhstan
x Ms Natalya Turekulova, President of ICOMOS Kazakhstan

Page | xi
 x Dr Dimitry Voyakin, Director of Archaeological Expertise Scientific Organisation,
Kazakhstan
x Dr Mukhammed Mamedov, Chairman, Department for the Protection and
Restoration of the Historical and Cultural Monuments, Ministry of Culture and TV
and Radio Broadcasting, Turkmenistan
x Dr Bakhyt Amanbayeva, Head of Department of Cultural Heritage, Institute of History
and Cultural Heritage, National Academy of Sciences, Kyrgyz Republic
x Ms Ainura Tentieva, Expert, ICOMOS Kyrgyz Republic
x Dr Sherali Khodzhaev, Senior specialist, Department of Historical and Cultural
x Heritage of the Ministry of Culture, Tajikistan
x Prof Rustam Mukimov, President of ICOMOS Tajikistan
x Dr Abduvali Sharipov, Director, National Museum named after K. Behzod, Tajikistan
x Dr Saidmurod Bobomulloev, Director of the National Museum of Antiquities,
Tajikistan
x Dr Mohammad Hassan Talebian, Member of Scientific Board of Iranian Cultural
Heritage, Handicrafts and Tourism, Iran
x Dr Adel Farhanghi, Director Research Center for Silk Routs and Advisor to the Head
of the Research Institute of Cultural Heritage, Handicrafts and Tourism Organization,
Iran
x Prof Aziz Ahmad Panjshiri, Member of the Board of Policy, Ministry of Information
and Culture, Afghanistan
x Dr Kosh Prasad Acharya, Former Director General, Department of Archaeology
x Government of Nepal
x Dr Bishnu Raj Karki, Director General, Department of Archaeology, Government of
Nepal
x Dr Buddha Rashmi Mani, Joint Director General, Archaeological Survey of India
x Prof Lena Kim, Hongik University & Cultural Heritage Committee Member
x Prof Juhyung Rhi, Professor of Art History, Seoul National University
x Ms Hyosang Jo, International Affairs Division, Cultural Heritage Administration,
Republic of Korea
x Dr Shahin Mustafayev, Director of International Institute for Central Asian Studies
(IICAS)
x Mr Yuri Peshkov, Culture Specialist, UNESCO Almaty Cluster Office
x Mr Sanjarbek Allayarov, UNESCO Cluster Office in Tashkent
x Prof Kazuya Yamauchi, Center for International Cooperation in Conservation,
National research Institute for Cultural Properties, Tokyo, Japan
x Dr Susan Whitfield, Director of the International Dunhuang Project, The British
Library, UK
x Professor Henry Cleere, Institute of Archaeology, University College London (former
World Heritage Coordinator for ICOMOS from 1992-2002, and former Special
Advisor to the State Administration of Cultural Heritage of China)
x Dr Matthew Ciolek, National Institute for Asia and the Pacific, Australian National
University, Canberra, Australia
x Prof Koenraad van Balen and Ona Vileikis, Raymond Lemaire International Centre for
Conservation, Katholieke Universiteit Leuven, Belgium
x Prof Philippe De Maeyer, Ghent University

Page | xii
1 Context for this thematic study
1.1 The purpose of the study
ICOMOS thematic studies are a synthesis of current research and knowledge on a specific
theme. The aim of this study is to provide an analysis of sites along the Silk Roads that could
be used by States Parties participating in the Serial transnational World Heritage
nominations of the Silk Roads as a basis for comparative analyses when nominating series of
sites.

The scale of existing research on this topic is extensive, ranging from synthetic works and
maps, to studies of individual regions and sites1. The inclusion of sites in this paper is not an
indication of their individual significance, or their potential as part of World Heritage serial
properties, but rather as part of a wider overview of the nature of sites along these vast
routes.

The study does not attempt to recommend any specific series of sites for Tentative Lists or
for nomination.

This study specifically aims to:

x Provide an analysis of sites along the Silk Roads that could be used by States Parties
participating in the Serial transnational World Heritage nominations of the Silk Roads
as a basis for comparative analyses when nominating serial properties.
x Profile the distribution and distinctiveness of Silk Roads sites in order to understand
how sites are manifestations of the shifting systems of power and patronage that
prevailed over time along the Silk Roads, in relation to the organisation of flourishing
trade and the protection of trade routes.
x Define the distribution of Silk Roads sites, in order to understand:
ƒ What sites are common to the whole extent of the Roads
ƒ What sites are specific to the whole Silk Roads or to certain parts of the
Roads
ƒ What sites are unique or exceptional
ƒ Which sites are plentiful and how their form varies in time and space
ƒ What sites are persistent over time
ƒ What sites reflect specific periods of history, power systems or cultural
traditions
ƒ Consider whether certain sections or corridors of the Silk Roads, through the
assembly of sites within them, are distinctive from other sections of the Silk
Roads, in terms of being manifestations of particular geo-cultural systems,
and whether a case could be made for considering the Silk Roads as a
collection of World Heritage properties, linked by a concept, instead of one
single World Heritage serial property.

1
See Annex 5 on bibliographic sources.

1|Page
The Thematic Study was directed by Tim Williams, Institute of Archaeology, University
College London (UCL), with the support of members of UCL, and carried out in collaboration
with Susan Denyer (ICOMOS) and Feng Jing (UNESCO World Heritage Centre).

1.2 Background to this study

1.2.1 Global Strategy
In 1988 UNESCO launched its Integral Study of the Silk Roads: Roads of Dialogue project, to
highlight the complex cultural interactions which arose from the encounters along the Silk
Roads. This prompted a number of regional meetings and scientific expeditions.

At broadly the same time a number of key debates were taking place regarding cultural
routes and the development of a global strategy for addressing the representivity of the
World Heritage list. Key papers included:

x Desk study on the Asia Region (ICOMOS 2002)

x Draft Framework for World Heritage Cultural Representation (Jokilehto et al. 2003).

In 2004, ICOMOS published The World Heritage List: Filling the Gaps – An Action Plan for the
Future (ICOMOS 2004). The review analysed cultural sites on the World Heritage List and
Tentative Lists, using regional, chronological, geographical and thematic frameworks. The
intention was to assess the then current representation of sites, likely short- to medium-
term trends, and identify under-represented categories.

The review has a number of ramifications for the present study. In particular, it identified
that, “in spite of its remarkable historical background and numerous historic/cultural sites in
the sub-region, Central Asia remains one of the most under-represented regions on the
World Heritage List” (UNESCO 2008a).

1.2.2 Cultural routes

The work of the ICOMOS International Scientific Committee on Cultural Routes (CIIC)2 is also
important in this context. An expert meeting in Madrid in November 1994 stated that:

“The concept of a cultural route or itinerary refers to a set of values whose whole is
greater than the sum of its parts and through which it gains its meaning;
identification of the cultural itinerary is based on an array of important points and
tangible elements that attest to the significance of the itinerary itself. To recognize
that a cultural itinerary or route as such necessarily includes a number of material
elements and objects linked to other values of an intangible nature by the
connecting thread of a civilizing process of decisive importance at a given time in
history for a particular society or group” (ICOMOS International Scientific Committee
on Cultural Routes 1994).

2
http://www.icomos-ciic.org/CIIC/CIIC.htm

2|Page
Development on the theory and implementation of cultural routes led to modifications to
the World Heritage Operation Guidelines (UNESCO 2005; updated in 2008b; and 2012), in
which the concept of ‘heritage routes’ was defined as:

“based on the dynamics of movement and the idea of exchanges, with continuity in
space and time; refers to a whole, where the route has a worth over and above the
sum of the elements making it up and through which it gains its cultural significance;
highlights exchange and dialogue between countries or between regions; is multi-
dimensional, with different aspects developing and adding to its prime purpose
which may be religious, commercial, administrative or otherwise. … The
identification of a heritage route is based on a collection of strengths and tangible
elements, testimony to the significance of the route itself.” (UNESCO 2008b, 91)

1.2.3 Serial transnational World Heritage nominations of the Silk Roads

From as early as 2003 consideration was being given to the development of a transboundary
nomination for the Silk Roads (Jing & van Oers 2004). A series of sub-regional workshops
were then developed, in Almaty in November 2005, Turfan in August 2006, Samarkand in
October 2006 (UNESCO 2006), Dushanbe in April 2007, Xi’an in June 2008, and Almaty in
May 2009 (UNESCO 2009a), all of which culminated in the first meeting of the Coordinating
Committee for the Silk Roads Serial Nomination, in Xian, China, 3-6 November 2009
(UNESCO 2009b).

Importantly, as a result of these meetings a Silk Roads concept paper was developed,
initially in 2006 by Professor Henry Cleere (2006) and refined at the subsequent meeting at
Dushanbe (UNESCO 2007). This defined the geographical and historical scope of the Silk
Roads in relation to possible World Heritage nominations.

The need for this Thematic Study was identified at the first meeting of the Coordinating
Committee for the Silk Roads Serial Nomination in Xian in November 2009 and the study
was subsequently commissioned by ICOMOS in June 2010.

1.2.4 Ittingen expert meeting 2010

In February 2010 an International World Heritage expert meeting on serial nominations and
properties was held in Ittingen, Switzerland (http://whc.unesco.org/archive/2010/whc10-
34Com-9Be.pdf). The conclusions of that meeting are directly relevant to the nomination
strategy for the Silk Roads. Specifically:

The 2008 operational guidelines stated that the components of a serial property should
be of: “a) the same historico-cultural [sic] group; b) the same type of property which is
characteristic of the geographical zone”; but the Ittingen meeting concluded that this
“may lead to a mere catalogue of sites without an adequate definition of the functional
links between the component parts, or an explanation as to how they contribute to the
Outstanding Universal Value (OUV) of the property as a whole”.

This has fundamental implications for the Silk Roads project: clearly the Silk Roads do not
represent the same historical/cultural group, or the same type of property in a geographical
zone: indeed the Silk Roads are an anathema to both these concepts. The Ittingen
recommendations, therefore, are very helpful in this context.

3|Page
The main recommendations of the Ittingen expert meeting were recognised by the World
Heritage Committee at its 34th session in Brasília, Brazil, 2010 (Decision 34 COM 9B) 3 and
modifications to the Operational Guidelines were subsequently agreed at the 35th session in
20114:

137. Serial properties will include two or more component parts related by clearly
defined links:

a) Component parts should reflect cultural, social or functional links over time that
provide, where relevant, landscape, ecological, evolutionary or habitat
connectivity.
b) Each component part should contribute to the Outstanding Universal Value of
the property as a whole in a substantial, scientific, readily defined and
discernible way, and may include, inter alia, intangible attributes. The resulting
Outstanding Universal Value should be easily understood and communicated.
c) Consistently, and in order to avoid an excessive fragmentation of component
parts, the process of nomination of the property, including the selection of the
component parts, should take fully into account the overall manageability and
coherence of the property.

This is an important issue for the Silk Roads. The scale of the Silk Roads is vast and it is vital
to maintain a sense of spatial integrity whilst finding a manageable sub-set of sites and
landscapes that can be effectively managed.

The Ittingen Expert Meeting also recommended that:

“the format for Tentative List submissions for serial transnational and transboundary
sites … would indicate the agreement between all of the States Parties involved. The
proposed property would therefore only be registered within the Tentative Lists as a
serial transnational site when all States Parties involved have included their
component parts in their national Tentative Lists” (Martin & Gendre 2010, 69).

The meeting also stressed “the opportunities for international assistance and regional
training workshops” (loc cit). This is a crucial aspect of the Silk Roads project (see section
7.3).

Helpfully, the meeting also noted that:

1.6 The Meeting recommended promoting and implementing the concept of

transnational serial nominations as a tool for international cooperation, shared
approaches and thus better management and conservation practice – for example,
by using new technologies such as establishing virtual visitor centres which, in
particular, facilitate the understanding of the concept and scientific framework of
such nominations.

3
Available at: http://whc.unesco.org/archive/2010/whc10-34Com-9Be.pdf
4
Available at: http://whc.unesco.org/archive/2011/whc11-35com-20e.pdf

4|Page
 2.3.2 When a series of sites is nominated, each State Party should be aware of the
implications (both in terms of opportunities and benefits, as well as specific
additional challenges) in relation to the nomination strategy they choose to adopt.
Examples of different nomination strategies include:

a. Nominating single properties, including series of national sites;

b. Extensions to existing World Heritage properties (both single or serial);

c. Nominating a series of single properties under a common framework (but not

constituting a single property);

d. Nominating a single serial transnational property.

5|Page
2 The Silk Roads: background, definitions and scope
2.1 The Silk Roads
The Silk Roads5 were an interconnected web of routes linking the ancient societies of East,
South, Central, and Western Asia, and the Mediterranean. It contributed to the
development of many of the world's great civilizations and enabled the exchange of
technologies and ideas that reshaped the known world6. This combination of routes
represents one of the world’s preeminent long-distance communication networks (Figure
1).

Figure 1. The vast geographical extent of the terrestrial Silk Roads, showing major routes (in red) and other significant
routes (orange).

There were a number of major impacts from this extensive network of interactions:

x The development of cities along these routes, which gained power and wealth from
the trade, providing the infrastructure of production and redistribution, and policing
its routes. Many became major cultural and artistic centres, where peoples of
different ethnic and cultural backgrounds intermingled.

5 th
A term coined by the 19 century German geographer Ferdinand von Richthofen.
6
For recent general works on the Silk Roads (Whitfield & Sims-Williams 2004; Boulnois 2008; de la Vaissière
2002; Hansen 2012; Liu 2010; Liu & Shaffer 2007; Millward 2013; Nebenzahl 2004; Tucker 2003; Waugh 2008;
Whitfield 2009; Wood 2002).

6|Page
 x The development of religious centres, which benefitted from the patronage of
political systems and wealthy individuals.
x The movement of technologies, artistic styles, languages, social practices and
religious beliefs, transmitted by people moving along the Silk Roads.

The Silk Roads spans a vast geographic area (section 2.4), over a protracted timeframe (see
section 2.5), that it presents significant problems with scale and resolution. The sheer scale
of these routes has meant that their complexity and impact is still under-researched and
often simplified into anecdotes. Broad-scale maps are common; large pen strokes across a
map of half the world, from Japan to the Mediterranean; there have been some excellent
regional studies; some useful historical accounts of sections of the route; and numerous
points of mapped data (such as the work on caravanserais, way-stations and monasteries in
Matthew Ciolek’s brilliant OWTRAD website)7. But there remains a problem of the sheer
scale of the routes and their impact, both spatially and chronologically.

2.2 Silk Roads’ misconceptions: neither silk nor roads

The term ‘Silk Roads’ (Seidenstrassen) was first coined by the 19th century German
geographer Ferdinand von Richthofen in a lecture in 1877. However, he also recognised that
there was a wider phenomenon of trans-Eurasian exchanges which are “now encompassed
by the shorthand we know as the silk road” (Millward 2013).

The first misconception, which the name ‘Silk Roads’ embodies, is that silk was the primary
commodity or driver of the exchange systems. In reality there were a great number of goods
being moved, over both short and long distances, and through a variety of different
mechanisms (section 2.3). Many had a far greater impact than silk, reflecting the spread of
technologies (such as gunpowder, paper or cotton production 8), or were moved in far
greater volumes (e.g. salt, tea, copper or iron).

Elements of this inter-connected system have sometimes been labelled with other single-
commodity terms, such as the salt route or the tea-horse road, or split by topographic
terms, such as the desert route: again, these do not really capture the complexity of the
trade goods or their impacts. These names also obscure the complexity of goods, material,
peoples and interactions; they blur the interconnections between routes and their
stimuli/drivers and serve to compartmentalise a complex systems of interactions. Where did
the tea-horse road stop and the salt route start, and how do these relate to the Silk Roads?
Nevertheless, these terms, perhaps especially the ‘Silk Roads’, have passed into common
parlance: they have become evocative labels. As a result this study, and the UNESCO Serial
transnational World Heritage nominations of the Silk Roads, uses the term ‘Silk Roads’, but
adopts the very broadest possible definition of the routes and materials exchanged. What is
of interest is the complexity of the interactions and their impact upon the development of
past and present societies. So the term ‘Silk Roads’ here is used to encompass this broad
network, and encompasses routes that many might be more familiar with by other names,
such as the tea-horse route.

7
http://www.ciolek.com/owtrad.html
8
The spread of the cotton industry, for example, was to have a major impact upon world economies (Riello
2013).

7|Page
A second misconception is that the principal impact of the Silk Roads was as an economic
trading network. There is no doubt that the movement of goods, and the volume of
exchange (and the taxation systems that extracted wealth from this process; section 2.3),
had a massive impact on many of the societies along the Silk Roads. It shaped their ability to
construct the cities, religious monuments and elite structures for which the Silk Roads are
famous. But the most significant outcome of these complex networks of interactions was
the movement of people and ideas: the spread of religions, social customs, languages,
political ideas, agricultural practices, scientific knowledge and technological innovations
(section 6.3.5) (e.g. Dani 2002).

Another further misconception is that the Silk Roads are primarily about the exchange of
material between East and West, as typified by silk: initially produced in China and
transported as far west as Europe. But the movements are much more complex: goods and
people did not simply flow from east to west and vice-a-versa. Many of the materials traded
emanated in Central or South Asia (section 2.3), and many of the ideas that moved along the
Silk Roads started in neither the east nor west (e.g. Buddhism). As a result the patterns of
routes from north to south are also vital elements of the complex system that is being
explored (see section 2.4). There is no doubt that understanding the routes across Central
and South Asia is essential to understanding the Silk Roads. The Silk Roads have often been
defined by their impact upon the ends, China and Europe, but this significantly underplays
the complexity of their overall impacts, almost making them seem of more relevance to the
ends than the rest of the regions through which they passed. The reality is that the Silk
Roads were essential in shaping the societies along their whole length.

Goods and ideas moved north-south as well as east-west. Crucial exchanges took place
across the Tibetan plateau and through the high mountain passes of the Karakoram
mountain range (e.g. via the Khunjerab Pass), linking Xinjiang with modern day Nepal,
Bhutan, Pakistan, India and Bangladesh (Freeman & Ahmed 2011; Fuchs 2008; Millward
2007; Tong 2013). The complex links between Central Asia and South Asia, through
Afghanistan and Pakistan, was also a vital aspect of the cultural exchange (e.g. see the
recent study on the 'Great Indian Road' by Edvard Rtveladze 2012). The much commented
upon routes travelled by Buddhist monks, so vital in the movement of ideas from South to
East Asia, who all used south-north paths that connected South and Central Asia (Chandra &
Banerjee 2007; Bernstein 2002; Boulnois 2008; Faxian 1886; Neelis 2011; Wriggins 2004).
These were crucial aspects of the Silk Roads and it is vital that the complexity of movements
does not become over-simplified into an east-west dialogue.

There are also misconceptions regarding the number of routes and their variability in space
and time. There was no single road winding from Asia to Europe. Rather there was a
diversity of paths, tracks and roads, changing not only over time, but also fluctuating
seasonal as river crossings and mountain passes became impassable, and shifting as
travellers passed across wide valleys and steppes, choosing different routes across the
landscape (section 5.1). Regional conflicts, changing markets and shifting political power
meant that the Silk Roads were never static. The importance of specific sections of the
routes changed over time, in response to a variety of socio-political factors. It is these
spatial and temporal changes in exchange and trade, and the scale of movement of people
and ideas, which articulates the significance of the Silk Roads. As a result this report always
uses the plural ‘Silk Roads’, rather than the singular.
8|Page
Another issue to note is that the Silk Roads were never a systematically planned network of
routes, or infrastructure, over its entire length; in certain regions, and at specific times, it
was very well planned and supported (for example, the system of Seljuk caravanserais (e.g.
Silverstein 2007), or Han dynasty forts (e.g. Millward 2007)), but it was often very local
responses to specific geo-political circumstances (see below).

2.3 The nature of exchange: high, medium and low value goods
The exchange of goods, and with them ideas and knowledge, played a major role in the
cultural, religious, and artistic exchanges that took place between Europe and Asia during
antiquity. The long-distance movement of materials and goods involved trade, exchange,
gift giving, and the payment of tribute. The latter considerations may have sometimes
overshadowed material motives in the exchange of goods (Christian 2000, 3), but the Silk
Roads undoubtedly facilitated the transfer of large quantities raw materials, foodstuffs, and
luxury goods.

Silk was one amongst several high-value, relatively low-weight, goods that moved along the
Silk Roads. Rhubarb, Spice, Musk & Tea - at various times all these commodities have been
used to describe part of the routes - reflecting their importance. These high value goods
certainly moved over considerable
distances, exchanged from one
merchant to the next.
Periplus Maris Erythraei, a Greek work describing
The variety of high- or medium-value navigation and trading from Roman Egyptian ports to
goods that moved through the Silk the Red Sea, North Africa and India, probably dating
Roads networks was truly remarkable to the mid-1st century CE, described the many goods
(see right) (e.g. Allsen 1997; Boulnois exchanged. At the market-town of Barigaza (modern
2008; de la Vaissière 2002; Hansen Bharuch, Gujarat State) in India, for example, these
2012; Liu 1988; Liu & Shaffer 2007; included “wine, Italian preferred, also Laodicean and
Schafer 1963; Tucker 2003). Some of Arabian; copper, tin, and lead; coral and topaz; thin
these relied on their manufacturing clothing and inferior sorts of all kinds; bright-coloured
skill to create value, as with the Song girdles a cubit wide; storax, sweet clover, flint glass,
dynasty ceramics (Kessler 2012), realgar, antimony, gold and silver coin, on which there
whilst many other commodities is a profit when exchanged for the money of the
would have been shipped as raw country; and ointment, but not very costly and not
materials, rather than finished goods: much. And for the King there are brought into those
for example, the famous Damas steel, places very costly vessels of silver, singing boys,
produced in Central Asia, was beautiful maidens for the harem, fine wines, thin
probably moved as ingots before clothing of the finest weaves, and the choicest
being worked by craftsmen in ointments. There are exported from these places
Damascus (Feuerbach 2002a; 2002b). spikenard, costus, bdellium, ivory, agate and
carnelian, lycium, cotton cloth of all kinds, silk cloth,
Many have argued, particularly in the
mallow cloth, yarn, long pepper and such other things
classical world, that overland trade
as are brought here from the various market-towns”
was too expensive for low-value
(Casson 1989, Chapter 49).
commodities (Garnsey et al. 1983;
Garnsey & Whittaker 1998; Jones
1964, 841). While it is certainly likely

9|Page
that cities did not rely on long-distance trade to feed themselves, a more complex pattern of
ancient economic activity is becoming apparent (Mattingly & Salmon 2000; Paterson 1998),
including responses to risk and the movement of agricultural supplies (Decker 2009, 229-
33). It is likely that not all the goods being moved were destined to be transported vast
distances. Much of the material, perhaps the bulk of caravans, was probably shorter
distance goods: bulkier, lower-value goods moving to regional markets.

It is probable that large parts of this were driven through the mechanisms of government
and/or military supply (Duncan-Jones 1990, 49; Hansen 2012, 107-8), but there is also a
greater understanding of carrying capacities and ‘costs’ (Decker 2009, 248-57). The scale of
this movement, of domestic consumption rather than elite exchange, and the scale of social
impact that this suggests, is currently becoming clearer form the archaeological evidence
(Smith 1999). More work on the movement of materials such as ceramics (Khakimov 2011;
Bernsted 2003; Henderson et al. 2005; Jenkins-Madina 2006; Kennet 2004), glass (Fuxi et al.
2009; Zorn & Hilgner 2010; Baypakov 2011; Carboni 2001; Henderson et al. 2005) and
metalwork (Park & Voyakin 2009; Sevillano-López & González 2011) will help to build a more
complex picture of this interaction and its scale.

The existence of dedicated merchants (Boulnois 2008; Decker 2009, 233-7; Whitfield 1999),
and purpose-built market centres/bazaars (e.g. Bang 2008; Daryaee 2010; Morrisson 2012;
Weiss & Westermann 1998), testify to the scale of activity along the Silk Roads. It is perhaps
hard to see this as solely explained through small quantity high-value goods. Some have
argued that, without the high-value and portable commodities, such as silk, there would
have been insufficient profits to be made from other goods for the long-distance land routes
to Central Asia and beyond to be sustainable. But “trade in silk did not happen in isolation
but was built on a foundation of trade in other commodities, not all of it long distance and
much of it already well-established. It would be misleading also to assume that all silk from
China was for a Roman market” (Whitfield 2006). Indeed, “the demand for ordinary goods
provides an explanation for the development, success, and long-term viability of regional
trade networks. These ordinary goods—household furnishings, containers, and utensils—
are valued for their social as well as for their functional content, where social content is
expressed through decoration, form, and choice of material type. The use of goods in social
interactions provides a way for individuals to proclaim their group identity; this identity is a
key factor in creating and maintaining social bonds within a group” (Smith 1999, 109).

The scale of these interactions also impacts upon the nature of the networks and polities
that were created (for example, see McLaughlin 2009; or Young 2001). Andrew Sherratt
(2004, Fig 2), for example, examined a core, periphery and margin model that explored the
scale and nature of interactions.

Nevertheless, it may be that trade in high-value commodities is the best explanation for the
creation of sufficient wealth to fund some of the major outcomes of the Silk Roads, such as
the impressive sacred sites. At the Mogao Caves, for example, Susan Whitfield argues that:
“Dunhuang was not a large town and the basic livelihood for its long-term residents was
agriculture, unlikely to have resulted in more than a subsistence income for the farmers.
There was no known large-scale industry, nor any local mineral wealth. Yet the caves and
locally produced paintings on silk and manuscripts on paper were expensive to produce and
the colophons and inscriptions show that local people and rulers were among their major

10 | P a g e
patrons. … Wealth would also have been created by the services needed to accommodate,
feed and entertain passing merchants” (Whitfield 2006). It will be interesting to see how
these issues can be reflected in the selection of sites for nomination.

2.4 Geographical scope

This study focuses on the land routes (Figure 1). The Silk Roads spans such a large
geographic expanse (Figure 2), and a long chronological timeframe, that it presents some
significant problems with scale and resolution.

2
Figure 2. The broad study area (in green), covering some 25 million km .

Clearly the land and maritime routes were interlinked, especially through the interaction of
entrepôts and terrestrial routes bringing material to and from the ports, but there were also
a number of basic differences: the maritime routes developed later, and profoundly
impacted on the significance of the land routes; the two route-systems required
fundamentally different infrastructures; they enabled very different goods to be
transported; and very different political and economic organisations grew up to exploit
them.

An exception to this is short sea routes, for example between Korea and Japan and mainland
China, or across the Caspian Sea between Turkmenistan and Azerbaijan. These were not
about linking into the long seaborne journeys of the ‘Spice routes’ that were to be so
important in the later periods, but rather about short sea crossings that enabled these areas
to connect to the main land-based networks of the Silk Roads. In addition, the exploitation
of water-born transport in places was inter-linked with the ‘land routes’, as in the
movement of people and goods along the Syr Darya in Central Asia.

11 | P a g e
The study focused on a broad zone of movement from China to the Mediterranean through
Central Asia, but extending southward to encompass the crucial routes connecting South
Asia with this zone (Figure 3).

Figure 3. The study area (green) with major (red) and other significant (orange) routes.

