A historical region, not a timeless border Mithila, Vajji, and Anga belong to different kinds of historical geography. Mithila is at once a remembered kingdom, a scholastic-cultural region, and a modern language homeland spanning north Bihar and Nepal's eastern and central Tarai. Vajji names an ancient confederated political formation centred on the north Gangetic plain. Anga was an eastern early-historic polity associated in textual tradition with Champa and the lower middle Ganga corridor. Their boundaries shifted; none maps neatly onto a modern state or district. This book therefore follows networks, rivers, institutions, and remembered centres rather than drawing a single permanent frontier (Singh 2008; Thapar 2002). Water as archive and agent The Himalayan rivers that cross the Tarai and north Bihar deposit silt, abandon channels, flood settlements, and erase or bury archaeological surfaces. The Ganga simultaneously linked Anga and the north-bank regions to long-distance traffic. Rivers were not a static backdrop: they structured cultivation, ferries, markets, pilgrimage, revenue assessment, and disaster. Later chapters show how attempts to stabilize these rivers also redistributed risk. Rules of evidence Archaeological strata establish material sequences, not ethnic labels. Texts disclose categories and memories at the time of composition and transmission; they do not automatically prove every narrated event. Inscriptions are usually firmer chronological anchors but are public acts of power and piety, not neutral chronicles. Gazetteers contain valuable observation together with colonial classification. This layered method is essential in a region where sacred geography, dynastic memory, and political identity remain active. Three spatial scales The narrative moves among three scales that should not be confused. The first is the site or locality: an excavated mound, a river crossing, a settlement cluster, or a named shrine. The second is the historical region, reconstructed from distributions of objects, texts, routes, institutions, and remembered centres. The third is the state, whose administrative reach varied and whose documents often describe claims more readily than everyday practice. Mithila may be meaningful at all three scales, but not in the same way. Vajji is most securely approached as an early political field known through textual and archaeological evidence around Vaishali, whereas Anga is reconstructed from eastern political geography and the Champa-Bhagalpur corridor. Keeping these scales separate prevents a capital, a language area, and an ancient polity from being treated as interchangeable. Names as historical evidence A place name is evidence, but it is never self-interpreting. The survival or revival of a name may register memory, political aspiration, pilgrimage, literary authority, or administrative convenience. Conversely, the disappearance of a name from official records need not mean that the cultural region vanished. This study therefore asks who used a name, in which language, for what territory, and in what genre. The method is especially important for Mithila and Tirhut, for Vaishali and Vesali, and for Champa and modern Bhagalpur. Similar-looking names can preserve genuine continuities, yet continuity must be demonstrated through dated records and material contexts rather than assumed from resemblance alone (Singh 2008; Thakur 1988). Periodization without sharp walls The five-volume division is an editorial instrument, not a claim that social life changed everywhere at one date. Prehistory gives way unevenly to protohistory and early history as texts, inscriptions, coins, and datable institutions become available. Medieval and early-modern boundaries likewise depend on the question being asked: dynastic change, agrarian institutions, literary production, and religious practice followed different rhythms. The modern period begins here with the transformation of revenue and political authority under Company and colonial rule, while the contemporary period follows Indian independence and the end of Rana oligarchy in Nepal. Transitional overlap is therefore expected rather than treated as a defect in the chronology. The modern border as an analytical problem The India-Nepal boundary is indispensable for the history of modern sovereignty, citizenship, law, censuses, and infrastructure. It is anachronistic, however, when projected unchanged into prehistoric or ancient landscapes. Rivers, foothill routes, marriage networks, pilgrimage, trade, and political influence crossed the line that later became an international frontier. The book consequently uses present national locations to orient the reader while refusing to divide earlier evidence into sealed national containers. This is not an argument that borders are unreal; it is an argument that their historical force must be dated. A source-controlled research practice No single class of source can carry this regional history. Archaeological reports require attention to trenches, layers, sampling, and later disturbance. Texts require attention to language, genre, redaction, and manuscript transmission. Inscriptions demand accurate reading, provenance, and calendrical conversion. Colonial surveys and gazetteers preserve valuable observations while organizing people and