Full chapter text
horizons
A long transition
Across the middle Gangetic plain, the later second and first
millennia BCE saw changes in settlement density, ceramics,
metal use, and cultivation. These did not occur everywhere at
once. Iron objects are part of the sequence, but older models
that made iron technology a single cause of forest clearance,
surplus, cities, and states now require qualification. Political
centralization depended on labour, exchange, agrarian
relations, ritual authority, and ecological opportunity as well as
tools (Singh 2008; Thapar 2002).
Ceramic labels are chronological tools
Black-and-red ware, northern black polished ware, and other
ceramic categories are archaeological classifications. They
cannot by themselves identify Vedic, Vajji, or Anga populations.
Northern black polished ware is associated with the early-
historic urban horizon in many Gangetic sites, but local
sequences and dates differ. The proper question is how
settlements participated in larger circuits, not which text-name
can be attached to every sherd.
Protohistory and restraint
The threshold between prehistory and history is uneven
because writing, recoverable inscriptions, and datable textual
references appear gradually. Literary works preserved orally
may remember older settings while taking form over long
periods. This volume therefore stops short of constructing
dynastic lists for times unsupported by contemporary records.
Iron within a larger technological system
Iron mattered because it expanded the range of durable
edges, points, fittings, and implements available to communities,
but its effects depended on access to ore, fuel, specialist
knowledge, exchange, repair, and labour. The presence of an
iron object in a layer does not show that iron tools were
common in every household, nor that they immediately
transformed cultivation. Bronze, copper, stone, bone, wood, and
clay continued to be used alongside iron. Technological change
is therefore best reconstructed as a changing assemblage rather
than a sudden replacement of one age by another.
Settlement growth and political concentration
Larger settlements and denser site distributions can indicate
demographic growth, intensified production, exchange, and
administrative concentration. They do not, by themselves,
reveal the constitutional form of a polity. Fortifications may
defend inhabitants, control access, display collective labour, or
mark authority; they require contextual study rather than a
single explanation. In the first millennium BCE, the eastern
Gangetic plain saw the emergence of towns and states within a
landscape that still contained villages, mobile groups, forests,
wetlands, and cultivated clearings. Urbanization was a regional
process linking different kinds of settlement, not the
disappearance of rural life (Singh 2008; Thapar 2002).
Exchange before coinage
Long-distance movement did not begin with minted money.
Stone, metal, ornaments, salt, forest products, animals, textiles,
and food could circulate through barter, reciprocal obligation,
itinerant specialists, tribute, gift, and market exchange.
Archaeological similarities may reflect such connections without
implying political unity. When punch-marked coins and other
monetary evidence appear in early-historic contexts, they add
another medium of valuation and payment; they do not replace
all earlier forms at once. The river corridors of the Ganga
system made the eastern regions participants in networks larger
than any single polity.
Texts and material sequences
Vedic and later texts cannot simply be laid over a pottery
map. A textual name belongs to a compositional and transmitted
discourse, while a ceramic type belongs to an archaeological
classification. The two records may illuminate the same broad
historical transformation, but correlation requires independent
arguments about date and geography. Witzel's reconstruction of
early Sanskritization and Bronkhorst's model of Greater
Magadha offer influential ways of understanding the cultural
difference and interaction between western and eastern
regions. They are interpretive models to be tested, not labels
that identify the makers of particular vessels (Witzel 1995;
Bronkhorst 2007).
The threshold of the early-historic city
Northern Black Polished Ware, increasing settlement
complexity, coinage, inscriptions, and textual references form
part of the evidentiary horizon within which early cities and
states become more visible. Their chronologies vary by site, and
the appearance of one indicator should not stand for the entire
transformation. A securely excavated sequence is more valuable
than a loose object with an impressive label. The following
volume therefore begins its political narrative only where
textual traditions and material evidence can be compared with
explicit caution about their different dates and purposes.
