Craft history asks how knowledge, materials, labour and exchange were combined to make useful and valued objects. The archaeological record of Mithila, Vajji and Anga preserves pottery, metal tools, ornaments, beads, terracottas, bone and ivory objects, bricks and other durable remains, but it preserves wood, fibre, leather and many workshop installations much less reliably. The presence of a finished knife, bead or fine pot therefore cannot automatically be converted into proof of a local workshop. Conversely, the absence of textiles or carpentry debris cannot be treated as proof that spinning, weaving or woodworking were unimportant. This chapter reconstructs craft production by matching each claim to the evidence actually available at Chirand, Vaiśālī, Panda, Champa and related Middle Ganga sites, while using better- preserved comparator sites only to clarify technological possibilities rather than to fill local gaps (Sinha and Roy 1969; Verma 1971; Chakrabarti 2009). 11.1 Technology was social knowledge, not a list of inventions A technology is a sequence of learned actions: selecting material, preparing it, controlling heat or force, shaping the object, finishing it, maintaining tools and judging quality. These skills were transmitted through observation, apprenticeship and repeated practice. For early historic society, the most defensible questions are therefore not simply when iron, the potter’s wheel or glass appeared, but who controlled particular techniques, how raw materials reached settlements, whether production occurred in households or specialist settings, and how finished goods circulated. Archaeology can sometimes recover this operational sequence, but only when production debris and installations survive alongside finished artefacts. 11.2 A finished object does not prove a workshop This distinction is crucial for the present region. Excavations at Vaiśālī yielded pottery, terracotta and stone ornaments, metal objects and other manufactured goods, while Panda produced iron tools and weapons, semi-precious-stone beads, copper ornaments and ivory objects in different phases. Such finds securely establish use and circulation. Local manufacture becomes a stronger claim only when excavators recover raw material, blanks or unfinished pieces, specialised tools, moulds or crucibles, furnaces, slag, wasters or concentrations of production waste. In their absence, the object may have been made elsewhere and exchanged into the settlement (Sinha and Roy 1969; Archaeological Survey of India 1993–94). 119119 GAJENDRA THAKUR Figure 40 — The craft production chain: raw material, preparation, transformation, finishing, circulation and repair/discard form a sequence of technological decisions. 11.3 Craft specialisation existed in degrees Specialisation should not be imagined as a binary contrast between self-sufficient households and factory- like workshops. Some tasks could be seasonal household crafts; others required part-time specialists; still others demanded expensive pyrotechnology, trained drilling or polishing, or access to distant raw materials. Standardised products may suggest routinised skill, but standardisation alone does not reveal whether production was controlled by a state, guild, merchant, household or itinerant artisan. Comparative archaeology across early historic South Asia shows that craft production could be spatially dispersed even when towns created dense markets for specialist goods (Allchin 1995; Chakrabarti 2009). HISTORY OF MITHILA, VAJJI & ANGA — VOLUME II Figure 41 — Evidence for local production becomes stronger when finished objects are joined by tools, raw material, unfinished pieces, waste and production features. 11.4 Pottery was the largest surviving craft archive Ceramics dominate excavated assemblages because fired clay survives where baskets, wooden vessels and cloth decay. The Middle Ganga sequence includes black-and-red ware, black slipped ware, red and grey wares and, in the early historic horizon, Northern Black Polished Ware. Variation in fabric, firing, surface treatment, wall thickness and vessel form reflects technological choices. Yet a ceramic category is not equivalent to an ethnic group or single production centre. Similar wares could be made at multiple places, exchanged across regions, copied locally or produced to shared standards. The archaeological challenge is to distinguish style circulation from workshop organisation. 11.5 Northern Black Polished Ware reveals skill but not one centralised industry NBPW required unusually careful clay preparation, forming, surface treatment and firing to create its characteristic fine fabric and lustrous surfaces. Its wide distribution makes it an important marker of intensified interregional interaction, but it should not be treated as the product of one royal industry or as a mechanical index of urban status. At Panda, the NBPW phase was accompanied by black slipped, grey, black and red wares; the assemblage therefore represents a ceramic economy with several production traditions operating together rather than one luxury ware replacing everyday pottery (Archaeological Survey of India 1993–94). 