land for rule. Modern censuses and statutes are authoritative for their own categories and dates, not for timeless identity. The cumulative method compares these records, marks disagreement, and leaves gaps visible. A responsible large history is built by increasing the density of checkable evidence, not by filling silence with seamless narrative. The Himalayan foreland as a historical problem The north Bihar plain and the adjoining Nepal Tarai form part of the Himalayan foreland, a zone in which relief, sediment supply, monsoon discharge, and low downstream gradients combine to make water and sediment historically consequential. The region cannot be represented adequately as a stable sheet of cultivated land crossed by rivers whose channels merely carry water from north to south. Modern geomorphological work describes river systems with markedly different source areas and sedimentary behaviours. Rajiv Sinha and Peter F. Friend distinguished mountain-fed, foothills-fed, plains-fed, and mixed- fed systems in northern Bihar, emphasizing that their discharge, sediment concentrations, channel forms, and depositional settings differ. The value of that classification for history is methodological rather than deterministic: it warns against assuming that every named river created the same kind of floodplain or that the same crossing, bank, or channel retained the same significance through long periods (Sinha and Friend 1994). For the history developed in this book, the Himalayan foreland therefore has to be treated simultaneously as terrain, archive, and agent. Terrain matters because routes, settlement choices, cultivation, ferries, and political jurisdictions operated through particular landforms. It is an archive because abandoned channels, levees, flood basins, buried soils, settlement mounds, and archaeological deposits preserve traces of earlier conditions, although often incompletely. It is an agent because channel movement, inundation, erosion, and sedimentation could alter access to land, move a market away from a crossing, expose an old mound, bury another, or transform the practical meaning of an administrative boundary. None of these processes determines political or cultural outcomes by itself. Their importance lies in changing the field within which households, states, religious institutions, and traders acted (Hill 1997; Sinha and Friend 1994). River types, megafans, and interfan plains The distinction between mountain-fed and smaller foothill- or plains-fed rivers is especially important for avoiding a generic history of “the flood.” Large Himalayan rivers may construct broad depositional systems, while smaller rivers can occupy interfan areas with different patterns of channel instability and waterlogging. Sinha and Friend's study of northern Bihar linked source area to contrasting hydrological and sedimentary regimes and noted widespread evidence of channel movement. These are modern scientific observations, not direct reconstructions of prehistoric channel positions, but they provide a disciplined physical framework for interpreting why a site distribution or historical route may be uneven. A locality situated on a favourable surface in one period could later become difficult to reach, eroded, or deeply buried without any corresponding cultural collapse (Sinha and Friend 1994). The Kosi, Gandak, Bagmati, Kamla and other rivers must consequently be handled at two levels. At the first level, each river has a physical history that can be investigated through geomorphology, old maps, satellite imagery, hydrology, and field observation. At the second, each acquires historically specific social meanings through revenue boundaries, irrigation works, embankments, pilgrimage, local memory, litigation, transport, and disaster. The two histories overlap but are not interchangeable. A modern channel map cannot be projected backward as if it were an ancient itinerary, and a textual river name cannot be fixed to a modern course without independent argument. Later chapters follow this rule when discussing the Sadanira, the Kosi, the Bagmati, and rivers used in treaty or administrative boundary clauses. Avulsion and the instability of apparently fixed geography Channel avulsion—the abandonment of one course in favour of another—is not a marginal exception in the north Bihar plains. Vikrant Jain and Rajiv Sinha reconstructed a long migration history for the Bagmati from maps and satellite images and identified eight major and several minor avulsion events over roughly 230 years. Their result does not allow a historian to infer where the river flowed in a much earlier century without evidence. It does, however, establish that the assumption of channel permanence is unsafe even within the documentary period. Historical maps that disagree about a river may therefore record cartographic error, seasonal variation, different survey dates, or genuine geomorphic change; those possibilities must be tested rather than collapsed into one “correct” line (Jain and Sinha 2003). This matters for political and legal history because rivers frequently appear as boundaries, routes, taxable resources, and markers of jurisdiction. Where a boundary follows a moving channel, a legal phrase that looks geometrically precise may generate practical ambiguity on the ground. Where an