without false unity Three names, many scales Mithila, Vajji, and Anga do not form a single continuous state, ethnicity, language zone, or administrative unit. They belong to different historical scales and to different moments of political vocabulary. Their histories intersect through the middle and lower Ganga plain, Himalayan foothills, river systems, religious routes, courts, languages, manuscript cultures, landed institutions, markets, migration, and the modern India–Nepal border. A defensible synthesis therefore begins with relation rather than with presumed identity. The names themselves have histories. Videha and Vajji belong to early textual and political worlds; Anga appears as an eastern polity in several ancient traditions; Tirabhukti and Tirhut have administrative histories; Mithila develops overlapping political, literary, ritual, and regional meanings; Madhesh and the Tarai operate at yet other geographical and political scales. Historical connection is strongest when these terms are allowed to remain specific rather than being made interchangeable. Connected history is not a claim of sameness A connected history asks how places, institutions, and communities became linked, what moved across those links, and what changed because of the movement. It does not require the connected units to have been politically united. Merchants can trade across rival jurisdictions, religious specialists can circulate across borders, literary forms can travel between courts, and river systems can bind landscapes that governments divide. Connection is therefore an empirical question about relations, not a rhetorical device for turning difference into unity. This approach is especially important for north Bihar and the Nepal plains because cultural continuities coexist with durable political distinctions. The same pilgrimage route can be sacred, commercial, and international; the same language can have speakers under two constitutions; the same river can be governed by several agencies. The history becomes clearer when those layers are studied together without collapsing one into another. No single territorial line can contain the whole history The maps appropriate to one period can mislead in another. An ancient polity reconstructed from textual references cannot be drawn as though it possessed a modern surveyed boundary. A medieval fiscal designation does not automatically define a cultural region. A colonial district, a zamindari estate, an independent-state province, and a present federal unit are different territorial technologies. Their borders may overlap, but the institutions that produced those borders are not equivalent. For that reason, the book treats territory as evidence that must be dated. Where boundaries are known from administrative records, they may be mapped with relative precision; where they are inferred from sites, routes, or textual geography, the uncertainty must remain visible. Historical maps should show what the evidence permits, not what later identities would prefer the past to have contained. Region is a relationship before it is a polygon A historical region can be produced by repeated relations even when its edges remain variable. Marriage networks, pilgrimage circuits, shared markets, scholarly lineages, manuscript copying, dialect continua, revenue institutions, seasonal labour, and river crossings can generate a recognizable regional field without a single permanent border. This is one reason Mithila can be historically meaningful while its cartographic limits vary across texts and periods. The same principle helps separate Anga from every later administrative map around Bhagalpur and prevents Vajji from becoming a timeless synonym for all of north Bihar. A region is not made unreal by changing boundaries; it is made historical by them. The task is to identify the practices and institutions through which regional coherence was produced in each period. Difference is evidence, not an obstacle to synthesis Contradictions among sources are often treated as defects to be eliminated. In regional history they can instead reveal changes in classification, political interest, memory, or scale. A courtly text, a village genealogy, a census schedule, and a linguistic survey may name the same population differently because they were created for different purposes. Their disagreement is itself part of the history. Synthesis should therefore not mean averaging incompatible statements into a smooth narrative. It should show which claims belong to which evidentiary worlds and explain why categories diverge. A history that preserves difference can be more coherent than one that forces all records into a single vocabulary. Evidence has a hierarchy, but no source is neutral A dated inscription, excavated layer, statute, treaty, coin, court order, or contemporary administrative record can provide a firmer chronological anchor than a much later legend. Yet evidentiary strength in one dimension does not make a source socially complete. A revenue record may be exact about assessment while nearly silent about unpaid labour; a census may count speakers while simplifying multilingual practice; a law may define rights that are unevenly realized. The appropriate hierarchy is therefore claim-specific. Archaeology is strongest for material sequences; inscriptions for particular acts, titles, donations, and dated political assertions; legal texts for formal rules; oral traditions for memory and social meaning; literature for imagined worlds as well as historically situated language and patronage. No single source type should be made to answer every historical question. Provenance precedes interpretation Before a source is used, its provenance must be established as far as possible: who produced it, when, where, for what audience, through what institutional process, and in what surviving form. A manuscript copied centuries after a work's composition carries evidence about both the earlier text and the later transmission. A photograph of a painting records an object but also the circumstances of collection, cataloguing, and reproduction. A digitized document retains the historical record while adding a modern layer of selection and metadata. Provenance is not bureaucratic housekeeping. It