Scope and Themes

South Asia, also known as the Indian Subcontinent, covers 4.5 million square kilometres and contains 109 of the world’s mountains that rise over 7,000 metres (Figure 1.1). This region is home to one-third of the world’s population and encompasses several hundred local languages and dialects and is the site of the emergence of four major world religions and one of the four Old World Civilisations. It now accounts for a massive US$ 1.854 trillion of the world’s gross domestic product and is the source of a diaspora of some 30 million people. Given the economic and political significance of contemporary South Asia, it is no surprise that this vast geographical region has a matching richness within its archaeological and historical record. It is so vast and rich that it is correct to question whether it is even possible to present a volume which draws together such disparate topics as hunter-gathers from western India, the major urban forms of the Indus Civilisation, the Iron Age megaliths of Peninsular India and the imperial ideology of the Mauryans. We believe that this is possible but also believe that in order to do so, it is important to present this information through the medium of a major narrative theme in order to structure our material. Rather than just pursuing a route of describing site sequences and moving from one chronological building block to another, encyclopaedically detailing all the different cultures that have been identified across the region or focusing on technical descriptions of pottery or stone tools in an attempt to define archaeological cultures, we have chosen to take a site and regional-based themed approach structured within a distinct developmental framework.

Figure 1.1. Map of South Asia showing modern nation states.

Whilst fully conscious of the multiplicity of narratives, identities, approaches and paradigms present within contemporary South Asian archaeology, or rather archaeologies, our selected theme involves the direct comparison of South Asia’s two largely urban-focused developments, generally termed the Indus or Harappan civilisation and the Early Historic or Indo-Gangetic civilisation. We will also undertake a detailed consideration of the people and settlements belonging to the period between the two, which has frequently been presented and interpreted as a distinct cultural, political and social transformation. We have chosen to do this for two main reasons. The first is that there were many similarities in the internal sequences and cycles of both these developments and the time lapse between them has now been reduced to a matter of centuries. The second reason is that research now establishes clearly that the origins of both the Indus and the Early Historic urban-focused developments were much older and that both developed far more slowly than has often been presented in the past and, as such, have formed distinct traditions. Within this volume, we will also explore a range of different theories about state formation and social organisation in relation to South Asia, and then test them against a range of archaeological and, where appropriate, historical evidence. This process will serve to demonstrate how much our understanding and perspectives have changed archaeological theory and fieldwork in South Asia since Cambridge University Press’s foundation publication of Raymond and Bridget Allchin’s The Rise of Civilization in India and Pakistan in 1982 in the Cambridge World Archaeology series.

Whilst we will closely examine the dynamics of both of these urban-focused populations in turn and consider issues such as continuity and transformation, similarity and difference, it is also important to remember that few regions have ever existed in a vacuum. South Asia has always influenced and been influenced by its near neighbours and more distant trading partners. Recognising this perspective is critical for understanding questions of diffusion and indigenous development as these two fundamental issues of continuity and transformation dominate discussions of archaeological explanation in South Asian archaeology. By exploring the development, character and ultimate transformation of each of the two main urban-focused sequences in depth, we will present a range of past and current theoretical explanations. We will also demonstrate how these have influenced the development of past and contemporary archaeological and historical interpretations, which in turn have resulted in a number of enduring social and political narratives. We would also stress that this volume is not focused solely on urban forms or urbanism but on settlements and communities more broadly and their networks and connections. Although, of course, chapters and debates on the urban-focused development of the Indus and Early Historic societies receive considerable coverage. As such, we believe that the title of the volume reflects its contents, which consider the archaeologies of urban development and their spheres of influence as well as non-urban communities and non-urbanised regions and their populations between the Indus and Asoka.

Traditional synthetic archaeological studies of South Asia have tended to either follow a chronological narrative introducing the main events and developments across the whole region, or present the developmental sequence of either the Indus or the Early Historic civilisations. Whilst some of these general chronological or synthetic narratives provide invaluable sources of evidence, such as Settar and Korrisettar (Reference Settar and Korisettar2002) and Singh (Reference Singh2008), they remain largely separate from theoretical concerns or explanations of change. Eltsov has recently contributed to this cohort of scholarship with a volume exploring concepts of the ancient South Asian city as gleaned from heavily edited textual sources but remains largely urban-focused and controversial in his application of later texts to the third millennium BCE (2008). Some of the works that have explored either the Indus or the Early Historic urban and rural sequences have provided innovative approaches for the analysis of those complex societies, for example Shaffer’s (Reference Shaffer and Ehrich1992) concept of an ‘Indus Valley Tradition’ to which we return later. However, most have focused on either one tradition or the other, thus continuing the long-standing division between the Indus and Early Historic, for example Wright (Reference Wright2010), Sengupta and Chakraborty(Reference Sengupta and Chakraborty2008), McIntosh (Reference McIntosh2002) and Kenoyer (Reference Kenoyer1998). This division can be broadly traced back to the later years of European colonial influence in South Asia and the impact of individuals such as Mortimer Wheeler (Reference Wheeler1950), Gordon Childe (Reference Childe1934) and Stuart Piggott (Reference Piggott1950) with their claims that a distinct cultural, linguistic and social transformation lay between the Indus Civilisation and the Early Historic. This is not to suggest that this was purely a colonial concept as a number of post-Independence South Asian scholars also adopted and adapted it, including Dani (Reference Dani1967), Banerjee (Reference Bandaranayake, Mogren and Epitawatte1965) and Lal (Reference Lal1955). Furthermore, some scholars have viewed the Indus through a prism influenced by the archaeology of Mesopotamia, such as Wright Reference Wright2010. As this volume considers merchant populations within the Arabian Sea and Indo-Iranian Plateau, we also feel justified in citing relevant comparative models and concepts associated with those regions and beyond if they help us advance our understanding of the sequences and processes under discussion (Trigger Reference Trigger2003).

