Full chapter text
horizon
Anga as a historical problem
Anga is one of the clearest examples of why political
geography must be reconstructed from several different kinds of
evidence. Buddhist lists place Anga among the major polities of
northern India, usually beside Magadha; Jain texts also preserve
Anga within an eastern political and religious horizon. Later
Brahmanical and epic traditions give the region additional
genealogies and heroes, while archaeology at Champa
documents a long-lived fortified settlement whose material
sequence cannot be equated automatically with any one literary
story. The historical task is therefore not to choose one archive
and make all others conform to it. It is to identify what each
source can securely establish and where the categories of
kingdom, city, sacred place, and remembered homeland diverge
(Singh 2008; Thapar 2002).
Mahajanapada lists and political scale
In the Anguttara Nikaya, Anga appears in the conventional
list of sixteen major countries or mahajanapadas. The list is
important because it places Anga in a remembered system of
large-scale political units, but the form is schematic rather than
cartographic. The names are not accompanied by surveyed
borders, population totals, or a constitutional description.
Different Buddhist and Jain texts transmit partly different lists,
which shows that the number and sequence served
classificatory as well as geographical purposes. Anga's
appearance is therefore strong evidence for political
prominence in the eastern Gangetic world, not a warrant for
drawing a fixed frontier around modern districts (Singh 2008;
Thapar 2002).
Early Brahmanical references and the danger of ethnic reading
Anga is also named in early Brahmanical textual traditions,
including the Atharvavedic corpus, where eastern peoples can
be represented through the cultural judgments of authors
located elsewhere. Such references are evidence for the
existence of names and for perceptions of distance or difference;
they should not be turned into timeless ethnic descriptions. A
stigmatizing or exoticizing passage tells us about the
classificatory world of the text as well as about the people
named. The emergence of Anga in several textual traditions is
historically significant, but linguistic labels, ritual status,
political membership, and archaeological culture must remain
separate unless independent evidence connects them
(Bronkhorst 2007; Singh 2008).
Epic Anga and Karna: tradition rather than regnal history
The Mahabharata's powerful association of Karna with Anga
has had an immense afterlife in regional memory. It is
indispensable for the history of literature, performance, place-
making, and modern Anga identity. It is not, however, a securely
dated administrative record of an early kingdom. Epic
composition and redaction extended over centuries, and the
political geography of a narrative can preserve older memories,
contemporary realities, later editorial knowledge, or
combinations of all three. The responsible formulation is
therefore that epic tradition makes Anga central to Karna's
kingship narrative; archaeology at Champa or Bhagalpur cannot
be assigned to Karna without independent evidence (Bengal
Government 1911; Singh 2008).
Champa as the capital in Buddhist textual geography
Buddhist sources repeatedly connect the country of the Angas
with the city of Champa. In the Sonadanda Sutta, the Buddha is
represented as travelling among the Angas and arriving at
Champa, where he stays beside Gaggara's lotus pond. This is
unusually useful political geography because country and city
occur in the same narrative frame. Even so, the text does not
give the urban perimeter, coordinates, or a district map. The
association should be combined with archaeology and later
itinerary evidence rather than treated as a self-sufficient
location proof. The modern identification near Bhagalpur
becomes persuasive through cumulative topography, surviving
place names, excavated fortification and occupation, and the
long history of scholarly comparison (Walshe 1995; Chakrabarti
2001).
Sonadanda's grant and Bimbisara's authority
The Sonadanda Sutta provides a firmer anchor for Magadhan
power than many later conquest stories. It describes the
brahmin Sonadanda as residing at Champa in a prosperous
holding granted by King Seniya Bimbisara of Magadha as a royal
gift with royal authority. Whatever the precise institutional
translation of the Pali terms, the narrative presupposes that the
Magadhan king could make a prestigious landed grant at
Champa. That does not supply the date or military sequence by
which Anga came under his authority, but it is direct textual
evidence for remembered Magadhan control within the city's
political economy. It also links state power to landed
endowment, brahmin status, grain and resources rather than
reducing expansion to battlefield conquest alone (Walshe 1995).