determines what kind of claim can responsibly be made. When provenance is weak, the narrative should narrow rather than compensate with confidence. Archaeology cannot name prehistoric peoples by itself The archaeological chapters repeatedly show the value and limit of material evidence. Ceramics, settlement layers, iron- working residues, architecture, botanical remains, and landscape change can establish sequences of occupation and economic practice. They do not, without independent evidence, assign a pot type to a language, caste, clan, or named polity. The temptation to convert material culture directly into ethnic identity is especially strong where later traditions are vivid. The safer method is convergence. Archaeological evidence may be compared with securely dated inscriptions, textual geography, environmental reconstruction, and later settlement history, while keeping the distinct evidentiary status of each source visible. Texts are historical evidence even when they are not chronicles Vedic, Upanishadic, Buddhist, Jain, epic, philosophical, devotional, legal, and courtly texts preserve indispensable evidence. But genres speak differently. A philosophical dialogue is not a gazetteer; a pilgrimage narrative is not a cadastral survey; a praise poem is not a neutral administrative report. Their historical value comes partly from understanding the conventions through which they represent persons, places, institutions, and ideals. Genre criticism prevents two opposite errors: dismissing literary texts as unusable because they are not modern histories, and reading them literally as transparent records. The strongest interpretations ask what a text can establish, what it presupposes, and what remains beyond its reach. Inscription, coin, manuscript, and archive must be read together Political chronology is most secure where independent anchors converge. Inscriptions can date patronage or rulership; coins may fix regnal horizons; manuscripts can reveal copying, ownership, and intellectual circulation; archival records can reconstruct administration. None necessarily supplies a complete dynastic narrative. The history of Karnata and Oiniwar Mithila illustrates why a sequence assembled from mixed evidence is stronger than a genealogy repeated from late compilations alone. The same discipline applies to colonial and modern history. Official files become more informative when placed beside newspapers, petitions, court cases, literary publications, private records, and material landscapes. An archive is not one voice but a contested field of documents. Censuses classify; they do not freeze society Modern censuses are powerful because they provide large- scale demographic evidence within specified administrative categories. They are also historical institutions that ask people and enumerators to select among named options. Mother tongue, caste, religion, residence, occupation, and migration become countable through rules that can change between censuses. The resulting tables are essential evidence, but they should not be mistaken for timeless social ontology. This distinction is central to Maithili, Angika, Bajjika, Bhojpuri, and other language histories, and equally central to contemporary Nepalese categories. A census category may be politically consequential and statistically real while still simplifying multilingual, mobile, or overlapping identities. Memory and tradition belong to history without replacing chronology Sacred geography, genealogical memory, festival narratives, oral histories, local legends, and commemorative practices reveal how communities inhabit the past. They can preserve place associations and social experiences omitted from formal archives. Their evidentiary power is greatest when they are studied as traditions with histories of transmission rather than converted automatically into proof of an ancient event. This approach does not diminish tradition. It gives memory a historical life of its own: who remembers, what is emphasized, how a narrative changes, when a site acquires new significance, and how ritual repetition creates durable belonging. Diagram 54: from source to historical claim Diagram 54. From source to historical claim: provenance, claim fit, cross- checking, confidence labels, and explicit narrative limits form a claim-control sequence. Uncertainty is part of the result Historical writing often becomes less accurate when uncertainty is treated as embarrassment. Terms such as secure, probable, plausible, contested, conventional, reconstructed, and traditional perform analytical work. They allow the reader to distinguish a dated anchor from a scholarly inference and a later memory. The book's chronology uses these differences because precision about confidence is more informative than a single continuous line of apparently exact dates. Uncertainty also creates a revision pathway. A new inscription, excavation report, manuscript witness, court file, or census release can alter one conclusion without requiring the entire history to collapse. A narrative built from graded claims is more durable because it can absorb better evidence. Chronology should be anchored, not overfilled A long chronology can create the illusion that every year is equally known. The more useful chronology identifies turning points whose evidentiary basis is clear: settlement horizons, inscriptions, dynastic transitions, treaties, legal reforms, institutional foundations, flood disasters, linguistic recognition, constitutional changes, and major cultural interventions. Empty intervals are not necessarily gaps in history; they may be gaps in surviving evidence. The chronological framework should therefore be read as a set of anchors connected by analysis, not as a substitute for analysis. Prehistory requires restraint in naming The earliest sections of the book deliberately resist the urge to project later regional names into periods before those names are securely attested. Environmental change, settlement, subsistence, craft production, and exchange can be described archaeologically without calling every early community Maithil, Vajji, Anga, or any later caste or linguistic name. Such restraint is not agnosticism about human life; it is precision about evidence. Once named polities