old channel remains visible as a depression, wetland, or local name, later communities may remember a geography different from that on an official map. The book therefore treats historical river courses as dated propositions. A statement such as “the Bagmati flowed here” requires a date and source; a statement such as “this is the Bagmati” in a literary text requires attention to the text's date, transmission, and geographical conventions. Alluvium, site formation, and archaeological visibility Alluvial landscapes complicate the relation between where people lived and where archaeologists can now see them. Flood deposition can seal occupation surfaces beneath later sediment; erosion can remove them; cultivation and construction can truncate mounds; and a shifting channel can redeposit objects away from their original context. The resulting archaeological map is therefore partly a map of preservation and investigation. A dense cluster of recorded sites may reflect genuine settlement concentration, better-surviving mounds, road access, a history of archaeological survey, or some combination of these factors. Conversely, a blank area cannot be treated automatically as an uninhabited zone (Singh 2008). The implication is particularly important for attempts to map ancient Mithila, Vajji, or Anga from selected excavated places. A securely identified site can anchor a local history; it cannot by itself draw the frontier of a polity. The archaeological record becomes more informative when site sequence, ceramics, inscriptions, environmental setting, and route geography are compared across a wider distribution. This is why the later chapters on Vaishali and Champa distinguish the evidentiary strength of an excavated mound, a textual place name, and a reconstructed territorial extent. Historical geography is not produced by joining a handful of famous sites with a boundary line. Historical geography is a reconstruction, not a recovered political map Historical geography combines topography, texts, archaeology, epigraphy, routes, settlement distributions, and later records in order to reconstruct spatial relationships at particular times. Mithila Sharan Pandey's historical geography of Bihar remains useful because it treats river systems, regions, place names, routes, and topography as connected problems rather than as an appendix to dynastic history. Dilip K. Chakrabarti similarly stresses the need to relate urban centres, geographical units, and routes rather than discuss cities as isolated points. The approach adopted here follows that relational principle while applying stricter labels to the evidentiary status of each reconstruction (Pandey 1963; Chakrabarti 2006). A historical region is therefore neither imaginary nor necessarily bounded like a modern district. It becomes historically observable through repeated relationships: administrative names, court patronage, markets, marriage networks, pilgrimage, language practices, land grants, routes, learned institutions, and remembered centres. Different relationships may produce different spatial extents in the same century. A revenue division called Tirhut, a literary invocation of Mithila, and a Maithili-speaking social field may overlap substantially without being coextensive. The analytical task is to identify the source and function of each spatial claim before comparing them. Modern orientation without retrospective boundaries Readers need modern geographical reference points even when the object of study predates modern states. The orientation map below therefore shows selected present-day places in India and Nepal without drawing a proposed frontier for Videha, Vajji, Anga, Mithila, or Tirhut. Its purpose is navigational: it makes clear, for example, that Janakpur and Simraungadh are now in Nepal, while Darbhanga, Vaishali/Basarh, Patna, Purnia, and Bhagalpur are now in India. Those modern locations help readers follow subsequent chapters, but they do not decide the territorial extent of earlier polities or cultural regions. Map 1. Selected present-day reference points used for orientation in this book. Modern state boundaries are shown only to locate places; no frontier of Videha, Vajji, Anga, Mithila, or Tirhut is implied. Editorial map prepared for this book, 2026. Underlying country polygons: Natural Earth 1:110m public-domain cultural vectors; selected point locations checked against GeoNames/open geographic gazetteers. Rights: Natural Earth data are public domain; editorial overlay © Gajendra Thakur, 2026. Source links: Natural Earth ; GeoNames Videha, Mithila, Tirabhukti, and Tirhut: related names with different histories The names Videha, Mithila, Tirabhukti, and Tirhut are connected in regional historical writing, but the connection should not be flattened into permanent synonymy. Videha belongs prominently to Vedic, Upanishadic, epic, Buddhist, Jain, and later historical memory; Mithila functions in different periods as a royal, sacred, scholastic, literary, linguistic, and cultural name; Tirabhukti appears as an administrative- territorial designation in early-medieval contexts; and Tirhut becomes a durable later regional and administrative name. The survival of one name alongside another can indicate continuity of memory or regional association without proving that their borders were