Bimbisara's conquest: what can and cannot be said
Buddhist and later historical traditions remember Bimbisara
as the Magadhan ruler who subdued Anga. Modern histories
commonly describe a defeat of an Anga king, often named
Brahmadatta, followed by annexation. The relative political
development - Anga's incorporation into the Magadhan sphere
before Ajatasattu's later expansion - is plausible and consistent
with the Sonadanda Sutta's representation of Bimbisara
granting land at Champa. The exact battle narrative, regnal
dates, and sequence are less secure because early Magadhan
chronology is reconstructed from texts transmitted long after
the events they purport to describe. This chapter therefore
distinguishes the well-supported transformation in political
control from conventional precision about a single campaign
(Thapar 2002; Singh 2008).
A city ranked among great cities
The Mahaparinibbana Sutta gives Champa an important place
in remembered urban hierarchy. When Ananda asks that the
Buddha not die at small Kusinara, he names Champa together
with Rajagaha, Savatthi, Saketa, Kosambi, and Varanasi as major
cities where wealthy nobles, brahmins, and householders could
honour the Buddha's remains. The passage is not an urban
census, and its rhetorical purpose is to contrast great centres
with an apparently minor town. Yet it is valuable evidence that
the text's compilers imagined Champa among cities
characterized by substantial elite and household wealth. That
literary urban rank can be compared with, but not mechanically
dated by, the archaeological evidence for fortification, wells,
coins, craft goods, and dense occupation (Walshe 1995;
Chakrabarti 2009).
Gaggara's lotus pond and urban ecology
The repeated association of Champa with Gaggara's lotus
pond reminds us that early cities were ecological landscapes, not
only walls and streets. Ponds, wells, river channels, marshes,
gardens, and seasonal water regimes affected residence, ritual,
transport, waste disposal, and food supply. The literary pond
cannot be identified with a modern water body merely because
of proximity or name resemblance, but its narrative role is
historically useful: audiences could imagine a major city whose
religious gatherings took place beside managed or recognized
water. Excavated ring-wells, brick wells, and drains add an
independent material dimension to that urban-water history,
though they belong to particular archaeological phases rather
than automatically to the lifetime of the Buddha (Indian
Archaeology 1970-71; 1982-83).
Jain Campa and Mahavira's itinerary
Jain tradition gives Campa a second major sacred geography.
The Kalpa Sutra, in Hermann Jacobi's translation, lists rainy
seasons spent by Mahavira in Campa and Prishticampa,
alongside longer stays at Vaishali-Vanijagrama, Rajagriha-
Nalanda, and Mithila. The list is a sectarian biographical
tradition rather than a travel diary, yet it establishes the
importance of Campa within the remembered circuit of Jain
teaching and monastic residence. Its value increases when
compared with Buddhist evidence that also treats the city as a
major eastern centre. Convergence between independent
religious traditions supports urban significance; it does not
mean that both transmit identical chronology, topography, or
institutional details (Jacobi 1884; Dundas 2002).
Merchants in narrative and the problem of economic inference
Later Buddhist narrative traditions repeatedly use Champa as
a point of departure for merchants. The Mahajanaka Jataka, for
example, sends its protagonist on a trading voyage from
Champa toward Suvannabhumi before shipwreck. Such stories
are valuable evidence for the cultural imagination of commerce,
river-to-sea connectivity, risk, wealth, and distant horizons.
They are not shipping registers and cannot prove that every
narrated route operated in the same form at the story's dramatic
date. The appropriate use is comparative: literary traditions of
merchants and long-distance movement are set beside the city's
Ganga location, excavated coinage and craft goods, and the
wider archaeology of eastern exchange. Convergence supports
commercial importance while leaving the scale and destinations
of actual trade open to further evidence (Cowell and Rouse
1907).
Riverine commerce before a seaport narrative
Champa's position in the Ganga system made river transport
a central possibility, but the city should not automatically be
called a maritime port. Goods and travellers could move down
the Ganga toward deltaic and coastal centres without ocean-
going vessels sailing directly from the urban riverbank. The
distinction matters because later literary narratives may
compress several stages of transport into one journey.