Archaeological research in South Asia has of course moved far beyond these simplistic models, but the influence that such early interpretations of key sites and materials had on the development of archaeological explanation has been immense, and one which we will explore, along with other archaeological discussions and theories throughout this volume. Although elements of continuity between the two periods have been recognised by an increasing number of scholars (e.g. Agrawal Reference Agrawal2007; Upadhyay Reference Upadhyay, Sengupta and Chakraborty2008; Eltsov Reference Eltsov2008; Coningham Reference Coningham and Allchin1995a; Shaffer Reference Shaffer, Spodek and Srinivasan1993; Kenoyer Reference Kenoyer1991b; Chakrabarti Reference Chakrabarti1999), the techniques, theories and methodologies for studying these two urban-focused developments have tended to remain separate – as indeed do most of the archaeologists working on them. Indeed, one recent comparative study of South Asian cities from 2500 BCE until after the ninth century CE has even stated that their configurations appear to have been quite separate: “The Indus, Early Historic and Medieval urban phases were independent developments” (Smith Reference Smith2006a: 130). It is not the intention of this volume to lionise the contributions of colonial scholars but to join other scholars in acknowledging that their theoretical and methodological influences are still distinctly traceable (Basant Reference Basant, Sengupta and Chakraborty2008: 191); therefore addressing this artificial divide is one of the cornerstones of the present volume.

Chronological and Geographical Span

Bridget and Raymond Allchin presented three major synthetic texts to South Asian archaeologists; The Birth of Indian Civilisation (1968), which began with the Early Stone Age, continued through the Indus Civilisation and terminated with the Iron Age and what the Allchins called the beginnings of history. In parallel, The Rise of Civilisation in India and Pakistan (Reference Allchin and Allchin1982) began with a discussion of hunter-gathers and nomadic pastoralists, moved through early sedentary, agricultural populations to the main focus of the book, the Indus Civilisation. Finally, Raymond Allchin’s edited The Archaeology of Early Historic South Asia (Reference Allchin1995a) revisited the transitional end of the Indus Civilisation, and then concentrated on the emergence and regional development of the second urban period, concluding with the Mauryan Empire. Sharing a similar title with Allchin’s edited volume, Gautam Sengupta and Sharmi Chakraborty’s book contains a number of contributors who question the usage and very definition of the term ‘Early Historic’ (Reference Basant, Sengupta and Chakraborty2008). Dilip Chakrabarti’s text India: An Archaeological History: Palaeolithic Beginnings to Early Historic Foundations (Reference Chakrabarti1999) primarily covered the archaeology of the modern state of India from the Palaeolithic to AD 300, and Upinder Singh’s A History of Ancient and Early Medieval India, up to the twelfth century AD (Reference Singh2008). Our own text falls between these approaches; we aim to be less wide ranging chronologically than Chakrabarti and Singh’s volumes, which allows us to look in greater detail at sites and issues, and we draw together the two main urban-focused South Asian developments which formed the subject of separate Allchin volumes (Timeline 1.1).

Timeline 1.1. General timeline for the Indus Valley and Early Historic Traditions.

The very term ‘South Asian’ as a description of people from the geographical region of South Asia is contested by some, and there is current debate surrounding the suitability of this term to describe people or groups of people who have originated from the countries of Bangladesh, India, Nepal, Pakistan, Sri Lanka and associated states, or are descended from citizens of these places. To many, ‘South Asia’ is considered a colonial construct, a blanket term that oversimplifies the geographical and cultural complexity of the region, and thus reduces the people so described to a uniform ethnicity. In place of ‘South Asian’, it has sometimes been proposed that people and groups of people are better referred to by their religion, such as Sikh, Muslim, Hindu or Buddhist. While there are clearly many issues with this (and other) suggested classificatory and descriptive system, the main point here is that many of the archaeological and cultural terms that we use within South Asia have been developed externally and may not always be appropriate. In many cases, it is important to realise that forcing the fit of such terms and names is not only inappropriate but may also have been a means of masking internal or indigenous activity. There are also a number of terms and related issues that are used commonly in South Asian archaeology, about which we need to make our own position and understanding clear. Notwithstanding these points, we will continue to use ‘South Asia/n’ as a geographically descriptive term, a form of shorthand, for the nation states outlined in Chapter 2. However, as we make clear in Chapter 2, this is not intended to mask differences, whether physical or cultural, as these differences are integral to our understanding of the prehistory and early history in this region. Rather, it is intended as an overview term, which we feel is relatively free from ethnic, religious or other content whilst reflecting the strong cultural and historical connections of this region and distinguishing it from West and South East Asia.