The “Iron Age” is an archaeological period label, not a people
The term Iron Age is convenient only when its scale is
specified. In South Asian archaeology it usually marks a horizon
in which iron becomes part of a material assemblage, but the
first appearance, frequency, uses, and social consequences of
iron differ sharply by region and site. A period label based on
one technology therefore cannot be allowed to carry an ethnic,
linguistic, or constitutional meaning. An excavated iron point,
nail, sickle, or blade may be securely associated with a layer, but
it does not identify the inhabitants as Vedic, Vajji, Videha, Anga,
or any other named community unless independent evidence
supports that identification. The same caution applies to older
expressions such as “Iron Age culture,” which can obscure the
coexistence of iron with stone, bone, copper, wood, clay, and
organic technologies (Banerjee 1965; Chakrabarti 1992; Singh
2008).
How the chronology debate changed
Older syntheses often placed the effective beginning of iron
use in the Gangetic region close to the opening centuries of the
first millennium BCE and then linked its spread to the
emergence of agriculture, towns, and states. Later discoveries
complicated that sequence. Rakesh Tewari’s published report
from the central Ganga plain and the eastern Vindhyas
described iron artefacts, furnaces, tuyeres, and slag in contexts
with radiocarbon determinations falling broadly between 1800
and 1000 BCE. The importance of this evidence lies in the
combination of artefacts and production debris rather than in a
single early object. It reopened the chronology of iron working
in northern India and made diffusionist narratives more
difficult to sustain as simple one-directional stories (Tewari
2003).
Why an early date elsewhere cannot date Mithila or Anga
A comparative date is not a regional date. Tewari’s excavated
sites are in present Uttar Pradesh, and their chronology cannot
be transferred automatically to north Bihar, the Nepal Tarai, or
the Anga-Bhagalpur corridor. The correct historical use of such
evidence is comparative: it demonstrates that iron-working
knowledge existed in parts of the wider Ganga-Vindhya world
earlier than some older schemes allowed, and it therefore
changes the range of possibilities that archaeologists must test in
Bihar and Nepal. It does not prove that every eastern Gangetic
settlement was making iron at the same time. Each locality still
requires its own stratigraphy, diagnostic assemblage, technical
debris, and dating programme (Chakrabarti 1992; Tewari 2003).
Iron use, forging, and smelting are different archaeological
propositions
The archaeological vocabulary of metallurgy should be kept
precise. A finished iron artefact establishes use or deposition.
Hammer scale, smithing slag, hearths, and related residues may
indicate secondary working or repair. Smelting slag, furnace
structures, ore, and tuyeres can support an argument for
primary production when their identification and association
are secure. Even then, production scale remains a separate
question: a few technical residues do not prove an industrial
quarter or centralized state workshop. Metallurgical analysis
can distinguish mineral phases, manufacturing histories, and
thermal processes that are invisible to typology alone. The
analytical sequence therefore moves from context to
identification to technical interpretation, not from the mere
presence of iron to a generalized “iron industry” (Chakrabarti
1992; Sahu 2006).
Diagram 4. Iron-working evidence hierarchy. The diagram distinguishes what
an artefact, slag or technical debris, production features, raw-material
evidence, and stratigraphic dating can each establish. It is an analytical
schematic, not a reconstruction of an ancient workshop.
Original editorial diagram prepared for this book, 2026. © Gajendra Thakur,
2026. Conceptual distinctions synthesized from Chakrabarti (1992), Tewari
(2003), and the historiographical debate collected by Sahu (2006).
Source: Tewari article DOI
Black-and-red ware: a firing category with a long history
Black-and-red ware (BRW) is especially vulnerable to over-
interpretation because its name sounds like a bounded
archaeological culture while describing a visible ceramic effect
produced through firing conditions. Archaeological Survey of
India syntheses documented black-and-red ware at sites in
many parts of the subcontinent and across very different
chronological contexts, including Sonpur in Bihar. Vessel forms,
fabrics, slips, decorations, associated wares, and stratigraphic
positions vary. The category is therefore useful for describing
and comparing pottery, but it cannot by itself mark one
migration, one language, or one polity. Its historical value comes
from local technological study and stratigraphic association, not
from treating all black-and-red vessels as the signature of one
population (Banerjee 1965; Roy 1983).