For the purposes of this study we initially took the east-west extent to be Chang’an
(modern day Xi’an in China) to the Eastern Mediterranean (e.g. Antioch in modern day
Turkey). This is a distance of 6,461km (Figure 4), but closer to 7,250km by the shortest
practical route (Figure 5). There is no doubt that the study could be extended westward,
through Turkey to Istanbul, and on to Italy; or eastward, to Korea and Japan (Nara lies some
8,700km from the eastern Mediterranean). However, as a start, this was considered a
sufficiently broad study area.

Defining the north-south extent of the study was also problematic. As already noted, the
complex exchange and movement patterns of the Silk Roads includes considerable
movement north-south as well as east-west (section 2.2). To practically constrain the
enormous scale of this study it was decided to restrict the northern extent to routes along
the Tian Shan mountain piedmont and along the major river systems of the Syr Darya and
Amu Darya: this effectively excluded the so-called ‘steppe routes’ further north (Bálint
1989), which means that the routes through Mongolia9 and northern Kazakhstan, for

9
“For a time in the mid-thirteenth century the Mongol capital of Karakorum, deep in Mongolia, was the most
important single stopping point on the Silk Roads.” (Christian 2000, 18)

12 | P a g e
example, have not been explored. Similarly the study did not extend northward into
Armenia, Azerbaijan, etc. or cover the areas north of the Caspian or Black Seas, although
these are clearly relevant.

There has also been no attempt to extend southward in the Mediterranean zone, to
included Jordan, Palestine, etc., despite the obvious trade routes radiating though these
regions and into Africa (e.g. Mattingly 2013).

Figure 4. Chang’an (China) to Antioch (Turkey): 6,478 km ‘as the crow flies’ (Google Earth image Landsat © Orion-ME
2014 © Google 2014 [accessed April 2014]).

13 | P a g e
Figure 5. Chang’an in China to Antioch in Turkey, by perhaps the most direct route taking into account geographic
obstacles: closer to 7,250 km (Google Earth image Landsat © Orion-ME 2014 © Google 2014 [accessed April 2014]).

Even given these limitations the current area encompassed within the study comprises
some 25 million km2 (Figure 2), and within this area we have mapped over 75,000 km of
significant routes (Figure 3; sections 5.1 and 6.1).

2.5 Chronological scope

There is evidence for long-distance trade from the 4th millennium BCE; for example, lapis
lazuli from the Chagai mountains, in modern-day Pakistan, was traded 2,000 km to lowland
Mesopotamia (Sherratt 2004), and by c 2500-2000 BCE a web of trade routes connected the
cities of the Mediterranean with those in western India (Tomber 2008). From mid-3rd
millennium BCE onwards there were also complex movements of agricultural crops (Fuller
2009; Spengler et al. 2014). By the 6th century BCE onwards there is evidence of the
movement of some goods between China and the Mediterranean zone (Christian 2000, 5-6).

However, quantitative change seems to have occurred in the last centuries of the 1st
millennium BCE, with sustained exchange/trade across the Central Asia region, linking China
with these pre-existing Afro-Eurasian exchange networks. I would also argue that there is a
qualitative change as well – away from socially-embedded exchange dominated activities
and towards trade.

During the reign of the First Chinese Emperor, Shi Huangdi (reigned 221–210 BCE), policies
were developed that began a new era of more frequent and longer journeys (Gosch &
Stearns 2008). From the 2nd century BCE, China was unified under the Qin and then
particularly the Han rulers. This coincided with, and was probably in part a response to, the
unification of many of the northern nomads in the Xiongnu (Hsiung-nu) confederation.

14 | P a g e
Emperor Wudi augmented this by dispatching Zhang Qian on a mission to Central Asia in
138 BCE, thereby opening up further contacts.

The Silk roads gained momentum in the 1st BCE to 3rd century CE with four contiguous
empire systems - the Roman, Parthian, Kushan, and Han – along with the nomadic
confederation of the Xiongnu, developing long-term connections. The latter are significant,
as the interactions between nomadic and settled agricultural communities was an important
component of the dynamics of the Silk Roads throughout its history. The relationships
between major civilizations, and between these and complex nomadic societies, is woven
into the processes of empire creation and destruction along the Silk Roads. These
relationships were not static and fluctuations in connections sustained and changed
societies and empires fortunes.

Activity peaked in the 8-9th centuries CE, once again with large empires across its span
aiding the process: including the Islamic empires of the Middle East and Central Asia, the
Tang Dynasty in China, and the Byzantine Empire in the Mediterranean. From about this
time, however, seaborne trade from ports in Fujian and Guangdong began to flourish.
Nevertheless, the Silk Roads experienced another surge in activity under the Pax Mongolica
in the 13th and 14th centuries CE, but the importance of the land routes declined again after
the break-up of the Timurid Empire in the early 16th century CE, with the seas routes
becoming dominant. They never entirely died; indeed, the quantity of Chinese goods still
transported to this day overland into Central Asian markets is evidence of this.

2.6 Geography, climate & vegetation

2.6.1 Ecological zones
The area of the Silk Roads encompasses a wide variety of topographic, hydrographic,
climatic and ecological regions, including the physical and climatic zones of mountain,
steppe, grasslands, deserts, river valleys and deltas (Figure 6). These geo-ecological zones
lay at the core of some of the most important Silk Roads interactions, and their significance
is reflected in the naming of sections of the route; such as the grass route, the steppe route,
and the oasis route. These factors had a significant impact upon the nature of the routes:
including the specific choice of routes (mountain passes, desert margins, river crossings,
etc.), the seasonality of routes, and the ability to sustain major population centres along the
routes. The “trans-ecological” nature of the routes (Christian 2000, 22) may have also been
of major significance, providing an impetus for the movement of goods and materials
between different zones. Silk Roads empires and societies adapted to the ecological context
of regions, but they also exploited the contrasts between them to develop long-distance
exchanges.

Defining geographic and ecological regions within the Silk Roads is a complex issue,
accentuated by changing climate, hydrology and ecology over time. However, a
fundamental issue has always been access to water: both directly to sustain life and to
irrigate pasture and crops to support the development of populations. Significant
populations were supported, both in a dispersed form in the grasslands and in sedentary
concentrations in the oases, river valleys and deltas. Recent research has highlighted the
close interaction between pastoralists and agriculturalists (Waugh 2008).

15 | P a g e
Figure 6. The diversity of ecological and topographic zones that the Silk Roads passed through.

There were also significant fluctuations in temperature across these zones: within major
seasonal variations days, and significant daily fluctuations.

2.6.2 Topography
The high mountains, such as the Pamirs, the Altai and the Tian Shan, and the extensive
deserts along the routes (Figure 7), were formidable obstacles to travellers. There were
significant variations in elevation, which is seldom considered except in the context of
mountain passes. For example, the route from Anxi to Kashgar through the Turfan
depression would have required a significant climb c 1,400m.

2.6.3 Hydrology
Central Asia is a region of interior drainage, with steppes and mountains, desert floodplains
(Turania) and their surrounding borders (Caspian Sea and Kopet-Dag, Pamir, and Tian Shan
mountain ranges) (Lewis 1965). “The entire region is in the process of progressive
aridization, which started with the middle Pleistocene and continues to the present, with
occasional minor wetter phases between 1400-1000 BCE, 600-250 BCE, 900-1200 CE and
1600-1800 CE” (Sala 2003, 3).

16 | P a g e
Figure 7. The major desert areas along the Silk Roads. Even here, there is a complexity of desert forms, from the high
plateau deserts surrounded by chains of high mountains, such as the Gobi, the Djungarian, and the Taklamakan, as
opposed to the lowland dunal deserts such as the Karakum.

The climatic conditions mean that “dry farming can only be implemented in a narrow band
of the middle mountain zone (between 800 and 1700 m) of the Kopet Dag, Pamir and Tian
Shan ranges, where precipitation is mainly in the form of rain, averaging yearly 500-700 mm
and peaking in winter and spring. … In spite of the arid climate of the Turanian plains, the
surrounding high mountain ranges of Kopet-Dag (max 2900m), Pamir (max 7000 m) and Tian
Shan (max 7000 m) act as collectors of precipitation, mainly in the form of snow and ice,
which is discharged by a few rivers across piedmonts and desertic flood plains down to
interior reservoirs. … Even in desertic lowlands with yearly precipitation less than 50-100
mm, basins seasonally moistened by floods exist along the lower course of small piedmonts
streams and around mild distributaries of deltas of the large rivers” (Sala 2003, 3).
Importantly, however, “perennial and seasonal streams, with peaking regimes in spring and
summer, make the irrigation potential of Middle Asia very high, the water being sufficient
for the reclamation of vast areas for agriculture” (Sala 2003, 3-4).

Tolstov and Andrianov (Tolstov & Andrianov 1957), in their reconstruction of the water
systems in the Amudarya and Syrdarya deltas, identify a period of maximum development
between the 4th century BCE and the 2nd century CE. They estimate that a total of 2 million
ha was being irrigated in the two deltas; half with small basin schemes, the other half by
permanent reservoirs (4 times the area currently under permanent irrigation). These
represent an example of the outstanding adaptation of local conditions to support the cities

17 | P a g e
of the Silk Roads. Sala (2003, 3-4) established a variety of basic geo-hydrological conditions
and landscapes:

Irrigated landscapes:

x along piedmont streams

x around lowland river deltas

Stream fed water regimes:

x fed by water deposits (springs, ice): depending on stored waters, these are
characterized by stable water discharges throughout the year and by non-destructive
transgressions: their stability makes them of easier to exploit.
x fed by precipitation (snow, rain): they depend on climatic conditions which result in
tremendous variations in discharge throughout the year and year-on-year. They also
tend to destructive mudflows. Their control requires more labour and a greater skill-
base. Example: Hissar, Surkan-Darya and Kafirnigan rivers (Dyakonov 1953).

Each of these situations requires different hydrological management and different irrigation
techniques.

18 | P a g e
3 Issues
3.1 Scale of the overall study
This study is a synthesis of existing information: it was necessarily a broad sweep and so
undoubtedly there are many inaccuracies – including mis-located site - but sufficient to
establish broader patterns. The work was intended to act as a platform for more informed
local researchers in the countries to add material and refine misunderstandings. As such it
should be seen as very much a work in progress: it aims to provide a framework for debate,
discussion and addition.

3.2 Scale of existing mapping

There have been many attempts at mapping the Silk Roads, often at a very broad scale, with
many thousands of kilometres reduced to a single page in a book. The recent excellent map
by Odyssey (2011), for example, reproduces the whole route at a scale of approximately
1:19,000,000, with detailed maps of three zones, each at approximately 1:12,000,000. Most
maps in books are at considerably larger scales. In these cases mapping at this scale need
not involve more than joining the dots, from one major city to another. Necessarily such
maps also have to reproduce rather generalised courses for the routes, often roughly
straight lines between one major nodal city and another, with relatively little ability to
reflect the subtly of landscapes (except major mountain passes) or the impact of the
location of minor settlements and way-stations.

Very occasionally more detailed regional studies have produced more detailed maps of
suggested routes (e.g. Baipakov & Pidayev 2011 in Central Asia; Dussaud 1927 in Syria;
Freeman & Ahmed 2011 for the Tibetean area; Siroux 1949 in Iran; Wood 2002 for the area
around the Taklamakan).

For this study, despite the scale of the landscape involved, it was felt that we needed to
attempt to provide more accurate locations of sites, and more detail of the complexity of
the routes. There were a number of reasons for this:

i) The questions as to the distinctive nature of the routes would require a degree of
confidence about the association of place, and route, with wider landscapes and
ecology.
ii) That State Parties would wish to be able to explore routes at a more local (higher
resolution) scale, considering the inclusion of sites and landscapes.

This has significant implications for the way that we have approached plotting the routes in
the GIS mapping of the Silk Roads for this project – see section 5.1.

3.3 Sites included - and omitted

There is a problem that extensive national inventories, for many of the countries along the
routes, are not easily accessible. This means that it is very difficult to construct a picture of
the range and diversity of surviving sites and we have to rely on accessible published
sources. Given the scale of this study, inevitably there is a bias towards the more well-

19 | P a g e
known and substantial sites (perhaps arguably the more significant sites) that have been
identified and collated by regional, local and thematic surveys.

This introduces a number of variables and biases into the study:

x Areas for which there are published detailed regional surveys or inventories, such as
the caravanserai of Iran (Siroux 1949), were better represented in the data. This may
also be reflected in a greater complexity of mapped routes in these areas.
x Conversely, those areas that have not been well-studied may appear as ‘blank’, or
less intensive areas. In some cases this may be accurate, but for many others this
may simply reflect the state of archaeological knowledge. For example, the upper
Chuy Valley in Kyrgyzstan, or the routes through western Nepal – in both of these
areas there are undoubtedly many more sites and routes, representing far greater
complexity than it is currently possible to reflect on the basis of published data.

3.4 The quality of information on sites and their dating

Comparatively few sites along the length of the Silk Roads have been well explored
archaeologically. Indeed, a great many lack even basic surveys, let alone an understanding
of sequence, buried deposit survival, dating, etc.

The quality of dating evidence in particular is very variable along the routes, and often over-
simplified in the published accounts. Not only is there a paucity of excavated sequences, but
work on regional ceramic typologies, across much of the region, is still in its infancy. This,
combined with a lack of absolute dating, means that many sites have vague chronologies,
often based around historical accounts. A particular problem is the foundation dates for
many deeply stratified cities, which may be considerably more complex than currently
presented. Furthermore, published summaries tend to highlight major phases/periods of
occupation, often omitting other activities and so not providing a definitive
presence/absence record.

As a result, we need to be careful that we are not simply selecting the ‘well-studied’ and
under-estimating the potential of many other sites. Overall, it is likely to be much more
subtle and complex process of development than we are currently able to articulate.

3.5 Site locations

Many sites lack accurate locations. In part this is due to the scale of maps that the sites are
published on (see 3.2) and partly the lack of recent survey. The advent of cheap handheld
Geographic Positioning Systems (GPS) systems is likely to significantly improve this in the
coming years.

In the short-term, however, it has been possible to use widely available satellite imagery to
improve the location of sites. Where high-resolution imagery (1m resolution or better) is
available via Google Earth it has been often been possible to identify specific sites and
provide a more accurate coordinate. However, some sites fall in areas with only broad scale
imagery (10m+ resolution), while in other cases it is not always clear where the ancient site
lies beneath modern settlements. Where new coordinates have been suggested for sites

20 | P a g e
(and used in the mapping for this project) this is recorded in the project database (see
Annex 4.2 & 4.3.7.3).

3.6 Place names: ancient & modern

Many places have been known by multiple names, and in multiple languages - Turkic,
Chinese, Arabic, Hindi, Persian, Latin, etc. Many place names have changed over time, with
some historic names not securely located to modern places. There are also often variable
spellings of modern place names, with different customs and transliterations. This can lead
to considerable confusion over the association of places with specific locations, both ancient
and modern.

Different State Parties have adopted different approaches to using ancient or modern
names to identify sites. In general, the project has adopted the name and spelling used by
the State Party, with other names recorded in the database (see Annex 4.2). There are
issues in terms of the future development of this database (see 7.3.5.3).

21 | P a g e
4 Preparation of a Geographical Information System
4.1 Introduction
To assist in exploring the spread and range of sites along the Silk Roads a Geographical
Information System (GIS) was established for the Thematic Study (see Annex 4 for details).
This was necessarily a rapid exercise, largely drawing upon existing data sources, although
not all of these were available digitally and time was spent digitising some material from
existing maps and publications.

The aim was not to provide a massive new database on the Silk Roads, but rather to draw
attention to the existing sources of material and use these to help understand the range and
diversity of the sites along this massive and complex set of routes. It would be useful, in the
future, to establish a portal to provide links to this data, and to develop and adopt standards
for the geo-referencing, naming and information sources (for example, see Ciolek 2006).
These issues are considered further in section 7.3.5.3.

4.2 Data sources

Data has been combined within the GIS from a variety of sources, including published works
(4.2.1), historical accounts (4.2.2), Tentative Lists (4.2.3), published maps (4.2.4), and online
data (4.2.5).

In addition, a variety of other digital datasets were also drawn upon as part of this study,
covering climate (4.2.6), ecology & hydrology (4.2.7) and topography & physical data (4.2.8).
Selections of these data sets have been added to the GIS to provide base-maps or to help
with the characterisation of sites by landscape/climatic/ecological zones.

4.2.1 Published works on the Silk Roads

The research drew upon a wide range of existing published material, in the form of books,
journal articles, conference proceedings, etc. (see Annex 5). This was largely collated from
international sources and is dominated by works in the English language, although it
includes some key texts in French, German, Italian, Russian, etc. This was supported and
enhanced by the State Parties supplying bibliographies of key works relating to the Silk
Roads in their country. Bibliographies were received from Afghanistan, India, Iran, Japan10,
Kazakhstan11, Kyrgyzstan12, Republic of Korea, and Uzbekistan13. In addition, a number of
internet resources were used, generated by Universities, research institutions, and special
interest groups.

4.2.2 Historical accounts

Many historical accounts of the Silk Roads exist:

“For instance there are over 70,000 extant manuscripts for the Chinese section of
the Eastern Silk Roads alone including diplomatic and military reports in the Chinese

10
In Japanese and not yet translated.
11
In Russian.
12
In Russian and translated by the project into English.
13
In Russian and translated by the project into English.

22 | P a g e
 written histories of the period. Hundreds of Buddhist monks from China made
pilgrimages to India to bring back sacred texts, and their travel diaries are an
invaluable source of information. The diary of Fa Xian, for example, describes a 14-
year voyage between 399 and 414, whilst the 25 year journal of Xuan Zang lasted
from 629 to 654. There are also accounts by Persians and Turkic travellers of the
period. The Arab traveller ‘Ibn Battutah was in Balkh and Samarkand in the mid-14th
century. The most readable account of the trade in silk and its preparation is to be
found in Marco Polo’s account of his travels in China and its neighbouring countries
from 1271 to 1292. Other 13th century European visitors were Giovanni da Pian del
Carpini, sent by Pope Innocent IV in 1245–47, and William of Rubruck, a Flemish
Franciscan monk sent by Saint Louis from 1253 to 1255. In addition to these
accounts, there is a vast treasury of archaeological artefacts, including tens of
thousands of manuscripts in over twenty languages and scripts and hundreds of
inscriptions.” (UNESCO 2008a, 7-8)

For details of these references see Moore & Wendelken (2010). For a discussion of Han
sources see Hill (2009).

4.2.3 State Party Tentative Lists

The original core group of countries, the five Central Asian countries and China, have all
produced Tentative Lists focused upon the specifics of the Silk Roads. Of the countries
joining the programme subsequently only India has been able to produce a Silk Roads
Tentative List, although Iran has flagged up the intention with a broad Silk Roads Tentative
List entry (Table 1).

Submitted Date submitted Total number Detailed site

TL of sites descriptions
China 9 28/03/2008 46 8
Iran 9 05/02/2008 ? 8
India 9 20/01/2010 12 9
Kazakhstan 9 05/03/2012 31 9
Kyrgyzstan 9 19/02/2010 14 9
Tajikistan14 9 15/01/2013 8 9
Turkmenistan 9 01/03/2010 29 9
Uzbekistan 9 19/02/2010 18 9
All 158
Table 1. Summary of current status of State Party Silk Roads Tentative Lists.

14
The Tajik Silk Roads TL was originally submitted in June 2010, but did not appear on the UNESCO World
Heritage Centre website, although some of the individual sites, which were already on the Tentative List, did.
The full list was added in 15/01/2013.

23 | P a g e
Afghanistan, Iraq, Lebanon, Nepal, Pakistan, Syrian Arab Republic and Turkey have not yet
submitted a Tentative List for specific Silk Roads sites. However, where existing Tentative
List sites fit the chronology and geographical scope of the Silk Roads study 15, they have been
included (Annex 2).

4.2.4 Published map data

In addition to digitally available data (see 4.2.5 below), there have been a number of
conventionally published maps encompassing the Silk Roads. While most are not particularly
detailed, some do provide good overviews (e.g. Odyssey 2011) or more detailed accounts of
specific areas (e.g. Goskartografia 2002).

One of the most useful sources for Silk Roads data is the Historical Atlas of Central Asia
(Bregel 2003) whose collated maps form an excellent basis for further investigation. Bregel
defines a number of specific routes between sites, allowing for comparison with other
delineated maps of the Silk Roads.

The Barrington Atlas of the Greek and Roman World (Talbert 2000) is a major source of
information on the classical and late antique world. It also provides the basis for the
Pleiades digital project (see below).

There have also been detailed maps of specific areas or topics, such as The archaeological
map of Iraq (Directorate General of Antiquities of Iraq 1967) or the amazing survey by
Maxime Siroux of the Caravanserai of Iran (Siroux 1949).

In addition, some books have produced detailed maps of segments of the routes (e.g. de la
Vaissière 2005; Tucker 2003; Wood 2002).

Some of these maps have been scanned, or information from them digitised, to form part of
the GIS platform (see Annex 4.3 for details).

4.2.5 Digital data sources

A variety of digital resources have been gathered together for the Silk Roads project. Full
details of the material collected, and its availability, is provided in Annex 4.3.4.

The most important sources were:

x Historical Atlas of Eurasia Online (Annex 4.3.4.1)

x Digital Silk Roads Project (Annex 4.3.4.2)
x The Old World Trade Routes Project (Annex 4.3.4.3)
x ArchAtlas (Annex 4.3.4.4)
x Silk Roads on Google Maps (Annex 4.3.4.5)
x Ancient Cities DataBase (Annex 4.3.4.6)
x Pleiades (Annex 4.3.4.7)
x Project HESTIA (Annex 4.3.4.8)

15
For example, in Turkey: Konya, the 12th and 13th century capital of the Seljuks, and the Seljuk caravanserais
on the route from Denizli to Doğubeyazıt.

24 | P a g e
 x The International Dunhuang Project at the British Library16
x The threat to world heritage in Iraq (Annex 4.3.4.9)
x UNESCO World Heritage List 2010 (Annex 4.3.4.10).

Most of this data is freely available online. Where appropriate copies are provided as part of
the data distribution in support of this Thematic Study (Williams forthcoming a). Some of
the information platforms, such as the excellent OWTRAD Electronic Atlas of Buddhist
Monasteries, are also linked to online data platforms (in this case, a very good Wiki site
http://monastic-asia.wikidot.com/): in these cases there has been no attempt to download
all of the data but rather users are directed to the online resources (see Annex 4.3.4 for
details).

4.2.6 Climate data

The frequently used Köppen climate classification system (the most commonly used version
is the 1961 edition by Rudolf Geiger, and thus referred to as the Köppen-Geiger system) is
based on the concept that native vegetation is the best expression of climate. Climate zone
boundaries were selected on the basis of vegetation distributions, and the system combines
average annual and monthly temperatures and precipitation, with the seasonality of
precipitation (McKnight & Hess 2000, 200–1).

Based on recent data sets from the Climatic Research Unit of the University of East Anglia
and the Global Precipitation Climatology Centre at the German Weather Service, a new
digital version of the Köppen-Geiger world map of climate classification for the second half
of the 20th century has been produced (see Annex 0 for details).

There is also a range of detailed contemporary information available at WorldClim - Global

Climate Data (see Annex 0).

4.2.7 Ecological and hydrographic data

Some basic data was collected to examine broad scale ecological data. Most of the available
data works best at a regional or local level, but some broader attempts have been made to
characterise ecological zones. Given that the development of the Silk Roads clearly reflected
adaptions to ecological and climatic circumstances, and that the civilizations that flourished
along the routes did so in a dynamic relationship to such zones, it is interesting to explore
these wider relationships.

There are basic problems with this, not least that changing irrigation systems and climatic
changes have radically altered the distribution and scale of crop and pasture land within the
study area. Thus extrapolation from modern landuse data is problematic.

The best available large-scale data is the AS cropland & pasture data (Ramankutty et al.
2008), which gives a broad pattern for crop land and pasture in the year 2000 (see Annex 0
for details). Obviously this is not directly readable on to the complex development of the Silk
Roads landscapes over the past two and a half millennia, but nevertheless it does give a feel
for the broad ecological zones of fertile arable land, pasture land and desert landscapes.

16
Although no specific GIS data the resources were very helpful, especially regarding sites explored by Stein in
the Taklamakan Desert: see http://idp.bl.uk/

25 | P a g e
 Figure 8. A portion of Central Asian routes against a backdrop of the AS cropland
data.

General characterisations of deserts (see Annex 4.3.6.6) are also useful in understanding the
broader patterns of the Silk Roads.

There are also data sets of hydrology, such as World Linear Water (see Annex 0) represents
the narrow rivers and streams of the world, and World Drainage Systems (see Annex 0)
which represents major water catchment areas.

4.2.8 Topographic & physical geography

Within ArcGIS (the software used by the project – see Annex 4.3) there are a number of
freely available base maps which provide a variety of physical topographic data, at a variety
of resolutions (see Annexes 4.3.5 and 4.3.6). The maps used in the compilation of this report
primarily used ESRI ArcGIS online World Imagery (Annex 4.3.5.6), under a Creative
Commons Attribution Non-commercial Share Alike License.

The project also used the DeLorme World Base Map (Annex 4.3.5.7), World Shaded Relief
(Annex 4.3.5.4), World SRTM (Annex 4.3.6.1), and World Elevation contours (Annex 0) along
with some reference overlay data, such as World reference overlay (Annex 4.3.5.2), World
Boundaries and Places (Annex 4.3.5.3).

Furthermore, there are some useful reference layers of modern cities, such as World
Gazetteer (Annex 0) and World Cities (Annex 0).

26 | P a g e
5 Conceptual approach to mapping: nodes, route segments &
corridors
5.1 Mapping the routes
The advantage of a computer-based mapping system, over maps reproduced in books and
wall charts (section 3.2), is that it is extensively scalable, thus enabling the researcher to
zoom into a very detailed exploration of one small segment of a larger route. Thus it would
be possible to zoom in to examine a route entering a specific gate within a city, or which
side of a river the route ran along. For the purposes of this study, however, it was not
possible to drill down to such a local level for tens of thousands of kilometres of routes; to
make suggestions regarding every alternative ford or river crossing, every possible deviation
and alternative, would be many years work and in most cases unfeasible without local
knowledge. This level of refinement must await local research.

There are also some significant conceptual problems with drawing a single route:

x In some areas, especially extensive fertile zones or wide areas of grassland, there
were probably a number of alternative routes through the landscape; all still
negotiating the passage from node A to B.
x Paths may have changed, both over the centuries and seasonally: for example,
fording a river at a different point depending upon whether it was the wet or dry
season.

For this study, therefore, we adopted the approach of identifying major nodes (large cities)
along the Silk Roads; identifying segments of routes between these; and broadening these
out to represent the corridors of ‘movement and impact’ that took place between the nodes
(rather than suggesting specific ‘roads’).

5.2 Definitions: nodes, route segments & corridors

5.2.1 Node
Node: a major urban centre and usually a place where a number of routes intersect. These
centres are likely to have played an important role in administration, re-supplying,
redistribution/marketing, and production; and in reflecting the outcomes of the Silk Roads
in terms of architecture, arts, society and beliefs.