11.6 Pottery manufacture linked countryside resources to urban demand Potters required clay sources, temper or other inclusions, water, fuel, drying space and firing facilities. Large settlements increased demand for cooking pots, storage jars, serving vessels, lamps, lids, ritual containers and construction ceramics. The resulting industry was therefore tied to the ecology around towns as much as to urban consumers. Even when a kiln has not been excavated, the quantity and diversity of ceramic vessels can indicate sustained production somewhere within the settlement’s economic catchment. Where kiln wasters or firing installations are absent, however, the exact place of manufacture should remain open. 11.7 Iron working must be separated into smelting, smithing and use Iron is often discussed as though every iron artefact records local metallurgy. In fact, ore smelting, bloom consolidation, forging, sharpening and repair are distinct stages that may occur at different places. The strongest evidence for primary smelting includes ore, furnaces and slag; smithing is indicated by hearths, hammer scale, slag and unfinished or repaired objects; finished tools alone establish use. Early iron in the Middle Ganga plain predates the full early historic urban phase at several sites, while Bihar sites including Chirand have iron in Black-and-Red-Ware contexts. This long chronology cautions against an ‘iron revolution’ in which urban crafts suddenly appear at one date (Chakrabarti 2009). 11.8 Panda shows a diversified iron-tool repertoire without proving local smelting The NBPW phase at Panda in Samastipur produced iron knives, sickles, daggers and spearheads, together with fine pottery and beads. These objects reveal agricultural, domestic and martial uses of iron and show that iron had become embedded in daily and political economies. But the published summary does not, by 121121 GAJENDRA THAKUR itself, document ore reduction or a smelting furnace at the site. The correct historical statement is therefore that Panda participated in a developed iron-using economy; whether every object was smelted and forged locally remains a separate archaeological question (Archaeological Survey of India 1993–94). 11.9 Copper and copper alloys occupied different technological niches Copper did not disappear when iron spread. Hooks, earrings, rods, bangles, vessels and coins could exploit copper’s malleability, appearance and resistance to corrosion. At Panda, a copper hook occurs in the Chalcolithic phase and copper earrings in a later urban horizon. Vaiśālī also yielded copper objects and ornaments. These finds illustrate technological coexistence: material choice depended on function, cost, social value and available skill. A stronger metal did not automatically replace a more workable or visually valued one. 11.10 Chirand preserves an exceptional tradition of bone and antler working Chirand is indispensable because its early levels preserve a wide range of bone and antler tools—celts, scrapers, chisels, hammers, needles, points, borers, awls, diggers and arrowheads—as well as ornaments. Verma also reported unfinished implements, demonstrating production rather than mere consumption. This evidence belongs primarily to the Neolithic history treated in earlier chapters, but it establishes a deep regional repertoire of cutting, scraping, drilling, polishing and shaping organic materials. It would be unsafe to assume unchanged artisan lineages into the early historic period; the defensible continuity is technological possibility and long familiarity with animal-derived raw materials (Verma 1971). 11.11 Bone and ivory continued beside metals because materials were functionally different Bone and ivory could be cut, polished, drilled and carved into points, handles, needles, ornaments, combs and small implements. Their continued use need not be explained simply as a cheap substitute for metal. Lightness, elasticity, colour, availability and suitability for fine carving made organic materials useful in their own right. Panda’s later levels yielded ivory objects, while other early historic sites in the broader Ganga system show bone points and carved ivory. As with metal, finished pieces establish use; unfinished blanks and production debris are required before claiming a local ivory workshop. 11.12 Stone beads reveal both skill and long-distance material movement Carnelian, agate, chalcedony, crystal and other semi-precious stones were transformed through chipping, grinding, drilling, heat treatment in some traditions and polishing. Their geological sources were unevenly distributed, so bead assemblages can signal exchange even when production occurred elsewhere. Vaiśālī and Panda yielded stone beads; their presence links consumers to wider raw-material and craft networks. Ethnoarchaeological studies of Indian beadmaking demonstrate how many specialised stages may lie behind a small finished bead and warn against identifying a workshop without manufacturing waste (Kenoyer, Vidale and Bhan 1991). 