## From Small States to Universalism in the Pre-Islamic Near East

## Mark Altaweel Andrea Squitieri

**REVOLUTIONIZING** 

**REVOLUTIONIZING A WORLD**

Mark Altaweel and Andrea Squitieri

This book investigates the long-term continuity of large-scale states and empires, and its effect on the Near East's social fabric, including the fundamental changes that occurred to major social institutions. Its geographical coverage spans, from east to west, modernday Libya and Egypt to Central Asia, and from north to south, Anatolia to southern Arabia, incorporating modern-day Oman and Yemen. Its temporal coverage spans from the late eighth century BCE to the seventh century CE during the rise of Islam and

The authors argue that the persistence of large states and empires starting in the eighth/ seventh centuries BCE, which continued for many centuries, led to new socio-political structures and institutions emerging in the Near East. The primary processes that enabled this emergence were large-scale and long-distance movements, or population migrations. These patterns of social developments are analysed under different aspects: settlement patterns, urban structure, material culture, trade, governance, language spread and religion, all pointing at population movement as the main catalyst for social change. This book's argument is framed within a larger theoretical framework termed as 'universalism', a theory that explains many of the social transformations that happened to societies in the Near East, starting from the Neo-Assyrian period and continuing for centuries. Among other infl uences, the effects of these transformations are today manifested in modern languages, concepts of government,

Mark Altaweel is a Reader in Near Eastern Archaeology at the UCL Institute of Archaeology having held previous appointments and joint appointments at the University of Chicago, University of Alaska, and Argonne National Laboratory.

Andrea Squitieri studied Archaeology of the Ancient Near East at the University of Turin and at UCL (PhD 2015). He has been a post-doc researcher since 2015 working in the Peshdar Plain Project, based at the Ludwig-Maximilian University of Munich,

> **Back cover image:**  Photograph by A.Davey,

**Front cover image:**  Photograph by Mike Peel, CC-BY-SA-4.0

> CC BY 2.0 **Cover design:** www.ironicitalics.com

focusing on the study of the eastern border of the Neo-Assyrian Empire.

universal religions and monetized and globalized economies.

collapse of the Sasanian Empire.

Free open access versions available from

www.ucl.ac.uk/ucl-press

**A WORLD**

## Revolutionizing a World

# Revolutionizing a World

*From Small States to Universalism in the Pre-Islamic Near East*

Mark Altaweel and Andrea Squitieri

First published in 2018 by UCL Press University College London Gower Street London WC1E 6BT

Available to download free: www.ucl.ac.uk/ucl-press

Text © Authors, 2018 Images © Authors and copyright holders named in the captions, 2018

The authors have asserted their rights under the Copyright, Designs and Patents Act 1988 to be identified as the authors of this work.

A CIP catalogue record for this book is available from The British Library.

This book is published under a Creative Commons 4.0 International license (CC BY 4.0).This license allows you to share, copy, distribute and transmit the work; to adapt the work and to make commercial use of the work providing attribution is made to the authors (but not in any way that suggests that they endorse you or your use of the work). Attribution should include the following information:

Altaweel M. & Squitieri A. 2018. *Revolutionizing a World: From Small States to Universalism in the Pre-Islamic Near East*. London: UCL Press. DOI: https://doi.org/ 10.14324/111.9781911576631

Further details about Creative Commons licenses are available at http://creativecommons.org/licenses/

ISBN: 978–1–911576–65–5 (Hbk.) ISBN: 978–1–911576–64–8 (Pbk.) ISBN: 978–1–911576–63–1 (PDF) ISBN: 978–1–911576–66–2 (epub) ISBN: 978–1–911576–67–9 (mobi) ISBN: 978–1–911576–68–6 (html) DOI: https://doi.org/10.14324/111.9781911576631

## Preface

Many scholars and even some of the wider public recognize how the ancient Near East and Egypt contributed to modern societies. Whether it was through the sciences, literature, writing, arts, governing systems or even religion, the imprint is clear. However, some periods in the ancient Near East and Egypt look more alien to us than others. In particular, many features of the third and second millennia BCE, including ethnic groups, religions, governments, languages, and even the media for economic exchange, appear very strange to us. In the first millennium BCE, circumstances began to change and we begin to see facets, such as languages, population groups, government and social institutions, and ideas, that we find more familiar. A simple argument is that over time things change, and therefore cultures closer to today should be more familiar to us. But could there be a process that demonstrates why societies shifted to create some of the cultural traits we are more familiar with? As these changes were occurring, another clear pattern emerged, in that large states had become common. Are these two phenomena related? We think that there *is* a link, and we propose a process that we term *universalism* to explain such changes. We are also aware that such terms are often criticized, and perhaps too many terms are used to describe different cultural developments. Nevertheless, the utility of this term is that it helps to explain a process of commonalities that forms in the first millennium BCE. The evidence of such wider common attributes is clear. Hellenism is one such development: in effect, it is a merger of different cultural trends that included Greek and Near Eastern styles and cultural traits. Like Hellenism, universalism is an older term, but we provide a different way of understanding it: we look at the core attributes and qualities that made common traits emerge. Thus, universalism serves to decode a process that explains elements we see as combined, that is, the formation of new social and cultural phenomena, the creation and continuity of large states, and the fundamental process that enabled such change, which we see as population movement.

There are often clear and stark divides, in teaching about the ancient Near East, between the period before Alexander's conquest of the Near East and that which followed it. Magically, it often seems, Alexander's invasion caused some seismic change in the ancient Near East that brought about a process whereby the region became so often dominated by foreign entities that the old religions and customs began to wither away. Before the events of 334 BCE and Alexander's great invasion – that is, in the late Neo-Assyrian and later periods – what is telling is that empires had already become very large. If we look at what should have happened after the fall of the Neo-Assyrian Empire, between 612 and 605 BCE, the Near East should have reverted to a pattern of small states or even city-states, as it so often did in the Bronze Age. While some areas did indeed fragment, in general the Neo-Babylonians and the Medes created their own large political entities on the removal of the Neo-Assyrians. Not only did the region not fragment politically, but also states became even larger and, even after their scale reached a peak in the Achaemenid period or even in that of Alexander's empire, for millennia empires continued to be large, often spanning large parts of Eurasia. There has been little discussion of the topic of the continuity of large-scale empires in a single region. The process that enables large states and empires to become the political norm is not well understood in the context of preceding periods, which often showed a reversion to small, fragmented states after the collapse of major dynasties.

Our inquisitiveness about large states becoming the norm, along with our noticing major institutional and cultural changes such as those indicated above, helped us start the project of writing this book. At first, it did not lead to many ideas. One key factor, though, stood out as our investigation unfolded, which was settlement patterns: data provided information on the size and distribution of archaeological sites across a region. The size and distribution of settlements across periods from the Neo-Assyrian and into later periods showed structurally similar patterns. Yes, survey data often vary and results are not always certain, given the quality of the data captured. But these weaknesses do not hide some clear facts. Some regions, such as Southern Mesopotamia, developed extremely large cities, far larger than in previous periods. Other regions, such as Northern Mesopotamia, showed a contrast: sites were much smaller, and the larger urban patterns of the Bronze Age seemed to largely disappear. Rather than comparing settlements with some absolute size (e.g., 100, 200, 300 hectares), we compared the sites with each other, which made the patterns clearer. This pattern of urban change did not occur simultaneously throughout the Near East. Initially, we found that major settlement pattern shifts occurred in Mesopotamia at the end of the Iron Age. However, as other regions were assessed, patterns comparable to those found in Mesopotamia began to emerge, even if they occurred in later periods. What caused this change became an important question in our minds, and this is where the story of this book began. As settlement patterns shifted so too did other social patterns evident in historical and archaeological records. This then became our main area of exploration, and population movement emerged as the common theme in the data we had examined.

The methods we used to investigate changes in settlement patterns and other social and cultural phenomena, with a view to addressing the larger issue of why the Near East fundamentally changed (in our minds, from the late Neo-Assyrian period), are not typical in Near East archaeology. They do, however, demonstrate some key differences in what happens before and after the development of large-scale empires. The data used include settlement patterns, material culture and textual sources. We cover a long time span in this volume, inevitably diluting a focused look into any one period, but that long view helps to show whether subsequent patterns look generally similar or different, an important feature in our view. We look particularly at the periods from the Neo-Assyrian to the Sasanian; however, we compare this era with the earlier Bronze and Iron Ages. We will inevitably miss many aspects and details because of this focus, but it is critical to demonstrating the larger patterns of social change in this volume. This is why, throughout, we discuss what happens before the development of continuous large states and empires and what happens in the Neo-Assyrian period and after.

Social change itself is not the most important factor in our investigation; rather it is population movement, the main dynamic that enables this social change, that focuses this book. Other volumes have looked at how government, religion and other social phenomena change in periods they consider 'globalized', but a key difference here is our focus on the changes that are evident prior to major institutional changes becoming prominent. Migration has been present throughout human history, but the scale of movement, and how populations integrate, engender the ways in which subsequent social change unfolds. This focus on movement underlies the discussion provided in the chapters throughout this volume.

This type of work develops neither over a narrow time span nor without influence from colleagues. In fact, years of influence from scholars we have interacted with have shaped this research, just as much as our own work and experiences. It is these people we have to thank for their inspiration. The late Tony Wilkinson, John Christiansen, McGuire Gibson, Muzahim Mahmoud Hussein, Hussein Ali Hamza**,** Andrew Bevan, Karen Radner, Alessio Palmisano, Simone Mühl, Peter Miglus, Stephen Shennan, Alan Wilson, David Wengrow, Kris Lockyear, Paolo Fiorina and St John Simpson have provided encouragement or inspiration to parts of this volume. Numerous others, including undergraduate and graduate students, have listened to parts of the book's ideas; their feedback has often been incorporated in this work. Undoubtedly, such a book will have errors; we hope they are minimal but they are entirely our fault.


**Table 0.1 Major historical periods, states and empires and their approximate dates**

## Contents


## List of figures and tables











newgenprepdf

## 1 Introduction

## 1.1 Why this book?

Few works have looked at the effects of the long-term continuity of large-scale states and empires on a region's social fabric and what, if any, fundamental changes occurred to major social institutions in the context of these political forms. In the ancient Near East, there is a pattern in the Bronze (3000–1200 BCE) and Early Iron (ca. 1200 to the ninth/eighth centuries BCE) Ages whereby city-states and small states were the political norm, punctuated by periods of larger territorial states and empires. Populations and regions were generally politically fragmented, even when cultural interactions became common. At times, empires such as those of the Akkadians or Hittites arose, but once these states collapsed the pattern generally reverted to small territorial states. The nature of political organization changed with the Neo-Assyrian Empire, particularly in the late stages of the empire in the late eighth and seventh centuries BCE. From this period, and into time spans beyond the rise of Islam in the seventh century CE and lasting until the end of the Ottoman period in 1922 CE, large territorial empires became common, or even the political norm, throughout the Near East (Finkel 2006; Cline and Graham 2011; Peacock and Yildiz 2013). The size of empires based in the Near East peaked in the eighth century CE, with the Islamic Caliphate stretching from Spain to Central Asia. This long era of empires, when these entities were common, can be termed an 'Age of Empires', or AoE.

In the eighth and seventh centuries BCE, the Near East underwent changes that affected social institutions such as settlements, the economy, artistic expression, social interactions, political structures, religion and languages. From the seventh century BCE to the seventh century CE, very large cities, far larger than any seen earlier, developed. The more condensed pattern of urban centres and settlements found in the Bronze Age in parts of the Near East disappeared, and greater disparities between the largest cities and second-tier towns emerged in the wider region, as people gathered in more restricted regions (e.g., see Mazzoni 1991–2). Much larger cities in the AoE were generally established along coastal regions or major rivers, or in locations affiliated with the homeland regions of large ruling powers, while interior regions in the Near East became less settled or had smaller settlements. Cities such as Babylon (Gibson 1972; Pedersén 2011), Seleucia (Invernizzi 1976; Hannestad 2012), Antioch (Kondoleon 2000), Alexandria (Haas 2006) and the Ctesiphon urban area (Christensen 1993; Lee 2006) were among the largest in antiquity, and much larger than many earlier Bronze Age cities. Such a development, it could be argued, was a new form of urban revolution, in which primate cities (cities much larger than other cities in their region) reached unprecedented sizes, had trade interactions spreading to very distant areas, contained religious institutions that originated from different regions, and had socially and ethnically diverse populations. Other regions became more intensively settled as they became associated with increased trade and other interactions spanning much greater distances across the Old World. These changes were contemporary or nearly contemporary with such social transformations as the emergence of universal governments that controlled vast areas, the spread of coinage, more direct and intensive long-distance trade, shared iconographic and artistic elements, increased use of common languages, more diverse cultural groups living together, and eventually the rise of religions termed universal, whose doctrine is intended to be relevant for all people in larger empires and beyond.

## 1.2 Central argument: universalism and its social foundations

The region this book examines covers, from east to west, modern-day Libya and Egypt to Central Asia; from north to south it covers Anatolia to southern Arabia, incorporating modern-day Oman and Yemen. The period focused on, the AoE, extends from the late eighth century BCE to the seventh century CE during the rise of Islam and the collapse of the Sasanian Empire. However, earlier periods, termed pre-AoE, are discussed and are compared with this time span. While the wide spatial coverage means we cannot look at all these regions in detail, and some data covering the time span will be neglected, we recognize the importance of an extensive time and spatial outlook to an understanding of

the long-term patterns and major social change that this book addresses. Figure 1.1 shows the region and the areas within it that will be discussed throughout this work. In general, we will refer to this large area as the 'Near East'; we recognize that it covers areas beyond the traditional boundaries of the ancient Near East, but the term is convenient for simplifying the wide spatial coverage. Some parts of the book will cover areas even wider than this primary area of focus, spanning the breadth of the Old World from Europe to East Asia. This scope is intended to demonstrate the change to social institutions relevant to the ancient Near East covered in this volume.

## 1.2.1 Definition of empire

Before we present our central focus, we provide a basic definition of what we mean by empire. A number of definitions can be used, which include politically, economically and even informally based actors in these political entities. For our work, we mean any interaction between two or more political entities whereby one entity exerts political control, including of internal and external policies, over at least one other state or territory outside the area it had controlled in an earlier period. In short, one government has sovereignty over another government or region, largely following a definition given by Doyle (1986: 12). Cline and Graham (2011: 4), in their assessment of empires across a long period, apply a similar definition. For this volume, empires have areas in which they exert political control, and are referenced by the territorial extent in which they exert such control over a given time. Thus, as Sinopoli (1994: 160) states, they are expansive and incorporative of given regions.

### 1.2.2 Research argument

We argue that the persistence over many centuries of large states and empires, from the eighth/seventh centuries BCE, led to the emergence of new socio-political structures and institutions in the Near East. The primary processes that enabled this emergence were large-scale and longdistance movements, or population migrations. By movement we mean both forced and voluntary migration, including deportation and movement because of new opportunities created by large states and empires that led to the concentration of people, usually from different cultures, and ultimately of wealth and power, in large cities or high-population regions. In contrast, the scale of movement in the Bronze and Early Iron Ages was more limited. Movement, as defined here, is how populations from distant locations could spread or disperse in a landscape, and interact between settlements and regions. While we cannot easily determine absolute population concentration in regions, differences between regions allow a determination of how populations shifted over time. Ultimately, the study of movement investigates how people from distant regions lived in new areas and with different population groups. Such movement was certainly present long before the rise of major empires; however, the scale, spread and speed of the movement of populations during periods of large territorial states were at a qualitatively different level. Movement also became characterized by the integration of varied cultural traits rather than by one or only a few strands of cultural expression. While cultural diversity becomes evident, and was maintained, in the AoE, the amalgamation over time of varied cultural traits helped lead to the emergence of new, common and even universal cultural expressions that were shared by many different groups. This is evident from the shared ideas and material cultural characteristics found in various regions.

For our purposes, a long-term, persistent or evident pattern that enabled population movement or dispersal is of interest. Some movement events, such as invasions, may be temporary or easily reversed, leaving few material traces for analysis. Movement such as this could have occurred at different paces, as a single occurrence or throughout specific periods; however, the pattern or its result had to persist for long enough for it to be measurable. For instance, settlement structures that persisted over long periods can indicate how populations were configured in a region. A key measure is to determine if there was a process by which large numbers of people could cross long distances more easily than in earlier periods. Such movements should leave traces, including different settlement patterns and settlement sizes, and new material culture. A particular period may have less movement, but the long-term trend of population movement is important to our argument. Interactions among different populations, including those of different social groups living together and beginning to blend ideas and practices, facilitate an emergence of commonalities, shared ideas and new institutions among these disparate groups.

Movement potentially enables and perpetuates large-scale sociopolitical integration and cohesion, which will be discussed in later chapters. The population movement afforded by the large states and empires facilitated the development of new economic opportunities, social interactions, government structures, shared language on a large scale, and even new religious ideas. Furthermore, these new institutions themselves probably attracted more or greater movement, as new opportunities arose. In effect, positive feedback developed that allowed a system to attract greater population concentration into limited regions and cities. Work by Cavalli-Sforza (2001) discusses movement, or population diffusion, in such periods as the Neolithic in Europe as a demic process by which people carried farming cultural practices with them to different parts of Europe rather than teaching agricultural practices to neighbouring populations. Similarly, although cultural influence can occur through teaching and the diffusion of ideas, the movement of people in the AoE, we argue, played a critical role in the reshaping of key social institutions and facilitated the establishment of a pattern of long-lived, large empires in the Near East.

Once disproportionate population concentrations in large centres or regions developed, along with a more rural countryside in areas that once had large populations, the emerging social and political patterns were not easily reversed, as established patterns became self-reinforcing and control of major cities facilitated the creation of large territorial empires. This emergent system created incentives for its persistence, even after the fall of specific dynasties or states, and the renewal of large states and empires became easier, which, along with the retention of various and distinct cultural groups, made the Near East socially more cohesive. In other words, even as ethnic groups retained their unique identities, they formed new social bonds with other groups. This does not mean that the developed institutional structures were static. On the contrary, they were changing and adaptive, as the process was transformed when new cultures brought new ideas and influences. At times the blending and acceptance of multiple cultural ideas was evident, while at others a single, universal political, philosophical or religious idea emerged. Both developments indicate a shared identity or commonality that facilitated the integration and interaction of diverse population groups that lived within large states. Even as social institutions changed, political adaptation still facilitated and attempted to perpetuate large states.

Figure 1.2 demonstrates the conceptual model used and argued for in this volume. Movement of population is shown as the fundamental process that leads to new institutional developments. However, those institutions, once developed, may also facilitate greater movement over time, through, for example, trade contacts, including the establishment of colonies, government policies or even a common language that allows easier social integration. Demonstrating the presence of population movement, and that developed institutions are related to movement, is the key focus of this volume. Many of the ensuing changes may not have been planned; they were probably the results of an increased presence of mixed populations living together and interacting.

**Figure 1.2** Movement of people to new settings influences and restructures institutions such as those indicated. These changes facilitate greater movement

We will present data and analyses that support our stated position. Additionally, we believe there is a larger theoretical framework in which we can formulate our argument. This framework is *universalism*, a term that has been used to describe political and religious structures in which disparate population groups are considered to be under a unitary authority (see, e.g., J. Assmann 2010; Cline and Graham 2011; Bang and Kołodziejczyk 2012). The prior existence of this term suggests that our work is not novel: others have argued that empires transformed social institutions, and presented information on how they did that, but our work is new in that it looks at long periods in a single region and the ways in which institutions were transformed by empires and movement. Rather than comparing empires from different regions, we take a long-term perspective on a specific and wide area, demonstrating how long-term patterns of change allowed the development of new social structures that permitted the persistence of large states and empires. We also modify the definition of universalism to encompass other social factors affecting individuals and households in these large territorial states. We see universalism as a socially holistic and pervasive social transformation that deeply affected society at its fundamental levels, encompassing more than just political or religious attitudes. In effect, political and religious change may reflect broader social commonalities in trade, art, language patterns, knowledge, and other forms of social interaction. We also look at how these changes helped larger political entities to become the political norm. Our description and definition of universalism is:

The process whereby social transformation enables socio-cultural commonality and sharing across many different populations over vast distances (hundreds or thousands of kilometres). While individual social groups often retain their own identities and distinctiveness, they also form common social bonds with other populations that are manifested in economic, artistic, linguistic, political, religious and other social forms, through, for example, syncretism or universal philosophies. These commonalities are evident from material culture and historical records.

We argue that universalism is an appropriate term for such periods of large empires because it was the circumstances and continuity of a system in which people were able to move to major centres and population regions or be dispersed into the countryside that made universalism a pervasive social phenomenon that affected many aspects of life in the Near East during the AoE. Universalism is still dependent on common cultural interactions and on populations having the ability to move or be moved to established centres. Within the framework of universalism, movement and interactions are on a scale at which new social developments emerged and became established. This book does not focus on the exact time at which specific ideas or institutions arose; rather the focus is on their establishment and continuity over the AoE.

Overall, we see universalism as a theory that explains many of the social transformations that happened to societies in the Near East from the Neo-Assyrian period onward. It frames our understanding of important transformations that occurred in the AoE, from which syncretistic developments or universal concepts emerged. It is a social phenomenon that affected individuals as well as larger state structures and institutions. The importance of universalism as a theoretical idea is evident, as it provides a useful way to explain how the new institutions and social norms that developed in the first millennia BCE and CE differed from those of the earlier Bronze and Early Iron Age cultures. While universalism helped to perpetuate empires and social institutions, the persistence of large states and empires also facilitated the long-lived effects of universalism on society. Long-lived patterns of empire and universalism that enhanced each other became established as a feedback system in the AoE. This is relevant because we see universalism as a key social transformation that eventually had a profound influence on many modern societies and institutions, including those in the West and other regions. As an example, the spread and use of common languages is a process that universalism helps to explain, while the benefits of their use, to both states and individuals, reinforce their utility to subsequent states and individuals, even in modern societies. These common languages, including Aramaic and Greek, have subsequently influenced numerous other, modern languages and helped to spread shared ideas.

## 1.3 Universalism and ancient globalization

The concept most closely related to universalism is globalization; however, we believe there are key differences between these concepts. Recent works have looked at the idea that certain periods present greater social interaction, and even integration, between distant societies. Often this is in the form of trade, although cultural and social influences occur as part of these interactions. Migration also forms part of this interaction: the flow of people brings ideas and concepts into close contact. Works that have investigated concepts of globalization, in the frame of so-called 'Big History' or *longue durée* perspectives, include LaBianca and Scham (2006), A. Assmann and Conrad (2010), Stearns (2010), Grinin (2011) and Cunliffe (2015). These works have looked at how increased contacts, cultural syncretism and continuity of empires contribute to globalized, long-distance interactions. In our opinion, however, a distinction should be made between globalization and universalism. If we use the term globalization (see, e.g., Boudreaux 2008 for definitions), we see that factors of economics, migration, disease transmission, culture and trade are common drivers of, or the products or factors affecting, globalization. Furthermore, modern communication technologies and mass media play a major role in how populations begin to share cultural concepts today. Movement of people, in its use with globalization, is not a required state.

What is distinct about movement, compared with other factors that lead to social interaction, is that when a common social phenomenon develops it has to accommodate or address the diversity found in society. Globalized societies can be influenced by very distant ideas and concepts, but closed to major migration or integration of foreigners (e.g., see Ritzer 2010: 208 on Japan). Societies can be globalized without being universal. Changes within such societies do not have to address the presence of new population groups. On the other hand, migrating populations from different ethnic groups will bring their own social norms and practices, and these populations will have more daily and common interactions with other population groups. As social groups are incorporated into another society, they may continue the social practices brought with them or develop new social practices adapted to an intermixed population. We argue that both are in evidence in the AoE. Furthermore, there is nothing in the ancient past that easily replicates today's mass media, which suggests that closer personal interaction was vital for many of the evident types of social change. What is necessary is the development of new cultural traits and institutions that accommodate a variety of people living together as migration increases, particularly as people share similar ideas or even religions. This was the case as states became ever more expansive, which facilitated the movement of people to new places from distant regions. The continuity of large states and empires not only allowed easier movement, it also provided time for new cultural phenomena to become ingrained and dominate people's lives.

Movement was not only affected by economic incentives, although the latter were one of the reasons people moved. At times, movement had political or religious causes. In fact, one of the initial triggers of universalism in the AoE may have been forced migrations that helped to blend populations in the Near East. The central importance of movement, whereby people migrate from distant places and begin to live with ethnically diverse populations, for any reason, makes universalism different. The term globalization is simply not sufficient to demonstrate that the process of movement is the key driver. Universalism, we feel, explains a more pervasive process that affected many aspects of social change relevant to the ways in which cultural processes became shared over time.

Although various works have looked at the ancient roots or the concepts of globalization, few have looked at the fundamental effects of increased social contacts and transformed institutions based on population movement. This type of analysis requires that we start at the level of settlements: settlement structures, or hierarchies, are among the best evidence of the large population shifts or distributions that occurred in particular periods and which indicate that something more than natural population growth had occurred. Cities should also show different characteristics: evidence of multiple ethnic groups and material culture shows that movement over a large distance was a key driver of the changes that occurred. Globalization could be seen as overlapping with universalism, or related to universalism, as long-distance contacts did increase in the AoE, but we believe the concept does not address the key dynamics that allowed the Near East to change from its pre-AoE characteristics. Nor does it fully explain why the pattern of large states persisted in the Near East.

### 1.4 Structure of presentation

To capture the essence of the argument and the theory, we examine a period long enough to show that a new socio-political pattern had emerged and was perpetuated in the Near East. We present an overview of earlier patterns for societies, including settlement structures, economy, material cultures, state organization, language patterns and religions in the Bronze and Early Iron Ages. The work investigates how the process of universalism began in the Neo-Assyrian state and then continued through the rise of the Babylonian, Achaemenid, Hellenistic, Seleucid, Parthian, Roman, Byzantine and Sasanian states, a time span covering ca. 800 BCE to 651 CE. While we believe universalism continued after the rise of Islam, the period addressed is sufficiently long to show that the trends of universalism had become well established as a socio-political norm, so that it could be perpetuated even in later states. The length of this period affords us the opportunity to demonstrate that important patterns of social development occurred during the AoE and that these developments affected individuals and households, and created institutions that show that universalism had indeed become pervasive at many levels of society, forever transforming the Middle East, and the Western and other societies that inherited some of these traits. We will look at the same social measures for the Bronze and Early Iron Ages and in the AoE period, showing how social patterns we attribute to universalism changed the Near East. Both the historical and the archaeological records from the timescale investigated provide us with sufficient data to demonstrate a long-term process that shaped universalism, even though short-term periods within that span might show short-lived or more minor social trends. The aim, therefore, is to show what overall and general patterns of socio-political development occurred.

Chapter 2 provides an overview of the historical data, which form a set of information used to demonstrate key societal transformations between the Bronze and Early Iron Ages and the later periods up to the fall of the Sasanian Empire. We present a general historical summary of events in the region across the periods covered. The trends discussed include the size of territories in different periods and evidence of increased political integration across the time span that show a transforming Near East.

In Chapter 3, we outline the methodology we used for demonstrating population movement. The intent of this chapter is to show how the methods used demonstrate that settlement patterns and material culture changed in the AoE, and that movement is a likely reason for such change. Other data obtained and used in chapters will be discussed.

In Chapter 4 we assess settlement patterns, using the methodology described in Chapter 3. We demonstrate how urban patterns shifted, using qualitative, quantitative and modelling and simulation approaches that show which conditions and scenarios facilitate population movement and the concentration of people in larger cities or specific regions. We will show that the large cities became nodes that drew people to them, particularly as movement over long distances became easier during periods of large states and empires.

In Chapter 5 we will look at the nature of urbanism in the AoE, and discuss how it differed from the urbanism of earlier periods. Urban population centres in the AoE began to have diverse approaches to art, integrating influences from many regions; populations began to be more multi-ethnic, a greater diversity of gods became evident, and a variety of shared ideas and expressions began to characterize life in cities. The chapter will also look at how villas or estates replaced areas where larger settlements or cities were once found, demonstrating how the countryside, too, changed during the AoE.

Chapter 6 demonstrates the shift in trade patterns that meant that more direct, large-scale and long-distance trade became the norm by the end of the Iron Age and during the pre-Islamic periods. The diffusion of coinage and incense, for example, helped to connect distant places both within the Near Eastern empires and beyond, as far as China and Europe in Late Antiquity. A more unified economic system connected distant regions, which took advantage of a new-found socio-political integration as movement facilitated interaction. As the historical records demonstrate, this provided opportunities not just for states but also for individuals and private enterprise.

In Chapter 7, we focus on types of material culture that became similar or common across vast distances during the AoE. New hybrid styles emerged under the AoE, as a result of shared ideas, in designs blending stimuli from the Mediterranean, the Near East, Central Asia and India. During the AoE, such material culture hybridization permeated all strata of society, unlike in the pre-AoE when the 'intercultural style' was mainly an elite phenomenon. The AoE hybrid material culture reflects a multiethnic population that lived in cities and towns; at the same time, the spread of the same stylistic features over a vast area is a proxy for greater movements of people across and between empires than that which had occurred in the Bronze and Iron Ages.

Chapter 8 will show that the AoE provides strong evidence that governments attempted to unite new and disparate populations. Large states and empires created opportunities or circumstances in which people migrated, and began to present themselves as unifiers of different population groups. This is in contrast to earlier Bronze and Early Iron Age states and cultures, which were predominantly focused on displaying and promoting their cultural differences, or cultural aspects distinct from surrounding groups. Using various sources, we will analyse how such actions and policies in the AoE enabled both propaganda and actions that facilitated a new form of unity across cultures that persisted.

The role of language, and shifts from diverse language groups to a pattern of common languages over large areas in the AoE, are discussed in Chapter 9. In the Bronze Age we see the spread of written language, but many linguistic differences and barriers remained. While Akkadian was the first common language in diplomacy and correspondence, Aramaic and Greek became the first languages that many levels of society across vast regions and populations were able to speak, read and write. We examine how the use of a common language affected social interaction and integration in societies.

In Chapter 10, religions as proxies for larger shared ideas that became common or more accepted are discussed. We examine how shared ideas in some of the polytheistic faiths foreshadowed the development of the monotheistic and universal faiths, whose ideas claim validity for all. Universal religions are also discussed. This is in contrast to the Bronze and Early Iron Age religions, mythologies and gods, which were predominantly associated with specific cities or small states.

Chapter 11 integrates the preceding chapters to demonstrate how universalism can be seen as an overarching, holistic, theoretical perspective that helps to explain many of the evident changes discussed in the earlier chapters. We demonstrate the factors that enabled universalism to be a force for social change, which enabled its own continuity as a longlived pattern for large states and empires that ultimately facilitated deep social change.

While we do not focus our analysis on the periods after the rise of Islam, Chapter 12 demonstrates why universalism is an important concept for understanding today's events, institutions and ideas. The chapter extends the idea of universalism to later periods, covering the wider Middle East and briefly examining how it rarely returned to earlier social patterns of more fragmented political entities and populations. On the other hand, could modern events in the Middle East, including the rise of groups such as so-called Islamic State, demonstrate a reversal of universalism, so that fragmentation has begun to emerge in the twentieth and twenty-first centuries? This helps to extend the value of universalism as a theoretical framework that explains how the cycle of modern events is unfolding, and why they differ from those in the past. More recent and long-term patterns of political fragmentation in the Middle East could lead to future developments in which new socio-political institutions arise.

## 2 Historical overview

Many scholars have extensively covered the political history of the Near East and surrounding areas between the third millennium BCE and the first millennium CE. Here we provide a general historical overview, giving the background of the periods covered and displaying key, long-term historical trends related to size of states, speed of conquests and other key events or factors that socially and politically shaped the region during the pre-AoE and the AoE. Different chronologies have been applied to much of the Bronze Age, in particular from the third to the second millennium BCE. We use the Middle Chronology for events, and timelines for states, dating Babylon's fall to a Hittite invasion fixed to 1595 BCE.

## 2.1 From the Early Bronze Age to the Early Iron Age

#### 2.1.1 Early Bronze Age (3000–2000 BCE)

For the first half of the third millennium BCE in Mesopotamia, our knowledge of key historical periods is limited to brief episodes. Writing, for the most part, was limited to Southern Mesopotamia, Elam, a few places in Northern Mesopotamia, including Tell Beydar and Mari, Ebla and Egypt. Very few other sites have yielded any texts from this period in Northern Mesopotamia, and many of these are fragmentary, or insufficient to piece together a wider picture. Figure 2.1 shows the region and the key sites found in the third millennium BCE.

It has been argued that the historical and settlement data for Southern Mesopotamia in the first half of the third millennium BCE show a politically fragmented region (R. McC. Adams 1981; Van De Mieroop 2016). The rise of city-states began in the Early Dynastic period (ca. 2900–2350 BCE), although much of our data either relies on later copies or is derived from the end of the Early Dynastic period. From this fragmentary picture, border conflicts were evident (for example between Umma and Lagash), signifying a pattern of conflict between nearby cities.

**Figure 2.1** Map of the region, cities and states from the Early Bronze Age. Names of regions or states (e.g., Subaru, Elam) are in larger type. The boundaries indicate the approximate maximum extents of the larger states and empires during the third millennium BCE

Geopolitically, the historical data from Northern Mesopotamia have been interpreted to show competition between small states and economies (Akkermans and Schwartz 2003). The kingdom of Nagar is one such example of a pre-Akkadian third-millennium-BCE state (Archi 1998).

In Elam, a geographic designation constructed by Southern Mesopotamian scribes, and nearby locations that covered the regions or cities of Susa, Awan, Shimashki and Anshan, we do have texts that cover this period. However, the very earliest phases, from ca. 3200 to 2700 BCE, are enigmatic, as Proto-Elamite has not been deciphered. Most of the historical documents in the first half of the third millennium are from the point of view of Southern Mesopotamia, and have an unclear and biased perspective for much of this time span. Elam in the third millennium did not appear to be a unified political entity. During the Akkadian Empire (ca. 2334–2154 BCE) and later, there were more references to Elam and to conflict between Elamite cities and Southern Mesopotamia. These conflicts with Mesopotamia may have enabled the rise of an Awanbased dynasty during the Akkadian period and eventually one based in Shimashki, perhaps near or just north of Elam, in the Ur III period (Vallat 1980; Carter and Stolper 1984; Potts 1999).

During the period of the Akkadian Empire, a large state emerged in Southern Mesopotamia; it was the first documented political state to dominate wide areas of the Near East. More primary texts are found, and a clear imperial presence of the Akkadian state could be seen in distant regions, including at settlements such as Tell Brak (ancient Nagar; Oates, Oates and McDonald 2001) and Tell Leilan (ancient Shehna; de Lillis Forrest, Milano and Mori 2007). The presence of an Akkadian imperial administration in northern regions suggests a territorial state that was centrally administered. However, there is no clear indication that any major population shifts had occurred whereby large population groups had intermixed, such as movement of Akkadians to Northern Mesopotamia. Eventually, the Akkadian state collapsed in its core regions, which historically has been attributed to a Gutian invasion (Liverani 2014).

After this, a period of political instability or fragmentation occurred, which emerged as a common pattern in the Bronze and Early Iron Ages. Another Southern Mesopotamian state that unified much of Southern Mesopotamia appeared in the form of the Ur III state (ca. 2112–2004 BCE). Its territorial extent was not as vast as the Akkadians', as it had different degrees of control in parts of Northern Mesopotamia and Iran. Vast numbers of economic and administrative documents written during this state's short, nearly 100-year, reign, have greatly informed historians about how its government operated. The state administered a provincial system of different levels of tax commitment to the central government at Ur. These so-called *bala* and *gun mada* systems were tax/tributedistribution systems that helped to finance the Ur III state and its various obligations to cities, individuals and temples (Steinkeller 1986; Sharlach 2004). By 2004 BCE, the Ur III state had collapsed because of an Elamite invasion, and Ur itself was sacked (Liverani 2014).

Throughout this time, the other historical state was Egypt, lasting most of the third millennium BCE. In fact, for a relatively large state of this period, it showed remarkable stability, lasting from ca. 3100 to 2181 BCE, with the First Intermediate Period (ca. 2181–2055 BCE) putting an end to the unified state. Egypt was united under Narmer, or soon after his reign, by ca. 3100 BCE during the First Dynasty, a formative period throughout Egypt that lasted until 2650 BCE (Wengrow 2006; Wenke 2009). This time was marked by civil war and periods of upheaval, but Egypt largely remained one unified state. From that period on, Egypt developed extensive trade in western Asia, and to the south in East Africa, which greatly enriched its elites. A period of environmental and political instability, which increased the power held by nomarchs, or local governors, probably put an end to the Old Kingdom phase (ca. 2650– 2181 BCE) and the first great period of Egyptian unity (J. Thompson 2009). Although Egypt was generally more politically stable, it was not immune to political fragmentation similar to that in the wider Near East.

#### 2.1.2 Middle Bronze Age (2000–1600 BCE)

While the Early Bronze Age period was characterized by limited historical data, in the Middle Bronze Age writing had spread to more areas throughout the Near East. With the dawning of this period, a repeating cycle of aggregating larger states and collapse of political entities, often leading to small city-states, emerged once again. Figure 2.2 summarizes some of the key settlements and states from this period.

In Southern Mesopotamia, this time span is often collectively called the Old Babylonian period. It began with political competition, primarily among the cities of Isin, Larsa and, to a lesser extent, Uruk. The region was dominated by Amorite elites who had migrated from within the Near East and ruled many of the cities (Charpin, Edzard and Stol 2004). The city of Eshnunna, in the Diyala region, was powerful for a brief period and was able to exert control in Northern and Southern Mesopotamia. In Southern Mesopotamia, the earlier half of the time frame is known as the Isin-Larsa period (ca. 2004–1764 BCE), which ended with the conquest by Hammurabi of Babylon (Frayne 1990). Hammurabi incorporated Southern Mesopotamia and northern areas up to Mari into a single

**Figure 2.2** Map of the region, states and cities during the Middle Bronze Age. The boundaries indicate the approximate maximum extents of the major states and empires during the period political entity; however, this proved to be short-lived, and by the reign of his son, Samsu-iluna, the empire had begun to break up. Nevertheless, Babylon lasted as an important political power until the sack of the city in 1595 BCE (Frayne 1990; Van De Mieroop 2016). While the Babylonian Empire was short-lived, the rise of Babylon proved to be a long-term trend, as this city dominated Southern Mesopotamian politics in the following centuries and into the next millennium.

To the north, Assyria is known to us through historical documents which indicate the importance of the city of Ashur (*Aššur* or Assur**)**. The kings of Assyria belonged to an Akkadian-speaking population that resisted the Amorite incursions that occurred in the late third and early second millennia BCE. Mari was another important small state centred on the Euphrates, and one of the old cities from the third millennium BCE to have continued to exert political influence (Veenhof and Eidem 2008). For much of this period, the Old Assyrian trading colonies (*kārum)* connected towns and cities in Central Anatolia, including the major trading centre of Kanesh (*Kaneš*), with Ashur. Assyrian merchants conducted a seemingly mutually beneficial trade with local Anatolian populations; this trade network extended across the Eastern Mediterranean and elsewhere (Barjamovic 2011). There is now evidence that Assyrian merchants sometimes lived and had families in Anatolian cities, which indicates movement between Anatolia and Assyria.

By 1808 BCE, the Amorite king Shamshi-Adad had conquered Ashur and incorporated this city within his kingdom. He also incorporated the small Amorite and Hurrian kingdoms of the Khabur region and Mari into his kingdom (Charpin and Ziegler 2003; Eidem 2012). Rival kingdoms in this period also appeared to the west, such as Aleppo (Yamhad; Klengel 1997). To the east, the Hurrians played an important role, having established a small kingdom around Arrapha, which was also incorporated within Shamshi-Adad's kingdom and is located under modern-day Kirkuk (Grayson 1987: 64). These areas controlled important routes to Iran and the central Zagros, Nuzi being one of the chief towns in this small kingdom.

The period witnessed thriving trade activity between Central Anatolia and Northern Mesopotamia, this trade network connecting Southern Mesopotamia, Central Asia, the Levant, the Eastern Mediterranean and the Indus (Larsen 1987; Rahmstorf 2006). Anatolia was not alone in benefiting from foreign trade; other cities in the Levantine region, such as Hazor (Ilan 1998), also benefited. While this trade transported Central Asian products, specifically tin and semi-precious stones, to other regions, trade was not direct for much of this distance; there were numerous intermediaries. During the period, Indo-European-speaking groups, including the Hittite (or Nesite), Luwian and Palaic populations became established in Anatolia, these groups being either newly arrived or, possibly, present in the region by the third millennium BCE. The Hittites conducted numerous campaigns against small kingdoms in Anatolia and Syria, where they reached Babylon in 1595 BCE under Mursili I (Bryce 1999: 103), although this was not so much a conquest as a raid, as they quickly withdrew.

Elamites, meanwhile, had been deported from Southern Mesopotamia with the rise of Isin by the beginning of the twentieth century BCE. However, after this period, Elam began once again to rival the Southern Mesopotamian cities, under the Epartid Dynasty (ca. 1950–1600 BCE) or Sukkalmahs, for power not just in Southern Mesopotamia but also in the Persian Gulf. Areas along modern Bahrain (Dilmun) and Oman (Magan), which had been of particular interest to Mesopotamia and Elam probably from the fourth millennium BCE, were important for copper resources, stones and access to trade to the east. Additionally, Elam controlled access to Central Asia and Afghanistan, which had much-sought-after tin and lapis lazuli resources that were vital to trade (Potts 1999; De Graef 2012).

Similarly to much of the third millennium BCE, Egypt after the First Intermediate Period was once again a strong centralized state during the late Eleventh Dynasty (ca. 2061–1991 BCE). While some of the dates of events are in dispute, what is clear is that Egypt expanded to the south under the Twelfth-Dynasty pharaohs (ca. 1991–1803 BCE), and reached its apex of power, with areas between the First and Second Cataracts under its control. Egypt controlled important transit points to the Sinai, possibly regions in the Levant, and quarries in the East Desert (Willems 2010; Wilkinson 2011; Van De Mieroop 2011). By the Thirteenth (ca. 1803–1649 BCE) and Fourteenth (ca. 1725–1650 BCE) Dynasties, Egypt was much weaker and had fragmented. This fragmentation continued with the arrival of the so-called Hyksos, or foreign populations, which probably consisted of Semites from the Levant. This period of political fragmentation is generally known as the Second Intermediate Period (ca. 1782–1550 BCE or 1650–1550 BCE; Ryholt 1997; Booth 2005).

The historical data, in a similar manner to that which is apparent for the Early Bronze Age, indicate a mostly politically fragmented Near East, which saw the establishment of political dynasties that formed, for a time, larger states (e.g., Hammurabi, Shamshi-Adad). However, all these dynasties quickly faded or were reduced in power. Power appeared to depend on the strong leadership of individual rulers. Furthermore, the territorial extents of the larger states that formed were not replicated by the dynasties or powers that replaced them. In other words, political boundaries did not endure. Political fragmentation and small states were the norm during the Middle Bronze Age (Charpin and Ziegler 2003; Veenhof and Eidem 2008; Barjamovic 2011).

#### 2.1.3 Late Bronze Age and Early Iron Age (1600–1050 BCE)

In the Late Bronze Age (Figure 2.3), larger states developed and became more stable, to some extent, across the Near East. These included large territorial states such as New Kingdom Egypt, Kassite Babylonia, Mitanni, Elam, the Hittites and later Assyria. Perhaps as a result of greater stability, trade and diplomatic interactions flourished. The Levantine region was an area of active political, and probably economic, competition between the Mitanni, the Hittites, Egypt and Assyria from the fourteenth to the eleventh centuries BCE (Van De Mieroop 2016).

After the fall of Babylon in 1595 BCE, the First Sealand Dynasty ruled at least parts of Mesopotamia, particularly in the southernmost regions. By about 1500 BCE, the Kassites had become more dominant and were able once again to unite Southern Mesopotamia, for a period of more than 300 years. Their power extended into the Persian Gulf, and parts of Bahrain were incorporated into the state (Magee 2014: 178). In the thirteenth century BCE, conflict with Assyria became more pronounced; there were border treaties between Assyria and the Kassites (see, e.g., Fuchs 2011: 253), but these did not prove to be long-lasting. Increasingly, Assyria became stronger than its southern neighbour (Sicker 2000: 44).

The state of Mitanni lasted roughly from the sixteenth to the fourteenth century BCE in parts of Northern Mesopotamia and southern Anatolia. While there are relatively few records from within the state, it appears that the state consisted of ruling Hurrian elites who controlled a multi-ethnic empire (Liverani 2014: 291). Regions and cities within the state had regional autonomy. One famous example is King Idrimi (fifteenth century BCE) of Alalakh, who established his state and wrote a famous inscription on a statue, now in the British Museum, that describes how he ultimately ascended the throne, giving allegiance to Mitanni (Greenstein and Marcus 1976; Collins 2008: 33). While the city, at the time, belonged in the sphere of Mitanni's power, Idrimi serves as an example of a local king relying on, or having to become a vassal of, a larger state's king, Parshatatar of Mitanni in this case. Such kings established a local power base, in which a fair degree of autonomy was achieved. This system of maintaining a larger state through local autonomy appears to have worked for Mitanni for a period; however, by the fourteenth century BCE, Assyria had fully broken away, under the reign of Ashur-uballit I (1365–1330 BCE), and was able to fully subdue Mitanni by the first half of the thirteenth century BCE, during the reign of Shalmaneser I (ca. 1274–1243 BCE).

As Assyria became more aggressive as it expanded westward, the Hittites increasingly saw it as their main threat (Bryce 1999). Assyria largely

**Figure 2.3** Key cities, regions and states of the Late Bronze Age and the eleventh century BCE. The approximate maximum extents of some of the larger states and empires are displayed stayed to the east of the Euphrates as it broke away from Mitanni rule; however, in the thirteenth century BCE, during the reigns of Shalmaneser I and Tukulti-Ninurta I (ca. 1243–1207 BCE), Assyria advanced into Anatolia and the Levant. After the fall of the Hittites, particularly during the reign of Tiglath-Pileser I (1114–1076 BCE), the Middle Assyrian Empire reached its peak: it reached the shores of the Mediterranean and deep into Anatolia, and both Cilicia and Cappadocia were subdued. For a while, the Assyrians benefited from the Sea People incursions and events during the end of the Late Bronze Age, but by the mid-eleventh century Assyria had diminished in power, although it never completely fell (Grayson 1976, 2000; Liverani 2014). Assyria's conflict with Babylon also began to shape those two regions' histories. During the reign of Tukulti-Ninurta I, the Assyrians briefly conquered Babylonia. Another important trend was the beginning of deportations of foreign populations to the Assyrian realm, particularly during the reign of Shalmaneser I (Stieglitz 1993: 269).

The Hittites, at the beginning of this period, from the sixteenth to the early fifteenth century BCE, were a weak power, particularly as Mitanni and Anatolian powers such as the Kaska limited them. With the rise of Tudhaliya I in about 1430 BCE, the Hittites expanded not only throughout much of Central Anatolia, incorporating regions found there, but also into Syria and the Levant, gaining access to the wealthy trade cities along and near the Mediterranean coast (Gurney 1990). While there were short periods of weakness after the rise of Tidhaliya I, from the reign of Suppiluliuma I in the fourteenth century BCE the Hittites began to access and control key trading cities in Syria and the Levant (Bryce 1999). The treaty signed by Ramses II and Muwatalli II after the battle of Kadesh probably consolidated the border between the Hittites and the Egyptians in the thirteenth century BCE.

By the Late Bronze Age or Middle Elamite Period (ca. 1500–1100 BCE), Elam had become unified and was one of the strongest powers in the region. Key trade still flowed in the Persian Gulf, and the Elamites and Kassites maintained peace with each other for a time, although conflict occurred periodically. Political marriages are documented that helped link the two states (Potts 2006: 119). In the twelfth century BCE, pronounced political problems between the states led to the demise of Kassite Babylonia (Arnold 2004: 75). The Elamite state was able to undertake major building projects, demonstrating its power, including the construction of new cities by, for example, Untash-Naprisha, probably in the fourteenth century BCE (Dur-Untash; Potts 1999: 230). At the end of the twelfth century BCE, Elam disappears from historical records for roughly 300 years (Van De Mieroop 2016: 189).

After emerging from the Second Intermediate Period, Egypt began to assert itself militarily in the Levant during the Eighteenth Dynasty (ca. 1550–1292 BCE); in the reign of Thutmose III (1479–1425 BCE), the Egyptians campaigned as far as northern Syria (Hoffmeier 2004: 125). This had the dual benefit of protecting Egypt from future invasions from the Levant and allowing it to control trade along the coast and Levantine land corridors. Egypt controlled the Nile as far as the Fourth Cataract in Nubia (Bard 2007: 64). This control in much of Nubia gave Egypt an enormous amount of gold and other kinds of wealth, which it used to leverage its economic and diplomatic position in the Near East, as probably demonstrated by the Amarna foreign diplomatic letters. The military and diplomatic policies of Egypt appear less direct in the later part of the Late Bronze Age in the Levant, with increased dependence on vassals and local rulers (Strange 2000: 74). Similarly to Idrimi, during the Amarna period (ca. 1353–1336 BCE) local kings who owed their allegiance to Egypt displayed a fair degree of autonomy, and even launched wars against each other.

In many respects, while the great powers competed for dominance, particularly of the lucrative trade routes that connected maritime trade along the Levantine coast and the Persian Gulf, cities along these routes thrived (Wachsmann 2009). What ultimately ended this system of trade, which was protected by the larger states, was the period attributed to the Sea Peoples. Although it is still unclear what happened or who these people were, as most of our sources derive from Egyptian texts, there appear to have been several groups or populations that invaded, or conducted incursions throughout, the coastal regions of the Eastern Mediterranean that included the Levant and Anatolia (Killebrew and Lehmann 2013). The Hittite kingdom and many cities in the Levant were destroyed or much reduced in power around 1200 BCE and later, although it is likely that invasions were not the only reason for the weakening of state power. The incursions or disruptions may have lasted for about two hundred years before and after 1200 BCE, which suggests there were several waves of invasions or political upheaval (Drews 1995). Climate change has been posited as a main contributor to the demise of the Late Bronze Age political and economic systems (Devillers, Brown and Morhange 2015). More critically for this work, it is evident that the Late Bronze Age system began to develop larger states that lasted longer than those of previous periods. However, the system was not enduring, as the disruptions associated with the Sea Peoples attest. Once again, social and political fragmentation followed after a period of larger states. States of comparable size and extent to those in the Late Bronze Age did not re-emerge until the ninth century BCE.

#### 2.1.4 The Early Iron Age (1050–800 BCE)

Historical data become more common in the tenth and ninth centuries BCE; once again the data show that the Near East had reverted to a pattern of small states and political fragmentation after a period in which there were several large states (Figure 2.4). New cultural groups migrated to or emerged from within the region, and began to shape some of the early small states.

Babylonia continued to exist as a political entity after the collapse of the Kassites, but new cultural groups vied for dominance in this power vacuum. The Second Sealand Dynasty ruled parts of the southern part of Babylonia, while the Second Isin Dynasty controlled Babylon and the northern part. The region appeared to be politically weak, while the new cultural group of Chaldeans began to play an important role in governing. Aramean groups also began to settle in parts of the region, particularly along the Tigris (Van De Mieroop 2016: 211–12).

To the east, in Elam, very little is known about the early centuries after 1200 BCE. Few inscriptions survive from this period, the first significant sources appearing in the eighth century BCE as Babylonia and Assyria increasingly came into conflict. Migrations by Persians, Medes and Mannaeans probably occurred at the beginning of the first millennium BCE or earlier (Van De Mieroop 2016: 215; Waters 2014: 21), although these are generally obscure to us since most sources about these cultures come from the Assyrian records. The Persians initially seemed to have lived in the highland regions near Anshan but near the Elamite population. The Medes lived alongside or near the Mannaeans in northern and northwest Iran.

By the start of the first millennium BCE, Assyria's territorial holdings had been pushed back to a small strip of land along the River Tigris. In the late tenth century BCE, Adad-nirari II (911–891 BCE) began to reclaim some of the lost territory of Assyria, particularly in the Khabur region of Northern Mesopotamia, the region being formally annexed by 867 BCE (Radner 2011). This marked the beginning of the Neo-Assyrian period (ca. 900–612 BCE), during which Assyria emerged as an increasingly powerful territorial empire that eventually controlled much of the ancient Near East (Cline and Graham 2011: 38). The reign of Assurnasirpal II (883–859 BCE) led to more expansion for Assyria. From his accession, Assurnasirpal II pursued a policy of establishing Assyrian political dominance and consolidating the conquests initiated by his grandfather Adad-nirari II. We now know of no fewer than 14 military campaigns during his 24 years on the throne, many of which were fought in the early part of his reign (Grayson 1982: 253). These campaigns focused on expansions to the west and east. As discussed

**Figure 2.4** Cities and small states of the Early Iron Age ca. 860 BCE with some of the population groups and regions indicated

later in this chapter, the total area over which these campaigns took place is small compared with later campaigns in the AoE. In part, this probably reflected the political fragmentation in the region, as the many political entities required separate, spatially restricted campaigns.

The state of Urartu existed in eastern Anatolia and in the regions of Lake Van. The area directly north of Assyria was already called Uruatri, an archaic form of Urartu, by the Late Bronze Age, but the region became more politically unified in the ninth century BCE. Shalmaneser III (859– 824 BCE) of Assyria is known to have campaigned in this region against the Urartian kings Arame and Sheduri. It is very likely that, in the period between the Late Bronze Age and the Early Iron Age, the Urartian state was beginning to unify local small states or entities into a larger political entity (Liverani 2014: 521). With a mountainous landscape, irrigated valleys and difficult-to-access settlements, the state of Urartu became a powerful political actor in the region, and continued to be so until the late seventh or early sixth century BCE.

The early history of Phrygia, an Indo-European-speaking kingdom in Central Anatolia with its capital at Gordion, is not clear to historians and it is quite possible that it was inhabited or even established by one or more groups referred to in the Egyptian records as the Sea Peoples. Classical sources suggest populations from Europe lived in Central Anatolia, perhaps indicating the very same people who were, at least in part, responsible for the destruction of the Hittites; some of the archaeological data may indicate this (Voigt and Henrickson 2000: 354; Robbins 2001: 173). In any case, Phrygia is mostly known from much later historical records, mainly from the Assyrian sources and Herodotus. Midas, the mythical figure in Herodotus, was probably a king called Mita in Assyrian records who ruled perhaps from the late eighth to the early seventh century BCE (Rose 2012: 217). In the west of Anatolia was Lydia, which developed out of the old region of Arzawa and around the Hermus valley. The extensive later remains of Sardis, the political capital of Lydia, prevent a full understanding of how this state and city developed in the Early Iron Age (Stafford-Deitsch 2010: 66).

The Syrian and Levantine states and political entities in the Early Iron Age show an even more politically fractured picture than regions to the east. The region was composed of Aramean, Neo-Hittite (Indo-European-speaking) and West Semitic-speaking groups, particularly those related to earlier Canaanite groups and Phoenicians. Several of these small states were relatively strong, including the Neo-Hittite state of Charchemish, which attempted to display the former power of the Hittites through its art and inscriptions. In fact, Hittite princes lived in the city during the Late Bronze Age, which suggests that the city's political links to the past may have made it more influential in the Iron Age. Other important states were Tabal, Melid (Malataya) and Quwê (Bryce 2012). The Aramean and West Semitic-speaking states were similar in that they were small and competed for local power in the Early Iron Age among themselves and with surrounding cultural groups. City-states or small states such as Bît-Agushi, Bît-Bahaiani and Bît-Adini were among the states in the region (Sader 2014).

In the Southern Levant, West Semitic-speaking cultures and polities such as Ammon, Moab, Edom, Judah, Israel and Phoenicia were found. The origin of the polities of Philistia have been debated among archaeologists, particularly as to whether the culture could be traced to some of the movements, around the twelfth century BCE, of the Sea Peoples, who possibly settled in the region, which would indicate that the culture may have derived from the Aegean or Eastern Mediterranean region (Yasur-Landau 2010). While it is not clear exactly when and how all these Southern Levantine polities formed, by the ninth century BCE they were the primary powers that ruled the region (Porter 2012: 42). Most of these cultures and states can be characterized as having derived, or probably having derived, from earlier cultures in the Late Bronze Age.

In Egypt, the state still appeared strong until the mid-twelfth century BCE during the reign of Ramses III (1186–1155 BCE), from which period inscriptions depict the defeat of Sea Peoples and Libyan incursions (Morenz and Popko 2010). However, after Ramses II's death there was a gradual decline in power in Egypt, which led it to be effectively split into two regions, centred on Tanis in the north and Thebes in the south. Eventually, Egypt became even more politically fragmented, as Libyan populations became influential in Lower Egypt. This division of power and fragmentation came to characterize Egypt until the eighth century BCE (O'Connor 2001: 233; Naunton 2010).

What is clear is that the Near East in the first few hundred years after the Late Bronze Age and Early Iron Age was composed of many small states, on a scale similar to the Middle Bronze Age and earlier periods. The cycle of political fragmentation or small states and the emergence and decline of larger territorial empires continued until the eighth century BCE.