South Asia today is a highly complex region with multiple religions, ideologies and belief systems, languages, ethnic groups and social identities, and this was also true in the past. As a result, we cannot offer a ‘one size fits all’ approach to understanding the past here because very different processes were running at the same time in different parts of the region. For example, when the first iron-using farmers sailed from Peninsula India to Sri Lanka they appear to have coexisted for a while alongside lithic tool-using hunter-gatherering populations, apparently bypassing Neolithic and copper artefact-using phases. Rather than starting this volume by presenting the earliest evidence for human activity within South Asia, and moving chronologically through each region, we will draw out core themes and processes and follow a comparative approach. For this reason, rather than beginning with early communities of hunter-gatherer-foragers, we will begin by considering South Asia’s first food producers and analysing their material culture, in order to both understand change and organisation within these populations, and to present them as the roots of increasing complexity and incipient urbanism. We are also aware of the great contrasts between available data sets, primarily chronometric dates ranges and published sites, across South Asia. While it is clear that there are foci of excellence, such as the pioneering work of Siran Deraniyagala (Reference Deraniyagala1992) exploring microlithic tool-using populations within the tropical rainforests of central Sri Lanka, comparative perspectives from elsewhere are not yet available. Until such data are more systematically investigated and approached across South Asia as a whole, it is unlikely to be systematically synthesised and presented. This situation is changing, as seen in publications such as those of Robin Dennell (Reference Dennell2009), Sheila Mishra (Reference Mishra1995; Mishra et al. Reference Altekar and Mishra2013) and Ravi Korisettar and Mike Petraglia’s teams in the Deccan (Reference Korisettar and Petraglia1999), so we may anticipate a greater degree of knowledge and information in the near future.

Similarly, the decision about where to finish the narratives within this volume was as difficult as determining the starting point, and we debated whether we should end with the opening of the Gupta ‘Golden Age’ or the era of Kanishka or possibly even as early as the movement of the Macedonians into the far west of the South Asian region. However, we have chosen to end it with the reign of Asoka (r. 269–232 BCE), the great Mauryan Emperor who had details of his life and ideology recorded in a variety of sources, including primary historical texts, and inscribed stone pillars and boulders. We have chosen this point to finish as we suggest that the Mauryan Empire brought together for the first time much of South Asia under a single hegemony, one which formed the basis of the state traditions which held sway for the next two millennia. Modern South Asia draws heavily on the time period covered in this book for many of its economic, social and cultural narratives, and these issues of identity and recognition will be discussed in our next chapters, where we consider the role of archaeology, identity and nationalism within the modern nation states of South Asia.

Given the great range of people and cultural markers within a single country such as India, Nepal or Sri Lanka, it is reasonable to ask how we consider it possible to explore the prehistory and early history of a number of countries over a six thousand year timescale. We argue that it is precisely because of this time depth that we can consider the countries of South Asia as a larger entity existing beyond modern geopolitical boundaries. By exploring the development of the two major urban-focused traditions in this region, we are able to examine both similarities and differences across a wide range of environmental, social, ideological and cultural groupings. In Chapter 2, we will discuss the geographic boundaries which both unite and define the modern states of South Asia, and we will also situate the study region within its wider setting of the Indian Ocean and the Himalayan and Hindu Kush mountain barriers. In so doing, we will ensure that modern geo-political boundaries do not artificially constrain our discussion.

Whilst there are a number of convincing geographical and cultural elements that make this region a coherent whole for the purposes of long-term study, there are of course many links with regions outside the immediate boundaries of study that can be elucidated through archaeological analysis. Historical and art historical sources inform us about contact with the Classical Mediterranean, the Red Sea and the Near East, Eurasia and, of course, Achaemenid Persia to the west. Indeed, we have accounts of Megasthenes, the Seleucid Ambassador to the Mauryan court, and the later records of the Chinese Buddhist pilgrims with records of contact with China and Central Asia, as well as South East Asia. However, in order to understand the nature and dynamics of such contact in earlier periods, we need to turn to archaeology, and we will explore these contacts in greater depth in relevant sections. For example, we will examine the evidence for the reported presence of Indus merchants and entrepots in Mesopotamia, northern Afghanistan and along the Persian Gulf, along with evidence from Indus sites in South Asia indicating external contact in Chapter 7. The impact of South Asia itself on surrounding regions is also important, not only with respect to trade and exchange, but also in the spread of ideologies such as Hinduism and Buddhism to various parts of South-east and Central Asia. In turn, pilgrims from these areas to South Asia have also had impact on developments in the region (e.g. Bellina and Glover Reference Bellina, Glover, Glover and Bellwood2004; Indrawooth Reference Indrawooth, Glover and Bellwood2004).

Key Concepts and Their Archaeological Indicators

There are many concepts in modern archaeology which are frequently used, although different scholars may attach somewhat different understandings as to their exact meanings and applications. For example, differing definitions of urbanism in South Asia have hindered comparisons between the Indus and Early Historic Traditions and, as a result, we believe that it is important to provide definitive explanations of potentially controversial terms and concepts from the outset. This section will therefore present and consider a wide range of different archaeological concepts relevant to our broader discussions, and offer definitions or outlines which will be of value to readers as well as helping to ensure that misunderstandings and misinterpretations do not arise. We will also draw on the definitions presented here to underpin current understanding (and misunderstandings) of the main chronological and cultural events within the region.

Many of the following terms and concepts are closely linked, and there is often a degree of overlap between definitions, but they are all part of our search for greater understanding of the origins of urban-focused communities and their populations. We have also explicitly engaged with the concept and definition of ‘civilisation’ as it is frequently identified as one of the fundamental questions to be addressed by archaeology (Gamble Reference Gamble2001: 157), albeit one of the most debated and contested. Just as important as understanding the origins of any archaeological phenomenon, is an understanding of the dynamics and processes which carried that phenomenon forward giving rise to tangible evidence which we as archaeologists have recovered in the present. Indeed, understanding and explaining issues of cultural resilience and stability may also shed light on differing trajectories of adaptation.