and textual geographies become visible, archaeological and textual records can be compared. Even then, the boundary between material sequence and political name should remain explicit. Videha, Vajji, and Anga belong to different ancient political grammars Ancient sources do not present a single constitutional model for the eastern Ganga plain. Videha is embedded in Vedic and later traditions of kingship and intellectual prestige; Vajji is remembered through corporate institutions and confederated political practices; Anga appears as an eastern polity linked with Champa and with the competitive interstate world described in Buddhist, Jain, and Brahmanical traditions. These are related histories, not interchangeable labels. Their later incorporation into larger political formations also differs. The narrative gains explanatory power when those differences are retained rather than treated as stages in a predetermined regional unity. Gana-sangha should not be translated automatically as modern republic The language of assembly, council, corporate lineage, and collective decision in early sources is historically significant. Yet the modern word republic carries assumptions about individual citizenship, electoral equality, territorial sovereignty, and universal political membership that ancient gana-sanghas did not necessarily share. Translation must therefore preserve institutional difference. Using gana-sangha or corporate polity where appropriate allows comparison with other political forms without turning analogy into identity. It also makes ancient political plurality visible on its own terms. Sacred prestige and political power are related but not identical Mithila's long association with Janaka, Sita, philosophical debate, pilgrimage, and learned traditions produced a powerful sacred and intellectual geography. That geography could reinforce courts, lineages, temples, and modern regional identity, but sacred prestige does not map automatically onto continuous sovereignty. Political capitals, revenue centres, pilgrimage centres, and literary centres can occupy different places and different scales. A connected history therefore studies the interaction of sacred and political geography while allowing each to change independently. Medieval Mithila is not one unbroken dynasty The Karnata, Oiniwar, Sultanate, and later regional histories require attention to changing forms of authority, tribute, military pressure, courtly patronage, local landed power, and wider north Indian politics. Dynastic succession is only one axis. Political history must also ask how institutions survived conquest, how scholarly communities moved, how manuscripts circulated, and how local elites negotiated larger states. This perspective replaces the image of a sealed regional kingdom with a history of courts embedded in transregional systems. The two Vidyapatis must remain distinct One of the book's governing corrections is the separation of two Vidyapatis who should not be conflated. Mahakavi Vidyapati belongs to the earlier Maithili Padavali horizon before Jyotirisvara, while the later Vidyapati Thakkura is associated with the Oiniwar court and with Sanskrit and Avahatta works such as Kirtilata, Kirtipataka, Purusapariksa, Goraksavijaya, and Likhanavali. Treating them as one person distorts literary chronology, patronage history, and the development of Maithili literature. The importance of the distinction extends beyond biography. It demonstrates why a famous name cannot be allowed to override chronology, genre, manuscript evidence, and historical context. Literary history requires the same claim control as political history. Caste classifications must be dated rather than naturalized Later caste and sub-caste classifications cannot be projected backward merely because they became powerful in modern social organization. The book therefore does not treat Srotriya as a fixed medieval Maithil Brahmin sub-caste. The Panji and classification evidence used here places its distinct consolidation in the modern period, around 1800, rather than as a timeless medieval category. This principle applies more broadly. Genealogies, ritual ranks, colonial schedules, reform movements, association politics, and modern reservation categories can all harden social labels that had earlier meanings or looser boundaries. Caste history must explain classification rather than assume it. Persianate and Sanskritic worlds were not sealed civilizations The Sultanate and Mughal periods are best understood through interaction among Persianate administration, Sanskrit learning, vernacular literary production, local landed authority, trade, and religious institutions. Languages and scribal practices could coexist within the same political economy. Conflict and patronage were both possible; neither a simple story of rupture nor one of unbroken continuity is adequate. Attention to translation, revenue terminology, manuscript transmission, and courtly movement reveals a region connected to Jaunpur, Bengal, Delhi, and other centres while maintaining local intellectual traditions. Land is where sovereignty becomes everyday practice Revenue demand, tenancy, survey, inheritance, mortgage, litigation, estate management, and agrarian labour convert abstract sovereignty into daily relations. Mughal fiscal systems, Company diwani, Permanent Settlement, Darbhanga estate administration, zamindari abolition, and Nepalese land regimes changed who could claim, record, transfer, and tax land. Those changes were never only legal; they reshaped credit, caste power, household strategies, and migration. There is no single linear movement from 'feudal' land to modern private property. Rights remained layered and were repeatedly translated into new documentary forms. Darbhanga Raj demonstrates the difference between estate and state Darbhanga Raj was a major landed estate with immense economic, cultural, and political influence, but it was not a princely state under British paramountcy. That distinction matters because estate power operated through property, revenue collection, patronage, litigation, and social authority inside a colonial province. Calling it a kingdom in the constitutional sense obscures the legal mechanisms that made its power possible. The