## 2.2 The Neo-Assyrian Empire (c. 800–612 BCE)

Although the Neo-Assyrian state began to reassert itself in the late tenth century BCE, it faced renewed weakness in the late ninth century, which continued through the mid-eighth century BCE, particularly after the death of Adad-nirari III in 783. Urartu, in particular, took advantage of this and began to expand in the late ninth century BCE, while local governors within the Neo-Assyrian Empire displayed greater independence from the central government (Grayson 1982; Liverani 2014). The situation changed with the ascension of Tiglath-Pileser III (744–727 BCE), who reformed the Assyrian army and reinvigorated campaigns that saw the empire rapidly expand from this period until the end of Ashurbanipal's reign in 627 BCE (Dubovský 2004/2005; Fales 2005). Following the short reign of Shalmaneser V (726–722 BCE), which saw most of Israel incorporated into the Neo-Assyrian Empire, Sargon II (721–705 BCE) continued to expand the empire and began to construct a new capital city called Dur-Sharrukin (Radner 2003/2004; Fuchs 2009). Sargon's reign ended earlier than it might otherwise have done because of his death in battle while he was in the region of Tabal in Anatolia. During the eighth century BCE, and the reign of Sargon II, which lasted into the 630s, the Cimmerians invaded the Near East from the Caucasus, which led to the downfall of the Phrygian state and the probable sacking of numerous Urartian and other settlements (Kristensen 1988).

During the eighth and seventh centuries BCE, the rate of deportation of subject populations increased, and many families and individuals were deported for economic purposes by the Assyrians (Oded 1979; Gallagher 1994). Long-distance population movement occurred at a greater rate in historical records, although these were mainly forced migrations. The Neo-Assyrian state directly incorporated Southern Mesopotamia, western Iran and southern and Central Anatolia, including areas held by Urartu, the Neo-Hittite states, the Aramaean states and most of the Southern Levant, while some kingdoms (e.g., Judah) may have become vassals. Although the image of Assyria is generally as an oppressive state, the empire actually incorporated subject population groups into key state enterprises such as the military, by utilizing mercenaries (Dalley 1985). The use of mercenaries shows that population groups were probably spreading within the Assyrian state. At the beginning of the seventh century BCE, in the reign of Sennacherib (705–681 BCE), the Assyrians extended their state from the borders of Egypt to western Iran and from Central Anatolia to northern Arabia. Massive building projects, probably fuelled by the excess labour now coming into the empire because of deportations or economic interest, increased during the reign of Sennacherib; the present outline of the capital, Nineveh, visible on satellite imagery, is attributed to this king (Altaweel 2008: Plate 16). With Nineveh expanding to roughly 800 hectares, it was supplied with water by irrigation projects similar to those at Kalhu (Nimrud) in the ninth century BCE, but on a larger scale (Bagg 2000; Altaweel 2008). Royal roads connected various parts of the empire, including distant provinces and key provincial cities (Altaweel 2003).

In the reign of Esarhaddon (681–669 BCE), Assyria successfully expanded into Egypt, the first time that a Mesopotamia-based state had done so. Just before this time, in the late eighth century BCE, Egypt's Twenty-Fifth Dynasty, formed of Kushite rulers from Napata, had unified Egypt and expanded into Southwest Asia (Kitchen 2009). In fact, Egypt reached its greatest territorial extent since the New Kingdom Period. Despite this strength, Esarhaddon not only succeeded in campaigning into Egypt, but also strengthened the realm along its frontiers and expanded it into other areas, including north-central Iran, near modern-day Tehran, and further into Anatolia (Leichty 2011). In the reign of Ashurbanipal (668–627 BCE), the empire reached its apogee in territorial extent with the conquest of Elam and Upper Egypt (Figure 2.5). It also fought a costly war in 652–646 BCE against Babylon and other rebellious vassals (Grayson 1980). It is probable that the civil wars and unrest that occurred after the death of Ashurbanipal weakened the Neo-Assyrian state. The Babylonians, this time with allies such as the Medes, pushed the Assyrians out of Southern Mesopotamia, while the Medes invaded from Iran. They destroyed the capitals of Ashur (614 BCE), Nineveh (612 BCE) and Kalhu (612 BCE). The Assyrians attempted to hold on to power for some time after the sacking of their core cities, particularly in Harran, but they ultimately failed, and by 605 BCE the Assyrian state had disappeared from historical texts (Zawadzki 1988; Radner 2015).

One of the key groups in the decline of the Assyrians was the Medes, who had probably formed a state by the late seventh century BCE. The Median state may be an example of secondary state formation (the formation of a state as a result of the influence of another through war or interaction): incursions by the Assyrians or other groups in the seventh century gave impetus for the unification of Median tribes and groups into larger political entities (Brown 1986), possibly helping to sow the seeds of Assyria's destruction. Although there are few records to confirm this, in the later half of the seventh century BCE (Radner 2003) repeated wars waged by the Assyrians probably weakened the empire, making it ripe to fall. Despite the fact that the Neo-Assyrian Empire can be considered the largest of the states so far discussed, its hold on much of its territory was tenuous or short-term. Regions and countries such as Egypt, Elam and even Babylonia, were often rebellious. Nevertheless, rather than a

**Figure 2.5** Approximate borders of the Neo-Assyrian Empire at its height in the seventh century BCE

reversion to city-states or small states after the fall of the Neo-Assyrian Empire, still larger states soon began to form. A new political pattern had emerged.

## 2.3 Neo-Babylonians, Medes and others (626–550 BCE)

After the fall of the Neo-Assyrian Empire, the map of the Near East shows the Neo-Babylonian Empire (626–539 BCE) occupying most of the areas once held by the Neo-Assyrians. Initially, some of the old citystates, particularly along the Levant, declared independence or tried to become independent, but most were quickly conquered or submitted to Babylonian rule (Fitzpatrick-McKinley 2015: 42). Western Anatolia was dominated by Cilicia, Caria, Lycia and Lydia, while the Median state that had grown in the seventh century began to span the eastern half of Anatolia, occupying much of Iran and regions to the east as well (Bryce 2009). Egypt, under the Twenty-Sixth Dynasty or Saite pharaohs, was once again unified and able to mount expansionist campaigns in the Near East (Lloyd 2001; Figure 2.6).

Conflicts in the late seventh and early sixth centuries BCE centred on Egyptian and Babylonian contests for supremacy in the Levant. While the Egyptians did not succeed in establishing a base in the Levant (which they had done in the Late Bronze Age), the Babylonians' attempts to incorporate Egypt into their empire were equally unsuccessful (Schipper 2011: 285). The city of Babylon, during this period and in the reign of Nebuchadnezzar II (604–562 BCE), became the largest city in the world and several large-scale building projects were undertaken, such as the reconstruction of Marduk's temple and the Ishtar Gate (Seymour 2014). To the east, the Medes, by the period of Cyaxares (Uvaxshtra in Akkadian sources; 625–585 BCE), had consolidated their control of Iran, including the Elamite and Persian populations, although the Elamites would continue to influence Iranian culture long after they lost their political power (Potts 1999: 4).

Lydia was able to unify much of western Anatolia; the kingdom reached its greatest extent in the reign of Croesus (ca. 560–547 BCE). Coinage may already have begun to spread in Anatolia by the seventh to sixth centuries BCE (Horesh and Kim 2011: 287). By 585 BCE, the border between Lydia and Media to the east was fixed, perhaps along the River Halys, as stated by Herodotus (Wood 1972: 27). By the mid-sixth century BCE, the Near East was dominated by four, mostly large, states. This was similar to the situation in the Late Bronze Age but the states were now larger and the collapse of one state did not lead to a new pattern of citystates or small states.

## 2.4 The Achaemenid Empire (559–330 BCE)

While the accuracy of the historical sources is not entirely certain, Cyrus II (559–530 BCE) established what would become the Achaemenid Empire, which was able to unite the large Median Empire with that of the now independent Persian state by 549 BCE. Cyrus proceeded to conquer the Lydian state (ca. 540s BCE) and then the Babylonian Empire (539 BCE; Waters 2014: 41). In his conquest of Babylon, Cyrus portrayed himself as a legitimized Babylonian king, relieving the population from the oppressive Babylonian Nabonidus (556–539 BCE), rather than as a foreign conqueror, through the text on the so-called Cyrus Cylinder, in which Marduk, the god of Babylon, justifies his actions. At some point, either before or after the conquest of Babylon, Cyrus began to incorporate eastern Iran and Central Asia into his realm. After the conquest of Babylon, it is likely that the intent was to continue into Egypt, unifying the Near East for the first time, but this had to wait, as Cyrus died in 530 BCE during a campaign against the Massagetae in Central Asia (Briant 2002: 49). Although all the battles and conquests of Cyrus are not fully known, what we do know is that he created the largest empire and conquered the most territory of any ruler up to that point, doing so at a relatively fast rate. The Achaemenids unified a large, diverse population across their vast realm. Cyrus adopted the title 'King of Kings', an old Mesopotamian title that was also used by later imperial peoples, such as the Sasanians who emulated the Achaemenids (Dandamaev 1989: 55). The title 'King of Kings' reflects the Persian idea of a high king having dominion over subsidiary or vassal kings of regions within the empire. Rather than being portrayed as oppressed by the great king, the vassal kings and populations were used to reflect glory onto the realm by showing its diversity. Perhaps for the first time, a political philosophy began to appear that took pride in the ethnically and socially diverse nature of the empire.

Even though the empire that Cyrus left was already enormous, expansion continued in the reign of Cambyses II (530–522 BCE) with the conquest of Egypt. The expansion incorporated Libya, but attempts to conquer Carthage and Kush failed. After the death of Cambyses, and Darius I's (522–486 BCE) eventual accession to the throne, the first task of the empire was to put down revolts in several provinces, including Babylonia, Elam and Media. Successful campaigns were also launched in Central Asia and along the Indus. Additionally, for the first time, a Near Eastern empire had expanded into Europe, conquering large parts of Thrace and southeast Europe. After subduing the Ionian cities in 493 BCE, Darius was able to focus on Greece. But this initial attempt failed after the Battle of Marathon in 490 BCE (Briant 2002).

Although Darius I is known for failing to defeat Athens and conquer Greece, he is also known to have undertaken major administrative and economic reforms, while practising a religious tolerance that helped to internationalize the empire even more. In Egypt, he was depicted as Pharaoh, as Cambyses was, showing attempts to justify his rule to a local region's governing culture (Briant 2002). He further developed the satrapies, following Cyrus' example. With the exception of Persis, satrapies were now responsible for providing taxes to the central government. Regular checks were made on satraps to avoid any one of them gaining too much power. Important advances occurred in the economic sphere. Darius introduced the *daric* as a single currency for the empire. The royal highway system, similar to and building on the Neo-Assyrian royal roads, was implemented, although this clearly provided an economic benefit by facilitating long-distance movement and making it more direct. An important canal linking the Nile with the Red Sea was built, further aiding trade. *Qanats*, or underground channels, were built to stimulate agriculture (Poolos 2008). Additionally, large-scale private enterprise, in the form of investment firms or banks, had developed in multiple cities. The one that is best known to us is that of the Murashu family in Babylonia during the fifth century BCE, although earlier Babylonian families had developed similar firms (Stolper 1985; Kuhrt 2007: 12).

It is likely that, by the time of Darius I, parts of the Eastern Mediterranean coastal regions had begun to develop greater population concentrations, trade having been a likely motivation (Mazzoni 1991–2). In contrast, we know far less about the interior regions of the Near East during this and subsequent periods, which reflects a settlement decline or at least an abandonment of major cities in eastern Syria, Northern Mesopotamia and other areas. Finally, Darius was the first of the Achaemenid rulers to create a tomb at Naqsh-i Rustam; this site would become important not just for the Achaemenids but also for later Sasanian rulers who emulated them (Davies 1932).

At the time of the death of Darius I, the expansion phase of the Achaemenid Empire had reached its peak (Figure 2.7). The next king, Xerxes I (486–465 BCE), is known for his attempt to conquer Greece, which ultimately failed at the decisive Battles of Salamis and Mycale, although he briefly took Athens. After his failure to conquer Greece, Xerxes appeared to be content with completing major construction projects at Susa and Persepolis, which symbolized and incorporated the diverse cultural influences in architecture and populations of the Achaemenid Empire (Briant 2002).

In the reign of Artaxerxes I (465–424 BCE), there were rebellions in the vast Achaemenid Empire, particularly in Egypt, where the Athenians

were active in supporting the rebellion. However, these rebellions did not prove to be effective. In general, the last few decades of the fifth century BCE were less stable for the Achaemenid kings, as shorter-reigning kings ruled and more threatening revolts emerged. The two most famous rebellions are that which led to the loss of Egypt in 404 BCE, and the revolt of Cyrus the Younger (401 BCE), the brother of Artaxerxes II (404–358 BCE). The general perception has been that Achaemenid power began to decline in the late fifth and early fourth centuries BCE. In Egypt, the Achaemenids had been seen as oppressive, but this could be later propaganda. Nevertheless, the Achaemenid Empire was still a strong power and was able to put down rebellions and launch major campaigns (Kuhrt 2007: 347; Waters 2014). Artaxerxes III (358–338 BCE) was, in fact, able to reconquer Egypt, although he initially failed in this quest.

The next two kings of the Achaemenid state were Artaxerxes IV (Arses; ca. 338–336 BCE) and Darius III (336–330 BCE), the exact reign and dates for the former king being less certain. The Achaemenid throne may have been contested at this time; Artaxerxes IV's hold on power was probably tenuous, and he was killed after a brief reign, possibly by his vizier Bagoas. Certainly the key event at this time was the invasion of the Achaemenid Empire by Alexander of Macedonia (336–323 BCE). By 336, during the reign of Alexander's father Philip II (359–336 BCE), a bridgehead had been established in Anatolia by the Macedonians in preparation for a major invasion. In 334 BCE, Alexander began his campaign and to expand on his father's gains, having first put down revolts. Historically, the advance of Alexander is seen as swift. Surprisingly, relatively few major battles, perhaps only five, and several sieges were fought to conquer the Achaemenid Empire (Briant 2002; Heckel and Yardley 2004; Kuhrt 2007: 419–21). This is in stark contrast to earlier conquests by the Assyrians in the ninth century BCE, when 14 campaigns are recorded in the reign of Assurnasirpal II over a much smaller territorial area covering parts of Syria, Anatolia and Mesopotamia (Grayson 1982: 253).

## 2.5 The Seleucid Empire and its contemporaries (312–64 BCE)

Alexander's achievement in unifying the Near East, Egypt and Greece under the same empire opened a new phase in the history of this region, characterized by the spread of Greek material culture, language and populations across the Near East, Central Asia and India. New cities, founded in a Greek grid pattern like that of Alexandria, began to appear. This phenomenon is usually called Hellenization, a variegated social process that displayed persistence of local cultures, and resulted eventually in hybridization between Greek and other cultures (P. Green 2007). This, as will be demonstrated, was a period of increased syncretism between Greek and Near Eastern cultures, demonstrated not just in art, but also in religion, urbanism and other social manifestations. Alexander's premature death in Babylon, in 323 BCE, plunged his newly created empire into a series of wars fought among Alexander's commanders as they contended for supremacy (Waterfield 2011). Despite these wars, few states succeeded the downfall of Alexander's realm. Among the feuding commanders was Seleucus I (called 'Nicator'), a Macedonian officer who had accompanied Alexander during his military campaigns, and who eventually prevailed. In 312 BCE, Seleucus gathered his troops in Harran and marched towards Babylon, entering the city in triumph in 311 BCE, where he was welcomed by the local population (Grayson 2000; Grainger 2014: 41–54). At the end of the same year, Seleucus conquered Ecbatana, capital of Media, and Susa, capital of Susiana, thus becoming the ruler of Mesopotamia and west Iran (Grayson 2000; Diodorus 1954: book 19.92.5). While another of Alexander's generals, Antigonus the One-Eyed, occupied Syria and Anatolia, Seleucus, in 308 BCE, set about extending his empire into east Iran and Central Asia. Seleucus subdued Sogdiana and Bactria, crossed the River Indus, and in 305 BCE sealed a peace treaty with King Chandragupta of the Indian royal dynasty Maurayas (Appianus 1999: book 11.55; Grainger 2014: 54–69). After his eastern campaigns, Seleucus headed westwards to fight Antigonus, who was in Phrygia (Central Anatolia). In 301 BCE, with Ptolemy I Soter (another of Alexander's generals, who founded the Ptolemaic Dynasty in Egypt) as an ally, Seleucus defeated Antigonus (at the Battle of Ipsus), gaining control over a territory stretching from Phrygia and Syria to the Indus (Diodorus 1954: book 20.107–13; Grainger 2014: 77). Lebanon and Palestine were then added to Seleucus' possessions but soon after they were ceded to Ptolemy of Egypt.

When in 281 BCE Seleucus defeated Lysimachus, who ruled Lydia, Seleucus became the sovereign of virtually all the Near East; his empire included Anatolia, Syria, Armenia, Mesopotamia, Iran, Bactria, Sogdiana and the territories up to the Indus (Appianus 1999: book 11.55; Figure 2.8). As a supreme lord of the Near East with Greek origins, Seleucus could realize Alexander's dream of unifying the Greek and Near Eastern cultures. One of Seleucus' achievements was the foundation throughout his empire of many cities, which became the major vehicle for the penetration of Greek cultural elements and population into the Near East (Grainger 1990). In 305 BCE, soon after his eastern campaigns, Seleucus founded a new capital city

in Mesopotamia, called Seleucia on the Tigris, whose ruins today lie not far from modern Baghdad (Invernizzi 1976). Being located at the crossroads of trade routes connecting Iran with Anatolia and the Mediterranean, Seleucia on the Tigris was intended to be a Greek city, but its population was a mixture of varied ethnic groups – Syrians, Babylonians, Greeks and Jews. Large cities throughout the Near East now commonly had very diverse ethnic groups. Seleucia became one of the major cities of the Near East and one of the largest metropolises; historical records suggest it reached about 600,000 inhabitants in the first century CE (Pliny 2006: book 6.122). Around 300 BCE, Seleucus founded another Seleucia, called Seleucia in Pieria, located near the River Orontes in northern Syria; soon after, he founded Antioch (modern Antakya), also located by the Orontes. Antioch was to become another major city of the empire, populated by a diverse population that included Syrian, Aramaean, Greek and Jewish settlers from another city, Antigoneia, north of Antioch (Diodorus 1954: book 20.47.5– 6). Several other cities were founded around the same time: Dura Europos on the Euphrates in eastern Syria, Apamea on the Orontes, and Laodicea (modern Latakya) on the Syrian coast; these too became important trade hubs (Grainger 1990). We therefore see an increased trend of the creation of important trade locations centred on towns along major rivers and coastal Mediterranean regions (see Chapter 6).

Seleucus' attempt at conquering Thrace and Macedonia ended with his death in 281 BCE, and the burden of preserving the vast empire he had created fell on Antiochus I, his half-Iranian son (Bryce 2014: 170–1). After many campaigns against his adversaries, Antiochus eventually defeated the Galatians in Anatolia (275 BCE) and signed a treaty with Ptolemy of Egypt in 270 BCE, reaffirming his control over the territories conquered by his father, whereas Ptolemy maintained his authority over Egypt, Lebanon and Palestine (Bryce 2014: 172). New cities were founded, such as Ai Khanum on the River Oxus (within modern Afghanistan) which had a mixture of local and Greek cultures (Martinez-Sève 2014). Sardis, in Lydia, became the third capital of the empire, along with Seleucia on the Tigris and Antioch. Antiochus promoted building activities in Borsippa and Babylonia (Oelsner 2002: 187), showing a tolerant attitude towards the long-lasting traditions of these cities.

After Antiochus' death in 261 BCE, the Seleucid Dynasty was hit by internal divisions, and at the same time it had to face the expansionist goals of Ptolemy of Egypt, whose aim was to conquer Syria, and of Eumenes I king of Pergamum, who was carving out his own kingdom in western Anatolia and proclaiming independence from the Seleucids (Bryce 2014: 173–8). Despite the dynastic squabbles, the Seleucids maintained most of their territory for about 40 years, but lost the eastern provinces of Parthia and Bactria as well as the area north of Taurus, which was ceded to Attalus king of Pergamum in 228 BCE (Polybius 2010: book 4.48).

The greatness of the Seleucid Empire was restored, although briefly, by Antiochus III (222–187 BCE), who subdued Bactria (Battle of Arius in 208 BCE; Polybius 2011: book 10.49) and Parthia, and then crossed the Indus, like his predecessor Seleucus I, forging an alliance with the Indian king Sophagasenus in 205 BCE (Polybius 2011: book 11.34; Grainger 2014: 186–200). The former limits of the Seleucid Empire were now completely restored, although after Antiochus left eastern Iran, local kingdoms (Parthia and Bactria in particular) quickly reaffirmed their independence from the Seleucids (Sherwin-White and Kuhrt 1993: 200).

In 200 BCE, Antiochus III won the Battle of Panium, located in the Golan Heights, against Ptolemy's troops, thus obtaining the latter's territories outside Egypt, that is, Lebanon and Palestine, as far as Gaza (Polybius 2012: book 16.18). Antiochus III was now the master of all the Near East, outside of Egypt, which earned him the title of 'the Great'. However, his expansionist goals were to clash with the new power that had arisen in the Mediterranean: Rome. Much like his predecessor Seleucus I, Antiochus III launched an attack against mainland Greece, whose cities were under Rome's protection, thus declaring war on the latter. Eventually, Antiochus' troops were defeated by the Romans at the Battle of Magnesia (southwest Anatolia) in 189 BCE; the Seleucid king was forced to give up his possessions in Anatolia and pay an indemnity (Gruen 1984: 640–3).

The Battle of Magnesia represented a watershed in the history of the Seleucid Empire, as it marked the increasing involvement of Rome in the political affairs of the Seleucids, with the aim of limiting their expansionist goals. After Antiochus III's death in 187 BCE, civil wars divided the Seleucid Dynasty (Gruen 1984: 667); as a result the territories in Iran and Mesopotamia went to the Parthians (between 148 and 138 BCE), while Ptolemy VI of Egypt established control over Syria and Palestine, though only for a short period (Bryce 2014: 209–10). Antiochus VII (139–129 BCE) was the last king to attempt to restore the Seleucid Empire's grandeur, but after his death – and until 64 BCE – the Seleucids' territory was reduced to northern Syria, around the city of Antioch, while the eastern territories were in the hands of the Parthians, and Palestine was under the Judean kings of the Hasmonean Dynasty (Figure 2.9; Bryce 2014: 214–17, 222). In the meantime, the Ptolemaic kingdom progressively lost its territories in the Levant as well as Cyrenaica. The years 64 and 63 BCE finally saw Syria and Palestine becoming Roman provinces, and the kingdom of the Nabateans, an Arab population devoted to trade, extended from southern Jordan (Petra) to Damascus across the Transjordan, becoming a vassal kingdom of the Romans (Millar 1994: 27–43).

The other major Hellenistic state in the Mediterranean was the Ptolemaic kingdom, founded by Ptolemy I Soter in 305 BCE, which lasted until the Roman annexation of Egypt in 30 BCE, after Cleopatra VII's death (Lloyd 2000). The Ptolemaic kingdom extended across Egypt and Cyrenaica (northeast Libya) as well as Cyprus; it also included the coasts of southwest Anatolia and the Southern Levant (Cisjordan) as far as Tyre, although these territories were constantly threatened by the Seleucids, with whom they clashed in several battles, as previously mentioned. The Ptolemaic kingdom was relatively stable, and its dynasty was the longest Egypt had ever had. Similarly to the Seleucids in the Near East, the Ptolemaic kings promoted the diffusion of Greek culture in Egypt, favouring its blending with the local long-lasting culture so as to encourage the emergence of a hybrid Greco-Egyptian style, visible in their royal iconography (see Chapter 7). The Ptolemies also favoured a new syncretic religion centred on the figure of the god Serapis, who blended Egyptian and Greek deities (see Chapter 10). The main city, Alexandria, became one of the most important trade hubs of the Mediterranean, and perhaps the most influent cultural centre of its time (see Chapter 5). Here, Ptolemy I founded the famous library of Alexandria, along with the *Musaeum*, a literary and scientific research centre where some of the most important scientific achievements of the ancient world were attained, for example in astronomy (Manning 2013).

## 2.6 The Parthians and the Romans

During the third century BCE, two eastern provinces of the Seleucid Empire, namely Parthia and Bactria, became independent following the rebellions of their respective satraps, Andragoras and Diodotus. These events are usually thought to have happened during the first years of Seleucus II Callinicus' reign (246–225 BCE), although some scholars prefer a more remote date (Wolski 1993: 47–50). While Bactria became an independent kingdom under the rule of Diodotus and his dynasty, Parthia was occupied in 238 BCE, soon after Andragoras' rebellion, by a semi-nomadic population from Central Asia known as the Parni (Strabo 2001: book 9.7.1). After entering Parthia, the Parni took the language spoken in that area as well as the name of Parthians (Debevoise 1938: 1– 2). Their leader was Arsaces, who rose by 247 BCE, the year the Parthian period began (V. S. Curtis 2007). Under the king Phraate I (176–171 BCE), the Parthians expanded as far as the region south of the Caspian Sea inhabited by the Mardian tribes, who were subsequently deported and forcibly settled in Charax near the Caspian gates (Debevoise 1938: 19). In doing so, the Parthians followed the practices of the Assyrians and Achaemenids by displacing conquered populations (Wolski 1993: 74).

Phraate I's successor, Mithradates I (171–138 BCE), was to become the great conqueror who made the Parthians the masters of Iran and Mesopotamia. The details of the Parthian expansion in Iran are not well known; however, some dates can be deduced. First, the Parthian king conquered the regions west of the River Hari which were under the Bactrian Empire (Strabo 2000: book 11.11.20). In 148 BCE, the Parthians took Media by conquering its capital, Ecbatana, where the local satrap had rebelled against Seleucid rule (Wolski 1993: 79). In 141 BCE, Mithradates conquered Babylon and Seleucia, where he was crowned with the now ancient title of King of Kings, following the Achaemenid tradition (Wolski 1993: 81). On the left bank of the river he founded another capital, Ctesiphon, near Seleucia (Invernizzi 1976). Mithradates' last campaigns were against the Seleucid king Demetrius II, who intended to claim back the lost territories but was defeated by the Parthian king in Hyrcania, and against Susa and Elymais, whom Mithradates subdued in 138 BCE (Wolski 1993: 81–3). These events are recorded in a relief at Hung-I Nauruzi. In the same year, Mithradates died, leaving an empire that extended across Parthia, Hyrcania, Media, Babylonia, Assyria, Elymais and, perhaps, Persis, which were unified within a ten year period (Debevoise 1938: 27). Once again, a large empire developed quite quickly after the weakening of another.

After Mithradates' death, his successors Phraate II (138–129 BCE) and Artabanus I (129–124 BCE) struggled to maintain the empire (Wolski 1993: 83–8). In the meantime, the Bactrian kingdom, extending over roughly the area of present-day north Afghanistan, weakened because of several nomadic invasions (Strabo 2000: book 11.8.1). The Bactrian kings had diplomatic and trade relations with China, and promoted the spread of Greek culture in Central Asia and its blending with local traditions through the foundation of cities (e.g., Ai-Khanum and Bactra), coinage and figurative art (Bernard 1994). The imprint of Greek art in this area remained even after the collapse of the Bactrian kingdom in 125 BCE. The territory was then settled by the population of the Yuezhi, who adopted the Greek alphabet and Greek-style iconography in their coinage; in about 30 BCE, the Yuezhi founded the Kushan Empire, extending across Bactria, the Hindu-Kush and northwest India (Puri 1994).

The political situation of the Parthian Empire changed when the new king Mithradates II, son of Artabanus I, rose to the throne (124–87 BCE) and became one of the greatest sovereigns of the Arsacid Dynasty. In 122 BCE, Mithradates II subdued Babylonia, which had rebelled against Parthian rule; he conquered the fortress of Dura Europos on the Euphrates in 113 BCE, soon after he took control of the regions of Adiabene and Osrohene in Northern Mesopotamia and transformed them into vassal kingdoms (Wolski 1993: 89; Figure 2.9). Mithradates II restored the Parthian Empire to its glory, stretching now across Iran and Mesopotamia.

Under this empire, the fusion between Iranian (Achaemenid) and Greek traditions became more visible. Following the Achaemenid custom, Parthian kings favoured the use of Aramaic as a lingua franca, while also favouring the Iranian language Phalavi and maintaining the use of the cuneiform script along with Elamite (Wolski 1993: 98–9). However, because part of the population they controlled was of Greek origin, the Parthian kings from Mithradates I onwards adopted the title of Philoellenos ('friends of the Greeks') on their coins (V. S. Curtis 2007), evidently as a propagandistic act for their diverse population.

As we have seen under the Seleucid Empire, elements of Greek culture (figurative art in particular) spread across Mesopotamia and Iran. This phenomenon continued under the Parthians despite their non-Greek origins; the new capital city of the Parthians, Parthian Nisa, probably founded by Mithradates II (today in Turkmenistan), showed a mixture of Iranian and Greek influences in its material culture (Invernizzi 2004, 2007; see also Chapter 7). At the same time, there was an emergence of new material cultural styles in architecture and crafts. New figurative styles were evident in cities such as Dura Europos, Hatra, Assur and Uruk, defining what is known as the Parthian style (Colledge 1977). The establishment of the Parthian Empire not only favoured the spread of these new cultural stimuli but also facilitated trade contacts with distant cultures.

Mithradates II is known to have established political and diplomatic relations with a Chinese emperor, to whom the Parthian king sent an ambassador (Wolski 1993: 94–5). This act paved the way for the establishment of long-distance trade contacts with China and laid the basis for the Silk Road, along which silk and other goods were traded from China to the Mediterranean, crossing Parthian lands. A maritime route was also opened through the Indian Ocean (Debevoise 1938: 43–4), perhaps facilitated by the extension of Parthian control over the western shores of the Persian Gulf, though this extension is suggested only by some archaeological remains (see Grajetzki 2011: 85–91). Another of Mithradates II's important achievements was the conquest around 100 BCE of Armenia, in eastern Anatolia, where the Parthians installed the Armenian prince Tigranes as a vassal king (Wolski 1993: 91). This move,

**Figure 2.9** The Parthian Empire and major states ca. 100 BCE

however, aroused Rome's concerns about the expansion of the Parthians, especially because the Romans had become an active political force in Anatolia and increasingly had interests in the Near East. In the aftermath of Mithradates' death, the Parthians maintained their territories, though they clashed on several occasions with the Romans over the control of Armenia (Wolski 1993: 122–8). As mentioned before, in 64 BCE Syria and Palestine became Roman provinces; hence the Euphrates became the natural border between the Roman and Parthian Empires.

During the first century CE, the Romans consolidated their control over Egypt and the entire Levant, from Anatolia down to the territories of the Nabateans, in south Jordan (Millar 1994); they favoured the construction of roads to make communications easier, and they promoted grandiose architectural programmes in many cities, where theatres, baths and other Romanstyle monuments were erected (Sartre 2007). Egypt and the Levant under Rome underwent a period of economic growth witnessed by the intensification of international trade networks connecting the Mediterranean with India (see Chapter 6). The security granted by the Roman Empire certainly stimulated the economic growth of the Levant and Egypt at this time, despite frequent military confrontations with the Parthians (Debevoise 1938).

The Parthians maintained their control over Mesopotamia and Iran, Armenia being contested with the Romans (Debevoise 1938: 121–212). East of the Parthian Empire, the Kushan Empire arose in 30 CE and lasted until about the fourth century. This empire was characterized by the blend of Greek style and Indian and Buddhist traditions, visible in the art of Gandhara (see Chapter 7). The Kushans were actively involved in international trade, maintaining contacts with the Parthians and the Romans to the west as well as with the Chinese Han dynasty to the east (Puri 1994). During these years, trade relations between the Parthians and China were also maintained; Chinese written sources dated to 97 CE mention the Parthian king Pacorus sending lions and ostriches from Charax (in the Persian Gulf) to China (Debevoise 1938: 216–17).

This political balance between the Romans and the Parthians remained quite stable until the Roman emperors of the second century CE started an aggressive policy against the Parthians. The Roman emperor Trajan sailed towards Antioch, where he arrived in 114 (Debevoise 1938: 219). From there, he moved towards Armenia, which he transformed into a province (Debevoise 1938: 223). In 115, Trajan moved south towards Adiabene and Osrohene in Upper Mesopotamia, which passed to the Romans (Debevoise 1938: 226).

In 116 Trajan conquered Dura Europos and moved against Ctesiphon. Between Dura Europos and Ctesiphon the Roman emperor encountered no major cities offering resistance. Ctesiphon, the capital city of the Parthians, fell into the hands of the Romans. After the conquest of Ctesiphon, Trajan sailed south and occupied the cities of Southern Mesopotamia, Akra, Oratha and Apamea, thus receiving tribute from the king of Characene, formerly a vassal of the Parthians (Debevoise 1938). On his way back, in winter 116, Trajan entered Babylon (Wolski 1993: 180).

Trajan's efforts brought Mesopotamia within the borders of the Roman Empire, but after his death in 117 the new emperor Hadrian withdrew Roman troops from Mesopotamia; thus, the Euphrates was restored as the border between the Romans and Parthians, and Armenia was once again under Parthian control (Wolski 1993: 182–3). In 197, Septimius Severus crossed the Euphrates and conquered Nisibis, and with it Adiabene. He went on to conquer Seleucia, Ctesiphon and Babylon in 198. On their way back, the Roman troops laid siege to Hatra, but without success (Debevoise 1938).

After Septimius' campaigns, the collapse of the Parthian Empire was inevitable. In 208, Vologases VI became the new Parthian king, but his brother Artabanus V (216–224) rebelled against him and conquered Iran and Media while Northern Mesopotamia was still under Roman control (Wolski 1993: 191–2). In 217, Artabanus V fought the Romans at Nisibis, but he was eventually defeated in 224 by Ardashir I, who belonged to the Sasanian Dynasty originating in Persis, southwest Iran. With the death of Artabanus V, the Parthian Empire came to an end, and a new dynasty arose.

## 2.7 The Sasanian Empire and its contemporaries

The origins of Ardashir I (224–242) and his family are not clear, because the sources offer different and contradicting versions.1 According to most scholars, Ardashir I was the son of the Anahit priest Papak, who had dethroned the local ruler in Persia in 205–6 and begun to strike coins portraying himself as king. At the death of Papak and his elder son Shabur, power passed to Ardashir I (Daryaee 2010: 243–4), who defeated the Parthian king Artabanus V in 224 on the plain of Hormozgan, taking control of Mesopotamia (as far as the Tigris), the Iranian plateau, and the eastern side of the Persian Gulf (Herodian 1970: book 6.2.2). In the same year, he was crowned at Ctesiphon, which became the capital of the empire, as a King of Kings, thus resuming the Achaemenid kings' title (Daryaee 2010: 252). Soon after these events, Ardashir confronted the Romans, who controlled the Levant and North Mesopotamia (Herodian 1970: book 6.5); on the death of the emperor Alexander Severus in 235, Ardashir succeeded in annexing all of Mesopotamia by seizing Dura Europos, Carrahae, Nisibis and Hatra (Kettenhofen 1982). In the east, Ardashir I expanded his empire by conquering Khorasan, Margiana and Chorasmia (Frye 1993).

Ardashir glorified his achievements by means of several rock reliefs, one of which, at Naqsh-i Rustam, shows the Sasanian king on his horse, which is stepping over the body of Artabanus V; the god Ahura Mazda gives Ardashir the symbols of power (Herrmann and Curtis 2002). This relief, and the inscription that accompanies it, are of particular importance: they show that Ardashir considered himself a 'Mazda worshipper' and 'descendent from the gods' (Wiesehöfer 1986), which denotes the devotion of the Sasanians to Zoroastrianism, Ahura Mazda being the main deity of this religion. Moreover, the core territory over which Ardashir governs is called in the inscription *Iranshahr*, that is, the 'realm of the Iranians', and the people are named *Eran*, that is, Iranian, following a tradition attested in the Avesta, the sacred book of Zoroastrianism (Wiesehöfer 1996: 165–71; Daryaee 2013: 5). Another important aspect of Ardashir's ideology is his reference to the Achaemenid legacy. Not only did he resume the Achaemenid title of King of Kings, but also by choosing Naqsh-i Rustam for his reliefs he clearly connected himself to the Achaemenid past, as this site had been chosen by the Achaemenid kings for their monumental tombs. Finally, Ardashir's name itself reminds us of the name of the Achaemenid king Artaxerxes, the form 'Ardashir' being a later version of it (Daryaee 2013: 2).

It seems, therefore, that at first the Sasanians tried to wipe out the Parthian past by connecting themselves to the Achaemenid rulers; however, many aspects of material culture continued from the Parthian period well into the Sasanian era, demonstrating a strong element of continuity (J. Curtis 2000). At the administrative level, Ardashir and the later rulers assigned to the easternmost regions (e.g., Margiana) the status of semiindependent kingdoms governed by kings loyal to the Sasanian emperor (Wiesehöfer 1996: 183–91), similarly to the way in which the Parthian Empire governed.

Ardashir I's son, Shapur I (242–270), who became coregent in 240, enlarged the empire even further at the expense of the Romans. He commissioned at Naqsh-i Rustam a trilingual inscription, in Middle Persian, Parthian and Greek (Shapur I's Ka'ba-ye Zartosht inscription, abbreviated as SKZ2 ), in which he listed the regions under his control and recorded his victories over three Roman emperors: Gordian III, Philip the Arab and Valerian (Herrmann and Curtis 2002). Shapur defeated, and perhaps killed, the emperor Gordian at Misikhe in 244 (SKZ, 6). He also founded two cities: Bishapur, in Persia, where mosaics blend Iranian and Roman styles (Keall 1989), and Nishapur, in Khorasan (northeast Iran; Honigmann and Bosworth 2012), located in a strategic position that controlled trade routes connecting Mesopotamia and China.

Although the economy of the Sasanian Empire was predominantly based on agriculture, long-distance trade played a major role. The strategic position of the Sasanian Empire made it a crossroads for trade routes connecting China and India to the Mediterranean. Among the products imported from China along the now well-developed Silk Road were raw silk yarns, luxury ceramics and glassware, with spices and aromatics coming from South Arabia (Chegini and Nikitin 1996: 43). The intensification of trade contacts with China is also witnessed by the discovery of Sasanian coins at Chinese sites (Bivar 1970; Skaff 1998; see Chapter 6).

Shapur adopted a tolerant religious attitude, perhaps following the example of the Achaemenid kings. Although Zoroastrianism remained the official religion, Shapur indicates in his inscriptions the occurrence of rituals and animal sacrifices that had been banned by Zoroastrianism, and he appears to have had a welcoming attitude towards Mani, the founder of another universal religion called Manichaeism (Daryaee 2013: 9; Wiesehöfer 1996: 199–208; Boyce 1979: 111–12).

Following the death of Shapur I in 270, the rivalry between the Romans and the Sasanians was destined to become more intense, especially over the control of Armenia. In the meantime, Zenobia, queen of Palmyra, taking advantage of a period of weakness in the Roman Empire, took from the latter a large portion of territory stretching from southern Anatolia to North Arabia and Egypt, from 270 to 273. In the latter year, however, the Roman emperor Aurelian reconquered all the territories and destroyed Palmyra (Millar 1994: 159–74). With the Roman East finally restored, and by the treaty of 299 (the treaty of Nisibis), signed by the Sasanian emperor Narseh, the Tigris became the border between the two empires (Millar 1994: 209).

The 299 arrangements remained in place until the military campaigns of Shapur II (309–379). After re-establishing Sasanian control over eastern Arabia and deporting some Arab tribes from within the empire, Shapur II attempted in vain to attack Roman garrisons such as Nisibis. He then turned his attention to the east, where he reaffirmed control over the eastern regions, which were being threatened by the invasions of the Hunni and the Kushans (Chegini and Nikitin 1996: 38– 9; Daryaee 2013: 17). According to Ammianus Marcellinus (1940: book 23.6.14), the Sasanian Empire now extended over Mesopotamia (as far as the Tigris), Iran, Margiana, Bactriana, Arachosia (south Afghanistan) and Gedrosia (today's Beluchistan).

Shapur II's reign was the longest in Sasanian history, and the king devoted much effort not only to military campaigns but also to reinforcing the empire's structure. In religious affairs, Shapur II persecuted the Christians, whom he considered to be potential allies of the Romans, as Rome had become predominantly Christian after Constantine's reforms at the beginning of the fourth century. He also tried to bring order to the Zoroastrian religious order by favouring the organization of a council of Zoroastrian theologians. Shapur II also founded several cities in Iran and established propagandistic art through different media, including silver bowls, stuccos and rock reliefs, in which he coded a figurative Sasanian court language (J. Curtis 2000; Daryaee 2013: 20). The rock reliefs of Shapur II and his successors show the kings motionless, standing frontally (thus continuing the Parthian style) next to the divine figure (Mithra), who became the most prominent god, and not displayed as equal to the king as he was in Ardashir's reliefs, perhaps indicating the growing power of the religious elite over the emperor (Daryaee 2013: 20).

Perhaps to balance the power of the Zoroastrian priests, Yazdgerd I (399–420) adopted a tolerant policy towards religious minorities, making Christianity a recognized religion within the empire and promoting (in 410) the first council of the Nestorian Church (Wiesehöfer 1996: 204). Yazdgerd's reign is said to have been a peaceful one, as the emperor never waged war against the Romans and established good relations with the Roman emperor Arcadius (Procopius 2006: book I.ii.1–10).

The emperors who followed had to face several incursions by nomadic groups, among which were the Hephthalites, who encroached on the empire from the east and from the Caucasus. The Sasanian emperors therefore engaged in several battles against these populations, on some occasions with the help of the Romans (Daryaee 2013: 24–5; Chegini and Nikitin 1996: 39).

During the fifth century CE the Western Roman Empire weakened until it collapsed in 476 CE, whereas the Eastern Roman Empire, called the Byzantine Empire, continued for about a thousand years, until it collapsed under the attacks of the Ottoman Turks in 1453. Although Greek was the official language, and Orthodox Christianity the official religion, the Byzantine Empire retained most of its Roman traditions and administrative structures (including a revised version of Roman law; Ostrogorsky 1956). It reached its apex under the emperor Justinian I (527–565), when the Byzantine Empire extended across North Mesopotamia, the Levant, North Africa as far as southern Spain, Greece, the Balkans and Italy. Justinian I strongly promoted Orthodox Christianity against paganism and Christian heresies; he favoured religious art and architecture (the church of Hagia Sophia in modern Istanbul being one example); he also protected international trade relations and encouraged maritime routes towards India and China to bypass the Sasanian Empire (J. A. Evans 2005).

Around the same time, the Sasanian Empire was ruled by Khosrow I (531–579), who was seen as wise and just, a type of philosopher-king. He reformed the empire, its administrative and military structure, promoted trade with both the Byzantine Empire and China, welcomed Western philosophers who abandoned Athens after the philosophical school was closed by Justinian I, and favoured the arrival in his empire of intellectual works from India (Daryaee 2013: 29–30; Wiesehöfer 1996: 216–21). Khosrow succeeded in repulsing attacks by nomads against the eastern border of his empire, signing a treaty with the Byzantine emperor Justinian I in 532 (the 'Eternal Peace'), whereby the Sasanians obtained Armenia and Georgia and the Byzantines definitively left their garrisons in Mesopotamia (Farrokh 2007: 230; Figure 2.10). Soon after this treaty, however, Khosrow resumed an aggressive policy against the Byzantine Empire by attacking the Caucasus and Syria. He was also successful in invading Yemen (Daryaee 2013: 31; Figure 2.10). Khosrow I's conquests were consolidated by his successor Khosrow II (590–628), who reinforced his control of the Persian Gulf and conquered Anatolia, Syria, Palestine and Egypt as far as Libya in 619 (Daryaee 2013: 33). He was eventually deposed by the nobility in 629, and the territories of Anatolia, Syria–Palestine and Egypt returned to the Byzantines in 630.

After Khosrow II's reign, the Sasanian Empire was devastated by dynastic squabbles and eventually succumbed to the Arabs, who had united under the religion of Islam. In 636, the Arabs took the capital Ctesiphon and in 642 they took Khuzistan and Media, with Persia falling in 650 (Daryaee 2013: 37); these victories ensured their grip on the core territories of the Sasanian Empire and opened a new phase in the history of the Near East.

## 2.8 Towards cohesion

In this overview, larger political entities developed after the eighth century BCE and these states often succeeded one another in a near-continuous fashion through the seventh century CE. At times, as new powers arose, campaigns had to be fought to unite various groups; however, large states formed quite quickly after the fall of an empire. Comparison of some of the larger states from the periods discussed clearly shows the trend for large states to become the norm in the AoE (Figure 2.11). The size of the

**Figure 2.11** Approximate total area (in millions of square kilometres) of empires' maximum extent in different pre-AoE and AoE periods. The *x*-axis indicates territory for the Akkadian (AK), Middle Kingdom Egypt (MKE), New Kingdom Egypt (NKE), Neo-Assyrian (NAE), Achaemenid (AE), Seleucid (SE), Parthian (PE) and Sasanian (SAE) Empires

largest third- to second-millennium BCE states indicates that they did not average more than 0.4–0.5 million square kilometres, while the largest AoE states in each period discussed averaged closer to 3.7 million square kilometres. Even if we remove the Achaemenid Empire from the AoE calculation, the average was still about 2.6 million square kilometres, or more than five times the size of the pre-AoE average for the largest states. Furthermore, many of the third- and second-millennium BCE empires in the pre-AoE did not last long, although the Late Bronze Age states were generally longer-lasting. When the large Bronze Age states collapsed, a reversion to city-states or small states is generally evident. Empires were not only smaller in the pre-AoE, but also even these smaller entities fragmented after the collapse of a state or empire.

**Figure 2.12** Territories conquered or fought over in different periods: (a) 883–859 BCE, (b) 626–601 BCE, (c) 553–522 BCE, (d) 334–323 BCE, (e) 114–117 CE and (f) 250–259 CE

Another potential proxy that expresses greater political cohesion during this time is the fact that larger areas were conquered or fought over for less time in the AoE than in the pre-AoE. We have, for example, considered the number of battles Assurnasirpal fought in comparison to Alexander, and the area of territory the latter conquered compared to that won by the former. Figures 2.12 and 2.13 display the total territory conquered and the territory conquered or fought over per campaign year for six different periods spanning the ninth century BCE to the third century CE. For the later battles, there were fewer power centres or regional interests to contest, which meant that each victory yielded more land and conquest was thus quicker. As wealth and power were concentrated in fewer places, greater political integration of territory across the Near East became a possibility.

**Figure 2.13** Territory (in millions of square kilometres) conquered or fought over per campaign year in (a) 883–865 BCE, (b) 626–601 BCE, (c) 553–522 BCE, (d) 334–323 BCE, (e) 114–117 CE and (f) 250–259 CE

The proxy data showing territory size and land conquered, we believe, reflect the fact that universalism had begun to transform the Near East. There were fewer major political and economic centres in the region that dominated territory as people moved to larger centres. It was also easier to create much larger empires, as resources could be saved for fewer key battles. New political capitals and trade hubs were becoming far larger than their pre-AoE predecessors, and their socio-economic interactions across very large regions were often more centralized. In the remaining chapters, key changes in the AoE are explored in more depth, using concepts presented in this chapter, as well as other information which demonstrates social, economic and political cohesion, including how the process of universalism took place.

### Notes


## 3 Methods of analysis

Here we present the key methods that will be used in subsequent chapters. The primary data of analyses are settlement, urban, material cultural and historical data. Some of these will be analysed in descriptive or qualitative ways that are discussed here and in subsequent chapters. Chapter 1 stated that population movement is a fundamental driver through which universalism is enabled. As populations began to move to more distant locations, often to very large cities and even specific regions, mixing with new social groups, the basis of universalism was both established and perpetuated. To demonstrate this, good proxies are ancient settlement patterns found in the Near East, which are best understood from archaeological surveys, from which site sizes and hierarchies can be estimated. These allow us to represent and understand overall settlement structures and how they change between periods. Material culture and historical data support the movement analysis and demonstrate that social institutions adapted to the newly evident social changes that helped to perpetuate a pattern of larger empires and states. The analyses require more explanation before they are applied, which we now focus on.

## 3.1 Archaeological surveys and measuring settlement structures

In Chapter 4, both qualitative, statistical methods and quantitative modelling will be applied to the measurement of probable patterns of population movement. These methods are applied to relative or estimated settlement sizes during specific archaeological periods. Before these methods are applied, however, it is acknowledged that there are obvious problems in interpreting settlement size in any given period. As an example, estimates of site size may not be accurate for the entire period. Nevertheless, the spatial extent of sites in a given period may reflect the maximum size a settlement reached at a given time within that period, or at least indicate whether a site is larger than its neighbours. Therefore, despite its flaws, interpreted settlement size is one of the better measures for providing information on relative population concentrations, even if exact populations are difficult to determine. The relative size of a site is more significant than its exact size, where more minor or major differences between site sizes influence results. These patterns are critical for demonstrating how settlement structures shift from one period to the next.

#### 3.1.1 Methodology: quantitative and qualitative interpretation

Archaeological data are often patchy and not easily interpreted. Ideally, an extensive area with detailed archaeological site-size estimates would provide us with the best data to give information about population concentration in the analysed region. Surveys are often conducted quickly, or are limited by the extent they can cover and the intensity they can achieve. This makes it difficult to use survey data from all regions. Summary statistics, including measures for rank-size hierarchies (Savage 1997), are used to see how settlement structures change over time (from the Early Bronze Age to the Iron Age, for example). We also use a Gini coefficient, which is a general measure of disparity or inequality (Dixon, Weiner, Mitchell-Olds and Woodley 1988). Rather than using it to measure income distribution, its traditional application, here we use it to assess differences in site sizes in different periods. Major changes from one period to another could indicate major shifts in differences between settlement sizes. The measure can show if there is a larger proportional population concentration in the largest site(s) than in other sites. Gini coefficients can therefore be used to measure relative population distribution, or inequality in distribution, in the measured settlements in a region. A larger Gini coefficient demonstrates a greater difference in site size between the largest settlement and other sites.

There are often biases in the recording of survey data: some periods are better represented simply because the material culture is better known or more visible during surveys. In such cases, differences between the ten largest sites in surveys are assessed using the Gini coefficient, as this removes from analysis smaller sites that are often missed in surveys because of a lack of visibility or of a lack of knowledge of the material culture. In other regions, it is evident that settlement structures change, but systematic surveys have not been conducted, which leaves only a qualitative understanding of how large sites are in comparison to other settlements around them. Some areas have been assessed using statistical analyses of settlement structures. In fact, this has been done in publications that are directly used by this work (e.g., Falconer and Savage 2009). Finally, publication quality varies from region to region. For some regions it is easy to reconstruct the location and estimated size of sites, while in other regions these data, even when surveys have been conducted, are difficult to obtain or interpret. These difficulties necessitate an approach that combines quantitative and qualitative methods, in which the chosen method is based on the quality and nature of the data.

### 3.1.2 Methodology: measuring settlement interaction

Where there are expansive areas of archaeological survey data, a method that incorporates spatial interaction and movement could be used to show how settlement structures are formed in different periods. One can combine this method with those that incorporate more qualitative and quantitative statistical summaries to show that there are comparable patterns, even if the data are less clear in some locations.

As stated above, a key measurement outlined at the beginning of this volume is that of population movement and how empires shape such movement. By movement, we mean dispersion and concentration of people in relation to each other. How people interact and move in a landscape generally shapes where and how they can settle; movement then affects overall settlement structure, so that the sizes of sites are influenced by where people can move to (Altaweel, Palmisano and Hritz 2015). While overall population may indicate whether given periods had more or fewer people, the measure of movement allows us to tell which sites attracted more people than other settlements in a region. A method that has proved useful for measuring movement or dispersion of population between sites is spatial interaction entropy maximization (SIEM; Wilson 1970; Davies, Fry, Wilson, Palmisano, Altaweel and Radner 2014). Because this method is not much used in archaeology, we present further discussion and a background description to explain how it can be used, for example to show how movement shapes settlement structures.

#### *3.1.2.1 Background: approaches to spatial interaction modelling*

Applications of SIEM have traditionally focused on modern economic interactions (Wilson 1970; Harris and Wilson 1978), including those between retailers and communities. It has also been applied to settlement structures in different archaeological settings (Wilson 2012; Bevan and Wilson 2013), including the ancient Near East during the Bronze and Iron Ages (Davies *et al*. 2014; Altaweel *et al*. 2015). At its most fundamental level, the approach is applied to help explain the structure of settlement sizes and their distribution in a spatial setting. This includes how location benefits and settlement attractiveness, regardless of the reasons why specific places might be attractive or beneficial, affect why specific settlements become larger or smaller.

The wide range of factors that make settlements attractive include economic, political, religious and environmental benefits. In addition to these features, the method is employed to look at how settlements are affected by transport and at how the presence or absence of constraints on movement affects where people choose to settle. As with settlement attractiveness, factors that affect transport or movement are varied: they may be cultural, political or even environmental. What the approach does is to use the spatial extent and distribution of sites and their sizes to estimate factors that may have allowed such settlement distributions to develop; difficulty or ease of movement is used to investigate interactions.

The benefit of the method is that one can determine whether areas of population growth or decline might be based on distance, the capacity to move in a given landscape, or social-ecological factors that make settlements attractive, which can be termed pull factors (Altaweel *et al*. 2015). The method is general and many factors could affect settlement attractiveness and transport, which allows us to apply this method without full knowledge of all the factors that may have affected settlement structures. The method is therefore useful for the focus outlined in Chapter 1, as the analysis can look at how population movement and interaction would allow given settlement sizes and distributions to develop.

The methodology applied is a spatial interaction model used in a simulation. This means time is part of the analysis, and the analysis looks at how settlement systems change over time until they reach relative equilibrium, or a state in which change is limited. This state allows one to measure how attractiveness and movement enable the settlement structure observed at that state. For this model, return of attractiveness, designated α, controls how much feedback site advantages affect settlement growth for a given region. The presence of relatively large sites indicates areas in which settlement has produced greater benefits. The incentive could increase over time as populations continue to move to specific settlements, creating more growth or positive feedback (R. McC. Adams 2001; Persson 2010). However, site advantages could be altered by events such as war, famine and economic change, or by other social-environmental factors that limit population growth (Cowgill 1975). Additionally, one settlement's benefit is potentially another settlement's loss in a given region: cities or towns may benefit at the expense of other settlements, which leads to less desired places having less overall attractiveness for settlement and potentially diminishing in size over time (e.g., see Van De Mieroop 2004: 38).

Regardless of the overall pattern or trajectory that shaped settlements in a region, α allows one to quantify benefit feedback and determine how important such feedback is. Determining values of α and how they match known site-size hierarchies and structures is one way of establishing how settlement structures change from one period to the next and between regions.

The other key variable is β, which controls how easy it is to migrate to given sites. A clear benefit of increased mobility is that it enables ideas, economic benefits and general interactions to increase rapidly (Braudel 1995). Mobility can limit or increase settlement options for populations, enabling people to choose where to disperse and settle according to different factors (Fox 1971; Desrochers 2001). While people may want to migrate to or live in a particular place, they may not be able to make this choice. Despite the advantages present in a particular settlement, economic, physical, political, religious or other reasons may constrain a person's choice to live there. Cities may reach the maximum population they can support in terms of food or infrastructure. Therefore, while settlements may have attractive factors that pull people to them, there may also be push factors that limit or hinder population movements from one settlement to another.

Overall, the effects of α and β on sites lead to macro-level patterns that represent the regional settlement hierarchies and structure in any period, whereby simple choices to move are facilitated or constrained by circumstances. Intriguingly, a major factor that facilitates or constrains movement is political integration or fragmentation (G. A. Johnson 1980; R. McC. Adams 2001; Altaweel *et al*. 2015). In some cases, political fragmentation may limit options for settlement, creating more numerous, relatively large settlements in small areas, while in other periods a politically integrated pattern may result in fewer larger cities or even in one primate city far larger than other settlements.

#### *3.1.2.2 Spatial interaction entropy maximization details*

The methodology could be applied to reflect the role of complexity theory on settlement structures, and agent-based or individual-based methods (Bonabeau 2002) could be used to allow bottom-up choices to shape settlement hierarchies (Altaweel 2015). Here, however, SIEM is applied, because the intent is to quantify and assess differences in settlement structures between periods that may reflect site advantage feedback and movement differences.

Site advantage feedback and movement can be measured by choosing population for each settlement as their key output effect. While we cannot know what the actual population was for any site in these periods, from site size we can determine whether a site was likely to have had a greater or smaller population than surrounding settlements. As it is used here, population is a proxy that reflects site size, not the actual population of a site in any period. The number of hectares occupied by a site is estimated from survey results, and then the settlement population is scaled in proportion to the site size. As an example, one hectare could represent 1–100 people. The results can then be used to determine the ranges of the values of return of attractiveness (α) and movement (β), in order to create population and simulated settlement hierarchies which are comparable to the empirical record. While the factors discussed above form the core of the methodology, several variables are used to determine settlement structures and simulated populations; they are given here:


To summarize the behaviours of the simulation model, α, or return of attractiveness, enables a settlement's advantages (*Z*) to increase or decrease in relation to other sites through feedback. β controls the effect of distance (*d*); in some cases it is less significant in affecting how sites grow, while in other cases it becomes important in affecting how easily people are able to move. Higher values of α create site populations (*X*) that are larger or more varied for specific sites; lower values put less emphasis on site advantages, which leads to less differentiation in settlement population. Flow (*S*) acts as a proxy for the population value in a given place. It is necessary to obtain the site location, which is used to measure the cost surface distance between sites, before the simulation is used. Site size estimated from empirical data is compared with how well it fits with the final simulated population; in this case, population is measured proportionally to site size. With the exception of site location, input variables can be made to vary during simulations. Overall, site advantage feedback, ease of movement in a landscape and spatial location influence what settlement structures and interactions between sites are possible. These interactions are reflected in a quantitative form within the model, and these dynamics map to fundamental behaviours (e.g., political interaction) that shape settlement hierarchies in any period.