Nomenclature and Terminologies

The purpose of this section is to establish a number of key definitions for the terminologies and chronologies used in this book, so that readers are aware of how and why we use particular terms and phrases, and the basis for our dating of sites and events across South Asia. Although it might seem self-evident, we believe that it is worth stating that language is not only very powerful, but is also evolving and changing. Terms become fashionable or unfashionable and they can take on positive or negative meanings of their own, often largely dependent on context. A good example of this can be found in discussions of neo-evolutionary language associated with the development of complex or state-level societies (see e.g. Chapman Reference Chapman2003; Yoffee Reference Yoffee2005). As we shall see in Chapter 3, the British antiquarian movement and early British archaeological activity had a great influence on the practice of archaeology in South Asia, therefore it is not really surprising that British and European terms and concepts were imported to describe and explain South Asian material culture and cultural developments, although many of these early models are no longer credible. An example of this is the ways in which chronologies and time in South Asia have been dealt with by colonial archaeologists, and then later challenged by South Asian archaeologists after Independence. The standard European ‘Three Age’ system of dividing prehistory into progressive chronological sequences based on materials (i.e. stone, bronze, iron) was enthusiastically adopted by early archaeologists such as Robert Bruce Foote (Pappu Reference Pappu2008), Wheeler (Reference Wheeler1948), Piggott (Reference Piggott1950), and Gordon (Reference Gordon1960) for application across South Asia. Of course, the Three Age system has since been challenged across the globe and South Asia is no exception – the presence of microlithic stone tools alongside evidence for settlements, domestication, and other craft working at numerous sites shows how difficult it is to categorise past human activity according to a very narrow material culture definition (Box 1.2). Challenges to the European understanding of time have come from a number of scholars, with analyses of the ways in which precolonial Indian time has been understood as cyclic, unchanging and ahistoric and European time has been understood as linear and progressive (Sen Reference Sen2002: 349; Thapar Reference Thapar and Ridderbos2002: 27–28). New analyses of Early Historic texts has allowed the argument that ‘traditional’ time was (and is) both cyclic and linear. These types of challenges to pervasive Colonial scholarship and interpretation are very important for highlighting the layers and nuances of South Asian prehistory and early history.

Another example of the way in which traditional, European archaeological classification, description, analysis and ultimately interpretation are neither useful nor appropriate within South Asia is centred on the lifeways of hunter-gatherer-foragers. There was demonstrably a great diversity of population and communities present within the Indus or Harappan civilisation, traditionally characterised as urban-focused and literate with a uniform material culture spread over an area of almost 3,133,886 square kilometres (Wright Reference Wright2010). The nature of many early finds and their comparison with Mesopotamian material culture placed the whole discovery firmly within the category of Old World Civilisation (Marshall Reference Marshall1931a: Wheeler Reference Wheeler1953), and that is how the civilisation is still presented in much of popular culture and some academic writing today (Kenoyer Reference Kenoyer and Kenoyer1994). However, it is becoming increasingly clear that the Indus cities closely interacted with sizable populations reliant on technology that would traditionally be associated with ‘Neolithic’ or even ‘Mesolithic’ lifeways (Mughal Reference Mughal and Allchin1997; Possehl Reference Possehl and Taddei1979). At the site of Bagor in Rajasthan, for example, we can see this duality in the archaeological remains, where characteristically Indus artefacts, such as bronze fishing hooks and drilled carnelian beads, have been recovered alongside microlithic tool assemblages. Langhnaj appears to mark a similar site as it yielded burials, a copper knife and microlithic tools (Wright Reference Wright2010: 175). In order to articulate these populations with those of the cities, M. K. Dhavalikar (Reference Dhavalikar1995) has suggested that the inhabitants of such sites may have co-ordinated the collection of nodules of semi-precious stones and exchanged these at sophisticated processing centres, such as Kuntasi or Lothal, and Possehl (Reference Possehl2002a, Reference Possehl, Morrison and Junker2002b) and Morrison (Reference Morrison and Stark2006) have both stressed the potential of such symbiotic exchanges.

A further example is the impact of the appearance of iron within the sequences of sites across South Asia. Across Europe, early archaeologists assumed a major social and economic transformation to have accompanied this innovation. However, in South Asia, iron working appeared within the sequence at the site of Pirak in Baluchistan where it was not accompanied by other fundamental changes in material culture. Indeed, this new metal technology was practiced immediately adjacent to the already established copper and bone-working areas (Jarrige and Santoni Reference Jarrige, Quivron and Raven1979). A further example is provided by the transition from the production and use of only stone tools (‘Mesolithic’) to the production and use of iron (‘Iron Age’) in Sri Lanka. Unlike the sequences across the Palk Straits in South India, where it is possible to trace a series of cultural transformations from a ‘Neolithic’ to a phase of using copper tools and then to an ‘Iron Age’, Sri Lanka’s sequence appears to have shifted from an established ‘Mesolithic’ to the abrupt appearance of iron tools and associated ceramic types at the beginning of the first millennium BC (Deraniyagala Reference Deraniyagala1992, Reference Deraniyagala1990). Finally, it is worth noting the difficulty of using the term ‘Bronze Age’ to refer to the Indus Civilisation and ‘Chalcolithic’ to discuss some of the contemporary and later farming communities in the Deccan and Peninsular India. This is because although copper and bronze objects were utilised in both regions and during both phases, stone tools were also utilised and appear to have retained an important position. For this reason, we shall refer to both the Indus Civilisation and the later farming communities of the Deccan and Peninsular India as ‘Chalcolithic’. This list is by no means exhaustive, and there are many other examples which demonstrate that South Asian cultures and people did not always adopt or select linear progressions in technological and social change.