identical (Pandey 1963; Thakur 1988). For this reason the book uses each term where the source warrants it. It does not rename every early reference to Videha as “Tirhut,” nor does it call every colonial district “Mithila” without qualification. When a modern author reconstructs Mithila as extending between named rivers or across particular districts, that reconstruction is cited as a scholarly claim and compared with the underlying texts and records. The methodological distinction is essential because regional identity remains socially and politically meaningful in the present. Historical criticism is strongest when it protects that significance from false precision rather than attempting to validate it through unsupported ancient cartography. Vajji and Anga are not subdivisions of one timeless Mithila Vajji and Anga enter the record through different political and textual histories. Vajji is most securely approached as an ancient confederated political field associated especially with Vaishali and the Licchavis. Anga is an eastern polity associated with Champa and the Ganga corridor toward what is now the Bhagalpur region. Later regional identities can remember or incorporate these names, but the book does not treat them as administrative subdivisions of a single timeless Mithila. Doing so would erase the very political plurality that makes the eastern Gangetic plain historically important (Singh 2008; Thapar 2002). The same caution applies in reverse. The fact that Vajji and Anga were distinct early political formations does not imply that later economic, linguistic, devotional, or literary networks respected their ancient separateness. Historical space changes. By tracing transformations rather than assuming inherited compartments, the book can ask how an ancient political name persisted, disappeared, revived, or acquired a new cultural meaning. This is a history of changing spatial vocabularies as much as a history of rulers. Routes, crossings, and nodal places Political boundaries alone are poor guides to regional connectivity. The Ganga linked the Anga-Bhagalpur corridor to Patna, Bengal, and the wider north Indian river system; northern routes connected the plains with the Tarai and Himalayan valleys; ferries and seasonal crossings linked opposite banks; and market, pilgrimage, and scholastic circuits created repeated movement among places that belonged to different jurisdictions. Chakrabarti's work on early-historic geographical units and trade routes is useful here because it treats urban centres as nodes within larger relational fields rather than as self-contained capitals (Chakrabarti 2006). Route history also requires chronological discipline. A railway line documented in a colonial atlas cannot be used to prove an ancient commercial road. A modern highway may follow an older corridor, but continuity has to be demonstrated through intermediate evidence. Conversely, a route can remain important while its technology changes from footpath and ferry to metalled road or rail. Later chapters therefore distinguish corridor continuity from infrastructure continuity. This distinction is especially significant on the India-Nepal border, where kinship, pilgrimage, and market movement can persist through successive regimes of sovereignty and documentation. The modern border after Sugauli The international boundary between India and Nepal has a history that must be dated rather than projected backward. The Treaty of Sugauli was concluded after the Anglo-Nepal War: the text was dated 2 December 1815 and the exchange of ratification recorded on 4 March 1816. Its provisions included Nepalese renunciation or cession of territories described through rivers and lowland tracts. The treaty is therefore a primary legal anchor for the nineteenth-century reordering of sovereignty, but it was not the last word in every sector of the boundary, and later territorial adjustments and demarcation practices also mattered (Treaty of Sugauli 1815/1816; Whelpton 2005). For earlier centuries, that line is anachronistic. Courts, estates, pilgrimage centres, kin networks, forests, pasture, and trade routes connected the Tarai and north Bihar through relationships that did not depend on modern passports or national censuses. For later periods, however, the border became increasingly consequential because law, taxation, policing, citizenship, land registration, infrastructure, and state statistics were organized nationally. Borderland scholarship usefully holds both facts together: states make real territorial distinctions, while connected communities continue to produce social spaces that cross them (Ibrahim and Kothiyal 2021). Colonial and archaeological maps are historical arguments Maps made under colonial rule or early archaeological survey are invaluable sources, but they are not transparent windows onto ancient geography. Alexander Cunningham's 1880-81 survey of north and south Bihar, for example, combined field observation, inscriptions, textual identifications, routes, and earlier scholarship in an effort to locate ancient places. The map associated with that survey is evidence for nineteenth-century archaeological reasoning and for the topography known to its makers. It is not a contemporary map of ancient Vajji, Videha, or Magadha (Cunningham and Garrick 1883). This