Riverboats, ferries, transshipment points, caravan routes, and
downstream ports could form one commercial chain. Historical
geography should therefore reconstruct a corridor from inland
production and markets through the Ganga to eastern outlets,
rather than imagine a single unchanging sea harbour at Champa
(Chakrabarti 2001; Singh 2008).
Diagram 14. Anga and Champa: four evidence streams. Buddhist and Jain
textual geography, archaeology, and later survey or memory are treated as
distinct archives whose convergences strengthen an argument but whose
differences must remain visible.
Credit: Editorial diagram prepared for this book, 2026. Based
on Digha Nikaya 4 and 16, Kalpa Sutra, published Champa
excavation reports, and modern archaeological/historical
synthesis. Rights: © Gajendra Thakur, 2026.
Source links: https://www.palicanon.org/index.php/sutta-
pitaka/digha-nikaya/dn4-so-ada-a-sutta-about-so-ada-a ;
https://sacred-texts.com/jai/sbe22/index.htm ;
https://nmma.nic.in/nmma/IAReview.do
Identifying ancient Champa near Bhagalpur
The identification of ancient Champa with the Champanagar-
Nathnagar area near modern Bhagalpur is cumulative rather
than dependent on name resemblance alone. Buddhist and Jain
texts place Champa in eastern Gangetic political and sacred
geography; later pilgrims and antiquarians attempted to relate
those descriptions to the modern landscape; the survival of
Champa-related place names kept the association visible; and
excavation exposed a large, long-lived settlement with
fortification and early-historic material. Colonial gazetteers
themselves recorded uncertainty about exact topographic
equations, which is a useful warning. The secure historical claim
is the identification of the broader archaeological zone; the
exact position of every ancient neighbourhood, gate, pond, or
palace remains a more difficult question (Bengal Government
1911; Chakrabarti 2001).
Excavation history: a city recovered by campaigns
Modern archaeological knowledge of Champa was built
through repeated field seasons rather than a single definitive
excavation. Patna University's Department of Ancient Indian
History and Archaeology investigated the site across several
campaigns beginning in the late 1960s and continuing in later
decades, with results summarized in the Archaeological Survey
of India's annual Indian Archaeology - A Review. Different
trenches targeted ramparts, habitation, areas outside the
defence, and the relationship between structural phases.
Because objectives and trench locations changed, the published
sequence must be read trench by trench before being
generalized to the whole settlement. A composite history is
possible, but it should always state when a phase is local to one
cutting rather than city-wide (Indian Archaeology 1970-71; 1971-
72; 1974-75; 1982-83).
Pre-NBPW occupation below the urban horizon
Later campaigns pushed the sequence earlier. The 1974-75
review reports a deposit beneath the first mud-rampart
containing coarse black-and-red ware forms and explicitly
describes this as the earliest occupation at the site prior to the
NBPW phase. The 1982-83 review, undertaken partly to check
earlier results concerning the pre-NBPW phase, likewise records
about a metre of coarse black-and-red, black, and red ware over
natural soil. These reports do not by themselves produce an
absolute date for the earliest settlement. Their importance is
stratigraphic: Champa's material history begins before the
better-known NBPW-bearing urban horizon (Indian
Archaeology 1974-75; 1982-83).
NBPW is a horizon, not a city label
Northern Black Polished Ware is prominent in Champa's
early-historic deposits, but it cannot be treated as a synonym for
the Anga kingdom or for urbanism itself. Excavation reports
record NBPW together with black, black-and-red, red, and grey
wares and associate later portions of the horizon with brick
architecture, wells, iron tools, coins, terracottas, and other
objects. These associations make the NBPW horizon
archaeologically rich; they do not establish the political identity
of every pot user. Ceramic chronology and political history must
be correlated rather than collapsed (Chakrabarti 2009; Indian
Archaeology 1971-72; 1982-83).
Fortification: more than one defensive episode
Champa's mud ramparts are among its most important
archaeological features. Excavators distinguished multiple
defensive phases, including massive earthen ramparts,
brickwork, and traces of wooden posts in different contexts.