This complex picture suggests that models derived from Europe may not always be applicable to the diversity of South Asia, parts of which did not become ‘historic’ until recent times. This is not to advocate the presence of residual Pleistocene populations as some scholars have attempted to do based on comparisons of modern distributions of tribal languages with scatters of microlithic tools (Parpola Reference Parpola1994). Nor should it be suggested that such communities may be viewed as conservative or unchanging; rather that some groups of people have made choices to exploit the lucrative resources of forest and jungle and, like the Veddas of Sri Lanka, such groups have often been in contact with literate, state-level populations for centuries (Fox Reference Fox1969). This inability to categorise archaeological sequences on the basis of monothetic classificatory schemes is a theme to which we will return.

Chronologies in South Asia

Chronologies and dating, two of the pivotal axes of archaeological analysis and interpretation, are also far from straightforward in South Asia, and we will examine the impact of early historical geography on the development of relative chronologies and the way they became entrenched in archaeological practice. Clearly this position owes much to the personalities and interests of the influential early British antiquarians such as Cunningham and Prinsep, as discussed in more detail in the following chapter. Radiocarbon date estimates have been exploited by archaeologists with increasing frequency in many parts of the world throughout the second half of the twentieth century, and samples suitable for processing are now routinely collected. We will look at scientific dating across the different regions of South Asia and discuss the dearth of radiocarbon dates from archaeologically crucial periods, such as that between the two major urban-focused developments. As discussed earlier, this period has often been designated a transformation, and we suggest that this commonly accepted concept (and others, such as diffusion) is in many ways the product of lack of research, archaeological visibility and poor relative and scientific dating. This problem is illustrated very clearly when we examine the Indus cities, where despite nearly 150 years of archaeological exploration and research on various sites by numerous archaeological teams, there is still no absolute agreement about dates, and even within a single key site such as Harappa, there remain gaps in the radiocarbon chronologies (Kenoyer Reference Kenoyer1997a). Despite this, Kenoyer has correctly described Harappa as one of the most important sources of information about the whole Indus Civilisation and it has been the focus of numerous collaborative research projects.

The point we wish to make here is that although scientific dates for specific sites and different cultures in South Asia do exist, they often stand in isolation, and even when dealing with one of the most studied sites in South Asia, there are gaps and uncertainties. While Kenoyer has been able to draw on more than 70 radiocarbon date estimates, along with stratigraphy, architectural analysis and diagnostic artefacts in order to construct this chronology for Harappa (Reference Kenoyer1997a: 266), few other Indus sites have been sampled so intensively. For example, Sonawane has presented radiocarbon date estimates from ten sites in Gujarat, which are believed to fall into what is termed the ‘post-urban’ Indus, or the period falling between circa 1900 and 1400 BCE (Reference Sonawane, Settar and Korisettar2002: 168). Yet of the ten sites covered, only one (Rojdi) had more than five date estimates, while three sites had only a single date estimate (Sonawane Reference Sonawane, Settar and Korisettar2002). Similarly, a synthesis of ‘pre-Indus’ cultures in Gujarat has presented material from eight sites, of which one has four date estimates, while five sites have two or only a single date estimate (Ajithprasad Reference Ajithprasad, Settar and Korisettar2002: 133). Using a single date to evaluate the sequence of an entire site is always challenging but there are ways of mitigating the risk. For example, evaluations of the radiocarbon measurement ‘chronological hygiene’ have been developed in other parts of the world in order to separate more reliable dates (e.g. Pettitt et al. Reference Pettitt, Davies, Gamble and Richards2003), but such studies have not been adopted broadly within South Asia. Moreover, it is possible to apply OxCal to sequences of radiocarbon measurements from single sites to obtain a greater chronological resolution through the application of Bayesian statistics, but few projects have chosen this route. While radiocarbon measurements and Optically Stimulated Luminescence dating remain so expensive, and thus out of the range of many projects within South Asia with notable exceptions (Haricharan et al. Reference Haricharan, Achyuthan and Suresh2013), more traditional techniques of typologies will continue to be used to develop relative dating frameworks. We will return to specific chronologies for sites, regions and cultural developments in following chapters, but these examples serve to illustrate some of the difficulties archaeologists face when there is an absence or scarcity of scientific dates around which to construct a chronological framework. We will also consider the problems which can arise when archaeologists attempt to link material culture to known historical dates and sites identified through historical geography.

Traditions, Eras and Phases

One of the tasks archaeologists frequently undertake is the subdivision and differentiation of the past, and within South Asia this is no different. As noted previously, many of the pioneering archaeologists and antiquarians to work within South Asia brought with them existing systems from elsewhere, such as the Three Age System. Additionally, archaeologists have often adopted the application of a tripartite division to the civilisations of the world, and this includes the Indus. In the mid-twentieth century, Wheeler divided the Indus Civilisation into three main periods: the Early, the Mature, and the Late Indus. The highly influential British archaeologists Raymond and Bridget Allchin used similar subdivisions in their work (Reference Allchin and Allchin1982), and this largely cemented the chronological nomenclature in common use (Singh Reference Singh2008). We believe that the continued use of these descriptive, limited chronological terms has contributed to the restricted approaches to understanding the development and decline of the major urban-focused developments in South Asia, and we hope that by moving beyond these traditional chronologies we may begin to provide a framework to enable us to look at alternative ways of exploring and discussing key events and processes.