estate also shows why institutional categories should be kept precise even when popular memory uses royal language. Titles, ceremony, and prestige can coexist with a different formal constitutional status. Markets connect regions without erasing hierarchy Bazaars, fairs, grain trade, cattle movement, credit, railways, roads, border posts, and urban commerce linked Mithila and Anga to wider economies. Markets were not neutral spaces of exchange. Access depended on transport, caste and community networks, capital, information, gender norms, policing, and state taxation. Colonial commercialization could widen opportunity while also deepening debt or exposing producers to distant price shocks. The history of markets therefore belongs alongside the history of land and labour. It reveals how regional economies were integrated unevenly rather than simply modernized. Migration is a structure, not an epilogue Seasonal labour, education, military service, clerical employment, marriage, trade, professional migration, and overseas mobility repeatedly carried people beyond the region. Remittances and return migration altered households, consumption, landholding, language use, and political expectation. Migration also built diasporic readerships and cultural associations that later became important to print and digital Maithili worlds. A regional history that studies only those who remained within a boundary will therefore miss one of the principal mechanisms through which the region was reproduced. Rivers are historical institutions as well as natural systems The Kosi, Gandak, Bagmati, Kamla, Ganga, and other rivers shape settlement, agriculture, transport, pilgrimage, erosion, sediment, and disaster. Their channels do not respect administrative boundaries. Governments respond with embankments, barrages, drainage works, bridges, forecasts, relief systems, and treaties, thereby turning hydrology into institutional history. River history makes visible a central theme of the book: nature and state are not separate spheres. Infrastructure alters the distribution of water, risk, access, and responsibility. The Kosi shows why protection and risk must be studied together The Kosi chapters distinguish routine inundation, embankment breach, and river avulsion because each has different causes and consequences. Embankments can protect particular areas from frequent flooding while increasing drainage congestion, concentrating consequences when structures fail, or transferring risk across space. A high discharge is not by itself a complete explanation of disaster; maintenance, sediment, channel behaviour, settlement, warning, and institutional response matter as well. The 2008 avulsion and later flood episodes demonstrate why the history of engineering must include the people living both outside and between protective structures. Infrastructure redistributes responsibility Barrages, railways, roads, embankments, check posts, irrigation works, and digital forecasting systems are often narrated as technical objects. Historically they are also agreements about jurisdiction, maintenance, finance, expertise, and acceptable risk. India–Nepal river mechanisms make this especially clear because physical structures and catchments cross sovereignty. When infrastructure fails or produces uneven benefits, disputes over responsibility reveal the political assumptions built into design. Studying those disputes is part of infrastructure history, not an external commentary on it. Diagram 55: connected histories across scales Diagram 55. Connected histories without false unity: scale changes from site and region to river basin, state, borderland, and transregional network. Boundaries and categories change while relations can persist. Language is an institution, not a biological boundary Maithili's literary depth and contemporary recognition are major historical facts, but language history cannot be reduced to a map of immutable populations. Languages are reproduced through households, manuscripts, schools, printing, examinations, broadcasting, theatre, migration, dictionaries, digital keyboards, social media, and state recognition. These institutions change the incentives and possibilities of language use. The same principle permits a historically serious account of modern Angika identity without erasing Grierson's earlier classification of Chhika-Chhiki or Bhagalpuri within the Maithili field. Classification and self-identification belong to different evidentiary moments and can both be documented. Recognition is plural The chapters on Maithili, Angika, and Nepalese languages show that recognition has several distinct meanings: linguistic classification, census coding, literary institutions, school use, university teaching, academies, state official-language provisions, constitutional scheduling, publication infrastructure, and digital encoding. Achievement in one field does not automatically produce recognition in another. This plurality prevents misleading comparisons. Maithili's inclusion in India's Eighth Schedule, its literary institutions, its census position, its constitutional presence in Nepal, and Tirhuta's Unicode encoding are separate institutional facts. Angika claims and institutions must likewise be assessed field by field. Script history belongs to institutional history Mithilakshar or Tirhuta is not merely an ornamental emblem. Manuscript practice, scribal training, printing constraints, font development, keyboard support, Unicode encoding, and digital publication determine whether a script can circulate in modern institutions. The shift toward Devanagari in many Maithili contexts did not erase the historical script, but it changed its everyday functional domain. Digital encoding creates new possibilities while not guaranteeing widespread literacy or publishing. Technical recognition becomes historically effective only when communities, schools, publishers, archives, and software ecosystems use it. Literary canon is made through institutions A canon is not simply the list of the best texts waiting to be discovered. Manuscript survival, editing, school curricula, university syllabi, awards, translation, anthologies, criticism, publishing economics, and digital access all influence which authors become