11.13 Terracotta beads and ornaments represent a different economy from rare stones Terracotta beads, ear ornaments, bangles and pendants could be made from locally available clay with lower raw-material costs than carnelian, crystal or glass. Vaiśālī produced numerous terracotta ornaments alongside stone examples. This coexistence suggests differentiated consumption rather than a simple HISTORY OF MITHILA, VAJJI & ANGA — VOLUME II evolution from cheap to expensive materials. Shape, finish and context mattered as much as raw material. Terracotta also demonstrates how craft categories overlapped: a potter or clay specialist could produce vessels, figurines, discs, toys, beads and architectural ceramics using related but not identical skills. 11.14 Glass was technologically demanding, but beads do not automatically prove glassmaking Glass production required high-temperature pyrotechnology and control over silica, fluxes and colourants; bead manufacture could then involve winding, drawing, cutting or reheating pre-existing glass. Early historic eastern India participated in expanding glass-bead networks, but the distinction between making glass and making beads from glass is fundamental. A settlement with glass beads may have imported finished beads, imported glass cane or produced beads locally. Production is best demonstrated by rods, tubes, cullet, failed beads, furnaces or characteristic waste. The broader South Asian literature shows why manufacturing technique must be reconstructed from debris and context, not bead presence alone (Kanungo 2004; Kanungo 2021). 11.15 Shell working connected riverine interiors to distant ecological zones Shell ornaments are valuable because raw material origin can differ sharply from the place of use. Freshwater shell was locally accessible in the riverine plains, while marine conch required longer-distance movement. Bangles, discs and ornaments therefore combined craft skill with exchange. Sawn shell, blanks or large quantities of working waste are stronger evidence for local manufacture than finished bangles. The presence of shell in inland towns belongs to the same history of circulation that moved semi-precious stone, metal and perhaps specialised glass products. 11.16 Textile technology is archaeologically underrepresented Clothing, cordage, sacks, nets, mats and household textiles were fundamental to daily life, but plant and animal fibres survive poorly in humid alluvial environments. Spindle whorls, loom weights, needles, impressions and occasional mineralised fibres can provide indirect evidence, while texts attest the social importance of cloth and robe donation. The evidentiary rule must remain strict: a textual reference to cloth in early Buddhist society does not prove a specific loom technology at Vaiśālī, and the absence of preserved fabric does not prove that weaving was absent. Textile history therefore requires a combined archaeological and textual method with explicit uncertainty. 11.17 Spinning and weaving could cross household and specialist domains Fibre preparation and spinning are tasks that can be embedded in household routines, while certain forms of weaving, dyeing or finishing may require dedicated skill and equipment. Production could therefore shift between domestic and specialist settings without leaving a clear architectural boundary. This matters for social history because craft specialisation is not synonymous with male workshop labour. Where the evidence does not identify the workers, gender should not be assigned automatically. Later ethnographic analogies may generate questions, but they cannot identify the gender of early historic spinners or weavers at a particular site. 11.18 Carpentry was essential precisely where archaeology is weakest Posts, rafters, doors, carts, boats, agricultural implements, furniture, handles and storage structures required woodworking, yet tropical wood usually decays. Post-holes and impressions may preserve the 123123 GAJENDRA THAKUR location of timber architecture while iron nails or clamps survive after the wood disappears. Riverine transport also presupposed boat-building and repair skills, but surviving boats are rare. A material history based only on durable artefacts therefore systematically underestimates carpenters and other workers in perishable media. 