As noted earlier, many works on South Asian archaeology have either followed a broad narrative concentrating on India and Pakistan to the large exclusion of neighbouring territories or have studied the developmental sequence of only the Indus or that of the Early Historic. Reinforcing a division between the two which dates back to the early years of archaeology in South Asia, and despite evidence of continuity to the contrary (Sengupta and Chakraborty Reference Sengupta and Chakraborty2008; Agrawal Reference Agrawal2007; Shaffer Reference Shaffer, Spodek and Srinivasan1993; Coningham Reference Coningham and Allchin1995a; Kenoyer Reference Kenoyer1991b), the techniques, theories and methodologies for studying the two traditions continue to remain separate – as do most the majority of their practitioners. A good example of this dichotomy is illustrated by a comparison of Shaffer’s Reference Shaffer and Ehrich1992 study of the Indus Valley, Baluchistan and the Helmand chronologies with the chronologies of Possehl and Rissman’s Reference Possehl, Rissman and Ehrich1992 study for the Early Historic period of the same region. Although both of these chronologies were presented within the same volume, the ‘Chronologies in Old World Archaeology’ (Ehrich Reference Ehrich1992), and despite sharing the same geographical region, they followed entirely separate frameworks and therefore there were no links between the two. Whilst significantly different from Possehl’s more traditional approach, Shaffer utilised his chapter to pioneer the establishment of an innovative developmental framework for the north-western areas of South Asia which drew on three general archaeological structures: Tradition, Era and Phase (Reference Shaffer and Ehrich1992: 411).

Rather than restricting focus to the Indus cities or even the ‘Early Harappan period’, Shaffer and Lichtenstein argued that it was possible to perceive a broader ‘Indus Valley Cultural Tradition’; identifiable as “persistent configurations of economic adaptations, basic technologies and other cultural systems within the context of temporal and geographical continuity” (Reference Shaffer, Lichtenstein and Kenoyer1989: 119). This definition allowed these scholars to argue for the “integration of both stylistically similar and diverse patterned sets of archaeological assemblages into a single analytical unit which implies the existence of cultural and chronological relationships” (Shaffer and Lichtenstein Reference Shaffer, Lichtenstein and Kenoyer1989). This larger analytical unit was made up of Phases grouped within a number of Eras, in which the Phase represented “the smallest analytical unit; and its main feature is a diagnostic ceramic style located at one or more sites during a particular time” (Shaffer Reference Shaffer and Ehrich1992: 442). The Era was, in turn, defined as forming “a sequential series proceeding in the same order and connoting changes in general cultural organisation within the areal traditions” (Shaffer Reference Shaffer and Ehrich1992: 442). He further identified the presence of four major Eras within the Indus Valley Tradition: Early Food Producing, Regionalisation, Integration and Localisation. The first of these was defined as an Era characterised by “an economy based on food production and an absence of ceramics” and the second as an Era of “distinct artefact styles, essentially ceramics, which cluster in time and space, and interaction networks which link dispersed social groups” (Shaffer Reference Shaffer and Ehrich1992: 442). The third, Integration, was defined as “pronounced homogeneity in material culture distributed over a large area reflecting an intense level of interaction between social groups” and the fourth, Localisation, as being “comparable to regionalisation except that there is a more generalised similarity in artefact styles, including continued, but altered, presence of interaction networks” (Shaffer Reference Shaffer and Ehrich1992: 442). We recognise that other scholars have used ‘Integration’ to denote “the political process in which differentiated social groups come to exist within an institutionalised framework” and that “States have the power to disembed resources from the differentiated groups for their own ends and glorification, not least because symbols of incorporation are so critical in establishing the legitimacy of societies.” (Yoffee Reference Yoffee2005: 32–3). However, it is important to note that at no stage did Shaffer suggest that integration implied a complete social, economic and political consolidation within a single state, and we follow this understanding.

Associated with the concept of interaction systems with avenues of social communication within and between social groups across traditions and phases, the framework was developed to examine longer-term sociocultural developments, and Shaffer suggested that the exploration of these structures “not only may adumbrate social interrelationships for which we have no specific information as to their nature, but they are also highly important in establishing relative chronologies when radiocarbon determinations are inadequate and in reinforcing those dates that do exist” (Reference Shaffer and Ehrich1992: 442). A critical feature of Shaffer’s developmental framework was replacing the traditional Mesolithic/Neolithic, ‘Chalcolithic’/Early Harappan, Mature Harappan and Late Harappan terminology with Eras which were intended to reflect the longer-term changes or processes which provided the platform for eventual complexity and urbanisation. Thus, in Shaffer’s scheme, there was a development from the Food Producing Era to the Regionalisation Era, then to the Integration Era and finally the Localisation Era (Reference Shaffer and Ehrich1992). Notably, Shaffer’s categorisation also allowed scholars to frame sites such as Mehrgarh, accepted by all as partly ancestral to the Indus cities within a distinctly pervasive Indus Tradition rather than lying outside a Pre-Urban or incipient urban phase.