Painted Grey Ware is a comparative horizon, not the template
for the east
Painted Grey Ware (PGW) has been central to archaeological
discussion of the upper Ganga-Yamuna plain, especially because
excavations at sites such as Hastinapura and Atranjikhera made
it important to debates about first-millennium BCE chronology
and the arrival of iron. Vibha Tripathi’s synthesis established
the ware as a major archaeological problem in its own right. For
a history centred on Mithila, Vajji, and Anga, however, PGW is
primarily comparative evidence. Its distribution is not identical
to the middle and lower Gangetic trajectories that produced the
principal archaeological record of Bihar. Using PGW as a
universal chronological ladder would reproduce the very
western-to-eastern template that regional archaeology is meant
to test (Tripathi 1976; Roy 1983).
Northern Black Polished Ware: name, technology, and
distribution
Northern Black Polished Ware (NBPW) is a fine ceramic
category whose highly finished surface attracted archaeological
attention early in the discipline’s history. Deva and Wheeler’s
1946 discussion already assembled examples from a broad
north Indian distribution, including Patna and Rajgir in Bihar,
while later work greatly enlarged the corpus. A recent
archaeological synthesis emphasizes the ware’s limited
production, distinctive surface treatment, chronological
usefulness, and wide circulation, while also showing that its
colours and technical characteristics are more varied than the
older name suggests. The important point for regional history is
not that every NBPW sherd belongs to one political system, but
that the ware can help trace site sequences, craft choices,
exchange, and social differentiation when securely stratified
(Deva and Wheeler 1946; Kanungo et al. 2021).
A fine ware is not a universal social class marker
The fineness and lustre of NBPW have often encouraged the
label “luxury” or “elite” ware. That inference can be useful as a
hypothesis, but it should be tested against quantity, vessel form,
find context, repair, association with architecture, and
comparison with coarser wares. A small amount of technically
demanding pottery in a settlement may indicate valued
tableware, exchange, or restricted consumption, yet it does not
identify the caste or office of the user. Nor does wide
distribution prove that every vessel travelled from one
manufacturing centre. Petrography, compositional analysis,
firing experiments, and close study of fabric are required to
separate local imitation, multiple production zones, and long-
distance movement (Kanungo et al. 2021).
Bihar within the NBPW archaeological field
The NBPW horizon is directly relevant to Bihar because early
and later archaeological literature records the ware at major
sites including Rajgir, Patna, Sonepur, Chirand, and other
centres of the middle Gangetic plain. That distribution helps
define a connected archaeological field, but the sites do not all
have the same stratigraphic depth, chronology, or urban
character. A sherd from a city mound, a village, a religious
complex, or a disturbed surface context carries a different
historical weight. The strongest regional argument therefore
comes from comparing published sequences: when NBPW
appears, what preceded it, which iron objects or production
residues accompany it, how settlement form changed, and
whether coins, writing, fortification, or specialized crafts occur
in the same phase (Roy 1983; Sinha 1995; Kanungo et al. 2021).
Ceramic horizons and political geography should not be
collapsed
It is tempting to map the spread of NBPW directly onto the
expansion of Magadha because the ware is prominent in the
Ganga plain during the period in which Magadhan power
became historically consequential. That correlation may be
historically meaningful at some scales, but it is not an identity.
Pottery can move through exchange, emulation, itinerant
craftspeople, marriage networks, pilgrimage, and changing
consumption without political annexation. Conversely, a polity
can exercise authority over communities that retain diverse
local ceramic traditions. The political history of Anga, Vajji, and
Magadha must therefore be reconstructed from texts,
archaeology, coinage, inscriptions, and settlement evidence
together rather than from a ceramic distribution map alone
(Thapar 2002; Singh 2008).