There were difficulties in systematically selecting nodes:

x It was not easy to select cities simply on the basis of their size; in part, because we
have too little data on many of them to be sure of their size at any given moment in
their histories; in part, because size relates to other aspects of urbanism: functions;
agriculturally productive capacity of the hinterland to support the population; scale
of production/access to raw materials; ideological significance (ruler’s home town,
major religious connections); etc.
x It is also problematic to select cities solely on the basis of the number of connecting
routes: some major cities developed because they control a limited number of route

27 | P a g e
 options through a region (e.g. Merv with its strategic position in the Murghab delta
controlling routes from the Oxus to the northern Iranian plain – there were few
other routes and none as direct).

These issues with the selection of nodes leave us with a number of choices regarding how to
segment routes; broadly the three models are:

x Selecting long segments, between nodes at major (‘mega’) cities. Selecting major
urban centres for the nodes has the advantage that it is likely that most of these
have already at least been identified and in many cases there has been some
archaeological work in developing their chronologies;
x Splitting routes into numerous segments between each reasonably large town;
x A combination of using urban size and route connectivity: focusing on substantial
urban centres that also act as inter-connections between routes.

See Figure 9 for an example of the approaches.

EXAMPLE: the route from Loulan to Khotan around the southern margins of the Taklamakan
Desert, a distance of just short of 1,000 km.

In model (i) this might be considered to be a single corridor between two very major cities.

In model (ii) Charklick to Khotan segment could be broken down, at Waxxari, Charchan,
Endere and Niya (so a total of 7 segments between Loulan and Khotan). The problem with
model (ii) is that we know very little about these smaller towns (indeed, in some cases, even
their location is disputed), and if we adopted this approach, we would end up with many
hundreds of segments along the Silk Roads.

In model (iii) we would have three segments17: Loulan to Miran (c 170 km); Miran to
Charklick (c 70 km); and Charklick to Khotan (c 740 km): each a substantial settlement, and
each acting as intersections of major route junctions.

17
Which is effectively what OWTRAD adopted, based on Whitfield’s work (1999).

28 | P a g e
Figure 9. Selecting nodes and segments. The principal sites between Khotan (left) and Lop Nor/Loulan (right). The green
lines are the segments identified in the OWTRAD dataset, and simply link nodes with straight lines. In red, a more
complex picture, with more settlements and a route digitised to reflect local topography (rivers, oases, etc) (from the
Historical Atlas of Eurasia).

Model (i) might seems to offer the best approach at present – given the tremendous scale of
the project, the fact that the major nodes are more likely to have been identified, located
and (at least partially) documented - combined with the concept of the junction of major
routes. Of course, segments can be grouped together, or split, in terms of selecting corridors
for inscription (see 7.2.3).

5.2.2 Segment
Route segment: the route between two major nodes, taking into account known
topographic and cultural features (see 5.3 below on how these were digitised). For example,
in Figure 9, the red route reflects the complexity of the route utilising the rivers and oases of
the region. The routes defined are not trying to reflect every path and track, but rather
provide insight into major movements of people and goods.

5.2.3 Corridor
Corridor: a uniform buffer applied to all the digitised route segments. This is intended to
cope with the potential variability of specific paths and roads between the nodes, and to
capture sites/landscapes along the segments: the latter reflecting the impact of the route in
terms of the development of way-stations, forts, bridges, smaller towns, religious sites, etc.

We tested a variety of buffer sizes against the known sites recorded in the GIS and a buffer
of 30km on either side of the route segment (i.e. a 60km wide corridor) worked most
effectively in capturing most key elements18.

The process of buffering the route segments also highlighted places where groups of sites
lay outside any designated corridor, suggesting the presence of subsidiary routes that had
not been identified from other sources.

18
It is a simple operation in the GIS, using ArcMap, to change the size of the buffer to model alternative
patterns. A buffer of 2km was used by Del & Tavernari (2009) when exploring caravanserai routes, but the
purpose here is to capture a more complex set of movements and impacts through the landscape.

29 | P a g e
5.2.4 Adapting the corridor buffer
What will be needed in the future is the local refinement of the generic 30km buffer,
replacing it with a more considered boundary polygon encompassing the specifics of the
individual segment (Figure 10). This will be particularly relevant if we use a selection of
corridors to drive forward the nomination strategy (section 7.2).

(a) (b) (c)

Figure 10. Adjusting the boundary of specific corridors. In (a) all the principal sites lie within a corridor defined by a
generic buffer 30 km to either side of the main segment defined between two nodes (yellow). In (b) some sites (in red)
lie outside the corridor. In (c) the buffer is redrawn to take into account the local topography/ecology – which constricts
the landscape available in some places, and broadens it out in others.

5.2.5 Conclusions
Obviously this is a relatively coarse approach: it cannot take into account all of the subtleties
of local topography, environment or vegetation, and undoubtedly local knowledge and
research will add to, and deepen, this picture. However, this was a viable approach when
dealing with in excess of 50,000 km of routes. We would argue that it creates a reasonably
robust tool, in terms of identifying key elements of the overall Silk Roads and patterning
their diversity (section 6.1).

5.3 Digitising routes

We digitised the routes at an approximate scale of 1:25,000.

In drawing the routes we took account of obvious topographic features (mountain passes,
well-known fords and river crossings, known bridging points, etc.), along with known way-
stations, forts, and smaller towns. Major shrine complexes were also taken into account,
although it is also clear that these were sometimes located off the main routes of
movement, perhaps taking advantage of the siting (on a hill or mountainside overlooking
such routes), or reflecting the specific desire for remote locations.

30 | P a g e
6 Analysis
The aim of this study is to provide an analysis of sites along the Silk Roads that could be used
by States Parties participating in the Serial transnational World Heritage nominations of the
Silk Roads as a basis for comparative analyses when nominating series of sites. This may
take two forms:

x Profile the distribution and distinctiveness of Silk Roads sites in order to understand
how sites are manifestations of the shifting systems of power and patronage that
prevailed over time along the Silk Roads, in relation to the organisation of flourishing
trade and the protection of trade routes.
x Define the distribution of Silk Roads sites, in order to understand:
ƒ What sites are common to the whole extent of the Roads
ƒ What sites are specific to the whole Silk Roads or to certain parts of the
Roads
ƒ What sites are unique or exceptional
ƒ Which sites are plentiful and how their form varies in time and space
ƒ What sites are persistent over time
ƒ What sites reflect specific period of history, power systems or cultural
traditions.

6.1 Routes & corridors

As discussed (section 2.4) the scale of the routes across the vast area of the Silk Roads as
defined in this study, has led to more than 75,000 km of routes being plotted. The scale of
resolution on many of these is necessarily broad-brush (section 5.3), and undoubtedly these
can and will be augmented in the future19.

In many cases there has been insufficient field research to convincingly argue the detailed
chronology of specific segments or routes. We know that different routes rise to
prominence at different times - for example, the shifting emphasis on the routes around the
Taklamakan Desert20 – but often without a great deal of precision about the fluctuations.
The Tea Horse road is just one example of the complexity of route development and
chronology (see panel below).

19
The corridors through Iraq and Syria, for example, need elaboration both in terms of character and
delineation, as and when the State Parties are able to engage with the Silk Roads process. Currently the routes
have been defined by reviewing existing literature (e.g. Dussaud 1927) or maps (e.g. a map of the Antiquities
of Iraq, dated 1967, prepared by the Directorate General of Antiquities). These give a reasonable overview of
the major settlements and monumental archaeology, but do not include way stations and smaller
settlements/structures.
20
Especially between the northern Tianshan bei lu route and the southern Nan Shan Bei Lu route (Baumer
2003; Baumer 2008; Wood 2002).

31 | P a g e
 The Tea-Horse Road

The Tea-Horse Road (sometimes called Southern Silk Road), comprised a network of horse, mule and
human caravan paths through the mountains from Yunnan Province in southwest China to South Asia.
These are argued as having developed, at least in part, around tribute rather than trade goods, but the
routes encompassed trade, migration and strategic military activity. “Salts, medicines, silver, pelts, jewels
and all manner of other goods would in time find their way along the Tea Horse Road making it what some
locals called a ‘conductor of economies’” (Fuchs 2008, 64).

The routes are often thought to have started in the early T’ang Dynasty (618-907 CE) and by the 10th
century CE a complex networks of routes linked Yunnan to India, Nepal, Tibet and central China (via Sichuan
Province) (Rapoport 2008). However, this should not mask the fact that routes between southwest China
and South Asia had operated over considerable periods of time: in the 2nd century BCE Zhang Qian
observed bamboo canes and cloth from southwest China in a market of Balkh (Afghanistan), and was
informed that merchants had bought them from India, indicating trade routes from southwest China to
India, and then on to Central Asia, were operating at that time (Gosch & Stearns 2008).

As a result, it is not useful, in the context of either this study or the nomination process, to
try to substantially refine the chronology of each route at this time. Rather the nomination
process is better focused on reflecting the main routes, over the agreed timescale (the 3 rd
century BCE to the 16th century CE), and thus attempting to reflect the scale and diversity of
the routes and their internal chronologies. The routes into South Asia, including those across
the Tibetan plateau (Tong 2013), should be a priority for inclusion, to reflect this complex
pattern.

6.2 WHS and TL distribution

6.2.1 Existing WHS
A total of 35 major sites along the Silk Roads have already been successfully nominated as
WHS (Table 2). All of these sites justified inscription in their own right, but many explicitly
drew upon their relationship to the Silk Roads (e.g. Samarkand or Palmyra), or were very
closely associated with the outcomes of the routes (e.g. Mogao Caves, Dunhuang).

Country UNESCO Name (i) (ii) (iii) (iv) (v) (vi)

Afghanistan Cultural Landscape and Archaeological Remains of the X X X X X
Bamiyan Valley
Afghanistan The Minaret of Jam X X X
China Historic Ensemble of the Potala Palace, Lhasa X X X
China Mogao Caves (Dunhuang) X X X X X X
China Mount Qingcheng and the Dujiangyan Irrigation System X X X
Iran Bam and its Cultural Landscape X X X X
Iran Bistun X X
Iran Dome of Soltaniyeh X X X
Iran Meidan Emam, Esfahan X X X
Iran Shushtar Historical Hydraulic System X X X

32 | P a g e
Iran Tabriz Historic Bazaar Complex X X X
Iran Takht-e Soleyma X X X X X
Iran The Armenian monasteries of Iran X X X
Iraq Hatra X X X X
Iraq Samarra Archaeological City X X X
Kazakhstan Ahmed Jasawi architectural complex and archaeological site X X X X
of the medieval town of Yasi-Turkestan
Lebanon Anjar X X
Lebanon Baalbeck X X
Lebanon Tyre X X
Nepal Kathmandu X X X
Pakistan Buddhist Ruins of Takht-i-Bahi and Neighbouring City X
Remains at Sahr-i-Bahlol
Pakistan Historical Monuments at Makli, Thatta X
Pakistan Taxila X X
Syria Ancient City of Aleppo X X
Syria Ancient City of Bosra X X X
Syria Ancient City of Damascus X X X X X
Syria Site of Palmyra X X X
Turkey Great Mosque and Hospital of Divriği X X
Turkmenistan Archaeological Park of Ancient Merv X X
Turkmenistan Kunya Urgench X X
Turkmenistan Parthian Fortresses of Nisa X X
Uzbekistan Historic Centre of Bukhara X X X
Uzbekistan Historic Centre of Shakhrisyab X X
Uzbekistan Khiva (Itchan Kala) X X X
Uzbekistan Samarkand – Crossroads of Cultures X X X
Totals 12 20 25 24 6 15
Table 2. Current World Heritage Sites within the study area, with some connection to the Silk Roads: showing criteria for
inscription.

6.2.2 Tentative list sites

At present there are 221 relevant sites on Tentative Lists across the study area (Annex 2).
These reflect a broad range of site types and landscapes, but with a strong emphasis on the
‘Outcomes’ of the Silk Roads (see below, section 6.3).

Geographically the sites are widely spread, although there are obvious clusters and gaps
(Figure 11).

33 | P a g e
Figure 11. Distribution of current Tentative List sites (yellow triangles) across the study area.

6.3 Silk Roads site categories

6.3.1 Definitions
The Silk Roads concept paper (UNESCO 2008a) and the draft statement of OUV (UNESCO
2009a) (reproduced here in Annex 1) proposed that the types of monuments, sites and
cultural landscapes found along the Silk Roads could be categorized under:

x Category 1 Infrastructure - facilitating trade and transportation (including

caravanserais and inns; military posts, garrison stations and fortifications; bridges;
irrigation systems; natural and cultural landmarks).
x Category 2 Production - of trading goods (including mining, metal working,
manufacturing and handicrafts, and other industrial and production sites).
x Category 3 Outcomes - such as cities, art, knowledge as a result of contact and
exchange (including trade cities, urban centres and settlements; religious, spiritual
and ceremonial sites (including shrines, caves, tombs, sites of pilgrimage); and places
of associations with political events, transfer of ideas, language, music, dance,
poetry, etc.).

6.3.2 Number of sites and their distribution

Taking a very broad classification of the TL sites against the Silk Roads categories, 29
represent aspects of the infrastructure of the routes, 183 outcomes, and none production.
Looking at the Silk Roads Thematic Study project database (see Annex 4.2), of 473 sites, 64

34 | P a g e
might be classified as infrastructure, while 409 are associated with the outcomes. Again, no
sites are specifically associated with production (Table 1 & Table 3).

Type OWTRAD Ball SR On TL TL Country

database
Category 1
Infrastructure
Bridge 1 11 3 2 IR, TM
Caravanserais/inns 684 18 16 8 KZ, KY, IR, TK,
/etc TM, UZ
Dam 1 - -
Fort 24 5 KY, TJ, TM
Bazaar 1 1 IR
Irrigation systems 1 2 CH
Outpost/station 3 1 CH
Pass/Path/Track 135 14 1 CH
Watermill 1
Category 2
Production
Category 3
Outcomes
Bas-reliefs 4
Petroglyphs 1
Religious sites
Monasteries 592
Burial Mounds 6 4 KZ
Cemetery 3 1 CH
Grotto complex 18 15 CH, TM
Mausoleum 22 13 CH, KY, TM, UZ
Religious complex 65 15 CH, IN, TJ, TM,
UZ
Caves 120
Settlements 2,112 365
Citadel 7 1 KZ
City 268 86 ALL
Others
Köshk 1 - -
Palace 5 - -
Village 1
Cultural Landscape 8 2 KY
21
Table 3. Numbers of sites within the study area, by broad type, from three different data sources: OWTRAD (perhaps
the most extensive general Silk Road dataset, with detail on smaller halting places, settlements and monasteries ), Ball’s
Afghanistan gazetteer (as an example of an intensive regional study), and the Silk Roads database complied for this
study (which focused on extracting major sites from all datasets/published sources).

21
OWTRAD nodes database identifies 2,971 places within the study area.

35 | P a g e
The distribution of these types of sites across the study area is considered below.

6.3.3 Category 1 Infrastructure

6.3.3.1 Caravanserai & way stations

The scale of way stations, caravanserai, khans, funduqs, etc.22, along the Silk Roads was
enormous. OWTRAD has identified 684 significant stopping places supporting long-distance
communication routes within the study area, including 529 caravanserais/khans, 5 funduqs,
7 hospices and 143 halting places (Figure 12).

Figure 12. Stopping places as mapped from the OWTRAD data, showing something of the scale of the sites in the central
area of the Silk Roads.

However, OWTRAD is very much a work in progress23, and detailed local inventories show
that the picture is likely to be even more complex: Siroux (1949), for example, identifies 259
caravanserais in Iran alone (compared to the 175 identified in OWTRAD). Recent research in
Turkmenistan (Williams & Wordsworth 2010) suggests that there are at least 30-40% more
stations than previously recognised. The OWTRAD coverage is patchy at present, with little
work done on China, Kazakhstan, Kyrgyzstan, Nepal, etc. (see Table 4). So the density of
sites currently known probably largely reflects where there has been intensive work

22
Variety of terms and forms: see terminology in http://www.ciolek.com/OWTRAD/trade-routes-glossary.html
23
OWTRAD has a variety of data concerning way-points along routes: some substantial settlements, some
way-stations (see, for example, the Bulianov_1999_Turkmen dataset). These are as yet unsorted and are not
included in the above figures.

36 | P a g e
(Turkey, Syria, Iran, India, etc.), or the inaccessibility or lack of data on way-stations in other
areas, and thus reflects research rather than original distributions or concentrations.

Country Number of caravanserai &

halting places listed in OWTRAD
Afghanistan 13
Bangladesh 10
China 3
Israel 9
India 162
Iraq 11
Iran 175
Jordan 3
Kyrgyzstan 3
Kazakhstan 3
Lebanon 15
Pakistan 8
Palestine 12
Syria 51
Tajikistan 6
Turkmenistan 7
Turkey 154
Uzbekistan 15
Table 4. Numbers of caravanserais and halting places listed in OWTRAD, by country.

Nevertheless, there were clearly some distinct local variations in the density and chronology
of buildings (Constable 2003; Schutyser 2012; Silverstein 2007). For example, it has been
suggested that “during the Middle Ages caravanserais were apparently not built along
Syrian desert roads, apart from rare cases. Some stopovers were built only along the
Damascus-Palmyra axes and, even in this case, hardly ever on the first stages of the road,
the nearest to Damascus. … Caravanserais concentrated along the Aleppo-Damascus road
during both the Ayyubid and Mamluk period alike” (Tavernari 2009). (This distribution is
reflected in Figure 13). However, the problem is that often only later, more substantial, way-
stations have been identified and it is probable that there was a range of earlier structures
along the routes through the Syrian deserts 24. To the east of Palmyra, however, there
appears to be a dearth of way-stations on the routes to the Euphrates25.

24
Current survey work is being conducted by the University of Bergen, augmented the map of Syrian sites
associated with trade around the Palmyra region, but this is still incomplete and the routes from Palmyra
across the desert need to be defined further.
25
Although there are a number of desert castles in the east, such as Qasr al-Hayr al-Sharqi (on Syria’s Tentative
List – a very good example of this form of structure): these probably performed a variety of administrative,
military and palatial functions.

37 | P a g e
Figure 13. Distribution of caravanserai (orange circles: based on OWTRAD data) in Syria, with the concentration of
known sites between Damascus and Palmyra, and the paucity of sites in the eastern Syrian desert.

Overall, however, some form of way-station seems to have been prevalent across most of
the study area. Indeed, the functional need was almost inescapable. Way-stations enabled
the exchange, distribution, and storage of goods, and the re-victualing of travellers with
food, fodder and drinking water. The distance between way-stations, and the types of
services they provided, was conditioned by a number of factors:

x Terrain (e.g. the differences between flat valley bottoms and steep mountain
climbs26).
x Aridity and the availability of fodder/grazing 27.
x The type of shelter required (e.g. inclement mountain terrain).
x Political context: the balance of private and state sponsorship, and issues of security,
taxation and control.

26
The mountain halts in the OWTRAD database show the complexity of the patterns and distribution: we are
only just beginning to get a handle on this.
27
The carrying capacity of any caravan was heavily conditioned by the availability of water supplies and
fodder: “Where both fodder and water were plentiful so that only food needed to be carried, and where the
provisions represented no more than 70% of the total load, one metric ton of merchandise could be
transported by 133 porters walking some 270 km in nine days. Under the same conditions, a team of four
drivers and 23 pack camels could move the same cargo across some 1060 km and do so in 38 days. However,
where neither fodder nor water were available en route, and the weather was hot, the maximum operating
ranges would shrink dramatically. Porters would walk only one day and cover a mere 23 km, while the camels
and their handlers would walk only for five days and cover no more than 150 km. … land routes were governed
ruthlessly by the spacing of dependable sources of drinking water and the prevailing temperatures.”(Ciolek
2006).

38 | P a g e
As a result, we see different types of networks develop - with different forms of
architecture, and intervals between stations – changing over both times and space -
reflecting these changing ideas and responses to control and local eco-climatic factors.
However, while the posts around the Taklamakan, for example, are physically very different
to the caravanserais of Central Asia and the Middle East, they probably performed many of
the same functions – protection, refuelling, and the control of the flow of goods and taxes.
The scale of shelter, the degree of transhipment and storage, the scale of the refuelling
(food, water, fodder), all varied – but the basic need for such way-stations did not: they
occur along the majority of the Silk Roads routes.

Where there is detailed evidence available, way-stations

often seem to have lain about one-day’s travel apart
(Siroux 1949) - where there was a need for (and Isodorus of Charax, a geographer of
infrastructure to provide) these services, they existed. the 1st century BCE/1st century CE,
However, in other areas they were more widely spaced described various posts and routes,
(more than a single day’s journey), when it was possible stretching from the Mediterranean
to camp, where supplies of water and fodder could be coast and across the Persian plateau
gleaned from the landscape, and/or where security was to the border with India, the
good. distances between them specified,
in a book entitled Parthian Stations
Two specific caravan routes are already on Tentative (Isidore of Charax - trans 1914).
Lists: Seljuk Caravanserais on the route from Denizli to
Doğubeyazıt (Turkey) and the Ghaznavid-Seljuk
caravanserai in Khorasan (Iran). These cover the Anatolian landscape and the Iranian
plateau; but additional priorities for nomination will certainly need to reflect not only the
well-developed Near Eastern systems (the Syrian Damascus-Palmyra group is another well
preserved example), but also needs to capture the desert routes in Central Asia and China,
the piedmont and valley-systems through Central Asia and western China, the broadly east-
west Sub-Continent routes of Afghanistan-Pakistan-India (&?Bangladesh), the high
mountain routes of India-Nepal-China, the complex system of way-stations in eastern China,
and the responses in the Korean peninsula and Japan.

6.3.3.2 Military posts, garrison stations and forts

Similar to caravanserai, the control of routes, whether for protection or extraction/taxation,
was a dominate feature of most sectors of the routes. In places these functions were
inseparably from way-stations/caravanserai: for example in some of the high mountain
passes. Elsewhere, specific military control points were established, for example with the
forts and watchtowers of the Taklamakan region (both to north and south of the desert).
These types of sites reflect specific regional political and social responses to the organisation
and infrastructure of the routes, and as such are an important component of capturing the
complexity and diversity of the Silk Roads.

At present the problem is that few of the smaller sites (watchtowers, small forts, military
staging posts/postal stations, etc.) have been accurately mapped and published (at least in
internationally accessible sources). State Parties are probably much more aware of the
diversity and range of such sites within their territories, but these will be extremely
important of the routes to capture within the selected corridors.

39 | P a g e
6.3.3.3 Bridges
Relatively few historic bridges have survived in anything like an authentic form along the Silk
Roads. The main concentration of published material comes from Afghanistan (11
mentioned in Ball’s gazetteer, including the monumental structure at Pol-i Malan), Iran
(including a group of some 60 bridges in Lorestan, included on the TL under ‘The Collection
of Historical Bridges’), southern Turkmenistan (1 site, Pulkhatyn, on the TL; others known),
Syria (good examples at Cyrrhus & Qanawat) and Lebanon. To some extent this reflects the
geographic conditions of this region, from western Central Asia to the Mediterranean, with
relatively deeply incised and un-fordable rivers requiring more permanent crossing points,
combined with the availability of stone building materials. Across this region, where such
structures survive, with an acceptable degree of authenticity28, they should be a priority for
inclusion in appropriate corridors.

It is likely that bridges where needed in other parts of the Silk Road: however,
documentation is remarkably scare. This may reflect different local adaptions; such as
fordable rivers, crossing points significantly changing to reflect seasonal flood waters, or
timber bridges which have not survived.

6.3.3.4 Irrigation systems

Irrigation was vital in many regions along the Silk Roads, in shaping the landscape and
providing an agriculturally productive base to sustain significant population densities. The
flourishing of urban centres, and the scale of craft specialisation and artistic outputs that
they achieved, would not have been possible in many places without complex hydrological
management.

Attention has been paid to some of the systems, particularly where they have left major
structural legacies such as the qanāts and Karez systems (see below 6.3.3.4.2), but there has
been insufficient research in many areas into the development of canals, dams and other
water-management systems.

6.3.3.4.1 Rivers, canals and & dams

Unsurprisingly there is a very wide variety of river and stream systems within the study area
(see Figure 14; and also Table 1, as an example of the range of sources in the Central Asian
region).

There are some very major drainage systems, such as the Tigris and Euphrates, the Indus, or
the Wei He, which were major sources of irrigation and urban development from well
before the rise of the Silk Roads. These river systems were often relatively easily exploited,
with large alluvial and fertile lands, and these areas saw the rise of many of the major pre-
Silk Roads civilizations. These areas were to become crucial staging points along the Silk
Roads, with important social and cultural developments taking place though the stimuli of
the Silk Roads, but it is clear that their hydraulic management, nor the resultant
urbanisation, was not a unique feature of the development of the Silk Roads.

28
Some have been substantially rebuilt in later times, given the continuity of some crossing points, and have
little historical authenticity.

40 | P a g e
Figure 14. Major linear water systems (in blue) and some of the major drainage systems (in green, such as the
Tigris/Euphrates and the Ob), against the Silk Roads (red and orange).

In contrast, a number of major glacier or mountain rainfall-fed river systems – for example
the Murghab (Turkmenistan), Amu Darya (ancient Oxus) (Turkmenistan, Uzbekistan,
Kazakhstan), Zarafshan (Uzbekistan), Syr Darya (Kazakhstan), and Yarkand and Ak-su rivers
in Taklamakan - crossed the Silk Roads and the ability to manage and exploit these became
an important facet of sustaining large-scale populations along the Silk Roads, supporting the
development of intensive urbanisation along its routes. Many of these river systems are
deeply incised and thus hard to use for irrigation along large stretches of their courses. As a
result, they often required complex engineering in order to effectively exploit them.
Crucially, although many were exploited much earlier on a smaller scale, or at their margins
(like the Murghab), it could be argued that it is the development of complex hydraulic
civilizations that enable these areas to sustain large-scale agricultural production and thus
large populations and urban centres. This in turn was a crucial feature of the development
of the Silk Roads, for without these centres of urbanism, both in terms of production and
consumption, the volume of movement along the Silk Roads would have been considerably
less, as would their overall impact.

In many cases dams were probably used to manage these many of the irrigation systems:
normally located at the headwater of a delta or upstream of the land to be irrigated, and
used to control the flow of water into channels and canals radiating from those points. But
relatively little archaeological work has been done on these. For example, we are aware of
systems at Soltan Band (Turkmenistan) to control the Murghab River and the
canals/channels of the delta, but as yet we know little of their date or construction. In part,
this is because these strategic points in the landscape are oft reused, and the traces of early

41 | P a g e
hydraulic systems has been destroyed or disappeared under, more modern adaptions.
Recent work in the Zarafshan Valley in Uzbekistan (Stride et al. 2009) demonstrates that the
construction of canal systems can be explored to good effect and the complexity of the
hydraulic management reconstructed. Any good evidence for this form hydraulic
management (possible candidates might include the Zarafshan and possibly the Chu
valleys), where dams/canal management can be effectively documented and protected,
would be a priority for State Party inclusion in the nomination process.