distinction is more than a warning about old scholarship. Every map selects features, fixes names, chooses a scale, and suppresses uncertainty. Administrative maps may privilege district limits, police stations, railways, or revenue units; archaeological maps may privilege mounds, inscriptions, or identifications; linguistic maps may privilege survey categories. The cumulative history retains useful historical maps when their provenance is clear, but captions state what the map actually documents. Where an ancient boundary is uncertain, the preferred cartographic language is a zone, corridor, point distribution, or explicitly hypothetical line rather than an authoritative-looking frontier. The three spatial scales, supplemented by corridor and borderland The three spatial scales defined above—site, historical region, and state—remain the book's basic hierarchy. Two relational lenses supplement them. The first is the corridor: a river, route, or chain of markets and crossings through which movement can be studied without assuming political unity. The second is the borderland: a zone in which different jurisdictions meet and in which people may organize livelihood, kinship, language, or worship across a formal boundary. Neither lens replaces the state. They make it possible to ask how state claims interacted with movement and locality. These scales should not be collapsed into one another. A site can be securely identified even when the region around it is uncertain. A cultural region can persist through several states. A state can claim a territory without homogenizing language or social practice. A corridor can connect places that belong to rival states. A borderland can be socially integrated in some respects while legally divided in others. Much of the historical confusion surrounding Mithila, Vajji, Anga, Tirhut, and the Tarai comes from treating evidence produced at one scale as if it automatically answered questions at another. Asynchronous change and the problem of period labels The volume divisions used in this book organize a large manuscript; they are not claims that all institutions changed simultaneously. A political conquest may be rapid while agrarian tenure changes slowly. A court language may shift while household speech remains stable. A railway may transform the time required to reach a market without erasing older pilgrimage routes. A constitutional change may redefine citizenship while kinship networks continue across the same border. Periodization is therefore a set of questions about change, not a rigid partition of lived time. This is particularly important at transitions conventionally called ancient-medieval, medieval-early modern, colonial- modern, and postcolonial-contemporary. Different chapters use different chronological anchors according to the evidence being interpreted. The emergence of inscriptions, the rise of particular dynasties, the expansion of Sultanate or Mughal institutions, the Company's acquisition of diwani, Indian independence, and the end of Rana oligarchy are all important. None should be treated as a single date on which every village, institution, language practice, or environmental relationship became “new.” Uneven archives on the two sides of a modern border The evidentiary record is institutionally uneven. Archaeological survey histories, excavation publication, district gazetteers, revenue records, court archives, census series, and manuscript catalogues developed differently in British India, princely and zamindari domains, and Nepal. Modern research access also differs among archives and languages. A denser documentary record for one district or one side of the present border can therefore create an illusion of greater historical activity. The book treats density of surviving documentation as a property of the archive that must be explained, not as a direct measure of the density of past life. Cross-border comparison consequently requires symmetry of method rather than identical sources. A Nepalese land grant may have to be compared with an Indian revenue record at the level of institutional function rather than document form. An archaeological site in the Tarai may be known through a different survey tradition from a site in Bihar. A Maithili manuscript preserved in one library may illuminate networks extending far beyond the place where the surviving copy was catalogued. The task is to keep provenance visible while reconstructing relationships across archives. Evidence hierarchy without a single privileged source The governing principle of this book is source control, but source control does not mean that one class of evidence is always superior. A stratified archaeological context can establish material association and sequence more securely than a late legend, yet it may not name the people who used an object. An inscription can provide a firm date and title, yet its praise may exaggerate power. A literary text can preserve sophisticated political or social categories, yet its compositional history may span generations. A colonial survey can describe crops, castes, rivers, and markets in extraordinary detail, yet its categories were formed within administrative projects. Each source becomes stronger when its question matches what it can actually show. The workflow diagram below formalizes that