From site size to settlement hierarchy
The growth of larger settlements is one of the central
archaeological changes of the first millennium BCE, but size by
itself is not equivalent to urbanism. A large mound may
represent long duration, repeated rebuilding, clustered villages,
administrative concentration, craft production, or a town; only
excavation and survey can distinguish among these possibilities.
Settlement hierarchy is stronger evidence when a region shows
sites of different scales linked by roads, rivers, craft
distributions, storage, fortification, or specialized functions.
George Erdosy’s work on early historic urbanization helped shift
attention from isolated “cities” toward settlement systems and
the processes that sustained them (Erdosy 1988).
Fortification as labour, defence, and political display
Ramparts, ditches, gates, and defensive walls are among the
most visible signs of concentrated labour in early historic
archaeology. Their interpretation is nevertheless multi-valued. A
fortification may protect inhabitants, control access, channel
traffic, mark jurisdiction, contain animals, protect stored goods,
or display authority. Construction and repeated repair reveal
the mobilization of labour and materials, but not automatically
the constitutional form of the authority that organized them. In
alluvial landscapes, the relationship between ramparts and
water management also deserves attention: elevated earthworks
can interact with drainage and flood behaviour as well as
military needs. Chronological claims should follow excavation of
construction phases, not the later fame of a city name (Erdosy
1988; Singh 2008).
Iron and forest clearance: from master cause to testable
mechanism
A major historiographical model linked iron axes and
ploughshares to forest clearance, agricultural surplus, social
stratification, and the rise of states in the middle Gangetic basin.
The model was influential because it joined technology to a large
social transformation. Its weakness appears when iron is treated
as a sufficient cause. Forests differ in density and species; land
can be cleared with fire, stone, copper, and collective labour;
floodplains include grasslands, wetlands, levees, and already
cultivated patches; and iron implements vary in availability and
quality. The appropriate question is therefore not whether iron
“caused” agriculture, but where particular iron tools altered
labour productivity, repair cycles, cultivation depth, woodland
use, or transport in demonstrable contexts (Chakrabarti 1992;
Sahu 2006).
Agrarian intensification was a social process
Cultivation expands only through people, institutions, and
access to land and water. Seed selection, seasonal calendars,
draught animals, field preparation, transplantation, harvesting,
storage, redistribution, credit, and claims on labour can change
independently of metal technology. Riverine environments add
another variable because fresh alluvium may be fertile while
erosion, waterlogging, and avulsion repeatedly reorganize
fields. An iron sickle or plough component can matter
profoundly within such a system, but the archaeological object
must be connected to plant remains, settlement, field ecology,
and the organization of labour before it becomes evidence for
an agrarian transformation. This is why the history of iron
belongs inside social and environmental history rather than in a
technological appendix (Sahu 2006; Singh 2008).
Craft specialization without imagining modern factories
Iron working, fine pottery, bead production, carpentry,
textiles, construction, and food processing all required learned
skills. Archaeological concentrations of debris, standardized
products, repeated installations, and raw-material stocks can
reveal specialization, but specialization exists on a spectrum. A
household specialist working seasonally, an itinerant smith, a
workshop attached to a large settlement, and a politically
supervised production unit are different institutions. The record
rarely permits a leap directly from technical skill to guild, caste,
or state monopoly. Production should be reconstructed from
spatial pattern, scale, chaîne opératoire, and distribution rather
than from the modern assumption that complex objects require
factory-like organization (Chakrabarti 1992; Sahu 2006).
Exchange networks before and alongside coined money
The first-millennium BCE expansion of exchange should not
be dated solely by the first coin found in a trench. Communities
moved salt, metals, stone, beads, textiles, animals, food, wood,
and crafted objects through reciprocity, gift, tribute, itinerant
trade, periodic markets, and river transport before coinage
became archaeologically visible. The Ganga and its tributaries
lowered transport costs for bulky goods and linked Anga,
Magadha, and north-bank regions to wider circuits. NBPW and
other non-local materials can illuminate such movement, but
provenance must be demonstrated rather than inferred from
visual similarity alone. Coinage added a powerful medium of
valuation and payment without abolishing older forms of
exchange (Erdosy 1988; Singh 2008).