Area Flow Description Approx. first Researcher(s)

irrigation

Piedmonts streams

Kopet-Dag Spring-fed, Small and stable 5000 BCE Masson

max. flow in water discharges,
spring, min. easy to control
in January

Nuratau range Spring-fed, 1200 BCE Gulyamov

max. flow in
spring, min.
in January

Karatau range Spring-fed, Gulyamov

max. flow in
spring, min.
in January

Ferghana Snow and 1500 BCE Bernshtam,

ice fed, max. Latynin
flow in July.
Also
numerous
streams
spring and
rain fed.

In all the above, during the medieval period, ground-waters were exploited by creating
artificial springs through underground systems of qanāts (karez) (Smagulov & Sala 2003)

Hissar (Surkan- Snow fed, 800 BCE Dyakonov

Darya and max. flow in
Kafirnigan rivers) June, min. in
January

Usrushana Snow and 600 BCE Negmatov

ice fed, max.

42 | P a g e
 flow in July,
min. in
December

Kaunchi-Karatau Snow and In Tashkent Terenozhkin

(Churchik and ice fed, max. region 400 BCE
Arys, right flow in July,
tributaries of the min. in Arys river area
Syrdarya) December 100 BCE

Podushkin

Lowland and delta streams/rivers

Murghab and Rain fed, Both deltas lose 2200-2000 BCE Masson, Gubaev,
Tedjen deltas max. flow in water in the desert Tosi, Ceseratti
(Margiana) May, min. in floodplains. Most
August unstable water
regimes of large
rivers of Central
Asia. The Tedjen,
most irregular,
carried 26 m²/sec in
1925 and 1000
m²/sec in 1956.

Murghab delta
where first large
scale perennial
reservoir irrigation
systems

Dahistan (Atrek, Rain fed, Similar to above - 1200 BCE Masson

Gorgan) max. flow in unstable water
May, min. in regimes
August

Northern Bactria Snow fed, 2000 BCE Dyakonov,

(Surkhandarya) max. flow in Kabanov,
June, min. in Sarianidi, and
January Askarov

Southern Bactria Snow fed, 2000 BCE Dyakonov,

(Balkhab, max. flow in Kabanov,
Kunduz) June, min. in Sarianidi, and
January Askarov

43 | P a g e
Kashka-Darya Snow fed, The delta loses its ?300 BCE Dyakonov
delta (Hissar) max. flow in waters in the desert
June, min. in floodplain. The
January region is poorly
studied and
irrigation possibly
started as early as in
northern Bactria

Zerafshan river Snow and The delta 1800 BCE in the Latynin,
delta and middle ice fed, max. distributaries, and is delta, 1200 BCE Terenozhkin,
course flow in very irregular. Dies on the middle Besenval, Tosi
(Sogdiana) August, min. out 30 km before course
in January reaching the Amu-
Darya

Amu-darya (with Snow and The Amu-Darya is Akcha 1300 BCE, Tolstov,
Akcha-daria and ice fed, max. one of the muddiest Sarykamysh 600 Andrianov, Ittina
Sarykamysh flow in July, and transporting BCE
deltas) min. in rivers; average
January water discharge of
2000 m²/sec at the
exit from
mountains; 500
m²/sec lost in
evaporation; much
lost today in
irrigation practices

Syrdarya (with Snow and Syrdarya forms an 700 BCE Tolstov,

Kuvan-daria and ice fed, max. ill-defined Andrianov,
Zhana-daria flow in April floodplain; Middle course c Maksimova,
deltas) and July, numerous meanders 200 CE at Levina
min. in Chardara/Otrar
December

Semirechie Snow and Both rivers form 200 CE Senigova,

streams and ice fed, max. deltas which lose Kozhemyanko
deltas (Talas and flow in July, waters in the
Chu) min. in alluvial-aeolian
December plains of the
Moinkym desert.

Table 5. Water systems in Central Asia: rivers, streams and deltas. (After Sala 2003, 4-5.)

6.3.3.4.2 Qanāts/Karez
One particular form of irrigation and water-supply system that has received some attention
are the underground systems variously referred to as Qanāts, Karez and al-Falaj systems

44 | P a g e
(Figure 15). These were very effective in managing water resources, especially in hot
climates where open systems are prone to massive evaporation loss. They are complex and
require considerable labour to construct and to maintain.

There is extensive survival of these features across the study area. Obviously their
distribution is in part dictated by specific topographic/climate circumstances: water sources,
the need to protect against evaporation loss, etc. They are well known in Jordan, Syria, Iraq,
Iran (e.g. Bam & Yazd), Afghanistan (e.g. Herat), Turkmenistan (e.g. Nisa), Pakistan, Central
Asia (Smagulov & Sala 2003) and China (e.g. Turfan - Bertrand 2010). They are also present
in areas outside the study area, such as Azerbaijan.

Once again, there are some problems with the scale of research that has been undertaken
into these features. In many places there is inconclusive evidence for their original
introduction and construction; which is often complicated by their longevity, with some
antique examples still in use today (e.g. Herat in Afghanistan). Their use, in a developed
form, is argued from c. 5th century BCE in Achaemenid Empire (Iran, Turkmenistan and
Afghanistan), which may suggest that the technology spread from this region, although the
dating elsewhere is not well researched. However, the spread and development of this form
of water-management seems to have been a crucial factor in the development of the Silk
Roads urban centres in many regions from the eastern Mediterranean to China. At present
we know too little about subtle variations in the engineering and organisation of these
systems, but this would be a useful facet of inter-State collaboration on nominations – any
comparative analysis of these features, to support specific corridors, would be extremely
useful.

In terms of priorities: Bam (Iran) has already been inscribed on the World Heritage List and
includes an extensive qanāt system. Unfortunately Nisa (Turkmenistan) was inscribed on the
list without including the extensive qanāt systems that dominates the surrounding
landscape. It is not clear from the current Tentative List whether the Jiaohe site (China)
encompasses any of the extensive Karez systems of the region. Given the importance of
these features to the development of the Silk Roads, prioritised corridors should actively
explore the ability to encompass evidence of these systems within their nominations.

This is also a potentially fruitful area of inter-State Party collaboration. These systems will
have very specific documentation, management and interpretative needs. A new ICOMOS
thematic study regarding Water Management Heritage in desert and dry countries will
commence shortly and this may substantially develop the priorities in this area.

45 | P a g e
 Figure 15. Qanāts crossing a now desertified landscape in Central Asia (Google Earth).

6.3.3.5 Mountain passes

There are a number of mountain passes that were an integral part of the infrastructure of
the Silk Roads, enabling passage through key mountain ranges, especially in Central and
Southern Asia. Many of these, these have also become iconic to the routes: such as the Iron
Gates (between Termez and Sharisiyabz/Samarkand in Uzbekistan); Takhta-Karacha Pass
(from Samarkand to Shakhrisiabz, Uzbekistan); Irkeshtam Pass (between Kyrgyzstan’s Alay
valley and China: shortest route between China and the Ferghana valley); the Hajigak, Unai,
Shibar and Salang Passes (Afghanistan); Khyber Pass (connecting Afghanistan and Pakistan);
Khunjerab Pass (between China and Pakistan through Kashmir: linked to the Karakorum
Highway); Yangguan Pass (China: opened up by Emperor Wudi in the 2nd century BCE to
access areas west of the Yellow River in Gansu: strategic point through which passed the
caravans when travelling westward from Dunhuang to follow the southern route of the Silk
Route into Central Asia); Sangju and Han'gu Passes (China). In some places evidence of the
complexity of the physical landscapes survive without major modern interventions,
particularly in the more remote mountain passes of the Pamirs, Tian Shan, and Himalayas.

Not only were these passes integral to the routes, but the control of these key strategic
points attracted a number of empire systems and led to the establishment of a number of
forts and watchtowers (c.f. 6.3.3.2). There is, unsurprisingly, considerable diversity in the
nature of these controlling mechanisms, based upon empires and local adaption. There is
some archaeological evidence that reflects the use of these strategic points: for example,
the watchtowers at Yangguan (China), Baltit Fort (on the Karakuram Highway from Gilgit)
(Pakistan), or the mountain forts and settlements of Ladak (India).

Priorities for nomination might be corridors that encompass good evidence of the physical
landscape and adaption of the mountain passes. This would be particularly useful if this can

46 | P a g e
be combined with good evidence of local adaptions of control (forts, watchtowers, types of
settlements).

6.3.3.6 Landmarks, markers & pillar stones

There are likely to have been many forms of markers along the Silk Roads, commemorating
events, recording empires, providing guidance on routes, edicts of control or warning. But
by their very nature, many were relatively ephemeral and there is a paucity in the
archaeological record of what was probably once a common feature of the Silk Roads
landscape.

Prominent landmarks, in the form of wayside shrines and religious sites, forts and
watchtowers, way-stations and watering-points, would have made up an important part of
the visual reference of the routes for many travellers. There are some notable examples of
locating other forms of monuments to specifically impact upon the traveller, such as the
bas-reliefs rock edicts at Shahbazgarhi and Mansehra (Pakistan), which were clearly
carefully located alongside the trade routes connecting the Vale of Peshawar with the
valleys of Swat, Dir and Chitral to the north and the city of Taxila to the south east.

Perhaps the most amazing survival and regional adaption are the pillar stones of Nepal and
India. These provide both route markers but also reinforce the political patronage of the
routes. Interspersed with small forts, these routes are a remarkable survival of the
complexity of route infrastructure and control. These, and other regional manifestations of
way-markers, would be extremely useful to capture in the nomination process.

6.3.4 Category 2 Production

As mentioned in the discussion of the current Tentative Lists (6.2.2), no sites were put
forward solely on the basis of being production sites. In reality, most productive activities –
crafts, textiles, dying, ceramics, metalworking, wood-working, etc. - were embedded within
other forms of settlement; especially within, or on the periphery of, urban centres. As such
these activities are being captured in the nomination process, but not as overtly as this
categorisation perhaps intended.

To counter-balance this, it would

perhaps be useful for nomination Example: Merv (Turkmenistan) has good evidence of
corridors to actively consider the nature metal-working (copper-alloy, crucible steel, etc), glass,
of the productive elements in the specific pottery types, etc., all of which can be
settlements along the corridor: to demonstrated to have been significant impact upon the
emphasise, and provide supporting scale of regional, and in some cases long-distance,
evidence for, specific productive trade. These can be linked to documented and extant
elements that characterise the socio- industrial zones surviving in the archaeological deposits
political manifestation of the corridor – of the site.
such as particular craft specialisations
(types of pottery industry, metal-
working, etc.).

There are, of course, many productive activities of the Silk Roads that are not adequately
captured by urban settlements. One obvious omission is agricultural production, not simply
staple food stuffs, but the production of goods that travelled many hundreds or thousands

47 | P a g e
of kilometres: viticulture, sericulture (the very basis of the Silk Roads name), dried fruits
(such as the famous dried melons), horses, cotton textiles, etc. (see section 6.3.5.3.2 below).
Some of these may have been processed in settlements and can leave tangible remains (silk
production vats, wine presses, tanning and dying vats, etc.), but the productive landscapes
are very difficult to capture in the archaeological record.

Another obvious omission is the mining and extractive industries. The extraction of
minerals, to support craft production on the Silk Roads, was an important element in
developing specific industries and products. At present these sites/landscapes are poorly
understood and certainly under-represented in both academic studies and heritage
protection.

6.3.5 Category 3 Outcomes

6.3.5.1 Cities/towns
Cites are integral to capturing the complexity of the outcomes, infrastructure and often the
modes of production of the Silk Roads.

The scale and complexity of cities along the Silk Roads is phenomenal. Within the currently
defined geographic and chronological scope there are c 276 major cities and towns in the SR
database, ranging from mega-cities like Antioch, Merv and Chang’an, to substantial regional
centres. Some of the larger/best surviving cities are already nominated (Table 2), but these
only represent a small sub-sample of the extraordinary range and scale of urbanism along
the Silk Roads and filling the gaps will be complex. The forms of urban space varied
considerably over both time and space.

Spatially, there are differences between the eastern Mediterranean, with its classical
traditions, as opposed to the cities found in western and central Asia, created by the
interplay between the spread of classical influences, initially during the Hellenistic
expansion, with a complex mix of rectilinear street networks, classic public buildings and
spaces, and densely packed urban housing, but also reflecting features of pre-existing Asian
urbanism, such as large open areas in the corners of cities. The character of the cities also
reflected local adaption to building materials, particularly the use of earthen architecture,
and in many areas the lack of good building stone. They also reflect differing styles of
architecture and religious buildings. All of this produced some significant crossovers and
blends. Further east, we see very different forms of urbanism, including the growth of
Chinese urbanism with a very clear demarcation and organisation of urban space from the
Han dynasty onwards (Sit 2010; Wheatley 1971). There were also different attitudes to
urban and suburban, the location of industrial production, the location of the elite in the
urban space, etc.

The nature of the cities along the Silk Roads also changed over time, including the growth of
discrete neighbourhood planning in the Islamic city (Bennison & Gascoigne 2007; Wheatley
2001; Whitcomb 2007), the increasing role of order, uniformity and rank in the Chinese city
(Sit 2010; Wheatley 1971), or the changing location of palatial complexes in many forms of
urbanism, away from the core to periphery of the urban area.

48 | P a g e
It will be difficult to break this up to try to capture the Hellenised/classical city; the Central
Asia qala, the early Islamic city, the Imperial heavenly city, etc. However, that is exactly
what a thematic approach will need to do: selecting corridors with a sufficient geographic
spread, it will be possible to reflect the spread of the forms and expressions of urbanism
along the Silk Roads, and their changes over time. This needs to capture all of the key
elements/forms, and also the key chronological and social developments: so a selection or
corridors across different geo-climatic areas, reflecting the varying chronological shifts in
emphasis along the routes, will help to ensure that the wider range of responses are
encompassed.

The complexity of the urban process along the Silk Roads - for example see the complexity
of cities and settlements around the Taklamakan Desert/Tarim Basin (Di Cosmo 2000) - will
also not be reflected by simply selecting major cities: these will tend to reflect well the elite
structuring of political space, but will not necessarily reflect the full complexity of the
adaption of urban planning to local responses. There is also a need for the selected urban
centres to reflect the manifestation of craft, industrial and artistic outputs, and again this
will not always be best reflected in simply the largest urban centres: indeed, in many
instances the larger centres are performing very different functions, and reflecting very
different types of social and political elite display, to the smaller urban centres along the
routes, and thus it is vital that the selection strategy reflects and encompasses these
diversities. Again, the corridor approach should help to reflect the range of scales and
attributes.

6.3.5.2 Religions
The spread of ideas and beliefs along the Silk Roads is a key feature of the significance and
impact of these routes on world culture. The scale and diversity of ideological penetration
into societies is complex to chart from the physical/tangible remains, but chronology, speed
of movement, and the scale of institutions, can give some proxy indications of the range and
depth of the impacts. The issue of elite patronage, which supported many of the more
elaborate structures (and thus the ones most likely to be nominated), is also an important
aspect to be reflected in both nomination and interpretation strategies.

It should also be noted that a range of smaller sites, sometimes with less substantial
patronage, will also be captured within the urban landscapes and sequences (see section
6.6).

6.3.5.2.1 Buddhism
The spread of Buddhism form South Asia along the Silk Roads is well known, following the
movement of monks, merchants, and other travellers. Literary, epigraphic, and
archaeological sources reveal the growth of Buddhist monasteries from c the 5th century
BCE to the end of the first millennium CE. Dynamic mobility was closely linked to trans-
regional trade networks, extending from South Asia through the upper Indus into the
Central Asian Silk Roads and through the high passes and the Tibetan plateau, to Tarim
Basin and beyond (see Neelis 2011).

There is good surviving evidence of this aspect of the Silk Roads, both in terms of specific
monuments (from the westernmost Buddhist stupa at Merv in Turkmenistan to the
impressive monastic complexes in Japan) and artistic outcomes (such as the extensive and

49 | P a g e
complex wall paintings within the grottoes, shrines and monasteries of the Silk Roads).
There is also good evidence for the complex development of the religion, with different
branches of traditions and philosophies, reflected both in the artwork and the monastic
monuments (see OWTRAD monasteries - Figure 16 - see Annex 4.3.4.3). There are also
important interactions with other religions in the first millennium CE: for example, leading
to the Hindu-Buddhist interactions at monasteries such as Krakravihar in western Nepal.

Figure 16. A sample of the Buddhist monasteries from the OWTRAD data sets, displayed in Google Earth. See Annex
4.3.4.3 for details.

6.3.5.2.2 Islam
The spread of Islam was initially a very different process to that of Buddhism, with the
conquest of large portions of western and central Asia taking place during the rapid
expansion of the 7th century CE. After that early diffusion, however, came a longer period of
acculturation and interaction, with Islamic beliefs spreading through contact and travel, and
with communities developing throughout the area of the Silk Roads.

The development of Islamic beliefs is well represented in the art and architecture of the
existing Tentative Lists and other sites of the region. These tend to focus upon the later
developments of the architecture, such as the Timurid mosques and mausolea of Central
Asia, because of their outstanding survival. Early Islamic sites are much less well known,
although they are represented in the archaeology of many of the settlement sites and would
be well reflected if a range of major and minor cities, with good below-ground
archaeological survival, are included in the list (see section 6.6).

50 | P a g e
6.3.5.2.3 Other religions & beliefs
Many other religions and beliefs, such as Judaism, Manichaeanism, Nestorian Christianity,
Shintoism and Zoroastrianism also spread along the Silk Roads and had extensive impacts
upon the development of the societies along its routes. Surviving evidence for many of
these is less obvious in the existing range of archaeological sites identified in the Tentative
Lists. As with the early Islamic sites, there will be an extensive range of evidence surviving in
the archaeological record and the inclusion of a range of major and minor cities along the
length of the Silk Roads will help to capture the diversity of these impacts.

6.3.5.3 The movement of ideas

6.3.5.3.1 Technologies and inventions

The Silk Roads were also the routes along which new technologies and inventions could
travel. From the east came, printing, gunpowder, cast iron, the crossbow, the magnetic
compass, whilst from the west came engineering developments (such as bridge-building),
tapestry weaving, calendrical sciences, vine cultivation, as well as certain glazing and metal
working techniques. The transmission of glass manufacture from west to east (Fuxi et al.
2009; Razi 2010); the cobalt that the Chinese eventually used to produce blue and white
Ming ware originated in Persia; one of the most studied transferred art motif, the pearl-
edged medallion or roundel, began in Sassanid Persia and moved eastward, going all the
way to Japan. And there was also a substantial exchange of medical knowledge and
medicines.

“The Roman craftsmen who incorporated Persian

textile and stucco ornament into the mosaics of
Justinian I's Hagia Sophia, the weavers of Sogdian
textiles who used Persian and Roman motifs, as well as
Chinese ceramicists who adapted Persian shapes and Soon after the return of Zhang Qian,
motifs all provide ready examples of this phenomenon” Sima Qian states that grapes and
(Canepa 2008, 139). alfalfa from the Ferghana valley
were introduced into China: “The
6.3.5.3.2 The spread of agricultural practices people [of Ferghana] love their wine
A major impact of the Silk Roads was the movement of and the horses love their alfalfa. The
agricultural practices, and the spread of varieties and Han envoys brought back grape and
species of plants and animals. For example, the alfalfa seeds to China and the
cultivation and working of silk itself (Frederico 1997), emperor for the first time tried
cotton (Riello 2013) and vines (see panel to right). growing these plants in areas of rich
soil. Later, when the Han acquired
Archaeologically these processes can be studied, large numbers of the 'heavenly
through botanical and faunal remains from well horses' and the envoys from foreign
stratified and dated archaeological sequences. There states began to arrive with their
has been considerable research into this field (e.g. retinues, the lands on all sides of
Aubaile 2012; d'Alpoim Guedes et al. 2013; Zhang et al. the emperor's summer palaces and
2013), which is now being supplemented by research pleasure towers were planted with
into genetics into the movement and origins of both grapes and alfalfa as far as the eye
plant and animal species. For example, animal exchange could see” Sima Qian, Han Dynasty,
networks are known to have played “an important role II, 245.
in determining gene flow among domestic animal

51 | P a g e
populations. The Silk Road is one of the oldest continuous exchange networks in human
history … facilitating animal exchange across large geographical distances and
topographically challenging landscapes … Horses are known to have been traded along the
Silk Roads; however, extensive movement of horses in connection with other human
activities may have obscured the genetic signature of the Silk Roads” (Warmuth et al. 2013,
5340). The potential for future work in this area is strong: see, for example, the doctoral
research currently underway on the movement of rose species along the Silk Roads29.

In terms of reflecting them in the Tentative Lists and nomination process, this is more
difficult. Once again, protecting and conserving major sequences of urban deposits, from
both major and minor settlements, will be important here and the corridor/node approach
outlined in section 5 will assist with this.

6.3.5.3.3 Artistic & architectural

A significant impact of the movement of peoples and ideas along the Silk Roads was the
transfer of architectural and artistic styles, often resulting in complex adaptions and
reworkings. Under the Kushans, for example, Gandharan and Mathuran sculptural styles,
and early Buddhist stone-carving practices, spread throughout Asia. This gave rise to the
distinctive development of Buddhist art and architecture, but there were also significant
amalgamations with local styles.

The interplay of architectural styles is also evident in the Islamic empires, where
architectural forms incorporated many design elements for neighbouring areas, such as
Hindu stone-working, Hindu naturalistic plant motifs, Chinese linear patterning and
Byzantine domes and mosaic decoration; while influencing those areas as well, for example
the spread of the pointed arch and arabesque decoration.

6.3.5.3.4 The movement of peoples

Empire systems (see below) provide only one framework for viewing the significance of
archaeological evidence. Ethnicity and the movement of peoples provide another important
selection criterion. The various merchant diaspora, such as the Sogdians, form a vital
element of the Silk Roads. While the Sogdians never established a territorial empire outside
their homeland (in the oases and river systems of modern-day Uzbekistan and the
Tajikistan) their influence, from the 3rd to 8th centuries, came from the extensive colonies
and trading enclaves they established in the regions to the east and south, from China and
to northern Pakistan (de la Vaissière 2003; de la Vaissière 2005; Grenet 2003; Skaff 2003).

6.3.6 Conclusions: Silk Roads Categories

It seems evident that while Categories 1 and 3 are robust, with strong and diverse
archaeological and architectural evidence along the Silk Roads, Category 2 does not hold up
as well. There are virtually no specifically identified Category 2 sites, and there are no
examples on any of the current Tentative Lists. The counterpoint, however, is that many of
the Silk Roads cities contained major production areas, and this especially applies to smaller
regional centres which provided the basis for craft production (e.g. Talgar in Kazakhstan,
Park & Voyakin 2009). This makes it vital that the nomination strategy, including the

29
Robert Mattock’s research: see http://www.plantresearch-bath.org/tropical-biotech/robert-mattock/
[accessed 1 November 2013]

52 | P a g e
selection of corridors and the sites/landscapes within corridors, captures the complex
aspects of the Silk Roads economy and production.

Category 3, when examined in detail, is dominated by cities and religious monuments.

Places of association with political events, the transfer of ideas, language, music, dance,
poetry, etc., are generally poorly represented at present. However, the movement of ideas
is reasonably well represented: in art (e.g. wall paintings) & architectural innovations (e.g.
dome designs), with some evidence for specific industrial technologies (e.g. glass and glaze
manufacture), the construction of observatories (e.g. astronomical knowledge), etc. The
frescoes at Kizil, for example, showing flying musicians playing flutes and four-stringed lutes,
points to some of the intangible aspects of the social exchanges that took place along the
routes. Generally, however, specific attention needs to be paid in the site/landscape
selection, and in the articulation of OUV, to the transfer of ideas (beyond religions) and
spread of technologies.

6.4 Empire systems

The changing dynamics of empires that spanned the Silk Roads is hugely complex. Major
empires - such as the Han, Tang, Kushan, Mauryan, Sassanid, Roman, Parthian, Abbasid,
Seljuk, to name but a few - held sway over considerable territories for protracted periods of
time. The importance of these systems, and their interaction, are fundamental attributes of
the Silk Roads. Empire systems not only shaped the development of settlements and
societies within their boundaries, but also the interaction and exchanges between empires
was vital in shaping identities: for example, the “processes of cross-cultural interaction
between the courts of Rome, Sassanian Iran, and Sui-Tang China were primarily concerned
with the formation and maintenance of imperial identity” (Canepa 2008, 144).

There were also numerous smaller polities, which controlled significant areas for varying
lengths of time, such as the Kasa Malla in Nepal, the Himalayan kingdom of Ladakh, Kangars
tribal confederation in Central Asia, the Nabataean kingdom in Jordan, or the Buddhist
kingdom of Khotan. And beyond this, in many places and times smaller city states held local
control, existing within, or often between, the dynamics of larger empire systems.

Not only did empire boundaries change over time, expanding and contracting in struggles
for control of neighbouring polities, but also they were often more fluid than the simple
boundaries that are often drawn on maps30. Sometimes borders were marked by hard
boundaries, such as forts, walls, boarder posts, etc.: one can hardly fail to acknowledge the
Great Wall as a tangible demarcation of imperial space, but even here the influence of the
empire systems was not always contained by these physical markers (Gaubatz 1996). Often
boarder zones were complex areas of negotiated power, without hard edges, reflecting
sometimes rapidly changing spheres of influence (e.g. Daryaee 2005; Nokandeh et al. 2012).
The tensions at the boundaries/interfaces of polities also sometimes led to architectural
innovations.

30
The concept of ‘fuzzy edges’ might also be useful here, as the sharp lines on conventional maps seldom
reflect the changing temporal boundaries, or the complex nature of some empire systems.

53 | P a g e
Empires, with their specific needs for administrative, political and military centres, and the
role of borders (including the concepts of liminality and ‘edge of empire’), has had a
profound impact upon the nature and development the urban nodes along the Silk Roads. In
addition, regional conflicts and changing markets mean that the Silk Roads were never
static. For example, we know of attempts by Rome to find routes that avoided Parthian
control, because of the steep duties imposed by them on overland trade (McLaughlin 2009,
97ff).

These variations, over space and time, needs to reflected in the selection of corridors so
that they encompass the both the major empire systems, and the borders/tensions
between them. This will also impacts upon the selection of sites/landscapes within corridors
to ensure that these reflect the diversity of settlement outcomes.

6.5 Nomads
The nomadic communities of Asia are a dynamic part of the interactions and productive
systems of the Silk Roads. The tensions between nomadic and sedentary communities are
often emphasised: usually characterised as the shift from subsistence to surplus farming,
with changing land use leading to the displacement of nomadic peoples. Agricultural
development, sedentarisation and urbanisation are seen as interlocking elements of this
process. However, recent research has tended to suggest a more complex set of
interactions, with nomadic and sedentary communities involved in the exploitation of
different resources, and with exchange systems existing between them (Juliano & Lerner
2003; Baipakov 1990; Barfield 1989; Stride et al. 2009). Indeed, Xuanzang observed at
Samarkand that nomads from the steppes further north brought their fur, cattle, and hides
to trade (Wriggins 2004, 38) – feeding into the wider Silk Roads network.

6.6 Buried archaeological sites

The case for including sites on the World Heritage List on the basis of buried archaeological
remains alone, with few visible surface signs, is always complex. There are often problems in
understanding and demonstrating the scale of survival and attributes (values) of buried
sites. Kashgar, for example, is one of the preeminent cities of the Silk Roads, a fulcrum
through which much of the east-west traffic passed: but the problem is our lack of
understanding regarding the scale of its buried archaeological resource and whether
sufficient remains survive to constitute a viable nomination.