The steps of interaction in the simulation are presented here. First, flow (*Sij*), used to measure flow of people between any two sites (*i* and *j*), is calculated:

$$S\_g = X\_i \frac{Z\_j^{\alpha} e^{-\beta d\_0}}{\sum\_k Z\_k^{\alpha} e^{-\beta d\_k}} \tag{1}$$

What this indicates is that *S* between sites *i* and *j* is affected by any benefits (*Z*), return of attractiveness (α) affecting such benefit's impact, and ability to move (β) within a given distance (or cost surface in this case; *d*) between *i* and *j*. Population (*X*) affects the level of flow between sites (that is, greater population leads to more flow). Total summed interactions for all sites (*k*) and dividing this provides a way to measure any two settlements' interactions. All these interaction flows are summed (*Dj* ) for each site:

$$D\_{\parallel} = \sum\_{\ell} S\_{\parallel} \tag{2}$$

Then *Zj* , or site advantages, at the next time step (i.e., *Zt+*δ<sup>t</sup> ) is calculated:

$$Z\_{\rangle}^{t \leftrightarrow \delta t} = Z\_{\rangle}^{t} + \varepsilon \{D\_{\rangle} - kZ\_{\rangle}^{t}\} \tag{3}$$

The speed at which changes happen to *Z* is affected by ε. Total advantages for sites are therefore adjusted by looking at the total interactions of a given settlement with all sites. In this case, *k* is simply used as a constant that can scale *Zj* . With site advantages evolved based on total flow, that is, sites that gain more people become more attractive, the next step is to evolve site population to reflect the results of interactions:

$$X\_{l}^{t \times \delta t} = \mathbf{n} \frac{Z\_{l}^{t \times \delta t}}{\sum\_{k} Z\_{k}^{t \times \delta t}} \tag{4}$$

Site population in the next time step (*Xt+*δ<sup>t</sup> ) is calculated by taking the new site advantages value (*Zt+*δ<sup>t</sup> ), relative to all sites (*k*), and then scaling each site's population according to the total population for sites (*n*), making advantages, and by extension flow, proportional to population. Once this step is completed, the simulation goes back to (1) and repeats until the end of the simulation, which is generally when results largely stabilize or reach equilibrium in affecting population. Overall, this method is the same as the one expounded in Altaweel *et al*. (2015) and Palmisano and Altaweel (2015), and has close similarities to that in Davies *et al*. (2014). Simulation runs for 100 time ticks are used, giving an idea of how settlement structures, or hierarchies, are affected by α and β values. The time length of simulations represents the length of the historical and archaeological periods presented in the results in Chapter 4.

Three types of scenario are studied. The first measures how settlement structures develop if there is an equal chance that all sites will become large. This scenario requires no initial input other than site location: results are measured against empirical site sizes from surveys to see what values of α and β create settlement structures. The second scenario gives certain settlements advantages using site sizes estimated from surveys. It measures how site advantages affect sites and overall settlement structure. The scenario is used to study the effect of interactions between sites, including which specific sites have greater interaction dominance through flow of people and goods. In this case, α and β values are less of a focus in the results provided, as sites do not have equal advantages, which makes it more complex to compare results from different periods. On the other hand, interactions between sites help to illustrate how effective sites are in drawing people to them, and, by extension, demonstrate movement. A third scenario applies a bootstrapping technique to study how robust or sensitive results are for the first two scenarios. As settlement surveys contain a degree of uncertainty because long archaeological periods mean that many settlements may not have been contemporary, bootstrapping provides a means to test different combinations of sites by removing some sites and detecting whether results from previous scenarios remain consistent.

To demonstrate the model, some conceptual possibilities are discussed. For some cases, it is possible that β, a measure of more or fewer restrictions on transport or movement, is able to lead to comparable results at different range values. If movement is very easy then populations are able to move to settlements and create site-size hierarchies that are less varied or even in population. When movement is more constrained, it may become less direct as populations try to access sites. Intermediate sites may become more important when short-distance movements become the norm and the population begins to stabilize. This creates a site-size hierarchy that has more varied settlement sizes. Some restrictions in movement direct people to specific sites, creating local hubs. Even greater restrictions also result in more equal populations for sites, as the lack of migration means that people stay near to their places of origin, at least in cases where people have equally distributed starting points. Very different reasons could therefore result in comparable settlement structures. However, this is where α has a key role. As this value becomes greater, larger returns for site populations and advantages become possible, which allows one or a few settlements to become far larger than others through positive feedback growth that attracts people to a few centres. As α increases there are fewer possibilities where very low or high β values can lead to comparable results. This means that greater α ranges generally have greater difference between the largest and smallest sites, where larger α helps lead to larger site size, and the possible causes of these structures have a more restricted range. Values of β, assuming all sites have no initial endogenous or exogenous advantages other than their initial locations, in the middle range enable larger sites. Figure 3.1 illustrates this conceptually, along with other possibilities, including how variance in site populations is based on values of α and β when all sites have equal initial advantages.

The model presented allows one to measure and compare return of attractiveness and movement for urban structure growth. Attractiveness and movement result in urban spaces growing or losing population at variable rates. Growth and decline can have rapid effects based on feedback growth, in which change can be exponential. Slow change is possible as the limits of α's and β's effects have less impact and overall population limits begin to influence results. Figure 3.2 shows conceptual outputs that the model can produce, reflecting different types of population scenarios for cities that rapidly or slowly grow or lose population over variable time.

**Figure 3.1** Conceptual ranges of α and β leading to site size similarity or difference and ranges in which sites generally become small or large when they have equal advantages

#### *3.1.2.3 Further analysis of spatial interactions*

Outputs from modelling show interactions or movement between settlements. Such interactions enable growth and decline cycles such as those in Figure 3.2. Interactions are links that show where people migrate from and to, forming a network structure. This allows a graph to be created that can be further processed by different approaches that analyse network interactions. One approach is Markov Clustering (MCL) (van Dongen 2000; Enright, van Dongen and Ouzounis 2002). The algorithm uses a Markov chain that makes links with more interactions more evident. A Nystuen–Dacey (N–D) graph (Nystuen and Dacey 1961) is another relevant approach, as this graph outputs links that have the greatest interactions to a given node from all possibilities, showing where the greatest movement occurs. The MCL and N–D methods allow one to see which settlements become dominant in interactions as hubs. The frequency and proportion of interactions are used to indicate differences in the movement of people between different periods and settlements. These methods are particularly used to study the second scenario in modelling, in

**Figure 3.2** Conceptual examples of growth and decline curves for urban populations that could be produced by the SIEM model

which sites are given initial advantages, as that scenario provides results that allow the rank and size of settlements to be closely replicated and the population interactions that create these results to be observed. These types of outputs not only help to show the trend of interactions between sites but also are used to assess how socio-economic or political cohesiveness could be represented in given regions through settlements.

#### *3.1.2.4 Physical differences in settlements*

While the SIEM method is used to show changes to settlement structure that demonstrate movement, Chapter 5 demonstrates how large settlements physically changed in the AoE as population movement occurred. Physical changes take place in types of religious institutions, size of cities, wealth, art, knowledge repositories, population diversity, languages and other characteristics found in AoE cities in contrast to earlier periods. Small settlements in the AoE are also investigated for their physical and architectural characteristics to see if they represent possible evidence of movement. In effect, here we investigate how settlements change in their makeup as movement and interactions shape the AoE.

## 3.2 Material culture and measuring cultural change

Chapters 6 and 7 will deal with another proxy that can be used to detect and measure population movement, namely material culture, by focusing on how far and how quickly specific objects travelled. This treatment includes how far specific stylistic elements spread. Such displacement of objects and diffusion of stylistic elements often implies movement of people who travelled or were dispersed across the area for different reasons, for example as merchants, emissaries, artisans, deportees, refugees or soldiers, and took their ideas with them.

In order to assess the impact of empires on object trade and diffusion of stylistic elements, how far and how quickly objects and stylistic elements spread during the Bronze Age and the Early Iron Age will be compared and contrasted with how far and how quickly they spread during the AoE. Focusing on distance and the time taken to cross that distance is important, because this will show that in the AoE people could travel further, and often in a much shorter time, than in earlier periods, taking advantage of the political and economic cohesiveness brought about by empires and large states in the Near East. This reflects some of the interactions that will be demonstrated by modelling in Chapter 4.

Travel speed in antiquity was of course affected by many factors, such as topography and means of transport (donkey, camel, wagon, etc.; see Veenhof 1969: 1; Dorsey 1991; Moorey 1994: 12; C. Adams 2007). It should be noted that innovations in transport technology did not always lead to great increases in transport speed over long distances, as social or political limitations may have prevented more rapid movement across landscapes. Improvements such as camel domestication by the tenth century BCE (Sapir-Hen and Ben-Yosef 2013) and the discovery of the monsoon wind in the Hellenistic period (see Chapter 6) facilitated new trade routes with South Arabia and across the Indian Ocean. Apart from these two innovations, however, land and maritime transport in the AoE did not differ much from that in the pre-AoE.

Tracking down the origin of an object is not always an easy task in archaeology; however, considerations regarding raw materials and style can help. For example, in the pre-AoE, we will focus on objects made of chlorite, lapis lazuli and carnelian, because these raw materials were sourced in Iran, Afghanistan and India during this period (Barthélémy de Saizieu Casanova and Casanova 1993; Casanova 1995; Pinnock 1988). For the AoE, the focus will be on incense burner, coins and black pepper. Incense burners were used to burn frankincense (also called frankincense oil and olibanum) extracted from *Boswellia* trees, which can only grow in South Arabia and eastern Africa (Evershed, van Bergen, Peakman, Leigh-Firbank, Horton, Edwards *et al*. 1997; Groom 1981, 2002); incense burners are therefore excellent indicators of how far frankincense was traded. Frankincense will be distinguished from other forms of incense. Black pepper is a good indicator of the extent of the trade network, as this commodity was sourced in antiquity only from India (Tomber 2008). The other object category focused on for the AoE is coins (e.g., Mildenberg 1993). First limited to restricted areas, coins spread across the Near East and beyond, especially after the Hellenistic period; many Near East cities struck their own coins, which allows us, in many cases, to identify their general provenance. In examining the longdistance trade networks before and during the AoE, we will pay particular attention to the movements of merchants and the presence of merchant colonies, so there is a focus on the movement of the people responsible for trade rather than on the indirect and 'down-the-line' movement of objects across distances.

As mentioned above, beyond traded objects, another way to use material culture to reveal population movement is by focusing on stylistic elements. Some features of objects, architecture and works of art (e.g., terracotta figurines), including clothing styles, architectural decoration and iconographic elements, can be ascribed to the particular areas in which these features first appeared. One of these areas is third-millennium BCE Mesopotamia (T. C. Wilkinson 2014); another area that originated a distinctive and characteristic style in the pre-AoE is Egypt (see, e.g., Roaf 1983; Mumford 2013). We will also focus on Greek stylistic elements that can be found across the Near East and Central Asia during the AoE, in both elite and non-elite art. Similarly to objects, analysing how far stylistic elements travelled, and, by looking at non-elite forms of art, how pervasive they were across all strata of society will demonstrate the extent of the long-distance movement of people, in particular artisans. Our analysis will compare the pre-AoE with the AoE and evaluate the results in the light of the political landscapes established by the universal and large empires. Our focus will be on evidence suggesting the actual movement of craftsmen behind the spread of specific stylistic elements.

## 3.3 Other measures

To demonstrate how other important social and cultural elements changed as populations began to move, Chapters 8 to 10 investigate government, language and religious changes respectively between the pre-AoE and the AoE. Most of the data are historical, but archaeological data are also used. The methods will be qualitative, demonstrating distinct differences through comparisons between the two periods. Chapter 8 will show how governments accommodated increasingly diverse areas and their strategies for governing large regions, which facilitated greater movement and created more socially cohesive regions, or at least responded to such socially diverse areas. These actions and institutions also demonstrate how large states were perpetuated, so that after the collapse of one state another large state arose more easily. Policies, in essence, began to reflect cultural and ethnic diversity, while helping to forge long-term bonds between populations.

Chapter 9 investigates common languages, looking at how and where AoE common languages became more widely spoken and written, spanning wide areas across Europe, the Near East, northern Africa and Central Asia. This created many possibilities that allowed easier movement and allowed people from very different backgrounds to live together more easily. In other words, common languages facilitated movement across larger distances as well as the social integration of populations. The use of historical texts demonstrates this.

Chapter 10 applies a comparison of religions, looking at how common ideas arose in the AoE. While pre-AoE religions and religious ideas showed more regionally limited similarities, AoE religious ideas showed commonalities across a wide area even before the rise of universal faiths. The presence and mapping of specific mystery cults shows how the popularity of particular gods spread as empires dominated the political landscape in the Near East and the Mediterranean. The establishment of universal faiths also provided states with tools to help unify different populations, even as they led to new conflicts. Texts and archaeological data are used to show these patterns qualitatively.

## 4 Settlement patterns and spatial interaction modelling

To understand settlement structure and hierarchy, and by extension population movement, we assess settlement sizes and survey data from different parts of the Near East. By movement, we mean population spread or concentration in a landscape and likely interaction across settlements. The interest here is in determining disproportional population change and differences in settlements, where some sites become far larger than surrounding places. To show how settlement structures change between the pre-AoE and AoE periods, the methods discussed in Chapter 3 are used. These include qualitative, statistical and quantitative modelling, including the spatial interaction entropy maximization (SIEM) method described earlier and its associated analytical methods. Clearly settlement data are not perfect, as sites are often destroyed, buried or misinterpreted, or not investigated because of their invisibility in the archaeological record. Therefore, the intent in this chapter is to obtain information on regions in which relative population shifts are noticeable and settlement organization is more clearly evident. The following chapters will incorporate some of the results recorded here and use them to explain other phenomena related to universalism.

Figure 4.1 indicates various regions which were assessed using the methodologies indicated in Chapter 3. The analysis, that is, the choice of which sites to study, is affected by ease of access to data, including whether data are available in a particular spatial format (e.g., geographic information system (GIS) shapefiles), whether there are size estimates for sites and periods, and whether it is relatively easy to digitize and obtain precise locations for sites.

### 4.1 Case study: Southern Mesopotamia

A region with wide-ranging settlement data, where surveys have been extensive and site-size estimates are available, is Southern Mesopotamia (Figure 4.1: 1). This is in large part due to the pioneering archaeological survey work led by Bob Adams and his colleagues, who conducted several extensive surveys of areas nearly abutting each other. Roughly 34,950 km2 have been covered by these surveys in a critical part of the Near East, where many large cities once existed through various periods. While these results were compiled decades ago, and undoubtedly the surveys would have benefited from more recent advances in satellite imagery and mapping, including survey methodology, the large number of sites over a broad area gives us an idea of shifting settlement patterns from the prehistoric to the Islamic periods (R. McC. Adams 1965, 1972, 1981; Adams and Nissen 1972; Gibson 1972; Wright 1981). For our purposes, site-size estimates were sometimes given as a range (e.g., 5–10 hectares); therefore we randomly select a size from the provided site-size ranges or use satellite imagery (Hritz 2005) to estimate the sizes of sites for which full occupation is indicated. More intense surface surveys at Southern Mesopotamian sites, specifically Uruk (Finkbeiner 1991), Kish (Gibson 1972), Mashkan-shapir (Stone and Zimansky 2004) and Lagash (Carter 1989–90), allow us to refine some of the site sizes used in the analysis.

During the Bronze Age (ca. 3000–1200 BCE), as indicated in Chapter 2, Southern Mesopotamia was often fragmented into city-states, although by the Kassite period (after 1600 BCE) the region begins to be more integrated into one larger state for longer periods. Settlement size from the Bronze Age can be reflected statistically, using rank-size curves that demonstrate any significant changes through the Bronze Age. Figure 4.2 a–c reflect rank size for some of the Bronze Age periods, while Figure 4.2d–f show rank size for the AoE (i.e., the Neo-Babylonian/Achaemenid, Seleucid/Parthian and Sasanian periods). In Figure 4.2a–c the greatest difference between the top- and second-ranked sites in the Bronze Age is about 275 hectares (during the Kassite period); the top-ranked site is about double the size of the second-ranked site, and that period has the highest Gini value for the pre-AoE. The Gini index indicates disparities in size between the ten largest sites; its values range between 0.38 and 0.46 in the Bronze Age. In effect, there is greater disparity in site sizes for the largest sites in the Kassite period than in the other pre-AoE periods. In the Neo-Babylonian and Achaemenid periods, the Gini coefficient is far larger than in earlier periods, indicating even greater disparity between the largest sites. This is primarily due to Babylon's great size (about 1000 hectares). In the Neo-Babylonian and Achaemenid

**Figure 4.2** Natural log rank-size plot of settlements in Southern Mesopotamia during the Early Dynastic, Old Babylonian, Kassite, Neo-Babylonian/Achaemenid, Seleucid/Parthian and Sasanian periods, a–f respectively. The values at the top right of each graph (G) are Gini coefficients that show inequality in settlement sizes

periods, there is roughly an order of magnitude difference between the size of Babylon and the next-largest sites (Uruk, Nippur, Larsa and Adams Site #1439). In the Seleucid and Parthian periods, the disparity between toptiered sites is still high, but it decreases more substantially and then rises again in the Sasanian period so that it is similar to the Neo-Babylonian and Achaemenid periods.

It is not clear how large the largest cities were in the later part of the AoE, that is, between the Seleucid/Parthian and Sasanian periods. In these cases, Adams (1965) indicates that Seleucia and Ctesiphon were about 550 and 540 hectares respectively. Seleucia might have been closer to 1000 hectares in size, similar to Babylon in the Neo-Babylonian and Achaemenid periods (Grainger 2014: 39). During the Sasanian period, or at least in the later part of the period, Ctesiphon was not so much a single city as part of a large urban area of sites abutting or near each other. The ruins in this area are called Madāʾen in Arabic, meaning 'cities', indicating multiple cities next to each other (Adams 1965; Invernizzi 1976; Negro Ponzi 2005)*.* In fact, historical sources mention seven cities (although only four or five were major cities, or perhaps some of the cities mentioned were the same city with different names) that abutted or were near each other and together covered about 1500 hectares or more (Lee 2006: 157; Morony 2009). Seleucia and Ctesiphon have not had substantial surface survey: in each case the walled area was assumed to be the total area of the site, so that it is difficult to be certain of its true size. In the case of Seleucia one can use Adams's results as a minimum value, while for Ctesiphon in the Sasanian period historical texts support the possibility that the site is part of other urban sites and formed a district within a larger urban area, which suggests that a site, or more accurately a group of sites, of nearly 1500 hectares is plausible. That is, the Ctesiphon area is more appropriately considered as a conurbation than as one city.

Figure 4.3 indicates the total settled area for the top 100 sites for the six periods investigated for Southern Mesopotamia. While there was an upturn in settlement area in the Old Babylonian period, overall there was an increasing trend towards a greater settled area during the AoE (the Neo-Babylonian to Sasanian periods). These data are used with caution, as site preservation and understanding of ceramics used for different periods in site recognition can vary greatly. Only the largest 100 sites are used here, as the smallest sites, from earlier periods in particular, may be less visible or less well preserved. The average size of the occupied area of the largest 100 pre-AoE sites was 2249 hectares , while for the AoE the average was 2713 hectares. Overall, occupation in the Sasanian period was far more substantial than in earlier periods. The AoE period generally had about 20 per cent

**Figure 4.3** Total area occupied (in hectares) for the largest 100 sites in Southern Mesopotamia for the Early Dynastic (ED), Old Babylonian (OB), Kassite, Neo- Babylonian/ Achaemenid (NB/AC), Seleucid/ Parthian (SEL/PA), and Sasanian periods (SAS)



more occupied area than the pre-AoE, with an increasing trend for larger, top-tier settlements. Combining these results with Figure 4.2 indicates that as the total settled area became larger in the AoE, much of that growth was concentrated in fewer, larger sites and disparity in site size increased.

To look at how settlement structures may have formed in different periods, and how population may have been dispersed or moved across a given landscape, according to settlement distribution, SIEM is employed, using the parameters listed in Table 4.1. Scenario 1 is applied to see what the major differences in settlement structures were between the Bronze Age and AoE.

Figure 4.4 shows the results of applying SIEM to the case study for the Early Dynastic to Sasanian periods. Overall, the results displayed reflect a parameter sweep (North and Macal 2007) (that is, a test of the different parameter values in Table 4.1), to see which values most closely replicate the empirical settlement structures. A linear leastsquares regression is applied that looks at how well a simulated urban population (using population as a proxy for simulated settlement size) compares or fits with empirical settlement size. The results reflect outputs that show the empirical settlement size versus the simulated population of settlements. The proportion of surveyed settlement hectares and simulated population, that is, the size and population of each site divided by the total size and population, allows us to apply the regression and compare the two sets of values. This informs us what values of α (return of site attractiveness) and β (ease of movement) create urban structures comparable to the survey data. The dark regions in Figure 4.4 indicate areas of good fit (r2 > 0.9). The settlement structures assessed indicate that there was generally less emphasis on very large primate sites (sites that are far larger than lower-ranked sites), from the Early Dynastic to the Kassite period (Figure 4.4a–c). In other words, α is relatively low, as multiple sites that were large are evident and the largest sites were not as disproportionately large. Additionally, β ranges between 0.5 and 0.6 and 0.15 and 0.20 for good fit in Figure 4.4a–c when α = 2.1. For the pre-AoE periods, good-fit β values when α > 3.1 are not evident. The best-fit settlement structures for the pre-AoE periods (Figure 4.5a–c) show that generally when β > 0.13 there is closer agreement between the empirical and the simulated results.

For the Neo-Babylonian period, greater fit for values of α at ranges often greater than 3.1 is evident. For Figure 4.4d, good β fits are seen for values of less than 0.8 when α > 3.1 and at 0.01 for α = 3.1. Figure 4.5d shows a good-fit result between empirical survey data and simulated population. For our purposes, what β shows in Figures 4.4d and 4.5d is that to create the settlement structure that is evident, movement may have become easier as a primate site such as Babylon grew in the Neo-Babylonian and Achaemenid periods. This β indicates that people could aggregate more easily in the advantaged site. When α > 3.1, there is also a good fit for β > 0.8. This means that another way to create the urban structures for the period is to restrict general movement but disproportionately concentrate what access there is into the largest site by giving it far greater advantages through high α. In other words, greater restrictions on movement would need to be compensated for by more advantages in order to create settlement structures comparable to those of the Neo-Babylonian and Achaemenid periods. For the Seleucid/ Parthian period (Figure 4.4e; Figure 4.5e), the results show a better fit when α = 2.1 and β > 0.15, indicating a pattern closer to the pre-AoE results. This is affected by the fact that Seleucia was only 550 hectares, in a period when few other large sites existed. This period saw more conflict than the Achaemenid and later periods, with repeated invasions, which could be another reason why the urban hierarchy was not similar to the Neo-Babylonian/Achaemenid and Sasanian periods. However, the size for Seleucia may be incorrect, since it only uses the walled area. By the Sasanian period (Figure 4.4f, Figure 4.5f), the results are once again closer to the Neo-Babylonian/Achaemenid results, where there are goodfit results when α > 3.1. Additionally, β has good fits at very low values (< 0), the best-fit result being α = 2.1 and β = –0.03 (r2 > 0.95).

This scenario shows how urban centres and structures develop when the population has an equal chance to move to any settlement. Another possibility is that a situation in which particular urban centres have advantages over others enables a greater concentration of population in specific sites. These advantages could be an already larger population or other benefits given to the city. To test this possibility, the empirical site sizes from surveys and size estimates are used to give different values for *Z* (the site advantage value). This scenario (Scenario 2) shows the degree to which larger sites influence mobility through their advantages. The results provide a possible insight into the ability of sites to socially integrate particular regions or sites through regional interactions (Altaweel *et al*. 2015). Movements at high volume across the full breadth of a region suggest a region in which movement is easier; these movements allow one to determine whether sites became major hubs for movement from surrounding regions, indicating a likelihood that a site had greater social and political dominance in a region.

Figure 4.6 shows the results of this scenario. These results reflect what factors of α (return of site attractiveness) and β (ease of movement) are needed to develop or maintain rank-size order of sites, as well as the correct proportion of site sizes. Spearman's rank order correlation and linear least squares are used so that the rank order of site population in the simulation is compared to the rank order of site sizes in the empirical data; this allows us to see whether the simulation has more closely determined the correct rank order from settlement survey data. Linear least squares are still used, since this approach allows us to see whether the proportions of site sizes and population between the empirical and simulated data are similar and form a close fit. Overall, what we see is that very comparable α and β values are needed to develop or maintain settlement size and rank for different cases. This reflects situations in which settlements are leveraging advantages in site sizes. For instance, if a site has greater social relevance it may draw more people to it even if its location is not optimal. In the Old Babylonian period, for example, Babylon is already large. In the Kassite period, the

**Figure 4.5** Results comparing empirical and simulated site sizes showing α and β values that have the best or nearly the best fit (i.e., r2 > 0.94) to the empirical survey record. The size portion reflects the area occupied by a site relative to all sites in the surveys. Graphs a–f represent the Early Dynastic, Old Babylonian, Kassite, Neo-Babylonian/Achaemenid, Seleucid/Parthian and Sasanian periods respectively

**Figure 4.6** Results of a parameter sweep applied to α and β for Early Dynastic, Old Babylonian, Kassite, Neo-Babylonian/ Achaemenid, Seleucid/Parthian and Sasanian periods (a–f) respectively in which different advantages, using empirical site size, are given to settlements. Values indicate r2 fit using Spearman's rank order correlation and linear least squares (greyscale shading for r2 > 0.7) together, which compare the simulated population results with the empirical settlement size data

settlement is still large, which may simply reflect a continuity of the settlement's attractiveness from the previous period (for example, its economy could draw many people). The scenario reflects what site attractiveness feedback and movement capabilities are needed to develop or maintain the overall settlement structure in the periods assessed, and accounts for initial settlement advantages. This also has the benefit of addressing edge effects, as these reasons could be endogenous or exogenous.

Because the results reflect settlements in which the initial advantages are based on the size of the settlement from the survey, α and β are less relevant for this scenario. In effect, initial advantages given to larger sites do not require higher α values to make them larger, while β can simply reflect the maintenance of the advantages, and by extension the population, that sites have. In fact, α and β are more difficult to compare because advantages for sites are different for each case. On the other hand, a benefit of this scenario is that it is informative about the level of interaction between settlements that helps maintain rank and size. Such interactions help demonstrate the intensity and distance of travel between sites. Figures 4.7 and 4.8 indicate the centrality of major settlements for the assessed periods in Scenario 2. Centrality is defined as the number of weighted links that connect to a site from a given site. The weights reflect flow output from the SIEM model, which represents a value for people coming to a site from another site. The flow links to sites provide a proxy for determining how influential a site might be in a given region with regard to its ability to attract people. For this case, graphs are studied using the MCL algorithm (see Chapter 3), which emphasizes influential sites in given regions, indicating where key hubs might be located. Figure 4.7 highlights some sites that are central or influential in interactions. These sites are: Uruk (159), Seleucia (185), Lagash (305), Umma (310), Babylon (137), Aqar Quf (ancient Dur-Kurigalzu; 436), Nippur (444), Adams's Site 004 (484) and the Ctesiphon region (803). What is apparent (Figures 4.7a–c and 4.8a–c) is that pre-AoE largest sites are not overly dominant in interactions, and multiple hubs emerge for interactions. Figures 4.7d–f and 4.8d–f show that in the Neo-Babylonian/ Achaemenid and Sasanian periods, Babylon and the Ctesiphon area respectively occupy very dominant and central positions in interactions. The Seleucid/Parthian periods (Figures 4.7e and 4.8e) show Seleucia as less dominant in interactions. In Figure 4.8d and f, flow is heavily concentrated in primate sites; Babylon has about five times greater flow and the Ctesiphon area has about six times more flow than the site with the second-greatest number of interactions. The pre-AoE (Figure 4.8a–c) top two or three sites have far fewer differences in their portion of link flow; no city has more than double that of the second-largest city in flow. For Babylon (d) and Ctesiphon

**Figure 4.7** Pre-AoE (a–c) and AoE (d–f) sites in Southern Mesopotamia for the periods mentioned in the caption to Figure 4.6 and interactions that enable given settlement structures and hierarchies observed. The colour bands indicate the flow intensity (in standard deviation) of the given links between sites. The circles indicate some hub sites that have proportionally higher interaction flow

(f) in the AoE, the portion of the total flow these cities have (20 per cent or more of the total) is far greater than second-ranked sites.

To assess the validity of the previous results, a bootstrapping method that tests the robustness and sensitivity of scenarios and what happens when only a percentage of sites exist is applied. The intent is to look at the entire period and see how multiple combinations of settlements using only part of the dataset at any given time would affect the overall settlement structure and hierarchy. This provides an idea of how well surveys have captured the general settlement structure in scenarios, while helping to show the strength of modelling results if settlements were not contemporary in any given period (see, e.g., Palmisano and Altaweel 2015). It is possible that many sites were not contemporary within the periods studied, as dating generally uses ceramics that are less precise in chronology. To address this issue, and see what may result if different combinations of sites existed in any one period within each archaeological period, sampling is done by removing a ratio of sites (e.g., 0.05) and then selecting sites for a given simulation run. This is repeated 500 times for each ratio; an average fit value with different sets of sites in each run is then determined. While one cannot be sure which sites were contemporary at a given time, this provides more confidence in the results, as these indicate what levels of sampling drastically change results from the previously tested scenarios. The results of this method are presented in the Appendix (Table A.1–2). Table A.1 tests the robustness of Figure 4.5's parameters, that is, the best-fit results in Scenario 1, using least-squares fitting. The results generally show that the Neo-Babylonian period was less sensitive to change and robust at least at the 0.05 sample ratio level (that is, 5 per cent of sites were removed from simulation runs). The results show more weakness at 0.15 and above for sampling. The other cases are more robust at all levels; where moderate weakness in results, however, is more evident at the 0.5 sampling levels. In general, this indicates that, even if a large number of sites were not contemporary in any period, for the pre-AoE sites in particular the overall structures suggested by the results in Scenario 1 are more likely to be representative of what existed. It is possible that the ratios studied may not adequately sample the correct sites that were contemporary, which means the structure simulated might not be accurate, but the simulated ratios are intended to provide a greater measure of confidence for the sites simulated. For Scenario 2's bootstrap test (Table A.2), which tests the best-fit results from Scenario 2, the results are an even better fit at all levels, using least squares and Spearman's rho for all periods tested. The results for Scenario 2 are generally better, since the sites with settlement advantages have remained the same even if different combinations of sites are used for each scenario run.

What is evident in the Southern Mesopotamia case is a shift from multiple large settlements in the pre-AoE to one dominant, primate centre by the Neo-Babylonian and Achaemenid periods. The pattern of one very dominant site in the Sasanian period is also evident. It is possible that this is also the case in the Seleucid/Parthian case; however, Seleucia's size has been estimated to be 550 hectares, although intensive survey was not applied to the site. The results for Seleucia are likely to be less certain, since only the walled area was included in the site's size. Overall, Scenario 1 suggests that it is possible that freer movement initially enables cases in which one large or primate city can develop, whereas in the pre-AoE greater hindrances to movement are evident in the settlement structures. Such cities, through their interactions, become dominant: they have long-distance contacts and interactions that make them dominant in the region. Ease of movement and site attractiveness allow these cities to grow far larger through positive feedback. It is also possible that cities with great advantages obtained through higher α could become very large despite greater restrictions to movement (β). Scenario 2 (Figure 4.7) demonstrates how dominant sites such as Babylon and Ctesiphon are not only developed but also maintained. This means that movement could become either relatively restricted or not after a city has gained initial advantages over its neighbours. The scenario also shows that flow, and subsequently population, proportionally concentrate further in one area, as these sites dominate the region in their total interactions. In other words, large cities develop the ability to draw people and resources from more distant regions as they become more dominant as economic or social centres. Greater advantages of sites in the AoE allow them to draw people from far more distant places than in the pre-AoE. For pre-AoE cities, intense interactions were dispersed among multiple settlements and more localized.

Historical texts relating to Babylon suggest that this city was able to use its influence to attract greater foreign wealth and even foreigners (Jursa 2009; Moukarzel 2014). The ability to draw people and resources from more distant lands would suggest that movement became easier, which allowed resources to be concentrated in the city of Babylon at greater proportions. It is possible that movement was restricted, but this could have happened after Babylon achieved its dominant position; restrictions after Babylon became relatively large would have preserved the city's high proportional population. Ctesiphon shows similar patterns; there, a mixture of ethnic groups suggests immigration into or movement to the city was high (Ṭabarī 1989). Chapter 5 will further discuss documents and material records in relation to Babylon and Ctesiphon, including how such records may reflect regional socio-economic dominance and indicate whether relatively easy movement to these cities enabled their growth and dominance. Our results suggest that Babylon and Ctesiphon were able to become dominant in size through ease of movement and through leveraging their advantages when the flow of people became concentrated in these cities. The AoE may have afforded opportunities for ease of movement and leveraging advantageous situations for these cities. One measure of social or political cohesion may be in the form of settlement hierarchies (G. A. Johnson 1980; Steponaitis 1981; R. McC. Adams 2001). In the AoE, differences in rank-size hierarchies and Gini values indicate greater proportional concentration of population into primate cities.

## 4.2 Case study: the Khabur Triangle

Southern Mesopotamia shows a progression from fragmented cities and interactions in the pre-AoE to more centralized ones in the AoE, during which the largest cities grew more quickly. We now explore other regions. This will help demonstrate whether the phenomenon noticed in one part of the Near East is comparable to patterns seen elsewhere. One region of the Near East that has been relatively well surveyed, in which many sites have been located by using satellite imagery and surveys are nearly contiguous, is the Khabur Triangle in Northern Mesopotamia. The proximity of surveys allows a wider area to be assessed, which will help us to understand whether the regional interactions noticed in Southern Mesopotamia are similar to what occurred in Northern Mesopotamia. This case study's data derive from the settlement surveys highlighted in Figure 4.1(2) (Meijer 1986; Eidem and Warburton 1996; Lyonnet 2000; Ristvet 2005; Wright, Rupley, Ur, Oates and Ganem 2007; Ur and Wilkinson 2008; Ur 2010; Ur, Karsgaard and Oates 2011).

Similarly to the previous case, the analysis begins by looking at the general settlement patterns and hierarchies found for different periods. Figure 4.9 lists pre-AoE settlement rank-size hierarchies from the Early (a) and Middle (b) Bronze Ages and the Iron Age (c) at the beginning of the AoE. What is immediately noticeable in the rank-size graphs is the decrease in the size of sites, whereby the largest sites became smaller from the Bronze to the Iron Ages. The rank-size curves flatten later in time, as the largest sites were no longer very large and there is a more general evenness in site sizes (T. J. Wilkinson and Barbanes

88

2000; T. J. Wilkinson, Barbanes, Ur and Altaweel 2005). The Gini coefficients indicate a far greater equality in site sizes for the top ten largest sites in the Iron Age than for the other periods. In fact, the largest site in the Iron Age is only 20 per cent of the size of the largest site in the Early Bronze Age. The total occupied area is roughly 797 hectares, 1418 hectares and 697 hectares for the Early Bronze Age, the Middle Bronze Age and the Iron Age respectively.

Such results suggest that a very different phenomenon occurred in the Khabur Triangle than in Southern Mesopotamia. While in Southern Mesopotamia the top site became larger from the Early Bronze Age to the first half of the first millennium BCE, here the exact opposite is true. What is argued here, however, is that this could represent the same dynamic as in Southern Mesopotamia. To demonstrate this, a SIEM model is applied to determine possible factors that enable such settlement structures. Similarly to the procedure in Scenario 1 in Section 4.1, Table 4.1 is used to test factors of α (return of site attractiveness) and β (movement) that shape observed settlement structures. Figure 4.10 shows the results.

When α is 2.1, the upper range of good fit for β is 0.44–0.48 in the Early Bronze Age (Figure 4.10a). For the Khabur Triangle in the Middle Bronze Age, Figure 4.10b shows that the upper range of good fit for β is between 0.61 and 0.65 when α = 2.1. Additionally, the lower ranges of good fit for β when α = 2.1 are 0.1–0.15 in the Khabur Triangle in the Early Bronze Age and 0.1–0.15 in the Middle Bronze Age. Graphs a and b in Figure 4.11 indicate the best fit for these Bronze Age settlements. For the Khabur Triangle in the Early Bronze Age, the best-fit β results are between 0.1 and 0.15 in the Middle Bronze Age the best-fit results are when β > 0.6. These results could suggest that the Khabur Triangle in the Middle Bronze Age experienced more impediments to migration or movement interaction. For the Khabur Triangle in the Iron Age (Figure 4.10c), there are greater differences. One result shows that α could be comparable to the earlier cases (i.e., at 2.1), although there are many good fits when α is lower, at 1.1. As Iron Age sites are generally small, the result is expected. Unlike the other cases, however, the best-fitting results are when β is lower, specifically between 0.03 and 0.06 when α = 2.1. There are also β values comparable to or higher than the other periods' upper ranges of good fit. This apparent contradiction is explained by the fact that easier and more restricted movement create settlement sizes comparable to the empirical record, and no site easily gains a larger population as α is lower (for example, see Figure 3.1). If movement is easy or facilitated and α is relatively low, then population spreads at a relatively even rate, while very restricted movement restrains

**Figure 4.11** The α and β values that show a very good fit (r2 > 0.94) between the empirical survey site sizes and simulated site populations. Graphs a–c represent the Early Bronze Age, the Middle Bronze Age and the Iron Age respectively

site sizes differences. In effect, a similar result is achieved under different circumstances. Figure 4.11c shows a good-fit result for the Iron Age. The significance of these results is discussed later in this section.

As previously applied to Southern Mesopotamia, Scenario 2 for the Khabur Triangle can be used to study interactions among settlement structures evident in the region for different periods. This time a Nystuen–Dacey (N–D) graph is used to emphasize which hubs attracted more flow or were more central (Figure 4.12). As before, the graphs show not just the intensity of interactions but also to what extent and from what locations sites are able to attract flow based on their relative importance in interactions, which suggests sites' relative importance in regional interactions. Results show that the Early Bronze Age (α = 0.8, β = 0.4) and Middle Bronze Age (α = 0.9, β = 0.7; Figure 4.12a and b) patterns are similar to those of Southern Mesopotamia, in that multiple settlements (e.g., 156 (Tell Mozan), 62 (Tell Brak), 60 (Tell Leilan) and 61 (Tell Farfara)) appear to be hubs, or locally important and central in interactions, and overall interactions for the region are not dominated by one site. For the Iron Age (α = 0.9, β = 0.6; Figure 4.12c), something similar is noticeable. However, what is evident from the volume of interactions and the central nodes (Figure 4.13) is that the proportion of interactions for the top sites decreases through time, from the Early Bronze Age, through the Middle Bronze Age, to the Iron Age. By the Iron Age, the portion of interactions by the largest site (Tell Hamidiya) is only about 5 per cent of the total flow, whereas it is 14 per cent in the Early Bronze Age. The results show less dominance by any one site in the Iron Age and greater distribution of interactions among all sites, leading to more even site sizes. Although the Bronze Age sites do not have a single dominant site, a few sites are evident as centres. While there are multiple centres of interactions in the Iron Age, none of these attract a high portion of interactions, so that no dominant hub emerges.

The low β range (indicating easy movement) in Scenario 1 and the more equal flow demonstrated in Scenario 2 for the Iron Age is supported by the empirical data. When the region was integrated into the large Neo-Assyrian Empire, long-distance roads appeared in the landscape that connected key Neo-Assyrian cities and provinces (Altaweel 2008). This suggests that movements occurred over long distances, and that the Assyrian centres to the east of the Khabur Triangle integrated politically and interacted with the Khabur Triangle in the Iron Age (Radner 2006, 2011). This is in contrast to the Bronze Age, when the region is known to have consisted of small, fractured states that held territory across the entire region for only short periods (e.g., Eidem 2008, 2012). The

newgenrtpdf

**Figure 4.13** Interaction flow portions for sites in the Khabur Triangle during the Early Bronze Age (a), the Middle Bronze Age (b) and the Iron Age (c)

Bronze Age is also known for numerous short-distance hollow ways or roads connecting sites (T. J. Wilkinson 1994; Ur 2003). Such a preponderance of short-range interactions emphasizes how they could have shaped Bronze Age communities. During periods when communities were politically fractured, movement tended to be more constrained; it occurred primarily between neighbouring sites. In the Iron Age, the fact that long-distance roads become more apparent suggests that movement became easier and occurred over longer distances (Altaweel 2008). Similarly to Southern Mesopotamia, therefore, the early AoE showed relatively easy mobility, facilitating the development of more even site sizes, the main difference from Southern Mesopotamia's Neo-Babylonian and Achaemenid periods being the absence of a primate site that attracts much greater flow. Rather, the sites' populations concentrated in areas much farther away than the Khabur Triangle in the Iron Age. Southern Mesopotamia becomes a region of population concentration with a large urban area, while in the Khabur Triangle populations are drawn away from larger centres.

To demonstrate the strength of these results, a bootstrapping scenario similar to that applied earlier is used. These results are in the Appendix (Tables A.3 and A.4 for Scenarios 1–2 respectively). In general, Scenarios 1 and 2 for the Khabur Triangle show a strong likelihood that the simulation results are meaningful, even when 50 per cent of the sites are removed from scenarios. These results may reflect the greater intensity of surveys in the Khabur Triangle than in Southern Mesopotamia; removal of sites from specific simulation runs may not affect results as much, since there are many sites in a relatively small regional area. Nevertheless, the strength shown in the results is an average, meaning that any individual combination of sites may indicate some significant differences from what is evident in simulation results.

While the Iron Age results for the Khabur Triangle show that settlements remained small and dispersed, which suggests that conditions of low α and β could lead to the observed empirical patterns, the question arises as to whether this pattern persisted for the later AoE. Other cases are therefore needed. Figure 4.14a–c show settlement rank-size hierarchies in the Hellenistic to Sasanian periods for the North Jazira Survey (NJS) and the area of Hamoukar (T. J. Wilkinson and Tucker 1995; Ur 2010). While this does not represent the entire Khabur Triangle, much of the region appears to show developments comparable to these two areas (Meijer 1986; Eidem and Warburton 1996; Lyonnet 2000). In general, as in the Iron Age, sites are small in the later AoE periods. Although the Gini coefficient results in Figure 4.14 show differences from the Khabur

**Figure 4.14**and North Jazira regions for the Hellenistic (a, d), Roman/Parthian (b, e) and Sasanian (c, f) periods. The best-fit results are all r2 > 0.94

Triangle in the Iron Age, the number of sites sampled is far smaller, probably affecting this measure's utility in this case. Simulation results for these AoE periods, in which settlements have equal initial advantages (as in Scenario 1, described above), show α at ≤ 2.1 and β at < 0.04, demonstrating that easy movement within the region may have persisted after the Iron Age (Figure 4.14d–f). Figure 4.15, applying Scenario 2 where *Z* equals site size, emphasizes flow and movement in the region. Because the region is smaller than that which was modelled for the whole Khabur Triangle in the Bronze and Iron Ages, the results show one or two dominant sites in the modelled area. The key output here, nevertheless, is not proportion of flow but to demonstrate that movement is generally easy across the landscape, as flow is directed across the entire area except for Figure 4.15a (the Hellenistic period), which has two main hubs. The fact that the surveyed areas are near to but not actually abutting each other may have affected the results somewhat. While some sites in the scenario appear locally dominant in attracting flow, no site is large or has an overwhelming ability to attract flow, as indicated in Figure 4.14. Generally, the population is low-density and spread across the region. For the North Jazira Survey, Bronze Age results from the application of modelling similar to that described here indicate that more restrictions to population movement are probably shaping settlement structures (Altaweel 2015). Larger and more differentiated sizes are also evident for that period. In summary, after the Bronze Age, large urban settlements mostly disappear from the Khabur Triangle and the North Jazira Survey, the regions becoming characterized more by dispersed, small settlements throughout the AoE. The scenarios demonstrate that in the AoE easy or unhindered regional movement, or at least less hindered than in the pre-AoE, may have affected the development of smaller settlement structures, in which pattern no site becomes overly dominant in size.

## 4.3 Case study: southwest Iran

The previous cases show seemingly divergent patterns of one region having increasingly large top-tier sites, specifically in Southern Mesopotamia, while the Khabur Triangle is characterized by smaller and more dispersed sites in the AoE periods. This section applies some of the methods used above to southwest Iran in the Susiana Plain to discover how the region compares with others (Figure 4.1: 3. Key data sources

newgenrtpdf

**Figure 4.15**2010: sites 60 and 25) and 3 and 70 (Wilkinson and Tucker 1995) are indicated as the sites with the highest flow

are Adams (1962), Schacht (1987) and Wenke (1975–6, 1987), who conducted survey and quantitative analysis of different periods to look at key settlement transformations in the region. Wenke (1987) sees this region as having become an area of high-intensity settlement by the Sasanian period, suggesting that this was made possible by major investment in irrigation by the Sasanians. To sample some of the general trends between the pre-AoE and AoE periods, Figure 4.16 shows the settlement rank-size trends present in the region. What is evident is that secondmillennium BCE settlements (Figure 4.16a and b) and Seleucid-Sasanian settlement patterns show that the size of the largest settlement increases through time. Additionally, the discrepancies between the first- and lower-order settlements become greater later on in the AoE, as indicated

**Figure 4.16** Log size-rank settlement hierarchies and Gini coefficients in the Susiana Plain from (a) the Sukkalmah (2000–1500 BCE), (b) the Middle Elamite (1500–1200 BC), (c) the Seleucid/Parthian and (d) the Sasanian periods

by the Gini coefficients in Figure 4.16. In terms of total occupied area, there are no great differences between the periods until the Sasanian period. The Sukkalmah and Middle Elamite periods appear to have nearly 270 hectares and 320 hectares occupied respectively, compared with about 260 hectares and 870 hectares for the Seleucid/Parthian and Sasanian periods. In the Seleucid period, rather than being characterized by large built-up areas that had increased relative to the pre-AoE, the region showed a greater concentration of population into one chief city. In the Sasanian period, numerous urban areas are present; however, the increase is far greater for one site as overall settled area increased.

Figure 4.17 applies a SIEM model to investigate settlement structural changes, using the same settings as in Table 4.1 (Scenario

**Figure 4.17** Scenario 1 r2 results showing a parameter sweep applied to α and β for the Susiana Plain in (a) the Sukkalmah (2000–1500 BCE), (b) the Middle Elamite (1500–1200 BC), (c) the Seleucid/Parthian and (d) the Sasanian periods

1); however, here β ranges between –1 and 1, as these were found to have a better fit. The best-fit Sukkalmah results (Figure 4.17a) show α and β at higher ranges, greater than 1.1 and 0.5 respectively. For the Middle Elamite results (Figure 4.17b), the best-fit results are found when α = 2.1 and β < 0, at around –0.1 to –0.15. For the AoE periods (Figure 4.17c–d), β < 0 is apparent for the better-fit results, suggesting relatively easy mobility. For the Seleucid/Parthian period, best-fit β is around –0.05, while for the Sasanian it is near –0.4.

The results for Scenario 1 suggest that relatively easy mobility may be evident as early as the pre-AoE periods, with this pattern continuing in the AoE. This is perhaps not surprising, as the area modelled here (2600 km2 ) is far smaller than Southern Mesopotamia and the Khabur Triangle. Using Scenario 2 may provide other insights. Looking at interactions where *Z* equals settlement size (Figure 4.18), in all periods one site largely dominates. In this case, Susa (indicated by '1' in the figure) is the largest and most dominant in interactions in all periods except the Sasanian, when Jundishapur ('98') is the largest. The main difference between what is happening in the pre-AoE (Figure 4.18a, b, e and f) and in the AoE (Figure 4.18c, d, g and h) periods is that the dominant site in the AoE is larger and attracts far more flow. Historically, the region formed parts of Elam and corresponded to Susa's territory during the Bronze Age (Potts 1999). The results for the pre-AoE may simply support the political integration that occurred during that time. In the AoE period, the region is part of much larger empires; this period, and in particular the Sasanian period, were the zenith of economic and population growth for the region (Christensen 1993: 107). The population growth, therefore, could very well be because of high mobility in the AoE that enabled the Susiana Plain to be more intensively settled, leading to greater differences in site sizes between the largest and smallest sites, even though the settlement pattern suggests that single-site dominance was already occurring in the Bronze Age. Overall, the results suggest more intensive interaction, with easier mobility or greater advantage of a single urban site in the AoE periods than in the pre-AoE, which can lead to greater site-size differences between these periods, demonstrated in the Gini coefficients and settlement rank-size distributions.

In the Appendix, Tables A.5 and A.6 demonstrate the bootstrapping results, similar to those of previous cases, where ratios (0.05, 0.15, 0.25 and 0.5) for sites removed from runs are applied. Table A.5 is applied to the best-fitting parameters for Figure 4.17 (Scenario 1), while Figure 4.18's (Scenario 2) results are tested in Table A.6. The Seleucid and Parthian periods show the weakest results once sites are removed,

**Figure 4.18** Interaction relationships using N-D graphs (a–d) and flows coming to sites as modelled using MCL (e–h) for the Sukkalmah (a and e; α = 1.5 and β = 0.7), Middle Elamite (b and f; α = 1.3 and β = 0.6), Seleucid/Parthian (c and g; α = 1.2 and β = 0.7) and Sasanian (d and h; α = 0.9 and β = 0.5) periods' settlement patterns

at greater than a 0.05 rate in Table A.5. As the ratio of sites removed increases, the results become weaker. The results here tell us that, at least at the 0.05 levels (the ratio at which sampled sites may not have been contemporary), more confidence in results is justified. Similarly to before, Table A.6 shows that using site size for *Z* leads to mostly robust results in the bootstrapping method; some weakness in the results are found at the 0.5 level, but generally less than in Table A.5. However, this is an average, which suggests that there could be larger deviations in the robustness of results for any single scenario setting of settlements.

## 4.4 Case study: Central Anatolia

For Central Anatolia, data become patchy in regions and periods, so that it is generally more difficult to have a broad spatial overview of the area and its settlement structure. However, for the Middle Bronze Age, settlement data have been digitized from previous studies and earlier SIEM modelling already applied (Palmisano and Altaweel 2015; Figure 4.1: 4). This work is based on surveys and relevant research conducted in the region that are summarized in Table 4.2; these data allow us to reconstruct settlement patterns and hierarchy for this period (Figure 4.19a). Overall, in the Middle Bronze Age the Gini coefficients are the same (0.28) in Central Anatolia as they are in the Khabur Triangle, showing comparable site-size disparity among the ten largest sites in these regions. Overall, about 1209 hectares are occupied in the Middle Bronze Age.

Once again, SIEM is applied using Table 4.1 parameters to investigate the factors of α (site advantage feedback) and β (movement) that affect overall settlement structure (Figure 4.19b). In this case, the results show that if all sites have equal levels of advantages, then the best results are α = 2.1 and β = 0.051–0.061 and 0.651 for linear leastsquares fits that are r2 > 0.94 between the surveyed and the simulated data. To further determine the social and economic dominance of sites, a second scenario that looks at site advantages using the empirical site size is applied. The result of the best Spearman's rank order correlation and least-squares fit (α = 0.8, β = 0.5) is also indicated (Figure 4.19c). An N–D graph indicates eight main centres, one site being slightly more dominant (Boğazköy; Site 70); these results are similar to those in Palmisano and Altaweel (2015). Scholars indicate that the Middle Bronze Age, and the second millennium BCE in general, was a period of localized conflict (Glatz*,* Matthews and Schachner 2009), which could affect settlement structure by restricting population migration across the landscape. While some good-fit results are seen in the first scenario for cases where movement is less restricted, the results also indicate that more restricted movement is also possible. According to Scenario 2, which is similar to previous cases, no site is able to completely dominate the region – eight hubs are found – in part because of the nature of political competition in the region in the Middle Bronze Age (Palmisano and Altaweel 2015).


**Table 4.2 Sources reflecting surveys from Central Anatolia**

While the above results indicate the settlement structure for the Middle Bronze Age, after the Late Bronze Age settlement sizes and overall occupation may have declined. However, from the Iron Age to the later AoE periods, the overall settled area shows an increase in the number and size of settlements in such regions as north Central Anatolia (i.e., Paphlagonia). In the Hellenistic era, a period characterized by conflict

**Figure 4.19** (a) Rank-size hierarchy for settlements, with the Gini coefficient (G), for CA during the Middle Bronze Age (2000–1600 BC); (b) Scenario 1 results; (c) an N-D graph for Scenario 2

in Anatolia, where the wider region was more fragmented than other parts of the Near East, overall settlement in Paphlagonia was low. In the Roman and early Byzantine periods (ca. second century BCE–seventh century CE), a steady increase in settlements, small and large, is evident (Matthews, Metcalfe and Cottica 2009: 178, 189). This observation is comparable to those for other parts of Anatolia, including Phrygia (Kealhofer 2005: 148), Lydia (Pleket 2003: 89), the Konya region (D. Baird 2004: 232), Sagalassos (Vanhaverbeke, Martens, Waelkens and Poblome 2004: 255), Cilicia (Blanton 2000: 60), and western coastal regions (Izdebski 2013). Overall, much of Anatolia became more intensively settled by the Roman period. One possibility is that as greater socio-economic integration and less internal warfare occurred in the region there was greater opportunity for settlement and economic potential (Köse 2005). Some of this growth could have been migration-driven.

In the next set of runs, SIEM is used to investigate parts of Central Anatolia for the AoE periods where data are present. Before we apply this, however, Figure 4.20 shows settlement rank size from several surveys in the region (Brown 1967; Efe 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997; Ökse 1994, 1995, 1996, 1997, 1999, 2000). The results show an increasing trend in overall settled hectares, from about 135 hectares for the Iron Age to 180 hectares in the Hellenistic/early Roman and 240 hectares in the late Roman/Byzantine periods. The full extent of site sizes given in the survey results suggests that sites did not reach the largest settlement sizes seen in the pre-AoE. However, the surveys are more problematic than the earlier cases, because site area rather than occupied area is provided for each period. If the site areas are an indication of period occupation, then the pattern shows a greater number of larger sites later in the AoE than in the earlier AoE, similarly to the trends seen in other regions of Anatolia. While these results are less reliable than others because of the survey data, with modern cities such as Ankara probably obscuring some of the ancient sites, it is evident that there is a settlement pattern of more equal site sizes in the Iron Age. Even with the less reliable results, this is likely to be true since full site sizes are generally small. There is more differentiation in site size in the Hellenistic/early Roman and late Roman/Byzantine periods. Figure 4.21a shows the results of applying SIEM using Scenario 1 parameters (Table 4.1); it indicates that the best-fit results are obtained when α = 2.1 and β = 0.05–0.1 and 0.4–0.42. For Figure 4.21b, the best results are α = 2.1 and β = 0.11–0.14 and 0.39–0.42 for the Hellenistic/ early Roman periods. The best-fit results are α = 2.1 and β = 0.03–0.1 and 0.39–0.431 for the late Roman/Byzantine periods (Figure 4.12c).

Scenario 2, allowing *Z* to be equal to settlement size, demonstrates flow to settlements similar to other cases discussed previously. In Figure 4.22a and d, the Iron Age (Phrygia period) shows more dispersed interactions; Figure 4.22b and e show Hellenistic/early Roman interactions, mostly focusing on the largest site (Site 35; Harabe, about 40 hectares), where the site has the greatest portion of interactions. Figure 4.22c and f show interactions for the late Roman/Byzantine periods, showing similar central flow, but in this case to Site 81 (Porsuk 1; about 26 hectares). According to these data, from the Hellenistic period and later, there is increasing settlement size but also centralization of flow to the largest site. Given these results and what is known about the region politically and economically between the

e

**Figure 4.21**Roman/Byzantine periods (c)

newgenrtpdf

**Figure 4.22**periods (c); interactions shown for the IA (d), Hellenistic/early Roman (e) and late Roman/Byzantine (f) periods Iron Age and Byzantine periods, movement may have been more restricted in the Iron Age because of warring states in Anatolia (that is, higher β in Figure 4.21a is plausible). By the Hellenistic/early Roman periods, movement may have been less restricted in the scenario. The Hellenistic period also witnessed conflict in Anatolia, as it was less integrated into larger states similar to other regions in the Near East at the time. In the late Roman/ Byzantine periods, however, movement appears to be the easiest or most facilitated of all cases, with β somewhat lower in this period for Scenario 1, and proportionally greater flow towards the largest site (i.e., Scenario 2). To summarize, the results suggest that movement in the region becomes much easier by the Roman and Byzantine periods, as this was a time when the region was well integrated into larger empires for long periods. Flow towards the largest site in the late Roman and Byzantine periods may have been concentrated towards a single site rather than to more dispersed settlements. The largest sites in Central Anatolia during the AoE also appear not to reach the level seen in the Middle Bronze Age, despite having greater regional dominance in interactions. Nevertheless, larger sites may have been present in the AoE, but they may be obscured or have been destroyed by more recent or modern construction.

Similarly to previous cases, a bootstrapping methodology is applied to test the robustness of the best-fit results from Scenarios 1 and 2 (Tables A.7 and A.8). The results for both scenarios show that the results are not very sensitive to change. Weaker results, that is, < 0.9 r2 fits, are seen at the 0.5 sampling ratios for all periods except the Middle Bronze Age. This suggests that, even if many sites were not contemporary, the results observed may represent the known settlement structures, although, for any individual case, settlement structure may have been different in parts of the period represented. While the trends in these results appear to largely parallel what has been described for other regions in Anatolia, size estimates for sites are less clear for Central Anatolia, since surveys were often general and not intensive. Although the results are relatively robust, as demonstrated in the bootstrapping results, the lack of intensive survey in the region means that sites and empirical site-size estimates may have been missed, which may have adversely affected our understanding of the true settlement structures.