practice. Claims are tested for provenance, date, scale, genre, and purpose before evidence from different classes is compared. Agreement strengthens an inference, but disagreement is not edited away for the sake of narrative smoothness. Instead, the prose labels a claim as established, probable, disputed, traditional, commemorative, or otherwise limited. This is especially important in subjects where sacred geography, dynastic memory, regional pride, colonial classification, and modern identity politics all remain active. Diagram 1. Evidence-status sequence used throughout the book. Original editorial diagram prepared for this book, 2026. © Gajendra Thakur, 2026. It summarizes the source-control method stated in this chapter; it is not itself historical evidence. Rules for maps and spatial claims in this book Four cartographic rules follow from the method. First, every map must state its date or the date-range of the evidence from which it is reconstructed. Second, modern state and district boundaries are orientation devices unless the chapter is actually discussing those modern jurisdictions. Third, textual frontiers are shown as claims derived from particular sources, not as surveyed lines. Fourth, archaeological distributions are represented as known points or zones and must not be converted automatically into ethnic or political territory. These rules apply equally to maps reproduced from older scholarship and to new editorial maps prepared for the cumulative edition. A related rule concerns names. Where historical sources use variant forms—Vesali/Vaishali, Campa/Champa, Tirabhukti/Tirhut, Koshi/Kosi—the map and caption should indicate the convention being followed. Modern place labels orient readers but should not silently settle disputed identifications. Where a river identification such as Sadanira = Gandak is conventional but not absolutely demonstrable, the caption or text must retain the qualification. Cartography should display uncertainty rather than hide it behind graphic confidence. Questions carried forward into the remaining volumes This opening chapter establishes the questions that the rest of the book repeatedly asks. Which spatial name is contemporary with the evidence? At what scale does a political claim operate? What river course or route was relevant at that date? Is a monument in its original context or a later reuse? Does a legal text prescribe a norm or record an enforced practice? Does an administrative category describe how people identified themselves, how the state classified them, or both? When later memory associates a place with an ancient figure, what is securely historical about the memory even if the ancient localization remains unproven? The cumulative approach means that these questions are not settled once and forgotten. As each chapter is expanded, new evidence can refine the chronology, alter the map of known sites, sharpen a place-name concordance, or expose a contradiction in older scholarship. The book is therefore designed to grow through verified additions rather than through narrative padding. Expansion toward a very large book is meaningful only when the additional pages increase the density of checkable evidence, make disagreements more explicit, and bring previously underrepresented social groups, places, and source traditions into view. Chapter-specific bibliography Singh, Upinder. A History of Ancient and Early Medieval India: From the Stone Age to the 12th Century. Pearson Education India, 2008. Thapar, Romila. Early India: From the Origins to AD 1300. Allen Lane, 2002. Thakur, Upendra. History of Mithila: From the Earliest Times to 1556 A.D. 2nd ed. Mithila Institute, 1988. Hill, Christopher V. River of Sorrow: Environment and Social Control in Riparian North India, 1770-1994. Association for Asian Studies, 1997. Pandey, Mithila Sharan. The Historical Geography and Topography of Bihar. Motilal Banarsidass, 1963. Sinha, Rajiv, and Peter F. Friend. 'River Systems and Their Sediment Flux, Indo-Gangetic Plains, Northern Bihar, India.' Sedimentology 41, no. 4 (1994): 825-845. https://doi.org/10.1111/j.1365-3091.1994.tb01426.x. Jain, Vikrant, and Rajiv Sinha. 'Hyperavulsive-anabranching Baghmati River System, North Bihar Plains, Eastern India.' Zeitschrift für Geomorphologie 47, no. 1 (2003): 101-116. https://doi.org/10.1127/zfg/47/2003/101. Chakrabarti, Dilip K. 'Relating History to the Land: Urban Centers, Geographical Units, and Trade Routes in the Gangetic and Central India of circa 200 BCE.' In Between the Empires: Society in India 300 BCE to 400 CE, edited by Patrick Olivelle, 4-31. Oxford University Press, 2006. https://doi.org/10.1093/acprof:oso/9780195305326.003.0001. Cunningham, Alexander, and H. B. W. Garrick. Report of Tours in North and South Bihar in 1880-81. Archaeological Survey of India, vol. XVI. Office of the Superintendent of Government Printing, 1883. East India Company and Kingdom of Nepal. Treaty of Sugauli. Dated 2 December 1815; ratification exchanged 4 March 1816. Public-domain treaty text: https://en.wikisource.org/wiki/Treaty_of_Sugauli. Ibrahim, Farhana, and Tanuja Kothiyal, eds. South Asian Borderlands: Mobility, History, Affect. Cambridge University Press, 2021. https://doi.org/10.1017/9781108951500. Whelpton, John. A History of Nepal. Cambridge University Press, 2005.