11.19 Brick and earthen construction were technologies of organised labour Early historic settlements increasingly used burnt brick, rammed floors, plaster, drains, wells and other built features alongside timber and earth. Brick production required clay preparation, moulding, drying, fuel, firing and transport; construction required measurement and coordinated labour. At Vaiśālī, burnt-brick structures, floors, fortification and water-management features show sustained investment in built space. Such structures reveal technical organisation, but they do not by themselves tell us whether labour was corvée, hired, household-based or controlled by a corporate body (Sinha and Roy 1969). 11.20 Ring-wells and drains show ceramic engineering, but chronology matters Ring-wells are composed of standardised fired ceramic rings fitted vertically to manage water or soakage in dense settlements. Their manufacture demanded moulding or forming large repeated units and successful firing. Yet ring-wells belong to different dates at different sites. At Panda, four ring-wells are specifically reported in the Kushana horizon, later than the NBPW phase. They must therefore not be backdated simply because ring-wells are broadly associated with early historic urbanism elsewhere (Archaeological Survey of India 1993–94). 11.21 Standardisation does not automatically imply state control Repeated brick sizes, vessel forms, beads or metal objects can indicate shared conventions, templates, specialist training or market demand. They do not automatically demonstrate a royal department or guild. Production can become standardised through apprenticeship networks, consumer expectation, itinerant artisans or competition among workshops. The same caution applies to NBPW, coins and architectural units. Centralisation is a political-economic claim that requires evidence beyond formal similarity. 11.22 Workshops were workplaces, but not all craft labour occurred in workshops Archaeologists often search for furnaces, kilns and concentrated waste because these features make production visible. Yet many crafts could occur in courtyards, house fronts, seasonal work areas or temporary shelters. Repair work might be mobile. Pottery could be fired beyond dense habitation because of smoke and fire risk. A town may therefore depend heavily on craft production while preserving only part of its productive landscape inside excavated trenches. Urban excavation must be connected to suburban and rural survey before craft geography can be reconstructed confidently. 11.23 Household production and specialist production were interconnected Households consumed tools, pots, ornaments and cloth, but they could also produce thread, food- processing equipment, simple clay objects or repaired goods. Specialist artisans depended in turn on household demand, rural raw materials and food supplied by cultivators. This interdependence prevents a sharp analytical division between ‘craft economy’ and ‘rural economy’. Chapter 10 traced surplus from countryside to towns; craft history shows the reverse flow of tools, containers and valued goods back into rural life. HISTORY OF MITHILA, VAJJI & ANGA — VOLUME II 11.24 Gendered craft labour is often hidden by the archive Craft histories can reproduce textual and archaeological bias by assigning visible tools to men and invisible fibre or household production to women. The evidence rarely supports such certainty. Women certainly participated in household production and textile work in many later South Asian contexts, but a direct early historic attribution requires local evidence. The safer reconstruction recognises that craft labour was organised through households, dependence, apprenticeship and specialist networks while leaving worker identity open where sources are silent. 11.25 Raw materials made crafts regional before finished goods reached markets No major settlement possessed every needed raw material. Fuel, clay, timber, bone, fibres, stone, ores, shell and colouring agents came from different ecological zones. This distributed geography meant that craft production depended on transport and exchange before a finished object entered a market. River corridors were particularly important in the study region. The Ganga and its tributaries could move bulky materials and finished goods, linking Champa and other eastern centres to wider networks while also connecting north-bank settlements to crossing points and market towns. Figure 42 — Craft production as a distributed regional network linking resources, households, specialist workshops, markets, urban consumers, institutions and repair/recycling. 11.26 Vaiśālī was a major consumer and craft environment, but workshop claims must remain specific The Vaiśālī excavations document ceramics, terracottas, ornaments, metal objects, beads and substantial architecture across a long early historic sequence. The diversity is consistent with a large settlement in which specialist production, exchange and repair were important. Yet the published corpus also demonstrates the methodological difference between artefact diversity and manufacturing proof. Unless raw material, unfinished pieces, tools and waste are tied to a specific context, the safest conclusion is that Vaiśālī participated in extensive craft and exchange systems rather than that every recovered object was made inside the city (Sinha and Roy 1969). 