However, the inclusion of buried sites is fundamental to representing the diversity of the
Silk Roads. In terms of geographic spread and local adaptation, a wide range of sites are
needed to reflect many aspects of the Silk Roads. In particularly, such sites provide the
opportunity to include sites that reflect:

x The spread of ideologies and religions. For example, the initial spread of Islam
(section 6.3.5.2.2) is poorly represented in the standing monuments of the Silk
Roads, but the buried remains of cities of the Silk Roads contain an unrivalled
testimony to the development of this religion. The Friday Mosque at the centre of
the city of Gyaur Kala at Merv (Turkmenistan), perhaps the first congregational
mosque constructed in Central Asia, is already on the World Heritage List by virtue of

54 | P a g e
 the inclusion of the whole city – but demonstrates the potential to reflect the spread
of religions through their impact on urban settlements, and through specific
architectures.
x Production (Category 2) and the transfer of technologies (part of Category 3).
Buried urban settlements that contain complex evidence of crafts, industries, and
production. These can both bear witness to the importance of these processes to
regional and inter-regional trade and exchange, and also to the transfer of
technologies along the Silk Roads.

55 | P a g e
7 The way forward
7.1 A single property?
The first Coordinating Committee of the Serial Transboundary World Heritage Nomination of
the Silk Roads meeting raised the question: “Should we nominate sections of the overall
cultural routes that reflect the necessary overall attributes” (UNESCO 2009b)?

The main purposes of this Thematic Study was to consider whether certain sections of the
Silk Roads, through the assembly of sites within them, were distinctive from other sections
of the Silk Roads, in terms of being manifestations of particular geo-cultural systems, and
thus whether a case could be made for nominating Silk Roads sites as a collection of World
Heritage serial properties, linked by a concept, instead of one single serial World Heritage
property.

In order to address these questions, the Thematic Study has conducted a rapid review of the
Silk Roads, using evidence that is available in an accessible published form, and compared
this to basic cartographic, topographic, hydrographic and ecological data.

What this data has shown is that there are specific geographic aspects to the distribution of
some site types along the Silk Roads (see section 6.3), reflecting very specific responses to
their geo-topographic setting (e.g. grotto sites), or the extent of different empire systems or
beliefs (e.g. the distribution of Buddhist stupa). Most site types – such as cities (6.3.5.1),
way-stations (6.3.3.1), or religious complexes (6.3.5.2) – existed over the whole vast area of
the Silk Roads, but they do differ considerably in terms of their specific form: reflecting local
building materials, architectural styles, responses to climatic conditions, ideas about the
organisation of urban space, the scale of centralised administration, and different political,
ideological and cultural regimes.

This Thematic Study has attempted to identify different corridors along the Silk Roads where
different responses, driven by both geo-climatic and socio-political forces, have resulted in
different manifestations of infrastructure, production and outcomes (categories 1-3). It
suggests (see below) that a number of the corridors could perhaps form the basis for
separate Silk Roads serial nominations, reflecting the cultural diversity of the overall Silk
Roads, whilst enabling the smaller sites of the Silk Roads to be captured within a nomination
strategy (section 7.2.3).

In practical terms, some way of compartmentalising the Silk Roads into a number of World
Heritage properties, linked by an overall concept, could be seen as highly desirable. It would
produce more manageable serial nominations and would enable these to progress at
differing paces, within an overall framework, and still maintain the concept of trans-national
cooperation that lies at the core of this endeavour.

7.2 Nomination strategy

7.2.1 Introduction
It is important to establish a viable way forward with the project which will create an
environment in which State Parties can both participate in trans-boundary activities and also

56 | P a g e
achieve clarity regarding site selection, and thus focus upon the next stages of this
nomination process.

When the World Heritage Committee considers the first Silk Roads nominations, it will need
to understand the overall nomination strategy, so that it may understand how individual
nominations fit into the larger picture.

A nomination strategy should therefore enable a clear overview of the selection of sites and
landscapes within the Silk Roads, and the scale of that selection, so that a cohesive Silk
Roads framework can be presented to the World Heritage Committee. This would then
enable separate serial nominations to be developed, each developing along their own
timetables, and involving different combinations of State Parties, with the Committee
understanding the relationship between them.

This study considers three options for selecting sites (section 7.2.2) related to themes, sites
and corridors, and concludes that a selection of corridors would provide the most effective
way forward. These corridors could be selected so that each manifests OUV through a
selection of sites that reflect the specific manifestations of the Silk Roads in that particular
corridor.

Assuming that this is a viable way forward, the study has then considered the criteria for
selecting these corridors and offers a provisional list (section 7.2.3). The selection of specific
sites and landscapes within each corridor is then discussed (section 7.2.4).

Issues regarding the compilation of nomination dossiers (section 7.2.5) and the harmonizing
of Tentative Lists (section 7.2.6) are then reviewed.

To support this process, it is suggested that there is also a need for:

x Further research (section 7.3.1).

x Documentation and national inventories (section 7.3.2).
x A variety of capacity building exercises to support this process (section 7.3.3).
x Developing the role of expert groups (section 7.3.4).
x Data sharing and accessibility (section 7.3.5).
x Funding and support (section 7.3.6).

7.2.2 Themes, individual sites or corridors?

The scale of the Silk Roads, a vast landscape covering thousands of square kilometres, is
such that it is impossible either to nominate all the surviving sites along the roads or all the
corridors.

Three possible approaches to selecting a smaller number of sites/landscapes for possible

nomination were considered:

x using a thematic approach to reflect the main attributes of the potential OUV of the
overall Silk roads;
x selecting individual sites/landscapes against the overall Silk Roads’ potential OUV;

57 | P a g e
 x or selecting a number of spatial corridors along the routes that reflect the diversity
of geo-political responses and outcomes, each of which could be considered as serial
nominations with their own OUV.

The following summarised the arguments for each of those possibilities and explains why
the corridor approach was considered the best way forward.

Themes: taking different attributes of the Silk Roads, and selecting the best sites and
landscapes to represent the potential OUV of that aspect. For example, exploring the
‘Infrastructure’ of the Silk Roads through a selection of sites that reflected way-stations,
irrigation systems, etc. Or exploring the ‘Outcomes’ of the Silk Roads by selecting sites that
reflected the great cities of the Silk Roads; the Buddhist grottoes of the Silk Roads; the
Islamic shrines of the Silk Roads; the music of the Silk Roads; etc. A range of such sites,
perhaps c 20-30 in all, would be needed to fully reflect the range of attributes. Such a
selection of sites would have the advantage of promoting trans-boundary working and the
exchange of specific expertise (for example, in approaches to the management and
conservation of Buddhist grottoes).

The obvious difficulties with this approach are that:

x It breaks up the understanding of the overall Silk Roads landscape: the

interconnectivity of the large cities, the smaller market towns, the way-stations, the
shrines and monasteries, etc. A collection of way-stations from along the route,
divorced from their landscapes, their contexts of smaller towns, hydraulic
management, natural passes, etc. seems less important.
x It will tend to omit the smaller categories of sites, going against the idea that the
whole is greater than the sum of the individual parts.
x Some of the themes – the great cities for example – span the entire route and a
single nomination of these sites would be a very complex process involving many
State Parties.
x The selection of the most appropriate sites for each theme would require a much
more detailed comparative analysis, justifying the selection. This Thematic Study
offers some evidence to begin that process, for example highlighting the diversity of
city forms along the Silk Roads (section 6.3.5.1), but these would need to be
developed in greater depth for each thematic area. This would be very time-
consuming and complex.
x Perhaps most importantly, the World Heritage Committee might not look favourably
on this approach as the Ittingen meeting (Martin & Gendre 2010) noted the need for
component parts of cultural properties to reflect clearly defined cultural, social,
historical or functional links over time and not to be a mere catalogue of sites
without an adequate definition of the functional links between the component
parts31.

31
An amendment to paragraph 137 of the Operational Guidelines has been proposed to cover this.

58 | P a g e
 Individual sites within a serial nomination: selecting individual sites within countries to
reflect the potential OUV of the overall Silk Roads (effectively the existing strategy in the
compilation of the Silk Roads Tentative Lists).

The obvious difficulties with this approach are that:

x The Silk Roads as one serial nomination is of enormous scope and extent and
presents manifold management issues.
x It is hard to find cohesion in the site selection – the comparative analysis of the Silk
Roads presented in this study helps to argue for specific selections within themes
(above), but it is difficult to select, for example, one caravanserai here and another
there and make a compelling comparative analysis for their inclusion. Should there
be 5, 10, 20 or 100 along the route? How do they reflect political and social
variations, as well as responses to different hydrological and topographic needs?
Themes or corridors give a focus to such decision-making.
x Effectively thematic studies would be needed for each type of site/landscape, as
above: these would need to be undertaken to justify why any given city,
caravanserai, mosque, etc. had been selected.
x As a result, the nomination procedures would be very complex: it would effectively
require agreement on all facets of selection before there was a clear picture of the
overall nomination strategy and before any State Party would be able to proceed to
a nomination.
x This complexity and difficulty has already been understood and demonstrated by the
States Parties and this is why this Thematic Study has been undertaken.

Corridors: Selecting a number of corridors that reflect the Silk Roads’ diversity of geo-
political contexts, and each of which could demonstrate OUV, might provide an approach
that encompasses serial nominations of manageable scale. It could also reduce the scope of
comparative analyses, and should also enable the fairly rapid selection of sites.

The overall shape of such a nomination strategy could be laid out now (see section 7.2.3)
and, once agreed by the State Parties, it would be possible to progress different corridors on
different timescales. Most of these would have the advantage of promoting trans-boundary
working and the exchange of specific expertise.

Corridors focus attention both upon the major nodes of any particular segment, which will
include some of the most important cities with all their complex evidence, and upon a range
of smaller sites that capture the concept of movement through the landscape, its support
and infrastructure, and the outcomes of religious, artistic and social change. It also supports
the concept of the wider picture – that the whole is more than the parts – by capturing the
smaller scale elements of the Silk Roads.

One of the difficulties with this approach is that:

x As with any selection process, some sites/landscapes may be excluded, due to

insufficient sites having been identified. Also because of the inherent geographic
focus of corridors, it is inevitable that some very significant sites may fall within
segments not selected. In the two approaches above (themes and individual sites)

59 | P a g e
 there is a meritocracy in the selection of sites, the most important sites can be
included: in the corridor approach, a major site lying in a corridor not included in the
prioritised list is potentially excluded from the nomination process.

However this approach need not preclude individual sites being nominated where they
individually can be seen to manifest an exceptional response to the Silk Roads.

7.2.3 Selecting specific corridors

The corridors are attempting to provide an intellectual framework to discuss the selection of
a series of sites along the Silk Roads, based upon their chronological, geo-climatic,
topographic and societal factors. It is important to recognise that:

i) Just because thousands of kilometres of corridors have been identified (Figure 3) it

does not mean that all of these will be nominated. Indeed, it is envisaged that only a
subset of the geospatial coverage of the Silk Roads will ever be nominated. The
selection of which these are to be is the role of the State Parties and the Silk Roads
Co-ordinating Committee.
ii) It is possible to combine more than one corridor to form a single trans-boundary
serial nomination. The start and finish points of the corridors as defined here are
major nodal points (normally cities), but for the purpose of specific trans-boundary
project, the start and end point of any nomination project can incorporate more
than one corridor as defined below. The selection of start and end points for
nominations has to be the role of the State Parties and the Silk Roads Co-ordinating
Committee.
iii) Similarly, a nomination project can also decide to only take forward part of corridor.

This Study has identified some of the corridors that might be taken forward as serial
nominations (Annex 3Error! Reference source not found.), but this is not a definitive list.
The selection of corridors in this list tried to reflect the Silk Roads’ variations in (1) the
topographic, climatic and ecological landscapes, and also (2) its historical cultural diversity.

Selecting (1) corridors that reflect the cultural responses to topographic, climatic and
ecological variations of the Silk Roads are perhaps easiest: we are looking at observable
characteristics of contemporary landscapes and ecologies, with some extrapolation to
historic conditions (e.g. the processes of desertification). We will need to include corridors
that run through high plateau and mountain passes, fertile valleys and oasis, deserts and
their margins, coastal littorals, major river crossings, etc.

Capturing the (2) the empires and geo-politics of the Silk Roads is harder. This is partly
because we do not understand the full quality of archaeological sequences at many of the
sites, or their dating. It is not always obvious from surface remains, for example, what the
foundation date of the settlement was. For example, in Central Asia we have numerous sites
with surface remains of late 12/13th century CE (the Mongol conquest), but whether deep
beneath lie good examples of Kushan urbanism is more difficult to conclusively answer.
Nevertheless, we do actually understand much of the scale, distribution and chronology of
such empire systems (see section 6.3.5.2), and thus it is possible to argue that we can
capture attributes of these by ensuring that buried sites, with good survival, are included
(see section 6.6). We need to try to capture the major empire systems that flourished and

60 | P a g e
competed over the Silk Roads, as well as the development of the smaller kingdoms and city-
states that interspersed them.

Nevertheless, we can suggest routes that capture the diversity of the responses to the geo-
political organisation and the wider manifestations of the Silk Roads. A suggested list of
corridors for the nomination strategy is given in Annex 3.

7.2.4 Site selection within chosen corridors

The concept is that the chosen corridors will each be able to manifest their own OUV as a
significant and distinctive serial nomination within the overall Silk Roads framework. Sites
selected within the corridors will be those that reflect, in a direct and tangible way, the
specific character of the corridors as a geo-cultural ensemble. These would include sites that
can be seen to manifest (see Annex 1):

x Responses to topographical and natural features

x Urban patterns and architectural designs
x Socio-economic development
x Political events that shaped the corridor
x Religious and spiritual activities
x Achievements in science and technology
x Specific achievements in the arts (sculpture, painting, carving, etc.)

It is not suggested that all sites within any given corridor could be selected for World
Heritage nomination. Most of the corridors still represent very large landscapes (600-1,400
km in length) and the issue of scale is still problematic. It would not be realistic to include
every way-station, watch tower, fort, small town, grottoes, monastery and major city along
each corridor: it would not be feasible to document, protect, manage and conserve all the
sites along such corridors, let alone ensure effective access and interpretation. Rather, sites
would be selected on the basis that they make an outstanding contribution to the attributes
of the OUV of a particular corridor.

For those corridors selected the ‘generic’ 60km buffer (section 5.2.3) should be replaced
with a more nuanced boundary, reflecting the local topographic conditions (section 5.2.4).
This would enable sites that fall close to, but currently outside corridors, to be included. It
could also reflect more complex responses to regional conditions: for example the broader
adaptation to wide fertile river systems such as the Ganges (Figure 17).

61 | P a g e
Figure 17. The wide fertile river system of the Ganges enabled sites to develop over a broad area and there were
multiple routes across this landscape. At present two corridors have been defined, encompassing the main urban
developments and monuments, but a single wider corridor probably would be more appropriate.

Similarly, in cases where movement was constricted by the natural topography, as with
mountain passes and steep valley systems, a narrower buffer would be appropriate.

Figure 18. Mountain passes and steep-sided valleys would require a spatially much tighter definition of the corridor,
which might, in places, only extend a few hundred metres from a very discreet route. Here the Karakoram highway
(corridor 14) and through the high Karakoram (corridor 43) have been deliberately left un-buffered at this stage.

There is still a tension regarding the selection (or non-selection) of smaller sites. Shorter
corridors might allow the project to capture a greater complexity of sites; for example the
Merv to Amul section (200 km), where it could be argued that every small way-station could

62 | P a g e
be managed and protected. But overall, longer corridors capture a wider range of important
sites.

7.2.5 Compilation of nomination dossiers & an overall Silk Roads framework

The Coordinating Committee for the Silk Roads Serial Nomination, in discussion with the
UNESCO World Heritage Centre and ICOMOS advisers, will consider whether a single
nomination dossier, or a number of dossiers, should be complied for each corridor.

An overall Silk Roads concept is needed to articulate how each selected corridor contributes
to the overall Silk Roads concept (Annex 1). It also needs to be clear as to why these
corridors (as opposed to other corridors) were selected. This effectively exists, in draft, in
this report, but can only be finalised on the basis of decisions made by the Coordinating
Committee.

Each selected corridor, in its nomination dossier, would then need to provide a comparative
analysis that explained how the sites within that corridor were selected against that Silk
Roads framework and the overall Silk Roads attributes.

7.2.6 Harmonizing and developing Tentative Lists

The Tentative Lists for the various State Parties along the Silk Roads are not yet complete.
Previously, in the absence of a clear overall Silk Roads framework, the compilation of
Tentative Lists and individual site selection has proved to be very difficult. The exercise thus
far has been very useful, in identifying individual State Parties’ aspirations and in helping to
focus on the range of sites available and their perceived significance.

New Tentative Lists could now be compiled against the framework presented here, focused
on site selection within the chosen corridors. This should be a much quicker and easier
process, requiring a more focused comparative analysis within the corridor (see above). This
stage could be rapidly achieved and reviewed for each potentially nominated corridor (there
would be no need to await all corridors TLs). Attention could then be rapidly focused on the
substantial work of compiling appropriate documentation, management strategies and
nomination dossiers.

7.3 Additional recommendations

7.3.1 Further research
There are a number of significantly under-represented regions in the current study (which
has been focused on the original members of the consortium: China and the five Central
Asian republics). Further work is needed to draw in more material from:

x Afghanistan; South Asia (especially Nepal, Pakistan, Bangladesh); Mongolia; the

eastern Mediterranean/Middle East; the Caucasus; and the eastern extent of the
routes, into Korea & Japan.

This research will undoubtedly produce a more complex picture of corridors, potentially
contributing to the selection of further corridors in those areas.

63 | P a g e
7.3.2 Documentation and national inventories
There is an urgent need to promote work on national inventories 32 in support of site
selection within the selected corridors.

Many of the national inventory systems are quite detailed (for example, the Monument
Passport system in Central Asia): but access to this information is very difficult - few are
computerised. A programme to computerise the monument passport system across the five
Central Asian countries, including an analysis of the structure of the current record,
inventories of databases in use, and broader standards, would be a useful platform for this
region.

In addition, large quantities of early records, vital to understanding the significance of the
sites, are not easily accessible. There are problems with: under-curated archives; archive
dispersal (e.g. the importance of the Tashkent archive, and holdings in the Institute of
Material Culture in St Petersburg, for Central Asia); and difficult formats (e.g. large scale
plans, old photographic prints, etc.).

The lack of accessible national inventories is a major limitation regarding the confidence of
the current study. One hopes that most of the important sites have been identified – but
that is not always the case.

In addition to archival inventory work, there is a need for:

x Developing documentation standards and shared approaches.

x Targeted survey/geophysics work to develop the documentation at some of the key
sites and landscapes in each selected corridor.
x The development of an integrated programme of satellite imagery acquisition (for
the nominated corridors) and its analysis.

7.3.3 Capacity building

There are a variety of areas that would be useful to support to assist the nomination
process.

Specifically:

x The compilation of nomination dossiers.

x Translation of recent UNESCO guidance document (especially into Russian for the
Central Asian partners)33.
x Archaeological archives: approaches and standards.
x GIS and spatial data: storage and analysis.
x Satellite imagery: analysis.

7.3.4 Developing the role of expert groups

The establishment of expert groups would be useful to support the development of the Silk
Roads nominations. To be effective, these should perhaps be focused around specific issues,

32
The recently agreed JFIT project in the Central Asian region is welcomed here.
33
To include: the International World Heritage expert meeting on serial nominations and properties held in
Ittingen, Switzerland (Martin & Gendre 2010).

64 | P a g e
such as data warehousing, site survey/documentation, satellite imagery, nomination
dossiers, etc. Such focused groups might provide a means of encouraging dialogue, debate
and shared practices. There is significant local enthusiasm for the Silk Roads project, and this
needs to be mobilised.

It is likely that funding for meetings will be limited, but ‘virtual’ groups, connected via email
and utilising an effective platform of data warehousing (see 7.3.5.2) to exchange
information and drafts, could make considerable progress.

There are issues here of administration of such facilities, and translation of material, but
these are surmountable (see section 7.3.6).

7.3.5 Digital data dissemination

7.3.5.1 Introduction
“Expert groups formed by the States Parties may provide detailed information about
proposed sites to the Information Centre at the Secretariat office at Xi'an for a data base
which could be shared by all the States Parties for harmonizing the Tentative List” (UNESCO
2009a, Annex 1). This aspiration has failed to develop, for a variety of reasons, but mainly
the lack of focus regarding the next steps with the nomination strategy, and thus clarity as
to what was needed.

It is now evident that it would be extremely useful to develop a centre, or centres34, to:

x Disseminate existing bibliographic data (see Annex 5), digital databases and GIS data
(see Annex 4), and other existing information relevant to the Silk Roads;
x Enable State Parties to add to/enhance the existing data, be that bibliographic (see
section 7.3.5.4), site locations, site information, etc.;
x Exchange material amongst expert groups;
x Exchange material amongst State Parties as part of the nomination dossier
compilation.

7.3.5.2 Data exchange & warehousing

Key issues in developing such a platform include:

x Security for storage and backup.

x Procedures for adding and updating information.
x Staff/time implications for the host(s).
x The degree of functionality provided (ranging from simply warehousing to more
elaborate front-ends, maps, etc.).
x Language/translations.

34
The European Commission's Central Asia Research and Education Network (CAREN) recently provided €5m
to increase the internet capacity available to researchers in the Central Asia region (Kyrgyzstan, Tajikistan and
Turkmenistan) which may make web based delivery of Silk Roads data more achievable: for the Press release
see:
http://europa.eu/rapid/pressReleasesAction.do?reference=IP/10/1152&format=HTML&aged=0&language=EN&guiLangua
ge=en

65 | P a g e
Options (not exclusive) for hosting data currently consist of:

ICOMOS International Conservation Centre in Xi’an (IICC-X)35: established as a Silk Roads

data centre and the Silk Roads Cloud, although this is not currently publicly available. There
are password protected areas and it was initially designed to provide support to China and
the five Central Asian republics.

International Institute for Central Asian Studies (IICAS)36: based in Samarkand, Uzbekistan, it
was established in 1995 as a direct outcome of the UNESCO Silk Roads Expeditions. It aims
to bring historical and cultural issues about Central Asia to the attention of the international
community and to strengthen collaboration between local scholars and colleagues abroad.
In 2008 Azerbaijan, China, Iran, Kazakhstan, Korea, Kyrgyzstan, Pakistan, Tajikistan, Turkey
and Uzbekistan were full members of the consortium.

Silk Roads Cultural Heritage Resource Information System (CHRIS) 37: Development of an
Information System to assist in the nomination of the serial World Heritage Site, specifically
focused on the Central Asian countries. The site has a variety of access and protection
levels, which would be ideal for the identified activities. It is not designed, at present, to be
a long-term repository for data (although website and data externally hosted with good
security and backup in place), but it may act as an effective short-term delivery system until
a longer-term platform can be established.

7.3.5.3 Site names and referencing

There is a need to develop standards for the referencing of sites and other features of the
Silk Roads. Not least are the problems of the multiple names and spellings of individual sites,
and understanding the accuracy with which they are located.

Data standards are essential, in terms of metadata and terminology, if the data being
developed across many research projects and initiatives are to be useful to the widest range
of researchers. This is not about shaping the content, or monitoring the accuracy, of specific
research databases, inventories, etc.: this is about establishing a means of sharing and using
information, while encouraging individual researchers and groups to develop new
resources. Matthew Ciolek (2006) has already gone some way to establishing a baseline of
information regarding nodes and routes which could provide a platform for future
development.

7.3.5.4 Bibliographic data

The bibliographic information (Annex 5) will grow and develop in the coming years. Many of
the State Parties have not yet had an opportunity to add their material and undoubtedly
their input will ensure a wider range of source material.

It would be very useful, therefore, to hold a centralised version of a bibliographic database,

in one of the data warehouses discussed above, to which new material could be added.

Two enhancements might be considered:

35
http://iicc.org.cn
36
http://portal.unesco.org/culture/en/ev.php-URL_ID=35410&URL_DO=DO_TOPIC&URL_SECTION=201.html
37
http://www.silkroad-infosystem.org/

66 | P a g e
 x The bibliography is currently organised under a number of basic headings. Works
that pertain to wider Silk Roads issues, or cover more than one country, are not
listed under individual countries, to avoid repetition. A key-worded version, to
enable searching for themes, places or countries would be relatively simple and a
very useful additional project.
x At present the bibliography is primarily in English, but elements of it also exist in
Russian and Chinese (for example). Translation (especially of key words), would be
useful.

7.3.6 International funding and support mobilized by UNESCO

There needs to be careful consideration of the opportunities for international assistance of
the Silk Roads project; especially once the Silk Roads framework and nomination strategy
are agreed, as it is hoped that this will provide an effective focus for future action.

A number of capacity building and enhancement activities have been identified (above) and
these should be discussed and prioritised at forthcoming meetings of the Silk Roads
Coordinating Committee.

In addition, the first meeting of the Silk Roads Coordinating Committee identified the need
to support “the establishment of a separate post of ‘project manager’ to oversee the whole
nomination project in individual countries” (UNESCO 2009b). While these posts may be
created by local State Party resources, there is a need for their integrated training and
support. The suggestion of a nomination dossier mentor, made at the Almaty 2009 meeting,
is worthy of consideration.

In addition, the first meeting of the committee identified the need to support “the
establishment of a separate post of ‘project manager’ to oversee the whole nomination
project in individual countries” (loc cit). While these may be created by local State Party
resources, there is a need for their integrated training and support. The suggestion of a
nomination dossier mentor, made at the Almaty 2009 meeting, is worthy of consideration.

67 | P a g e
Annexes

68 | P a g e
1 Draft Concept Statement for the Silk Roads
(From: 5th UNESCO SUB-REGIONAL WORKSHOP ON THE SERIAL WORLD HERITAGE
NOMINATION OF THE SILK ROADS, 18 – 24 May 2009, Almaty, Kazakhstan: final report:
drafted as a Statement of Outstanding Universal Value (OUV) before the idea of a series of
corridors was suggested. It is here presented as an over-arching Concept for a series of serial
nominations that will ultimately reflect the scope and extent of the Silk roads).

The Silk Roads are routes of integration, exchange and dialogue between East and West that
have contributed greatly to the common prosperity of humankind for almost 2 millennia.
The whole of the route is more than the sum of its constituent parts.

Flourishing in particular between the 2nd century BC and end of the 16th century AD, this
network of routes, started initially from Chang’an (present-day Xi’an) and ultimately
stretching from East Asia to the Mediterranean in the west, and down into the Indian
subcontinent, facilitated and generated a two-way intercontinental trade in a dazzling array
of trading goods. Of these, Chinese silk was among the most valuable, but it included
materials such as precious metals and stones, ceramics, perfumes, ornamental woods, and
spices in return for cotton and wool textiles, glass, wine, amber, carpets and the celebrated
horses. This trade connected various civilizations, persisted over centuries and was
sustained by a system of caravanserais, commercial settlements, trade cities and forts along
its entire length of more than 10,000 km, which makes it arguably the longest cultural route
in the history of humanity.