## 4.5 Case study: western Syria, southern Anatolia and the Northern Levant

For regions covering southern Turkey and the Northern Levant, several surveys have been conducted. They include the Kurban Höyük (T. J. Wilkinson 1990), the Amuq (Casana and Wilkinson 2005), Land of Charchemish (Wilkinson, Peltenburg and Wilkinson 2016), Homs (Philip, Abdulkarim, Newson *et al*. 2005), Titriş Höyük (Algaze, Mısır and Wilkinson 1992) and Tell es-Sweyhat (T. J. Wilkinson 2004) regions (Figure 4.1). As before, one can look at known sites and apply Gini coefficients to the top ten settlements to obtain an idea of overall settlement inequality or differences between the larger and smaller sites, which tells us if there is much disparity between them. Overall, Figures 4.23 and 4.24 show increasing disparity over time until the Iron Age, whereas disparity decreases in the Hellenistic and Byzantine periods. The Roman period, on the other hand, sees an increase in disparity. The total occupied area declines after the Early Bronze Age and then recovers in the Iron Age. Only in the Roman period does the overall occupied area increase more than in the Early Bronze Age; the Byzantine period sees another decline.

While the results demonstrate general trends for the wider region, consideration of the surveys that have been mentioned may provide evidence of variation across different survey regions, as the larger results may mask geographically relevant developments. The graphs for the Kurban Höyük and Tell es-Sweyhat regions (Figure 4.25a and b) show that they never reached the same total occupied area after the Early Bronze Age in the periods investigated. On the other hand, in the Hellenistic or Roman period the total number of hectares occupied recovered, and even exceeded the Early Bronze Age in the Homs and Amuq regions (Figure 4.25c and d). Therefore, in most areas to the east and around the Euphrates, settled occupation never approached the level of the Early Bronze Age, while in the Levantine regions the area occupied exceeded that of the Early Bronze Age in parts of the AoE, starting in the Hellenistic period and continuing into the Byzantine period. Looking at this further using rank-size hierarchy, we see that in the inland regions, specifically those around Kurban Höyük (Figure 4.26a–d), not only did occupied area decline after the Early Bronze Age, but also settlement rank-size hierarchy became relatively even, as indicated by the Gini values. Even when total settlement area recovers in the Roman period, differences in size between settlements are minor. In the Homs region (Figure 4.26e–h) there are greater differences between the largest and smallest sites over time, particularly in the Roman period. In other words, the Homs region had more and larger sites, with greater size differences, in the Roman period than in the Early Bronze Age. In the Amuq (Figure 4.26i–l) the rank-size differences were no greater in the AoE than in the pre-AoE, but there were more settlements, and a greater area was occupied, in the Hellenistic and Roman periods.

**Figure 4.23** Settlement rank-size graphs with Gini coefficient (G) values for the top ten largest sites. The periods represented (a–g) are Early Bronze Age (EBA), Middle Bronze Age, Late Bronze Age (LBA), Iron Age (IA), Hellenistic (HEL), Roman (ROM) and Byzantine (BYZ)

**Figure 4.24** Total area (in hectares) occupied in the EBA, MBA, LBA, IA, HEL, ROM and BYZ periods

What appears to have been happening is a shift of settlement to regions closer to the Levantine coast later in time and during the AoE, whereas regions around the Euphrates or to the east were less occupied for most periods after the Early Bronze Age. This is largely in agreement with what Mazzoni (1991–2) has stated. In the Roman period, when more settlements in regions further inland are evident, the settlements were generally similar in size; that is, they were small, without the major urban centres of the Early Bronze Age. The dispersed and relatively flat ranksize hierarchy curves were very similar to those in the Khabur Triangle in the Iron Age and later periods (discussed earlier) for regions such as the North Jazira. This suggests that the population was generally smaller, but the settlement sizes suggest that movement was easier or facilitated, as in the Khabur Triangle regions in the Iron Age and later periods. In other words, such structures, of more even settlement sizes, suggest relatively easy movement across the landscape.

**Figure 4.25** Surveys from the Kurban Höyük (a), Tell es- Sweyhat (b), Homs (c) and Amuq (d) regions showing total occupied area (in hectares) for the EBA, MBA, IA, HEL and ROM periods

In fact, it is not just the less occupied regions but also the more densely occupied coastal regions, such as the Amuq during the AoE, that suggest that freer movement and interactions were affecting settlement structures. As the Amuq region is more expansive, and settlements have been recovered throughout the valley for all periods, one can use SIEM to investigate how settlement structures may have been created by factors of site benefit feedback and movement. Once again, Scenarios 1 and 2 are applied to test factors shaping settlement structures, this time focusing on the Roman period when widespread settlement and total area occupied were at their peak. The results (Figure 4.27) show good β fits for values < 0.1 when α = 2.1 (Figure 4.27a), some of the best fits being at α = 2.1 and β = 0.069 (r2 = 0.98; Figure 4.27b). Figure 4.27c and d show Scenario 2, which emphasizes how sites can maintain rank and size. The best results (α = 0.8 and β = 0.4; r2 > 0.98 and Spearman's rho > 0.94) show dispersed interactions where no site is able to draw

 (e–h) and Amuq regions (i–l), with Gini coefficient (G) values. The periods shown are the EBA (a, e, i), MBA (b, f, j), Hellenistic (c, g, k) and Roman (d, h, l)

**Figure 4.27** Results of SIEM (Scenarios 1 and 2) for the Amuq region in the Roman period. The results show r2 fit between empirical and simulated data for Scenario 1 (a) and the best-fit case (b). Scenario 2 shows an N-D graph (c) and a portion of interactions (d). (e) and (f) show a hypothetical case that adds Antioch using Scenario 2's approach

many more interactions, similarly to the Khabur Triangle in the Iron Age. As before, a sensitivity analysis was done using a bootstrapping method. The results are robust at all sampling levels (0.0–0.5 ratios) for both scenarios: the 0.5 ratio resulted in a least-squares fit between the empirical and simulation of r2 > 0.94 for Scenario 1; Scenario 2 shows r2 > 0.98 and Spearman's rho > 0.98.

An aspect missing from this analysis is the major city of Antioch, which is mostly obscured by modern occupation. Its incorporation in the analysis would certainly have affected the results, given the ancient city's very large size (perhaps over 1000 hectares). Figure 4.27e and f show hypothetical results under the assumption that during the Roman period Antioch reached 1000 or more hectares; the results show the city dominating the region's interactions. This city, in fact, was probably already several hundred hectares by the Hellenistic/Seleucid period, which would mean that even as early as that the city would probably have dominated the region in size (Aperghis 2004: 93; Cohen 2006: 93). This would suggest that the Amuq in the Hellenistic to Roman periods bore more similarity to Southern Mesopotamia during the Neo-Babylonian and Achaemenid periods (see Casana 2007); the Gini coefficient for sitesize difference should also be far greater than indicated in Figure 4.26k and l. The results probably show that in the Hellenistic to the Roman periods the Amuq mostly had small settlements, Antioch probably being far larger than anything nearby. Adding a large Antioch to the region shows it dominating interactions as Babylon did. This suggests that movement towards the centre of Antioch was relatively easy, which is similar to Scenario 2's results for Babylon's Neo-Babylonian/Achaemenid periods, indicating Antioch's regional socio-economic significance.

From these results, there are two possibilities for areas that have greater occupation by the Roman period, specifically the coastal and Levantine regions. One is increased occupation spread over many smaller sites, while the other is increased settlement over the entire region with a more dominant centre as in the Homs region, probably the city of Antioch. The Amuq region interactions show that movement could have been easier or less constrained in order to form the settlement structure observed. In fact, for the Amuq, the proximity of Antioch seems to have made many settlements around this site far smaller than they might otherwise have been. While one cannot know how many settlements were contemporary, the settlement structures suggest that, perhaps as early as the Hellenistic period for the Amuq, but certainly by the Roman period for both the Homs and Amuq regions, movement was easier, and population may have been concentrated in one large city. Unfortunately, much of the Achaemenid period is relatively unknown in these regions, but political integration may already have facilitated interactions and ease of movement by this period. In the Iron Age, however, smaller sites are known, suggesting that regional interactions more like those in the Khabur Triangle may have occurred earlier in the AoE, when movement may have been relatively easy as in the Khabur Triangle.

## 4.6 Case study: the Southern Levant

As quantitative analysis has already been applied by authors who have investigated settlement patterns in parts of the Southern Levant (Falconer and Savage 2009), some of the relevant results can be summarized here. In this case, cluster analysis and assessment of the types of rank-size curves indicate a Bronze Age landscape of shifting or multiple centres where the political landscape is interpreted as fractured and dynamic. The settlement patterns reflect this, in that multiple major centres arise in different periods that dominate specific but small regions, and settlement structure and hierarchy change throughout the third and most of the second millennia BCE. In particular, in the Middle Bronze and Late Bronze Ages, even when much of the Southern Levant was occupied by Egypt, the pattern of city-states is evident. Overall, little regional integration is evident in settlement patterns. These statistically based results largely support the similar conclusions already drawn for other regions such as Southern Mesopotamia and the Khabur Triangle during pre-AoE periods.

Other works can be used to summarize trends that can be compared with what has already been discussed. In the Iron Age I (ca. 1200–1000 BCE), small sites (254 from the survey) dominated much of the inland hilly regions of the Southern Levant (Finkelstein 1998; Levy and Holl 2002). New migrations or changes in settlement were already apparent by the Late Bronze Age. In the Iron Age I, the inland regions did not show a clear urban centre. The coastal region, on the other hand, probably showed the establishment of the Philistines, whose five main cities (Ekron, Gaza, Gath, Ashdod and Ashkelon) had already begun to develop, which may have led to the gradual squeezing out of the Canaanite populations of the region (Stager 2003). By the Iron Age II (ca. 1000–600 BCE), several small states had arisen, including the Philistines', Israel, Judah, Ammon, Edom and Moab. These states generally had a chief city associated with their territory, which was often the political capital. The eighth century BCE in particular saw the rise of many larger towns near each other, including Dan, Hazor, Megiddo, Jokneam, Dor, Samaria, Shechem, Jerusalem, Gezer, Beersheba and Lachish (Faust 2012: 259). Some of these, however, belonged to the same state. Nevertheless, given that several polities occupied the Southern Levant, this pattern is likely to be similar to that seen in the Bronze Age, in which multiple large towns existed near each other and multiple states existed in a small area. The ninth to eighth centuries BCE were a period of major conflict between small states in the region and, in particular, with Assyria. While Jerusalem was a large city in the seventh century BCE, its sacking in the sixth century BCE suggests that much of the region may have become devoid of large settlements by that time, although this is not universally agreed upon (Lipschits 2006).

In the Achaemenid period, surveys and scholars have indicated a decline in the total number of settlements in the inland hill country in Israel and Judah (Faust 2007). Where there are clear settlements they are generally small. This reflects a pattern similar to that seen in the Khabur Triangle, where the region became more rural and sparsely settled. Jerusalem may have served as a slightly bigger town, but it was still likely to have been no larger than 5 hectares, while other rural settlements were generally smaller than 1 hectare. On the other hand, evidence of Phoenician-influenced settlement appears to indicate some population increase or recovery along the coast. This probably reflects new commercial interests in the region, as international trade increased by the Achaemenid period (Lipschits 2006). In the late Achaemenid, or more clearly in the Hellenistic period, in the fourth century BCE and later, settlement began to increase and, probably, overall population (Lipschits and Oren 2007; Faust 2007). Some of this probably reflects Achaemenid construction of fortresses and other sites, possibly including administrative ones, in response to Egypt breaking away in the late fifth and continuing into the fourth centuries BCE. The trend of increased settlement continued throughout the Hellenistic and early Roman periods. By the first century BCE, it is likely that Jerusalem was far larger than any other regional town in the Southern Levant, although disagreements remain about the exact figure for that population (Levine 2002: 343; Geva 2014). Overall, a pattern comparable to that in the Northern Levant occurred, in that the settled area increased in the coastal regions and slightly inland, by the Hellenistic period, trade being a probable factor in this.

In Late Antiquity, that is, from the late Roman period until the Byzantine period, like the Northern Levant the Southern Levant experienced high population growth (Broshi 1979; Bar 2004). In this period, seven settlements were probably between 90 and 120 hectares (Ptolemais, Legio, Caesarea, Scythopolis, Aelia-Capitolina, Anthedom and Gaza). Four of these cities were on the coast, and the other three slightly inland. It is likely that this reflects the prosperity in the region with regard to the trade and other commercial activities that became more active across the Mediterranean, with the Roman and Byzantine Empires integrating the region. The settlement pattern largely reflects trends seen in places such as the Homs region, where larger sites emerge. However, settlement sizes are dissimilar to Antioch's, sites being probably not larger than 120 hectares, which indicates that the Northern Levant had a far larger city in the form of Antioch, which was probably more than 1000 hectares. In effect, while population grew and settlements became larger, size disparity between the largest and second-tier sites probably grew throughout the Levantine region. Population was more concentrated along the coast, but within this concentration it was skewed to a particular place.

Generally, the Southern Levant has a lot in common with the Northern Levant. The region moves from a fractured Bronze Age system to a more densely populated Iron Age one, although most of the Iron Age shows a fractured political landscape and probable settlement pattern. In the early Achaemenid, apparent low settlement numbers reflect perhaps either a lack of knowledge of Achaemenid material culture, or that it took some time for settlements to recover from the destruction and deportations of populations that occurred in the seventh to sixth centuries BCE. The settlements that are known are structurally similar to those we see in post-Iron Age regions of the Near East such as the Khabur Triangle. By the late Achaemenid, or at least by the Hellenistic period, and into Late Antiquity, many settlements and a large number of areas were occupied again, although in many cases settlement sizes between the largest sites were not very different in the Southern Levant. Unlike in the Northern Levant, very large cities like Antioch appear to have been missing. Nevertheless, the recovery of settlements by the Hellenistic era begins to reflect trade and other activities in the coastal regions. The trend of greater population shifts towards the coast in the Levant suggests movement of population to areas where active trade and other interests were growing along the Mediterranean. This begins to show a closer economic, and eventually political, integration across the wider Mediterranean. The dominance of Antioch suggests that by the Roman period it would have been in a category of its own in terms of urban scale, dominating in size the length of the Levantine coast, where it was possibly an order of magnitude larger than any other city. This would indicate that Antioch had a similar socio-economic dominance in the region to that of Babylon in the Neo-Babylonian/Achaemenid periods.

### 4.7 Summary

Throughout the AoE, a pattern of easier movement becomes evident in several regions. While we cannot define movement or migration as a short-term process, settlements, as a picture of where people lived in a given period, indicate that cities that were disproportionally larger were developing in some selected regions. Figure 4.28 shows relative patterns of urbanism in the AoE from the end of the Achaemenid/early Hellenistic to the Sasanian periods, when the trend of large and small urban areas shows regional variations. Total population may have increased, particularly in parts of the Roman, Parthian and Sasanian periods, but larger differences between city and site sizes indicate that greater population concentration was also happening in regions in relation to the overall population. Regions that include southwest Iran, Southern Mesopotamia and the Northern Levant developed cities that were far larger than anything near them. The Levant in general became a region of greater urbanization, and this is probably true of other places, such as the coastal regions of Anatolia. These changes were not uniform, and concentrations of populations were focused differently in the various regions and periods. Whereas Southern Mesopotamia developed an even larger capital in the Neo-Babylonian and Achaemenid periods, Antioch reached its peak size in the Roman period. In southwest Iran, Gundishapur became far larger in the Sasanian period. In the AoE, the Susiana Plain in Iran has large-scale settlements that not only dominate the immediate surroundings but also suggest that movement from distant areas could shape them; greater attraction to a single site is evident. Where movement across a landscape is facilitated, population can grow disproportionally. Such growth may not be explained by natural birth alone; it is likely that migration also explains why some regions, rather than just sites, gain in population. Nevertheless, population growth often favours those places that have natural or accrued advantages, such as through trade and wealth, and this leads to greater concentration in fewer places and creates more disparities in site size. This is possible when movement becomes unhindered, allowing concentration of population as people from distant regions are able to migrate. In the Khabur Triangle and Northern Mesopotamia during the Neo-Assyrian period, decreasing site sizes and a flattening of the settlement rank-size curve are evident. Both patterns, of one very large site and of sites that are relatively even in size, could be created by the greater ease of movement that was afforded during periods of broad political integration. For the Khabur Triangle and Northern Mesopotamia, greater population migration became focused on regions

**Figure 4.28** General representation of urban growth from the late Achaemenid to the Byzantine/Sasanian periods. The Levant, particularly in the Hellenistic-Roman period, experienced greater urban growth, while much of Northern Mesopotamia probably saw cities declining, or at least less abundant, during and after the Iron Age. Cities that are hundreds of hectares larger than nearby sites are indicated as very large (primate) cities

that were more distant. In general, these regions became less urban than in previous periods. Although larger cities, such as Nisibis and Hatra, still existed in Northern Mesopotamia in the AoE, they became fewer, and often farther apart.

The pattern of movement in Anatolia during the AoE may be different from that in other regions. For instance, the Hellenistic-period settlement structures in Anatolia suggest more politically fragmented patterns, in which multiple large centres and the lack of a very dominant site are evident. By the Roman period, the region appears to have great political unity, which might be evident in the settlement patterns. Other regions (e.g., the Kurban Höyük region) show similarity to the Khabur Triangle in the Iron Age: flat settlement hierarchies are evident and no settlement dominates. In effect, throughout much of the AoE, areas further from the coastal region in the Near East were characterized by a greater number of smaller sites. Nearer to the coastal regions, by the Hellenistic and later AoE, greater population concentration is present. Antioch, in a similar manner to the Ctesiphon area and Babylon, probably dominates the region, and no other city near it is likely to be similar in scale and population.

The settlement structures therefore begin to provide a picture of changing settlement sizes in the Near East as early as the Late Iron Age, these changes continuing into later periods of the AoE and leading some regions to have much larger cities or generally larger sites. Previous peaks of about 500 hectares, reached in the Bronze Age, were far surpassed in the AoE, when population movement became a possible mechanism for a lot of this growth. These changes may have not have been happening concurrently in all regions but do appear to have become pervasive throughout the Near East by the later periods of the AoE, that is, from the Roman to the Sasanian period. Some regions became more rural, such as the interior regions of the Near East; greater population movement may have made this possible. This process had begun by the Iron Age. Even the regions around Nineveh and Ashur, once the capitals of Assyria, probably became less populated after the fall of the Neo-Assyrian Empire. In the pre-AoE periods, restricted movement created more centres or relatively large top-tier sites, with a large number of sites similar in size and not very far from each other. In the next chapter, we present data from within cities, including some of the cities discussed here, that may provide further evidence of movement and social change during the AoE.

## 5 The changing nature of cities and other settlements

Chapter 4 showed that patterns of urbanism changed during the AoE. These patterns include increasingly small settlements in areas where site sizes were more diverse, as dispersion and easier movement of population spread people across the landscape. At the same time, much larger centres emerged where greater populations began to concentrate. These included the cities of Babylon, Antioch and the Ctesiphon area. Such urban patterns were fundamentally new in their time, representing a clear shift away from the pre-AoE urbanism that was often characterized by a greater number of larger settlements much closer together. Although there were fewer centres in areas where larger cities were once present in the interior regions of the Near East in Syria and Mesopotamia in the AoE, the Levant became more densely settled. The previous chapter investigated the underlying dynamics that shaped pre-AoE-to-AoE settlement patterns; this chapter focuses on the characteristics found within cities, including how cities changed and how their physical characteristics, such as their architecture and population, reflected greater migration. Cities are defined as urban areas greater than 20 hectares (Creekmore 2014: 35), where size reflects a larger concentration of population than in other types of settlements in the environs. Case studies of large cities, including many of the largest, from different periods and regions are presented to contrast urban characteristics between the pre-AoE and the AoE. These are used to show how movement of population may have resulted in noticeable change within cities and not only in the surrounding settlement structures. Additionally, the nature or characteristics of some small settlements, including those that replaced areas where larger cities were once found, are discussed to demonstrate a pattern of easier movement in the Near East during the AoE. While both the pre-AoE and the AoE periods show material cultural changes within cities, during the AoE forms of syncretism in material culture, ideas, government, religion and variety of languages show connections between more distant regions. These AoE changes began to reflect some of the consequences of movement across the Near East that helped to shape the region's emerging universalism.

## 5.1 Large pre-AoE cities

In the fourth millennium BCE, Uruk (R. McC. Adams 1981; Finkbeiner 1991) and Tell Brak (Ur, Karsgaard and Oates 2007) developed into large settlements in Mesopotamia. In fact, the trend of larger settlements continued into the Early Bronze Age in the third millennium BCE (Wilkinson, Philip, Bradbury *et al*., 2014), with urbanism spreading to more areas in the Near East in later parts of the Bronze Age. By roughly 2500 BCE, Northern and Southern Mesopotamia, as seen in Chapter 4, had multiple sites that were more than 100 hectares. Additionally, not only did the largest types of settlements become bigger during the pre-AoE, but also cities displayed key characteristics during this time. These included large temple complexes for the main deities, and the chief cities had palaces. Other distinctive features included city walls, monumental gates and upper towns that served as areas for government centres or major religious precincts, sometimes fortified or separated from the rest of the city. Lower towns were often enclosed within outer walls, private houses were found within condensed neighbourhoods, and smaller shrines or temples were located in neighbourhoods. Other characteristics within cities included shops, markets, open spaces, gardens and even places for goods manufacturing (Van De Mieroop 2004). Cities and regions were often multi-ethnic (Kamp and Yoffee 1980) even in the pre-AoE, although it is difficult to detect this using material culture, given that many groups either were assimilated or adopted local customs. This is not always the case, however, and groups sometimes brought very distinctive culture to the region (see, e.g., Kohl 2009). In some of these cases, there were rapid or radical changes in the material culture, for example in the pottery used. Nevertheless, most of these more rapid changes occurred in the less urbanized regions of the Near East in the Bronze Age, such as in parts of Anatolia or northern Iran.

#### 5.1.1 Uruk

An example of a large urban centre in the pre-AoE, and one of the largest cities during most of the third millennium BCE, was Uruk, which reached a size of roughly 400 hectares in the first half of the third millennium BCE (Finkbeiner 1991). By the mid- to late fourth millennium BCE, Uruk had already developed large religious complexes and temples (Figure 5.1). In the Ur III period (2112–2004 BCE), the Eanna district, which was one of the two main districts, continued to display a large temple complex to a major goddess, this time adding a multistage tower or ziggurat (van Ess 2001). Such complexes were religious centres, but they also had important economic roles, as temples often owned major landholdings and were involved with the production of goods. The temples of chief gods were critical to cities, as the identity of a city and its fortunes were seen as being related to the gods; upkeep of these temples and religious activity were intended to appease the gods that resided in these temples. In fact, in Mesopotamia, and probably in other ancient Near East regions in the Bronze Age, ideology, chiefly in relation to urban-based gods, and government power were tied closely together in cities that saw their political fortunes vary (Van De Mieroop 2004: 33). A god who did not favour his or her city might abandon it, leading to its decline, while a favoured city

**Figure 5.1** The Eanna district at Uruk during the late fourth millennium BCE (Eanna IVa and Eanna IVb; after Lamassu Design 2009)

could grow, flourish, and even become a centre for a larger state. Uruk devoted much of its space to its chief temples.

These temples are distinctive in representing the architecture prevalent in the region of Southern Mesopotamia. Distinctive features, such as large ziggurats surrounded by sacred precincts with large courtyards, characterize major cult centres in Southern Mesopotamian cities during the late third millennium BCE. The primacy of the city, and its regional culture, were indicated by the distinctive architecture of the cult centres. Even though foreign populations from distant regions had already begun to live in Southern Mesopotamia, the signatures of these cultures on Mesopotamian religious complexes were often not distinctive, although ideas from different regions were probably blended within established traditions. In other words, such populations probably became assimilated or kept their religions away from the major cult centres. Patterns of syncretism in religious architecture are less evident in the third millennium BCE. The basic form of major temples and temple complexes in Southern Mesopotamia had developed by the fifth to fourth millennium BCE (Safar, Mustafa and Lloyd 1981; Nissen 2002), multistage ziggurats being added in the third millennium. The pattern of major religious complexes dedicated to gods continued into the first millennium CE with minimal change.

#### 5.1.2 Ur

Bronze Age Ur was another major city, though far from being among the largest, as it reached a size of about 90 hectares. While Ur had many features similar to Uruk's, including chief temples (to the gods Nanna and Ningal) which were enclosed in a sacred precinct, what has been revealed at Ur is a substantial part of a residential district within the city (Woolley and Mallowan 1976). The Old Babylonian (ca. 2000–1600 BCE) residential area had houses, primarily courtyard and linear structures (Van De Mieroop 2004: 80–1), neighbourhood shrines or temples, open spaces, squares, shops and workshops. The houses have a typical Near Eastern pattern or even a Mediterranean-style configuration of dense housing, probably with relatives often living close to each other. Alleyways and streets are generally narrow, which is characteristic of the Mediterranean region as it keeps areas cool and shaded. Burials were underneath houses or in cemeteries (Leick 2002; Van De Mieroop 2004). In particular, Ur is a good example of larger residential districts within pre-AoE cities, as many other major Mesopotamian cities do not have large areas exposed. During the Old Babylonian period in the early second millennium BCE, households began to write more extensively, or to have access to writing, which allows us to learn about private economic activities, inheritance and family relations. Although multiple ethnic groups migrating into Mesopotamia are known, they are difficult to distinguish, given the similarities of material culture between many groups (Arnold 2004).

As one might expect from an important maritime city, many foreign objects from distant regions were present at Ur during the third and second millennia BCE, including carnelian, gold, silver, electrum, shell, various stones and lapis lazuli. These items originated from such places as Egypt, Anatolia, Iran, the Indus, Arabia and Central Asia (Figure 5.2; Woolley 1934; McIntosh 2005: 257). Despite the city's connection to maritime trade in the Bronze Age, indications of significant foreign populations from areas where luxury items were obtained (e.g., lapis lazuli) are not evident at Ur. While importing of exotic goods became common in Ur, the integration of foreign populations as part of trade colonies or general movement was not evident. There is evidence of individuals who may have been from Meluḫḫa, that is possibly the Indus region, residing in Sumerian lands, around Lagash, although if this did occur it does not seem to be a large settlement and some even had Sumerian names (Parpola *et al*. 1977: 150). Even if one assumes that trade with the areas the foreign objects found at Ur came from was direct, the longitudinal range the precious objects seem to span was from North Africa/Egypt to Central Asia/India, a distance that was surpassed in the AoE. While wealthy cities such as Ur developed tastes for exotic foreign goods, these goods were probably brought without any large-scale accompaniment of foreign populations.

**Figure 5.2** Reconstructed headdress and necklaces (a) and the socalled Standard of Ur showing combat (b). These works incorporate carnelian, lapis lazuli, gold and shell imported to Ur (after JMiall 2010; Standard of Ur 2016)

#### 5.1.3 Ebla

In the mid-third millennium BCE, one of the great cities in the Northern Levant and western Syria was Ebla (Matthiae 1981, 2010, 2013). This city dominated much of this region politically, in the period immediately before it was sacked by the Akkadians (ca. 2400–2200 BCE). Texts show that it had an important if not dominant economic role as well; the main palace (Palace G) was central to regional trade and exchange. The city had links with a long-distance trade network that connected Central Asia and Cyprus, although some of the goods that came from distant places, such as precious stones, may have arrived indirectly via various trade routes from the Gulf or Iran. There is a good understanding of the architecture of the famous Palace G, which had a large archive of texts (about 20,000 cuneiform tablets) and forms a large part of our historical understanding, from around 2400 to 2300 BCE, of Northern Mesopotamia and the Levant. The site itself was about 60 hectares at its peak, but few residential districts have been extensively studied (Figure 5.3). After its sacking, it continued to be occupied as a major centre into the first half of the second millennium BCE, but it was abandoned shortly thereafter. Large palatial structures, which were often the centre of government and administration, had an important economic function for the city, and were a characteristic of major cities such as Ebla by the third millennium BCE. Palaces such as Palace G reflect regional architectural traditions; in this case, the culture integrates Levantine and Mesopotamian traditions. The gods represented in texts and in the city come from Mesopotamia or the Levant (Snell 2011: 133). As in Uruk and Ur, wideranging ethnic diversity is not evident, although some of the gods, who are unknown to us, may have come from more distant regions. Exotic goods are found, but they reflect mostly the trade activities the palace was engaged with rather than the people who became part of the city's social fabric. Ebla was a regional centre but its gods and material culture show no evidence of having integrated diverse populations beyond Syria and the Levant.

#### 5.1.4 Mari

One of the best-known and most complete palaces in the ancient Near East is the palace at Mari renovated by Zimri-Lim (eighteenth century BCE), which was occupied during the early second millennium BCE (Parrot 1958; Margueron, Pierre-Muller and Renisio 1990). Mari was also one of the most important cities along the Euphrates north of

**Figure 5.3** The acropolis and lower mound (or lower city) of Ebla (about 60 hectares) with key areas within the site indicated, including Palace G, which was the main palace in the mid- to late third millennium BCE (after Barlemi74 2014)

Southern Mesopotamia, possibly reaching ca. 100 hectares at its peak occupation (Figure 5.4). The palace and its archives, of about 25,000 texts, form an important historical record of a period in which several

**Figure 5.4** The site of Mari showing key structures and temples of local or nearby Near Eastern gods (after Attar-Aram 2015)

dynasties had power in Northern Mesopotamia. The city's history spans much of the third millennium BCE, and conflicts with Ebla are attested in texts. Much of the palace archives relate to the eighteenth century BCE and the rulers Shamshi-Adad and Zimri-Lim; the latter ousted the dynasty of the former. In the Middle Bronze Age the city was destroyed by Hammurabi, although a small number of people remained in the city. The archives provide us with an understanding of how volatile politics were in the Middle Bronze Age; dynasties and larger states were often short-lived and not often replaced to a similar spatial extent (Durand 1997, 1998, 2000).

Mari is a case study of a city that shifted from being the centre of a small state to the centre of a small empire; the palace archives document this change. The culture of the surrounding region mostly remained similar during that time, with Hurrians, Amorites and Assyrians characterizing Northern Mesopotamian politics. Localized culture is expressed in the palatial architecture (a large courtyard with surrounding rectilinear rooms; McIntosh 2005: 154) and material culture found in the site. The temples were dedicated to the gods from the surrounding Levantine and Mesopotamian region, such as Ishtar, Dagan and Shamash; the statues of gods found reflect the surrounding region's artistic traditions. Material culture, such as seals, inlays, statues and architecture, is largely Mesopotamian or influenced by Southern Mesopotamia, building on traditions established before the second millennium BCE. Although the archives at Mari suggest that there were trade links between Crete and Central Asia, from whence luxury goods were obtained, the site, like Ur and despite its regional importance, mostly used local or nearby regional tradition in its most important and common art, material culture, religion and cultural influences.

#### 5.1.5 Dur-Untash

The Elamite cities, as stated in Chapter 2, were great rivals to Southern Mesopotamian cities. Susa became an important royal residence in the Achaemenid period, but long before this it was a chief capital within Elam. During the Middle Elamite period (ca. 1500–1100 BCE), Anshan became united with Susa, and Anshan reached a size of nearly 200 hectares (Sumner 1976; Carter and Deaver 1996). Additionally, in the fourteenth and thirteenth centuries BCE throughout the Near East, a new spate of royal cities, including Dur-Kurigalzu (Baqir 1946), Amarna (ancient Akhetaten, discussed below), Kar-Tikulti-Ninurta (Eickhoff 1985**)** and Elamite Dur-Untash, were built. Most of these cities were either abandoned or lost their significance after the death of the founding ruler. As much of Dur-Untash was built in a single period, the extensive remains allow one to see what much of the city was like.

Perhaps like the other royal cities in the Late Bronze Age, Dur-Untash ('the city of Untash', modern Choga Zanbil) may have served as a new power base established for the religious and political establishment during the reign of Untash-Napirisha (late fourteenth century BCE). Although the reasons for the construction of the city remain unclear, its remains show the large-scale establishment of a new city of over 100 hectares, at a time when the Elamite state was united and Elam was considered one of the great powers of the Near East (Carter and Stolper 1984: 37). After the reign of Untash-Napirisha, the royal capital appears to have moved back to Susa, which was continuously occupied until the seventh century BCE, when the Assyrians sacked it. The best-known structure in Dur-Untash is the largest known ziggurat in the ancient Near East, dedicated to the city's main deity Inshushinak (Figure 5.5). Features in the city include a large inner temple district surrounded by a wall, a royal quarter to the east of the ziggurat, and a large palace with burial chambers (Hypogeum Palace; Ghirshman 1966, 1968). What is evident at Dur-Untash is that it was a form of ceremonial city; this ceremonial city is distinct among Elamite cities; however, the architecture (e.g., ziggurat and temenos) and material culture do not show a syncretistic pattern or mixing with distant surrounding cultures. This city symbolic of Elamite power showed clear Elamite or Southern Mesopotamian cultural influences. In the Achaemenid period, a very different type of royal and ceremonial city developed, which began to import and incorporate various cultural elements and people. While it is hard to determine who may have occupied Dur-Untash, and whether the city was fully utilized as an urban centre or served strictly ceremonial functions, cultural elements generally resemble those found in the immediate surrounding area. In effect, as it was a symbol of the Elamite state, its cultural representations were mostly local.

**Figure 5.5** The ziggurat in Choga Zanbil (Dur-Untash), demonstrating localized architectural elements (after Nováková 2014)

## 5.1.6 Hattusha

In the second millennium BCE, the Hittites either emerged as a new ethnic group or developed from the local third-millennium BCE cultures. During the Late Bronze Age, Hattusha became a major city and was the most dominant in all of Anatolia by the late second millennium BCE. In fact, the growth of Boğazköy, the modern name of the site, is not only remarkably evident but also was rapid during the mid- and late second millennium BCE. The site may have spanned 180 hectares during the New Kingdom phase, with a lower city and an administrative acropolis, known today as Büyükkale, separated from the rest of the city by a wall (Bittel 1970; Neve 1996). As might be expected, the scale of palaces and living areas expanded as the success of the Hittite state increased. This, in part, can be attributed to the fact that more goods and resources could now be brought to the capital than in earlier periods, when the site was far smaller. Despite the success of the Hittites, major cities did not expand beyond the 400/500 hectare limit seen in Uruk and Babylon during the third and second millennia BCE. Sites such as Boğazköy and other major capitals in the pre-AoE were not able to surpass this limit even though they often expanded rapidly in periods of social, political and economic growth. Near Hattusha is found the significant Hittite shrine of Yazılıkaya, which famously depicts a procession of more than 90 deities and beings that represent Hittite, Hattic, Hurrian and Mesopotamian gods and figures (Seeher 2011; Figure 5.6). The shrine shows that the Hittites syncretized their neighbouring cultures' gods with their own beliefs. Even this

**Figure 5.6** Relief of the storm god Teshub and goddess Hebat, who are of Hurrian origin, at Yazılıkaya (after Gagnon 2014)

syncretism, however, was regionally limited compared with that which occurred in the AoE, as the figures displayed at Yazılıkaya were mostly those found in Anatolia, Syria and Mesopotamia.

## 5.1.7 Hazor and Southern Levant cities

During the Middle Bronze Age, 2000–1550 BCE, the Southern Levant witnessed a great urban expansion during which several large rival towns and cities emerged. In this city-states period, well-fortified sites with monumental gates appeared (Figure 5.7). Among the largest sites was Hazor, of nearly 80 hectares including its lower town and a monumental city wall, a moat, a revetment and a gate system (Yadin, Aharoni, Amiran *et al*. 1989; Ussishkin 1992). Like other major cities, Hazor developed important temple and palace complexes in the Middle Bronze Age. Throughout the Middle Bronze Age, sites including Tel Dan, Megiddo, Gezer and Shechem acquired large city walls with glacis and moats, presumably as intense city-state competition and local warfare developed (Burke 2008). Such urban patterns emerged during a period of conflicts between small states, during which large centres, often found near each other, became common. Not only were these cities fortified, but also their walls became symbols of their strength. Urban characteristics reflect the defensive traits spawned by conflict and competition between neighbouring small states. Cities such as Hazor, in their size, material culture

**Figure 5.7** City gate from Tel Dan's Middle Bronze Age (after Nimi 2011)

and display of fortifications, reflected the fragmented nature of politics in the period, which was similar to that of other Near East regions during the Middle Bronze Age.

#### 5.1.8 Amarna

A good example of a newly established major capital in the Late Bronze Age period is Amarna, known as Akhetaten in ancient Egyptian, which became the chief seat of government of Akhenaten (ca. 1353–1334 BCE) and the centre of diplomatic correspondence between states during his time (Kemp 2013). The city was large and extended into several areas along the east bank of the Nile, covering over 380 hectares (Kemp and Garfi 1993; Lacovara 1997: 82). Amarna was largely abandoned shortly after Akhenaten's death, perhaps because he was viewed as a heretic for his focus on Aten or monolatristic worship. The main districts were the north city and the central city, which included important palaces, temples to Aten and houses, and the main suburbs to the south, where the private houses of important nobles were located (Kemp 2013). The outer parts of the city were marked by boundary stelae describing the city and its foundation by Akhenaten; tombs of the city's nobles were also located in these outer areas.

The city was a large urban area founded by a specific king, and because there was little activity at the site after its abandonment the preservation is good. While Amarna gives us an idea of urbanism in ancient Egypt because it is extensively preserved, it also has the characteristics of a monumental city associated with a particular ruler, and the city did not remain politically significant beyond the ruler's reign. This is similar to Dur-Untash, discussed earlier, and other Late Bronze Age cities. Newly established cities such as Amarna may have been attempts to unify kingdoms around new ideas or political agendas. Furthermore, while the diplomatic correspondence centred on the city may imply that foreign dignitaries would periodically visit or be based in the city, there is no evidence that a large number of foreigners migrated to Amarna or to any of the other newly built centres. The new religion established by Akhenaten, with its cult and worship focused on the god Aten, was intended solely for the Egyptians and not universal. In contrast, the later monotheistic and universal faiths were intended to incorporate all people: the intention was greater social and cultural unity across different ethnic groups (Montserrat 2000). The tombs and other material culture from the site suggest it was a city built for the local Egyptian elite. While the city of Amarna was a new city, it was distinctively Egyptian at a time when Egypt controlled vast areas outside of Egypt, including some

**Figure 5.8** Although Akhenaten introduced new religious ideas to Egypt, including representation of the Aten as in this example, representation, incorporation and display of foreign influences and foreigners were not common (after Ollermann 2008)

in Nubia and the Levant. Few attempts seem to have been made to integrate the religion, the art or other cultural elements of conquered areas within the larger empire (Figure 5.8). As an example, rulers in the other major Near East states during the Late Bronze Age and in regions administered by Egypt could not marry Egyptian princesses, even though such marriages may have helped with diplomacy (Robins 1993: 32). Although foreigners may have been well respected privately, and were even potential allies, royal display and propaganda did not show foreign populations as equals or with the high regard afforded to Egyptians (Kemp 2006). While foreign populations had already begun living in Egypt long before the construction of Amarna, reflections of their cultures had not become prominent in the common material culture of Egypt even by the New Kingdom period. This would change in the AoE, particularly as more foreigners began to live in Egypt and as they blended and integrated their ideas with those of the Egyptians.

## 5.2 AoE cities

Many of the larger cities in the AoE continued to have the characteristics apparent in the pre-AoE. These include large temple districts, large palaces, city walls, gates, manufacturing areas and large residential districts. However, some important changes happened during the AoE; these were driven, at least in part, by the population movement discussed in the previous chapter that made cities noticeably different from their pre-AoE predecessors. Evidence of greater wealth and displays of power became more evident in the AoE. Not only were there much larger chief cities with far larger monumental structures and districts, but also large neighbourhoods of foreign populations began to be found. The records show that towns and cities had multiple temples dedicated to gods from distant regions. A large number of languages were spoken within cities, even as common languages developed to facilitate communication between populations. Material culture reflected not only the influence of local cultures but also that of much more distant cultures. As populations began to mix, new cultural trends, which included syncretism in art and ideas, including knowledge and philosophy, emerged. Below are descriptions of some cities that demonstrate key changes from pre-AoE cities.

### 5.2.1 Kalhu, Dur-Sharrukin and Nineveh

In the ninth century BCE a new type of ceremonial and capital city emerged in Northern Mesopotamia (Figure 5.9). The first of this type was Kalhu/Calah, or modern Nimrud, where the main citadel mound has been extensively investigated by Western and Iraqi archaeologists. The city was approximately 360 hectares, the main citadel being about 20 hectares (Oates and Oates 2001). In Northern Mesopotamia in the pre-AoE, it was rare for cities to be much larger than 100 hectares. The Neo-Assyrian capital cities far surpassed this limit. Furthermore, beginning at Nimrud, a new level of wealth emerged. Vast quantities of ivory, probably the greatest amount in the ancient world from a single site, have been found in the city, while the palace reliefs from the site are world-renowned. This wealth reflected the ability of the Neo-Assyrian Empire to exact or receive tribute from distant regions, that wealth being sent to the royal capitals. The royal treasures of the Assyrian queens have also been found; their splendid tombs represent a level of wealth previously unseen in royal graves in Northern Mesopotamia. These treasures indicate the reach of royal power that brought such wealth to the capital from distant regions. Furthermore, the Assyrians developed the skills to manufacture some of these luxury objects, as skills and workers from conquered territories were acquired and brought to their capitals (Oates and Oates 2001; Herrmann, Coffey and Laidlaw 2004; Hussein, Altaweel and Gibson 2016).

**Figure 5.9** The Assyrian royal cities of (a) Nimrud, (b) Dur-Sharrukin and (c) Nineveh. Temple, palaces and arsenals indicated (Kertai 2015; after Zunkir 2015a, 2015b; Fredarch 2016)

This pattern of wealth and grandeur continued with the later royal Assyrian cities of Dur-Sharrukin and Nineveh. Although Dur-Sharrukin was largely abandoned soon after its establishment, as Sargon II, the founder, was killed in battle, its wealth and position as a great capital are clear. The sheer size of the site, over 300 hectares, and major palaces and temples suggest a royal city that easily eclipsed most Bronze Age cities outside of Southern Mesopotamia (Loud and Altman 1938). The reliefs, such as the winged bulls (*lamassu*), from the site are among the largest Neo-Assyrian types. Dur-Sharrukin was eclipsed in its turn by Nineveh, which reached an unprecedented 800 hectares (Altaweel 2008). Within the city, Ashurbanipal created a royal library where scholars from Babylonia resided; the acquisition of scholarship from foreign lands, including Babylonia and Egypt, became a focus for Assyrian kings (Parpola 2007; Radner 2009). Workers, including artisans, from different areas of the empire became resident in the royal cities as they served in the construction and maintenance of some of the major monuments, including large irrigation projects and artworks (Oded 1979; Zaccagnini 1983). Although all the royal cities were very large, some of the space was taken up by new gardens that formed displays of the power and wealth of Assyrian royalty. The royal palaces, key media for wealth and power, were used to indicate Assyria's might to Assyrians and foreigners alike (Kertai 2015). The presence of arsenals in the royal cities also made the new Assyrian capitals important armouries and bases for the Assyrian army (Reade 2011). Direct and long-distance roads – 'royal roads' – were longer than Bronze Age roads and helped to connect regions to the Assyrian capitals. Movement to the Assyrian capitals from distant regions became direct, and probably more rapid, as the use of horses developed (Kessler 1997; Altaweel 2008; Radner 2014a). The key characteristics noticeable in Neo-Assyrian royal cities were their wealth, the presence of foreigners, including those brought to the cities, displays of power, the aggregation of knowledge, the use of long-distance roads and sizes that demonstrated a level that began to differentiate AoE cities from the pre-AoE. By expanding into regions far beyond their homeland, the Assyrians brought both physical objects and people to their royal cities, creating the conditions for the intermixing of populations and cultural ideas.

#### 5.2.2 Babylon

Babylon, which was already a great city by the second millennium BCE, perhaps as large as 500 hectares (Gibson 1972), reached nearly 1000 hectares during the Neo-Babylonian period, by the sixth century BCE (Figure 5.10). The city probably extended far beyond its city walls. Similarly to that of the Neo-Assyrian cities, the scale of the ceremonial, religious and palatial areas became far larger than in earlier periods. The temple of Ésagila and its enclosure alone, dedicated to the chief Babylonian god Marduk, occupy approximately 15 hectares. Babylon became the ceremonial, economic and political capital, reflecting not just its power but its central role in the Babylonian state and society (Koldewey 1914; Unger 1970; Jursa 2009; Seymour 2014: 9). Even after the fall of the Neo-Babylonian state, the city's importance continued for some time, until the Hellenistic period after the fourth century BCE, after it had served as one of the Achaemenid capitals. The presence of foreigners in Babylon and throughout Babylonia was already prominent in the Neo-Babylonian period, when Elamites, Egyptians, West Semites, Arabs and probably others from around the Neo-Babylonian Empire's territory, became evident in textual sources (Zadok 1979, 1981; Moukarzel 2014: 144). People either came to Babylon voluntarily or were brought forcibly. The exile of the Jews, mentioned in the Bible, brought another foreign element to Babylon, and much of this community remained in Iraq until the early 1950s CE. By the Achaemenid period, the Jewish community was

**Figure 5.10** Babylon's inner city indicating major structures and temples. The Greek theatre and the large temple of Ésagila are indicated (after Micro 2006)

thriving; it contributed to the rise of prominent banking and landholding corporations such as the Murashu, who were able to conduct business in various Babylonian cities (Stolper 1985).

As Babylon became very large, not only did it become an increasingly ethnically diverse city, but also the various cultural groups had opportunities to thrive socially. In the pre-AoE, Babylon incorporated foreigners such as Amorites and Kassites; however, in the AoE the diversity was probably greater or from more widespread regions, and there were opportunities for these groups to express their ethnic makeup. In the Seleucid period a Greek community was established, adding further ethnic diversity to the already diverse population. The presence in the city of a Greek theatre and gymnasium, among other structures, shows foreign and distant influences on Babylon (van der Spek 2009). Cultures expressed at Babylon did not simply reflect Babylonian elements, as they may have done in the Bronze Age, but the presence of various Near Eastern elements and, later, Greek elements began to be reflected in the city's architecture and material remains.

#### 5.2.3 Persepolis

The trend towards ceremonial capitals, seen in the Late Bronze Age, appears again in the Achaemenid period with the construction of Persepolis in the late sixth century BCE (Figure 5.11). While it is not clear how large the city was, several characteristics that contrast with those Bronze Age centres are evident. A remarkable aspect of Persepolis is the multiple iconographic and architectural elements incorporated within the central royal district and its key structures (Figure 5.12; Root 1979). Within 100 years of the collapse of the Neo-Assyrians, the multi-ethnic character of the Persian Empire had become evident, as we see in its remains today. Specific structures, such as the Apadana, demonstrate the incorporation of various populations which were paying homage to the Persian kings. Egyptian-style gateways, Hellenistic-style flowing robes and Assyrian-style winged human-headed bulls (*lamassu*) are among the artistic and architectural elements. In fact, Persepolis is not portrayed as having been founded only by Ahuramazda, the Persian god, but 'all' the gods are stated in the foundation inscription from the city to have participated (Schmidt 1953; Mousavi 2012; Babaie and Grigor 2015). It was not intended to be a city just for the Persians, but a place that represented the varied populations within the empire of the Achaemenids. Paradise, as envisioned by the Achaemenids, was embodied in the architecture and gardens of their royal cities (Boucharlat 2001). Included in this ideal were the multitudes and diverse populations found in their realm.

Persepolis began to represent the idea of universalism, in which people from different regions were symbolically united through the representation and presence of their gods in the metaphysical sense, but also in an earthly way through the architecture and art of the city. The architectural intent may have been to demonstrate a type of 'voluntary' subordination, as Khatchadourian (2016: 114) suggests, but the message was to display the diversity found in the city. This contrasted greatly with earlier Bronze Age and Iron Age cities and their iconography, in which the triumphant king was generally shown as being superior to his vanquished foes. The emphasis in pre-AoE cities was on the local, chief gods, while at Persepolis the inclusion of 'all' the gods represented the Achaemenids' different view, which incorporated others in their triumph rather than displaying them as victims. In Persepolis'

newgenrtpdf

reliefs, foreigners are not shown as inferior or vanquished but as individuals who supported and praised the Achaemenid king: they provide gifts to the court rather than having those items forcibly taken from them (Figure 5.13). Although the art at Persepolis certainly reflected official propaganda, where content foreigners came from different parts of the Achaemenid Empire (Dandamaev, Lukonin, Kohl and Dadson 2004: 293), the emphasis on inclusion of, rather than triumph over, foreigners indicates that the official message had begun to shift. Real policy implications became evident at places such as Persepolis. Foreigners are attested to have been based at Persepolis, including by texts from the Persepolis Fortification Archive (PFA). These foreigners included Arabs,

**Figure 5.12** Reliefs from Persepolis found in the Palace of Darius ((a) Kawiyati 2007) and the Gate of All Nations ((b) Farshied86 2006). Numbers 1–3 indicate Egyptian, Hellenistic and Assyrian influences

**Figure 5.13** Depiction in the Apadana of foreigners bringing wine to the Achaemenid court (Maiwald 2008)

Cappadocians, Indians, Babylonians, Bactrians, Egyptians and others who were civil servants or professionals who may have stayed temporarily, or lived permanently, in Persepolis and other royal cities, such as Susa. Languages in the PFA include Greek, Phrygian, Aramaic, Elamite, Persian and Babylonian (Stolper 1984; Dandamaev *et al*. 2004: 293). The population diversity at Persepolis may have been foreshadowed at Pasargadae under Cyrus, where foreign influences are evident. Evidence from Pasargadae suggests that foreigners, including Babylonians and Lydians, were incorporated into the population, and stylistic syncretism is evident in the art (Stronach 1997; Briant 2002: 77–8).

The very strategy of Achaemenid kingship, emphasizing the integration of foreign populations under the unifying power of the Achaemenids, was symbolized by Persepolis. Darius, Xerxes and some of their successors even use the title 'king of lands (or nations) containing all sorts of men' to show this diversity (G. Cameron 1973; Stolper 1984). At Persepolis, it is the foreign influences in the art and architecture and the incorporation of varied populations and their gods within the city that differentiate it from its pre-AoE predecessor ceremonial cities such as Dur-Untash and Amarna.

#### 5.2.4 Ctesiphon

At one time perhaps the largest city, or more accurately urban zone, anywhere, ancient Ctesiphon (Figure 5.14), about 35 km south of modern Baghdad's centre, served as one of the great capitals of the Parthian (Arsacid) and Sasanian states (ca. 247 BCE–651 CE). By the late Sasanian period, the cities in this urban zone were Aspanbur, Veh-Ardashir, Hanbu Shapur, Darzanidan, Veh Jondiu-Khosrow, Nawinabad and Kardakadh. However, it is likely that only four or five of these districts had large populations; some of the names may refer to the same place (Morony 2009; Davaran 2010: 59). Ctesiphon is best known for its famous archway, the largest freestanding vault until the last century, which is a remnant of a monumental Sasanian palace compound. The exact dimensions of the city are difficult to determine with certainty, and it is possible some of the site is missing because of erosion by the Tigris, but the city seems to have merged with Seleucia, the Seleucid capital, which was located nearby. Some of the districts and cities were created, in part, by deported populations; together they formed the area known as al-*Madāʾen* (Invernizzi 1976; Negro Ponzi 2005). This made Ctesiphon and its urban region part of a heavily populated urban zone that may have contained a population in the hundreds of thousands (Ṭabarī 1989). Although the walled area of

**Figure 5.14** Map of Ctesiphon and its urban region (after Lencer 2007; Negro Ponzi 2005: 167)

Ctesiphon is about 550 hectares, an estimated 1500 hectares is a reasonable estimate for the maximum extent of Ctesiphon and its urban environs. Given the effects of erosion, of the multiple urban districts, and of the site, or more accurately sites, not having been fully surveyed, this estimate is plausible (Lee 2006: 157).

Relevantly for demonstrating how such large cities came into being during the AoE, the population consisted of various ethnic groups, including Greeks, Persians, Jews, Assyrians, Arabs, Arameans, Babylonians, Syrians, Romans and probably others. In the Sasanian period, religions represented within the city included Judaism, Christianity and Zoroastrianism, while other cults existed, at various periods, of other gods that were associated with the various population groups (Ṭabarī 1989). The population therefore reflected the type of primate city the AoE helped to produce: it was a disproportionally large population made up of various ethnic groups, some of which had migrated or been brought to Southern Mesopotamia from distant regions; they included people who had arrived during earlier periods or in the lifespan of the city. Religion in the city represented the wide diversity of the population rather than just the local or regional beliefs. Large-scale manufacturing, dependent on foreign products from more distant parts, was increasingly important in the Sasanian period to large cities, where glass making and other production thrived, as it became possible to obtain resources from distant regions (Simpson 2014: 204). Great wealth flowed to Ctesiphon through long-distance trade, and the position of the city on the Silk Road routes allowed products from China and Europe to come to the city (Wagstaff 1985). Access to the Tigris and canals would have enabled it to benefit from seaborne trade from the Arabian Sea. Long-distance connections, easy movement and connections to international trade helped the surrounding countryside thrive economically and increase in population and population diversity, while the urban region itself developed into a major political and economic centre.

#### 5.2.5 Antioch

One of the great cities founded at the end of the fourth century BCE was Antioch on the Orontes (Figure 5.15). The city was established in a Hellenistic grid layout by Seleucus I. Much of the city is now underneath modern buildings or buried by sediment; however, it has been partially reconstructed from ruins and from historical texts. From its beginning, the city had a diverse population composed of people from the surrounding region in the Northern Levant, but also of Jews, Macedonians and Greeks (Malalas 1986). Antioch appears to have been founded as one city in a tetrapolis of Seleucid cities in Syria, the others being Laodicea, Apamea and Seleucia Pieria. It became a capital in the Seleucid period. By the Roman period, it dominated the Eastern Mediterranean coast

**Figure 5.15** Conjectural representation of Antioch (after Cristiano64 2010; Downey 1974: Fig. 11)

economically and culturally, far surpassing its nearby rivals in size and economic weight (Sandwell and Huskinson 2004).

Although the exact population is unknown, it is clear that the city was very large and had a diverse population in the Roman period. A reasonable estimate of the population is in the order of several hundred thousand; the city became one of the primate cities that greatly surpassed other Mediterranean cities (De Giorgi 2016: 180). The area of the city may have been only about 200–300 hectares during the Seleucid period, but it was far larger in the Roman/Byzantine period (Aperghis 2004: 93; Cohen 2006: 93). This growth probably contributed to its rise as an early seat of Christianity and a major centre for Judaism, which probably further diversified the already diverse population. Within the Christian community in the city, for instance, were missions from Armenia, Greece and Latin-speaking regions. The universal faiths began to use large and diverse cities such as Antioch as new bases, even though those cities had little to do with the origins of those faiths. Additionally, many temples to Greco-Roman gods, including Jupiter and Artemis, were found (Downey 2015). Large cities such as Antioch had influence that stretched over three continents. In the Roman period, although primary texts from Antioch itself are scarce, texts from other regions indicate the existence of individuals who identified themselves as having come from or lived in Antioch and its region. These include people from North Africa, southeast Europe and the Near East (De Giorgi 2016: 175). People were migrating to and emigrating from the city across many regions, and commerce from other cities throughout the Mediterranean and elsewhere became directly linked to the city.

#### 5.2.6 Alexandria

The best-known city founded by Alexander after his conquest of Egypt ca. 331 BCE is Alexandria. While the city's Jewish, Greek and Egyptian populations are well known, during the Ptolemaic period Syrians, Medes, Persians and other Asian populations also lived in the city. By the Roman period, if not earlier, various populations from different parts of Europe were intermixed with the already cosmopolitan population (Vrettos 2001: 7). In the first century BCE, the city was perhaps the second largest in the Roman Empire and served, through its great harbours, as the commercial entrepôt for the Eastern Mediterranean (Strabo 1967: book 17.1.31; Haas 1993: 234). Throughout its history in the AoE, Alexandria was an astounding mix of cultural ideas, ethnic groups and religions (Hinge and Krasilnikoff 2009). The intermixture of so many population groups made the city not only cosmopolitan but also a great example of how universalism transformed the urban makeup of primate centres in which a variety of cultural expressions and syncretism in art and ideas were found together. Whereas in the pre-AoE Egyptian thought and culture dominated, as in Amarna, in the AoE the cultural landscape became much more varied, even in respect of common material culture. Syncretism is expressed through the variety of artistic, theological, philosophical and religious ideas prevalent in the city, such as the worship of the Greco-Egyptian god Serapis (Figure 5.16) or the philosophy of Philo. Greco-Roman and Egyptian art and architecture commonly became fused (Vrettos 2001; McKenzie 2010). Such variety in ideas and material culture reflected the mixtures of cultures that were prevalent and the fact that they were able to intermix freely as they resided together.

Although it is not well preserved today and has been built over in many places by the modern city, our knowledge about Alexandria has been preserved in historical works. There were several unique structures during the history of this city, such as the lighthouse in Pharus. One of the best-known structures was the library of Alexandria, which functioned as part of Alexandria's *Musaeum*, an institution devoted to scholarly activity (Stephens 2010). The library epitomized the spread of knowledge

**Figure 5.16** The god Serapis (above), a syncretized Greco-Egyptian god, was worshipped in the Serapaeum, or temple to Serapis, at Alexandria (Nguyen 2009)

and information as travel and movement encompassed greater distances and became more direct during the AoE. Galen speaks of ships having to unload their written works to the library for copying. From what scholars can reconstruct, the library contained not only Greek and Egyptian knowledge but also knowledge originating from Babylonia. Although earlier cities, such as Nineveh, had established libraries, to which knowledge and scholars were brought from different regions, the knowledge held at Alexandria originated from even more diverse places and came to be collected in a central repository (MacLeod 2004; Potts 2004a; Barnes 2004). Alexandria's library showed that knowledge and learning became more mobile in the AoE.