125125 GAJENDRA THAKUR 11.27 Panda provides unusually useful evidence for technological coexistence Panda’s sequence brings several technologies into one stratigraphic narrative. The Chalcolithic phase included stone implements, a copper hook, bone arrowheads and pottery; the NBPW phase yielded finely prepared ceramics, iron knives, sickles, daggers and spearheads, and semi-precious-stone beads; later levels added brick construction, ring-wells, copper earrings and ivory objects. The sequence demonstrates coexistence and accumulation rather than simple replacement: stone, bone, copper, iron, clay and ivory remained useful in different technical roles (Archaeological Survey of India 1993–94). 11.28 Chirand shows why technological history requires the longue durée Chirand’s importance is not that one craft tradition remained unchanged for millennia, but that the site preserves successive technological repertoires across major cultural transitions. Its bone industry, ceramics, stone tools and later metals show repeated adaptation to new materials without erasing older skills immediately. This longue durée cautions against narratives in which a single innovation—copper, iron, wheel-made pottery or urban demand—suddenly transforms every productive practice. Technology accumulated, recombined and became socially reorganised over time. 11.29 Champa and Anga belonged to a craft-and-exchange corridor, but production locations remain unevenly known Excavations at Champa establish a substantial early historic settlement with NBPW and other ceramics, fortification, structural investment and a material assemblage connected to wider exchange. Its position near the Ganga made it well placed to receive raw materials and distribute finished goods. Yet the published archaeological summaries do not justify turning Champa into a universal manufacturing centre for every craft found in Anga. The correct scale of claim is that urban demand, transport connectivity and recovered artefacts make craft production and exchange integral to Champa’s economy, while the location and organisation of individual industries require craft-specific evidence (Archaeological Survey of India 1970– 76). Figure 43 — Preservation bias: pottery and stone are much more visible archaeologically than wood, textiles and fibres. The bars are schematic, not measured values. HISTORY OF MITHILA, VAJJI & ANGA — VOLUME II 11.30 Nepal-side Mithila and conclusion: technological absence may be an archive problem For early historic Nepal-side Mithila, archaeological publication remains uneven for many categories of workshop evidence. The absence of excavated furnaces, textile installations or bead workshops in a particular published record must therefore be distinguished from the absence of the craft itself. Across Mithila, Vajji and Anga, the strongest conclusion is that early historic technology formed a network of material knowledge linking households, specialists, rural resources, towns, markets and institutions. Pottery and brick survive abundantly; metal, stone, bone and glass survive selectively; wood and textiles are strongly underrepresented. A defensible craft history therefore ranks evidence by production stage and preservation potential, rather than filling silent parts of the record with assumed guilds, fixed artisan castes or state-controlled industries. Chapter 12 follows these objects beyond production into trade, merchants and markets. Table 11.1 — Craft evidence, what it can establish, and what it cannot establish alone Evidence Supports Does not establish by itself Pottery fabrics, forms, forming, firing, consumption; local ethnicity, political identity, or one wasters, kilns manufacture where kilns/wasters occur centralised production centre Iron tools and weapons use of iron and functional repertoire local smelting without ore/furnace/slag; local smithing without smithing debris Copper objects continued use of copper/copper alloys local casting or alloying without moulds, crucibles or waste Beads of stone or glass consumption, ornament, exchange and local beadmaking without blanks, drills, material connectivity rods/tubes, cullet or failed pieces Bone / ivory objects use and fine working of organic materials continuous artisan lineage or local workshop without unfinished pieces/waste Brick, drains, ring-wells construction skill, organised labour and a specific labour regime or universal water-management technology chronology across sites Text / iconographic social valuation, terminology and a particular archaeological technique or reference to cloth or craft possible practices workshop at a named settlement Absence of wood or textiles limits of preservation absence of woodworking, weaving, cordage or baskets 127127