But much more than trading goods was transported over the network of Silk Roads.
Buddhism, Judaism, Islam and Nestorian, Christianity, Zoroastrianism and Manichaeanism
spread over the Silk Roads, Scientific and technological developments were also diffused by
these routes, for example from China, paper, printing, gunpowder, cast iron, the crossbow,
the magnetic compass, and porcelain, whilst engineering developments (particularly bridge
building), the cultivation and working of cotton, tapestry weaving, calendrial sciences, vine
cultivation, as well as certain glazing and metal working techniques spread from Central
Asia, Middle East, Mediterranean and the west. There was also a substantial two-way
exchange of medical knowledge and medicines, as well as of what are now seen as universal
fruit and other food crops. As such, the Silk Roads generated outstanding manifestations of
global significance in the realms of economy, society, culture and the environment. The
types of monuments, sites and cultural landscapes found along the Silk Roads can be
categorized under:

x Infrastructure (facilitating trade and transportation);

x Production (of trading goods); and
x Outcomes (such as cities, art, knowledge as a result of contact and exchange).
The property includes outstanding examples of types of heritage under these categories.

Attributes include:

• Topographical and natural features

• Urban patterns and architectural designs
• Socio-economic development

69 | P a g e
 • Political events
• Religious and spiritual values
• Achievements in science and technology
• Achievements in the arts (sculpture, painting, carving, etc.)
• Intangible heritage

Under Category 1 Infrastructure, the sites among others, comprises caravanserais and inns;
military posts, garrison stations and fortifications; bridges; irrigation systems; natural and
cultural landmarks.

Under Category 2 Production the sites reflect mining, metal working, manufacturing and
handicrafts, and other industrial and production sites.

Under Category 3 Outcomes the sites include trade cities, urban centres and settlements;
religious, spiritual and ceremonial sites (including shrines, caves, tombs, sites of pilgrimage);
and places of associations with political events, transfer of ideas, language, music, dance,
poetry, etc.

70 | P a g e
2 Tentative list sites from the study area
This list below includes sites identified on specific Silk Roads Tentative Lists, produced by:
People's Republic of China, Republic of India, Republic of Kazakhstan, Kyrgyz Republic,
Republic of Tajikistan, Turkmenistan, and Republic of Uzbekistan. For countries who have
not yet submitted a Silk Roads list (Afghanistan, Islamic Republic of Iran1, Republic of Iraq,
Republic of Lebanon, Federal Democratic Republic of Nepal, Islamic Republic of Pakistan,
Syrian Arab Republic, and Republic of Turkey) sites already on the State Party Tentative Lists
that are of clear relevance (e.g. Seljuk Caravanserais on the route from Denizli to
Doğubeyazıt in Turkey) are also listed.

Tentative List sites from other countries with a potential interest in joining the project
(Republic of Armenia, Republic of Azerbaijan, People's Republic of Bangladesh, Kingdom of
Bhutan, The State of Israel, Japan, Hashemite Kingdom of Jordan, Republic of Korea, The
Democratic People's Republic of Korea, Mongolia and the State of Palestine) are not listed
here.

For full up-to-date listings see the World Heritage Centre Tentative List website page at:
http://whc.unesco.org/en/tentativelists/

Country Site Name

Afghanistan Herat Old City
Afghanistan City of Balkh (antique Bactria)
Afghanistan Bagh-e Babur
China Yulin Grottoes
China White Horse Temple (Luoyang)
China Luoyang city of Han and Wei Dynasties
China Gongyi Stone Cave Temple
China Site of the Chang'an City of Tang Dynasty
China Guoyuan - Xincheng Tomb Complex
China Ancient City of Loulan
China Buddha Subash Temple Site
China Taizang Tower
China Ancient City of Gaochang and Astana Cemetery
China Site of the Chang'an City of Han Dynasty
China Xuanquanzhi Site
China The Dagoba of Kumarajiva
China Cemetery of Northern Dynasties and Sui and Tang Dynasty in Guyuan
China Great Buddha Temple
China Niya
China Camel City Site and Tomb Complex
China Suoyang City Site and Tomb Complex
China Site of Yunen Gate and Hecang City
China Bingling Temple Grottoes (Xia Temple)
China Shuiliandong (Water Curtain Cave) Grottoes (Lashao Temple)
China Ancient City of Jiao River

1
Iran put in a broad descriptive statement regarding ‘Silk Route (Also as Silk Road)’ in 2008, but this does not
include any specific sites. However, a large number of sites on the Iranian Tentative List are relevant (see
below).

71 | P a g e
China Qian Imperial Mausoleum - Xianyang City
China Daqin Monastery Pagoda
China Kizil Grottoes
China Quanzhou City
China Ningbo City
China Mehmud Qeshqeri Tomb
China Toyuk Grottoes
China Bezeklik Grottoes
China Simsem Grottoes
China Mao Imperial Mausoleum of Han Dynasty and Tomb of Huo Qubing
China Tomb of Zhang Qian - Hanzhong City
China Maijishan Grottoes - Immortal Cliff Grottoes
China Historic City of Guyuan
China Xi'an Mosque
China Great Buddha Temple Grottoes in Bin County - Xianyang City
China Zhao Imperial Mausoleum -Xianyang City
China The Underground Chamber of Famen Temple - Baoji City
China Xingjiao Temple Pagoda
China Site of Kaicheng
China Luoyang City of the Sui and Tang Dynasties
China Mount Xumi Grottoes
China Kumtula Grottoes
China Han'gu Pass and Xiaohan Ancient Path in Han Dynasty
China Mati Temple Grottoes - Jiata Temple and Qianfo (thousand Buddhas) Cave
India Sravasti
India Arikamedu, Early Historic Site
India Ancient Site and Buddhist Stupa (Sanghol)
India Ruins of Ancient Vaishali
India Mound locally known as Burud Kot (Nalla Sopara Stupa) Maharashtra
India Buddhist remains of Kushinagar
India Kaushambi
India Remains of Vikramshila Ancient University
India Indraprastha
India Ahichhatra
India Excavated Remains of Kaveripattinam
India Ancient monastery and stupa together with adjacent land (Harwan)
India Excavated Remains at Nalanda
Iran The Historical Structure of Yazd
Iran Hegmataneh
Iran Jiroft
Iran The Collection of Historical Bridges
Iran Qanats of Gonabad
Iran Zozan
Iran Tape Sialk
Iran The Historical Village of Abyaneh
Iran Cultural Landscape of Alamout
Iran The Historical City of Maybod
Iran Historic monument of Kangavar
Iran Bazaar of Qaisariye in Laar
Iran Ribat-i Sharaf
Iran The Sepulcher Towers (Gonbad-e Ghabous)
Iran The Historical–Cultural Axis of Fin, Sialk, Kashan

72 | P a g e
Iran Firuzabad Ensemble
Iran Shush
Iran Jame' (Congregational) Mosque of Esfahan
Iran Historic ensemble of Qasr-e Shirin
Iran Taq-e Bostan
Iran Historic ensemble of Sheikh-Safi-ed-Din Ardebili
Iran St.Tatavoos Cathedral (Qara-Kelissa)
Iran Nasqsh-e Rostam and Naqsh-e Rajab
Iran Shahr-e Sukhteh
Iran Siraf
Iran Kerman Historical-Cultural Structure
Iran Kuh-e Khuaja
Iran The Historical Texture of Damghan
Iran Bastam and Kharghan
Iran Kaboud Mosque, Tabriz
Iran Tous Cultural Landscape
Iran The Historical City of Masouleh
Iran Historic-Natural Axis of Isfahan City
Iran Baba Loghman
Iran Ghaznavi- Seljukian Axis in Khorasan
Iran Ribat Mahi
Iran Khorrambad Valley
Iran The Complex of Izadkhast
Iraq Erbil Citadel
Iraq The Fortress of Al-Ukhaidar
Iraq The Site of Thilkifl
Iraq Wadi Al-Salam Cemetery in Najaf
Iraq Wasit
Kazakhstan Talgar
Kazakhstan Archaeological sites of Otrar Oasis
Kazakhstan Cultural landscape of Ulytau
Kazakhstan Karatau, Bes-shatyr (saka barrows site in Almaty region)
Kazakhstan Turkic sanctuary of Merke
Kyrgyzstan Barskoon
Kyrgyzstan Manas Ordo
Kyrgyzstan Krasnaya Rechka
Kyrgyzstan Uzgen
Kyrgyzstan Koshoi-Korgon
Kyrgyzstan Suyab
Kyrgyzstan Burana
Kyrgyzstan Tosor
Kyrgyzstan Safid Bulan
Kyrgyzstan Kochkor
Kyrgyzstan Shyrdakbek
Kyrgyzstan Shorobashat
Kyrgyzstan Manakeldy
Kyrgyzstan Khan Dobo
Lebanon Orontes Valley
Nepal Medieval Earthen Walled City of Lo Manthang
Nepal Bhurti Temple Complex of Dailekh
Nepal Medieval Settlement of Kirtipur
Nepal Sinja valley

73 | P a g e
Nepal The Medieval Town of Tansen
Nepal Vajrayogini and early settlement of Sankhu
Pakistan Archaeological Site of Ranigat
Pakistan Tomb of Bibi Jawindi, Baha'al-Halim and Ustead and the Tomb and Mosque of Jalaluddin
Bukhari
Pakistan Shahbazgarhi Rock Edicts
Pakistan Port of Banbhore
Pakistan Mansehra Rock Edicts
Pakistan Baltit Fort
Pakistan Tomb of Hazrat Rukn-e-Alam, Multan
Syria Noréas de Hama
Syria Apamée (Afamia)
Syria Tartus: la cité-citadelle des Croisés
Syria Dura Europos
Syria Raqqa-Ràfiqa: la cité abbasside
Syria Un Château du désert: Qasr al-Hayr ach-Charqi
Syria Simeon Citadel and Dead Cities
Tajikistan Yamchun Castle
Tajikistan Ancient town of Shahristan (Kahkakha)
Tajikistan Hissar Castle
Tajikistan The Site of Ancient Town of Takhti-Sangin
Tajikistan The Site of Ancient Town of Baitudasht IV
Tajikistan Mausoleum of "Khoja Mashkhad"
Tajikistan Buddhistic cloister of Ajina-Tepa
Tajikistan Palace of the governor of Khulbuk
Tajikistan Hodja Nashron Mausoleum
Tajikistan Amir Khamza Khasti Podshoh Mausoleum
Tajikistan The Site of Ancient Town of Pyanjekent
Tajikistan Mausoleum of "Mukhammad Bashoro"
Turkey Historic City of Ani
Turkey Harran
Turkey Mardin Cultural Landscape
Turkey Seljuk Caravanserais on the route from Denizli to Doğubeyazıt
Turkey The Citadel and the Walls of Diyarbakir
Turkey The Tombstones of Ahlat the Urartian and Ottoman citadel
Turkey Harran and Sanliurfa
Turkmenistan Kyone-Kaka
Turkmenistan Sakhsenem urban site (medieval Suburun) and suburb garden
Turkmenistan Mishrian
Turkmenistan Pulkhatyn bridge
Turkmenistan Old Sarakhs (urban site and Abul-Fazl mausoleum)
Turkmenistan Durun and Ak Ymam mausoleum
Turkmenistan Yzmykshir fortress (medieval Zamakhshar)
Turkmenistan Ygdy gala
Turkmenistan Amul
Turkmenistan Diyarbekir
Turkmenistan Akcha-kala
Turkmenistan Geok-Gumbez
Turkmenistan Chilburj
Turkmenistan Gebekly
Turkmenistan Khurmuzfarra (Uly Kishman)
Turkmenistan Gurtly-depe (medieval Bashan)

74 | P a g e
Turkmenistan Mele-Hairam temple complex
Turkmenistan Abiverd
Turkmenistan Devkesen (medieval Vazir)
Turkmenistan Odemerghen-gala (Rabat Suran)
Turkmenistan Injirli
Turkmenistan Anau urban site and ruins of the mosque of Said Jamal-ad-Din
Turkmenistan Meana-Baba architectural complex
Turkmenistan Dashrabat (Dandanakan)
Turkmenistan Astana-Baba architectural complex and Alamberdar mausoleum in Astana-Baba settlement
(medieval Maimarg)
Turkmenistan Dayakhatyn caravanserai
Turkmenistan Dargan urban site
Turkmenistan Ekedeshik
Turkmenistan Talkhatan-baba (mosque)
Uzbekistan Ahsiket
Uzbekistan Khazarasp
Uzbekistan Kanka
Uzbekistan Historic Center of Qoqon
Uzbekistan Ancient Termiz
Uzbekistan Shahruhiya
Uzbekistan Varakhsha
Uzbekistan Big Guldursun (Desert Castles of Ancient Khorezm)
Uzbekistan Pil Qala (Desert Castles of Ancient Khorezm)
Uzbekistan Anka Qala (Desert Castles of Ancient Khorezm)
Uzbekistan Djanbas Qala (Desert Castles of Ancient Khorezm)
Uzbekistan Ancient Pap
Uzbekistan Ak Astana-baba (mausoleum)
Uzbekistan Kurgashin Qala (Desert Castles of Ancient Khorezm)
Uzbekistan Poykent
Uzbekistan Minaret in Vobkent
Uzbekistan Chashma-Ayub Mausoleum
Uzbekistan Chor-Bakr
Uzbekistan Bahoutdin Architectural Complex
Uzbekistan Rabati Malik Caravanserai
Uzbekistan Mir-Sayid Bakhrom Mausoleum
Uzbekistan Complex of Sheikh Mukhtar-Vali (mausoleum)
Uzbekistan Andijon
Uzbekistan Koy-Kirilgan Qala (Desert Castles of Ancient Khorezm)
Uzbekistan Arab-Ata Mausoleum
Uzbekistan Toprak Qala (Desert Castles of Ancient Khorezm)
Uzbekistan Khanbandi (dam)
Uzbekistan Ayaz Qala (Desert Castles of Ancient Khorezm)
Table 6. Tentative List sites from the study area.

75 | P a g e
3 Selected corridors
A gazetteer of corridors of all the corridors identified in the study has been compiled,
drawing upon the information in the database (Annex 4.2) and GIS system (Annex 4.3). The
gazetteer will be distributed in a digital format (Williams forthcoming a & b).

3.1 Introduction
A few points of note:

x The corridors have been numbered in no particular order and numbering does not
reflect importance.
x The length of the corridor has no relationship to their potential significance, but
rather reflects their geo-political and geographic context (e.g. routes along desert
margins, along river valleys, or mountain piedmont).
x Nominations are likely to be of a sub-set of corridor, reflecting the most important
sites and landscapes that represent the OUV of the nomination.
x The principle routes have been defined in segments linking major nodes (cities).
Initially a buffer of 30km either side of the main route (i.e. a 60km wide corridor) has
been defined to capture sites/landscapes along the segments.

3.2 Selected corridors

Listed below are some of the potentially significant corridors for consideration in the
nomination strategy. The list is not meant to be exclusive.

Figure 19. Some selected corridors (or parts of corridors) highlighted in purple (details in next figures).

76 | P a g e
 Figure 20. Sample corridors in the west.

77 | P a g e
 Figure 21. Sample corridors in the central area.

78 | P a g e
 Figure 22. Sample corridors in the east.

79 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - State Parties
(km) eco-geographic cultural

1 (part) The Silk Roads: along 410 Narrow river Important route, Uzbekistan &
the Oxus Amu Darya) valley, in places with major Turkmenistan
- Dayahatyn to Kunya deeply incised, centres of urban
Urgench. with fertile zones and artistic
along littoral, development in
Case for expanding and wider fertile Khorezm.
corridor to lands in Khorezm Important range
encompass routes on oasis/Amu Darya of site types,
north bank and delta. including small
leading southwest towns, forts and
from Kunya Urgench caravanserai.
to Devkesen &
Shakhsenem.

80 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

4&5 The Silk Roads: 610 From wide fertile An iconic route. Uzbekistan,
(part) crossing the Oxus valleys, across the Strategically Tajikistan &
and the desert – desert and river, to crucial and Turkmenistan
Penjikent to Merv the Murghab supported large
delta. volume of
movement.

81 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

6 & 10 The Silk Roads: the 598 Piedmont of Interface Kyrgyzstan,

fertile valleys of Turkestan and between a Tajikistan &
Central Asia (the Pamir-Alay range, number of Uzbekistan
Zarafshan & Ferghana fertile zone empire
Valleys) – Osh to between Zarafshan systems (e.g.
Penjikent/Samarkand and Ferghana limits Parthian
valleys. Linked to Empire, with
Taklamakan desert regional capital
by high mountain at Khodzend -
passes, via Surkhab Antiochia in
valley and the Scythia), and
Alay/ Chyirchyk sometimes
pass. independent
states. Hugely
influential
cultural
exchanges and
strategic
locations

82 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

8&9 The Silk Roads: The 804 Piedmont of Fertile region China,
shores of Lake Issyk Kirghiz Mountain that supported Kyrgyzstan,
Kul & the Chu valley – range, Chu river a major Kazakhstan
Aksu to Taraz valley, and along development
southern Issyk-Kul of urban
shore line. Fertile centres and
lands with plentiful large scale
water supply. movements of
Upper section of people and
the Chu Valley has materials, over
fairly high a considerable
mountain passes, period of time.
connecting Issyk
Kul and Aksu.

83 | P a g e
No. Corridor Distance Reason (1) Reason (2) - State Parties
(km) – eco- cultural
geographi
c

11 & The Silk Roads: the heart of 255 Lower Crucial area of Afghanistan,
16 Central Asia – Hissar to Balkh Surkhon interaction, Tajikistan &
Darya with numerous Uzbekistan
River, joins important
Amu Darya cities
– a largely flourishing on
shallow north-south as
valley well as east-
route. west
Marginal connections.
desert to
fertile
oasis
Balkh.

84 | P a g e
No. Corridor Distance Reason (1) Reason (2) - State Parties
(km) – eco- cultural
geographi
c

14 The Silk Roads: the 681 Mountain Links between China &
Karakoram highway (via route, China and sub- Pakistan
the Khunjerab and Wakhir using continent
Passes) - Tashkurgan to narrow formalised under
Taxila river the Kushan Empire
valleys and – accessed chiefly
high through this
mountain corridor.
passes

85 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

17 The Silk Roads: the 387 Characterised by Reflects the China

Imperial cities – fertile plains key Han, Wei,
Louyang to Chang’an alongside major Sui and Tang
river systems, dynasty
supporting large capitals, with
urban populations. their major
impact on the
development
of Chinese
urbanism and
the arts.

86 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

18 The Silk Roads: the 1,600 Key landscape of Major China

Hexi corridor – control for most infrastructure
Chang’an to Dunhuang journeys between (way stations
the West and and forts) and
China. Narrow significant
passes, key outcomes,
strategic particularly
landscape. with regard to
religious sites.

Detail north-western section

87 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

19 The Silk Roads: the 668 Southern piedmont Crucial China

(part) northern margins of of the Tian Shan: landscape for
the Taklamakan desert largely desert infrastructure
– Kucha and then route punctuated of Silk Roads –
Kashgar with seasonal adaptation and
oases. control of this
essential to
many east-
west journeys.

88 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

20 The Silk Roads: The 1,017 Southern desert Important China

oases route south of fringe punctuated landscape for
the Taklamakan desert with seasonal control of
– Loulan to Miran and oases. routes to south
then to Khotan and west –
different
adaptations,
chronologies
and empire
systems to the
northern
routes (19).

Different
chronology of
use to
northern route.

89 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - State Parties
(km) eco-geographic cultural

22 The Silk Roads: Issyk- Complex Major irrigation China

Kul to Turfan, the landscape of systems and key
northern Tian-Shan deserts and arid development of
route basins. northern routes
and associated
socio-political
development.

90 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

26 The Silk Roads: along 729 Along the Syr Complex Kazakhstan
(part) the banks of the Darya, in a shallow development
ancient Jaxartes (Syr valley with a large of city-states,
Darya) – Zhankent to fertile zone and often quite
Shymkent via Otrar, the Aral Sea littoral independent
Sauran, etc. (much greater from southern
extent than today). areas:
important
Complex river social and
systems and runoff political
to Syr Darya. outcomes
along fertile
route, with
local
adaptations to
control.

91 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

27/54 The Silk Roads: 1,100 Fertile piedmont Characterised Afghanistan,

crossing the northern zone amongst the by well Iran
Iranian plateau foothills of the managed
Alborz mountains irrigation
and arid Northern systems to
Iranian plateau. support this
strategic and
heavily used
route.

92 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

30 The Silk Roads: 732 Temperate Important Iran & Iraq

through the mountains trans-Zargos
mountains and down punctuated by the route – heavily
to the wide valleys, down used over time.
Tigris/Euphrates – Ray to fertile zone east Substantial and
to Baghdad of the Euphrates. prolonged
urban
development.
Major political
and
administrative
centres.

93 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - State Parties
(km) eco-geographic cultural

34/36 The Silk Roads: 428 Desert route. Complex array of Syria
crossing the desert – caravanserai and
Euphrates to Palmyra routes.
and Damascus Outstanding
examples of way-
Case for two corridors stations and
between Palmyra and irrigation/water
Damascus, perhaps supply.
reflecting
chronological
variations

94 | P a g e
No. Corridor Distance Reason (1) – eco- Reason (2) - State Parties
(km) geographic cultural

35 The Silk Roads: upper 280 Eastern portion Vital route over Syria & Turkey
Euphrates to the follows the major long period of
Mediterranean - Euphrates river; time, leading
Raqqa to Antioch western section to some major
crosses fertile cities and large
plains and well areas of
irrigated broad agricultural
valleys. production.

95 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - cultural State Parties
(km) eco-
geographic

38 & The Silk Roads: down 1,392 Fertile river Issues of water India
39 the Ganges - system with management, and the
Indraprastha to wide role in this of monastic
Chandraketugarh agriculturally communities. Major
productive developments in
There is a good case zone. urbanism, sites of
for a much wider Important for learning and religious
corridor to encompass connections development/architect
38 & 39, and potential to northern ure. Major power base
a variety of other mountain for a variety of empire
routes/sites through routes and systems.
this broad fertile zone. eastward to
the sea.
Could also be
expanded to enable
trans-boundary
working with
Bangladesh.

96 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - State Parties
(km) eco-geographic cultural

40 The Silk Roads: the 980 The high Complex empire China, Nepal
(part) high Tibetan plateau plateau systems with & India
and down to the adaptation, periods of
Ganges – Lhasa to major valley isolation and
Pataliputra systems periods of major
through the connectivity; local
Complex route and Himalayas, and adaptations to
options which down into the climate and major
elements in India to fertile Ganges religious and
include. delta. urban outcomes.

97 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - State Parties
(km) eco-geographic cultural

41 & The Silk Roads: valleys 357 & 389 High Himalayan Crucial China?, Nepal
42 and mountains mountain adaptation of & India
(central and western passes, steep small routes;
Nepal) - Lo Manthang small trails and pillar stone
and Surkhet to the fertile high markers, forts
Ganges valleys. and way-stations,
along with blend
Two routes but on Hindu &
perhaps best Buddhist temples
combined as & iconography.
representing similar
issues of adaption.
Could be extended up
on Tibetan plateau.

98 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - State Parties
(km) eco-geographic cultural

43 The Silk Roads: 1,014 Winding Complex China & India

through the high mountain adaptation to
Karakoram – Srinagar, routes with harsh condition,
Leh and Sangju Pass significant producing unique
climatic and architecture,
seasonal issues cultural contacts
of adaptation. and
empire/political
controls.

99 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - State Parties
(km) eco-geographic cultural

47 The Silk Roads: Turfan 384 Turfan Complex China

to Hami depression: adaptation to
extremely hot harsh condition,
arid climate. producing
important
hydrological
management
systems.

100 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - State Parties
(km) eco-geographic cultural

51 The Silk Roads: 1,051 Complex Import Iran & Turkey

Soltaniyeh to the Black mountain and connection
Sea valley routes between Black
Sea and the heart
of Asia. Major
historic port,
caravanserai and
monastic sites.

101 | P a g e
No. Corridor Distance Reason (1) – Reason (2) - State Parties
(km) eco-geographic cultural

55 The Silk Roads: down 886 Strategic route Pakistan

(part) the Indus Valley – down to the
Multan to Banbhore coast, linking with
early maritime
connections.

102 | P a g e
4 Database & GIS
4.1 Introduction
The Silk Roads data set developed to support this Thematic Study is not a single integrated
one, but a link to the various existing data sources. As such it did not intend to replicate
what is currently available, but rather to draw attention to these sources and build upon
them.

There are two elements: a database (see Annex 4.2) and GIS digital graphic database of
maps, physical backdrops, routes, site locations, etc. (see Annex 4.3). The two are linked and
information from the database can be displayed in the GIS (see Annex 4.3.2).

The elements that have been created by the project, or that are freely available in the public
domain, will be made available as digital resources to support this publication (for details see
Williams forthcoming a).

4.2 Silk Roads Database

The database to support the GIS system was compiled in Microsoft Access 2012©.

4.2.1 The Sites table

A basic table (sites) which contains nearly 600 places at present.

Site selection: the sites included in the database reflect the most significant sites of the Silk
Roads, which fall within the chronological and geographic scope of the study, and include:

x All existing World Heritage Sites.

x All Tentative List sites – both those specifically supplied in relation to the Silk Roads
project and those put forward by State parties that have yet to compile a specific Silk
Roads Tentative List.
x Other key sites – most often major nodal settlements.
x There are many other sites which could be included in such a database, way-stations,
caravanserai, smaller towns, forts, etc. – but these are well covered in other datasets
drawn upon (e.g. OWTRAD – see Annex 4.3.4.3) and thus this data table has focused
on more major sites.

Fields: the table contains basic information on location, status, chronologies, criteria for
selection, description, links to web resources, etc. A full list of fields and their definitions is
available in the database (view the table in Design view).

Descriptive data: where available, the descriptions are taken from State Party Tentative Lists
(marked as TL) or World Heritage Site descriptions on the UNESCO website documents
(marked as WHS). These are supplemented by material from a variety of other sources and
where substantial blocks of information have been used the sources are provided.

Quality & variability: The quality of the information is currently very variable. The
descriptions are fairly rough and would ideally require considerable editing and checking.

103 | P a g e
They could also be significantly enhanced from a wider range of source material, but that is
beyond the scope of the current project.

Co-ordinates: decimal latitude and longitude used to interface with GIS.

4.2.2 Other tables

Tables of data derived from other sources have also been included. These are:

x Afghanistan_Ball (Annex 4.3.4.11)

x Iraq Antiquities Map (Annex 4.3.3.2)
x Siroux caravanserais (Annex 4.3.3.3)
x World Heritage List 2010 – trimmed to just those that fall within the study area
(Annex 4.3.4.10)
x Whitfield_1999_world_SR – part of the OWTRAD data set (see Annex 4.3.4.3)
x OWTRAD gazetteer nodes (see Annex 4.3.4.3)
x Stopping places – part of the OWTRAD data set (see Annex 4.3.4.3)
x Country codes – list of country codes
x Stein_ruins – part of the Digital Silk Roads project data set (Annex 4.3.4.2)

4.3 Silk Roads GIS

4.3.1 Introduction
Two primary tools were used for this study: Google Earth TM and Esri ArcGIS 10.2.

The existing data (see section 4.2) came in a variety of formats, most notably files in Keyhole
Markup Language (KML),or in a zipped format (KMZ) (both Google Earth formats1); or as
shapefiles or geo-referenced raster images, for use in GIS systems.