#### 5.2.7 Dura Europos

The town of Dura Europos (Figure 5.17), which has a Hellenistic-style grid layout, is found along the Euphrates in southeast Syria near the border with Iraq. The town was founded ca. 300 BCE and lasted until ca. 256/257 CE, the year in which it was destroyed by Shapur I (Matheson 1982). The site is small in comparison with the larger cities of the AoE, such as Alexandria and Antioch, as the city walls enclosed an area of only 75 hectares. At this time, as stated earlier, many of the great cities of the region were to be found along coastal areas or along major waterways, particularly in Southern Mesopotamia. Nonetheless, Dura Europos had many of the characteristics of a cosmopolitan city similar to the major urban centres that became more international. It contained places of worship for Jewish, Christian and polytheistic religions originating from Greco-Roman, Near Eastern and Indo-Aryan regions. Places of worship also contained temples dedicated to syncretized Greco-Near Eastern gods. The languages spoken and written in Dura Europos during the Roman period reflected the ethnic diversity found in the town: they included Aramaic (including Palmyrenean, Hatrean and Syriac), Hebrew, Parthian, Persian, Arabic, Greek and Latin (Kaizer 2009: 235). The famous art known from the town, including tempera wall paintings in the well-known synagogue (Figure 5.18) and church, indicates a mixture of local Semitic/Near Eastern and Greco-Roman stylistic influences, including dress and iconographic symbols from these varied cultures (Perkins 1973; J. Baird 2014). In short, the mixture of languages, cultural symbolism, religions and art styles reflects how people and ideas from distant regions came to characterize smaller towns such as Dura Europos and not simply large cities.

**Figure 5.17** Site plan of Dura Europos showing areas excavated (shaded). Areas uncovered include important religious structures from various religions and dedicated to Christian, Jewish, Roman, Near Eastern, Indo-Aryan and syncretized Greco-Near Eastern gods (after Marsyas 2016a; Gelin 1997)

## 5.3 Spaces in between: the ruralization of the countryside

Although most of this chapter focuses on major urban centres from the pre-AoE and the AoE, another transformation may have affected smallscale sites, that is, those sites that are less than a few hectares. Chapter 4 showed that in some areas new, dispersed and small sites were increasingly found in places where in the pre-AoE there would have been larger

**Figure 5.18**Examples of tempera wall paintings from the synagogue found at Dura Europos. Scenes a–d are: (a) from the Book of Esther (Duraeuropa 2016; (b) Moses being pulled from the Nile (Becklectic 2016a); (c) David anointed by Samuel (Marsyas 2016b); (d) the Exodus (Becklectic 2016b) sites, and a greater diversity of different-sized sites, including secondary and tertiary towns in a relatively small area. More proportional representation of varied settlement sizes was evident for different periods within the pre-AoE according to the rank-size curves shown in that chapter, including the Khabur Triangle region. There we also explained that small settlements were likely to become prevalent as movement became easier. The examples demonstrated that as populations were concentrated into fewer larger cities, some areas developed a greater proportion of small sites that had little area differentiation from each other. While ease of movement may explain how such patterns emerged in the AoE, it does not explain what exactly these small settlements were during the AoE.

As most archaeological excavations focus on larger sites in many periods, very small sites are often neglected or poorly understood. Archaeological surveys, although they often document small sites, generally do not adequately explain their functionality. Where there have been excavations or other investigations of small sites, large and isolated structures with relatively wealthy finds have been evident. This suggests that at least some of these sites may have been more than simple farming villages, hamlets or fortifications.

Examples of such small sites are Tell Boueid (Al-Maqdissi 1995) and Bir el-Haddad (Rouault and Masetti-Rouault 2014), both in eastern Syria, Tell es-Sa'idiyeh (Pritchard 1985) in the Southern Levant, and the 'palazzetto' at Tell Mardikh in western Syria (Mazzoni 1990; Figure 5.19a–d). Other sites, such as Khirbet al-Qasr (Altaweel 2006: 164–5) in the northern Jazirah of Iraq, are sub-hectare, singleperiod occupations that appear to have been newly established in the Iron Age or later (Figure 5.19e). Very small, sub-hectare sites such as Khirbet al-Qasr are often only noticed in areas surveyed intensively. Many other small or sub-hectare sites do not have any easily noticeable mounding, which makes them nearly invisible to archaeologists. The excavated sites mentioned above (Figure 5.19a–d) have isolated buildings that generally do not appear to abut or incorporate other buildings, although Bir el-Haddad's structure seems to abut a long wall that may be contemporary with the structure itself. The sites' structures are large, or larger than common houses, and have central courtyards. There is evidence of expensive goods (for example at Tell Boueid), such as wellmade incense burners and stone figurines. They could, in some cases, be fortifications, but evidence of common burials (e.g., at Tell Boueid) and a lack of large, wide walls and military installations (e.g., at Bir el-Haddad and Tell Mardikh) suggest they were more probably residential, civilian or administrative in nature. Administrative devices are not found in large

**Figure 5.19** Some examples of villas or large residences. These include (a) Tell Boueid (after Al-Maqdissi 1995: Fig. 8), (b) Bir el-Haddad (after Rouault and Masetti-Rouault 2014: Fig. 8), (c) Tell es-Sa'idiyeh (after Pritchard 1985: Fig. 185), (d) Tell Mardikh '*palazzetto*' (after Mazzoni 1990: Fig. 2) and (e) Khirbet al-Qasr (circled; after Altaweel 2006: Fig. 12). Figures are all reprinted with permission

quantities, although Bir el-Haddad had a cuneiform administrative text. The sites mentioned here all date to the late Neo-Assyrian and/or the Achaemenid period (that is, to about the eighth to fourth centuries BCE). Other structures, similar to those described above, often relatively isolated farmsteads, have been found from between the eighth and second centuries BCE, such as Tirat Yehuda, in the Levant (Faust 2006).

Historical data may support the idea that at least some parts of the landscape had villas or wealthy estates by at least the Neo-Assyrian and Achaemenid periods. In the Neo-Assyrian period, texts indicate royal land grants and estates given to individuals as a reward for their service to the Assyrian state (Fales 1990). The Assyrian/Akkadian word *kapru* is used, which suggests something comparable to a farmstead or large country villa. In the Achaemenid period, land tenure texts indicate that the royal family and the nobility owned large estates or wealthy agricultural holdings in some of the interior regions of the Near East, which larger Bronze Age centres once occupied (Sartre 1989). Types of agricultural estates similar to those of the Neo-Assyrian period seem to have continued into the Achaemenid period.

The settlement surveys referred to in Chapter 4, along with textual sources, suggest that some small sites could have developed into wealthy farming estates by the early AoE. Small sites show that as the larger settlements of the interior of the Near East were abandoned, for example in eastern Syria (Akkermans and Schwartz 2003: 391), the Jazirah in Iraq and parts of the Levant, small-scale settlements became common. This may mean that the countryside was transforming into a region where reduced violence and increased socio-political integration facilitated the rise of country estates for the wealthy. While, admittedly, this is an under-researched area in Near Eastern archaeology, such a process would be similar to that observed for the Roman Empire, where *villa rustica* sites, or country villas, appeared in areas of greater safety and integration into the Roman economy and political system (Garnsey and Saller 2014: 221). Using Roman villas in Gaul as examples, we show that these types of settlements became associated with the export of agricultural products to urban regions as the presence of Rome became more pronounced (King 1990). Villas in the Roman Empire may have been acquired as 'rewarded' estates given by the central government, in a process similar to that seen in the Neo-Assyrian and Achaemenid periods (Roymans 2011). In the case of Rome, the giving or awarding of land to foreign, non-Roman troops may have helped to Romanize the empire. Similar reasons for rewarding officials or military personnel may have occurred for the AoE states, where rewarded land may have been used to create greater loyalty to the central state, including from individuals from different ethnic or social backgrounds who served the state. This does, in fact, seem to be the case in the Neo-Assyrian and Achaemenid periods. Villas in the Near East may have become a type of settlement that reflected increased economic or even political integration of the countryside with the larger states in the AoE.

There is further archaeological and historical evidence of villas in different parts of the Near East and Egypt in Late Antiquity. In Egypt, texts from Oxyrhynchus demonstrate that the town had wealthy, large estates nearby that were operated by the Apion family, who leased land for profit and whose business activities resembled in many ways those of the Murashu family in the Achaemenid period (Sarris 2009: 85). Archaeological remains of villa architecture have also been found in the region of Caesarea, where the production of wine or oil would have been important to the local economy (Hirschfeld 1997: 46). Similar well-built and wealthy examples appear to have existed in northern Syria and in the region of Antioch at about the same time (Sarris 2009). This is not unexpected, as the Roman and Byzantine periods are well known for such remains. However, these estates, in the Near East, resemble structures that were found earlier in the AoE; the Roman or Late Antiquity villas were also large, isolated in cases, and indicated greater relative wealth. The model of large country estates run by wealthy families was an important economic component when the conditions of large empires permitted their widespread existence. Such settlements could spread in areas of the countryside that became pacified, and migration to larger cities may have depopulated, or at least deurbanized, some of the older settlements and regions, opening up more countryside for new owners or types of settlements. As larger cities developed in some regions, demand for agricultural goods would have required the rural regions to produce a greater supply for the more distant cities. In summary, the *villa rustica* model prevalent in Europe in the Roman period may be applicable to how Near Eastern rural places functioned early in the AoE, through their economic contribution and type of settlement, as larger empires emerged in the Neo-Assyrian period.

## 5.4 Conclusion

Contrasts between large urban centres in the pre-AoE and the AoE are evident. The largest sites in the pre-AoE became even larger in the AoE, particularly along or near the Mediterranean shore and major rivers such as the Tigris and the Euphrates. Whereas Uruk and Babylon were perhaps the largest pre-AoE cities, at 400–500 hectares, in the AoE the largest cities reached 1000–1500 hectares, Babylon, Antioch, Alexandria and the Ctesiphon conurbation being among the largest. Cities along international trade routes on the Mediterranean, the Tigris and the Euphrates now reached far greater sizes than in the pre-AoE.

However, it is not just size that differentiated AoE cities from their pre-AoE predecessors. This chapter demonstrates that in urban centres in the AoE, populations became far more international, not only coming from distant regions, as shown by their ethnicity, but also expressing their diversity through their religions, languages, art and ideas. This is what would be expected if movement had become a major driver of population shifts for cities. Neo-Assyrian, Neo-Babylonian and Achaemenid cities demonstrate that diverse populations had already characterized various cities in the Near East before the arrival of more Greek populations after Alexander's conquests. While natural population increase could accelerate the growth of some urban areas, it was the arrival of new populations from different areas that drove growth in many of the AoE's larger cities. Often it is not clear when such populations arrived, as some foreign populations may have migrated earlier than they were mentioned in texts, but for our purposes evidence shows that it became more common for people from geographically distributed origins to move or be moved to cities. Although migration is evident in the pre-AoE, for example the migration of Amorites or Kassites in the Near East, it is the scale and geographic spread that differentiates the AoE movements from earlier periods. Additionally, cultural expression, for example through architecture, art and religion, became more diversified and accepted: Greek, Egyptian, Roman, Indian and Near Eastern influences were found in major Near East cities in different periods. In contrast, major Bronze Age cities were more localized in architectural, artistic and religious expression, and imports of ideas and material goods were evident mostly in luxury objects. The major centres in the pre-AoE did not display foreign cultural influences as prominently as the AoE cities.

Antioch in the AoE had individuals from afar come to it, but, as well, people from Antioch began to spread and were found in a wider area, which reflects the city's influence. Knowledge became mobile, Alexandria's library probably collecting knowledge from Greek, Egyptian and Babylonian cultures. Religious worship became more diverse in the AoE, even in smaller cities such as Dura Europos. In contrast, major cities in the pre-AoE, such as Mari, appear to have more local or regional gods. Art influences in paintings, and new street or urban patterns, from Greece are introduced in the AoE in more parts of the Near East, as seen in Dur Europos, Antioch and Alexandria. Resources from distant regions made possible manufacturing areas, such as the ones at Ctesiphon, that created new types of goods, including types of glass, while international trade routes through AoE cities connected eastern Asia and Europe. Greater wealth from more distant areas was brought into large capitals in the Neo-Assyrian period, as demonstrated at Nineveh and Kalhu, where some of this wealth was extracted by force.

At Persepolis, the claim that 'all the gods' were important to the foundation of the city, rather than just the national or patron god of the Achaemenids, is evident. The ceremonial capital of the Achaemenids showed itself as giving a stake in the city to different populations through their representative gods. Foreigners from various places within the empire and beyond came to the city and were employed for their labour. The city itself developed architectural and artistic styles that integrated elements from various parts of the empire. Foreigners are shown bringing tribute and not simply as vanquished foes. Although forced migrations existed in the Achaemenid Empire, large population movements may have become increasingly voluntary as individuals recognized opportunities. Along with art, architecture, religions and diverse population groups, the foundation of new cities such as Alexandria, Antioch and Persepolis shows that blended cultures and various types of syncretism had become the norm. In the pre-AoE, the foundation of ceremonial cities such as Dur-Untash or Amarna glorified local chief gods, and local art displayed their greatness. In the AoE, the evident diversity and scale of change reflected population movement and influences from distant places that began to transform the social makeup and characteristics of cities in the wider Near East; multiple cultural groups now found expression as part of a larger whole.

The phenomenon of villas or large estates that developed in the countryside may be another factor that demonstrates increased movement and the socio-political and economic integration of the countryside. Although some of the cities of the pre-AoE became depopulated during the AoE, that urban landscape was being replaced with small sites in places. Where some small AoE sites have been excavated, structures that resemble Roman villas have emerged. It is possible that these types of compounds became of interest when the countryside became more pacified or perhaps more integrated into the economy and politics of the larger states in the AoE, similarly to regions such as Gaul in the Roman Empire. There are historical references to agriculturally based estates or farmsteads in the Neo-Assyrian and Achaemenid periods that seem to resemble country villas, as parts of the countryside were owned by wealthy individuals and those obtaining land gifts from royalty. The transition to small, villa-like sites in parts of the Near East, just as very large, primate cities began to emerge, shows that larger states and empires became the political norm.

## 6 Long-distance trade and economy before and during the age of empires

This chapter explores how long-distance trade was affected during the AoE. Similarly to earlier chapters, this is done by comparing patterns before and during the AoE. We use trade patterns based on the distribution of specific trade goods for different periods. The key result demonstrated is how trade networks enlarged under the AoE to an extent that was less possible during the Bronze and Early Iron Ages, as empires enabled easier access to goods and facilitated faster movement over longer distances. Thus, as a result of large-scale empires, the movement of people to cities increased and long-distance trade itself began to transform as the movement of goods became easier.

## 6.1 Long-distance trade in the pre-AoE

Both textual and archaeological evidence indicate that long-distance trade networks were a fundamental component of pre-AoE economies in the Near East. Various communities and city-states of Central Asia, the Indus Valley, Iran, South Arabia, Mesopotamia, the Levant and the Eastern Mediterranean were interconnected in the exchange of both finished objects and raw materials (T. C. Wilkinson 2014). In most cases, finished objects included highly prized items, such as beads, amulets and stone vessels, used by wealthy individuals and elites who boasted of their status in funerary or royal contexts; precious metals such as gold also travelled long distances to meet the demand of the Near Eastern elite. More utilitarian metals, such as copper and tin, were exchanged across a wide area and were essential for the development of bronze items (Figure 6.1). Bronze Age long-distance trade also permitted the development of skilled craftsmanship, which facilitated the production of

**Figure 6.1** Map showing the main sites and regions involved in Bronze Age long-distance trade, and the materials exchanged, in their regions of origin finished items, as witnessed by several artisanal quarters established in cities across the Near East (Steel 2013: 157–90).

In the third millennium BCE, semi-precious stones were among the most common items traded across long distances. These were lapis, carnelian, chlorite and steatite, whose geological sources were located in an area encompassing Afghanistan, eastern Iran and the Indus Valley (Moorey 1994: 77–103). These stones were traded as raw materials or were used to make beads, seals, amulets and small vessels which were distributed across Central Asia, Iran, Mesopotamia, the Levant, Egypt and, more sporadically, the Balkans (T. C. Wilkinson 2014: 125–37, 262–3, 282–3). In many cases, such objects were buried in royal tombs, the best example being the Royal Cemetery of Ur (Woolley 1934). Various commercial routes connected distant areas, some overland across Iran, and others via the Persian Gulf, which connected Arabia and the Indus region (Tosi 1974); as is often the case, trade contacts also entailed the transmission of ideas on styles. The spread of the so-called intercultural style chlorite vessels from Central Asia to Mesopotamia and the Gulf (Amiet 1986) is a good example of such a transmission of design ideas (Figure 6.2). Another example comes from double spiral-headed pins (Huot 2009), which are possible indicators of similar clothing styles spreading from Central Asia to Anatolia and

**Figure 6.2** Chlorite vessel of the so-called intercultural style showing a musical procession. Found at Bismaya (ancient Adab, Southern Mesopotamia) and dating to the Early Dynastic period (2700–2500 BCE; Oriental Institute Museum, University of Chicago; Daderot 2014)

the Balkans during the second millennium BCE and indicating that large trade networks had spanned these regions.

The Arabian Peninsula was included in this long-distance trade network, copper being particularly prized. According to Sumerian texts from the third millennium BCE, copper was sourced from Magan (corresponding to modern Oman), but in the early second millennium BCE other sources located in Northern Mesopotamia and Anatolia began to be exploited, and continued to be so at least until the eighteenth century BCE, when Cyprus became the main source of this metal for most of the Near East (Moorey 1994: 245–6). Copper was essential to meet the high demand of Mesopotamian and Syrian palaces for the production of bronze tools and weapons (Weeks 2004). In the early second millennium BCE, textual evidence from Kültepe (ancient Kanesh) in Central Anatolia informs us about donkey caravans departing from Ashur in Northern Mesopotamia, loaded with textiles from Babylon and tin (probably from Iran), that eventually reached Kanesh, where these products were exchanged for silver and gold (Veenhof 1969; Larsen 2015). Another important source of information about Bronze Age long-distance trade is the fourteenth-century BCE archive of Amarna, Egypt, whose tablets, written in Akkadian, cast light on an intricate gift-exchange network involving Egypt, the Aegean, the Levant and Mesopotamia, where precious metals and various objects bestowed with high value were exchanged (Cochavi-Rainey and Lilyquist 1999). Another product traded across long distances was amber, which was sourced in the Baltic area and reached the Levant through the Aegean, as witnessed by small amber objects found at Thebes in Tutankhamun's tomb and in the Royal Tomb of Qatna (Syria) in the fourteenth century BCE (Mukherjee, Roßberger, James *et al*. 2008). On the other hand, from what is observable so far, amber does not appear to reach areas east of the Levant, that is, Mesopotamia and Central Asia. Finally, relics found off the Eastern Mediterranean coast, for example near Uluburun and Cape Gelidonya, yielded a high quantity of ceramics, metal ingots and other luxury objects which were exchanged among different political entities of the Levant and the Eastern Mediterranean (Bass 1967; Pulak 1998).

The Bronze Age long-distance network encompassed an area stretching from Central Asia to the Eastern Mediterranean, including the Indus Valley and South Arabia, but beyond these limits trade for items was far more sporadic, and few items are found outside these areas. It is difficult to say how such a network was controlled or functioned, although it probably consisted of a set of overlapping trade routes that various parties organically developed, maintained and participated in because they had some stake in or benefit from the trade. Royal palaces and temples appear to have had a fundamental role in long-distance trade, in that they organized expeditions of emissaries or traders to foreign lands, thus framing the trade into political and diplomatic relations (Lipiński 1979). However, such evidence might be biased, since the archives of palaces and temples are the major sources of recorded information known to us and represent a limited view. The written evidence from Kanesh, on the contrary, indicates the existence of business and trade controlled by private families, with state intervention limited to the collection of taxes and participation through private families (Michel 2001). If we can assume that the Kanesh trade network was a typical example of long-distance trade in the Bronze Age, then such trade appears to have been a composite of both public and private spheres working together or independently. The Kanesh archive also informs us of the existence of an enclave of Assyrian merchants living in this city and managing their businesses there (Larsen 2015); it is possible that this was not an isolated case in the Bronze Age and that other communities of merchants lived far from their place of origin. One example comes from Late Bronze Age Ugarit, where tablets inform us of merchants from the Levantine coast, namely from Arwad, Byblos, Beirut, Tyre, Akko and Ashdod, who were stationed in Ugarit to run their businesses (Wachsmann 2009: 40).

It is noteworthy that the Bronze Age long-distance trade network depended heavily on the changing political landscape. In many instances, political turmoil, dynastic squabbles, conquests and population movements affecting one end of the network caused much of the trade system to break down or be disrupted. For example, at the end of the third millennium and the beginning of the second millennium BCE, trade connections between Mesopotamia, Central Asia and the Gulf appear to be drastically reduced; the causes of this phenomenon seem to vary, but the collapse of the complex societies in the Indus may have played a role (Moorey 1994: 245–6). The Hittite conquests of the late eighteenth century BCE possibly brought about the end of the Kanesh– Ashur trade system; the arrival of the Sea Peoples on the Levantine coasts at the start of the Iron Age (about 1200 BCE) coupled with the weakening and disruption of most Levantine palace economies caused the end of the Late Bronze Age political and long-distance trade system (Yasur-Landau 2010: 102; Van De Mieroop 2016). It seems that the Bronze Age long-distance and intercultural trade was far-reaching, but it was also heavily affected by political changes that occurred cyclically in the pre-AoE. Though this observation may be biased by the nature of our sources, which are often sporadic, the breakdown of long-distance trade suggests that these systems were fragile, and that political fragmentation could affect the trade networks that were active in the Bronze Age. What this demonstrates is that trade systems were dependent, to a great extent, on larger political systems. When political systems were relatively stable, as they were in periods of the Late Bronze Age, we see trade thriving. On the other hand, with so many different political entities, a major change in one area could have far-reaching repercussions for the overall trade system.

## 6.2 Long-distance trade during the AoE

During the AoE, long-distance trade changed, reflecting greater and more far-reaching movement. Some developments played decisive roles, such as the domestication of the camel, which opened up new routes across the Arabian desert (Sapir-Hen and Ben-Yosef 2013); similarly, improvements in astronomy, perhaps made by the Babylonians, and in navigation techniques, often attributed to the Phoenicians, in the first millennium BCE greatly improved long-distance and open-sea navigation (Wachsmann 2009: 299–300). Evidence of advanced technology, which could be used to improve navigation, comes from the so-called Antikythera mechanism dated to the third or second century BCE, which was a mechanical device that could predict astronomical positions (Carman and Evans 2014). Furthermore, the discovery of the monsoon wind in the Hellenistic period contributed to the intensification of maritime contacts with India (McLaughlin 2010: 41). Beyond the means of transport, an important innovation in the AoE was the introduction of coinage for trade transactions, which made them quicker and safer and meant that they were backed by government institutions. Although these innovations had a fundamental role in shaping long-distance trade during the AoE, large empires also, in our opinion, permitted the establishment of wider, longerlasting and faster connections between distant places than in previous eras. With fewer political entities over long distances, and hence fewer political actors, there were fewer political disruptions, and so transactions became simpler, at least politically. The movement of products to more distant locations and at greater intensities suggests that movement had become easier in general. Such trade acts as a possible proxy, along with the urban patterns discussed in Chapter 4, for the fact that populations could move more easily to more distant regions. To show how movement of products changed during the AoE, our analysis will focus on the trade of frankincense, myrrh and pepper, and coin distribution.

## 6.2.1 The frankincense and myrrh trade

Frankincense (*Boswellia*) and myrrh (*Commiphora*) trees grow only in Oman and Yemen and on the Somalian coasts. Their resins, if burnt, generate intense aromas, which were appreciated in many cultures for cultic and funerary rituals; they were also used for domestic, cosmetic and medical purposes (Groom 2002; C. Singer 2007: 6–8). In the Bronze Age, the frankincense used in Egypt was sourced from a region the Egyptians called the 'Land of Punt', whose identity is not clear, but it may have been northern Somalia (C. Singer 2007: 5). It is, however, only in the Iron Age, when camels became omnipresent in records, that caravans from Arabia loaded with frankincense and myrrh reached the Levant, where the merchants sold these products to a wide clientele. Assyrian written sources first mention Arabian caravans active in Syria, in the area of Damascus, during the reign of Shalmaneser III (858–824 BCE), and such activities became increasingly recorded in the Assyrian annals (Byrne 2003: 12; Potts 2010: 128–9; Zadok 1981).

Frankincense and myrrh were burnt in burners, also called censers, made of different materials, including metal, pottery, limestone and chalk (Figure 6.3); during the Late Iron Age and Achaemenid periods, such burners can be found in Arabia, the Southern Levant, Persia and Mesopotamia (Shea 1983; Millard 1984; Invernizzi 1997a; Hassell 2005). The Nabateans played a crucial role in controlling the aromatics trade from Arabia to the Levant, in particular between the first and third centuries CE (Erickson-Gini and Israel 2013). Some evidence suggests that by the Achaemenid period frankincense may have reached

**Figure 6.3** Relief from the 'Treasury' at Persepolis. The Great King Darius I (ca. 550–486 BCE) is shown on the throne with two incense burners (circled) on tall stands before him (after Davey 2010)

Central Asia, as indicated by textiles from the Pazyryk burials in the Altai Mountains between Khazakhstan and Mongolia, on which Persian-style censers are represented (Rubinson 1990).

In the first century CE, during the Parthian era, some censers were also found among the grave goods of Tappeh Hemat Salaleh in Azerbaijan (Curtis 2000: Fig. 8). During this period, the first indication of frankincense traded to China is found in written sources (Kauz 2010:131), and some incense burners have been recovered in China (Bulling 1972). This is also the period in which diplomatic contacts between Rome and China were established, which favoured trade exchange in several products along with silk; the trade routes were later called the Silk Road (J. Hill 2009; McLaughlin 2010).

Along these trade routes, many important caravan cities flourished. In the Near East, the most famous of these were Palmyra (in Syria), Hatra (Northern Mesopotamia) and Charax (Southern Mesopotamia; Frye 1992). The Parthian Empire, thanks to its location between the Mediterranean and Central Asia, had a crucial role in controlling these roads. *The Parthian Stations* (Isidore of Charax 1914), dated to the first century CE, describes the main land routes from Central Mesopotamia, where the main cities of Babylon, Seleucia and Ctesiphon (see Chapter 5) were located, to Central Asia. These roads passed to Iran, through Kermanshah, Hamadan and Tehran, then east towards the Parthian capital of Hecatompylos. From here, they continued through Khorasan to Herat, where they split into two branches, the northern branch leading to Merv and Sogdiana or northeast to Bactria and then China, the southern branch to north India (Frye 1992). Along these roads, the cities of Begram (in modern Afghanistan) and Taxila (in modern Pakistan) have yielded large quantities of Roman and Parthian material along with Indian and Chinese items (Tomber 2008: 122–4), which bear witness to the richness of commercial exchanges along the Silk Road. Under the Sasanian Empire, trade exchanges along the Silk Road continued, demonstrating an important economic dimension of the Empire (Daryaee 2013: 136–8). In fact, even as empires fell and were replaced, the Silk Road continued to thrive, with occasional disruptions, and it was only after the economic changes brought about by improved seaborne navigation, the discovery of the New World, and European repositioning in trade links in the fifteenth century CE and later that the Silk Road diminished in importance (Tucker 2015: 216).

The Silk Road, thus, was critical in the trade of aromatics, and these products became some of the most desired foreign products in many states and empires in the Old World. The popularity of frankincense and myrrh was especially great in the westernmost parts of the Old World; in the Roman Empire (C. Singer 2007: 7) one can trace evidence of their presence as far away as Britain. Chemical analyses of resins found in late Roman tombs confirm that these resins were indeed frankincense from South Arabian sources (Brettell, Schotsmans, Rogers *et al*. 2015).

In summary, the aromatics from Arabia demonstrate that these products became truly global by spanning the extent of the Old World, something not seen in earlier pre-AoE trade (Figure 6.4). The trade network in aromatics and other products connected distant regions stretching from China to Western Europe. This trade seems to have started as early as the Iron Age, and it continued and expanded throughout the AoE, reaching a zenith perhaps from the second century BCE to the second century CE, although heavy trade activity along international routes continued well after this period (C. Singer 2007: 7). It is noteworthy, however, that between the second century BCE and the second century CE, there were three or at most four large empires that spanned the area between Europe and China (i.e., the Roman, Parthian, Kushana and Han Empires). Merchants could therefore cover long distances yet cross few political borders, which may explain why trade activity throughout the Old World reached a high point. Also, travellers crossing such long distances could make themselves understood by people of different ethnic origins by use a small number of international languages. For example, Greek could be used throughout much of the Mediterranean, and Aramaic across the Near East to Central Asia (see Chapter 9). Moreover, the stability of the large empires granted this network long periods of activity. Even when the imperial political system changed (for example, from the Parthian to the Sasanian Empire, and from the Roman to the Byzantine Empire), the overall trade system was not impacted to the point where there was extensive disruption of trade contacts lasting for long periods, as was the case during the Bronze Age. The situation in Western Europe after the collapse of the Roman Empire in the fifth century CE may have caused a long-term partial disruption of trade in the Old World; however, trade connections between the Byzantine Empire and Western Europe continued despite the establishment of the Germanic kingdoms within the territory of the former Western Roman Empire (Drauschke 2007).

#### 6.2.2 Pepper and the Indian Ocean trade

Pepper was a sought-after spice in the AoE (see, e.g., Pliny 2006: book 12.14). It came from India and reached the Mediterranean via maritime routes along the Near Eastern and Egyptian coasts. In India, two species of pepper were available, black pepper (*Piper nigrum*) and long pepper (*Piper longum*; Tomber 2008: 55). Pepper, and other spices such as cinnamon and cassia (the former from west India, the latter from Sri Lanka, Southeast Asia and China; see Tomber 2008: 54), are also mentioned in the *Periplus Maris Erythraei*, a first-century CE text written by a merchant from Alexandria, Egypt, that provides first-hand information about the intense exchange of products from India via the Indian Ocean and the Red Sea. The trade accounts mention important ports and trade winds (Casson 1989). In the Far East, pepper eventually reached China during Late Antiquity, becoming part of culinary practice in that country (Adshead 1995: 93).

Archaeologically, trade between India and the Mediterranean is evidenced by the ports excavated along the Egyptian coast of the Red Sea, such as Myos Hormos (first to third centuries CE) and Berenike (third century BCE to the sixth century CE; Tomber 2008: 58–65; Sidebotham 2011). Remains from Berenike, in particular, are evidence of intense trade activities at this port, with imports from India, the Mediterranean and Arabia, including pots for pepper, and precious and semi-precious stones such as carnelian (Tomber 2008: 71). Of particular interest is a bead from Java recovered at this site, which indicates the extension of the trade network as far as Southeast Asia (Francis 2007). On the other side of the Red Sea, the ports of Clysma (modern Suez), Aila (modern Aqaba) and Leuke Kome also had important roles in longdistance exchange up to the fourth and fifth centuries CE (Tomber 2008: 66–70). Further east, in the Persian Gulf, ports were involved in the Indian Ocean trade, under the control of the Parthian and later the Sasanian Empires (Whitehouse and Williamson 1973). For example, the site of Ed Dur (first century BCE to early second century CE), located in the Oman Peninsula, yielded material such as coins and glass of Roman, Nabatean, Parthian and Indian origins; glass beads from Sri Lanka and Tanzania were also found (Haerinck 1998).

Finally, the Indian Ocean trade also brought Roman and Near Eastern materials to India and Sri Lanka. Here, Roman and Parthian/ Sasanian pottery, and other material from both the Mediterranean and Mesopotamia (e.g., glass, bullae), were recovered (Tomber 2008: 117–51; Frye 1992). This material spans the period from the first century BCE to the sixth century CE, with some short chronological gaps and changes in the distribution patterns across the region through time (Figure 6.4).

#### 6.2.3 Coinage

During the AoE, trade was increasingly conducted on a monetary basis. Coins were probably introduced in the mid-seventh century BCE in Lydia (west Anatolia) and quickly spread across the Near East and beyond, where several cities struck coins (Manning 2013: 2). Although it appears that coins were initially introduced to make the payment of mercenaries easier (Manning 2013: 2), they became more and more common in all economic transactions through time. Thanks to their great spread, coins also became a vehicle of imperial political propaganda.

Coin distribution can act as a proxy for the extension of trade exchanges during the AoE, though it should be borne in mind that coins, especially those made of precious metals, could circulate as valuable items rather than as a means of transactions (Figure 6.4). Within the Achaemenid Empire, coins appear to have been predominantly used for trade transactions with Greeks along the Levantine coasts, which would also explain why Persian coins are generally found towards the west, outside the reach of the Achaemenid Empire, in Italy, Macedonia and Greece (Alram 1994). Towards the east, Persian coins appear to have been used less frequently, though a hoard containing Greek, Persian and Indian coins was found near Khabul (Afghanistan) in 1933 which indicated that, albeit more sporadically, Persian coins too had travelled to Central Asia (Schlumberger 1953). It was during the Seleucid Empire that coins spread considerably across the Near East, because the Seleucids incentivized coin payments for their military, and therefore needed to collect taxes in coins. Many mints were established across their empire to guarantee coin supply, and, consequently, more coins were used in markets for everyday economic transactions (Aperghis 2004: 29–32). In the following periods this trend continued, reaching its peak under the Sasanian Empire. Sasanian coins were used for trade as far away as China, reaching the River Volga in Russia, and the Baltic Sea (Frye 1992; Thierry 1993). They were also found in Japan along with other Sasanian objects (Sugimura 2008), and along the coasts of the Persian Gulf and in India (Whitehouse and Williamson 1973; Potts 2010: 65–82). Sasanian merchants also engaged in trade with the rival Romans (and later Byzantines); in some cases international treaties were implemented to regulate exchange (Daryaee 2013:140; Drijvers 2009: 449). Sasanian coins in Britain, although rarely found, seem to indicate the wide scope of Sasanian–Roman trade relations (Herepath 2002). Finally, coin distribution gives us a glimpse into the wide reach of Roman commercial links, as witnessed by Roman and later Byzantine coins found in China, in Indonesia, and as far afield as Japan; Chinese texts that refer to trade relations with the Romans also suggest that coins should be expected this far east (Young 2001; J. Hill 2009; Kyodo 2016).

## 6.3 Private corporations during the AoE

Our information about private companies directly involved in longdistance trade and the economy increases with the AoE. The economy of the Achaemenid Empire has long been recognized as a mixture of the royal, temple and private sectors, the latter particularly attested on the Levantine coasts where Phoenician cities produced and traded mainly textile dye and glass (Dandamaev 1989). Important evidence for the private sector in the economy under the Neo-Babylonian and Achaemenid Empires comes from the private archives of families actively involved in different economic sectors and trade. Examples are the Egibi family of Babylon, the Ea-iluta-bani of Borsippa, and the fifth-century BCE Murashu family of Nippur (Nemet-Nejat 1998: 226). The tablets found in the Murashu archive indicate that this family was involved in a wide range of economic activities, including land management, money lending, the keeping of deposits, trade and tax collection (the last on behalf of the state); this family acted as a firm, hired agents, and stipulated agreements with several individuals, thus representing one of the first forms of a banking system that included not just one location or a branch but was found in many areas throughout Southern Mesopotamia and into Elam (Dandamayev 1988).

Information about the private sector in the economy and trade is again scarce during the Seleucid and Parthian periods; this is mostly because the number of documents that have survived is minute. Under the Sasanian Empire, however, we learn from written sources (e.g., Syriac law books) that trade was mainly under the control of private families and of companies regulated by state and religious laws (Frye 1992). Moreover, although in Persia itself merchants did not seem to be of high status, in Central Asia written texts reveal that merchants were at the top of the social ladder, particularly in areas such as Bukhara and Samarkand, where private initiative was essential to the construction of canals and other infrastructure with an economic purpose (Frye 1992). An interesting feature of the AoE economy is the importance of religious communities, in particular Christian ones, involved in longdistance exchange. The spread of universal religions under empires (see Chapter 10) appears to have allowed the creation of trusted links between distant communities and increased cohesion between different ethnic groups who shared the same religious beliefs, and helped to facilitate and set the conditions for international trade on a global scale (Dark 2007). This is evidenced by, for example, fifth- and sixth-century CE texts that indicate direct or indirect involvement of Christian churches and clergy in the trade with India (Tomber 2008: 168–9).

### 6.4 Merchant colonies

Some evidence regarding the intensification of population movement in the AoE comes from merchant colonies. In section 6.2 we saw that during the AoE the international trade network reached an unprecedented extent; hence one could wonder whether this phenomenon was mainly due to the movement of traded objects 'down the line', that is, through many intermediate passages, or via merchants who established colonies far from their homeland. In section 6.1 we spoke about the stable presence of Assyrian merchants in Kanesh, in Central Anatolia, located about 1000 km away from Assur, and their trade activities, dated to the early second millennium BCE. During the Late Bronze Age, evidence of the presence of merchant colonies comes from Ugarit, from where texts indicate the stable presence of merchants from Akko, Ashdod and Ashkelon (Vidal 2006). Canaanite merchants are attested as a stable presence in Crete (at Kommos) and Cyprus (at Hala Sultan Tekke), as are Assyrian merchants in Sidon and other cities on the Levantine coasts, for the trade of textiles (Aubet 2000).

These cases indicate that merchant colonies existed during the pre-AoE, though during the AoE the evidence increases and the phenomenon seems to spread to a larger area. In the first millennium BCE the phenomenon of Phoenician colonization greatly expanded the presence of Phoenician merchants across the Mediterranean (Aubet 2001), which allowed the Phoenician cities on the mainland to grow economically, thus becoming over time the target of Assyrian indirect or direct control; Greek merchants too were widely present across the Mediterranean: Greek emporia were founded in Egypt (for example at Naukratis), and a stable presence of Greek merchants can also be inferred in some Levantine cities, for example Al Mina (Fantalkin 2006; see also Chapter 7). This trend increases during the Hellenistic period, under the Seleucid Empire, when we discover many Greek cities being founded, across the Near East as far as Central Asia, in which elements of Greek and local Near Eastern cultures blend together (see Chapter 7). This trend continues under the Sasanian Empire, with Sasanian merchant colonies attested in the Persian Gulf, Sri Lanka and Malaysia, and as far as China (Daryaee 2013: 139; 2009: 64). Similarly, Roman, Jewish and Christian settlers from the Roman Empire are attested in India between the second and fifth centuries CE as being involved in commercial affairs (Curtin 1984: 90–109; see also above).

This evidence suggests that the phenomenon of merchant colonies, though not new in the AoE, greatly increased during this period, following the expansion of the trade network. This phenomenon should be framed within the establishment of the large empires of the AoE, which created more favourable conditions for merchants to move across wider areas and settle far from their homelands.

## 6.5 Speed of travel

Another difference between pre-AoE and AoE trade connections is the speed of travel. Evidence about the speed at which people travelled before and during the AoE comes from different sources, some dealing with long-distance trade and others (more numerous) with state correspondence and military campaigns. The Old Assyrian caravan trade connecting Assur to Kanesh in the second millennium BCE (see above) covered a distance of about 1100 km in 42 days (see Barjamovic 2011: 15; Larsen 2015: 175–6), which corresponds to about 26 km a day. About 30 km a day is the average distance travelled by pack donkeys in antiquity (Moorey 1994: 12).

The Old Assyrian caravan trade can be compared with later evidence regarding speed of travel during the AoE. In the Neo-Assyrian Empire (ca. 900–612 BCE), a system of relay was introduced to deliver letters. Estimates indicate that this system could cover a distance of about 700 km, from Que (in the modern Adana region) to the Assyrian heartland (northern Iraq), in about five days (Radner 2014b: 74), which corresponds to 140 km a day. Furthermore, it was mentioned earlier that long-distance routes were direct or relatively straight as they connected distant key cities with the Assyrian capitals (Altaweel 2008: Plates 16–17), while routes in the pre-AoE, such as the Assyrian trade route from Ashur, may have taken less direct routes (Larsen 2015: 179). The less direct Assyrian trade routes could have avoided taxation or even conflict in cities along the way. In fact, imagery showing remnants of ancient roads in the Jazira and Khabur Triangle regions of Northern Mesopotamia largely shows short or nearest-neighbour route connections between sites (Ur 2003). For other AoE cities, the trend towards trade routes that are direct and long-distance is evident from satellite imagery for sites such as Hatra (Altaweel and Hauser 2004: 64). In effect, trade, or movement of goods in general, in the AoE began to show physical evidence of being not only long-distance but also more direct than in earlier periods.

In the Achaemenid Empire, mounted couriers of the postal service, the Angarium, riding along the Royal Road from Susa to Sardis, could travel about 2700 km in seven days, according to Herodotus (Kia 2016: 127). This indicates an impressive speed of nearly 386 km a day. Colburn (2013) revised Herodotus' affirmation and calculated that the Angarium would probably take around 12 days to cover such a distance based on the parallel with the more modern Pony Express service. This gives 225 km a day, a value which, while not as high as Herodotus' figure, clearly indicates a swift connection between the cities.

Such high figures for travel speed in the Neo-Assyrian and Achaemenid Empires were due to different factors. First, the material transported by means of these very quick connections was essentially diplomatic, so it was crucial that it was transmitted as fast as possible; on the other hand, trade connections may have been slower. According to Herodotus, for example, the journey from Susa to Sardis took 90 days for a normal traveller, which is about 30 km a day, similar to the Old Assyrian caravans. Xenophon, however, informs us that the same distance could be covered in half the time by a normal traveller (which is 60 km a day; Colburn 2013: 42). After the Achaemenid Empire, information about travel speed comes from the rapidity with which the news of the king's or emperor's death spread. From an Idumaean ostracon bearing a date in the first year of the reign of Philip III (17 June 323 BCE), it can be inferred that the news of Alexander's death reached Idumaea (modern Negev, south Israel) from Babylon after one week (Colburn 2013: 42), covering about 1000 km in a straight line, that is, about 140 km a day. Such a high figure might be comparable with the speedy communication witnessed during the Neo-Assyrian Empire.

In the Roman Empire, it took 30 days to communicate the emperor's death from Italy to Egypt in summer when the sea was navigable, that is, about 62 km per day, which is slower than the postal service of the Achaemenid Empire (Colburn 2013: 47–8). On land, communication within the Roman Empire travelled at about 75 km per day, which meant that it could take about 17.5 days to travel from Rome to Colchester in Roman Britain (about 1300 km; Colburn 2013: 47–8). The figures for the Roman Empire's speed of travel are not as high as those of the Achaemenid Empire, which is probably due to the different geographies of the two empires, but they are still greater than the average Old Assyrian caravan trade speed of the second millennium BCE.

Chinese sources of the first and second centuries CE are also useful for inferences about speed travel. When referring to the Parthian Empire, they report that the route from Hecatompylos, in Parthia, to Chaldea was about 3580 km long and could be covered in about 60 days, giving 60 km a day (Hirth 1885: 36–40). This figure is close to that given by Xenophon for the Achaemenid Empire and those available for the Roman Empire.

While, clearly, horses would have made travel in the AoE far faster than the donkey-based caravan travel of the pre-AoE, differences in the route systems also made a difference. In the pre-AoE, routes were not direct and travel sometimes had to bypass particular areas. In the AoE, both maritime and land travel became direct. This was due both to technical changes, in navigation and the greater use of horses, and to the possibility of covering vast distances without encountering disruption from the fact that there were many states and political entities.

## 6.6 Conclusion: the factors that distinguish pre-AoE and AoE trade

In the previous sections, it was shown that exchange of goods across long distances is not an invention of the AoE; however, AoE trade shows globalized traits, in that goods moved through much of the known world, trade was run by private enterprise and government support, and the speed of trade probably increased. The scale of trade during the AoE became far larger than in previous periods, with trade connections crossing Eurasia, including South Asia, from west to east. Both land and maritime routes developed, and exchanges across these routes peaked during the AoE. The establishment of these long-distance trade corridors appears to be a more stable phenomenon, suffering only marginally from the collapses of empires and their replacement by other empires. As an example, both the Silk Road and the Indian Ocean trade routes remained active well beyond the AoE, into the modern era, when the discovery of the Americas, among other factors, drastically changed the trade scenario and opened up new opportunities for Western European countries in the form of long-distance trade across the Atlantic Ocean. What this shows is that movement in general probably became easier during the AoE, rarely being interrupted during this period. As the circulation of objects became easier, we should expect that population movement would also be easier. In effect, the movement of rare objects over geater distances shows that movement generally became easier as the political landscape favoured larger states. As Chapters 4 and 5 discussed, evidence of new urban settlement patterns and of foreign populations intermixing in cities demonstrates that there was probably movement of population. If such movement occurred, it is likely that objects, too, moved more easily along trade routes. This is exactly what one sees throughout the AoE, in places where political disruption did not have long-term effects on movement.

Although evidence about the speed of travel and trade is not abundant, where information is primarily derived from sources dealing mainly with couriers one can conclude that AoE states were able not only to move items to more distant areas, but also to send them more rapidly, probably because of the protection and stability that empires offered such transactions. One could conclude that, compared with the 30 km a day of the Old Assyrian caravans, under the AoE a speed of 60 km a day may have been more common, with very high peaks for the Achaemenid postal system; the transmission of very important information (e.g., the king's death) may have been even swifter. One of the main reasons we see higher communication speed was that large empires built roads and infrastructure (see, e.g., Altaweel 2008; Waters 2014: 111), and facilitated the movement of people across different regions by removing political borders. In the pre-AoE, political boundaries created more obstacles to extending the distance over which trade could take place, and made trade more vulnerable to political vicissitudes. As acknowledged by other scholars, quick connections across long distances were an essential part of an empire's communication strategy and internal cohesion (Radner 2014b: 1; Colburn 2013: 30). The benefit of these connections is that they helped create infrastructure that allowed trade to move more quickly and over greater distances.

The keys to the success of the AoE's long-distance trade, including its great scope and stability, are to be found in the effect of the imperial systems the Near East fostered. Incentives to trade were created for more individuals to participate in these systems as they became increasingly integrated into diverse societies. People from a variety of ethnic backgrounds now lived far from their original homelands, allowing trade connections to develop at more distant locations. More importantly, the political climate made movement not only possible but also easier, and allowed people to participate in trade interactions that were probably influenced by private individuals as well as government bodies. Movement of people and their concentration in larger cities created large markets that demanded staples as well as luxury products. Despite the consequent creation of more sparsely populated areas in parts of the Near East (see Chapter 4), empires facilitated rapid contacts over long distances by means of road systems and postal systems, and by drastically reducing or removing social barriers across the Near East. In Chapter 9, we will see that language, another barrier, became less of a factor in the AoE. Overall, these social possibilities and some technical innovation permitted movement of goods that reached a wider clientele, in more distant areas, much faster than in earlier periods. Thus, the basis of a globalized and intercultural trade was laid down during the AoE. Although the empires of the AoE were concerned with their economies and trade, and exercised firm control over coinage and taxation, private families organized themselves into firms that greatly developed during the AoE and laid the basis of a modern banking and financial system. The emergence of universal religions during the AoE probably facilitated the establishment and maintenance of long-distance, intercultural trade contacts by creating common faiths that could be shared and used for business.

## 7 Material culture hybridization

## 7.1 Background

This chapter focuses on the phenomenon of material culture hybridization or syncretism – that is, the blending of artistic styles that developed in different geographic areas and cultural spheres – comparing the pre-AoE with the AoE. It is worth briefly explaining what is intended by 'material culture hybridization'. In the literature on stylistic aspects of material culture, there is a great debate about how to define 'style' and what characterizes an 'international', 'intercultural' or 'hybrid' style as opposed to a 'regional' or 'local' one (e.g., Crowley 1989; Caubet 1998; Perrot and Madjidzadeh 2005; Fischer and Wicke 2011; Pfälzner 2015; Feldman 2015; Stockhammer 2013). 'Hybrid style' is defined here as the combination of local motifs and motifs borrowed from foreign cultural milieus (similar to Feldman 2006: 'international style', or Pfälzner 2015: 'hybrid regional style'), including reinterpretations and adaptations.

Material culture hybridization can be the consequence of different factors. For example, objects traded across long distances can inspire artisans to incorporate stylistic features that originated far away. There are cases, however, in which hybrid styles can be linked to the actual movement and mixing of populations, in particular of artisans from distant regions. In the archaeological literature, migrations, invasions and deportations have sometimes been connected to the emergence of hybrid styles (see, e.g., Chapman and Hamerow 1997; Burmeister 2000; Ben-Shlomo 2011; Knapp 2008), though it is often difficult to infer the transfer of craft skills from stylistic elements alone, without supporting texts. Pre-AoE texts mention artisans who travelled across the Eastern Mediterranean, the Levant and Mesopotamia. Some examples can be found in the texts from the palace of Mari of the second millennium BCE and the Amarna letters of the fourteenth century BCE (Zaccagnini 1983). These texts show that the movement of artisans was regulated mainly by royal administration, meaning that travelling craftsmen were dependent on the largesse of palaces and moved from one royal house to another, in a framework very similar to the exchange of gifts among royal houses (Zaccagnini 1983: 243–54). Evidence of the voluntary movement of skilled craftsmen across different regions and of their permanent residence in a foreign country can scarcely be found in pre-AoE written sources, though a few hints do exist (Bevan and Bloxam 2016; Zaccagnini 1983: 256–7).

The pattern of craftsman mobility continued throughout the AoE, as evidenced by contemporary written sources. Skilled artisans from Syria, Babylonia and Phoenicia are mentioned in the Neo-Assyrian texts as workers living in Kalhu (modern Nimrud) and Nineveh, among other foreigners (Luckenbill 1924: 73; Kinnier Wilson 1972; Radner 2007: 190–1). They were probably deported during the military campaigns, but some hints suggest that skilled craftsmen voluntarily moved to Assyria for job opportunities (Zaccagnini 1983: 260). Similarly, the Achaemenid texts mention Egyptian, Lydian, Ionian, Sardian, Babylonian, Median and other craftsmen recruited to build the capital cities of Susa and Persepolis, where a mix of Egyptian, Mesopotamian, Iranian and Greek styles is indeed visible in the architecture and architectural elements (Nylander 1972; Roaf 1983; see also Chapter 5). In the Hellenistic period, the availability of foreign (especially Greek) artisans across Egypt, the Near East and Central Asia as far as the River Oxus greatly increased because of the establishment in these areas of many Greek-Macedonian communities following Alexander the Great's conquests (see Chapter 2). At this point, the Hellenistic style merges more frequently with local styles in western and Central Asia in almost all categories of material culture, including common and non-elite crafts, as demonstrated below. Later, even during periods of state conflict there is greater evidence of craftsmen moving freely across empires and states. Despite the rivalry between the Byzantines and the Sasanians, for instance, texts inform us of the presence of Byzantine artisans at the Sasanian court, producing works of art that show Byzantine influences (Shahbazi 1990).

From what the texts tell us, material culture hybridization can depend on many factors, one of which is certainly the movement of skilled craftsmen trained in different cultural spheres. Texts suggest that this phenomenon occurred during both the pre-AoE and the AoE; however, here we show that in the AoE the geographical and chronological diffusion, as well as the pervasiveness across social strata of material culture hybridization, reached a much higher level than ever before. As empires facilitated and boosted the movement and blending of people from various cultural backgrounds, artisans moved at a much greater rate, which permitted an increasing blending of stylistic features and the emergence of hybrid styles. The sheer number of artisans moving, and the frequency with which they moved, also made it possible for very mundane objects to use new hybrid styles.

The following sections describe some examples of material culture hybridization, comparing the pre-AoE with the AoE. Particular attention will be given to more common, non-elite expressions of material culture, such as terracotta figurines and house architecture, as more mundane remains can tell us how pervasive the intermixing of styles was across society.

## 7.2 Material culture hybridization in the pre-AoE

During the Bronze and Early Iron Ages, from the Mediterranean to the Indus, various material culture styles developed in architecture, painting, sculpture and object design, including Minoan, Egyptian, Levantine, Mesopotamian, Elamite and Harappan. At this time, there was interregional borrowing of motifs and themes. Sumerian stylistic features that had developed in Southern Mesopotamia during the Early Bronze Age (the Proto-Dynastic period) expanded into Central and Northern Mesopotamia, as well as into north and west Syria, Anatolia and Iran (Matthiae 1981; Frankfort 1996: 83, 242). This is evident in statuary, architecture, glyptic art and portable objects such as chlorite vessels (Kohl 1974; Perrot and Madjidzadeh 2005: Fig. 6.1). Moving towards the Levantine coast, Cyprus and the Aegean, Egyptian motifs and stylistic features became predominant during the Middle and Late Bronze Ages (Figure 7.1). This is shown by finds from Megiddo, Byblos, Ras Shamra-Ugarit, Qatna and Enkomi, to name but a few sites (Markoe 1990; Frankfort 1996: 243–4; Moorey 2001; Feldman 2006; Pfälzner 2015). The prominent political and commercial role that Egypt had during this period in the Mediterranean and the Near East favoured the spread of Egyptian styles beyond Egypt's borders. Egyptian style, however, was not the only influence seen in the Levant and the Aegean. By this time patterns of stylistic influences formed an intricate network, reflecting the interconnected nature of political relations among the royal houses of the Eastern Mediterranean and the Near East. Aegean stylistic features also made their way into the Levant (Niemeier 1991; Hitchcock 2005; see also Figure 7.2) and into Egypt, where the best example is the Minoan

**Figure 7.1** Schematic map showing the spread of the main styles from the pre-AoE (Bronze Age) outside their places of origin

**Figure 7.2** Ivory lid from Minet el-Beidha (near Ugarit, northern Syria) showing the so-called 'mistress of animals', ca. 1250 BCE. This object merges a common Levantine and Mesopotamian iconographic theme with the Mycenaean-style dress of the mistress, who sits on an Aegean-style chair (see Caubet 1998; Rama 2016)

frescos of Tell el-Daba (ancient Avaris) (Cline 1998). Similarly, Syrian stylistic features can be recognized on some of the objects found in Tutankhamun's tomb in Thebes (Feldman 2006: Pl. 2).

This intense exchange of stylistic features brought about hybrid styles; however, it is difficult to say whether these hybrid styles were the result of the movement of populations (including artisans) across the regions or of cultural influences which spread across these areas through, for example, traded objects. As mentioned above, pre-AoE texts mention cases in which craftsmen worked in areas which were different from those of their origin. However, as highlighted, the long-term presence in cities of foreign artisans was often due more to the temporary lending of skilled labour by one royal house to another than to artisans travelling across the Near East and the Eastern Mediterranean (Zaccagnini 1983; Niemeier 1991; Muhly 2005; Hitchcock 2005).

It is possible to highlight some characteristics of the pre-AoE hybrid styles that mark a difference from the AoE. Firstly, the objects manufactured in a hybrid style during the pre-AoE retained a very local stylistic imprint (Pfälzner 2015); they were probably the products of local artisans incorporating foreign stimuli into their works. Secondly, hybridstyle objects circulated in most cases within elite circles, as the elites found that the exotic nature of these objects highlighted their prominent social status (Feldman 2006; T. C. Wilkinson 2014: 219). Additionally, hybridized architectural forms appear to have been mainly intended for the elites (Hitchcock 2005: 142). Only a limited part of society was influenced by such syncretism. The phenomenon of hybrid styles does not seem to have achieved a considerable geographic spread during the pre-AoE. Furthermore, it does not seem to reflect a major movement of artisans from one region to another. To demonstrate this better, we will focus on non-elite expressions of material culture, of which few examples of hybrid styles from the pre-AoE have been found.

Early and Middle Bronze Age terracotta figurines, for example, show a great stylistic diversity from the Aegean (the Cyclades) to the Indus, passing through Anatolia, Syria, Mesopotamia, Iran and Central Asia (T. C. Wilkinson 2014: Fig. 6.27, bibliography; Figure 7.3). These objects' stylistic details (e.g., body shape, body parts, eye decoration, hairstyle) allow one to pinpoint their regional origin with confidence, as each of the areas mentioned above produced its own stylistically distinctive figurine type. Limited stylistic borrowing from one region to another is observed in this non-elite class of objects.

Another area of non-elite material culture expression is domestic architecture, which reveals a strong regionalism during the pre-AoE. In Bronze Age Syria and Mesopotamia, the most common house type is the so-called courtyard house, usually found in dense agglomerations of houses separated by narrow streets (Akkermans and Schwartz 2003: 269). In the contemporary Southern Levant, various house types, showing regional differences between the Negev and the Mediterranean area, have been found that reflect regional social development and complexity (Miroschedji 2014). In Bronze Age west Anatolia, house plans similar to the 'megaron' houses typical of contemporary Greece appear alongside two- or three-room houses in the local style, which differ from those found in much of the rest of the Near East (Steadman 2011: 236). Moving eastwards, the Mesopotamian-style courtyard house is also found in Elam (Potts 2004b: 257–8), whereas in eastern Iran house plans show a different arrangement from the typical Mesopotamian style (for example at Shahr-i Sokhta; Tosi 1983).