Shaffer’s chronological framework has been successfully adapted and adopted by a number of scholars, such as Mark Kenoyer in his 1997 chapter on Early City-States in South Asia and his 1998 book Ancient Cities of the Indus Valley Civilisation, and by one of the present authors in his 2005 chapter on the archaeology of South Asia in Thames, and Hudson’s The Human Past. Other scholars, whilst not adopting it entirely, have developed parallel themes, and Rita Wright’s recent volume on the ancient Indus followed a similar framework with an Early Food Producing Phase, followed by a Pre-Urban Phase, an Urban Phase and a Post-Urban/Late Harappan Period (Reference Wright2010: 22). It is equally important to note that not all archaeologists have, by any means, adopted Shaffer’s framework. For example, the late Greg Possehl grouped archaeological phases into a seven stage development sequence from: Beginnings of Village Farming Communities and Pastoral camps (Stage One), Developed Village Farming Communities and Pastoral Societies, the Early Harappan, the Early Harappan-Mature Harappan Transition into the Mature Harappan. From the Mature Harappan, the sequence travelled through the Posturban Harappan, and the Early Iron Age of Northern India and Pakistan (Stage Seven) (Reference Possehl2002a: 29). Possehl’s mixture of older periodisation (Mature Harappan), artefact-based descriptive classifications (Early Iron Age) and socio-economic processes (Developed Village Farming Communities) is not unique and others, such as Singh (Reference Singh2008), have presented similar categories which treat the Indus Valley and the Early Historic Traditions in very different ways and thus reinforce established divisions which prevent easy comparative discussion.

We will adopt Shaffer’s framework in this volume in order to better understand and explore the processess which led to the two main urban-focused developments in South Asia and, in the following chapters, we first investigate the developmental sequence of the Indus Valley Tradition. This approach will also allow us to escape that paradox highlighted by B. D. Chattopadhyaya, whereby current divisions between the ‘Protohistoric’ cities of the Indus and ‘Historic’ cities of the Early Historic appear to pivot on interpretations of the nature of the Indus script and its use (Reference Chattopadhyaya, Sengupta and Chakraborty2008: 8). Following this new scheme, we will characterise the Indus urban-focused tradition, from its food-producing origins to its post-urban fragmentation and transformation, and we also compare these stages with the developmental trajectories in neighbouring regions. We then compare the Indus developmental character with the characteristics of the succeeding Early Historic Tradition utilising Shaffer’s approach. We also begin to replace the traditional terminologies of ‘Chalcolithic’, Iron Age, Proto-Historic, Early Historic and Mauryan with those of a ‘Localisation Era’ followed by an Era of ‘Regionalisation’ and an Era of ‘Integration’. We argue that Kenoyer’s (Reference Kenoyer1998) suggestion that the Era of Integration was only reached with the Mauryan period (c. 317 BCE) was overcautious and that such a cultural and economic stage became evident in the archaeological record as early as 600 BCE, although the actual stage of political integration is much later according to ‘historical’ sources.

This task is likely to be controversial and we acknowledge that not all scholars will be receptive. Indeed, we note that some academics have advocated even the abandonment of more traditional terminology and advocated the phasing of the archaeology of the Sarasvati River within a sequence which leads from Period I: Rigvedic through Period II: Brahmana Period and Period III: Mahabharata Period to Period IV: nineteenth century CE (Gupta Reference Gupta and Tiwari2010: 17). There are also a number of issues still to be refined, and it remains questionable whether there is sufficient difference and distinction between Shaffer’s definitions of Regionalisation and Localisation. Shaffer’s own definition (quoted earlier) observes the similarities of the two eras, with some differentiation in the form of contact between groups. In turn, we have retained this separation and nomenclature, although we recognise the overlap, and part of our aim in this volume is to further differentiate between the regionalisation (emerging complexity) eras and localisation (declining or contracting complexity) eras for both the Indus and Early Historic periods. Indeed, such a cyclical process had been successfully piloted by Louis Dupree, who referred to phases of ‘fusion’ and ‘fission’ when discussing the history of Afghanistan (Reference Dupree1973: 344). We can also question the relevance of the term Integration to refer to the period of Indus urban development as large swathes of northern and southern South Asia were unaffected by what was, on a subcontinental scale, a regional feature. This issue is easier to address as the Era of Integration is linked to the coverage of the Indus Valley Tradition and areas interacting with it, rather than being applicable to the entirety of the Subcontinent. Chase et al. have also questioned the extent to which the ‘borderlands’ were ever fully integrated and have suggested that in such areas “residents of various backgrounds and interests negotiated novel social identities in the context of ever-changing social, economic and political networks.” (Reference Chase, Prasad, Rajesh, Patel and Sharma2014: 64). The application of such a question to the Mauryan Empire is similarly complex, particularly if one endorses the suggestion of Monica Smith that cities were effectively linked, but ideological linkage suffered severe distance decay beyond the main networks between nodes (Reference Smith2003). Certainly, Sengupta and Chakraborty have stressed that the majority of the populations of Early Historic South Asia were rural rather than city dwellers (Reference Chakraborty, Sengupta and Chakraborty2008: xxiii). However, as Manuel has noted, it should be remembered that the temporal and spatial boundaries of Shaffer’s Phases should be considered flexible and possibly overlapping as “they are purely modern archaeological constructs derived from artefactual typologies and (admittedly few) scientifically obtained dates” (Reference Manuel, Gunawardhana, Adikari and Coningham2010: 148). Manuel has also focused on the beneficial flexibility offered by Shaffer’s concept, a flexibility which allows a focus on social processes and dynamics as well as the ability to integrate new discoveries, such as the ease with which the Ravi Phase was introduced (Manuel Reference Manuel, Gunawardhana, Adikari and Coningham2010: 151). These are all issues which we will investigate further throughout the volume, and we stress that there was a difference in scale, as the Indus Tradition was focused on the north-west of the subcontinent and the Early Historic on its entirety – in the words of Jaya Menon “Unlike the Harappan, the early historic does not offer us, either chronologically or geographically, a compact entity for analysis.” (Reference Menon, Sengupta and Chakraborty2008: 15). We feel strongly that we need to both utilise a single uniform yet flexible terminology to effectively bridge between both eras of development but, at the same time, are keen to avoid the creation of another new periodisation scheme.