representative. Parallel literary history is valuable because it can recover works and sequences obscured by later canonical simplification. The two-Vidyapati correction is one example of this broader task. Another is the need to distinguish literary prestige from documentary certainty: a celebrated attribution must still be tested against chronology, language, genre, and manuscript transmission. Mithila painting reveals how culture becomes an economy Mithila painting moved through several material and institutional worlds: ritual and domestic surfaces, paper, state and private collecting, exhibitions, museums, awards, tourism, fair trade, commercial reproduction, and digital sale. The transition did not simply turn an ancient practice into a commodity. It changed scale, authorship, audience, subject matter, and the distribution of income and recognition. The documented twentieth-century record therefore matters. It allows the history to respect ritual depth while avoiding unsupported claims that every present market style existed unchanged in antiquity. Gender belongs at the centre of regional history Women are not a supplementary social category added after political and economic history. Gender shapes inheritance, marriage migration, agricultural work, household labour, education, artistic production, citizenship documentation, disaster vulnerability, political representation, and access to markets. The history of Mithila painting makes this particularly visible because women's creative labour became both a source of international recognition and a field of unequal mediation. A gendered reading also changes the interpretation of land, migration, and citizenship. Formal rights matter, but so do the documents, household bargaining, mobility, and institutional access required to exercise them. Caste and class intersect without becoming the same thing Caste can structure occupation, status, marriage, access to land, ritual hierarchy, and discrimination; class shapes ownership, income, employment, education, and market power. They often reinforce one another but cannot be treated as synonyms. Regional history needs both because the same caste can contain substantial economic differentiation, while similar economic positions can be experienced differently across caste locations. Modern political categories add further layers. Reservation, census classification, association politics, and identity movements may reorganize how caste is publicly named. The history must track those changes rather than naturalize the latest vocabulary. Archive silence is itself structured People who leave few written records are not people without history. Landless labourers, women, itinerant workers, lower- status artisans, disaster-displaced households, and speakers of less institutionalized languages often appear in archives through the records of officials, landlords, courts, missionaries, collectors, or reformers. Their presence is filtered by the institution that documented them. Recovering these histories requires reading against the archive while remaining cautious about inference. Court testimony, petitions, wage records, material culture, oral history, songs, visual art, and local newspapers can widen the evidentiary field, but gaps should remain acknowledged rather than filled with invented detail. The border is both real and crossed The India–Nepal frontier is legally decisive for sovereignty, citizenship, policing, taxation, elections, property regulation, and official statistics. It is also routinely crossed through kinship, pilgrimage, education, trade, labour, health care, and cultural exchange. Treating it only as a hard line misses borderland society; treating it only as a cultural fiction misses the institutions that make nationality consequential. A connected history can hold both truths. It can study transborder Mithila and Madhesh networks while respecting the distinct constitutional histories of India and Nepal. Madhesh, Tarai, and Mithila must remain overlapping, not identical The modern politics of Madhesh cannot be read backward as a timeless territorial identity. Tarai is a lowland geographical term with changing administrative meanings; Madhesh has geographical, social, and political uses; Mithila is a historical- cultural region with major centres in both India and Nepal. Their overlap is substantial, but the terms answer different historical questions. Madhesh Province makes the distinction concrete: it is a federal unit created by Nepal's contemporary constitutional order, not a legal restoration of every historical conception of Mithila or Madhesh. Citizenship converts belonging into documents and rights Modern citizenship law gives political membership formal documentary routes. In Nepal, the history from the first national citizenship law through later acts, constitutional provisions, and recent amendments shows how descent, naturalization, parental status, gender, and documentation shape access. In border regions, these rules are especially consequential because ordinary kinship and mobility can cross an international boundary. Citizenship should therefore be distinguished from cultural belonging. A person can participate deeply in a cross-border cultural region while citizenship, voting rights, and constitutional office remain state-specific. Federalism turns recognition into institutional design Nepal's federal restructuring demonstrates that recognition is not completed by naming a province. Representation depends on constituency design, electoral systems, provincial competencies, local government, language rules, commissions, budgets, staffing, and the implementation of rights. Madhesh politics has consequently moved from protest over exclusion toward continuing disputes over how institutions operate. The broader lesson is that constitutional recognition is a beginning of administrative history, not its endpoint. Plural citizenship requires more than majority rule A province associated with a historically marginalized regional identity can still contain minorities and internal hierarchies. Madhesh Province includes Maithili, Bhojpuri, Bajjika, Tharu, Urdu, Nepali, Magahi, and other linguistic communities, as well as caste, religious, class, gender, and origin differences. Majority