Punch-marked coins: strong evidence, careful chronology
Punch-marked coins are important to the threshold of the
early historic period because they provide material evidence for
monetized transactions, political or mercantile symbols,
hoarding, and regional circulation. Their dating depends on
stratigraphy, hoard composition, metrology, typology, and
association with other securely dated material. A coin found in a
later disturbed deposit cannot date the entire layer below it,
while a hoard provides a terminus and a history of
accumulation rather than the mint date of every piece. Coinage
should therefore be integrated with settlement and exchange
evidence without being made the single criterion by which a
society becomes “historic.”
Writing, seals, and administrative visibility
Writing changes the surviving archive because names, titles,
donations, ownership claims, religious formulae, and
administrative acts can become materially recoverable. Yet the
appearance of inscriptions is not simultaneous with the first use
of writing, and the absence of an inscription from an excavated
layer does not prove illiteracy. Perishable writing materials can
disappear, while seals and sealings may preserve institutional
activity before long public inscriptions become common. The
transition from protohistory to history is therefore uneven:
some sites become visible through coins and seals, others
through texts copied later, and still others remain known chiefly
from archaeology. The methodological response is to keep these
evidentiary clocks separate until they can be securely correlated
(Singh 2008).
Diagram 5. The early-historic transition runs on several clocks. Iron use,
ceramic change, settlement growth, exchange, and money/writing overlap but
do not begin everywhere at one date. The diagram deliberately has no absolute
time scale.
Original editorial diagram prepared for this book, 2026. © Gajendra Thakur,
2026. Analytical model synthesized from Erdosy (1988), Chakrabarti (1992),
Singh (2008), and Kanungo et al. (2021).
Source: Kanungo et al. NBPW synthesis DOI
The “second urbanization” as a historiographical term
Historians often use “second urbanization” for the emergence
and growth of towns in the Ganga plain during the first
millennium BCE, distinguishing it from the much earlier
urbanism of the Indus tradition. The phrase is useful as
shorthand but can mislead if it suggests one synchronized event
or a single causal package. Urban centres varied in size, political
role, craft intensity, religious institutions, and relationship to
surrounding villages. Some older settlements expanded; others
were newly founded; some became fortified or administratively
important; and many rural communities continued outside
urban sites. The historical problem is therefore to reconstruct
networks of production, power, transport, and consumption, not
merely to count mounds above a size threshold (Erdosy 1988;
Sahu 2006).
Urbanism and inequality
Urban growth redistributed opportunities and burdens. Large
settlements could concentrate rulers, officials, merchants,
religious specialists, artisans, labourers, migrants, servants, and
dependants, while drawing food, fuel, building materials, and
revenue from wider hinterlands. The archaeological visibility of
substantial architecture and fine goods can bias interpretation
toward elites. Ordinary houses, temporary structures, refuse,
industrial debris, food remains, and skeletal evidence are
equally important for reconstructing work and inequality. Caste
categories known from later normative or literary texts should
not be projected directly onto an excavated quarter unless
independent evidence makes the connection plausible. Social
hierarchy is a historical question, not an automatic deduction
from urban scale (Singh 2008; Thapar 2002).
Rivers as infrastructure in the eastern Gangetic transition
The regional geography developed in Chapters 1 and 2
becomes especially important as settlement and exchange
intensify. The Ganga connected Anga and the middle Gangetic
basin to long-distance movement, while the Gandak, Kosi,
Bagmati, and other northern rivers connected the Himalayan
foreland and Tarai to the plains. River routes could move bulk
goods more efficiently than overland transport, but channels
shifted and landing points changed. Floods could destroy
settlements while also renewing soils or creating new
distributary access. The rise of towns must therefore be studied
together with river history, ferries, confluences, levees, and
seasonal navigability rather than on a map of permanent blue
lines (Erdosy 1988; Singh 2008).