**Figure 7.3** Some examples of terracotta figurines of the pre-AoE. Each area produced a stylistically distinctive figurine type, with very few stylistic borrowings from one region to another (Metropolitan Museum 2017: MET:31\_11\_3; MET\_74\_51\_1535; MET\_59\_125; MET\_ an64\_130\_R; MET\_59\_41\_20; MET\_1\_2250\_011; MET\_2001\_306)

In effect, there is not enough evidence for the pre-AoE to determine whether the hybrid styles that are visible in portable objects and some architecture were mainly due to interregional movement of artisans and people. Syncretized styles of the pre-AoE characterized mainly luxury objects and elite architecture, whereas they are virtually absent on common forms of material culture, including terracotta figurines and domestic architecture. This suggests that the dynamics of indirect cultural influences (e.g., through traded objects) and intercultural elite emulation were the main causes of material culture hybridization in the pre-AoE. This does not exclude the possibility that a few travelling artisans were present, but their input into the creation of hybrid styles did not leave a significant mark in the archaeological record.

## 7.3 Material culture hybridization during the AoE

#### 7.3.1 The Iron Age and the Persian periods

Material culture hybridization and the interregional borrowing of stylistic features increased throughout the first millennium BCE. The Late Bronze Age patterns of material culture hybridization continued into the Iron Age I–II (roughly the twelfth to eighth centuries BCE), despite the crisis that struck many political entities at the end of the Bronze Age (Markoe 1990). By this time, the Phoenicians had become a major factor in the spread of hybrid styles across the Mediterranean, especially following the establishment of Phoenician colonies along the Mediterranean coasts. Blending Levantine, Egyptian and Aegean styles, the Phoenician artisans created finely made items such as pieces of jewellery, glasswork and metal bowls, which were disseminated across the Mediterranean (Markoe 2000). This led to the addition of orientalizing motifs to objects from Greece, Etruria and Spain (Burkert 1992; Aubet 2001; Figure. 7.4). The movement of Phoenician merchants, settlers and artisans across the Mediterranean (López Castro 2006) was greatly encouraged by Neo-Assyrian rulers, who saw the economic benefit of the Phoenician trade networks in the Mediterranean (Aubet 2001).

The Phoenicians are not the only example of how the Neo-Assyrian Empire facilitated cross-cultural contacts and population movement. Following their military campaigns in the Levant, the Assyrians often made use of mass deportations not only to break local resistance but also to repopulate specific areas for political and economic purposes

**Figure 7.4** A Corinthian orientalizing jug, ca. 620 BCE. Note the two sphinxes on the top, which derive from a blend of Levantine and Egyptian motifs, whereas the rest of the depictions are in a Greek style (Unknown 2007)

(Oded 1979; Na'aman 1993). This inevitably caused a mixing of population in several areas of the empire. Moreover, Levantine craftsmen enjoyed greater possibility of movement across the Near East as much of the region became pacified by the Assyrians in the eighth and seventh centuries BCE (Zaccagnini 1983). In some cases, artisans were forcibly brought to Assyria to participate in the development of the royal cities such as Kalhu (Nimrud), Nineveh and Dur-Sharrukin, as indicated in Chapter 5.

The increased movement of people and craftsmen across the Neo-Assyrian Empire, either forced or voluntary, may have been one factor in the increased mixture of stylistic elements in material culture. Pottery traditions, for example, which had been regionally defined in the tenth to early eighth centuries BCE, began to merge, so that typical Syrian coastal types are even found inland and Assyrian and Assyrianizing pottery is found all over the Levant (Lehmann 1998). Typical Mesopotamian burial customs appear now in the Levant (Amiran 1959), and Syrian figurative arts acquire motifs and styles from Assyrian art and vice versa (I. Winter 2010).

This trend of material culture hybridization based on increased population movement becomes still more visible during the Achaemenid period. The generally tolerant policy of the Achaemenids towards multiple ethnic groups (see Chapter 8) allowed cultural influences to thrive and spread. The construction of long-distance road systems and the resettlement of people in parts of their empire for military purposes probably resulted in craftsmen from different cultures spreading to various areas (Stein 2014; Waters 2014: 100–7). The increased spread of a lingua franca (see Chapter 9), in the form of Aramaic, facilitated communication across their empire and made the voluntary movement of craftsmen much easier. All the while, as shown in Chapter 4, population concentration along the coastal regions and larger riverine urban areas led to increasingly multi-ethnic regions. Under these conditions, material culture showing syncretistic qualities began to be found all over the Achaemenid Empire, where development was not confined to one region or elite goods. For example, pottery, burial customs and the architecture of public buildings began to combine a variety of elements from distant regions (Lehmann 1998: 23–59; Wolff 2001; Akkermans and Schwartz 2003: 390–8).

The increase in syncretistic qualities probably related to population movement and increased intermixing of populations from different cultural backgrounds across the Achaemenid Empire. On the Levantine coast, Greek and local pottery types were common along with Cypriot figurines; Phoenician objects as well as Egyptian and Egyptianizing amulets and figurines were also typical (Nunn 2000; Betlyon 2005). The presence of graffiti in Greek, Phoenician and Aramaic strongly supports the existence of a multicultural population (Waldbaum 1997). A new type of hybrid material culture in the Levant was Phoenician sarcophagi (Figure 7.5), which blended the Egyptian custom of using sarcophagi with Greek elements in their decoration. The cultic statues from the Temple of Amrit, on the Lebanese coast, represented local gods in a Greek style, while the architecture of the temple itself merges Egyptian and local stimuli (Renan 1864). Burial customs point to the existence of mixed Levantine, Mesopotamian and Iranian populations in the Levant and other parts of the Near East (Stein 2014). Object design shows Persian influences all over the empire, from the Southern Levant, Anatolia and the Caucasus to Central Asia. In particular, Iranian-style weapons, metal and stone vases, metal plaques and rhyta (decorated

**Figure 7.5** Detail of a sarcophagus from Antarados, northern Lebanon, ca. fifth century BCE. The use of a sarcophagus is in the Egyptian tradition, but the decorative style is clearly Greek (McLeod 2007)

containers suitable for pouring liquids) were now found in the Levant (Ivantchik and Licheli 2007; Baumer 2012: 172–269; Betlyon 2005; Squitieri 2017). Style hybridization, as stated in Chapter 5, also reached the core of the Achaemenid Empire; Greek, Egyptian, Mesopotamian and Iranian figurative and architectural elements were employed in the construction of Persepolis by craftsmen who came from various regions (Roaf 1983; Nylander 1972).

Hence, examples of material culture hybridization became more frequent all over the Achaemenid Empire, and elite and imperial arts were not the only art styles affected. Terracotta figurines moulded to express popular religious and other iconography, for example, show increasing signs of hybridization, mixing Levantine, Greek and Egyptian styles (see, e.g., Oggiano 2009; El-Khouri 2011; Figure 7.6). Not only may such styles reflect the presence of increasing numbers of artisans from different regions, but also the objects may have been made to be more appealing to people from different cultural backgrounds.

**Figure 7.6** Persian terracotta figurines from the site of Kharayeb (northeast of Tyre, Lebanon). On the left, two females modelled and dressed in Greek fashion; on the right, the Egyptian god Bes represented in his Egyptian iconography with the naturalist elements of the Greek style (after Oggiano 2009: Figs 3 and 9)

Although, as shown in the Introduction, distinguishing between syncretistic styles due to cultural influences and those due to population movement is not easy, the integration of historical sources with changes in material culture that began in the Neo-Assyrian period and continued through the Achaemenid period suggests that material culture hybridization had increased in association with population movement. Mass deportations, the establishment of colonies, the construction of road systems, craftsmen relocating, merchant mobility (see Chapter 6) and soldier deployment are some of the types of movements documented that suggest that material cultural changes may be linked to some of these historical events. Intensified mixing of population is linked with an increasingly hybridized material culture that spanned all levels of society, in which more common artefacts began to show blended elements. This trend continued and increased in subsequent periods.

#### 7.3.2 The Hellenistic and Roman–Parthian periods

The phenomenon of material culture hybridization reached a peak in the period roughly between the third century BCE and third century CE. During this time, the style that had emerged in Greece became the common denominator of various material expressions of culture, from the Central Mediterranean to India. While this style is often called 'Hellenistic', in the Near East it would be more accurate to say that the style represented syncretistic combinations of local and Greek elements. Hybridized examples in architecture, urban planning, stone sculpture, reliefs, painting, metalworking, coinage, pottery and terracotta figurines are abundantly evident (Schlumberger 1970; Colledge 1987; Baumer 2012: 272–302; Invernizzi 2012). Over this vast area, the Hellenistic style blended with local, pre-existing styles to create new hybrid forms.

The Hellenistic style characterized the royal monumental art of the Hellenistic states in Anatolia (Kosmetatou 2003), the propagandistic art of the Ptolemaic kingdom in Egypt, in which the naturalistic forms of the Hellenistic style blended with the traditional pharaonic art (Figure 7.7), the imperial art of Rome, especially after the Roman conquest of Greece, the Levant and North Africa (Burn 2004), and the royal art and coins of the Seleucids and, after them, the Parthians (Invernizzi 1997b, 2012; Colledge 1977).

**Figure 7.7** Example of Greco-Egyptian style. Engraved ring with portrait in the Greek style of Ptolemy VI Philometor (ca. 186–145 BCE) wearing the traditional pharaonic double crown. Held in the Louvre Museum (PHGCOM 2009)

In the Near East, Egypt and Central Asia, the spread of Hellenism can be linked to the settlement of many Greeks in the multi-ethnic cities of these areas, in which Greek and other communities lived together (see Chapter 5). Among these cities, in which material culture blended local and Greek styles, are Antioch, Alexandria in Egypt, Palmyra, Dura Europos, Seleucia on the Tigris, Babylon and Susa, to name but a few (Figure 7.8). Ai-Khanum, in the Bactria, modern-day Afghanistan, is a vivid example of the penetration of Hellenistic culture as far as Central Asia following the settlement of Greeks. Here, typical Greek material cultural and architectural expressions (e.g., theatre, gymnasium, statues in the Greek style) stand alongside hybrid forms which mix local and Greek stimuli (Figure 7.9; Mairs 2014). Examples of Greco-Iranian style come from the Parthian capital of Nisa, in present-day Turkmenistan, where the architecture in Iranian style shows Greek-style decorative motifs, and some of the objects (e.g., statues, metalwork and rhyta) display a mixed style (Invernizzi 1997b, 2012).

Even further away, the Hellenistic style affected the monumental art of Gandhara, which developed in the Peshawar Plain in modern-day Pakistan, between the first and the third centuries CE, probably under the influence of Greek artisans present in the region (Behrendt 2007; Figure 7.10).

Although the Hellenistic style influenced elite and monumental art, it was also a pervasive phenomenon that affected stylistic expression at all levels of society. In other words, the Hellenistic style was not only the language of power, propaganda and monumentality, it also became an inspiration for the figurative art associated with the more common or lower strata of society. A good example of this is the production of terracotta figurines. From the Central Mediterranean to Bactria, terracotta figurines now show stylistic features that can be clearly connected to the Hellenistic style, such as naturalistic facial and hair details along with a Greekinspired cloth-folding style (Bailey 1983; Invernizzi 1985; Török 1995; Martinez-Sève 2002; Menegazzi 2012; Lo Muzio 2010). Despite some local variation, these figurines were now much more similar across the entire Near East (Figure 7.11), in striking contrast with the more diverse figurines of the pre-AoE (see above). The hybrid style of the Hellenistic terracotta figurines can ultimately be connected to the multicultural milieu that probably accelerated in the Hellenistic period, in which communities with different cultural backgrounds exchanged cultural elements by being in close proximity to each other (Langin-Hooper 2013).

Additionally, house architecture showed less pronounced regionalism, unlike during the pre-AoE. From the Hellenistic era onwards, it

**Figure 7.8** The Temple of Bel in Palmyra (Syria), first century CE. The temple shows a typical Greek-Roman peristyle around the central building, which, in contrast to the Greek-Roman tradition, presents an entrance on the long side as well as decorative merlons on the top, both features recalling Near Eastern traditions (Gagnon 2010)

**Figure 7.9** Silver and gold plaque from Ai-Khanum (Bactria), ca. second century BCE, depicting the goddess Cybele and a scarified scene. The dress of the two figures on the left and the god's face above are Greek in style; the astrologic symbolism at the top references Near Eastern religions; the priestess's robe on the right is in a local style (World Imaging 2006)

is possible to see common features in house plans across the Near East. The typical Hellenistic house plan that developed in Greece (e.g., in Olynthus), with a central courtyard surrounded by columns (a 'peristyle') and a broad room next to it (Winter 2006: 157–235), is evident in the Mediterranean and Mesopotamia alongside local forms (see, e.g., Hopkins 1972; Figure 7.12). From this house plan, the so-called *iwan* evolved in the Parthian period, consisting of a rectangular room open on one side to a courtyard (Colledge 1977). The *iwan* spread across the Near East, and in many cities *iwan* houses coexisted alongside peristyle houses (Hauser 2012). House architecture in the AoE, from the Parthian period onwards, showed the widespread presence of these two designs from the Eastern Mediterranean to Central Asia, in striking contrast with the regionalism in house designs present during the pre-AoE.

**Figure 7.10** Statue of Buddha from Gandhara showing Greek-style cloth folding and naturalistic facial details, ca. second century BCE, held in the Tokyo Museum (World Imaging 2010)

**Figure 7.11** Some examples of terracotta figurines from different areas of the Near East and the Eastern Mediterranean that share a common Hellenistic style, which is visible in the cloth-folding style and the naturalistic facial features. (A, B and C from Metropolitan Museum of Art 2017: inv. numbers: MET\_07\_287\_2; MET\_DP101765; MET\_32\_150\_176; Figure D courtesy of the Museum of Oriental Art 'G. Tucci' of Rome, inv. num. 13153/15644, see also D'Amore 1997)

**Figure 7.12** A house with a Hellenistic peristyle from the Seleucid level of Babylon (Reuther 1926: Fig. 65)

## 7.4 Conclusions

Our discussions have focused on the emergence of hybrid styles in material culture in the pre-AoE and the AoE. Hybrid or syncretized styles can mirror different social phenomena, such as cultural influence, elite emulation, or movement of people (e.g., artisans) across different regions. Hybrid styles clearly existed during the pre-AoE, but evidence of mass migration or the merging of very divergent cultures is difficult to find. In some cases, pre-AoE hybrid styles emerged because of crosscultural influences that spread across different regions through, for example, traded objects. Although this persists in the AoE, it is during this period that greater evidence of movement and more pervasive evidence of hybrid styles are found, particularly for cultural regions that were geographically distant. Larger empires recruited craftsmen from all over their territory and brought them together for public works. These states also created conditions, through either the forced movement created by deportations or voluntary action by migrants, including colonists, that allowed craftsmen from diverse regions to live together. This resulted in the spread of hybrid styles at all levels, from elite goods to more common, non-elite material culture, such as terracotta figurines and house architecture. The intensification of trade is not sufficient to explain the increased presence of hybrid styles, which became the main local style in some places. The movement of artisans is not only suggested by historical sources, but also material culture reflects a new scale of syncretistic developments in styles that suggest greater migration and intermixing of population, a process that increased substantially in the Hellenistic period.

## 8 The development of universal governments

This chapter focuses on governments and governance in pre-AoE and AoE states. The intent is to define how governance began to change during the AoE, as it either facilitated the social integration of multiple populations and ethnic groups or accommodated diverse populations within a state. These changes not only helped to facilitate social change but also enabled larger states to form more easily, and enabled the succession of large states in the AoE. Clearly, governments and policy in any period are complex; the chapter focuses on key trends that show how governing institutions treated their populations in response to facilitated movement. We seek to define emerging trends in approaches to governance as empires became the norm across the Near East. Developments in the AoE help account for the large, multi-ethnic states that began to become the norm and reflect the facts that populations were mobile and ethnic groups were found in places far from their traditional lands.

## 8.1 Pre-AoE governing

While there is little doubt that kingship became the norm with the rise of cities and states, how kings were viewed and how they ruled differed considerably between different states. In Bronze Age Mesopotamia, apart from a few notable exceptions, kings were not thought of as gods, and they often attempted to portray themselves as servants of their gods or even as having been chosen to do the work of their gods. While kings were powerful, their authority could be checked by the fact that they had to fulfil their royal obligations, such as maintaining the temples and being perceived as just (Saggs 2000; Tetlow 2004: 112). Justice, and providing for the oppressed, whereby the king acts as a shepherd of his people, were key characteristics in the official portrayal of kingship. This role included assistance with debts and exemption from taxes, and other decrees (Kraus 1958; Finkelstein 1969). Perhaps it is no surprise, then, that Shamash, the god of justice in Mesopotamia, often became associated with kings (Darling 2013: 23). In Egypt, during the third and second millennia BCE, pharaohs were seen as divine rulers. This may have made ruling easier, as a challenge to their authority could be considered an attack on the state religion. On the other hand, the pharaoh's power was checked by judgement and the concept of *Maat* (Karenga 2004). This means that ethics and trying to be a righteous and just ruler may have helped to check complete or absolute power, similarly to the situation in Mesopotamia. Justice, as a concept, began to emerge and be associated with kingship throughout the Near East and Egypt.

However, justice, for the ruler, was mostly restricted to those who already lived in the lands controlled by the kings or pharaohs. The extension of justice to all people, including those new to the state as it expanded, was rarely evident in the pre-AoE. Foreign lands were something that could be attacked, even ravaged, or fought, to show that the local gods supported the king or other ruler. Governments generally did not perceive it as their duty to help other than their own people; at least, they did not usually indicate that they felt any such duty (Darling 2013: 16). In fact, already by the late fourth millennium BCE, if not earlier, foreigners were forcibly moved as slaves or captives (McIntosh 2005: 167). Larger populations were also being deported, and sometimes the males seem to have been killed (Muscarella 2013: 278). Thus, forced migrations may have begun to reshape pre-AoE populations. However, few records indicate whether these captives retained some cultural identity beyond a single generation. There are few traces of any long-term influence of these populations on their captors. They were probably assimilated into the societies in which they were taken to.

#### 8.1.1 City-states

City-states formed the foundation of political structures throughout much of the Near East in the pre-AoE. This meant that many rulers, including vassals within larger states, were obliged to local constituents, many of whom, at least the members of leading families, the ruler would have known personally. Governing was, in other words, a very personal business. Kings played an important role as intercessors with the gods, and this role required them to perform important religious duties, such as participating in official ceremonies (for example the Akitu (New Year) festival). As conflict had become a near-constant reality for kings, they would have accrued power and their authority may have increased over time (McIntosh 2005: 173). With the rise of cities in the Near East and Mesopotamia, in particular in the third millennium BCE, multicultural populations became more common in urban environments. Texts show that, as new dynasties arose in the urban environments of the Near East, kings often did not emphasize their cultural background when it differed from that of the region they governed. Amorite populations that gained political ascendancy in the Near East in the early second millennium BCE demonstrate this in Mesopotamia. Although the Amorites were a West Semitic population, temples and religious institutions supported by royal authorities in the lands they controlled retained their original structures and gods, and kept East Semitic Akkadian as the primary language (Saggs 2000: 97; Charpin 2012).

#### 8.1.2 Empires

As empires emerged in the third millennium BCE, states had to incorporate larger and more diverse foreign populations. Governing foreign states that had distant populations and potentially varied interests and constituencies often proved difficult. Records from the pre-AoE show us some different strategies. One strategy was to install loyal officials, who originated from the empire's core areas, in foreign cities. For example, Shamshi-Adad, whose small empire was discussed in Chapter 2, installed his sons in different parts of his kingdom (Durand 1997; Charpin and Ziegler 2003). Another strategy was to place a vassal in a city, often a member of the conquered dynasty. Sometimes, however, the larger, imperial states placed a new claimant on the throne. This may have been a strategy by these larger states to gain the loyalty of a new local elite, which might take the view that the new ruler obtained his throne through the patronage of the larger state. A good example of this is King Idrimi, from the fifteenth century BCE, who ruled the city-state of Alalakh but was ultimately loyal to the Mitanni state, the regional power (Greenstein and Marcus 1976; Collins 2008: 33).

In fact, the system of vassal kings generally appeared to work well for kings in the Late Bronze Age: larger states, such as the Hittite, Babylonian, Elamite and Egyptian states, lasted longer than those of the earlier Bronze Age, even in cases where dynastic squabbles and internal conflicts were evident. The Amarna letters from the fourteenth century BCE indicate that vassal kings were often left so much to their own devices that they periodically quarrelled with neighbouring vassal kings (Moran 1992). The intent of the vassal system was to create mutually beneficial relationships, in which vassals would presumably be protected by more powerful kings and in return provide tribute, and troops in times of war. With this strategy, the larger empires were able to conquer and more easily administer territories well outside their traditional realms. But the vassal system also meant that vassal kings could easily change loyalty to a more powerful or beneficial state as it rose. Ugarit, for example, changed loyalties between Egypt and the Hittites in the Late Bronze Age (I. Singer 1999: 627).

Another strategy was raising an official, often a son, from a royal lineage or ruling line that was conquered in the court of the imperial state. This allowed the governed state and its future authorities to grow up within and learn the traditions of the ruling state and become, presumably, more pliable and amenable to the empire's interests. Examples of this occurred in Late Bronze Age Egypt, where Nubian princes were brought up in the pharaoh's own court before being returned to their native lands (O'Connor 1993: 64). This also made governing Nubia easier, because the princes could be held hostage to ensure the compliance of the local populations.

Propaganda in the pre-AoE largely does not show conquered populations or rival populations in a favourable light, even if, at a practical level, they were to some degree integrated into the state or helped to govern the state and its empire. While propaganda in art has to be seen in context, it shows an attitude that is intended to be displayed to people or elites in the homeland, to foreign dignitaries, or even to conquered populations. It also suggests that, in practical policies, strategies for integrating foreigners may have not been a major priority. For example, the victory stele of Naram-Sin and the battle relief of Thutmose III at Karnak have comparable themes: they show the triumphant king not only as much larger than anyone else but also as smiting foreign and defeated enemies (Figure 8.1). Rarely are foreign populations described by official documents or royal texts as being integrated politically and socially. Instead of rulers having pride in the diversity of the realm, their intent was to show foreigners as subjugated in order to demonstrate the king's power and, by extension, the power of his gods (Pu 2005: 53). Foreigners were often important in the maintenance of imperial provinces and were even accepted in the homeland regions of states and empires, where it is likely that they held important roles, for example in the military (108). There is, however, little praise for their role from the central governments, at least in the homeland regions of empires. In short, there was often little or no pride in the multi-ethnic makeup of the larger states

**Figure 8.1** (a) The Akkadian king Naram-Sin (after Jastrow 2005) and (b) Egyptian Pharaoh Thutmose III (Markh 2016) attacking and defeating their foreign enemies

as they grew. Foreigners in states, such as merchants, were sometimes forced to return to their native lands, or they may have had to assimilate (Heinz 1995; Beckman 2013: 206). While some foreigners were moved, in periods of conquest, to the homeland regions of an empire, there seems to have been little celebration of foreign cultures beyond the fact that they had been conquered by the ruling elites. In general, there is limited evidence of ostensible cohesion and openness to a wide array of foreigners from conquered lands, whereas those foreigners would have felt welcome in the core, homeland regions of an empire. Commonly, foreigners assimilated more closely to the existing culture (Beckman 2013: 211). Foreigners may have been welcome at some level, as traders for example, but this does not mean they stayed for long periods, or blended their cultures with local ones to any great extent. In many places, what was foreign was often distinct and had little chance of becoming an integral part of the state. Even where foreign integration into a state is evident, the scale to which it happened was probably limited.

## 8.2 Governing in the AoE

### 8.2.1 Neo-Assyrian and Neo-Babylonian governing

A cursory look at the propaganda of the Neo-Assyrian reliefs that adorned the royal places in the capitals shows imagery resembling what was seen earlier (e.g., in Figure 8.1), in which kings displayed their strength by subduing their enemies, foreign cultures were shown as being conquered, and the other populations depicted were less than equal to Assyrians. In other words, the Neo-Assyrian Empire was similar to the earlier states in the Bronze Age in how it viewed foreigners (Pu 2005: 105). However, key administrative differences began to emerge as the Neo-Assyrian state developed. From the ninth century BCE, Assyrian kings appear to have depended more on trained high officials, who were eunuchs, and on a host of other bureaucratic officials associated with the royal court and the provinces. The empire began to depend on officials, or 'great ones', who obtained their positions, in part, by merit and not simply through family ties or by being related to the royal family. These included governors who were appointed by the kings; people in states neighbouring Assyria would be appointed as local governors (Radner 2014a). Their responsibilities included collecting tribute and responding to the needs of the empire, although they often had a degree of local autonomy.

Assyrian governance generally allowed those who were outside the Assyrian court to work their way into positions of power. In addition to the Assyrian provincial system, vassal kings were able to rule their regions as long as they maintained loyalty to Assyria. In the eighth century BCE, local officials with demonstrated ability may have been increasingly appointed by the Assyrians to administer provinces. The aim of this practice may have been to minimize rebellions as well as to put more capable administrators in place (Mattila 2002; Radner 2015). Given that there were rebellions in the empire, this tactic probably did not always succeed in maintaining provincial order.

While improved administrative practices and better administrators could sometimes make an empire peaceful enough for greater travel, and by extension influence population movement around the empire, other policies may have had a more direct impact on population shifts. A key policy of the Neo-Assyrian Empire was the forced movement of populations, or deportations, to distant regions (Oded 1979; Postgate 1992; Gallagher 1994; Figure 8.2). Such deportations had been Assyrian policy since the late second millennium BCE, and had been used by other states as early as the third millennium BCE (Muscarella 2013: 278), but they became more common and intense in the Neo-Assyrian period. Assyrian deportation policy not only moved people, including elites, around the empire, it also moved them in order to improve the protection, structure and economy of the empire. The last-named includes its agricultural output. The intent seems to have been to integrate deported populations and mix them through marriages between different social groups, shifting their ethnic identities over time as they intermixed and new generations emerged (Luukko 2012: SAA 19 018). The Assyrians even provisioned the deportees. Construction and engineering around and within the

**Figure 8.2** Families being moved or deported (circled) shown in the Lachish Neo-Assyrian relief (after Peel 2010)

capitals, seen on a new and massive scale at Nineveh, can be attributed, in part, to a large number of foreigners being brought to Assyria's chief cities and surrounding regions. A noticeable feature of the people being relocated, as depicted on reliefs, was that they were often not in bonds (Radner 2014a).

Other evidence indicates the importance of deportees to various provinces. In the Harran Census, which recorded people and estates in the Harran region, deportees are evident (Fales and Postgate 1995: 30– 4). Foreign names at sites such as Tushan suggest that deportations were occurring outside the Assyrian heartland regions (MacGinnis 2012). A discovery in the Peshdar Plain in Iraqi Kurdistan, very near the Iranian border, shows that Aramean slaves were probably moved to that frontier region (Radner 2016: 19). In the reign of Tiglath-Pileser III in the late eighth century BCE, large-scale deportations to different provinces are evident: it was recorded that thousands were deported to various provinces (Tadmor and Yamada 2011: 27, Text 5, lines 9b–12). In this case, the deported populations were described as 'united' and even considered to be Assyrians, suggesting Assyrian attempts to assimilate the deportees into the state. Movement of populations to Samaria may have also become significant by the reign of Sargon II (Naʾaman and Zadok 2000). Assyrians were also moved, or migrated, to other provinces, where some may have been given *kapru* or rewarded estates (see Chapter 5).

There is evidence that Egyptian scholars and deportees were in the Assyrian capital Ashur, where the worship of Nabu and Horus, Mesopotamian and Egyptian gods respectively, seems to have been syncretized (Radner 1999: 74, 2009: 225). It is hard to be sure of the full effect of deportations on migration, but it may have been one of the earliest policies in the transformation of Iron Age settlement structures discussed in Chapter 4. In other words, the movement suggested by settlement patterns may have been shaped by at least some of these forced migrations across the empire. Small sites may have employed slaves or deportees to work agricultural fields, while other individuals, particularly skilled workers, were moved to the major cities. A key long-term outcome of deportation is that subsequent generations began to adapt to their new homes. This may explain, for instance, why some former cities were abandoned and not re-established after the fall of the Neo-Assyrian Empire. Another important change was that Aramaic, although not the official language of the court, began to play an important role as a common language throughout the empire and as a language of administration (Radner 2014a). The movement of people and the rise of Aramaic as a common language may have had the effect of facilitating cross-cultural social connections between different ethnic groups, in which shared ideas and commonalities could more easily form. Populations not only had a language that allowed relatively easy oral communication, but also its alphabetic script meant that it was far easier to write, which gave them greater access to writing and facilitated social and economic integration.

Key to maintaining order in the large empire was the 'royal road' system that the Neo-Assyrians exploited, which continued to be applied, although in modified forms, by later empires (Kessler 1997; Radner 2014b). These long-distance and direct roads connected the capitals and key cities, provided rapid access, and helped to maintain order through improved communications in the Neo-Assyrian Empire. Roads to the Assyrian cities of Nineveh and Ashur, for instance, are clearly visible on satellite imagery (Figure 8.3). Royal roads, in general, had road stations (*bit marditi*) that provided rest and facilities, such as extra horses, that enabled rapid communications and movement (Parpola 1987: xiv). These roads were key instruments in governing a growing empire, whereby the king could stay in communication with his dispersed army and governors. Long-distance roads were physical manifestations of more direct and rapid movement across the empire that was sponsored by state authority. Transport aided communication, and the movement of goods and people, within the empire, making it easier to govern a larger state. Improved bureaucracy by the late eighth century BCE, during the reign of Tiglath-Pileser III, probably also helped. This included the use of censuses, for instance in Northern Mesopotamia, and tax-related collections (Postgate 1974; Fales and Postgate 1995). Another key change was that the military became more professionalized, with full-time soldiers now forming the officer core, which allowed the military to campaign all year round. As the empire became ethnically diverse, mercenaries, foreign officers and soldiers were incorporated or conscripted into the military, where they had a greater influence on Assyrian campaigns (Oded 1979: 48–54; Dezső 2012). This military reform probably helped shape rapid gains in the empire in the late eighth century BCE and allowed some foreigners to benefit from Assyria's success. It also probably meant that soldiers stationed in various parts of the empire began to diversify the ethnic mix of areas conquered and occupied by the Assyrians.

During the Neo-Assyrian Empire, a more efficient bureaucracy was created that facilitated a structure that ruled multi-ethnic regions. While oppression and violence were used to stamp out rebellion, the integration of populations through movement, their relevance to the economy, administrative duties and employment in the army, along with the rise of

**Figure 8.3** Trajectories for ancient roads mapped from visible hollow ways (remains of roads) on satellite imagery. Roads that are related to the Assyrian (probably Neo-Assyrian; Nineveh and Ahsur; Altaweel 2008: Plates 16, 17), Hatra/Parthian (Altaweel and Hauser 2004: 64) and mixed Assyrian/Hatra (Parthian) periods are indicated a common language, began not only to mix populations but also to create a structure in which foreign populations could thrive. This was happening even as official propaganda often resembled Bronze Age depictions or showed foreigners as enemies. Populations were now on the move, while communication, administration, revenue collection and the military were organized into apparatuses that could preserve larger states and integrate, to some degree, multiple ethnic groups. Nevertheless, the Assyrians, generally, neither recognized other dynasties nor styled themselves as successors to kingdoms they conquered (a notable exception being Babylonia; Radner 2010), which shows their disregard of most other kingdoms. As indicated earlier, they do show evidence of trying to integrate or even assimilate some populations as Assyrians, including those deported; this policy may have been effective in bringing disparate ethnic groups and identities somewhat closer. While such assimilation may not have differed greatly from that of previous periods, the scale of movement was now much larger, which resulted in a greater mix of ethnic groups and more opportunities to blend different cultural traits.

After the fall of Assyria, some regions previously ruled by the Assyrians did not fragment into small independent states. In fact, it was in Harran, in Syria, that the Assyrian court made its stand in 609 BCE and was finally defeated in 605 BCE. This lack of fragmentation in Syria may have been because of the changed population makeup and the removal of the major towns in the interior regions of the Neo-Assyrian Empire before this. Many of the small states in the Levantine region conquered by the Assyrians, on the other hand, still retained local elites who took the fall of Assyria as an opportunity to seize a measure of independence before they were reconquered (Fitzpatrick-McKinley 2015: 42). While some regions, for example in Northern Mesopotamia, became more closely linked socially and politically, others still showed evidence of desiring more autonomy.

The policies of rule of the Neo-Babylonians were probably similar to those of the Neo-Assyrian Empire and its provinces, where court officials trained in Babylonia would have been important in the administration of the state. Unfortunately, however, few source materials are available from this period, which hinders our knowledge of how foreign regions were governed. In the homeland, Sumerian-Akkadian traditions from long before revived, and Babylonian Akkadian was still used as the main language of administration. However, the processes of change were underway and much of the population probably used Aramaic, which was also used as a language of administration by the Neo-Babylonians (Radner 2014a: 88). As it had done in Assyria, Aramaic served and facilitated communication between ethnically diverse populations within the Neo-Babylonian state. As before, the language may have facilitated social integration and interaction. Furthermore, the Neo-Babylonians continued the policy of deportations, further moving and mixing populations in the Near East, within which some populations were moved to Babylonia and Babylon (Vanderhooft 2003). As in Assyria, this policy had an economic and a security motive. Increasing evidence of diverse ethnic groups could be found in Babylon by the Neo-Babylonian period (see Chapter 5).

### 8.2.2 Achaemenid governing

Policies of deportation, carried out on a large scale by the Neo-Assyrians, never went away in the ancient Near East; in fact, they continued even during modern periods. After the Neo-Assyrian period, however, new styles emerged in how states approached foreigners. An indication of how the founder of the Achaemenid Empire wanted to rule is suggested by the Cyrus Cylinder, which was found at Babylon and was probably made by Cyrus soon after he conquered that city. The document focuses on Cyrus's portrayal of himself as doing the work of Marduk, the Babylonian god; it shows his actions as justified, as he had restored the sacred temples and improved people's lives. He also indicated that he repatriated displaced people and their gods to their lands. The shape of the cylinder, which is similar to other Babylonian foundation inscriptions found in temples, and the Akkadian used within it, indicate Cyrus's attempt to portray himself as a Babylonian king, despite being a Persian (Curtis and MacGregor 2013). Cyrus's efforts to gain favour from foreign populations are evident in the Hebrew Bible, in which he is one of the few Gentiles to be portrayed favourably for allowing Jews to return to Jerusalem and rebuild their temple to Yahweh. This text shows the type of governance and strategy that early Achaemenid rulers attempted to apply, in which a policy of respecting local cultural norms, acting as a local king, and having, or at least displaying, a tolerant attitude towards different cultures was evident.

A well-known and enduring administrative policy developed by the Achaemenids is the system of satrapies, more fully developed by Darius I but initiated by Cyrus, that represented a hierarchical provincial system. The term 'satrapy' is not a clear one, as the Persians used another term, *dahyu-*, for a more common form of administrative government of a country or province within the larger empire, while our views of satrapy have been largely influenced by Herodotus and other, even later works that reference the Persian system (Schmitt 1976; Khatchadourian 2016). Nevertheless, as aspects of the functionality of the Achaemenid system, satrapies, as scholars have come to call them, were intended as provinces that protected the empire, in which the satrap represented the king in parts of the empire, provided soldiers, paid taxes and managed affairs such as civil or judicial ones, on behalf of the state and the province. In effect, this was not unlike a system that the Neo-Assyrians had used. The king also bestowed the position of the satrap. Representatives of the king, including one known as 'the eye of the king', would be sent out to ensure that satrapies performed as the central state expected them to. The Achaemenids largely attempted to keep local bureaucracies where they already existed, which probably meant there was greater diversity in how each region was ruled. Satraps were not necessarily bound to specific administrative units. When governing, satraps sometimes attempted to attain relative power, or even autonomy, challenging direct royal authority. Not all satraps were equal: the central province of Persia seems to have been exempt from taxation (Briant 2002). Provinces within the Achaemenid state, such as Cilicia until 400 BCE, were given semiautonomous status, which provided them with more freedom to govern themselves (Dusinberre 2015: 46). The enduring nature of the satrapies or the hierarchical provincial system throughout the Achaemenid period proved to be successful, as the Achaemenid Empire maintained its very large size throughout almost all of its history. In effect, the system helped the provision of resources to the central state while also protecting it; it gave some degree of power to local rulers and continuity in the types of governance systems that regions may have been accustomed to.

A significant difference from the Neo- Assyrian period is that, rather than showing foreign populations as enemies, there seems to have been a greater effort to show them as willing subjects or even celebrating the fact that the state is now a large, multi-ethnic entity. This probably fed into Achaemenid propaganda that the rulers were unifiers of nations. The attempt to include various populations in the Achaemenid state is depicted in the reliefs at Persepolis, which show foreign populations bringing tribute to the Achaemenid king and the gods of various nations as having helped found the city (see Chapter 5 on Persepolis). Notably, these populations were not displayed simply as subjugated people. Achaemenid inscriptions began to celebrate good administration and the diversity of the realm, depicting the dominion as a fusion of cultures rather than showing the dominance of Achaemenid gods over others (Darling 2013: 34).

Building on Cyrus's initial, relatively tolerant, attitude towards other ethnic groups, official texts written at Naqsh-i Rustam, the royal burial ground, apply the word 'multicultural', or something comparable, to the empire (Schmitt 2000: 29; Daryaee and Rezakhani 2014: 10; Figure 8.4). The diversity of the empire had become the pride of the

**Figure 8.4** The royal burial ground of Naqsh-i Rostam, in connection with which the term 'multicultural' has been used to characterize the empire (Zolfaghary 2010)

Achaemenids, as well as its strength, and Persian identity was not the only one represented even in one of the state's most sacred places. The multi-ethnic character of the state became a way in which the ruling elites displayed the greatness of the state (Briant 2002: 78). The Achaemenids depicted themselves as bringers of happiness to the world and as restoring it to the way it once was, so that people could once again be united, as they should be (Lincoln 2012: 128).

An exception seems to be the Behistun inscription, in which conquered individuals are represented as showing the subjugation of foreign states. The context of that conflict is focused more on the rebellions against Darius's claimed legitimate rule than on seeing the people themselves as enemies. Not only nations are subject to Darius, in this case, but so is Persia: the inscription justifies Darius's claim over the entire empire. In general, references to the 'lands', or nations of people, and to diversity became a consistent theme in Achaemenid official writing (G. Cameron 1951, 1973). To facilitate this diversity, the Achaemenids promoted the standardization of Aramaic, called Imperial Aramaic, in a key policy that probably helped to facilitate communication between the diverse ethnic groups and officials within the large empire (Gzella 2015). This also allowed communities to live together more easily. It became possible to use Aramaic from Libya to Bactria.

Although the king was seen as the 'Great King' or 'King of Kings', titles that had appeared earlier in pre-AoE Mesopotamia, Persian kings did not simply do as they pleased. In addition to taxes, revenue for the royal court came from the king's own estate, which was administered like a household, and rental and other income were generated for the king (Llewellyn-Jones 2013: 78). In other words, the king also earned his money. Achaemenid depictions indicate kings who have taken the titles of pharaohs in Egypt or kings in Babylonia. These titles may not always have been used, but they reflect the fact that the Achaemenids attempted, even after Cyrus, to justify themselves as successors to existing major thrones (Briant 2002). Local elites from different ethnicities, for example in Anatolia, displayed both Achaemenid symbols of power and their own local languages and ethnic affiliations in official seals (Dusinberre 2015: 260). This fusion and syncretism of local and Achaemenid styles and governing could have been part of a larger strategy of holding important regions by justifying the presence of the Achaemenids, while accommodating local customs. Achaemenid architectural styles that demonstrate the aspirations of local governing elites towards a larger, symbolic unity of the empire are evident in different parts of the empire, such as modern northwest Azerbaijan at Karačamirli (Knauss, Gagoshidze and Babaev 2010; Khatchadourian 2016: 150). The policy of integration gave space for local, political display but also incorporated a wider Achaemenid ideal.

Titles and propaganda, however, may have not been sufficient to maintain order. Like that of the Neo-Assyrians, the Achaemenid army was a mix of various nationalities, and included foreign garrisons that were stationed in different regions, such as Elephantine in Egypt, which further helped to intermix populations (Dandamaev 1989: 114). As much of the royal income of the state depended on agricultural production and trade, including taxes, the Achaemenids created economic incentives for populations to maintain loyalty through the promotion of economic growth and gifts. For instance, the standardization of coinage, which may have been used mostly in the western part of the empire, and the creation of infrastructure projects, such as a canal connecting Egypt with the Red Sea, were policies enacted by Darius that promoted trade. Royal estates and silver were given to subjects. The royal roads that connected the various and distant regions of the empire facilitated trade between key cities from Anatolia to the eastern provinces, allowing many to benefit (Briant 2002; Dusinberre 2015: 49). The Greek colony of Naukratis in Egypt continued to thrive in the Achaemenid period, even becoming a major rival to Greek merchants in western Anatolia (Dandamaev 1989: 157; Bowden 1996). The Murashu family's banking enterprise, along with other banking and real estate enterprises from this period, such as those of the Egibi family, is another example of a local population benefiting from the Achaemenid state. The fact that the state did not interfere with these business activities, and even did business with firms without, as far as we know, abusing them, suggests that economic opportunities were available to private individuals from different ethnic backgrounds (Stolper 1985; Wunsch 2009). Entrepreneurs paid a price to the state, for example in the form of duties in a system called *hadru*. This was duty in the form of service, such as providing soldiers, or giving something of value in exchange for land (Kuhrt 2007: 671). However, nothing in this system indicates a heavy burden on landholders or business people. Overall, government policy encouraged economic benefit not just to the core of the empire but to regions beyond it, by acts which stimulated economic growth and created greater incentives for different populations to become part of a larger, multi-ethnic state, or at least to interact and trade with it. To populations within the state, these incentives may have been new, and they may have created opportunities that had never previously existed. They were mutually beneficial, since they created a sustainable tax-based system for the empire to operate under. The Achaemenids, through policies that stimulated the economy, whereby they could politically and economically stabilize regions, may have seen long-term benefits in such actions.

Despite the Achaemenids' generally more tolerant form of governance, they brutally suppressed rebellious enemies, as in the revolt of Sidon in the 340s BCE (Llewellyn-Jones 2013: 30). Other significant revolts occurred in Babylonia, against Xerxes in the fifth century and one led by satraps in the fourth century BCE, that challenged Achaemenid rule in Anatolia. Revolts in Ionia in the early fifth century BCE, and in Egypt near the end of the same century, which latter led to Egypt's brief independence, are two well-known examples, and many other minor or unsuccessful rebellions occurred (Waters 2014). Taxation in the satrapies led to open hostilities between the Achaemenids and the local population (Dandamaev 1989: 114; Briant 2002). Brutal repression and heavy taxation may, at times, have rallied people against the Achaemenids. Perhaps this was the case in Egypt after its reconquest by Artaxerxes III, at which time Egypt may have been looted in retribution, and heavily taxed so that it would be too weak to revolt again (Ruzicka 2012: 196). What the ruling system showed is that while it created the opportunity, through economic incentives and local power opportunities, for populations to benefit from a large state, parties within subject regions were not always satisfied.

The indication that the state endured and was largely intact immediately before the conquest of Alexander the Great suggests that the ruling system was relatively successful and created a system in which many, and not just Persians, thrived. In effect, the propaganda of the Achaemenids that attempted to show a happy unity of cultures did have the real effect of creating some social integration. Many aspects of the satrapy and of other economic and political administrative systems were used in subsequent periods, which may suggest that the core Achaemenid systems, which also build on the Assyrian and earlier models, began to be at least somewhat effective in ruling larger entities. Greek writers after the conquest of Alexander indicate that a number of Achaemenid institutions for governing, such as collecting taxes, were simply copied (Briant 2002: 389). The spread and standardization of Aramaic continued to play an important role in subsequent periods. Overall, the Achaemenid system would go on to be influential throughout the AoE.

#### 8.2.3 Hellenistic states

The Seleucid state showed a desire to maintain many of the systems that preceded it, despite their Hellenistic origin, in particular the key developments in the Achaemenid period that promoted the integration of multiple populations. Seleucid kings portrayed themselves as saviours, benefactors and shepherds of the people. While new cities such as Antioch were founded in a Greek rather than a traditional Near Eastern style, royal propaganda continued to use local cultural symbols (Darling 2013: 38). Important temples to the old Near Eastern deities were maintained by the state, and generally continued to be constructed in a Near Eastern style (Potts 1997: 289).

Rule in Greek-founded cities and colonies was direct, and the old cities of the Near East did not have Greek languages and customs, apart from taxation, imposed on them, although, as we saw earlier, Greek populations began to migrate in greater numbers to cities (Andrade 2013: 40). The satrapy system was adopted and modified, and subdivisions such as eparchy and hyparchy were developed along with military administration. Local law and administration were maintained, and Aramaic continued to play an important role as a common language. Despite centralized rule, semi-independent regions existed within the realm; regions had the right to self-government but had to assert the supremacy of the Seleucid ruler. These regions were probably expected to provide funds, and men for military purposes, to the central state. But these semi-independent regions sometimes proved hard to control; Parthia, for example, became fully independent later in the Seleucid period (Capdetrey 2007; see also Chapter 2).

On the whole, the Seleucid policy accepted that others could keep their traditional practices while blending Greek culture. Seleucus I, who founded the empire, even married a Persian noblewoman, being careful to style himself in a similar way to Achaemenid rulers (Venetis 2012: 153). Change in governance happened more gradually; in effect, governing became another syncretized, blended platform that included Greek, Achaemenid and other Near Eastern styles of administration. The Seleucids also accepted that other cultures could adopt Greek ways and join the elite classes even if they were not born to the system (Garthwaite 2007: 74). This acceptance also gave populations an opportunity to benefit from the new rulers, allowing a form of social mobility as well as physical mobility. Coinage substantially increased through being minted by the state during the Seleucid period, becoming not only widespread within the state but also used internationally. This helped to link populations in an increasingly international economy in which people throughout the Eastern Mediterranean, the Near East and Central Asia could use the same coins in transactions (Le Rider 2003; Meadows 2014).

The largely tolerant policies of the Seleucids were not always universally applied, as is seen when Jewish populations in the Southern Levant revolted against Seleucid rule and there were attempts to supress Judaic practice. Like previous states, the Seleucids applied force in dealing with rebellious populations and forcibly moved populations, including from Babylon to Seleucia on the Tigris, the new city built by them (Boiy 2004: 141). There are episodes of ethnic strife, as in Babylonia, where the heterogeneous populations sometimes clashed, the Greek population even abandoning Babylon at one point. Such events may even have led to increased commodity prices (Pirngruber 2017: 194). In general, however, the policy of syncretizing native cultures with those of the Hellenistic world, which had begun before the arrival of Alexander, accelerated in this period. Ethnic strife in Babylonia may in fact demonstrate the likely increased rate of ethnic diversity, which was now openly expressed. Despite such strife, historical attestations of ethnic conflicts are rare. Hellenism was not imposed, as the maintenance of older traditions and languages shows. The blending of populations continued, and movement was often voluntary and sometimes forced (Sherwin-White and Kuhrt 1993).

Like the Seleucid, the Ptolemaic state is generally seen as tolerant in its practice of integrating local customs with Hellenized practices. A Ptolemaic ruler was portrayed as a 'benefactor' and 'saviour', terms which were parts of the ruler's title. This was more than just propaganda: rulers showed significant investment in and focus on the state's economy and infrastructure. For example, investment by the state is evident in agriculture, as in the Fayum region (Monson 2012). The Ptolemaic court wanted to portray itself as 'Egyptian' to the people it ruled (Figure 8.5). Kings chose, for example, to be crowned at Memphis, a place important for Egyptian state unity from the time of the Old Kingdom (Hölbl 2001: 78). It was the early Ptolemaic rulers that commissioned Manetho to write a history of Egypt and create a list of Egyptian dynasties, which became the foundation of modern Egyptology's understanding of Egypt's royal dynasties. The new Ptolemaic rulers considered themselves to be successors to the older Egyptian dynasties and part of the ancient Egyptian traditions. They practised Egyptian customs and worshipped Egyptian deities, while syncretizing Greek gods with those from Egypt (Bowman 1996).

The Ptolemies created new, Greek elites, who eventually benefited from positions of power, but the old priestly classes and key parts of Egyptian society were not removed by the ruling structure. Meanwhile, Ptolemaic policies encouraged migration to Egypt by Greeks, who were

**Figure 8.5** The Ptolemies were careful to depict themselves as Egyptian rulers, as in this sample relief showing Ptolemy VI, which probably helped them to maintain order in their state (Hobbs 2007)

increasingly present, including in new cities such as Ptolemais (Fischer-Bovet 2007; Bevan 2014: 105). Other Near Eastern cultures were present in Egypt, which indicates that migration was occurring not only from Greece but also from the east and surrounding regions (D. Thompson 2011). Internal movement was taking place within Egypt, as populations were recorded as being brought to places such as the Fayum region. Migration within and to Egypt began to fuel Alexandria's rapid population rise above that of other cities, and the city may have contained more than 10 per cent of Egypt's population during the Ptolemaic period (Manning 2003: 49).

Although the Ptolemies attempted to bring revenues more under their authority, which may have led to revolts during their reign, including one serious enough to threaten their hold on power, they also enriched at least parts of the local population, in particular the priestly class. The Ptolemies, similarly to the Seleucids, attempted to build government institutions that incorporated local governing styles and concepts inherited from previous periods, such as those of the Achaemenids, even where they diverged from strictly Greek forms. Both the Seleucid and the Ptolemaic states emerged as more centralized entities, but granted autonomy in places and mostly attempted not to impose foreign customs on their subjects (Manning 2012; Monson 2012; Bevan 2014). This probably facilitated population integration and fusion between the various cultures.

### 8.2.4 Parthian governing

The Parthian conquest of Seleucid territory, like earlier empires' conquests in the region, led to the retention of many policies from the previous AoE rulers. To an extent, continuity in governing is a testament that policies affecting large regions were beginning to pay dividends for larger states by maintaining stability. Coinage minted by the state was initially conservative, particularly in regions in which substantial Greek populations were found (Colledge 1986: 23). Coins, although they bore the names of new Parthian rulers, retained stylistic elements and the language of Seleucid Greek coins (Figure 8.6). Once again, change was gradual, mostly occurring well after the Parthian conquest of Seleucid territory, and the new state often replicated the policies of the previous empire.

While coins are a tangible manifestation of an initially conservative policy, other policies also indicate the retention of previous styles of rule. In particular, the threefold hierarchy of provincial administration, similar to the Seleucid system of satrapies, eparchy and hyparchy, was retained (Widengren 1983: 1263). The smallest administrative unit consisted of a few towns or villages, and was similar to the Greek division (*stathmos*) introduced earlier. Local oligarchies of rich families from different, multiethnic backgrounds formed the backbone of local and city government (Koshelenko and Pilipko 1996: 141): Parthian policy was to give a stake in the politics, and probably the economy, of the Parthian state to locals. As stated in Chapter 2, the Parthians used the title *Philoellenos* ('friends of the Greeks') on their coins to accommodate their diverse population.

**Figure 8.6** Coins dating to the periods of (a) Antiochus V (a Seleucid king, 163–161 BCE; after CNG Coins 2006) and (b) Mithradates I (a Parthian king, 165–132 BCE; after Classic Numismatic Group 2006)

In addition to a centralized satrapy system, rule was decentralized in various parts of the realm where semi-independent kingdoms existed. Some important semi-independent kingdoms were Armenia, Hatra, Characene, regions in the Persian Gulf, and Edessa. These kingdoms were allowed to mint their own coins, govern their territories, and trade with distant regions (for example, see the roads for Hatra in Figure 8.3), but were required to acknowledge the supremacy of the Parthian king. This acknowledgement probably included the provision of soldiers in times of war, similarly to semi-autonomous satrapies in the Achaemenid period (Daryaee 2009: 57; Grajetzki 2011). While, very probably, there were more areas in which Parthian government may have adapted or even encouraged the variety and mixing of cultures found in the Near East and its increasingly diverse cities, unfortunately many primary sources from this period are absent. Overall, however, existent texts suggest that many earlier policies, particularly those started by the Achaemenids and others developed in the Hellenistic states, continued or were modified.

#### 8.2.5 Sasanian governing

There are few documents from the Sasanian period from which one can reconstruct first-hand accounts of how the empire's administrative apparatus functioned. Many of the texts that are known come from later, Islamic writers. Using the ancient title 'King of Kings', the Sasanian rulers considered themselves not just shepherds of the people but also guardians of the sacred religion. In fact, Shapur I's official title became 'King of Kings of Iran and non-Iran', reflecting and emphasizing the Sasanid view of a universal state and dominion over many ethnic groups during a time of great imperial expansion (Canepa 2009: 54). During the Sasanian period, not only had Zoroastrianism become the state religion, but also priestly classes, in particular the Magi priests, were powerful in influence and policy. Kings were seen as having been selected by Ohrmazd (or Ahura Mazda or Ahuramazda), the Sasanian god, whereby the office of king became divinely ordained (101). The king also had a vizier, or prime minister, a chief general of the armed forces, and ministers who were responsible for the economy and other aspects of the Sasanian state.

Powerful noble families facilitated the governance of different regions of the empire, some of the nobles being Iranian (many coming from the older Parthian aristocracy), and others non-Iranian, in origin. Some members of the aristocracy, the *Wuzurgan* or 'great ones', helped to choose rulers, and sometimes came into direct conflict with the king and his authority. This class held important positions as provincial rulers, and military positions within the empire (Nicolle 1996; Pourshariati 2008). Incentives created by the Sasanians appear to show that multiple groups could benefit from their system of governing, as elites in society included those of non-Iranian origin.

By the late Sasanian period under the reign of Khosrow I (531–579 CE), basic administration was divided into four regions – a quadripartition – for the entire empire. The regions were further subdivided for administrative purposes. The quadripartite system, however, may have been a late adoption. On the other hand, provinces (*shahr*) used the Achaemenid/Hellenic satrap-based provision. However, provinces were generally smaller than before (Daryaee 2013; Frye 1956). They seem to have been ruled by officials chosen by the king. Officials, including priests, played an influential role, which sometimes led to persecutions of non-Zoroastrian faiths, although tolerance of other faiths was also often evident (Daryaee 2013). Middle Persian served as the language of the court and was spoken by many native Persians, but Aramaic still played an important role as a common language between the various populations living in the Sasanian state (Mokhtarian 2015).

An important facet of Sasanian policy was deliberate urbanization, whereby populations were relocated to new cities that were built in this period in different regions. Populations seem to have come from throughout the realm and from rural regions, which reflects a diversity of ethnic groups moving into cities. Urbanization of the Iranian heartland, including the building of such cities as Bishapur, seems to have been a major development, whereby industries were promoted by the state and needed large workforces (Daryaee 2013: 135). Movement was, at least in part, motivated by economic developments in the Sasanian state. As before, state policies encouraged economic participation by many ethnic groups, and the process of urbanism reflected multiple communities that were economically integrated. As in earlier periods, forced deportations formed part of Sasanian policy. However, the act was not always malevolent, as populations were provided with land and living space. In effect, this policy had economic motives, but it also helped quell rebellions or control captured populations, similarly to earlier Neo-Assyrian deportations. For instance, Shapur I was known to have deported Christians to different areas in his empire, but he provided them with land and living spaces. This movement also helped Christianity to spread in parts of the Sasanian Empire, probably with Sasanian knowledge (Delehaye and Peeters 1925; Pigulevskaja 1963). The production of silk and other textiles also seemed to entail deporting or moving populations to the regions in which production was concentrated, such as Khuzestan (Huart and Delaporte 1943). Centres such as Gundishapur, which contained a wellknown academy, welcomed physicians and scholars from many regions, including Greece and India (D. Hill 1993: 4). When the Byzantines closed the famous Academy in Athens and the school in Edessa, philosophers, physicians and scientists were welcomed into the Sasanian state, and specifically to Gundishapur, to where they brought their works (Abivardi 2001: 450).

In the realm of religion, government policy varied, depending on the influence of the Zoroastrian priests and politics in relation to the Byzantine Empire. Jacobite and Nestorian Christianity thrived in much of the Sasanian period, during which Mesopotamia, the Persian Gulf and parts of Iranian Armenia were important regions for Christian populations and early Christianity in general. However, periodic persecutions against Christians persisted, some of them instigated by rival Christian sects rather than the Sasanian Zoroastrians (Walker 2006: 175). Other religions that were tolerated, and even thrived, were Judaism, Hinduism, Buddhism, Mandaeism and Manichaeism, although there were periods of persecution of all non-Zoroastrian faiths (Drijvers 2009; Daryaee 2013; Foltz 2016).

#### 8.2.6 Rome and the Byzantine Empire

While scholars probably know more about Roman administration than about some of the other contemporary empires, such as the Parthians and Sasanians, it is assumed that the picture is still far from complete. From what can be gathered about Rome's administrative presence from the first century BCE until the fourth century CE, that is, when power shifted to the Byzantine Empire in the Roman East, the Roman Empire adopted strategies similar to their predecessors', whereby tolerance and multiculturalism were part of the practice of administering the diverse lands. Some regions were ruled as vassal or semi-independent states. Rome also began to create colonies and to encourage populations to move to cities in distant regions, creating and adding to the region's diversity, which, by then, had well-established multicultural traits (Millar 1994). For example, Palmyra was at times ruled in a manner similar to that in which the Seleucids and the earlier Achaemenids ruled parts of their empires, that is, the region was given a fair deal of autonomy (McLaughlin 2010: 97). Palmyra, and other contemporary cities, such as Dura Europos, have temples to many gods, including those introduced by Rome, while trade and wealth grew to greater levels. Tribute and taxes to the empire, as they were in earlier empires, were of key significance and requirements for regions bordering Rome's eastern frontier (Millar 1994: 49). In Egypt, cities such as Alexandria thrived under Roman rule, as they offered opportunities for people from many areas to migrate to the cities, which helped to increase their importance in international trade through production and opportunities for employment (Koestner 2016).