recognition cannot substitute for mechanisms that protect access across these differences. Plural citizenship is therefore institutional: translation, fair representation, documentation access, public services, anti- discrimination, local government, and accountable commissions matter alongside symbolic status. Modern identity claims are historical evidence, not shortcuts to antiquity Language movements, regional flags, statues, commemorative maps, festivals, museums, social-media campaigns, and political speeches are valuable sources for the history of modern identity. They show what communities choose to remember and how they mobilize historical symbols. They do not, by themselves, establish that the same identity had identical boundaries or institutions in the ancient or medieval period. This distinction protects cultural history from two distortions: dismissing modern identity as invented and treating modern identity as proof of timeless continuity. Identities can be historically real precisely because they are made, transmitted, contested, and transformed. Colonial rule produced ruptures and reclassifications Company and Crown rule changed revenue institutions, law, policing, transport, education, print, statistics, and the scale of bureaucratic knowledge. Permanent Settlement reshaped landed authority; surveys and censuses generated new classifications; railways and roads reorganized markets and mobility. Yet colonial institutions worked through existing social relations and local intermediaries rather than replacing the region with an entirely new society. The result was neither pure continuity nor total rupture. Colonial history is most intelligible as the transformation of older relations through new legal, fiscal, technological, and classificatory systems. Independence did not end institutional history The end of colonial rule in India and the political transformations of Nepal opened new constitutional orders, but inherited districts, land records, infrastructures, bureaucratic practices, and social inequalities did not disappear. Zamindari abolition, linguistic recognition, electoral democracy, academies, education policy, flood management, and federalism created new institutions whose effects unfolded over decades. Contemporary history therefore belongs inside the same analytical frame as earlier periods. It asks how new legal forms interact with inherited structures and new social movements. Methodological nationalism is insufficient for a borderland region Modern archives and statistics are organized largely by states, which can make national boundaries appear to be the natural boundaries of every historical question. For Mithila and the Nepal plains, that assumption fails. Language, pilgrimage, marriage, rivers, trade, and migration cross the India–Nepal frontier, while citizenship, law, elections, and administration remain nationally specific. The solution is not to abandon national history but to nest it within other scales. State archives, provincial records, local sources, river basins, linguistic fields, and transborder networks illuminate different dimensions of the same historical landscape. Scale should change with the question A village-level land dispute, a regional language movement, an interstate river treaty, and a transregional literary tradition cannot be explained at the same scale. Historical analysis becomes distorted when the unit of explanation is fixed in advance. The appropriate scale may be household, settlement, district, estate, region, river basin, borderland, state, or literary network. Changing scale is not inconsistency. It is a method for matching explanation to evidence. The important discipline is to state when the scale changes and not to convert findings at one scale into automatic truths at another. What connects the histories is movement Across the longue duree, the strongest common theme is movement: rivers shift; capitals move; manuscripts travel; teachers and students circulate; armies and officials enter; pilgrims cross; markets expand; labour migrates; women move through marriage; languages change domains; paintings move from walls to paper and screens; political categories move from protest into law. Region is repeatedly remade through these movements. Movement does not eliminate place. It explains why particular places become nodes of unusual historical density: Vaishali, Janakpur, Simraongarh, Darbhanga, Bhagalpur, Champanagar, Birgunj, and many smaller sites matter because wider networks pass through them. What connects the histories is also classification States, scholars, religious communities, landlords, census officials, linguistic surveyors, courts, academies, political movements, and digital standards bodies all classify. They name territories, languages, castes, citizenship categories, heritage forms, protected areas, and administrative units. These classifications can allocate resources and rights, shape identity, and influence what later historians encounter as evidence. Classification should therefore be studied as an event. The question is not only 'What category is correct?' but also 'Who used this category, when, for what purpose, with what institutional effect?' False unity usually appears through anachronism The most persistent historical errors in regional writing arise when a later category is projected backward: a modern political boundary becomes an ancient kingdom; a contemporary caste label becomes a medieval sub-caste; a current linguistic identity becomes a prehistoric ethnicity; a constitutional province becomes a timeless cultural territory; or two authors with the same name become one figure despite incompatible chronologies. Anachronism control is therefore not a minor editorial preference. It is one of the central methods by which connected history preserves both continuity and change. False fragmentation is the opposite danger Avoiding false unity should not produce an equally misleading history of isolated boxes. India and Nepal cannot be studied as though the border ended kinship and commerce; Anga cannot be detached from Ganga networks; Mithila's literary history cannot be understood without Sanskrit, Avahatta, Persianate, Hindi, Bengali, Nepali, and wider print worlds; flood history cannot