Textual geography enters a material landscape already in motion
As texts begin to name eastern polities and communities
more frequently, they encounter a material landscape whose
settlement, technology, and exchange had already been
changing for centuries. Archaeology can test whether a named
region contained dense settlement, fortification, imported
materials, or specialized production; it cannot prove that the
makers of a particular pottery type used the same political
name. Conversely, texts can preserve political categories and
itineraries that artefacts cannot name. The strongest history
begins when the records can be compared without making one
subordinate to the other. A pottery map does not become a
political map by adding labels from a later text (Witzel 1995;
Bronkhorst 2007; Singh 2008).
The archive becomes denser, not complete
By the early historic centuries the historian possesses more
kinds of evidence than for the Neolithic: iron and other metals,
refined ceramics, settlement hierarchies, fortifications, coins,
seals, inscriptions, literary geographies, and eventually dated
imperial records. Greater density does not eliminate silence.
Organic materials decay, ordinary people commission fewer
durable texts, river processes erase sites, excavations sample
only small parts of large mounds, and early field methods
sometimes recorded contexts less precisely than present
standards require. Apparent certainty can therefore increase
faster than actual evidentiary control. The correct response is to
make the source base denser while keeping its asymmetries
visible.
A research programme for the full Chapter 3 edition
A publication-scale expansion should build a site-by-site
database for Bihar and the Nepal Tarai recording phase names,
calibrated and uncalibrated dates, ceramic frequencies, iron
artefacts, slag, furnaces or tuyeres, archaeobotanical samples,
faunal remains, fortification phases, coins, seals, inscriptions,
and published plans. Legacy excavation reports should be
reconciled with Indian Archaeology - A Review notices and
museum catalogues. Metallurgical claims should be separated
into smelting, smithing, and use; ceramic claims should record
fabric and context; and all maps should distinguish excavated
evidence from surface finds. The aim is not to create a smooth
line from iron to empire but to show where different processes
converged, where they did not, and where the archive is still too
thin for a secure conclusion.
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.
Witzel, Michael. 'Early Sanskritization: Origins and Development of the
Kuru State.' Electronic Journal of Vedic Studies 1, no. 4 (1995): 1-26.
Bronkhorst, Johannes. Greater Magadha: Studies in the Culture of Early
India. Brill, 2007.
Banerjee, N. R. The Iron Age in India. Munshiram Manoharlal, 1965.
Chakrabarti, Dilip K. The Early Use of Iron in India. Oxford University
Press, 1992.
Deva, Krishna, and R. E. M. Wheeler. 'Northern Polished Black Ware.'
Ancient India 1 (1946): 55-58.
Erdosy, George. Urbanisation in Early Historic India. BAR International
Series 430. B.A.R., 1988.
Kanungo, Alok Kumar, Chinmay Kulkarni, Varad Ingle, and Oishi Roy.
'Northern Black Polished Ware in Indian Archaeology: A Study of Spatial
and Chronological Distribution.' Encyclopedia of Global Archaeology.
Springer, 2021. https://doi.org/10.1007/978-3-319-51726-1_3555-1.
Roy, T. N. The Ganges Civilization: A Critical Archaeological Study of the
Painted Grey Ware and Northern Black Polished Ware Periods of the
Ganga Plains of India. Ramanand Vidya Bhawan, 1983.
Sahu, Bhairabi Prasad, ed. Iron and Social Change in Early India. Oxford
University Press, 2006.
Sinha, B. P. 'Ancient Cities of Bihar in Archaeology and Literature.' Man and
Environment 20 (1995): 7-20.
Tewari, Rakesh. 'The Origins of Iron Working in India: New Evidence from
the Central Ganga Plain and the Eastern Vindhyas.' Antiquity 77, no. 297
(2003): 536-544. https://doi.org/10.1017/S0003598X00092590.
Tripathi, Vibha. The Painted Grey Ware: An Iron Age Culture of Northern
India. Concept Publishing Company, 1976.