Some data sets are best suited, or only available, in one of these formats: for example, the
background physical data supplied by ESRI comes in ArcGIS (Annex 4.3.5); whereas the
Historical Atlas of Eurasia (Annex 4.3.4.1) is only available in Google Earth. It is possible to
convert some KML/KMZ material into ArcGIS, and similarly ArcGIS shapefiles can be
exported to KML – but at this stage we have not attempted to do that except for some
crucial datasets.

In addition, tabulated information held in the project database has been used to create point
data in the GIS (see 4.3.2.1), and routes have been directly digitised (see 4.3.2.2).

For a summary of the platform availability of data see Table 7. Details of each data set are
given below.

Data set ArcGIS Google Earth

Data drawn from the database or directly

digitised

1
KML was developed for use with Google Earth, and became an international standard of the Open Geospatial
Consortium in 2008 (http://www.opengeospatial.org/).

104 | P a g e
 Sites 9 8

Routes/corridors 9 8

Scanned and digitised data

Historical Atlas of Central Asia Partly 9

Archaeological map of Iraq 9 8

Siroux (1949) map of Caravanserai in Iran 9 Places only

Digital archaeological & historical data

Historical Atlas Eurasia Partly 9

Digital Silk Road project

Silk Road routes and places Places only 9

Aurel Stein expeditions 8 9

Innermost Asia – 4 volumes 8 9

Serindia – 5 volumes 8 9

The Old World Trade Routes Project

Stopping places 9 9

Georeferenced Nodes of Long-Distance Partly Partly

Communication Routes

Trade Routes 600-1000 CE 8 9

The Electronic Atlas of Buddhist 8 9

Monasteries

ArchAtlas 8 9

Silk Roads on Google Maps 8 9

Ancient Cities DataBase 8 9

Pleiades 8 9

Project HESTIA 8 9

The threat to world heritage in Iraq 8 9

105 | P a g e
 World Heritage List 2010 8 9

Physical data 9 8

Environmental data 9 8

Satellite imagery 8 9

Table 7. Summary of data available in ArcGIS, ArcExplorer and Google Earth formats

4.3.2 Data drawn from the database or directly digitised

4.3.2.1 Sites
This data is represented geo-spatially by drawing data from queries run on the Microsoft
Access© database. It includes sites on a Tentative List, existing World Heritage Sites and
other sites of interest, which are displayed using information directly derived from a query
of the database (Figure 23).

Layer Description Query

WH Existing World [UNESCOWHS] =True AND [UNESCO_REVIEW_SITE] =True

sites Heritage Sites

TL Tentative List sites [UNESCO_TL] =True AND [UNESCO_REVIEW_SITE] =True

Sites

Sites All other sites in the [UNESCO_REVIEW_SITE] =True AND [UNESCO_TL] =False
database AND [UNESCOWHS] =False

106 | P a g e
Figure 23. Example of sites in the database shown on the GIS. Key: Green = Existing World Heritage Sites; Yellow =
Tentative List Sites; White = Other sites of interest.

4.3.2.2 Digitised routes: corridors & their buffers

The process of selecting and digitising routes and corridors has been discussed in Section 5,
with the process of digitisation specifically highlighted in Section 5.3.

Figure 24. Example of digitised routes. Red = main routes; yellow = more minor routes.

107 | P a g e
Figure 25. Example of corridors: main routes buffered (red hatching extending 30 km either side of the route).

4.3.3 Scanned and digitised data

Some maps from conventional publications were scanned and information from them
digitised, to enable comparison with routes and sites along the Silk Roads. Some of this
material is included in the data distributed with the project, but some, such as the Bregel
maps, are copyright protected and were only used for internal project analysis.

4.3.3.1 Historical Atlas of Central Asia (Bregel 2003)

The following maps from this publication were scanned and geo-rectified:

x Geographic background (1_geography.jpg)

x Parthians and Kushans: 1st century BCE-2nd century CE (2_Parthians and Kushans
1CBC-2CAD.jpg)
x 3rd-6th century CE (3_3-6C.jpg)
x 10th century CE (4_10C.jpg)

However, these cannot be distributed as they are covered by copyright.

4.3.3.2 Archaeological map of Iraq

Source: The archaeological map of Iraq. Mudīrīyat al-Āthār al-Qadīmah al-'Āmmah. Baghdad:
Directorate General of Antiquities, 1967. Scale 1:1,500,000.

This map is freely available from “The threat to world heritage in Iraq”:
http://users.ox.ac.uk/~wolf0126/#major

108 | P a g e
Figure 26. A geo-rectified scan of The archaeological map of Iraq, overlain with digitised site locations.

A geo-rectified scan of the map from this publication, and the digitised location of each of
the sites, will be made available with the digital data in 2014.

4.3.3.3 Siroux (1949) map of Caravanserai in Iran

Source: Siroux 1949.

Figure 27. Geo-rectified scan of map from Siroux 1949.

A geo-rectified scan of the map from this publication (Figure 27), and the digitised location
of each of the sites and routes, will be made available with the digital data in 2014.

109 | P a g e
In addition, the GIS group Siroux Iranian Caravanserai contains the digitised locations (and
names) of each of the caravanserai (Siroux_carvanserais) and the routes (Siroux_routes) (see
Figure 28).

In Google Earth only the site locations and names are available.

Figure 28. Example of Siroux digitised caravanserai locations (named) and routes.

4.3.4 Digital archaeological & historical data

4.3.4.1 Historical Atlas of Eurasia Online

Historical Atlas of Eurasia (Third Edition) is an online atlas available at:
http://bbs.keyhole.com/ubb/ubbthreads.php?ubb=showflat&Number=1060464

It is organized into five independent parts, each with a complex array of data, including the
spread of Empires, routes and the location of archaeological sites through time. The most
relevant parts are:

x Part 2: ‘The Spread of Religion’, including the Silk Roads;

x Part 3: ‘The Great Eurasian Empires’;
x Part 4: ‘Asia Ante Portas’, including the Hun, Mongol, and Timurid Empires;
x Part 5: ‘The Explorers’, including the routes of early Chinese and European Travellers.

It is available in Google Earth (through the historical atlas 2010.kmz file).

110 | P a g e
Figure 29. The multiplicity of information available in the Google Earth version of the Historical Atlas of Eurasia.

Some of the Google Earth data has been translated into shape files for viewing in ArcGIS.
These include the Silk Roads routes (\Silk Roads\GIS\Google Earth\Historical Atlas
Eurasia\Eurasia_SR_Map.shp - see Figure 30) and Eurasian Empires (\Silk Roads\GIS\Google
Earth\Historical Atlas Eurasia\Shapefiles\Eurasian_Empires.shp – see Figure 31).

Figure 30. Example of the Historical Atlas of Eurasia ‘Silk Roads routes’ as a GIS shapefile.

111 | P a g e
 Figure 31. Example of the Historical Atlas of Eurasia ‘Eurasian Empires’ as a GIS shapefile:
any of the individual empires can be turned on or off.

4.3.4.2 Digital Silk Roads Project

The Digital Silk roads project is a research project on creating digital archives of cultural
heritage through collaboration between informatics and humanities at the National Institute
of Informatics in Tokyo, Japan, and linked to the Digital Archive of Toyo Bunko Rare Books. It
is available from http://dsr.nii.ac.jp. It comprises scans and data from a number of rare or
out-of print books, such as the database of Stein place names, and includes numerous
individual locations, 196 mountain passes and 189 ruins.

This is largely available only in Google Earth:

x Silk Road routes and places: site locations and routes in Silk Roads Places and Routes
- Digital Silk Road Project.kmz (in ArcGIS places only available in SR_places.shp).
x Aurel Stein expeditions: places, scanned maps of his expeditions, routes, etc – all in
Stein expeditions.kmz (one table of ‘ruins’ in database and accessible via ArcGIS).
x Innermost Asia – 4 volumes: Scanned and geo-rectified maps available in Innermost
Asia (T-VIII-5-A-a-3V-4) Maps - Digital Silk Road.kmz (see Figure 32 & Figure 33).

112 | P a g e
Figure 32. Coverage of Innermost Asia maps (from the Digital Silk Roads project) in Google Earth.

Figure 33. Detail of Innermost Asia maps (from the Digital Silk Roads project) in Google Earth

x Serindia – 5 volumes: Scanned and geo-rectified maps available in Serindia (VIII-5-B2-

9V-5) Maps - Digital Silk Road.kmz (see Figure 34 & Figure 35).

113 | P a g e
Figure 34. Coverage of Serinda maps (from the Digital Silk Roads project) in Google Earth.

Figure 35. Detail of Serinda maps (from the Digital Silk Roads project) in Google Earth.

4.3.4.3 The Old World Trade Routes Project

Developed by Dr T. Matthew Ciolek (Research School of Pacific and Asian Studies, Australian
National University, Canberra, Australia) this incredibly rich data includes point data for
specific sites and suggested routes. It is available at http://www.ciolek.com/owtrad.html

The data is primarily available as downloadable files (normally CSV format) and as Google
Earth KML or KMZ files.

114 | P a g e
The key elements include:

x Stopping places: A Catalogue of Georeferenced Caravanserais/Khans (version 5.0 -

Nov 2006). Contains 846 caravanserais/khans, funduks, and other rest houses
supporting long-distance communication routes.
x In Google Earth these are presented in Stopping Places.kml. This contains two earlier
kml files: tmc-nodes-RHSE-data.kml which covers funduks, hospices, caravanserai;
tmc-nodes-HALT-data.kml which contains other halting places.
x In ArcGIS these two are combined into a single shape file Stoppingplaces.shp
x From Afrosiab to Zucchabar: A Gazetteer of Georeferenced Nodes of Long-Distance
Communication Routes (version 10.1 - Jan 2007): contains 16,257 variant place
names, of 4,265 different places in Asia, Europe and Africa. A Silk Roads version of
this, only including nodes from within the current study area countries (but including
Azerbaijan, Armenia, the whole of Turkey, Israel, and Palestine), has 11,524 variant
names covering 2,523 separate places. The whole dataset is available in the
Microsoft Access© database, which also contains metadata on the fields.
These were used by OWTRAD to create a number of route files (see
http://www.ciolek.com/OWTRAD/DATA/oddda-asia.html), based around different
geographic areas or peoples work (such as Whitfield 1999); the nodes are simply
connected by straight lines (Figure 36).
x In Google Earth these appear under Nodes: the only ones converted to kml’s so far
are Tajik1 & 2, Persia, and Bulianov (1999) Turkmenistan & Iran.
x In ArcGIS these appear under the group layer Nodes. They currently include: NE &
Asia - 1400-1600; Asia - 1200-1400; China; Foltz 1999; Kyrgyzstan 1-5; Persia -
550BC-450BC; Tajikistan 1 & 2; Turkmenistan-Iran (Bulianov 1999); Whitfield 1999.

Figure 36. Example of OWTRAD nodes connected by straight lines – in this case from Whitfield 1999.

x Georeferenced historical transport/travel/communication routes and nodes - Old

World: a relatively simple set of route segments.

115 | P a g e
 x Only available in Google Earth, these are presented in input-geo-routes.kml: this
creates a group called ‘Trade Routes 600-1000 CE’, with separate named segments
that can be individually turned on/off.
x The Electronic Atlas of Buddhist Monasteries: this is a work in progress
(http://www.ciolek.com/GEO-MONASTIC/geo-monasteries-home.html). Details of
communication, contacts and affinities between as many as possible of the Buddhist
monasteries and convents known to have operated in South Asia, SE Asia, Central
Asia, and East Asia from c 200 CE until c 1200 CE. Fantastic data linked to a very good
Wiki site (http://monastic-asia.wikidot.com/).
x This data is only currently available in Google Earth this data set is represented by a
number of geographically defined kml files.

Figure 37. Example of OWTRAD monastic data in Google Earth. Legend - White circles: Clusters and groups of
monasteries; Yellow: Theravada monasteries; Green: Sammitiya, Sarvastivada & Mahasanghika monasteries; Red:
Mahayana monasteries; Magenta: Vajrayana monasteries; Blue: Indeterminate Buddhist tradition.

116 | P a g e
Figure 38. Example of OWTRAD wiki monastic entry.

x A trade routes' glossary: architectural, geographical, and historical terms relating to

trade, communication, and pilgrimage routes. Version 1.7 (Aug 2009). A very useful
thesaurus of terms is available at: http://www.ciolek.com/OWTRAD/trade-routes-
glossary.html.

4.3.4.4 ArchAtlas
ArchAtlas is a web-orientated archaeological mapping and research project, founded by the
late Prof. Andrew Sherratt. It is available at: http://www.archatlas.org/Home.php

Data from on-going archaeological mapping projects, across the Near East and Central Asia,
are discussed, with an emphasis on the Bronze Age although the site contains material from
all periods. It includes a number of thematic on-line essays.

A basic list of sites linked to essays and descriptions is available in Google Earth as
ArchAtlasGE2006.kml.

4.3.4.5 Silk Roads on Google Maps

Consists of a collection of place-markers and routes, in part derived from Digital Silk Roads
data (see Annex 4.3.4.2). Available at:
http://bbs.keyhole.com/ubb/download.php?Number=937387

The data set 552906.kmz contains routes and places. The alignment of the roads at high
resolution are somewhat arbitrary. The towns are mostly named after one of their historical
spellings. Some descriptions have been included.

4.3.4.6 Ancient Cities DataBase

The Ancient Cities DataBase contains information about cities and towns that were founded
before 400 CE. Data is available for Google Earth from:
http://sites.google.com/site/ancientcitiesdb/Home (Historical Cities.kmz).

117 | P a g e
Data includes: Original Name, Historic Name, Founding Date, Original Nation, and Location,
and can be selected by region. The emphasis is upon the classical world and the
Mediterranean, but sites from a wide area are covered.

4.3.4.7 Pleiades
Pleiades is a joint project of the Ancient World Mapping Center, the Stoa Consortium, and
the Institute for the Study of the Ancient World (http://pleiades.stoa.org/).

Pleiades aims to provide a worldwide ability to use, create, and share historical geographic
information about the Greek and Roman World in digital form. The project began by
digitising the Barrington Atlas (Talbert 2000).

The data is only available in Google Earth, and links to online resources/descriptions and
maps. It currently contains:
x 32,336 Ancient Locations
x 26,646 Ancient Place Names
x 31,559 Ancient Places
x Vocabularies covering: name-accuracy; association-certainty; place and feature
types; attestation-confidence; time periods; name-completeness; ancient-name-
languages; name-types.

At present, it has two drawbacks:

x It originally only provided grid square locations, although now there are more sites
with specific locations.
x Pleiades has only limited support for multiple toponyms for the same location.

A new Pleiades+ hopes to address these issues.

4.3.4.8 Project HESTIA: the Herodotus Encoded Space-Text-Imaging Archive

HESTIA provides an approach towards conceptions of space in the ancient world. It examines
the ways in which space is represented in Herodotus' History, in terms of places mentioned
and geographic features described. It is accessible from http://www.open.ac.uk/Arts/hestia/

KML file (hestia.kmz) has an embedded network link which creates an overlay of red squares
showing all the locations referred to by Herodotus. When zooming in low, the squares
become clickable points that provide information about the location as well as each
reference to it (in English and Greek).

4.3.4.9 The threat to world heritage in Iraq

An Oxford University website (http://users.ox.ac.uk/~wolf0126/) has three maps of Iraq
heritage:

x Upper Euphrates
x Upper Tigris
x Lower Mesopotamia

These have been geo-rectified in Google Earth and are available in the group Iraq at risk.

118 | P a g e
4.3.4.10 World Heritage List 2010
The current World Heritage List can be downloaded from http://whc.unesco.org/en/254
and a Google Earth KML file. It contains locations and links to online descriptions (Figure 39).

Figure 39. Example of UNESCO World Heritage Sites in Google Earth – with direct link to UNESCO World Heritage Centre
information.

4.3.4.11 Ball’s Gazetteer of sites in Afghanistan

Warwick Ball’s (1982) seminal gazetteer of sites in Afghanistan lists over 1,286 sites in
Afghanistan. The spatial data is in geodetic format (degrees and minutes). A digital version of
the data in Ball’s gazetteer was collated by Rebecca Beardmore, as part of the ArchAtlas
project (http://www.archatlas.org/Home.php).

x In Google Earth this is available as Af GE sites.kmz. There are also two sub-sets,
covering bridges (ball_bridges.kmz) and caravanserai (ball_caravanserais.kmz).
x In ArcGIS a copy of the data is available inside the Silk Roads database
(Afghanistan_Ball) and can be viewed through the ArcGIS layer Afghanistan_Ball.

119 | P a g e
Figure 40. Example of Warwick Ball’s (1982) gazetteer of sites in Afghanistan, as seen in ArcGIS.

4.3.5 Physical data: ArcGIS Online

Note: This data requires an internet connection (it is accessed live over the internet – it
cannot be downloaded – so you will need a reasonably high-speed internet connection to use
this).

The ESRI ArcGIS Online services provide a Natural Earth physical map at global and
continental scales. The collection also includes topographic maps, boundaries and places.

ArcGIS Online product page: http://www.esri.com/software/arcgis/arcgisonline/index.html

Additional information about ArcGIS Online: http://www.arcgis.com

ESRI ArcGIS Online. Source: US National Park Service, USGS, East View Cartographic.

120 | P a g e
4.3.5.1 World_Physical_Map

Figure 41. World Physical Map.

This map presents the Natural Earth physical map at 1.24km per pixel for the world.

For more information on this map, visit:

http://goto.arcgisonline.com/maps/World_Physical_Map

4.3.5.2 World_Reference_Overlay

Figure 42. World_Reference-Overlay on top of World_Physical _Map (see above).

121 | P a g e
This map is designed to overlay base maps and thematic maps, such as demographics or land
cover, for reference purposes. The reference map includes administrative boundaries, cities,
water features, physiographic features, parks, landmarks, highways, roads, railways, and
airports on a transparent background. The map was compiled from a variety of best
available sources from several data providers, including the U.S. Geological Survey, National
Park Service, Tele Atlas, AND, and ESRI. The reference map currently provides coverage for
the world down to a scale of ~1:1m and coverage for the continental United States and
Hawaii to a scale of ~1:70k.

For more information on this map, visit:

http://goto.arcgisonline.com/maps/Reference/World_Reference_Overlay

4.3.5.3 World Boundaries and Places

Figure 43. World Boundaries and Places on top of World_Physical _Map (see above).

This map presents country, state/province, and county or equivalent boundaries and place
names for the world. The map was developed by ESRI using administrative and cities data
from ESRI and AND Mapping. This map is designed for use with maps with lighter
backgrounds, such as World Shaded Relief (see below).

For more information on this map, visit:

http://goto.arcgisonline.com/maps/Reference/World_Boundaries_and_Places_Alternate

122 | P a g e
4.3.5.4 World Shaded Relief

Figure 44. World shaded relief.

This map presents a worldwide shaded relief map at approximately 1 kilometre or 90 meters
per pixel resolution for the world. The shaded relief map was developed by ESRI using
GTOPO30 and Shuttle Radar Topography Mission (SRTM) elevation data from the USGS and
National Geospatial-Intelligence Agency (NGA) overlaid with water bodies from Esri and
Automotive Navigation Data (AND).

For more information on this map, visit:

http://goto.arcgisonline.com/maps/World_Shaded_Relief

For more information on this map, visit: http://services.arcgisonline.com/ArcGIS/services.

4.3.5.5 World Topo Map

This map service is designed to be used as a basemap and as a reference map. The service
includes administrative boundaries, cities, water features, physiographic features, parks,
landmarks, highways, roads, railways, airports, and buildings overlaid on land cover and
shaded relief imagery for added context. This service is the default basemap used in the
ArcGIS.com map viewer and ArcGIS Explorer Online.

123 | P a g e
Figure 45. Example of World Topo Map.

Accessible via: World_Topo_Map.lyr

For more information on this map, visit: http://services.arcgisonline.com/ArcGIS/services.

4.3.5.6 World Imagery

Note: This is used as the backdrop for most of the maps produced in the report.

This map presents low-resolution imagery for the world and high-resolution imagery for the
United States and other areas around the world. The map includes NASA Blue Marble: Next
Generation 500m resolution imagery at small scales (above 1:1,000,000), i-cubed 15m eSAT
imagery at medium-to-large scales (down to 1:70,000) for the world, and USGS 15m Landsat
imagery for Antarctica. The map features i-cubed Nationwide Prime 1m or better resolution
imagery for the contiguous United States, Getmapping 1m resolution imagery for Great
Britain, AeroGRID 1m to 2m resolution imagery for several countries in Europe, IGP 1m
resolution imagery for Portugal, and GeoEye IKONOS 1m resolution imagery for Hawaii, parts
of Alaska, and several hundred metropolitan areas around the world. i-cubed Nationwide
Prime is a seamless, colour mosaic of various commercial and government imagery sources,

124 | P a g e
including Aerials Express 0.3 to 0.6m resolution imagery for metropolitan areas and the best
available United States Department of Agriculture (USDA) National Agriculture Imagery
Program (NAIP) imagery and enhanced versions of United States Geological Survey (USGS)
Digital Ortho Quarter Quad (DOQQ) imagery for other areas.

For more information on this map, visit: http://goto.arcgisonline.com/maps/World_Imagery

4.3.5.7 DeLorme World Base Map

Note: This data requires an internet connection (it is accessed live over the internet – it
cannot be downloaded – so you will need a reasonably high-speed internet connection to use
this).

Figure 46. DeLorme World basemap.

The map is a seamless global data set with horizontal accuracy of +/- 50 m. The map includes
major transportation layers, inland and shoreline hydrography, jurisdiction boundaries and
major geographic features. The map has a rich cartographic appearance.

Scale Range: down to 1:288,895.

Coordinate System: Web Mercator Auxiliary Sphere (WKID 102100).

For more information on this map, visit:

http://goto.arcgisonline.com/maps/Specialty/DeLorme_World_Base_Map or

http://www.delorme.com/digitalmapdata/world.htm

125 | P a g e
4.3.6 Physical & environmental data: downloadable

4.3.6.1 World SRTM shaded relief

Downloadable data: SRTM 90m Digital Elevation Data, from the CHIAR Consortium for
Spatial Information

Displays shaded relief of the world, between 60 degrees North latitude and 56 degrees
South latitude. This data was derived from the Global Digital Elevation Model (SRTM)
datasets from the U.S. Geological Survey's EROS Data Center. The resolution is 3 arc seconds
(90 meters). The version of the SRTM 3-arc-second data that served as the basis for the
SRTM Global Digital Elevation Model data is the Version 2 "finished" data in DTED Level 1
format that was created by NGA by subsampling SRTM 1-arc-second data.

Figure 47. World_SRTM, with gradual colour change representing relief.

126 | P a g e
 Figure 48. World_SRTM2 has a simplified colour scheme that represents the data in five
categories.

127 | P a g e
4.3.6.2 AS cropland & pasture
Downloadable data:

Raster data in: as_cropland.rrd

Geo tiff data in: as_pasture.tif

Figure 49. AS cropland

Figure 50. AS pasture.

128 | P a g e
NASA description: “The Global Croplands dataset represents the proportion of land areas
used as cropland (land used for the cultivation of food) in the year 2000. Satellite data from
Modetate Resolution Imaging Spectroradiometer (MODIS) and Satellite Pour l'Observation
de la Terre (SPOT) Image Vegetation sensor were combined with agricultural inventory data
to create a global data set. The visual presentation of this data demonstrates the extent to
which human land use for agriculture has changed the Earth and in which areas this change
is most intense. The data was compiled by Navin Ramankutty, et. al. (2008) and distributed
by the Columbia University Center for International Earth Science Information Network
(CIESIN). The global and regional data in Geographic Tagged Image File Format (GeoTIFF) and
ESRI GRID formats and maps in Joint Photographic Experts Group (JPEG) and Portable
Document Format (PDF) formats are available from the Socioeconomic Data and
Applications Center (SEDAC) via direct download at
http://sedac.ciesin.columbia.edu/es/aglands.html. To provide data on the extent of
croplands for research on human-environment interactions.”

The spatial data is provided in raster GeoTiff and ESRI Grid formats. Raster cell sizes are 5”,
or 0.08333 degree decimal (about 10 km at the equator). The data set is in geographic
projection. To download the zip files with accompanying metadata and documentation, click
on the links below. The downloaded compressed zipfiles contain raster data, a metadata
record, and a readme file.

Data distributed by the NASA Socioeconomic Data and Applications Center (SEDAC):
http://sedac.ciesin.columbia.edu/es/aglands.html . [Accessed 01 March 2011].

129 | P a g e
4.3.6.3 Köppen-Geiger climate classification
koeppen-geiger.shp

Köppen climate classification scheme divides world climates into five main groups and
several types and subtypes. Each particular climate type is represented by a 2 to 4 letter
symbol (see http://en.wikipedia.org/wiki/K%C3%B6ppen_climate_classification for a full
breakdown).

Figure 51. Köppen-Geiger climate classification (see key below).

GROUP A: Tropical/megathermal climates

GROUP B: Dry (arid and semiarid) climates

GROUP C: Temperate/mesothermal climates

GROUP D: Continental/microthermal climate

GROUP E: Polar & Alpine climates

130 | P a g e
4.3.6.4 WorldClim - Global Climate Data
Downloadable: but very large data sets

Format: ESRI grids

The data can be downloaded from http://www.worldclim.org/current

Downloadable data on current world climatic conditions (~1950-2000) at different

resolutions – from 30 arc-seconds to 10 arc-minutes. Data includes:

x Min. Temperature
x Max. Temperature
x Mean Temperature
x Precipitation
x Altitude

Each data set has 12 discreet files – one for each month of the year.

In addition, a Bioclim data set has bioclimatic variables derived from the monthly
temperature and rainfall values in order to generate more biologically meaningful variables.
These are often used in ecological niche modelling. The bioclimatic variables represent
annual trends (e.g. mean annual temperature, annual precipitation), seasonality (e.g. annual
range in temperature and precipitation) and extreme or limiting environmental factors (e.g.
temperature of the coldest and warmest month, and precipitation of the wet and dry
quarters). A quarter is a period of three months (1/4 of the year).

4.3.6.5 Natural Earth

Natural Earth is a public domain map dataset available at 1:10m, 1:50m, and 1:110 million
scales. Featuring tightly integrated vector and raster data, for cartography or GIS software.

Downloaded data:

x 1:10m Physical Vectors - A variety of physical regions can be displayed (see Figure
52), as well as a range of elevation points. More information and available data from:
http://www.naturalearthdata.com/downloads/10m-physical-vectors/
x 1:10m Admin Countries – country boundaries. More information and available data
from: http://www.naturalearthdata.com/downloads/10m-cultural-vectors/
x Natural Earth II with Shaded Relief, Water, and Drainages - a raster map dataset
that portrays the world environment in an idealized manner with little human
influence. A successor dataset to the original Natural Earth, Natural Earth II owes its
inspiration to the classic physical maps published by the National Geographic Society.
The softly blended colours of Natural Earth II make it a suitable base for general-
purpose mapping, including historical maps, because it shows the world much as it
looked before the modern era. Natural Earth II data comes in several versions for use
with cartography and GIS software. More information and available data from:
http://www.naturalearthdata.com/downloads/10m-raster-data/10m-natural-earth-
2/ Note: this is very similar to the World Physical Map (Annex 4.3.5.1) provided by
ESRI online. This version has the advantage of being downloadable and thus can

131 | P a g e
 operate without an internet connection, but the World Physical Map has
considerably better resolution.