By the first century CE, Rome was showing more direct interest in key places across the Near East by making its easternmost regions provinces. This move appears to have been influenced by competition with Parthia, which had developed as Rome expanded eastward. However, rather than being repressive, Roman policies produced more direct economic benefits to the Near East. In particular, irrigation projects, aqueducts, canals, roads and other infrastructure that stimulated local economies and helped to integrate trade across the Mediterranean and with Rome itself were developed (Millar 1994; Figure 8.7). This very probably helped the rise of Antioch as a leading city in the eastern part of the empire. In addition, locals became involved in the military of Rome, as the frontiers became militarized during the wars against Parthia and the later Sasanians. This provided an avenue for local populations to benefit from within the Roman structure and gain relative power. Perhaps one example of Roman rule in the Near East is how Judea became more 'Romanized' after its great rebellions against Rome, particularly after the Bar Kokhba revolt. The province of Judea became part of the new province Syria Palaestina, and Jews were forbidden to enter other regions or Jerusalem, at least for large parts of the year (Eck 1999). While this dramatic escalation in Judea shows the factious nature that still existed in the empire, it is also evident that the Near East became well integrated into the Roman economy and state. In effect, this meant regions along the Mediterranean basin could easily trade and interact, which encouraged easier movement of populations and goods. Rebellions and increased war in the Near East against the other large empires brought the region under more direct rule, which resulted in the greater presence of legions based in the Near East in the first and second centuries CE (Isaac 1998). As before, many of these troops came from different ethnic backgrounds, which again helped stimulate social intermixing.

In the third century CE, conflicts with the Sasanian state, and major rebellions, in particular by Palmyra, began to reorient Roman rule in the Near East. The restructuring of the provinces and a more detailed tax system had developed by the second century CE, for which records demonstrate that different types of property had different tax structures

**Figure 8.7** Aqueducts built in the Near East, such as this example in Caesarea Maritima, indicate Rome's attempts to encourage economic development in the region (after Mark87 2007) (Matthews 1984). Rome may have begun to focus even more on the Near East as a revenue-generating region; the rebellion of Zenobia and the establishment of a brief Palmyra-based empire in the second half of the third century BCE may have been motivated in part by economic interests, that is, to gain greater benefit from trade (Southern 2008: 103). During the reign of Caracalla (198–217 CE), Rome opened up citizenship to all its residents, not just elites, which allowed the state to integrate its multi-ethnic groups more fully under the same general status (Schott 2008). Additionally, structures relating to the economy, such as roads and aqueducts, were continually being built, which suggests continued investment by the state. Roman officials were also divided more clearly between civil and military areas, whereas in the past there had often been overlap between these positions. In fact, locals had important governing roles and became high officials within the state apparatus, demonstrating attempts at political as well as economic integration (Millar 1994).

Turning to religion, in the first decade of the fourth century CE the last great wave of Christian persecutions led to the deaths of many Christians. For instance, documents from Antioch attest persecution by the state, under direct orders from Rome (Mitchell 1982: 94). However, this situation changed dramatically when Constantine gained power in Rome in 312. Shortly after this, the state began to sponsor the construction of Christian religious sites and removed some non-Christian places of worship. Persecution of polytheists had begun in parts of the Near East, but this was gradual, or took some time to have a noticeable effect (Millar 1994). In the 380s and 390s, the Roman Emperor Theodosius I actively dismantled temples and banned non-Christian worship, although the bans were not universally applied (A. Cameron 1993: 76). A shift also began whereby the Roman emperor and later Byzantine rulers saw themselves as God's anointed rulers, their rule sanctified under the empire's one god (Canepa 2009: 114–15).

Despite occasional major upheavals, both internal and as a result of wars with the various Persian-based empires, long-distance trade connecting Rome with India and China flourished. Trade with Arabia was another sphere in which Rome and the Parthian and Sasanian Empires were active, perhaps because of the mutually beneficial results for these states (McLaughlin 2014). The continued presence within the region of so many temples to a variety of gods, and of many languages, including Latin, Greek, Aramaic-based dialects and Persian, suggests that Rome's policy was to remain tolerant of multiple ethnic groups, or at least to allow them to benefit from the wider economy even during major wars. Although at times Rome would persecute specific communities, particularly if it interpreted their behaviour as a threat to state authority, there were political or economic opportunities for many. Significant changes in societies were evident, but major social upheaval was usually avoided, despite some rebellions. Even the conversion of the empire's religion to Christianity led to only a gradual removal of the old gods in the fourth century CE, which gave time for change to affect populations gradually and minimized social disruption.

Within a few decades of the acceptance of Christianity in the empire in the fourth century CE, the Eastern Roman Empire, or Byzantine Empire, began to emerge and to control territory once held by Rome. The administration of this new empire largely mirrored that of the previous empire; however, the pace of Christianization intensified after the fourth century CE, and the old polytheistic religions, including the rival Zoroastrianism, were generally persecuted. The now common persecution and edicts to close temples may have helped to weaken the Byzantine Empire by the time of the Arab Islamic invasions in the seventh century CE, particularly in Syria and Egypt, where large populations of non-Christians and non-Orthodox Christians still existed and may initially have seen the Arabs as liberators (Luttwak 2011: 199). Judaism, although at times more tolerated than other religions, experienced less tolerance during the reign of Justinian in the sixth century CE (Kohen 2007).

Fundamentally, the Byzantine state ran on taxes, which were substantial by the seventh century CE (gold making up a large portion of revenues) and possibly served as another factor that weakened the state in parts of the Near East (Oikonomides 2002). On the other hand, the state was very efficient and had a large number of highly trained bureaucrats, and laws to check corruption. The vast bureaucracy and the efficient tax-collection system helped the empire to maintain not only a vast structure of government but also a fairly large army that could keep the population in check and thwart external threats (Luttwak 2011: 9). Like the earlier AoE empires, Byzantine policy promoted private trade, and historical accounts from Alexandria show a large network of trade that reached as far as Britain to the west and Sri Lanka to the east. For instance, trade ships from India docked in the port of Clysma on the Red Sea (Tsiamis, Poulakou-Rebelakou and Petridou 2009: 211). Trade and private enterprise may have promoted the creation of laws, at least in Egypt, to regulate or tax these successful enterprises. There were laws that taxed both wealthy estate owners and workers; the local, wealthy owners were responsible for tax collection, in a type of pagarchy system. This suggests that the Byzantine system may simply have used an existing local economic system, as centralized officials do not appear to have been involved at local levels. For local wealthy landowners this was a benefit, as it allowed them to extend their influence across the countryside, which gave them an incentive to maintain this system (Sarris 2009).

The Christian faith made it possible for those within the Byzantine Empire to rise to high levels in government or religion irrespective of their ethnic backgrounds. This helped to blend cultural groups, giving them opportunities to achieve power, and perhaps led to the merger of ethnic groups. In fact, few disputes within the state appear to be ethnic in character (Treadgold 2000). Although the variety of religions decreased, the unifying faith of Christianity allowed multiple ethnic groups to identify with the state, as the latter incorporated its own religiously based hierarchy and used it to bring disparate ethnic groups together. Perhaps this is best symbolized by the role of Hagia Sophia, a church so grand that it attracted pilgrims from many regions to the imperial capital of Constantinople (Luttwak 2011: 115).

With the emergence of one state, and of one universal faith supported by that state, as exemplified by the late Roman, Byzantine and Sasanian Empires, religion became part of international relations and a force to unify, but also to separate, people and states. State-sponsored persecutions became a way of forcibly making the state universal, or cohesive, through a common belief, regardless of people's cultural and ethnic backgrounds. Powerful armies supported by a large, bureaucratic tax structure facilitated this process. Nevertheless, this did not automatically mean unity, even within Christianity, where schisms between Christians, revolving in particular around the theological nature of Christ, existed. The schism of the Nestorians and their belief in the two distinct natures of Christ, for example, led to social conflict (Bell 2013: 133). This ultimately led the Nestorians to relocate to the more tolerant Sasanian Empire, which shows that universal faiths were a type of political game in which rival empires would host different versions of a religion as well as different religions. Although the Byzantine Empire suffered major defeats at the hands of the Sasanians and later the Islamic Empire in the seventh century CE, it was able to endure for over a thousand years and so become the longest-lived empire in the Near East.

## 8.4 Conclusion

Contrasts between how pre-AoE and AoE empires ruled are evident. On the one hand, early empires were multi-ethnic and diverse; on the other, governing larger territories proved difficult and keeping larger territories together was not an easy task, particularly after the fall of an empire. When large territorial empires emerged in the pre-AoE, they made little attempt to integrate conquered territories socially, economically or politically. Propaganda reflected the glories of the central state or of the rulers and their national gods; foreign populations were often displayed as inferior. There were attempts to transform some regions so that they became more politically amenable to the ruling state, as in Nubia, where subject princes were raised in the Egyptian court. However, integration of entire populations into a seamless political and economic system was not evident in the empires.

What changed in the AoE is that government began to reflect the multi-ethnic character of the state. The Neo-Assyrian and Neo-Babylonian states resembled the Bronze Age states in emphasizing their own cultures, achievements, gods and historical roots, but policies such as deportation and the incorporation of ethnic groups into the military mixed and moved populations, while also creating opportunities for foreigners. Long roads and the spread of Aramaic as a common language probably began to make it possible for distant and diverse populations to communicate, share cultural traits and integrate socially. The Neo-Assyrian state also supported intermarriage between different social groups, which helped to merge populations and create social bonds.

In the Achaemenid period, which was influenced by Neo-Assyrian imperial practices (for example, the establishment of royal roads), a large, multi-ethnic state was celebrated at the government level: the state took pride not in great conquests and the subjugation of foreign populations but in the variety of people found in the empire's territory. This represents an important transition in governance that not only reflected the diversity but also suggests that the state purposely gave new opportunities to various populations within the large state. This is evident from economic records, reforms to coinage, and infrastructure projects that reflected large-scale investment by the state in different areas within the empire. Achaemenid policy continued to move people to different parts of the empire. The Achaemenid system, of tolerating, using, helping, and even celebrating, multiple ethnic groups as a principle of policy and not just in propaganda, seems to have lasted beyond the state's demise and influenced subsequent governments. Policies of accepting foreign populations may have made it easier for populations to move, even on a voluntary basis, into territories. It would have meant that, rather than assimilation being the goal, as seen even in the Neo-Assyrian period, different populations could freely express their own cultures, so that, in time, cultural themes, and thus cultural amalgamation, shared between different populations could emerge. In the Achaemenid period, Greek populations, for instance, are found on the Levantine coast, where they began to blend their material culture with that of the surrounding region (Jigoulov 2010: 196; see Chapter 7).

The standardization of Aramaic, to be discussed in the next chapter, probably made communication between diverse populations even easier, and facilitated commerce and social integration between populations. The later AoE empires kept many of the tenets the Achaemenids had created, including some of the basic provincial and administrative structures. Rebellions during and after the Achaemenids indicate that the system did not always work well, but the evidence that multiple ethnic groups gained opportunities to thrive economically and socially, including through the military, in the large states indicates that empires and states benefited a wider range of groups. Economic incentives helped to create unity across larger states in the later AoE empires, as demonstrated by the pagarchy system in Late Antiquity. Larger empires needed taxes for their continuity, but this meant that they had to allow their populations and provinces to succeed economically within the systems they governed. A system that was beneficial for both the governed and the governing classes was established in the AoE. Regions were also given semi-autonomy to further incentivize their participation in the larger empire. The offer of citizenship, as seen in the Roman Empire, was another way in which states tried to integrate their diverse populations.

The emerging role of universal religions became a new force that was used to unify states; however, it also led to conflict within and between religions. Nevertheless, violence, strong armies, economic incentives and willing converts gradually helped to change the religious diversity once found in the Near East (see Chapter 10 on religion). Social identity based solely on ethnic affiliation had, in effect, a state-sponsored mechanism – that is, the unifying religion – for fading or becoming less central, and universal religion allowed the concept of one state and one god to become ingrained. AoE governments generally maintained and facilitated ethnic diversity, which allowed groups to migrate and to integrate into ethnically diverse regions. However, after the rise of universal faiths, the unifying power of religion became central in keeping social order.

Although the large territorial AoE states probably did not create perfect tolerance, or even durable systems of governance, what can be seen are policies that had success in incentivizing and maintaining their continuity. Policies either forced people to new regions or encouraged or facilitated migrations, which helped cities and regions to become more ethnically diverse. Policies were also reactions to increasing population diversity; opportunities to thrive in these systems emerged that enabled social and cultural integration and amalgamation. Ethnic groups, rather than assimilating to one culture, as often happened in the pre-AoE, began to blend cultural traits, which created commonalities between different populations. Governing even became another form of syncretism that showed this diversity, as ruling incorporated Near Eastern and Hellenic concepts and structures. Governance during the AoE not only reflected the increased diversity found in the Near East but also facilitated further diversity and social, political and economic integration.

## 9 The spread of common languages

The formation of common languages has long been seen as a by-product of empires. Take, for example, the spread of English as the British Empire expanded in the seventeenth to twentieth centuries CE across many parts of the planet (Black 2015: 244). In addition to conquests spreading language, the needs of diplomacy, commerce and day-to-day communication create a requirement between population groups to share a language of interaction. In the ancient Near East, it was only during the AoE that common languages became truly pervasive, in both spoken and written form, among populations, rather than being a limited language of a select group of scribes or officials, as Akkadian was in the pre-AoE. A widespread common language that was spoken and written helped populations to integrate the cultural, economic and political systems that developed in the AoE more closely. This development both facilitated and benefited from the population movement discussed previously and demonstrated below.

### 9.1 Pre-AoE common languages

Even during much of the early third millennium BCE, writing was limited to a few regions in the Near East. Outside of Southern Mesopotamia, a few cities in Northern Mesopotamia, Egypt and Elam, no area has shown substantial evidence of writing in that period. By the period of the Akkadian Empire the use of Akkadian had spread to other parts of the Near East, and this empire may have helped to establish the language as the dominant language of Southern Mesopotamia in regions where Sumerian had primacy (Hasselbach 2005). This picture changed substantially in the early and mid-second millennium BCE, as writing spread to the West Semitic regions and into Anatolia, probably driven by the Near East's increasingly interconnected economies and links to other areas (Liverani 2014: 233). Letters and household business documents become more common in the early second millennium BCE, which suggests that more households had access to writing, in part because of commerce and because there were more scribes in society. With the spread of writing, Akkadian appeared to be one of the first languages one can call common, or at least one can say it appeared in regions where it was not the primary spoken language.

The apogee of the spread of Akkadian was in the Late Bronze Age, during the Amarna Age in the fourteenth century BCE, when the Eastern Mediterranean communicated with the Near East states (Van De Mieroop 1999; Bryce 2003: 224). At this time, Akkadian was probably used from Cyprus to Iran, and from Anatolia to the Persian Gulf in Bahrain (Potts 2006). What is telling, however, is that at the height of the language's use in the region, it is likely that few people could write or speak the language. This should be expected, given the complexities of the written form, which had numerous logo- and phonograms. Interestingly, by the mid- to late second millennium BCE, two simpler writing systems existed, the Proto-Canaanite and Ugaritic alphabets, but for centuries they did not spread or become the main writing system across the region (Healey 1990). In fact, the Ugaritic alphabet died out with the fall of Ugarit. The complexities of Akkadian, and cuneiform more broadly, may have been a hindrance to interregional integration at the economic and social levels, although a very active trade system and network existed in the Late Bronze Age that spanned the Mediterranean and the Near East.

This picture changed after the arrival of the Sea Peoples, or at least during the period associated with their disruptions. In the Early Iron Age, new populations had emerged in the Near East, including Canaanite-derived West Semitic populations and Phoenicians who used an alphabetic script and spread its use. Furthermore, the arrival of the Arameans introduced a new population and a West Semitic language to the Near East (Lipiński 2000). This proved to have significant consequences for the region and beyond in the centuries to come. Groups in the Levant and Syria, in particular, saw the alphabet as beneficial and adopted it, while Aramaic began to facilitate interregional communication. With the expansion of Assyria into Syria and the Levant in the ninth and eighth centuries BCE, the Assyrians increasingly encountered Aramaic and other languages whose writing systems are alphabet-based (Radner 2014a).

## 9.2 Common languages in the AoE

#### 9.2.1 Aramaic

With the Assyrian conquests and the expansion of their empire in the Iron Age, particularly from the ninth century BCE, Aramaic became increasingly important to the Assyrian state as a language of administration (Radner 2014a: 84). The language, in various forms, continued to be either the most common in the Near East or used widely throughout the Near East for almost the entire AoE (Gzella 2015). Its alphabetical script had become convenient for written communication, including common correspondence, for many regions across the Neo-Assyrian Empire.

The use of Aramaic in the Assyrian court, along with population spread, whether through voluntary or forced migration, created and responded to a need not just for a common administrative language between rulers and subjects but also for a language for day-to-day communication and probably commerce. As populations began to intermix, including through Assyrian government policy, the need for a common language increased among populations whose languages often had common Semitic roots. This is evident in places such as Dur-Katlimmu, in the southern Khabur region of Syria, where legal and sale documents contain Aramaic annotations along with Akkadian cuneiform, showing that even areas that had a long-established and good knowledge of Akkadian began to use Aramaic more frequently (Radner 2002). This suggests that Aramaic speakers had spread into eastern areas of the Assyrian state. Aramaic was the first common language with an alphabet-based writing system, which made it more amenable to becoming widely used and employed in many areas beyond official and government-level communication. This is best demonstrated during the Achaemenid period, when Aramaic is used in regions between Libya and Afghanistan; evidence of the use of Aramaic emerges from Arabia and Anatolia as well. Thus, as the alphabet spread so too did Aramaic, throughout the Near East, North Africa and Central Asia (Driver 1957; Gzella 2015).

Although Aramaic was present in many parts of the Near East before the Achaemenids, the Achaemenids standardized the language, introducing one common dialect across their realm. This was not only important for official communication, but also it probably facilitated more common forms of communication such as business transactions and letters, as trained scribes would have been spread throughout the empire and were trained in the same dialect. Day-to-day use of Aramaic was evident in places such as Elephantine in Egypt, Mesopotamia and as far as Bactria (Kuhrt 2014: 113; Gzella 2015). The populations that used Aramaic were also diverse within the regions where Aramaic is found; the Elephantine archive, for example, shows that non-native populations of Jews and Arameans spread into Egypt and brought their languages with them as they migrated (Porten 2011). Commerce and other social activities are seen to be important topics within the inscriptions and documents that have been found. The common language of Aramaic not only made communication far easier, it also allowed different ethnic communities to integrate and communicate with each other as they moved to new areas, including larger cities, and began to conduct business and other activities together.

Widespread communication across the vast Achaemenid Empire created new opportunities that promoted not just commerce but also the transfer of knowledge (Cowley 2005). Papyri such as those from Elephantine show that Aramaic was used to transfer stories (e.g., *Story of Ahikar*), sayings and other knowledge. Examples of the transfer of knowledge through Aramaic must have been more common than our present evidence suggests: many documents written in Aramaic have not survived, as this language was often written on parchment or other more perishable media.

Under Achaemenid rule, writing became more widely accessible as it no longer had to depend on the parochial knowledge of cuneiform script or other non-alphabetic systems. This new development, of a widespread common language, is a key feature that differentiates the AoE from the pre-AoE, as, even in its apogee in the Late Bronze Age, Akkadian never reached widespread or common use among ordinary individuals or even elites.

Aramaic continued to be understood and used, sometimes widely, in Egypt, the Near East, Iran and parts of Central Asia long after the fall of the Achaemenids, although it was generally not an official language of the court. In fact, after the fall of the Achaemenids, more localized dialects of Aramaic emerged, in key cities such as Palmyra and Hatra, among others, which gave rise to two important dialects. Two main branches of Aramaic formed, the eastern and western branches, and they developed more prominent differences over time. The dialects continued to have many similarities, but eventually it may have become harder for those who used the two major divisions to understand each other (Beyer 1986; Healey 2009; Gzella 2015). Nevertheless, in contrast to earlier periods, the differences in the languages of much of the Near East diminished substantially in the AoE, when common languages covered much larger territories and probably had more speakers than in the pre-AoE. Even if some communication problems arose as Aramaic differentiated, the end product of long, successive empires that promoted Aramaic, or at least facilitated its use across large distances in the Near East, left a transformative mark on the region and beyond, where communities that had distinct differences could now communicate more easily, which allowed them to develop closer social links. At the very least, access to a common language that was much easier to master, in reading, writing, and perhaps even speaking, than Akkadian, provided many households with a way of participating in communication systems that had the potential to link ethnic communities spread over great distances.

#### 9.2.2 Greek

Even before the arrival of Alexander, Greek colonies, mercenaries and trade meant that different parts of Egypt and the Near East were already familiar with the Greek language. Clearly, the spread of Greek increased substantially after 330 BCE; specifically, a standardized version, Koine Greek, took root and spread as Greek cultural influence expanded. This emerging standardized language developed from and was influenced by Attic and Ionic Greek. The development of Koine is comparable to the standardized Aramaic that spread in the Achaemenid period, as populations that were spread over long distances now had a language and a dialect that they could all use. Similarly to Aramaic, population movement of Greeks to the Near East, particularly along the Levant, Anatolia, Mesopotamia and Egypt by the Iron Age, and accelerating afterwards, increased the spread of Greek (Tsetskhladze 1999; Horrocks 2010)*.* Earlier Greek populations would certainly have brought their own language, but presumably there may have been a variety of dialects from the various city-states of Greece. After the rise of a more common form of Greek, communication may have been even easier for Greek populations. While the spread of Greek made it another common language found in the Near East, it also spread from the Eastern Mediterranean to Central Asia and India, often coexisting with Aramaic (Strootman 2014). Although by the third century BCE Greek could be found in regions as distant as coastal parts of Spain, and India, at least as the language used by ruling governments and in official communications, it is likely that many populations in the Near East retained Aramaic as a more common secondary language. Greek was probably used more in some places in the Near East, particularly in the cities where more Greeks would have been found as they migrated and integrated into the region, while linguistic change may have been gradual or not as substantial in older Near Eastern cities (Vlassopoulos 2013: 3).

By the first century CE, Greek was spoken and written by sizeable numbers of people in regions between Britain, Western Europe, North Africa, Egypt and the Near East, although how much it was spoken in areas of the Near East outside of western Anatolia is unclear, since Aramaic would also have covered many regions in the Near East (Swain 2003). Certainly it is possible that both languages served as second languages for populations in the Near East. Greek also thrived alongside Latin, the language spread by the Romans. Latin spread throughout Europe and parts of the Near East controlled by the Romans, although it was limited to Roman outposts and official and military institutions. In other words, it is likely that fewer people used Latin than Greek or Aramaic dialects in the Roman-period Near East. Latin also faded from the Near East after the Roman period (Millar 1994: xiv; Leonhardt 2013).

Greek probably became less commonly used by the Sasanian period in Mesopotamia and regions to the east, but it was the official language of the Byzantine Empire in the Near East in the seventh century CE (Horrocks 2010). What is evident is that in periods of greater Greek influence, Greek often coexisted with Aramaic rather than replacing it. In fact, Greek began to influence Aramaic, and Syriac, an Aramaic language still spoken by Christians in the Near East and used also as a liturgical language, includes many Greek loan words. Similarly to the syncretism of Hellenic and Near East themes seen in artistic and cultural styles, Aramaic integrated and reflected Greek linguistic influences. Syriac arose as a later version of Aramaic that reflected the strong cultural intermixing of Greek and Near Eastern cultures as populations began to interact, and subsequently lived together for many centuries (Joosten 1996: 107; Brock 2015: 821). Syriac, at its peak in the seventh century CE, spread as far as India and China, to where, by that point, missionary zeal had spread the language farther than AoE empires ever did (Ji 2007: 41).

Reflecting the cultural and linguistic influence of Greek, Coptic emerged as another syncretized language, combining Demotic Egyptian with Greek (Brankaer 2010: 3). Similarly to Syriac, Coptic emerged during the Christian era as a liturgical language. Already in pre-Christian Greco-Roman Egypt, Old Coptic seems to have developed as Greek became ever more present (Bagnall 2011: 76). Pre-Old Coptic even suggests increased interaction between Egyptian and Greek communities that attempted to accommodate the non-alphabetical Egyptian script even as Demotic and Greek coexisted (Quaegebeur 1991). This early stage of Coptic represented a gradual transition to closer entanglement of Egyptian and Greek that ultimately led to the full development of Coptic.

## 9.3 Conclusion

The widespread use of Aramaic, and later Greek, created new opportunities for having common languages across the Near East and beyond. For the first time, populations from the Mediterranean region could communicate with those as far away as Central Asia; it was also the first time that many people could access writing, via this easier alphabetic communication. As people moved, language moved with them, and the borrowing of words demonstrates how language contact evolved as populations intermixed. In the Achaemenid period, populations that would have had at least some knowledge of Aramaic could be found between Libya and Afghanistan/Central Asia and from Arabia to the Caucasus, an area that covered nearly 6.1 million km2 (Gzella 2015). By the end of the third century BCE, Greek could be found in a territory covering something of the order of 8.7 million km2 , over which government-level communications, at least, used this language, but portions of the population that migrated did too (Siegel 1985: 358; Horrocks 2010). In the late Roman period, one could have used only two languages to communicate from Britain to Central Asia or even into India, namely Greek and Aramaic. The rise of widespread common languages would have opened up unprecedented opportunities for commerce and social interaction, facilitating integration across the Near East at economic, political and cultural levels. People who had social and linguistic differences may have spoken and written to each other far more easily than in earlier periods. In the pre-AoE, Akkadian, even at its peak, used a difficult script, which helped to make it less pervasive than the AoE common languages. Figure 9.1 shows regions where Akkadian, Aramaic and Greek would have been known or used by at least portions of populations in different periods. The clear differences not only reflect how far the AoE languages had spread but also demonstrate the opportunities that would have been created for some level of social and political integration in the AoE empires.

As previously seen, communication was not just becoming easier between populations but also becoming more rapid because of the presence of long-distance roads. With the rise of common languages, communications went farther, faster, and were read and written by far more people. This, along with government policies of more inclusiveness for disparate populations, probably paved the way for people, despite their ethnic affiliations, to participate in the state and a commercial system in a way that gave them a stake, and an opportunity to thrive and to move more easily across different regions. Movement to distant regions may have become easier as common languages developed, allowing

Seleucid/Ptolemaic periods respectively, where at least some speakers or scribes who knew these languages would have been found. The darker colour indicates where the language was primary, while lighter colours indicate government or official use or use as a secondary language (Van De Mieroop 1999; Bryce 2003; Potts 2006; Horrocks 2010; Gzella 2015) some cities to become attractors as language helped to facilitate social integration. Because the same common languages were found in distant regions and within different ethnic communities, moving to distant cities would not have been as difficult for population groups as in earlier periods. Although languages, such as Akkadian, Egyptian and other indigenous languages, continued for a time, bilingualism increasingly became a feature of the wider Near East and Mediterranean world. Aramaic, perhaps the first true lingua franca for the masses, covering widespread regions, thrived for well over a thousand years, and is still spoken by some Christian communities in the Near East and the wider diaspora today. Modern Greek is still similar to its ancient roots, demonstrating the resilience of that language. Remarkably, these first true common languages have never completely disappeared after their initial expansion, despite their replacement in many places. The modern durability of these languages might be a testament to their AoE success and pervasiveness.

## 10 The rise of shared and universal religions

Universalism has generally been associated with government and religious institutions, particularly the monotheistic faiths that arose in the Near East, including Judaism, Christianity and Islam. However, before and during the development of some of these faiths, the concept of shared or common god(s) began to arise as movement became more widespread and regions showed greater social integration and interaction. During the AoE, gods such as Mithras were worshipped in wide areas across Asia and Europe. The factors that made it possible to have common beliefs across distant regions, and may have facilitated the rise of universal faiths, are discussed in this chapter. Religious change helps demonstrate that population movement was increasing, and that it began to facilitate greater social and political integration as populations began living together and develop common bonds.

## 10.1 Pre-AoE religions

Many of the principal ancient gods of the Near East were worshipped over a continuous period that spanned the pre-AoE and AoE periods. Gods and goddesses such as Isis, Ba'al, Astarte, Shamash and Marduk were worshipped for millennia (Holland 2009). Some gods took on a national or even transnational character, but many began as deities simply for a particular place or city. In Mesopotamia, where the city-states played a dominant role in the pre-AoE political landscape, many gods and cities were closely linked. Key myths, such as the creation myth, sometimes interchanged different, local gods, and these gods would be given more prominent roles in different myths and versions of myths (Dalley 2008). The importance of local towns and cities affected a wide variety of mythological stories, because of the influence of the political landscape on early Mesopotamian religion.

Basic to Mesopotamian and wider Near Eastern religions, in general, were the major urban temples dedicated to the patron city gods. These temples were seen as the central dwellings of the patron gods, in which gods would establish themselves and from which they would protect the cities (Van De Mieroop 2004; Holland 2009). Misfortune and anguish for cities and kingdoms were portrayed as abandonment by the patron gods. If the city succeeded, it was because the patron deity had favoured it. Many of the gods that became important in different pantheons were associated with cities rather than states (Walton 2007: 277). Because of the fractured nature of early, pre-AoE politics, recurrent patterns of city-states and the independence of cities from larger authorities in many periods increased the importance of urban rather than national cults. The evidence of settlement patterns indicating restricted movement in the pre-AoE, particularly in the Early and Middle Bronze Ages, fits well with a model of localized deities being important to urban identities.

With the unification of larger states, a different trend in the worship of gods and their associations emerged. In Assyria and Babylonia, state gods soon developed from urban gods. Marduk, the patron god of Babylon, became the state's chief god and the head of the pantheon in Babylonia with the rise of his city's political fortunes (Seymour 2014: 31). Something similar probably happened in the city of Ashur, where the patron god, also Ashur, rose to become the chief god of the Assyrian state (Lambert 1983; Liverani 2014). These developments had occurred by the second millennium BCE, as larger political entities formed and became more stable over longer periods in the Middle to Late Bronze Age.

In Egypt, religious development differed from that in Mesopotamia, as state gods had already emerged by the early third millennium BCE; every nome (a subnational administrative division) had its own set of gods and temples. Because of the early unification of Egypt in comparison to much of the Near East, some gods took on larger, national significance in addition to their local roles. Horus and Osiris, in particular, became associated with Egyptian rulers, where they were significant to all those living under Pharaoh (Traunecker 2001). Amun and Ra, and eventually the combined god Amun-Ra, emerged as chief deities in Egypt. In the New Kingdom Period, Amun-Ra, in fact, was exported to areas outside Egypt, such as Nubia (Stookey 2004). As with Babylon, where the elevation of Marduk corresponded to the rise of the city, Egyptian politics shaped the elevation of deities, such as Amun-Ra, Mut and Montu, who began as patron gods of Thebes, into national gods, and this elevation reflected the importance of Thebes in unifying Egypt after the Second Intermediate Period (Van De Mieroop 2011: 145). Thus, while local deities continued to be worshipped, several gods became associated with the state; even these gods began as locally important deities or patron deities for cities. Egypt's journey towards the adoption of national gods began early as it was unified relatively early in its history.

By the Late Bronze Age, gods from different regions were increasingly worshipped outside their traditional homelands. The Hittites, for example, worshipped many deities of their own, while also syncretizing their gods with Hurrian (e.g., Teshub) or Mesopotamian (e.g., Enki; Bryce 2004) deities. In effect, the second millennium BCE showed that the gods were increasingly shared across cultures. In the Levant, Canaanite gods often intermixed with gods from neighbouring areas, such as the gods Teshub and Hebat who were Hurrian gods (Eliade 2000: 140). Mesopotamia itself also saw the introduction of foreign gods, possibly Indo-Aryan gods, in the Kassite period. In this case, the foreign gods were probably not publicly prominent, as they were rarely mentioned, and did not rise to such high official levels as the local Mesopotamian deities (Kynard 2015: 54). In general, syncretized or borrowed gods were originally found in neighbouring regions. Common types of gods shared in the Near East were storm or war gods, often seen as leader gods (A. R. W. Green 2003).

When gods moved to new areas they were often syncretized with local gods, or even retained the same names and functions in foreign areas. How far they moved is important for indicating how far ideas spread across regions. The sharing of gods, like the sharing of other social concepts and material culture, could reflect social integration or cultural commonalities as people from different ethnic backgrounds lived together. While in the pre-AoE gods were shared and moved across regional boundaries, gods in the AoE substantially increased the distance they travelled.

## 10.2 The spread of gods in the AoE

If movement of people, as has been argued, is critical to the changed nature of the Near East and the wider region, we should expect that in the area of religion movement would transform long-established, pre-AoE traditions. Characterizing religious ideas and change is not easy even for a short period. This section argues that the nature of worship and beliefs in some gods show that gods and belief systems spread in the AoE as people began to move and spread to more distant areas. This movement became more widespread than in earlier periods, reflecting people had moved farther distances. Gods no longer had shrines just in one state or region, as the extent of the worship of some gods greatly expanded during the AoE. Furthermore, the nature of worship was transformed in many regions. Many areas began to share ideas or religious themes as they came into increased contact. Beliefs and worship became more similar as the worship of the same gods spread across ethnic and territorial boundaries to an extent not seen previously.

#### 10.2.1 Religious syncretism

By the Iron Age, syncretism between Egyptian and Canaanite gods is evident, such as the integration of Hathor's qualities with Anat and Astarte (Ackerman 2003: 394). In the Neo-Assyrian period, religious syncretism includes the integration of Egyptian and Mesopotamian gods, specifically Horus and Nabu (see Chapter 8), as Egyptians were deported in the seventh century BCE. Greek populations increasingly came into contact with Egyptian and Near Eastern societies. Cultural influences, architecture and iconography depicting religious and cultural elements from throughout the Near East, Egypt and the Aegean regions are found in the Levant in particular, for example at Byblos (Jigoulov 2010: 81). This is probably related to the Phoenicians' trade networks and sailing activities, which brought them into contact with different populations, including those along the Aegean coast and in North Africa. However, migration and movement of people from the interior regions of the Near East, and increased population in the coastal regions, suggest people were moving their gods and cultural beliefs with them to areas that were active in international trade. Influence also went from the coast to inland regions in the Near East. Astarte, a Canaanite/Phoenician goddess, was prominent in the Achaemenid period, when other goddesses took on her qualities, probably because of the growing influence of the Phoenicians, or Sidon specifically (Orlin 2016: 721). There is also evidence of Greek gods in Anatolia under Achaemenid rule, including a temple to Artemis found in Sardis that combined Persian and Greek influences in its architecture and altar design (Dusinberre 2003: 63). In Elephantine, syncretism of West Semitic gods is evident in the foreign population that lived in that city during the Achaemenid period (Van Der Toorn 1992).

The process of religious syncretism greatly accelerated during the rise of the Hellenistic dynasties, and continued in the subsequent Roman period. Gods such as Isis from Egypt or Mithra from Iran either were worshipped in their own right or influenced the formation of similar gods. Some Near Eastern and Egyptian gods, for example Isis, Mithra and Serapis, developed into so-called mystery cults, for which the worship activities were not clearly written down in any religious text: their ceremonies and beliefs were passed to initiated members, who often came from a variety of cultural backgrounds (Bowden 2010). Many syncretized gods, including Zeus Ammon, Aphrodite-Isis, Serapis and Nabu Apollo, integrated aspects of their Greek and Egyptian or Mesopotamian origins. Other syncretized gods included Jupiter Dolichenus, Zeus Belos, Aphrodite-Astarte and Cybele, who combined Near Eastern with Greek or Roman gods (Beard, North and Price 1998; Ustinova 1999). Syncretism of deities was particularly evident in cities that had strong trade connections (Dirven 1999: 50; Demetriou 2012). While many of these cults spread far in Europe, North Africa, many parts of the Near East and Central Asia, beliefs and ideas underwent changes from the origin of these gods as new concepts were borrowed from other religions.

Generally, gods with similar attributes or functions, related to, for instance, power or love, were combined. Some of these, such as Jupiter Dolichenus, became the objects of mystery cults similar to that of Isis and Mithras (Beard *et al*. 1998: 275). For many of the mystery cults, soldiers, migrants and officials travelling in parts of the Near East and other regions helped foster syncretism as they took their own ideas and combined them with those of the areas they visited or lived in, particularly during the Roman period (Le Bohec 2001: 13; Tripolitis 2002: 56). Syncretism also reflected the style of governing seen in Chapter 8. By combining aspects of Near East religions, outside powers such as the Ptolemies made their governing policies easier to implement, as their attentiveness to local gods gave them legitimacy while they also incorporated their own cultures and gods. This allowed them to create or develop worship and cults that suited their governing interests, catering to populations that were more heterogeneous.

While the process of religious syncretism appears to have accelerated in the Ptolemaic and Seleucid periods, its presence in the Iron Age and Achaemenid periods reflected movement within the region. It is more than likely that the Hellenistic and Roman periods saw more religious syncretism as new populations increasingly migrated into the Near East.

#### 10.2.2 The spread of Mithras and Isis

Mystery faiths illustrate how beliefs spread much farther and intermixed with those of other cultures. Additionally, mystery faiths created ideas shared by many cultures in relation to a given cult. These types of faiths developed during the Hellenistic and Roman periods, many of them having borrowed from Near Eastern and Egyptian beliefs. They spread to the Near East, Central Asia, North Africa and Europe. Such faiths often spread through mobile populations or soldiers going to different regions in the larger empires. The general concept behind them is they had secret rites shared by the community of believers who were initiated into the cults. The faiths often revolved around gods that had already been worshipped for centuries or that were introduced in the Hellenistic or Roman periods, often intermixing pre-existing deities. Mystery cults included those of Serapis and Cybele, which were derived from Egyptian and Anatolian origins respectively but also integrated characteristics of Greek gods (Cumont and Showerman 1998).

Two of the most popular mystery cults were associated with Mithras and Isis. Both were modified and adapted to the cultures into which they spread. Mithras is known in India as Mitra, and is still worshipped by Hindus today. This god has had varying functions over many periods within Indian traditions. Mithra, a Persian god, is usually seen as a god of judgement, contracts and truth, and is often associated with the sun. He is of central importance to Manichaeism and Zoroastrianism. In Roman worship, Mithra was changed to Mithras and was often given a different iconography, as he was associated with the slaying of a bull. However, he continued to be associated with the sun, and with concepts of justice and judgement. Another common aspect of Mithraism is that the god was born from a rock. The clothing of Mithras, such as his hat, was borrowed from Near Eastern traditions or even Persian influences (Cooper 1996; Nabarz 2005). The Romans themselves mention that the Mithraic mysteries originated with the Persians (Nabarz 2005: 53; Foltz 2013: 24). Scholars, however, debate the degree of the influence, and whether it was superficial or had strong links to the Persian beliefs.

One evident aspect of Mithraic worship is how quickly the religion spread across vast areas of the Roman Empire and even beyond (Clauss 2001). Numerous Mithraea, or temples to Mithras, have been found throughout Europe, including in Britain, France and Germany and in the Near East. That they are typically small, underground, hidden structures, or cave temples, suggests that Mithraism was treated as a mystery cult in the areas in which the religion was practised. Most of these structures date to between the first century BCE and the fourth century CE (Figure 10.1). Scholars have suggested that, despite the rise of Christianity, the cult spread as far as Japan in 612 CE (Nabarz 2005: 67).

What the example of Mithras shows is that empires were fertile ground for the rapid spread of new religious ideas, including those that integrated foreign elements from distant regions. Ideas were adapted and cultures intermixed their own concepts with ideas obtained from other regions. However, commonalities also began to emerge and people with multiple or different ethnic backgrounds began to share the same basic beliefs in a common god such as Mithras (Stratton 2000: 306). The spread of Mithraism, and the likelihood that different cultures and regions had common religious elements, suggest that the concept of a shared god became established among widespread cultural groups.

Another mystery cult that became widespread was that of Isis; like Mithraism, it was probably open to many ethnic groups. In ancient Egypt, Isis was seen as the mother of Horus and the wife of Osiris, although her roles vary in different traditions and periods (Witt 1997). Unfortunately, few texts describe what the Isis mysteries entailed, but they combined Greco-Roman worship elements with those from Egypt (McCabe 2008). The mystery cult of Isis initiated by the Greeks applied her Egyptian qualities as a goddess but combined them with Greek concepts, such as those regarding initiation and cosmology

**Figure 10.1** A reconstructed Mithraeum near Saarbrücken, Germany (Anna16 2008)

(McCabe 2008). The influence of the goddess Isis rapidly became widespread during the Ptolemaic period. The Ptolemaic rulers began to identify themselves closely with Egyptian deities, in particular Horus (Ptolemy IV) and Isis (e.g., Cleopatra VII; McCabe 2008: 23). In the Roman period, Isis was worshipped as far west as Britain and as far east as Afghanistan (Witt 1997: 340). A temple to Isis, or an Iseum, could be found in many cities, where her worship was popular with a variety of ethnic groups. Her temples in Egypt lasted until the 530s CE, at which time the Byzantine edict to close all polytheistic temples was enforced (Dijkstra 2008). Much of the spread of Isis in the Greco-Roman world is attributed to her worship having become a mystery faith similar to Mithraism. The worship of Isis has similarities to that of Demeter, a goddess who was also worshipped through a mystery cult (Witt 1997). As in the worship of Mithras, there was a fusion of cultural ideas, and many ethnic groups participated in the worship of Isis.

Figure 10.2 indicates the locations of worship sites for Mithras and Isis throughout the Roman world and beyond. In contrast, the worship of Isis may not have extended far beyond the Levant, Nubia, the Aegean and Egypt in the Late Bronze Age (Lesko 1999). It is likely that few Near Eastern gods were worshipped outside the Near East and the Eastern Mediterranean in the pre-AoE. By the AoE, Mithras and Isis had spread to the western corners of Europe and to Central Asia, and even to Japan in the case of Mithras.

## 10.3 Universal faiths

Syncretism and the spread of the mystery cults reflect a scaling-up in the worship of gods, whereby faiths spread farther and combined with new elements in the new areas. They also show that it became more common for distant cultures and populations to share faith elements. The new universal faiths that arose in the Near East, on the other hand, reflected a new type of religious dynamic. During the AoE, no less than four major universal faiths arose or developed in known literature, namely Judaism, Zoroastrianism, Manichaeism and Christianity. Derivations or subbranches of these faiths also existed, such as Christian Gnosticism. While some of these faiths have branches today that may have less universal claims, during the AoE clear universalist claims either arose or became evident. These religions are defined as universal because they have religious philosophies that claim a central belief is applicable to all societies (Berkey 2003).

**Figure 10.2** Location of mystery cult temples to Mithras (Mithraeum 2016; Clauss 2001) and Isis (Witt 1997; Donalson 2003; Zanda 2011) during the Roman period

Scholars have attributed the origin or spread of universal faiths to empires (Burbank and Cooper 2010: 445). One discussion is about how empires, specifically the Neo-Assyrian and Neo-Babylonian states, helped shape the idea of one god in Judaism, perhaps the first universal faith, and certainly among the earliest (Smith 2001: 165). For instance, in the transition of Judaism to monotheism, universalism could be seen as a survival tactic. In this study the theological concepts that shaped universal faiths are not critical. The fact that universal concepts became more acceptable as empires developed does show that universal faiths have a relationship with the presence of empires. What is relevant here is that population movement helped shape and spread universal faiths and that universal faiths enabled large states to become more durable, even as they also created conflict.

Universal faiths show that shared ideas transcended ethnic and cultural boundaries, which suggests that the exchange and movement of ideas were important in the development of universal faiths, as we have already seen in relation to syncretic polytheistic faiths and mystery faiths. Concepts of divine judgement, good versus evil and the resurrection of the dead are some of the ideas shared by the universal faiths that arose in the AoE. In fact, even mystery faiths probably incorporated ideas such as good versus evil and resurrection, which suggests that universal faiths and other religions shared religious themes that now spanned much of the Near East and Europe (Ulansey 1991; McCabe 2008). Other beliefs common to some universal and non-universal faiths were baptism, shared sacred and communal meal and virgin births (Johnston 2004; Orlin 2016). Disparate and geographically widespread religions, at least regarding their places of origin, were sharing ideas as greater movement and spread of population occurred. Just as knowledge and commerce were moving farther, as discussed above, religious ideas were also moving greater distances as communication and long-distance movement became established. Followers of different universal and nonuniversal faiths were not just communicating their ideas but also moving to and living in new regions, spreading their faiths through proselytization (Ferguson 2003).

While shared ideas among universal and non-universal faiths are evident, the other noticeable factor in the AoE is how quickly religions spread. Like the growth of trade and exchange in the AoE, particularly during the Roman period, movement of religious ideas became more rapid. Mithraism, discussed earlier, spread rapidly across the Mediterranean basin and into Europe and the Middle East. With the aid of social networks and ease of movement, this religion moved easily, or at least fast, among populations. Similarly, the newer universal faiths such as Christianity also made rapid gains between the first and fourth centuries CE (Nabarz 2005; Drake 2005: 8). Both Mithraism and Christianity spread far, but were not dominant religions in most of the areas in which they were found. It took state authority to establish Christianity fully. Shared ideas would be a natural result of increased contact as populations moved and mixed. Thus, even universal and polytheistic faiths began to share common concepts (Fürst 2010: 89–90). Universal faiths also took on the characteristics of mobile populations, as these religions did not require a central temple but could be worshipped in many places (Seland 2013: 384). The deities, looked at in another way, became mobile and omnipresent rather than fixed to specific dwellings or a more limited range as was seen in the pre-AoE.

As discussed in Chapter 8, the emergence of universal faiths gave empires a new vehicle with which to integrate their populations. The concept of one king over many people from different ethnic backgrounds was already well established by the Achaemenid period. Universal religions, in their essence, reflect a philosophy that was already present politically. Although one can only speculate what effect a prolonged period of empires might have had on the concept of one god or religion over many different peoples or ethnic groups, the concept of a universal faith, or one god over many people, became less alien than it would have been in the Bronze Age. Because a unified social and political understanding for many ethnic groups was poorly developed in the pre-AoE, in which many cultural groups had their own clear religious hierarchies and national gods, shared religious ideas and gods over many would simply have been alien to many pre-AoE cultures. An increased possibility of people believing in universal concepts is evident in philosophy dated to the Achaemenid period. The principle and origin (or *archê*) of all things in a universal world-order (*kosmos*) and its manifestations are evident in Greek ideas and philosophers found in Ionia, western Turkey, at the edge of the Achaemenid Empire in the sixth century BCE (Algra 2006). Ideas about beliefs that were relevant to all people were established and even shared before most universal faiths developed, or at least became evident in the textual sources known to us.

Mass conversions and missionaries also became part of the cultural landscape in universal faiths, particularly Christianity. Zoroastrianism was practised by non-Iranians as well as Iranians, but the extent of proselytization is less clear, although evidence exists of Zoroastrian missionaries living in different regions (Buck 1999: 76; Boyce 1996: 255). Judaism also spread through conversions; sources show that adherents who had moved to cities such as Rome converted some parts of the population (Epstein 1994: 101). For religions in general, in addition to religious fervour or belief, force, incentives such as economic benefits, and even social constraints, facilitated mass conversions. Texts show that such methods were used during the rise of Islam to convert communities en masse, which further demonstrates how religions could have spread quickly across communities and landscapes. The conversion of some Zoroastrian and other communities to the new Islamic religion in the seventh century CE was, at least in part, based on incentives to continue or become actively engaged in economic or international trade networks such as those along the Gulf and the Arabian Sea (Lapidus 2014: 271). As another example, conversion to Zoroastrianism seems to have allowed slaves in the Sasanian Empire to become automatically free from Zoroastrian masters (Perikhanian 1983: 639).

While states used incentives, such as economic or social benefits, to obtain converts as a tool for social and political integration, one could also advance in social class as the universal faith was accepted. For states that used universal faiths as part of their national identity, such as the Byzantine Empire, religion became a vehicle to unify people from different social backgrounds. They also allowed people from different social backgrounds to participate in the religious hierarchy, giving opportunities for social benefit to many different ethnic groups. For example, Arab bishops are known to have participated in important ecclesiastical meetings in the state (Shahîd 1989: 523). Universal faiths, while at times divisive, made it possible for socially diverse states to become more politically cohesive, as a common religion transcended ethnic and class boundaries while allowing people from different backgrounds to benefit, within and outside of established religious institutions (Browning 1992: 127). Conversions may have been achieved through incentives or coercion, or by persuading the converts of the truth of the religion. Once the faiths were well established, they grew even faster than when they were new, and mass conversions were possible (Ebrey and Walthall 2014: 95).

### 10.4 Conclusion

While syncretism and the sharing of gods by cultures in the Bronze and Iron Ages are evident, the AoE gods, like other cultural features discussed previously, began to move even greater distances, particularly as the Near Eastern, Egyptian, Persian, Greek and Roman worlds came into increased contact with each other. The spread of gods and goddesses that were once worshipped only at city or national levels, such as Isis, became the norm in the AoE world. Syncretism between gods from more distant areas also increased, as we see with gods such as Jupiter Dolichenus and Zeus Belos. As people moved throughout the larger empires and states, religious ideas were shared across greater distances. Like non-universal faiths, universal religions demonstrated shared ideas, such as concepts of salvation and cosmic battles between good and evil. Mystery and universal faiths spread more easily, so that temples such as Mithraea and other houses of worship become evident in different regions in the archaeological record. Although it remains debatable when exactly monotheism or universal faiths truly arose, it is clear that universal faiths began to be more closely associated with empires. Shared religion and religious ideas between diverse cultural groups became increasingly possible as populations spread and migrated to new areas. Religions also helped states to be more cohesive and enabled a process of integrating diverse populations into economic and other social systems.

Conversion became a way for people from various social and ethnic backgrounds to adopt a common faith, particularly in the Roman and the Byzantine Empires, although this type of religious unity also caused friction within society, as the universalist nature of such religions probably meant that not everyone was going to accept the tenets. In some respects, tolerance of other faiths was found in certain periods, particularly in the Sasanian state, but even there persecution of Buddhism, Christianity and other faiths occurred. The process of Christianization of the Near East and Europe under Roman and Byzantine rule also took some time, and forced conversions seem not to have been immediate but rather to have happened slowly. Economic and social benefits rather than violence, although the latter was also used, probably enabled some of the larger conversions and the gradual transition to universal faiths after the development and establishment of these religions. Such benefits helped states to become more politically cohesive, as they allowed different cultural groups to see the utility of universal faiths. Gradual conversions, and the benefits given to different cultural groups, probably helped prevent greater social unrest. Both universal and non-universal religions allowed people to develop a closer social identity with each other, which allowed populations that had migrated to new cities and regions to live together more easily.

## 11 Characteristics of universalism

The introductory chapter presented the key focus of this work: to investigate why a pattern of long-lived empires became the norm in the Near East from the Late Iron Age. The concept of universalism, the idea that movement led to increased socio-cultural integration and commonalities for population groups, was posited as the result of this pattern of long-lived empires. Universalism also helped to maintain large states and empires as a political norm in the wider Near East, unlike in the pre-AoE. This chapter revisits the issues raised in the first chapter and demonstrates how the subsequent chapters have addressed the primary focus.

## 11.1 Chapter discussions: from pre-AoE to AoE

Chapter 2 demonstrated a historical pattern of small states punctuated by larger empires throughout the third and second millennia BCE. This pattern continued into the early first millennium BCE; however, in the late Neo-Assyrian period (i.e., the eighth and seventh centuries BCE), changes to the social and political landscape became evident. During and after the fall of the Neo-Assyrian Empire, larger states emerged and replaced one another in succession, in contrast with the city-states or small-states pattern that had previously followed the collapse of larger states and empires. The largest empires of the pre-AoE averaged less than 0.5 million square kilometres, while in the AoE larger empires averaged about 3.7 million square kilometres. Wars, as a proxy for political integration, showed much larger territories being conquered by armies, over shorter periods and often through fewer pitched battles. Regions that were once politically independent were now more commonly associated with one another; for example Iran and Mesopotamia often formed part of one state. Although, clearly, vassals and semi-independent kings persisted for much of the AoE, larger political entities dominated even when the new ruling elites were foreign to the Near East. Even though the pattern of conflicts and rebellions that characterized earlier periods continued, long-distance trade thrived and new policies emerged that accommodated multi-ethnic states, which was different from pre-AoE patterns. Chapters 4 to 10 present the key arguments of this book and show how settlement patterns, urban characteristics, long-distance economic exchange, material culture, governments, languages and religions all began to show key social transformations between the pre-AoE, or the Bronze and Early Iron Ages, and the AoE, which covers the period from around the eighth century BCE until the rise of Islam.

## 11.2 Movement and reflected changes in the AoE

The importance of Chapter 4 is that it demonstrates how settlement patterns showed key changes in the Late Iron Age that continued into later periods. Specifically, in some regions very large cities emerged, termed 'primate', meaning they were much larger than surrounding towns. Southern Mesopotamia, southwest Iran and the Levant were regions in the AoE that had one very large dominant city. In contrast, the countryside in some inland regions in the Near East, such as the Jazira and the Khabur Triangle, lost its relatively large urban centres, and small settlements, often less than one hectare or only a few hectares, were more typical. Populations, in the Hellenistic period, increasingly moved closer to the coastal regions; in the Roman period, the Levantine coast witnessed the emergence of a much larger city in the form of Antioch. This transformation, which showed a shift from numerous large and second-tier towns to one or just a few very large cities (primate cities) or to many small towns of roughly equal size (i.e., a relatively flat settlement-size hierarchy) in a region, is explained by migration or easy movement of population. Similar types of population movement can cause both types of the settlement structures described, as demonstrated in a hypothetical case (Figure 11.1). In order to create patterns of either one very large city or many small, similarly sized settlements, movement in the region may have been less constrained than in the pre-AoE. In the case under consideration, settlement patterns reflecting easier (a, c, d and f) and more constrained (b and e) movement demonstrate the types of variation found in regions across the Near East.

The movement outputs discussed in Chapter 4 demonstrate how the settlement patterns observed could have been possible; for most of the results achieved, easier movement than during the pre-AoE is

observed consistently across multiple regions. The interactions needed to create the settlement patterns observed demonstrate that longdistance movement would be a fair possibility for places where it is observed. Qualitatively, and from the evidence of other works, regions around the Levant also demonstrated increased overall settlement, which suggests that some regions attracted increased settlement in parts of the AoE. Movement that created such patterns was not ordinary day-to-day movement but rather demonstrated a pattern that became long-term and ingrained in settlement hierarchies. Even if some of the observed settlement hierarchies were created by migrations which were sudden or rapid, the results of these changes would have become relatively fixed or stable for long periods. Such changes were not uniform; Central Anatolia, during parts of the AoE, did not demonstrate this settlement pattern clearly, until, perhaps, in the later parts of the AoE, from the Roman period onwards. No clear very large, primate city dominates the Levant until the Hellenistic period or later. On the other hand, Southern Mesopotamia witnessed a long succession of primate cities, from the early periods (Babylon) within the AoE until the Sasanian (the Ctesiphon area) and even later periods (Baghdad). The rapid growth of primate cities also suggests that migration rather than natural population growth was responsible for much of the shift in where people lived.

### 11.3 Facilitators and reflections of movement

While Chapter 4 demonstrates that easier movement can explain major changes in settlement patterns across regions where primate cities emerged and other regions developed a flatter settlement/site-size hierarchy, of mostly small sites, Chapter 5 focuses on characteristics found within cities and on what some of the new, small settlements in the AoE may have been like. Both very large sites and small sites show the transformative effects of continuous empires, in which cities became ethnically diverse and new country estates formed. Cities in the pre-AoE were not often characterized by evidence of wide-reaching and expressed ethnic diversity, although cities did have multiple ethnic groups. The largest cities did not just become larger in the AoE, which they clearly did in places such as Babylon, Antioch, Alexandria and the Ctesiphon urban region, they became far more ethnically diverse, as witnessed by their historical data, their linguistic makeup, their temples to various gods, their syncretism of art and monumental architecture and even the knowledge found in them.

The process described above preceded the arrival of Alexander and Hellenistic governments in the late fourth century BCE, as it was already evident by the Late Iron Age. Patterns of universalism appeared in Mesopotamia and the Neo-Assyrian Empire before spreading to the wider Near East. In particular, the process of universalism probably spread during the Achaemenid period, when the Achaemenid Empire created new opportunities. After the arrival of Alexander and the later Romans, universalism accelerated in places such as Anatolia. In the later AoE periods, migration was often voluntary or motivated by religious or economic interests; conversely, the Neo-Assyrian period began to blend Near Eastern populations through large-scale forced migrations. In the Neo-Assyrian capitals, factors such as the great wealth transferred from distant regions, the presence of long-distance roads, and the foreigners brought to Assyria demonstrated the effects of easier movement across wider regions. By the Neo-Babylonian period, Babylon already had documented evidence of various foreign populations. In the Achaemenid period there was a shift towards celebrating ethnic diversity within the state, as witnessed in Persepolis and other places. Foreign influence was welcomed at the Persian court, and artistic styles from a variety of territories influenced palace architecture and art. Large numbers of artisans and foreigners came to live in the great Achaemenid capitals. Later, even smaller cities, such as Dura Europos, showed the effect of widespread, long-distance movement and many languages and gods from across Europe and Asia were found there. Another change in the AoE is that some of the small settlements appear to have developed into something akin to villas or country estates. The presence of empires afforded opportunities for wealthy landowners to build estates for business and residence. Interestingly, trends similar to those analysed in Chapter 4 can be observed in the Western Roman Empire, where few cities had inflated, disproportional populations, while the countryside was characterized by much smaller towns or wealthy estates (Woolf 1997).