stop at district boundaries. The alternative to false unity is connected differentiation: relations are traced precisely because the connected elements are not assumed to be the same. Historical explanation must separate cause from coincidence Regional histories often contain events close together in time: political change and literary production, floods and migration, recognition and institutional growth, transport and market expansion. Temporal proximity alone does not establish causation. A causal argument requires a mechanism supported by evidence: a law changes eligibility, a road changes travel cost, an embankment changes water distribution, a school policy changes language access, or a market intervention changes artistic production. Where mechanisms remain uncertain, the narrative should describe association without upgrading it to cause. Numbers require definitions Population counts, flood discharge, acreage, revenue, election shares, numbers of speakers, publication totals, and institutional budgets all appear objective because they are numerical. Their meaning depends on definitions, measurement procedures, territorial units, and dates. A 'protected area' is not the same as an area without flood risk; a census mother-tongue count is not a count of all languages a person can use; an electoral inclusion category is not a census of regional identity. Quantitative evidence becomes stronger when the denominator and classification rule are stated. Visual evidence requires the same discipline as text Maps, paintings, photographs, diagrams, architectural images, and scanned documents can make an argument appear self- evident. They too have provenance, scale, framing, and selection. A colonial map may encode administrative priorities; a museum photograph may detach an artwork from ritual context; a diagram may simplify a process to make relationships visible. The figures in this book are therefore explanatory aids, not substitutes for evidence. Their captions identify what is schematic, selected, or reconstructed so that visual clarity does not become false precision. Digital history expands access and creates new filters Digitized archives, Unicode text, searchable newspapers, online museum catalogues, census portals, GIS, satellite imagery, and digital publication have transformed the study of Mithila, Anga, and Nepal. Sources once confined to distant repositories can now be compared rapidly, and endangered scripts or small literary traditions can circulate globally. Digital availability is nevertheless selective. What is scanned, indexed, encoded, OCR-readable, or linked through search engines reflects institutional priorities and technical constraints. Digital abundance can reproduce archival silence unless historians actively seek what remains offline or poorly indexed. Public history should disclose uncertainty rather than hide it Regional history has audiences beyond universities: schools, museums, cultural organizations, language activists, descendants, local researchers, diaspora readers, and public institutions. These audiences deserve narratives that are accessible without pretending that every question is settled. Clear labels for reconstruction, tradition, debate, and secure evidence strengthen public history because they make the grounds of disagreement visible. The same standard helps cultural heritage. A site, language, art form, or memory does not require an exaggerated antiquity to be historically valuable. The archive of the future is being created now Websites, digital magazines, social media, online video, electronic government records, GIS dashboards, and born-digital literature are already primary sources for twenty-first-century history. Their fragility differs from paper archives: links decay, platforms close, interfaces change, and files may survive without metadata. Preservation decisions made now will determine what future historians can reconstruct about language movements, flood response, cultural production, and regional politics. Digital preservation therefore belongs to historical infrastructure. Stable repositories, metadata, open formats, and documented provenance are as important to future scholarship as the survival of manuscripts and printed books has been to earlier periods. Environmental change will reorder historical questions Flood regimes, heat, groundwater stress, sediment, urban growth, land-use change, and infrastructure will continue to alter the material conditions of the region. Historical knowledge cannot predict every future event, but it can show how vulnerability is produced: by settlement patterns, property, maintenance, social inequality, administrative coordination, and past engineering choices as well as by physical hazards. Environmental history is therefore not an appendix to political history. It is one of the places where long-term landscape processes and short-term governmental decisions meet most visibly. A research agenda should deepen evidence, not merely enlarge narrative The next advances in the field will come from denser evidence: stratified archaeological publication, epigraphic rechecking, manuscript stemmata, estate and court archives, village-level land records, oral histories with documented provenance, historical maps, women’s records, labour histories, multilingual newspapers, Nepalese provincial archives, and environmental data tied to precise locations. Expansion is valuable only when it increases verifiability or introduces voices and scales previously absent. Equally important is negative evidence: identifying where archives are missing, classifications change, or claims cannot yet be decided. A mature regional history includes an account of what remains unknown. Comparative history can test regional exceptionalism Mithila, Vajji, Anga, and Madhesh have distinctive histories, but many of their processes can be compared with other South Asian regions: corporate polities, Sanskrit cosmopolitanism and vernacularization, zamindari estates, language movements, river engineering, borderland citizenship, folk-art commercialization, and federal inclusion. Comparison helps identify what is genuinely specific and what belongs to wider institutional patterns. The purpose of comparison is not to dissolve the region