Figure 52. Example of 1:10m Physical Vectors from Natural Earth: showing mountain ranges, deserts, steppe, major
valley systems, etc, on the back drop of Natural Earth II.

4.3.6.6 Desert
A very simple representation of deserts (desert.shp).

Figure 53. Example of desert backdrop.

132 | P a g e
4.3.6.7 World Elevation Contours
Downloadable. World Elevation Contours.lyr

Feature Dataset: contours.sdc

Figure 54. Example of World Elevation Contours (with no backdrop). Individual contours lines can be displayed or re-
coloured.

World Contours with 600-meter interval contour lines of the world.

Source: DeLorme Publishing Company, Inc.

133 | P a g e
4.3.6.8 World Linear Water
Downloadable. World Linear Water.lyr

Feature Dataset: hydrolines.sdc

World Linear Water represents the narrower rivers and streams of the world.

Source: DeLorme Publishing Company, Inc.

Figure 55. Example of World Linear Water layer, with intermittent and perennial streams.

134 | P a g e
4.3.6.9 World Drainage Systems
Downloadable. World Drainage Systems.lyr

Feature Dataset: drainage.sdc

World Drainage Systems represents the major drainage systems of the world.

Figure 56. Example of simple World Drainage System map.

Figure 57. World Drainage Systems: can map and chart basin areas, discharge levels, sedimentation load, and distance.

135 | P a g e
Figure 58. World Drainage Systems: example of charting basin area, sedimentation and discharge.

136 | P a g e
4.3.6.10 World Gazetteer
Downloadable. World Gazetteer.lyr

Feature Dataset: gaz.sdc

Source: ESRI, CIA World Factbook, GMI

Figure 59. Example of World Gazetteer. Labels and symbols can be rescaled.

World Gazetteer represents the locations and proper names for map features around the
world. The gazetteer includes attribute and annotation name information from various
layers of the Digital Chart of the World.

137 | P a g e
4.3.6.11 World Cities
Downloadable. World Cities.lyr

Feature Dataset: cities.sdc

Source: ESRI, CIA World Factbook, GMI, NIMA, Times Atlas 10th

Figure 60. Example of World Cites.

World Cities represents the locations of major cities of the world, scaled according to
present day populations (or a variety of other fields, such as administrate status).

4.3.7 Software

4.3.7.1 ESRI ArcGIS

This framework has been created in ESRI ArcGIS, a commercial software GIS package which
allows the combination of scanned maps, imagery, topographic terrain models, coordinate
data and linear definition of routes.

The disadvantage with this package is that it is expensive, especially outside academic
communities, and relatively complex to use, requiring at least some basic training. There is

138 | P a g e
an initiative currently underway, via the International Scientific Committee on
Archaeological Heritage Management (ICAHM), to establish a license with ESRI for the supply
of free copies of the software in support of the World Heritage convention and the
protection of archaeological heritage. If this is forthcoming, it may be possible to roll-out the
more complex tools.

In the short-term, however, ArcGIS has been used as a data preparation tool, and while the
data will be available in this format, the main dissemination of the data has been planned in
ArcGIS Explorer and Google Earth (see below).

Figure 61. ESRI ArcMap.

4.3.7.2 ArcGIS Explorer

ESRI ArcGIS Explorer Desktop is a free to download and use – see:
http://www.esri.com/software/arcgis/explorer/index.html

It is available in English, French, German, Chinese, Spanish and Japanese.

It comes with a number of base maps, but the ‘World Imagery’ is the most effective for our
purposes.

The advantage of this platform is that you can easily view both KML/KMZ files and
shapefiles. But it has more significantly limited analytical tools compared to ArcGIS.

139 | P a g e
Figure 62. ESRI ArcGIS Explorer: showing a detail of an area from China.

4.3.7.3 Google Earth

Easily accessible and free to download, but requires good stable internet connection.

It has access to detailed satellite imagery (e.g. Figure 63).

Figure 63. Ayaz Tepe (Uzbekistan). On the left as seen through ArcGIS Explorer (with the World Imagery basemap) and on
the right in Google Earth.

However, there are problems with the patchy backdrop created by numerous satellite
images when looking at large areas (e.g. Figure 64).

140 | P a g e
Figure 64. Example of viewing large areas in Google Earth, with the patchy background created by numerous different
resolutions and exposures of satellite images.

141 | P a g e
5 Bibliographic survey
The scale of published resources on the Silk Roads and its archaeology and history is vast.
The project collected over 1,500 references to books, and there are many thousands more
relevant journal articles, only a few of the more important of which were included in the
bibliography. The list will undoubtedly continue to grow and develop in the coming years.
Sources were provided by Kazakhstan, Kyrgyz Republic and Uzbekistan (all in Russian),
Afghanistan, China, India and Iran (partially in Fārsi), while Japan and Korea also provided
some wider synthetic references. However, many of the State Parties have not yet had an
opportunity to add all of their material and undoubtedly they will be aware of a wider range
of source material which will need to be incorporated.

Rather than produce a static paper version of this resource, the project will release a digital
Silk Roads bibliography (see Williams forthcoming a)1. The aim will be to provide searchable
text, with some keyword search facilities.

1
An earlier version of the database was supplied to the Silk Roads CHRIS project and is available at:
http://arex.no-ip.org/specto/bin/view/home.

142 | P a g e
References cited
Allsen, T T (1997) Commodity & exchange in the Mongol Empire: a cultural history of Islamic
textiles. Cambridge: Cambridge University Press

Aubaile, F (2012) Voies de diffusion de quelques plantes et animaux entre l'Asie et la région
méditerranéenne [Pathways of diffusion of some plants and animals between Asia
and the Mediterranean region], Revue d'ethnoécologie 1: 2-26

Baipakov, K M & Pidayev, Sh R (eds) (2011) Prominent archaeological sites of Central Asia on
the Great Silk Road. Samarkand: ICAS

Baipakov, K M (1990) Po sledam drevnikh gorodov Kazakhstana. Alma-Ata:

Bálint, C (1989) Die Archäologie der Steppe : Steppenvölker zwischen Volga und Donau vom
6. bis zum 10. Jahrhundert. Böhlau

Ball, W (1982) Archaeological Gazetteer of Afghanistan. Paris: Documentation Française

Bang, P F (2008) The Roman bazaar: a comparative study of trade and markets in a tributary
empire. Cambridge: Cambridge University Press

Barfield, T J (1989) The perilous frontier: nomadic empires and China, 221 BC to AD 1757.
Cambridge, Mass. & Oxford: Blackwell

Baumer, C (2003) Southern Silk Road: In the Footsteps of Sir Aurel Stein and Sven Hedin. (2nd
ed) Bangkok: Orchid Press

Baumer, C (2008) Traces in the Desert: Journeys of Discovery Across Central Asia. London: I B
Tauris

Baypakov, K (ed) (2011) Artistic culture of Central Asia and Azebaijan in the 9th-15th
centuries: Volume II glass. Samarkand/Tashkent: International Institute for Central
Asian Studies

Bennison, A K & Gascoigne, A (eds) (2007) Cities in the pre-modern Islamic world: the urban
impact of religion, state and society. London: Routledge

Bernsted, A-M K (2003) Early Islamic Pottery: Materials and Techniques. Archetype

Bernstein, R (2002) Ultimate journey: retracing the path of an ancient Buddhist Monk who
crossed Asia in Search of Enlightenment.

Bertrand, A (2010) The Hydraulic Systems in Turfan (Xinjiang), The Silk Road 8: 27-41

Boulnois, L (2008) The Silk Road. Monks, warriors and merchants on the Silk Road. London:
Odyssey Publications

Bregel, Y (2003) An Historical Atlas of Central Asia. Leiden: Brill

143 | P a g e
Canepa, M (2008) Distant Displays of Power: Understanding Cross-Cultural Interaction
Among the Elites of Rome, Sasanian Iran, and Sui-Tang China, Ars orientalis 38: 121-
154

Carboni, S (2001) Glass from Islamic lands. London: Thames & Hudson

Casson, L (1989) The Periplus Maris Erythraei: Text With Introduction, Translation, and
Commentary. Princeton, N.J.: Princeton University Press

Chandra, L & Banerjee, R (eds) (2007) Xuanzang and the Silk Route. New Delhi: Munshiram
Manoharlal

Christian, D (2000) Silk Roads or Steppe Roads? The Silk Roads in World History, Journal of
World History 11 (1): 1-26

Ciolek, T M (2006) Old World Trade Routes (OWTRAD) Notation System: A method for
standardising and computerising geographical and logistical data about long-
distance transportation/communication routes. OWTRAD: Accessed 2/3/2011

Cleere, H (2006) Serial Nomination of the Silk Roads to the World Heritage List: A Concept
Paper.

Constable, O R (2003) Housing the Stranger in the Mediterranean World: Lodging, Trade, and
Travel in Late Antiquity and the Middle Ages. Cambridge: Cambridge University Press

d'Alpoim Guedes, J, Lu, H, Li, Y, Spengler, R, Wu, X, & Aldenderfer, M (2013) Moving
agriculture onto the Tibetan plateau: the archaeobotanical evidence, Archaeol
Anthropol Sci: 1-15

Dani, A H (2002) Significance of Silk Road to human civilization: Its cultural dimension,
Journal of Asian Civilizations 25 (1): 72-79

Daryaee, T (2005) Ethnic and territorial boundaries in Late Antique and early medieval Persia
(third to tenth century), in Curta, F (ed) Borders, barriers, and ethnogenesis: frontiers
in late Antiquity and the Middle Ages 123-138. Turnhout: Brepols

Daryaee, T (2010) Bazaars, Merchants, and Trade in Late Antique Iran, Comparative Studies
of South Asia, Africa and the Middle East 30 (3): 401-409

de la Vaissière, E (2002) Histoire des marchands sogdiens. Paris: Collège de France

de la Vaissière, È (2003) Sogdians in China: A Short History and Some New Discoveries, The
Silk Road 1 (2): 23-27

de la Vaissière, È (2005) Sogdian Traders: A History. Leiden: Brill

Decker, M (2009) Tilling the Hateful Earth: Agricultural Production and Trade in the Late
Antique East. Oxford: Oxford University Press

144 | P a g e
Di Cosmo, N (2000) Ancient City-States of the Tarim Basin, in Hansen, M H (ed) A
Comparative Study of Thirty City-State Cultures 393-407. Copenhagen: Royal Danish
Academy of Sciences and Letters

Directorate General of Antiquities of Iraq (1967) The archaeological map of Iraq (Mudi¯ri¯yat
al-A¯tha¯r al-Qadi¯mah al-'A¯mmah).

Duncan-Jones, R (1990) Structure and scale in the Roman economy. Cambridge: Cambridge
University press

Dussaud, R (1927) Topographie historique de la Syrie Antique et médiévale. Paris: Librairie

Orientaliste Paul Geuthner

Dyakonov, M M (1953) Archeologicheskeie raboti v niznem techenii reki Kafirnigana.

Materiali i Isledovania po Archeologii SSSR. Vol. 37

Faxian (translated by J. Legge) (1886) A record of Buddhistic Kingdoms: being an account by

the Chinese Monk Fâ-Hien of his travels in India and Ceylon (A.D.399-414). Oxford:
Clarendon Press (reprint Kessinger Publishing)

Feuerbach, A (2002a) Crucible steel in Central Asia: production, use and origins. Unpublished
thesis: University of London

Feuerbach, A (2002b) The Glitter of the Sword: The Fabrication of the Legendary Damascus
Steel Blades, Minerva 13 (4): 45-48

Frederico, G (1997) An economic history of the silk industry. Cambridge: Cambridge

University Press

Freeman, M & Ahmed, S (2011) Tea horse road: China's ancient trade road to Tibet. Bangkok:
River Books

Fuchs, J (2008) The Tea Horse Road, The Silk Road 6 (1): 63-71

Fuller, D (2009) Late Harappan "collapse", the opening of central Asia and long-distance crop
movements, in Endo, H (ed) Ethnogenesis of South and Central Asia (ESCA): 13th
Harvard University Round Table, Kyoto Session, Research Institute for Humanity and
Nature (RIHN), Kyoto, Japan, 30-31 May 2009 3-11. Kyoto: Research Institute for
Humanity and Nature

Fuxi, G, Brill, R H, & Shouyun, T (eds) (2009) Ancient glass research along the Silk Road. New
Jersey: World Scientific

Garnsey, P & Whittaker, C R (1998) Trade, industry and the urban economy, in Cameron, A &
Garnsey, P (eds) The Cambridge Ancient History, XIII: the late Empire AD 337-425 332-
335. Cambridge: Cambridge University Press

Garnsey, P, Hopkins, K, & Whittaker, C R (eds) (1983) Trade in the ancient economy. London:
Chatto & Windus

145 | P a g e
Gaubatz, P (1996) Beyond the Great Wall: Urban Form and Transformation on the Chinese
Frontiers. Stanford, Calif.: Stanford University Press

Gosch, S S & Stearns, P N (2008) Premodern Travel in World History. London: Routledge

Goskartografia (2002) Silk Road of Kyrgyzstan.

Grenet, F (2003) The Pre-Islamic Civilization of the Sogdians (seventh century BCE to eighth
century CE): A Bibliographic Essay (studies since 1986), The Silk Road 1 (2): 28-36

Hansen, V (2012) The Silk Road: A New History. Oxford: Oxford University Press

Henderson, J, Challis, K, O'Hara, S, McLaughlin, S, Gardner, A, & Priestnall, G (2005)

Experiment and innovation: early Islamic industry at al-Raqqa, Syria, Antiquity 79
(303): 130-145

Hill, J E (2009) Through the Jade Gate to Rome: A Study of the Silk Routes during the Later
Han Dynasty, 1st to 2nd Centuries CE. Charleston, South Carolina: Booksurge

ICOMOS (2002) Desk study on the Asia Region. ICOMOS

ICOMOS (2003) Analysis of the World Heritage List and Tentative Lists. ICOMOS

ICOMOS (2004) The World Heritage List: Filling the Gaps – An Action Plan for the Future.
Paris: ICOMOS

ICOMOS International Scientific Committee on Cultural Routes (1994) Routes as part of our
cultural heritage: report on the meeting of experts, Madrid 1994.

Isidore of Charax (1914) Parthian stations: an account of the overland trade route between
the Levant and India in the first century B.C. (Reprinted 1989) Chicago: Ares
(translation and commentary by Wilfred H. Schoff in 1914)

Jenkins-Madina, M (2006) Raqqa Revisted: Ceramics of Ayyubid Syria. The Metropolitan

Museum of Art & Yale University Press

Jing, F & van Oers, R (2004) UNESCO Mission to the Chinese Silk Road as World Cultural
Heritage Route. A systematic approach towards identification and nomination, from
21-31 August 2003.

Jokilehto, J, Solar, G, & Petzet, M (2003) Draft Framework for World Heritage Cultural
Representation. Unpublished report. Paris: ICOMOS

Jones, A H M (1964) Later Roman Empire, 284-602: a social, economic and administrative
survey. Oxford: Blackwell

Juliano, A L & Lerner, J A (eds) (2003) Nomads, Traders and Holy Men Along China's Silk
Road: Papers Presented at a Symposium Held at the Asia Society in New York,
November 9-10, 2001. Turnholti: Brepols

146 | P a g e
Kennet, D (2004) Sasanian and Islamic Pottery from Ras al-Khaimah. Oxford: Archaeopress

Kessler, A T (2012) Song blue and white porcelain on the Silk Road. Leiden: Brill

Khakimov, A (ed) (2011) The artistic culture of Central Asia and Azebaijan in the 9th-15th
centuries: Volume I ceramics. Samarkand/Tashkent: International Institute for Central
Asian Studies

Lewis, R A (1965) Early irrigation in west Turkestan, Annals of the Association of American
Geographers 56 (3): 467-491

Liu, X & Shaffer, L N (2007) Connections Across Eurasia: Transportation, Communication, and
Cultural Exchange Along the Silk Roads. McGraw-Hill

Liu, X (1988) Ancient India and Ancient China: trade and religious exchange AD 1-600. Delhi:
Oxford University Press

Liu, X (2010) The Silk Road in world history. Oxford: Oxford University Press

Martin, O & Gendre, S (eds) (2010) UNESCO World Heritage: Serial Properties and
Nominations. International expert meeting on World Heritage and serial properties
and nominations, Ittingen, Switzerland, 25–27 February 2010. Bern: Federal Office of
Culture, with UNESCO World Heritage Centre. Download available at:
http://whc.unesco.org/en/events/1097/

Mattingly, D J & Salmon, J (2000) Economies Beyond Agriculture in the Classical World.
London: Routledge

Mattingly, D J (2013) To south and north: Saharan trade in antiquity, in Eckardt, H & Rippon,
S (eds) Living and working in the Roman world: essays in honour of Michael Fulford on
his 65th birthday Portsmouth, R.I.: Journal of Roman Archaeology

McKnight, T L & Hess, D (2000) Climate Zones and Types: The Köppen System, Physical
Geography: A Landscape Appreciation Upper Saddle River, NJ: Prentice Hall

McLaughlin, R (2009) Rome and the Distant East: Trade Routes to the Ancient Lands of
Arabia, India and China. London: Hambledon Continuum

Millward, J A (2007) Eurasian Crossroads: A History of Xinjiang. London: Hurst & Company

Millward, J A (2013) The Silk Road: a very short introduction. Oxford: Oxford University Press

Moore, J M & Wendelken, R W (eds) (2010) Teaching the Silk Road: A Guide for College
Teachers. Albany: State University of New York Press

Morrisson, C (2012) Trade and markets in Byzantium. Washington, D.C.: Dumbarton Oaks
Research Library and Collection

Nebenzahl, K (2004) Mapping the Silk Road and beyond: 2,000 years of exploring the East.
London & New York: Phaidon

147 | P a g e
Neelis, J (2011) Early Buddhist Transmission and Trade Networks: Mobility and Exchange
within and beyond the Northwestern Borderlands of South Asia. Leiden: Brill

Nokandeh, J, Rekavandi, H O, Sauer, E W, & Wilkinson, T J (2012) Persia's Imperial Power in

Late Antiquity: The Great Wall of Gorgan and the Frontier Landscapes of Sasanian
Iran. Oxford: Oxbow Books

Odyssey (2011) The Ancient Silk Road., 2nd ed

Park, J S & Voyakin, D (2009) The key role of zinc, tin and lead in copper-base objects from
medieval Talgar in Kazakhstan, Journal of Archaeological Science 36 (3): 622-628

Paterson, J (1998) Trade and traders in the Roman world: scale, structure and organisation,
in Parkins, H & Smith, C (eds) Trade, traders and the ancient city 149-167. London:
Routledge

Ramankutty, N, Evan, A T, Monfreda, C, & Foley, J A (2008) Farming the planet: 1.

Geographic distribution of global agricultural lands in the year 2000, Global
Biogeochem Cycles 22

Razi, M (2010) Glass along the Silk Road in the Near East from the Persian Empire to the
Middle Age, in Zorn, B & Hilgner, A (eds) Glass along the silk road from 200 BC to AD
1000: international conference within the scope of the "Sino-German Project on
Cultural Heritage Preservation" of the RGZM and the Shaanxi Provincial Institute of
Archaeology, December 11th-12th 2008 Mainz: Verlag des Römisch-Germanischen
Zentralmuseums

Riello, G (2013) Cotton: The Fabric that Made the Modern World. Cambridge: Cambridge
University Press

Rtveladze, E V (2012) Velikii indiiskii put' [The Great Indian Road: from the history of the most
important trade routes of Eurasia]. St. Petersburg: Nestor-Istoriia

Sala, R (2003) Historical survey of irrigation practices in west central Asia.

http://www.lgakz.org/Publications/Publications.html: Accessed 24/3/2011

Schafer, E H (1963) The golden peaches of Samarkand: a study of T'ang exotics. Berkeley:
University of California Press

Schutyser, T (2012) Caravanserai: Traces, Places, Dialogue in the Middle East. Five
Continents Editions

Sevillano-López, D & González, F J (2011) Mining and minerals trade on the Silk Road to the
ancient literary sources: 2 BC to 10 AD centuries, in Ortiz, J E et al (eds) History of
Research in Mineral Resources Madrid: Instituto Geológico y Minero de España

Sherratt, A (2004) Trade Routes: The Growth of Global Trade. ArchAtlas: Accessed 6/4/2011

Silverstein, A J (2007) Postal Systems in the Pre-Modern Islamic World. Cambridge:

Cambridge University Press

148 | P a g e
Siroux, M (1949) Caravansérails d'Iran et petites constructions routières. Cairo: Institut
Français d'archéologie Orientale

Sit, V F S (2010) Chinese City and Urbanism: Evolution and Development. Singapore: World
Scientific Publishing

Skaff, J K (2003) The Sogdian Trade Diaspora in East Turkestan during the Seventh and Eighth
Centuries, Journal of the Economic and Social History of the Orient 46 (4): 475-524

Smagulov, E & Sala, R (2003) Karizi Turkestankogo oasika, Promislennost II

Smith, M L (1999) The role of ordinary goods in premodern exchange, Journal of

Archaeological Method & Theory 6 (2): 109-135

Spengler, R, Frachetti, M D, Doumani, P, Rouse, L, Cerasetti, B, Bullion, E, & Mar'yashe, A

(2014) Early agriculture and crop transmission among Bronze Age mobile pastoralists
of Central Eurasia, Proceedings of the Royal Society of London.Series B, Biological
Sciences 281 (1783): 20133382

Stride, S, Rondelli, B, & Mantellini, S (2009) Canals versus horses: political power in the oasis
of Samarkand, World Archaeology 41 (1): 73-87

Talbert, R J A (ed) (2000) Barrington Atlas of the Greek and Roman World. Princeton
University Press

Tavernari, C (2009) Roads & caravanserais in Medieval Syria. ArchAtlas: Accessed 30/3/2011

Tolstov, S P & Andrianov, B V (1957) Novie materiali po istorii razvitia Khorezm, Isnt
Ethnografii Kratkiie Soobschenia 26

Tomber, R (2008) Indo-Roman Trade: From Pots to Pepper. London: Duckworth

Tong, T (2013) The Silk Roads of the Northern Tibetan Plateau During the Early Middle Ages
(from the Han to Tang Dynasty). Oxford: Archaeopress

Tucker, J (2003) The Silk Road: art and history. London: Philip Wilson

UNESCO (2005) Operational Guidelines for the Implementation of the World Heritage
Convention. (WHC. 05/01) Paris: UNESCO

UNESCO (2006) The Silk Roads of Central Asia: Samarkand strategy. Final report, UNESCO
sub-regional workshop on serial nomination of the Central Asian Silk Roads,
Samarkand, Uzbekistan. 25-28 October 2006.

UNESCO (2007) A Concept for the Serial Nomination of the Silk Roads in Central Asia and
China to the World Heritage List. Unpublished report. Dushanbe: UNESCO

UNESCO (2008a) A Concept for the Serial Nomination of the Silk Roads in Central Asia and
China to the World Heritage List (updated text after the Consultation meeting in Xi'an
(China), June 2008). Unpublished report. Xi'an: UNESCO

149 | P a g e
UNESCO (2008b) Operational Guidelines for the Implementation of the World Heritage
Convention. (WHC. 08/01) Paris: UNESCO

UNESCO (2009a) 5th UNESCO sub-regional workshop on the serial World Heritage
nomination of the Silk Roads, 18 – 24 May 2009, Almaty, Kazakhstan: final report.

UNESCO (2009b) The first Coordinating Committee of the Serial Transboundary World
Heritage Nomination of the Silk Roads (3-5 November 2009, Xi'an, China). Final
Report.

UNESCO (2012) Operational Guidelines for the Implementation of the World Heritage
Convention. (WHC. 12/01) Paris: UNESCO

Warmuth, V M, Campana, M G, Eriksson, A, Bower, M, Barker, G, & Manica, A (2013) Ancient

trade routes shaped the genetic structure of horses in eastern Eurasia, Molecular
Ecology 22 (21): 5340-5351

Waugh, D C (2008) The Silk Roads and Eurasian Geography. Silk Road Seattle: Accessed
27/3/2011

Weiss, W M & Westermann, K M (1998) The bazaar: markets and merchants of the Islamic
world. London: Thames and Hudson

Wheatley, P (1971) The Pivot of the Four Quarters. Edinburgh: Edinburgh University Press

Wheatley, P (2001) The places where men pray together. Cities in Islamic Lands, Seventh
Through the Tenth Centuries. Chicago: University of Chicago Press

Whitcomb, D (2007) An urban structure for the early Islamic city: an archaeological
hypothesis, in Bennison, A K & Gascoigne, A (eds) Cities in the pre-modern Islamic
world: the urban impact of religion, state and society 15-26. London: Routledge

Whitfield, S & Sims-Williams, U (eds) (2004) The Silk Road: trade, travel, war and faith.
London: Serindia

Whitfield, S (1999) Life along the Silk Road. London: John Murray

Whitfield, S (2006) Was there a Silk Road?

http://www.ucl.ac.uk/histmed/centre_projects/medicine_asia/silk_roads: Accessed
28/3/2011

Whitfield, S (2009) La Route de la soie: Un voyage á travers la vie et la mort. Brussels: Fonds
Mercator & Europalia International

Williams, T & Wordsworth, P (2010) Merv to the Oxus: a desert survey of routes and
surviving archaeology, Archaeology International 12 (2008/09): 27-30

Williams, T (forthcoming a) The Silk Roads: data from the ICOMOS thematic study, Journal of
Open Archaeology Data

150 | P a g e
Williams, T (forthcoming b) A gazetteer of Silk Roads corridors, UCL Discovery
http://discovery.ucl.ac.uk/

Wood, F (2002) The Silk Road: Two Thousand Years in the Heart of Asia. Berkeley: University
of California Press

Wriggins, S H (2004) The Silk Road journey with Xuanzang. Oxford: Westview Press

Young, G K (2001) Rome's Eastern Trade: International Commerce and Imperial Policy 31 BC-
AD 305. London: Routledge

Zhang, J, Lu, H, Wu, N, Qin, X, & Wang, L (2013) Palaeoenvironment and agriculture of
ancient Loulan and Milan on the Silk Road, The Holocene 23 (2): 208-217

Zorn, B & Hilgner, A (eds) (2010) Glass along the silk road from 200 BC to AD 1000:
international conference within the scope of the "Sino-German Project on Cultural
Heritage Preservation" of the RGZM and the Shaanxi Provincial Institute of
Archaeology, December 11th-12th 2008. Mainz: Verlag des Römisch-Germanischen
Zentralmuseums

151 | P a g e
International Council of Monuments and Sites

11 rue du Séminaire de Conflans 94220 Charenton-le-Pont, France

www.icomos.org
secretariat@icomos.org

ISBN 978-2-918086-12-3

152 | P a g e