Later ring-wells and structures cut or overlie former defensive
ground, demonstrating changes of use. The sequence
demonstrates repeated investment in a bounded urban space. It
does not identify which named Anga ruler built a particular
wall, and the chronological range of construction must be
derived from associated layers rather than epic memory (Indian
Archaeology 1970-71; 1971-72).
Wells, drains, and household infrastructure
Water and waste infrastructure are among the strongest
material signs of dense occupation. Excavation documented
ring-wells, a deep brick well, drains, mud walls, ovens, and
paved or rammed surfaces in different phases. Such
installations can serve domestic, craft, ritual, or mixed purposes;
archaeology does not automatically tell us who owned them.
Their repetition nonetheless shows investment in controlled
water access, drainage, heat-based work, and everyday
maintenance within a long-lived settlement and supports the
interpretation of Champa as a substantial urban centre rather
than a purely literary city (Indian Archaeology 1970-71; 1974-75;
1982-83).
Coins, beads, ivory, and craft specialization
Cast copper coins and punch-marked coins, beads of semi-
precious stone, glass and terracotta, ivory objects, terracotta
figurines, bone tools, moulds, copper objects, and iron
implements occur in Champa's excavated sequences. The
assemblage implies access to diverse materials and specialized
skills in carving, moulding, metalworking, ornament
production, and exchange. It would be unsafe to infer a formally
organized guild from these finds alone, because the
archaeological record does not name institutional
arrangements. But the range of materials and technical
knowledge is compatible with a differentiated urban craft
economy (Indian Archaeology 1971-72; 1982-83).
Conflagration as an event, not a dynasty marker
One excavated phase ended with evidence of large-scale
burning. Archaeological burning may result from accidental
fire, conflict, craft processes, clearance, or deliberate
destruction. Without an inscription, weapon horizon, or
independent chronological link, a burnt layer cannot be
assigned to Bimbisara's conquest or another named political
event. This caution is particularly important at famous sites,
where narrative history invites dramatic correlation. At
Champa, the observation of burning is secure; its cause remains
an interpretive question (Indian Archaeology 1971-72).
Kushan, Gupta, and later Champa
Champa remained occupied and repeatedly reorganized after
the early-historic horizon. Excavation reports identify Kushan-
period construction or artefacts, Gupta habitation, sealings and
figurines, later post-Gupta material, medieval glazed sherds, and
later intrusions. This long sequence is historically important
because it demonstrates that the location's advantages did not
disappear with Anga's political absorption into Magadha. It also
makes attribution harder: a mound visible today may contain
layers separated by many centuries. The regional name, urban
focus, sacred associations, and settlement form each had their
own histories (Indian Archaeology 1970-71; 1982-83).
Diagram 15. Champa excavation sequence: a stratigraphic caution. Composite
summary of published excavation reviews; not every trench contains every
phase, and the diagram does not convert ceramic or structural phases into
dynastic labels without independent evidence.
Credit: Editorial diagram prepared for this book, 2026, from
Archaeological Survey of India, Indian Archaeology - A Review
reports for 1970-71, 1971-72, 1974-75, and 1982-83. Rights: ©
Gajendra Thakur, 2026.
Source links: https://nmma.nic.in/nmma/IAReview.do ;
https://zenodo.org/records/3367072
Agrarian hinterland and the city
No city of Champa's scale could have lived from commerce
alone. Grain, fodder, fuel, timber, fibres, animals, fish, clay, and
labour had to move from surrounding landscapes into the urban
centre. The Sonadanda Sutta's description of a prosperous grant
rich in grass, timber, water, and grain is literary, but it
highlights precisely the resources through which landed
authority connected countryside and city. Archaeological
storage, food remains, ovens, tools, and transport routes can
help reconstruct that relationship, although published annual
reports do not yet provide a complete quantitative agrarian
economy. The proper model is therefore an urban-hinterland
system, not a self-contained trading town (Walshe 1995;
Chakrabarti 2001).