into a national average. It is to sharpen explanation by showing which mechanisms recur elsewhere and which combinations are locally distinctive. The ethics of regional history require resisting prestige inflation Communities understandably seek recognition, antiquity, famous ancestors, and cultural continuity. Historians should take these aspirations seriously as social facts without adjusting chronology or evidence to satisfy them. Inflated claims may produce short-term prestige, but they weaken the credibility of the very traditions they are meant to honour. A stronger form of respect is evidentiary: document what is secure, state what is debated, preserve local memory, and allow cultural significance to stand without forcing every practice into the oldest possible date. The ethics of regional history also require resisting erasure Scepticism can become another distortion if it is applied only to marginalized traditions while official archives are accepted uncritically. State records, colonial ethnographies, elite genealogies, and scholarly classifications also carry interests and exclusions. Source criticism must be symmetrical: every claim, prestigious or subaltern, official or oral, is evaluated according to provenance and fit. This balance makes it possible to correct exaggeration without reproducing archival hierarchy. Diagram 55 and Diagram 54 together define the method The two concluding diagrams summarize complementary disciplines. The first changes scale: site, region, river basin, state, borderland, and transregional network are connected without being collapsed. The second controls claims: source, provenance, fit, cross-check, confidence, and narrative form a sequence that limits what can responsibly be said. Together they explain the book's title at its broadest level. The history is parallel because several trajectories remain distinct; it is connected because those trajectories repeatedly intersect. What the book ultimately argues The deepest continuity in the history of Mithila, Vajji, and Anga is not a single state or an unchanged population. It is the recurring production of regional relationships through landscape, institutions, language, texts, land, markets, sacred geography, art, migration, and political negotiation. Some relationships persist for centuries; others appear briefly and disappear. Their combination changes with every period. This conclusion permits both attachment and criticism. It recognizes the historical reality of regional cultures while refusing to make culture a substitute for evidence. It recognizes borders while tracing what crosses them. It recognizes modern identities while dating their institutional forms. It recognizes memory while distinguishing it from contemporary documentation. Connected histories without false unity A history of this region is most accurate when it accepts plurality as structure rather than as a problem to be solved. Mithila can be literary, ritual, linguistic, political, and diasporic at different scales. Vajji can illuminate ancient corporate politics without becoming a modern constitutional ancestor. Anga can remain a distinct eastern historical formation while participating in wider Ganga networks. Madhesh can be a powerful modern political space without being projected unchanged into antiquity. The result is not fragmentation. It is a more demanding unity of inquiry: one that connects evidence across periods while refusing to erase the differences that make historical change visible. Chapter-specific bibliography Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of Nationalism. Revised ed. Verso, 1991. Bloch, Marc. The Historian's Craft. Translated by Peter Putnam. Manchester University Press, 1954. Carr, E. H. What Is History? Macmillan, 1961. Chakrabarty, Dipesh. Provincializing Europe: Postcolonial Thought and Historical Difference. Princeton University Press, 2000. Ginzburg, Carlo. Clues, Myths, and the Historical Method. Translated by John and Anne C. Tedeschi. Johns Hopkins University Press, 1989. Guha, Ranajit. Dominance without Hegemony: History and Power in Colonial India. Harvard University Press, 1997. Scott, James C. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. Yale University Press, 1998. Stoler, Ann Laura. Along the Archival Grain: Epistemic Anxieties and Colonial Common Sense. Princeton University Press, 2009. Trouillot, Michel-Rolph. Silencing the Past: Power and the Production of History. Beacon Press, 1995. 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. Thakur, Gajendra. A Parallel History of Maithili Literature. Videha - First Maithili Fortnightly eJournal, ISSN 2229-547X. Grierson, George A. Linguistic Survey of India, vol. V, part II: Indo-Aryan Family, Eastern Group. Office of the Superintendent of Government Printing, India, 1903. Yadav, Ramawatar. A Reference Grammar of Maithili. Mouton de Gruyter, 1996. Yang, Anand A. Bazaar India: Markets, Society, and the Colonial State in Bihar. University of California Press, 1998. Hill, Christopher V. River of Sorrow: Environment and Social Control in Riparian North India, 1770-1994. Association for Asian Studies, 1997. Regmi, Mahesh C. Landownership in Nepal. University of California Press, 1976. Gaige, Frederick H. Regionalism and National Unity in Nepal. University of California Press, 1975. Whelpton, John. A History of Nepal. Cambridge University Press, 2005. Hachhethu, Krishna. Nation-Building and Federalism in Nepal: Contentions on Framework. Oxford University Press, 2023. Richman, Paula, and David L. Szanton, eds. India's Mithila Painting. University of Washington Press, 2025. Heinz, Carolyn Brown. 'Documenting the Image in Mithila Art.' Visual Anthropology Review 22, no. 2 (2006): 5-33. Sinha, Rajiv. 'The Great Avulsion of Kosi on 18 August 2008.' Current Science 97, no. 3 (2009): 429-433. Government of India and Government of Nepal. Agreement on the Kosi Project, 25 April 1954, and Revised Agreement, 19 December 1966. Constitution of Nepal. 2015, as amended. Nepal Law Commission. Office of the Registrar General and Census Commissioner, India. Census of India 2011, language tables. National Statistics Office, Nepal. National Population and Housing Census 2021 and Language in Nepal thematic report. Government of Nepal.