The effects of universalism on long-distance trade are analysed in Chapter 6, which shows that, during the AoE, goods moved throughout much of Eurasia, private enterprises with their bases in large cities became more involved in the management of such trade, seemingly with government support or at least approval, and the speed of trade was probably greater than during the pre-AoE. The establishment of long-distance trade corridors connecting distant trade hubs represented by cities was a consequence of the easier movement of people and goods enabled by the large states and empires of the AoE. Trade became more monetized during the AoE, which made economic transactions easier. Incentives to trade now applied to more individuals, so that merchants of different ethnic backgrounds could live in colonies or emporia established far from their homelands, which allowed trade connections to develop over further distances. The drastic reduction in, or even removal of, limiting political borders that had characterized the fragmented political landscape of the pre-AoE positively affected merchants' movements throughout a wider area during the AoE.

Chapter 7 shows that artisans from different cultural milieus converged on AoE cities, either because of forcible political action (for example, they were forced to construct capital cities such as Nimrud or Nineveh) or because of the new economic or social opportunities offered by cities. Artisans could move from one cultural milieu to another more easily under the AoE, thus becoming a primary factor in the diffusion of shared material cultural styles across much of Eurasia. Although the phenomenon of intercultural style borrowing was not new to the AoE, it became more pervasive during these later periods, affecting not only the monumental and luxury arts but also common and non-elite material culture expressions. Artisans brought the styles of their homelands with them, creating the conditions for the emergence of hybridized styles that blended Mediterranean, Levantine, Mesopotamian and Central Asian features. In many cases, texts provide evidence of such a movement of artisans across regions. Style hybridization peaked in the Hellenistic period, and this positive trend continued afterwards, as a consequence of Greek cities which were founded in the Near East, and where artisans mixed Greek and non-Greek stimuli. In particular, the wide diffusion of Greek-inspired artistic features in the AoE, as well as their persistence through time, for example in the art of Gandhara, showed how easy the movement of artisans had become and how multi-ethnic the AoE cities were. The new hybrid styles probably appealed to the diverse inhabitants of the multi-ethnic cities.

Chapter 8 showed that governments and governance transformed during the AoE, reflecting population change and also facilitating it. While, in the Neo-Assyrian and Neo-Babylonian periods, forced migrations and movement of populations diversified regions ethnically, policies encouraged ethnic intermixing through marriages between people of different ethnic groups. Foreigners served as mercenaries and participated in the military and governing affairs of the state at higher levels. Policies that moved people across regions on a larger scale and intermixed populations accelerated the emergence of commonalities between different groups. In the Achaemenid period, rulers portrayed themselves differently from their pre-AoE peers. Rather than celebrating their own conquests, the Achaemenids displayed themselves as unifiers of the cultures found within their empire. Happiness and perfection for all people were depicted as goals that the Achaemenid rulers aspired to, and the unification of different nations under their rule was seen as a means of achieving these goals. All the while, their policies, including forced and voluntary migrations, allowed ethnic intermixing and population movement. Governments still favoured their own elites, but they began to use the diversity within their states as a strength by investing in infrastructure and economic programmes. Rulers represented themselves as inheritors of the ancient crowns of places such as Egypt and Babylonia. The participation of foreigners in the economy became evident, which gave new opportunities to groups to succeed and benefit from larger empires that facilitated commerce through easier movement. Regional autonomy was given to some areas, but obligations were thereby incurred. Subsequent empires retained aspects of the Achaemenid system of governing.

The policies of later AoE empires show that they were ethnically more diverse; they had various national gods and temples, and acceptance of that diversity probably encouraged even more movement. Seleucid and Ptolemaic policies allowed the continuity of local cultural practices and built on many Achaemenid policies, and allowed the movement and integration of diverse populations into cities to accelerate. Greek populations that had migrated to the Near East demonstrate this accelerated process, but others from throughout the Near East were now also coming to such cities as Antioch. This is true of later states such as the Roman Empire, which respected local traditions in the Near East, as well as of the Sasanian Empire, whose rulers presented themselves as kings of both Iranians and non-Iranians, reflecting an inclusive attitude. Rebellions were often brutally put down, but the benefits obtained by local populations, as documented in texts, show that government policies allowed diverse groups to become established and thrive economically in some places. The rise of universal faiths presented new tools for integrating diverse populations by unifying them under one state religion.

The development of a common language, as described in Chapter 9, is another reflection of policies or social development that helped integrate some of the cultures found in larger empires, or at least make it easier for them to communicate. As populations moved across new areas, different groups began to share common languages. Aramaic, initially, and later Greek spanned large areas, connecting many regions from Europe to Central Asia. The continuity of empires in the AoE and their spread to new areas brought common languages to new regions. Shared languages facilitated long-distance movement and social integration, and newly arrived communities would have found it easier to integrate socially and economically into the new cities and other places they migrated to. The greater use of common languages facilitated more and easier communication, the spread of ideas and commerce. Households could now participate in extended trade and social contacts that might stretch across the Old World. Syncretism between Greek and Near Eastern and Egyptian languages was also seen in languages such as Coptic and Syriac. As Aramaic and Greek became the first truly globalized languages, many people, not just a restricted few as during the pre-AoE, were able to speak and write these languages. Shared languages also allowed the distribution of knowledge and the creation of centres of international academic research, such as the Library of Alexandria in Egypt and the Academy of Gundishapur in Iran.

Chapter 10 showed that the gods themselves and common religious ideas emerged and began to move in the AoE. No longer was worship of a specific god restricted to a small area within the Near East; some gods and their temples could be found hundreds of kilometres or more from their key shrines. Gods had already developed into national entities by the Bronze Age. During the AoE, shared religious ideas and faiths, such as the worship of Mithras and Isis, were evident on the eve of the rise of Christianity. The concept of shared faiths between many cultures had developed, and it incorporated more cultures as more populations came into contact. Zoroastrianism, the religion of the Achaemenid Empire, may have been the first of which the doctrine was meant to appeal to everyone, regardless of their ethnic background. At the very least, the rise of universal religions after the Achaemenid Empire reflected a continuity of what had already begun in the emergence of common cultural traits. In philosophy, too, universal concepts such as *arché* and *kosmos*, which had emerged in the Achaemenid period, spread across different areas.

By the rise of Christianity in the first century CE, populations had been widely dispersed, often living together in urban centres, and different ethnic groups were accustomed to having one ruler over them. Universal religions, like polytheistic beliefs such as the worship of Mithras and Isis, had integrated a heterogeneous mixture of religious ideas. These ideas, which included resurrection, final judgement and purification from sins, not only spread but also were shared by multiple, often universal religions, despite variations and different interpretations. The sharing of the beliefs and religious practices of widespread populations became another major cultural transformation as populations migrated and integrated. Furthermore, empires both facilitated the spread and sharing of ideas and helped to generate the concept of shared identity as different cultural groups lived together. Universal religions became for states a strategy for integrating their diverse populations, although this strategy did not always work, as universal religions often had narrower outlooks and more specific interpretations than non-universal ones. Persecutions on religion grounds became more frequent in the Sasanian, Roman and Byzantine states, but the religious hierarchy was a social ladder that different ethnic groups could climb. Universal religions, through proselytization, created new opportunities and facilitated links, such as economic networks, that probably helped to unite diverse populations.

## 11.4 Universalism as theory

The chapters presented are intended to show how universalism, as a pervasive social phenomenon that affected artistic, governmental, religious, linguistic and other forms of cultural expression, not only established itself but also developed to the point where it could perpetuate larger states and empires. The concept of universalism is better understood as a theoretical application that can explain how societies transform from fractured socio-political settings to more socially and culturally integrated societies, and how population movement made this change possible. What has been shown is that population movement is evident from the diversity of cultural traits and expressions found in the AoE. Over time, common cultural traits emerged in the diverse populations of the Near East, where universal ideas and material culture became evident. Universalism is made possible by the increased presence of diverse populations living together in the Near East and surrounding regions. Multiple cultural expressions are found as populations begin to live together, but those populations also begin to display common traits that emerged from an amalgamation of cultures.

Globalization, as discussed in Chapter 1, is a related concept. It is seen as a form of worldwide integration and interdependence of societies and cultures (e.g., see Ritzer 2010). However, movement of people to cities and towns is only one dimension of globalization. The study of globalization has generally focused on increased trade and social contacts, in which mass communication and technology play a critical role. While in the modern world this has led to greater commonalities, which have led to a type of universalism, how this is happening now is different from how it happened in the past. Globalization's definition includes migration, but it is not necessary. There is integration of technology, economy, language and other cultural features, but they are products of other kinds of interaction as well as migration. Furthermore, while globalization is affecting cultures throughout the world today, many regions are not witnessing migration and intermixing of populations. For instance, the influence of Western technologies and cultures on parts of Sub-Saharan Africa is not accompanied by large migrations of Westerners to Africa. Mass media and technologies have fundamentally shifted how the world is influenced by dominant political and economic powers today (Sparks 2007).

The key driver of universalism is movement that facilitates the development of common social phenomena. Close proximity of cultures allows more sharing and communication of cultural ideas. If a government is to be developed that addresses its diverse population, it has to provide space for other cultures to be openly expressed, allowing varied cultural identity to thrive in multi-ethnic cities. Close trade links, a common language, a shared religion and similar artistic forms were often products of close physical proximity and did not come into being only through long-distance communication or trade. As people from very different backgrounds lived together, for long periods and for various reasons, new institutions and common social bonds were needed that accommodated this diversity. Shared cultural attributes also provided new social opportunities to take advantage of population changes. Pre-AoE states did not address or facilitate social diversity in their states, although multiple cultures were often found. Official propaganda and the way regions were governed did not greatly encourage intercultural amalgamation and expression, though intermixing is evident on a smaller, regional scale in, for example the Late Bronze Age. Institutions in the pre-AoE did not fundamentally change to allow individuals from multi-ethnic backgrounds to gain a stake in society, or facilitate their movement to different and more distant areas within states and empires on the same scale as in the AoE. In effect, institutions and social developments in the pre-AoE did not assist larger states to continue, as societies within those states remained culturally and socially distinct.

#### 11.4.1 Movement and social change

Revisiting the conceptual model presented in Chapter 1 (Figure 1.2) allows a summary of our universalism framework. Movement brought about changes in the physical characteristics of cities, such as the presence of temples to a variety of gods, but institutions also made qualitative changes that adapted them to the new social reality. Social structures and norms that were unique became less distinct and began to blend, although a diversity of ideas continued to be found. Over time, many of these ideas became amalgamated, as expressed through art and knowledge. Cultural changes provided new opportunities, for example in trade, and incentivized further movement and new socio-cultural systems. Cultural change such as that expressed in the governments of the AoE, which accommodated more diverse populations, attracted those outside the larger states to migrate to areas that were more diverse. In other words, institutions made changes in response to people who were living in the diverse areas, and those changes encouraged others to move to these areas, so that a positive feedback loop of more social diversity and migration was established. The development of common languages made movement easier and facilitated the social integration of populations, as communication between diverse populations became easier.

We agree that something akin to globalization probably occurred in the AoE. Globalization, at least in some of its aspects, can be contemporary with or even complementary to universalism; some of the evidence in Chapter 6 on trade, for instance, may reflect long-distance trade that was not associated with major population movement. Trade on its own can increase the influence of one culture on another, through observation and through the incorporation of artefacts. However, a globalization perspective alone is not sufficient to explain how social expressions in the AoE changed in response to the diversity of populations. Nor does it explain why large states persisted after the collapse of one empire. For our purposes, population movement includes forced migration, colonization, military service, migration for religious reasons and movement for economic opportunity. Many factors, not just economic interests, created the conditions for universalism.

Migrations that fundamentally changed social structures such as those identified have occurred in other periods, including modern ones. Some forms of migration, such as the labour migrations that occurred in Europe in the 1960s, had little effect on wider cultural institutions, although changes to the law reflect society's way of addressing or accommodating the foreign populations now present in a state (Portes 2010). However, as migrations have become more large-scale, cultural change has become more evident and government and social actions focus more on developing policies in relation to new arrivals and adjusting to their presence (R. King 1993). Population movement can change existing institutions as they accommodate and respond to the newly arrived population and the ensuing social mixture. Subsequent changes to institutions may not only accommodate or reflect the existing migrants and a diverse population, but also encourage more migration to regions, as established communities arise that attract others from their places of origin. This, in many ways, describes some of the migration occurring today into Western states. Large neighbourhoods of foreign populations form in major Western cities as communities attract others from similar cultures. Institutions, social patterns and policies can reinforce patterns of cultural change as they encourage others from their groups to migrate.

Similar change is evident in post-contact South America after the fifteenth century CE: indigenous culture was not simply conquered, it was retained; it integrated foreign elements, and influenced the cultural development and institutions of the incoming Spanish migrants (Wightman 1990). A form of syncretism in native behaviour and actions, including religion, language and art, followed. Although many policies were forced on native populations, subtle cultural change was evident without any initiation from the colonial states. This demonstrates how universal or shared cultural developments can arise without policies that intended such outcomes. In the Islamic conquest of the seventh century CE, a wave of Arab migrants came to cities in the Near East, and this did in fact lead to noticeable changes not only in religion but also in the blending of new ideas in architecture, literature and art. In Persia, Arab migrants influenced Iranian culture, but Iranian culture also changed and influenced Arab migrants, so that the Arab and non-Arab populations showed greater integration with each other. In subsequent generations, the styles of Arab and non-Arab groups became less distinguishable. In effect, the traits became more universal. Islamic institutions, although brought by Arabs, were transformed by a century of contact with non-Arabs who had converted (Lapidus 2014). Learning institutions taught what was already well established in the Sasanian period, but blended it with Islamic thought and philosophy, including matter absorbed from other philosophies. Baghdad became one of the most cosmopolitan cities of the Abbasid period; an influx of diverse cultures and influences through migration helped its House of Wisdom become an international intellectual centre (Lyons 2009), and led also to new religious ideas within Islam, such as Sufism (Karamustafa 2007). As Baghdad emerged as an important centre for knowledge and Sufism, more migrants and adherents were attracted to the city. Movement to the city became a feedback mechanism that facilitated social change, and that social change led to further ease of movement, which encouraged more migrations from greater distances. This added to the positive feedback cycle, and other social changes occurred as more diverse ideas and influences became blended into the growing centre.

Conceptually, this type of change and growth in ideas and social change, along with population growth in the largest cities, such as Babylon, Antioch and Ctesiphon, shows similarity to Bettencourt *et al*. power law and scaling relationships. As population scales to higher levels, change becomes evident in other social characteristics that reflect that population change. As cities grow, they can displace greater and more distant trade and interactions, and lead to innovation, where more knowledge is shared and transmitted; such changes reflect new population levels. Limits to growth occur in the form of the exhaustion of available resources, changes in political or environmental circumstances, or even technology limitations that constrain how a city can grow. Typically, such growth cycles have been associated with economic growth, but feedback growth affecting social and cultural institutions based on migration is possible. As shown above, political and religious reasons can initiate movement and attraction to regions or cities. Conceptually, not only does what was shown in Figure 3.2 – that variable urban growth can be affected by return of attractiveness and ability to move – begin to have urban-specific influences, but also wider, regional transformations result as populations adjust to change. In other words, entire settlement structures and patterns respond to and interact with growth regions and centres.

Changes to urban patterns can happen at different rates; the costs of and limits to growth imposed by social and environmental constraints may reduce the capacity of large cities to attract people. Return of attractiveness and movement may reflect the limitations of resources. The feedback effects that create exponential growth or decline, or slow change in population, are reflected in the SIEM model presented in Chapter 3, which gives a formal model that explains how city populations can change under variable circumstances. In the Neo-Assyrian period, settlement structure change suggests migration. In the Achaemenid and later periods, movement was probably easier, as is evidenced by the more diverse populations found in cities and regions even further apart. Easier movement, and concentration of wealth and people, made it easier, over time, for regions to be incorporated into larger empires, which may explain why it took far fewer battles to conquer vast territories in later periods than in earlier campaigns such as those of Assurnasirpal II in the ninth century BCE. States concentrated their key resources, including their finances, in fewer cities over larger areas in which they held greater political and economic power, as easier movement attracted power and wealth to them. Empires, obviously, always had enemies, and this created borders, which at times made movement more constrained than it might have been. Perhaps this circumstance prevented some of the largest cities from growing even larger. Travel over great distances and migration were limited by the technologies available, including transport to bring food and resources to cities easily. Although, as demonstrated in Chapter 6, it became easier to move objects throughout the Old World in the Roman period and later, limitations to population movement probably restrained population growth in primate cities.

## 11.5 Complex systems theory: why large states continued in the Near East

The evidence for the growth of cities, and the limitations to that growth, along with the social change that occurred, can be explained using concepts from complex systems theory (N. Johnson 2009; J. H. Holland 2014). This theory can be useful to explain how cultural and social institutions changed as settlement and urban structures transformed. Complexity is not separate from the process of universalism described earlier. Rather, universalism, a process that explains the emergence of commonalities and that feedback into emergent systems, can be seen as a case of complexity. Complex systems can be understood as being composed of multiple actors and sub-groups that interact and that affect change and emergent behaviour and systems in a non-linear manner. The change is not proportional to the input provided by any actor or group; even simple interactions among actors at micro-levels, as in the case of migration, can lead to complex social outcomes, in a similar manner to what Bedau (2008) calls 'weak emergence'. For some universal concepts, such as common language or religion, population migration may initially have some limited effects, such as minor linguistic change, including shared words. However, as changes occur at some social levels, the combination of multiple changes may begin to affect other levels of society. When levels of migration are high or significantly intense, major institutional changes, including those that relate to governments, languages and religions, can emerge whereby those institutions begin to develop fundamentally different ideas that reflect social diversity.

What complex systems theory can help explain is how, as people interact through closer proximity, changes in larger social structures become evident. Bottom-up influences through small, personal interactions lead to larger, institutional changes. Such changes can occur at different timescales, but population infusion from different social groups, if its level is sufficiently high, is likely to lead to more pronounced institutional changes that cause traits to be shared by different cultural groups. These complex systems are also adaptive and can learn; they are 'complex adaptive systems'. In these systems, influxes of new social groups can influence how subsequent social, institutional changes occur. We saw this in cases such as the Hellenistic evolution of the satrapy system, in which new elements were added, and adapted, to a new social setting even though the system was learned from the previous style of government. Systems that create large states adapt and change to meet evolving conditions, often building on previous experiences and events that shaped the wider Near East. Many of the later empires adopted earlier methods of governing, modifying them, but not radically, as they learned from previous events. When universal religions became established there were more substantial social changes, but even in this field the change was gradual, and the polytheistic identities of populations changed only slowly – over centuries – across the Near East.

The continuity of large states and empires was not easily reversed, as populations, social change caused by movement, and concentrations of wealth made the Near East amenable to heterogeneous populations and to government by a single entity over vast distances. The political effect of universalism in the Near East was to facilitate the perpetuation of large states after the collapse of one state or empire. Institutions created incentives for populations to intermix and even to depend on each other, as in trade and military affairs. With fewer but larger significant centres, a smaller number of cities needed to be overcome than in the pre-AoE to gain political control over wider territory. Large areas of the Near East, such as the Jazira and the Khabur Triangle, had low-density settlements, while other regions, such as the Levant, grew into socially linked, important trade destinations that led to greater co-dependence between surrounding settlements. This made it far easier for larger territories to be associated politically. Socio-cultural changes, which led cultures to live closely together and to share many traits, became part of the social fabric over many generations as populations adapted to and learned from each other. Above, it was shown that governments had begun to facilitate and even enhance the multi-ethnic makeup of their states. This is another adaptive trait that was probably an outcome of increased population intermixing and encouraged even more migration. Common languages provided ways in which people could benefit more in a diverse society than if they were living in isolated communities. The rise of universal faiths created social bonds that went beyond ethnic identities. The emergence and continuity of universal concepts (e.g. religions) and large imperial administrations shows that institutions had adapted to the diverse populations within cities and across large states. Perhaps more importantly, those populations themselves adapted to the idea of one large state in which changes in the dynasty or state that governed did not lead to a re-emergence of the fragmented states common in the pre-AoE. The concept of 'King of Kings', or a ruler over many nations, became politically more palatable. Worshipping one's gods, speaking one's language and interacting with others from the same cultural background did not have to mean living in one's original homeland. After common social traits had begun to be established, empires and states had developed strategies for governing multi-ethnic and diverse states, and long-distance economic interdependencies had become the norm, few, but very large, urban areas arose, and universalism became a pattern not easily reversed.

The pattern of long-lived empires and large states represents a relative steady state or equilibrium which established itself after major changes caused by population movement. A part of complexity theory is the concept of dynamic equilibrium (Ramsey 2003: 87), in which a steady state emerges even during disruptions and changes that may, periodically, alter the wider system of larger states. Empires and large states are generally maintained, as the social systems within are quite resilient. Changes can happen over long historical cycles, in which a steady state emerges after a long period, but they can also occur quickly. In Near East empires, disruptions would reflect change, but the system was resilient, and regularly bounced back to its steadier state of large states throughout the AoE.

The equilibrium of empires and large states that occurred in the Near East was dynamic: political and social changes often occurred; new ideas and population groups were further integrated, along with new political powers, but larger states as a political reality in the Near East remained. Goldstone and Haldon (2009: 26) have indicated this for the wider Near East, as empires became the political norm. The political, social and wider economic systems that were created became self-perpetuating to the degree that, over time, which power was in charge did not matter. Change in the political order was continual, but states of similar size or sometimes even larger managed to emerge. Disruptions proved to be short-lived, as social and economic links now stretched across much wider regions, which enabled states to incorporate much wider territory. The incentives and structures that helped create larger states, including the spread of populations, wealth and trade, became too well established to be easily reversed. Larger states became something that populations got accustomed to; they became accepted by everyone from the lower social classes to the higher political elites, and the social bonds between the governed and the governing reflect an adaptation to more diverse populations.

During the pre-AoE there was a dynamic equilibrium, but that equilibrium point, during many periods within the pre-AoE, was of small or city-state entities, with the notable exception of Egypt, which was often unified. Empires emerged in the pre-AoE, but they did not remain stable for long periods. Settlement structures did not change substantially over many periods within the pre-AoE, during which patterns of material culture, trade, governance, language and religion do not suggest the population intermixing evident in the AoE. Not only was the emergent system in the AoE resilient, but also continued population movement reinforced an established pattern, which made large states and imperial systems the norm even during inevitable political change.

## 12 The impact of universalism

Given the abundance of data covering the pre-Islamic Near East, there does not seem to be a need to continue the argument of universalism into the Islamic period to demonstrate the significance of long-lived, large states having become the norm in the Near East and Egypt. One can, however, use information from the rise of Islam to show the continuity of universalism. Even as the old, polytheistic religions disappeared, cultures found new ways to create common social bonds that also allowed large states to redevelop. On the other hand, the modern Middle East has become far more politically fragmented than during the AoE.1 To conclude this book, and to demonstrate its wider relevance, this chapter discusses the continuity of universalism, the long-term significance of this concept for understanding later periods, and how current events in the Middle East may reflect an entirely different process.

## 12.1 Later evidence of universalism

With the rise of the Islamic Empire and the Umayyad Dynasty in the seventh century CE, an even larger entity than before emerged in the Near East and beyond. The apogee of empires originating in the Near East was reached in the eighth century CE, when the Umayyad Caliphate stretched from Spain, and at one point Islamic armies reached central France, to India and Central Asia (Egger 2016). The cities of Samarra and Baghdad were among the largest anywhere during the Abbasid period, and may have been the largest until as late as the nineteenth century. Baghdad grew to 7000 hectares or more (Kennedy 2006: 156), possibly seven times the size of Babylon at that city's peak in the AoE. Samarra was another of the great political capitals of the Abbasids, to the north of Baghdad, and is now one of the biggest archaeological sites anywhere, with an area of around 7000 hectares (Lassner 2000: 175).

Although the breakup of the Abbasid Empire was well underway by the ninth century CE, large political units characterized the Near East and North Africa until the end of the eleventh century CE. In the tenth and eleventh centuries, the Fatimids formed an empire based on Egypt that extended into parts of North Africa and the Levant (Brett 2017). At the advent of the Crusader period in the twelfth century, the wider Near East, in particular Anatolia and the Levant, had fragmented into several small states (MacEvitt 2008: 3). This return to small, fragmented states was probably ushered in by the combination of successor competition between the Seljuks (i.e., the Sultanate of Rûm**)** and the influence of European Crusader conquests (Tyerman 2008). The Crusader states became political players in the region, adding a new dynamic that altered the previous political balance, and influenced competition among neighbouring states. In the late twelfth century, the rise of the Ayyubid Dynasty, based in Egypt, enabled a large state to re-emerge across areas that were fragmented. This empire, including its dependencies, stretched from Libya to Iraq at its peak. In the thirteenth century the Ayyubids were succeeded by the Mamluk Dynasty, which lasted until 1517 (Holt 1986).

With the arrival of the Mongols, and the subsequent Black Death, the population of the Near East probably declined sharply, and urban life was disrupted in the thirteenth and fourteenth centuries. The population of Cairo, for example, was reduced, possibly by half, in this period (Lockard 2015: 292). Baghdad's population may not have recovered fully until the twentieth century, as the city was devastated by the Mongol invasion and then the Black Death. Nevertheless, the Ilkhanate formed a large state during the thirteenth and early fourteenth centuries, despite major disruptions. In fact, this period ushered in the Silk Road's last great period of relevance in international trade, which included the famous travels of Marco Polo (Barisitz 2017).

In the fifteenth and sixteenth centuries, large areas of the Near East were conquered by the Ottomans, and for the next four centuries the Near East was often characterized by competing Iranian and Ottoman dynasties (Selvik and Stenslie 2011: 8). In the sixteenth and seventeenth centuries, the Ottoman realm stretched from Morocco to Iran and from Vienna to Somalia. However, vassal states, such as Kurdish states in Mesopotamia and Iran (the Baban and Ardalan dynasties), arose that allowed the Ottomans and the Iranians to create buffer regions and govern their wide realms more easily (Eppel 2016: 35). Throughout this period, very large cities dominated in an era of large empires. Istanbul, for example, was the peak of the urban hierarchy in the Ottoman Empire; it was a primate city in a similar fashion to cities in earlier empires in the AoE such as Antioch or Babylon (Goffman 2002). From the sixteenth to eighteenth centuries, it is very possible that Istanbul was one of the largest cities in Europe and the Mediterranean region, only Cairo being anywhere close to it in population in the surrounding Near East and North Africa (Behar 2003: 1). Laws and actions were put in place to limit migration to Istanbul, as the city had become too large for the authorities to control (Kasaba 2009: 61). Movement was too easy, which allowed Istanbul's population to spiral to a level that created difficulties in infrastructure and policing. In the nineteenth century, cities throughout the Levant, Anatolia, Syria and Iraq were closely interlinked through trade and transport, movement facilitating the intermixing of populations and the transport of trade goods (McMeekin 2010; Schayegh 2014: 36). Such links displayed social interconnectivity in the region similar to that in the AoE.

Although the Ottoman Empire's dominance began to fray in the eighteenth century, the Near East around 1800 was still composed primarily of two large states, the Ottoman Empire and Persia, the latter of which had a succession of dynasties (the Safavid, Afsharid, Zand and Qajar dynasties). There were small states and tribal entities in the regions of Arabia, and some regions also acted as buffer states between the Persians and the Ottomans. Although smaller states did emerge, and regions in parts of the Near East sometimes lacked state authority, the longevity of the Umayyad, Abbasid, Seljuk, Ayyubid and Ilkhanate Empires*,* various Iranian empires and the Ottoman Empire showed the potential for very large states to continually emerge and politically dominate regions within the Near East and beyond for centuries after the AoE (Figure 12.1). The large Islamic-period empires, at their peak, reached areas comparable to or greater than the largest AoE states, such as the Achaemenid Empire. Other dynasties, such as the Mamluk and Persian dynasties, also produced large states. As in the AoE, patterns of very large, or primate, cities, which dominated large areas that contained few comparably sized cities, were found in some of these periods. Movement and socio-political integration appear to have continued in many areas, and many cities were ethnically diverse. The Fatimids, for example, had an ethnically and religiously diverse army that contained Christians, Africans, Armenians, Kurds and others (Lev 1991). Baghdad, as discussed in the last chapter, was another example of a cosmopolitan city. Cities, primate cities and large states continued to be largely multiethnic through the nineteenth century, albeit with some disruptions to this pattern, as seen during the Crusader period. Between the seventh and nineteenth centuries CE, the pattern of small states was still often

**Figure 12.1** Approximate areas of Islamic empires (in millions of square kilometres) and the Achaemenid Empire for comparison. Periods 1–5 represent the Umayyad, Abbasid, Fatimid, Seljuk and Ottoman Empires respectively

an exception in parts of North Africa and the Near East, similarly to the situation in the AoE.

## 12.2 Impacts of universalism

One of the observations that motivated this volume was that so much in the Bronze and Early Iron Age worlds of the ancient Near East seems unfamiliar to us, while the AoE reflects the prevalence of more familiar concepts and institutions. The complexities of the polytheistic cults that developed, the wide range of languages found in the ancient Near East, including those with difficult-to-read-and-write non-alphabetic scripts, and even the material culture that seems so distinct in different regions, are all characteristics seen clearly in the pre-AoE. We argue that, in the AoE, movement is not just a phenomenon that occurred but is fundamental to an understanding of the rise of institutions, norms, social integration and material culture that have a clear influence still seen today. Common bonds emerged that allowed populations to be more socially integrated, through political, economic, religious and other institutions.

Globally significant developments from the AoE that still shape our modern world – banking, coinage, investments, global trade, common alphabetic languages, universal-style government based on ethnically diverse states, ancient science, universal faiths, and even ideals such as multiculturalism – emerged, spread, or became more common, or greatly evolved, during the AoE. A simple comparison of the Bronze Age world and our own world today would show that the more ancient pre-AoE world was far more alien. In the AoE, large cities, comparable to our global metropolises, became places for multiple cultures to interact and live in, in which it was often accepted that multiple places of worship and belief systems could coexist. Cities could grow far beyond the physical limitations of the surrounding hinterland which were evident in the Bronze Age; they could receive goods on such a scale that their populations could reach unprecedented sizes. The urbanism of the AoE, in many places, did not resemble the urbanism of the pre-AoE. The use of coinage spread during the AoE; kings and emperors, sometimes calling themselves 'King of Kings', inscribed their faces on this medium of exchange, which became common across and between empires. People became accustomed to writing and speaking in multiple languages, while common languages facilitated multi-region and multi-ethnic interactions. The first two widespread common languages, Aramaic and Greek, survive, in different forms, today. The existence of libraries, including that at Alexandria, shows that knowledge could spread globally, as information accrued could be widely distributed and also be collected in a single central place. Governments attempted to realize universal ideals in the way they governed, and began to integrate and incorporate their foreign populations under one sovereign. Official propaganda celebrated diversity rather than showing different cultures as enemies; government institutions showed the hybrid effect of including multiple cultures. The economic systems of long-distance international trade by land and sea, which we have come to expect today, were also evident in the AoE. The surviving universal faiths are perhaps the clearest impact on our own world. Religious ideals and beliefs came to be shared across multiple cultures, something distinctly different from the pre-AoE.

The simple action of moving from a homeland in which one had lived for a long time and whose population was familiar to live in more ethnically mixed areas had a profound influence on societies in the AoE, as it does on our own societies. In Western states today, migration has become a major and politically divisive issue, as elections in mainland Europe, the UK and the US in 2016–17 have shown. Large-scale migrations are again evident and have begun to transform cities, including the world's primate and economically powerful cities. The elections indicated may be reactions to an increasingly multi-ethnic world. The AoE serves as a long-term example which could be useful for assessing the effects of migration in transforming societies and cultures. How will longterm migration affect Western states? While this is not an easy question to answer, the lesson from the AoE is that greater hybridization or syncretism is a strong possibility. Today's mass communication technologies make socio-cultural influences both different from and easier than those in the past, which may mean that an accelerated pace of change does not require as much population movement as previously seen (Friedman 2007). Another lesson from the AoE experience is that, despite the social tensions that population movement inevitably brings, large and resilient political entities, including new economic institutions and the knowledge that emerges from them, are possible outcomes of ethnically diversified and interconnected populations. This lesson may be particularly relevant to today's European Union and United States, which are at a crossroads between more integration, on the one hand, and more social fragmentation, on the other, and where the borders of states are being reasserted as harder boundaries.

## 12.3 Could today's Middle East reflect the reversal of universalism?

A benefit of this book is that it offers insight into how larger political entities can form while leading to more socially integrated societies. Although the AoE offers case studies that can be compared across time and for different societies, the region of study, at least politically, has now seemingly reversed many of the trends witnessed for the AoE.

As we look at the Middle East today, it is hard to imagine that this region once showed far stronger political cohesion. With the rise of the post-Ottoman states in the Middle East, connections between cities that were interconnected socially and economically as recently as the late nineteenth and early twentieth centuries (Özyüksel 2014) began to weaken. The new political realities of the twentieth century meant that smaller regions exercised greater political power. The 17 countries and authorities that make up the modern Middle East and Egypt represent the diversity of powers that have arisen, based often on Western concepts of national and political boundaries in the post-World War I era.

There are remnants of a world that was once more socially integrated. These include the widespread use of Arabic and the practice of the shared faith of Islam across much of North Africa and the Middle East. It is true that political boundaries often did not coincide with religious and linguistic boundaries in the AoE and more recent empires, but there were fewer barriers, as large states were common. In the late nineteenth century, nationalism, which affected populations' attitudes towards the Ottoman authorities, became prominent in the Middle East, fuelled in part by greater contact with European nationalism and the events that resulted in the emergence of some nation-states there (Anderson 2016: 262). Western expansion had also begun: the British Empire controlled Egypt and parts of the Persian Gulf region (Black 2015). At the fall of the Ottoman Empire, the creation of the modern Middle East and its boundaries began a period of internal strife in many Middle Eastern states. New forms of competition emerged, as new actors gained political power, after the demarcation of new borders by the victorious French and British after World War I. After states gained their full independence, political fragmentation became entrenched as monarchies, and later republics and dictatorships, competed for regional influence, and new power bases emerged in cities and capitals across the region (Lewis 1994).

The extension of an increasingly politically fractured Middle East that arose in the twentieth century has not only continued but also accelerated in the twenty-first. Localism within states is now the norm, particularly after the 'Arab Spring' of the early 2010s, this time manifested through ethnicity, religious sects and sub-state actors. The civil war in Syria has produced numerous political actors; analysts have identified five major groupings: the government, 'mainstream' rebels, extremist rebels, so-called Islamic State and Kurdish groups. One can divide such groups even further, particularly the rebel groups. Iraq has increasingly been seen as three separate regions composed of mainly Sunni, Shiite and Kurdish populations, and even within these regions significant political divisions are found. Even relatively small Lebanon has extremist groups, Hezbollah and the Lebanese government competing for influence. Instability, such as that seen in Egypt, and war within Yemen underscore the likelihood that power rivalries among larger states and militant actors within states will persist for some time and become widespread in the region. Within different political groups, radical and less radical elements compete, further clouding the picture. The presence of Islamic State (ISIS/ISIL/IS/

Da'esh) and other militant groups is perhaps an indication of how some modern states, such as Iraq, Syria, Lebanon and Yemen, have simply failed to control large parts of their countries politically. What is clear is that the Middle East today looks nothing like a politically integrated expanse, as ethnic groups and sub-state actors attempt to carve out their own regions of power, and political fragmentation has become more pronounced (Hazbun 2015). One may conclude that the region has become politically more similar to how it was in the pre-AoE period, when small, diverse states were found. Figure 12.2 compares a political map of Iraq, Syria and Lebanon in 2015 with one of the same area in the ninth century BCE, when political fragmentation was widespread, and shortly before the rise of universal states that persisted, arguably, into the Ottoman period. While the level of political fragmentation today is not as pronounced as in the ninth century BCE, we see that trends may be heading in that direction. If universalism characterized a period in the Near East that displayed more socially and politically integrated societies and cultural practices, are we seeing the reversal of this today?

The issue of 'reverse universalism' or the socio-political fragmentation of the modern Middle East is a large and complex topic. It should be discussed elsewhere in detail, but the long-term trend towards a modern Middle East broken into more independent and distinct entities is in stark contrast to the AoE periods. What are the processes that led to this trend and why are they happening? Clearly, foreign influences, whether the 2003 US-led invasion of Iraq, or the actions of the great powers after World War I in determining many of the modern state borders in the Middle East, have had some effect, which has been extensively covered elsewhere (see, e.g., Lowe and Dockrill 2002: 357). More time is probably needed to show whether long-term political fragmentation in the Middle East will transform social and cultural trends and institutions, which will, once again, alter the region's social and institutional fabric. The AoE stands as a drastically different case from the area as it is today. Perhaps, as a first step, monitoring how people move within the modern Middle East and how the region's urban centres and settlements are transformed, in size hierarchy and in social characteristics, could be key to understanding long-term, future socio-cultural change, as it was for the AoE.

### Note

<sup>1.</sup> For our purposes, the terms 'Near East' and 'Middle East' cover the same geographic area; however, we use 'Middle East' in relation to more recent periods, in particular those leading up to and after the fall of the Ottoman Empire.

**Figure 12.2** Map of Iraq, Syria and Lebanon in 2015 (top; after BlueHypercane761 2015) and the same region in 883 BCE with Assyria indicated (bottom; see Baudains *et al*. 2015: 6). The different colours indicate various competing political actors and entities in those periods

## Appendix

**Table A.1 Bootstrapping sampling method using well-fit least-squares Scenario 1 parameters for Southern Mesopotamia (SM) where the sampling probabilities are 0.05, 0.15, 0.25 and 0.5 for sites having a probability of not being simulated in a given run. Scenario 1's results (ratio is 0.0) are also included. Results reflect r2 averages for 500 runs. Cases include Early Dynastic (ED), Old Babylonian (OB), Kassite (KAS), Neo-Babylonian/ Achaemenid (NEO), Seleucid/Parthian (SEL) and Sasanian (SAS) settlements from Southern Mesopotamia**


**Table A.2 Bootstrapping sampling method using well-fit Spearman's rho (**ρ**) and least-squares Scenario 2 parameters for SM where the sampling ratios are 0.05, 0.15, 0.25 and 0.5. Scenario 2's results (ratio is 0.0) are also included. Results reflect averages for 500 runs for r2**


**Table A.3 Bootstrapping sampling method using well-fit least-squares Scenario 1 (Figure 4.11) parameters for the Khabur Triangle (KT) where the sampling probabilities are 0.05, 0.15, 0.25 and 0.5 for sites not being simulated in a given run. Scenario 1's results (ratio is 0.0) are also included. Results reflect averages for 500 runs. Cases include the Early Bronze Age (EBA), the Middle Bronze Age (MBA) and the Iron Age (IA)**


**Table A.4 Bootstrapping sampling method using well-fit Spearman's rho (**ρ**) and least-squares Scenario 2 (Figure 4.12) parameters for the KT where the sampling ratios are 0.05, 0.15, 0.25 and 0.5. Scenario 2's results (ratio is 0.0) are also included. Results reflect averages for 500 runs**


**Table A.5 Bootstrapping sampling results using well-fit least-squares Scenario 1 parameters for the Susiana Plain where the sampling probability is 0.05, 0.15, 0.25 and 0.5 for sites having a probability of not being simulated in a given run. Scenario 1's results (ratio is 0.0) are included. Results reflect averages for 500 runs. Cases include the Sukkalmah (SUK), Middle Elamite (MEL), Seleucid-Parthian (SEL) and Sasanian (SAS) periods**



**Table A.6 Bootstrapping sampling results using well-fit Spearman's rho (**ρ**) and least-squares Scenario 2 parameters for the Susiana Plain where the sampling ratio is 0.05, 0.15, 0.25 and 0.5. Scenario 2's results (ratio is** 

**Table A.7 Bootstrapping sampling results using well-fit least-squares Scenario 1 parameters for Central Anatolia (CA) where the sampling probability is 0.0, 0.05, 0.15, 0.25 and 0.5. Results reflect averages for 500 runs in the MBA, IA, Hellenistic/early Roman (HEL) and late Roman/Byzantine (BYZ) periods.**


**Table A.8 Bootstrapping results using well-fit Spearman's rho (**ρ**) and least-squares Scenario 2 parameters for CA using 0.0, 0.05, 0.15, 0.25 and 0.5 sampling ratios. Results reflect averages for 500 runs.**


## References


Bryce, Trevor. 2009. *The Routledge Handbook of the Peoples and Places of Ancient Western Asia: From the Early Bronze Age to the Fall of the Persian Empire*. London/New York: Routledge.


Bulling, A. Gutkind. 1972. China. *Expedition* 15(1): 23–39.


Caubet, Annie. 1998. The international style: A point of view from the Levant and Syria. In *The Aegean and the Orient in the Second Millennium: Proceedings of the 50th Anniversary symposium, Cincinnati, 18–20 April 1997*, edited by Eric Cline and Diane Harris-Cline, pp. 105–11. Liège/ Austin, TX: University of Liège.

Cavalli-Sforza, L. L. 2001. *Genes, Peoples, and Languages*. London: Penguin.


Fox, Edward W. 1971. *History in Geographic Perspective: The Other France*. New York: Norton.


Gibson, McGuire. 1972. *The City and Area of Kish*. Miami, FL: Field Research Projects.

Glatz, Claudia, Roger Matthews and Andreas Schachner. 2009. A landscape of conflict and control: Paphlagonia during the second millennium BC. In *At Empires' Edge: Project Paphlagonia*, edited by Roger Matthews and Claudia Glatz, pp. 107–48. London: British Institute at Ankara.


edited by St John Simpson. pp. 88–101. London: British Museum Press.


Gzella, Holger. 2015. *A Cultural History of Aramaic: From the Beginnings to the Advent of Islam*. Handbook of Oriental Studies, Section 1, The Near and Middle East, 111. Boston, MA: Brill.

Haas, Christopher. 1993. The Arians of Alexandria. *Vigiliae Christianae* 47(3): 234–45.


Invernizzi, Antonio. 1997a. Coppe e bracieri da incenso nell'Oriente classico. In *Profumi d'Arabia: Atti del Convegno*, edited by Alessandra Avanzini, pp. 121–46. Rome: L'Erma di Bretschneider. Invernizzi, Antonio. 1997b. Parthian Nisa: New lines of research. *Iran and the Caucasus* 1: 107–19. Invernizzi, Antonio. 2004. Thoughts on Parthian Nisa. *Parthica* 6: 133–43.


Koldewey*,* Robert*.* 1914. *The Excavations at Babylon*. Trans. Agnes S*.* Johns*.* London: Macmillan.


Kynard, Troy. 2015. *The Esoteric Codex: Mesopotamian Deities*. Raleigh, NC: Lulu.


Lamassu Design. 2009. Eanna of Uruk VI-IVa. https://commons.wikimedia.org/wiki/ File:Eanna4composite.svg. CC-BY-SA 3.0 license. Accessed 2 May 2017.

Lambert, W. G. 1983. The god Aššur. *Iraq* 45(1): 82–6.


Markoe, Glenn. 2000. *Phoenicians*. Berkeley, CA: University of California Press.


Miroschedji, Pierre de. 2014. The southern Levant (Cisjordan) during the Early Bronze Age. In *The Oxford Handbook of the Archaeology of the Levant, c. 8000–332 BCE*, edited by Margreet L. Steiner and Ann E. Killebrew, pp. 307–29. Oxford: Oxford University Press.


Negro Ponzi, Maria M. 2005. Al-Madā'in: Problemi di topografia. *Mesopotamia* 40: 145–69.

Nemet-Nejat, Karen R. 1998. *Daily Life in Ancient Mesopotamia.* Westport, CT: Greenwood Press.


Ostrogorsky, George. 1956. *History of the Byzantine State*. Trans. Joan Hussey. Oxford: Basil Blackwell*.*


byzantins, islamiques et slaves et leurs relations commerciales au Moyen Age, 6. Paris and the Hague: Mouton.


Renan, Ernst. 1864. *Mission de Phénicie*. Paris: Imprimerie Impériale.


Ritzer, George. 2010. *Globalization: A Basic Text*. 2nd edn. Malden, MA/Chichester: Wiley-Blackwell. Roaf, Michael. 1983. Sculptures and sculptors at Persepolis. *Iran* 21: 1–164.

Robbins, Manuel. 2001. *Collapse of the Bronze Age: The Story of Greece, Troy, Israel, Egypt, and the Peoples of the Sea*. San Jose, CA: Authors Choice Press.

Robins, Gay. 1993. *Women in Ancient Egypt*. Cambridge, MA: Harvard University Press.


Rubinson, Karen S. 1990. The textiles from Pazyryk. *Expedition* 32(1): 49–61.


Saggs, H. W. F. 2000. *Babylonians*. Peoples of the Past. Berkeley, CA: University of California Press.


by A. M. Davies and W. Meid, pp. 373–90. Innsbruck Beiträge zur Sprachwissenschaft, 16. Innsbruck:. Inst. für Sprachwissenschaft der Universität Innsbruck.


Seymour, Michael. 2014. *Babylon: Legend, History and the Ancient City*. London: I.B. Tauris.

Shahbazi, Shapur A. 1990. Byzantine-Iranian relations. *Encyclopaedia Iranica*, vol. IV/6: 588–99.


Sicker, Martin. 2000. *The Pre-Islamic Middle East*. Westport, CT: Praeger.

Sidebotham, Steven E. 2011. *Berenike and the Ancient Maritime Spice Route*. California World History, 18. Berkeley, CA: California University Press.

Siegel, Jeff. 1985. Koines and koineization. *Language in Society* 14(3): 357–78.


Squitieri, Andrea. 2017. *Stone Vessels in the Near East during the Iron Age and the Persian Period (c. 1200–330 BCE)*. Archaeopress Ancient Near Eastern Archaeology, 2. Oxford: Archaeopress.


Süel, A. 1990. 1988 yılı Çorum ili yüzey araştırmaları. *Araştırma Sonuçları Toplantısı* 7: 341–59.

Sugimura, Toh. 2008. Japan xi. Collections of Persian art in Japan. *Encyclopedia Iranica* 14(6): 571–74.

Sumner, William. 1976. Excavations at Tall-i Malyan (Anshan) 1974. *Iran* 14: 103–15.

Swain, Simon. 2003. *Hellenism and Empire: Language, Classicism, and Power in the Greek World, AD 50–250*. Oxford: Clarendon Press.


Thierry, François. 1993. Sur les monnaies sassanides trouvées en Chine. *Res Orientales* 5: 89–139.


Tomber, Roberta. 2008. *Indo-Roman Trade: From Pots to Pepper*. London: Bristol Classical Press.

Török, László. 1995. *Hellenistic and Roman Terracottas from Egypt*. Rome: L'Erma di Bretschneider.

Tosi, Maurizio. 1974. The lapis lazuli trade across the Iranian plateau in the 3rd millennium BC. In *Gururājamañjarikā*: *Studi in Onore di Giuseppe Tucci*, edited by R. Rubinacci, vol. 1, pp. 3–22. Naples: Istituto Universitario Orientale.

Tosi, Maurizio (ed.). 1983. *Prehistoric Sīstān*. Reports and memoirs, 19(1). Rome: IsMEO.


Tripolitis, Antonia. 2002. *Religions of the Hellenistic-Roman Age*. Grand Rapids, MI: W.B. Eerdmans. Tsetskhladze, Gocha. 1999. Introduction. In *Ancient Greeks West and East*, edited by Gocha

Tsestkhladze, pp. vii–xiv. Mnemosyne, Bibliotheca Classica Batava, 196. Leiden: Brill.


Wengrow, David. 2006. *The Archaeology of Early Egypt: Social Transformations in North-East Africa, 10,000 to 2650 BC*. Cambridge: Cambridge University Press.


Wilson, Alan G. 1970. *Entropy in Urban and Regional Modelling*. London: Pion.


Zawadzki, Stefan. 1988. *The Fall of Assyria and Median-Babylonian Relations in Light of the Nabopolassar Chronicle*. Seria Historia, 149. Poznań: Adam Mickiewicz University Press/ Delft: Eburon.

Zolfaghary, Amir Hussain. 2010. Naghshe Rostam in Persia. https://commons.wikimedia.org/ wiki/File:Naghshe\_Rostam\_ZPan.jpg. CC-BY 3.0 Unported. Accessed 2 May 2017.


## Index

Abbasid, 264, 269–70 Achaemenid Empire, viii, 35–8; *see also* government (AoE) Age of Empires (AoE), viii, 1–5 Alexander the Great, 38–9 Alexandria, 43, 149–51 Amarna (Akhetaten), 136–7 Anatolia, 2–3 Central, 20, 103–10 Western, 33–4 Antioch, 117, 147–9 Armenia, 39, 45, 47 Assurnasirpal, 26, 38, 55 Attractiveness, 60–2; *see also* spatial interaction entropy maximisation Ayyubid, 270 Babylon, 77, 140–2 Baghdad, 264 Britain (Trade with), 168, 172, 176, 226 Bronze Age, viii, 2, 15 Early, viii, 15–18 Late, 22–5 Middle, 18–21 Byzantine Empire, viii, 51–2; *see also* government (AoE) Central Asia, 1, 2, 13 (Trade with) 20–1, 160; *see also* trade China (Trade with), 45, 47, 50, 167; *see also* trade Coinage, 33, 165, 171–2 Complexity Theory, 61, 268 Complex Adaptive Systems, 266 Corporations, 141, 172–3 banking, 141, 172, 178, 214 Crusade, 270 Ctesiphon, 47–8, 75, 145–7 Cyrus, 35, 210 Darius I, 35–6, 210, 213–14 Dur-Sharrukin, 30, 138–40 Dur-Untash, 132–3 Dura Europos, 41, 45, 151–2 Ebla, 129 Egypt, 2, Pharaonic, 18, 21–2, 30–1, 33, 35–6; *see also* Ptolemaic Roman, and Byzantine Empire (Definition of), 4 Fatimid, 270–71 Frankincense, 68–9, 166–9

Gandhara, 192, 195, 258 Gini Coefficient, 58 Globalization (in relation to universalism), 9–11, 261, 263 Government, 199 AoE governing, 203–230 pre-AoE governing, 199–203 Hatra, 45, 48–9, 16, 175, 208 Hattusha, 134–5 Hazor, 135–6 Hellenistic Period, viii, 38–43, 68, 165, 174, 180, 190, 192, 198, 215–9, 244 Style, 142, 147, 151, 191–2, 194, 196, 197 Hollow ways, 95, 208 Hybridization, 12, 39, 43, 179–181 AoE, 186–198 Pre-AoE, 181–6 *see also* Hellenistic (Style) Ilkhanate, 270–1 India, 38–9, 44, 47; *see also* Trade Hybridization Iran, 20, 26, 30–1, 33, 35, 39, 41–2, 44–5, 47–51; *see also* Iranian Dynasties Iranian Dynasties, 270–1 (Afsharid, Qajar, Safavid, Zand) Iron Age, vii–viii, 11–13 Early, 11, 22–9 Late, 12, 29–34, 123, 166, 253 Islamic State, 14, 275–6 Kalhu, 31, 138–40 Kushan, viii, 44, 47, 50, 168 Language Akkadian, 13, 20, 156, 201, 209, 231–2 Aramaic, 168, 188, 206, 209, 213, 215–16, 221, 225, 232–5 common language, 2, 6, 206, 209, 216, 221, 228, 231, 262–3, 266–7 Coptic, 236, 260 Greek, 235–9 Levant, 20–2, 24–5, 33, 42, 47–8, 51, 113, 117, 122, 156 Lydia, viii, 28, 33, 35 Northern, 110–18, 129, 147 Southern, 118–20, 135, 154 Syriac, 151, 172, 236, 260 *see also* Trade

Mamluk, 270–1 Mari, 15, 18, 20, 129–32 Markov Clustering, 66 Medes, 26, 31, 33–4, 149 Mesopotamia, vi–vii, 15, 18, 22, 31, 39, 41–2, 45, 48–50, 122, 124, 125 Khabur Triangle, 87–97 Northern, 17–18, 20, 22, 26, 36, 38, 45, 51 Southern, 15, 17–18, 20–2, 30–1, 48, 72–87, 121, 123 *see also* Trade Middle East, 11, 14, 269, 274–6 Movement, v, vii, 4–7 constrained movement, 61, 65, 95 easy movement, 61, 65, 87, 89, 92, 95, 97, 101, 113, 117–8, 254, 258, 271 forced (or deportation), 4, 10, 24, 30, 44, 210, 217, 221, 230, 258–9, 263 travel speed, 68, 174–6 *see also* spatial interaction entropy maximisation Myrrh, 165–6, 168–9 Neo-Assyrian, history, 29–33, 54 cities, 138–40, 155–7, 159 government, 203–10, 253–4 long-distance trade, 174–5 material culture, 180, 186–7 settlement pattern, 90, 92, 121, 123 Neo-Babylonian, history, 33, 72 cities 140, 158 economy, 172 government, 228, 249, 257–8 language 210 settlement pattern, 74–86, 95, 117, 120–1 Nineveh, 30–1, 123 government, 206–8, 258 layout, 138–9, 151, 159, 180, 187 Nystuen and Dacey Graph, 66 Ottoman, 51, 270–2 post-Ottoman, 274 Palmyra, 50, 167, 192–3, 222–3, 225, 234 Parthian, history, 43–8 art, 190–4 cities, 145, 151 language, 219–20, 225 settlement pattern, 72, 74–5, 77, 79–82, 86, 96–102, 121 trade, 167–70, 172 Pepper, 68–9 trade of, 170 Persepolis, 142–5, 159, 180, 189, 211, 257 Persian Gulf, 22, 24–5, 45, 47–8, 52, 162 trade, 170–1, 220, 222, 232, 275 Pre-AoE (history), viii, 15–29 Ptolemaic, 43, 149, 191, 217–8, 247, 259 Religion, 240–52 AoE religion, 242–52 Apollo, 244

Christian schism, 227 Isis, 244–8, 260 Islam, 240, 251, 264, 269, 275 Judaism, 147, 149, 222, 226, 240, 247, 249–50 Manichaeism, 50, 222, 245, 247 Mithra, 51, 244–8 mystery cult, 245–58 pre-AoE religion, 240–2 syncretic religion, 242–7 universal, 247–51 Zeus, 244, 252 Zoroastrianism, 49–50, 147, 220, 226, 245, 247, 250–1, 260 Roads, 207–8 long-distance, 160–74 royal, 175, 207, 214, 228 short-distance, 65, 95 *see also* hollow ways Roman, 42–51, 98, 105–23 Sasanian, 48–52, 54, 72, 74–82, 86, 95–101, 121 Scaling, 64, 264 Seleucid, history, 38–43 cities, 141, 147–9 settlement pattern, 72–82, 86, 99–102, 117 trade, 171–4, 215–18 Seljuk, 270–2 Settlements, 71–123 AoE urbanism, 124–59 city, 124–59 hierarchy, 71, 78, 85, 103, 105, 111, 113, 118 interaction, 59, 107 large, 61, 67, 86, 119, 125, 128 pre-AoE, 124–37 small, 152–7 structure, 57–67, 71–123 Shapur, 49–51 Silk Road, 45, 50, 147, 167, 176 Spatial interaction entropy maximisation (SIEM), 59–60, 67, 71, 76, 82, 71–123 Syncretism, 8–9, 39, 125, 127, 145, 149, 159 AoE syncretism, 186–197 hybrid, 179–98 language, 236; *see also* Language material culture 179–198 pre-AoE syncretism 181–6 *see also* Religion Trade, 160–78 AoE trade, 165–76 merchant colonies, 173–4 pre-AoE trade, 160–5 Trajan, 47–8 Umayyad, 269–72 Universalism, v, 2, 7–14, 56–7, 125, 142, 149, 249, 253, 257, 261–76 reversal (fragmentation), 269, 274–7 Ur, 18, 127–8, 132, 162 Uruk, 18, 45, 72, 75, 82, 125–7, 129, 134, 157

Christianity, 247–52

This book investigates the long-term continuity of large-scale states and empires, and its effect on the Near East's social fabric, including the fundamental changes that occurred to major social institutions. Its geographical coverage spans, from east to west, modernday Libya and Egypt to Central Asia, and from north to south, Anatolia to southern Arabia, incorporating modern-day Oman and Yemen. Its temporal coverage spans from the late eighth century BCE to the seventh century CE during the rise of Islam and collapse of the Sasanian Empire.

The authors argue that the persistence of large states and empires starting in the eighth/ seventh centuries BCE, which continued for many centuries, led to new socio-political structures and institutions emerging in the Near East. The primary processes that enabled this emergence were large-scale and long-distance movements, or population migrations. These patterns of social developments are analysed under different aspects: settlement patterns, urban structure, material culture, trade, governance, language spread and religion, all pointing at population movement as the main catalyst for social change. This book's argument is framed within a larger theoretical framework termed as 'universalism', a theory that explains many of the social transformations that happened to societies in the Near East, starting from the Neo-Assyrian period and continuing for centuries. Among other infl uences, the effects of these transformations are today manifested in modern languages, concepts of government, universal religions and monetized and globalized economies.

Mark Altaweel is a Reader in Near Eastern Archaeology at the UCL Institute of Archaeology having held previous appointments and joint appointments at the University of Chicago, University of Alaska, and Argonne National Laboratory.

Andrea Squitieri studied Archaeology of the Ancient Near East at the University of Turin and at UCL (PhD 2015). He has been a post-doc researcher since 2015 working in the Peshdar Plain Project, based at the Ludwig-Maximilian University of Munich, focusing on the study of the eastern border of the Neo-Assyrian Empire.

**Front cover image:**  Photograph by Mike Peel, CC-BY-SA-4.0  **Back cover image:**  Photograph by A.Davey, CC BY 2.0

> **Cover design:** www.ironicitalics.com

**REVOLUTIONIZING** 

**REVOLUTIONIZING A WORLD**

Mark Altaweel and Andrea Squitieri

**A WORLD**

From Small States to

Pre-Islamic Near East

Mark Altaweel

Andrea Squitieri

Universalism in the