State revenue, landed grants, and intermediaries
Bimbisara's grant to Sonadanda is also evidence for political
mediation. A ruler governing an incorporated region need not
administer every village directly; land, revenue rights, honours,
and local authority could be delegated to elite beneficiaries who
linked court and locality. The Buddhist text presents such a
relationship through a prestigious brahmin recipient. It does not
tell us how cultivators experienced the grant, which dues were
collected, or how long the arrangement lasted. Nonetheless, it
offers a historically plausible mechanism for Magadhan
integration: conquest or submission followed by selective
redistribution of rights within an existing agrarian and urban
economy (Walshe 1995; Thapar 2002).
Anga and Magadha as a paired political geography
The recurrent pairing of Anga and Magadha in textual lists is
significant even after Anga's political subordination. Names of
former polities can survive as regional designations, cultural
memories, or administrative shorthand long after sovereignty
changes. This helps explain why Anga remains visible in
Buddhist and Jain geography even when Bimbisara is
represented as exercising authority at Champa. Political
incorporation need not erase regional identity. The same
principle appears throughout this book: a state boundary, a
cultural region, and a place-name tradition can have different
durations (Singh 2008).
Karna memory and archaeological restraint
Bhagalpur and Champanagar preserve strong traditions
connecting the landscape with Karna. Temples, place names,
festivals, and popular histories make that memory an important
part of the region's cultural past. Archaeological excavation,
however, has not produced an inscription securely identifying a
fortification or palace as Karna's. The correct historical
procedure is dual: document the epic and local tradition as
evidence for reception and identity, while refusing to use the
modern strength of the tradition as proof of an early regnal
event. This is the same evidentiary discipline applied elsewhere
in this book to figures whose cultural afterlives are historically
real even when event chronology remains unverified.
Evidentiary conclusions
A defensible history of Anga begins with a major eastern
polity remembered across Buddhist, Jain, and Brahmanical
traditions and a capital, Champa, whose broad identification
near modern Bhagalpur is supported by historical geography
and archaeology. Buddhist evidence places Bimbisara's
authority within Champa through a royal grant and remembers
the city among major urban centres. Jain tradition places Campa
within Mahavira's sacred itinerary. Excavation demonstrates
pre-NBPW occupation, NBPW-bearing early-historic phases,
substantial fortification, wells, drains, coins, crafts, and long
later reuse. What remains less secure includes exact early
borders, precise regnal dates, a complete battle narrative for the
Magadhan conquest, and the equation of epic Karna traditions
with excavated structures. Those distinctions allow Anga to
emerge as historically important without being made artificially
certain.
Chapter-specific bibliography
Walshe, Maurice, trans. The Long Discourses of the Buddha: A Translation
of the Digha Nikaya. Wisdom Publications, 1995.
Jacobi, Hermann, trans. Jaina Sutras, Part I: Acaranga Sutra and Kalpa
Sutra. Sacred Books of the East, vol. 22. Clarendon Press, 1884.
Cowell, Edward B., and W. H. D. Rouse, trans. The Jataka, or Stories of the
Buddha's Former Births. Vol. VI. Cambridge University Press, 1907.
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.
Bronkhorst, Johannes. Greater Magadha: Studies in the Culture of Early
India. Brill, 2007.
Dundas, Paul. The Jains. 2nd ed. Routledge, 2002.
Chakrabarti, Dilip K. Archaeological Geography of the Ganga Plain: The
Lower and the Middle Ganga. Permanent Black, 2001.
Chakrabarti, Dilip K. India: An Archaeological History: Palaeolithic
Beginnings to Early Historic Foundations. 2nd ed. Oxford University
Press, 2009. https://doi.org/10.1093/acprof:oso/9780198064121.001.0001.
Archaeological Survey of India. Indian Archaeology 1970-71 - A Review.
Government of India.
Archaeological Survey of India. Indian Archaeology 1971-72 - A Review.
Government of India.
Archaeological Survey of India. Indian Archaeology 1974-75 - A Review.
Government of India.
Archaeological Survey of India. Indian Archaeology 1982-83 - A Review.
Government of India. https://doi.org/10.5281/zenodo.3367072.
Bengal Government. Bengal District Gazetteers: Bhagalpur. Bengal
Secretariat Book Depot, 1911.