After considering the available data, we will also briefly address the ongoing debate about whether there is an inherently ‘South Asian’ character to these urban-focused developments, or whether they owe their characteristics to developments on their flanks. We will also reassess the nature of the intervening timespan by tracing strands of cultural continuity between the two before considering why South Asia’s first urban-focused development lost its integrated character. Once we have presented and analysed key material culture from a range of selected sites, we will carry out a critical comparison of the two traditions, which remains one of the fundamental themes of this volume. This focused approach will simplify the dated and overspecialized terminologies for the region while allowing us to link its sequence with those of neighbouring regions. Using Shaffer’s terminology will also allow us to demonstrate that not only are the sequences in the two urban-focused developments internally similar, but also that they are fundamentally different in character from those in neighbouring regions, such as West Asia. Moreover, we will argue that these sequences, and their material culture, do not respect many of the current generic archaeological models for explaining cultural change or evaluating cultural complexity. For example, the notable lack of evidence of ranking within the Indus cities, where the absence of royal burials, palaces and temples have suggested that models based on normative values of wealth and rank clearly fail (Miller Reference Miller1985; Possehl Reference Possehl, Feinman and Marcus1998; Rissman Reference Rissman1988).

Dating Conventions in the Text

What is the difference between the abbreviations BC and BCE or AD and CE when it comes to dating? BC and AD are the abbreviations most commonly used in archaeological, historical and general literature, and they are abbreviations for the terms ‘Before Christ’ and ‘Anno Domini’ respectively. ‘Anno Domini’ is a Latin phrase meaning ‘in the year of our Lord’ and has come to refer to a numbering system for years following after the birth of Jesus Christ. ‘Before Christ’ derives from the Ancient Greek word ‘Christos’ or ‘Anointed One’ and has come to refer to the years prior to the birth of Christ. Although a clear and widely accepted system for recording chronologies, the BC/AD system has some major drawbacks for many archaeologists – as it is western-derived, European-centric and bound up with Christianity. When working in areas outside Europe, we are of course working largely outside the Christian framework, and because of this, attempts have been made to find alternatives to the BC/AD notation.

The Islamic or Hijri calendar differs from the Western or Gregorian calendar in a number of ways. The Hijri calendar is a lunar calendar, meaning it is based on the moon, rather than on the solar Gregorian calendar. Because of this, the Hijri year is about ten or days days shorter than the solar year, which is why Islamic festivals are assigned different dates in each year. The Hijri calendar is also linked to the Prophet Muhammad, who established Islam. It begins with the Prophet’s journey from Mecca to Medina and is abbreviated to H or AH, the latter meaning ‘Anno Hegirae’, which is Latin for ‘In the year of the Hijri’. The Prophet’s journey is generally agreed to have taken place in the western calendar year of 570 AD, although there is some debate about this. However, there are many other groups in South Asia who are not Muslim and therefore do not follow the Hijri calendar. In an attempt to find a more universally acceptable and religiously neutral alternative, the abbreviations BCE and CE have been suggested. BCE stands for ‘Before Common Era’ and replaces BC, with the years being equivalent. CE stands for ‘Common Era’ and replaces AD, and again the years are equivalent. Although we recognise that there are still links to the Christian calendar, we will be using BCE and CE as dating conventions in this volume as they are relatively neutral, while still being widely understood and accepted. Where we draw directly from published scientific work, for example, palaeoecology studies relevant to the environmental background of our study region, we use the abbreviation BP as given in the original text in order to minimise the possibility of cumulative error through translation. BP refers to years before the standard benchmark date of 1950.

Conclusions

This volume is subdivided into four main parts, the first of which lays the foundations for the analytical and interpretative study of the other three. In this introductory chapter, we have outlined our main themes and approaches and, in the remaining chapters of Part I, we will introduce the geographical and environmental context of South Asia as well as a review of the historical development of South Asian archaeology as a discipline. Part II is concerned with the Indus Valley Tradition in the north-west of the subcontinent between circa 6500 and 1900 BCE and will present the four eras of the Indus urban-focused tradition, from food producers and regional proto-urban communities, through the emergence of an integrated urban and rural system, before considering the evidence for the apparent end of this tradition. Each of the three chapters in Part II will also draw on selected evidence from neighbouring communities in order to investigate the comparative development of this region and to test the extent to which diffusion or autochthonous development occurred.

The third part of this volume will consider the Early Historic Tradition between circa 1900 and 200 BCE and begins with a re-examination of the period which separates the two urban-focused traditions and will stress the elements of continuity and transformation through the sequence. We then outline the developmental stages by which urban forms re-emerged during the Early Historic throughout the subcontinent and compare them with Indus Valley Tradition and those of the contemporary developments to the west. In the fourth and final part, we will return to some of the fundamental issues raised by the volume such as the similarities and differences between the two major urban-focused developments, the role of indigenous development and the diffusion of ideas and innovations in South Asia, and the nature of complexity. We will also consider Tainter’s Reference Tainter1988 model of marginal returns, and other archaeological models and interpretations dealing with the collapse and transformation of civilisations, in order to consider why the first urban-focused tradition ended with such a loss of communal traits, whilst still passing on a number of fundamental continuities. We will also identify certain key areas and phases in South Asia which we believe should be investigated in more detail if we are to better understand the social and economic dynamics involved in these transformations.