1 5 7

I n t r o d u c t i o n

The study of bead manufacture and changing stylesof beaded ornaments is an important method for investi-gating the social and economic development of a society.Numerous studies, including our on going work atHarappa (Meadow & Kenoyer 2001) have demonstratedthat the careful documentation of bead manufacturingtechniques, raw material sourcing and stylistic analysiscan reveal valuable information about prehistoric cul-tures. The analysis of beads from different periods andareas of Harappa have made it possible to define specifictrade networks and the organization of production as wellas changing patterns of interaction over the history of thesite (Kenoyer 2000, 2001).

During the past 15 years of excavation and research atHarappa, all of the major mounds have been systemati-cally surveyed and selected areas have been both hori-zontally and vertically excavated. In the course of theseexcavations, large numbers of beads, drills and bead mak-ing debris have been recovered from different parts of thecity and from all of the major chronological periods(Fig. 1 & Table 1). The vast majority of the beads aremade of fired steatite (Table 2), which was a widely usedraw material, beginning with the Ravi Period and contin-uing through the Late Harappan Period. The long use ofa raw material combined with changing techniques ofproduction and bead morphology provide a unique per-spective on the technological tradition and the peoplewho used it. Terracotta is a locally available material thatwas also used to make beads throughout the history ofHarappa. The use of faience for making beads starts inthe Kot Diji phase and continues on through the Late

Harappan phase. Beads that are made of hard stone suchas agate, carnelian are relatively less common, with a sig-nificant drop in numbers for stones such as lapis lazuli,grossular garnet, serpentine and amazonite. Marine shell,which is relatively abundant at the site in the form of shellbangles and inlay, was not a common bead material, pos-sibly because of the common use of white-fired steatite.Precious metals such as copper alloys and gold wereprobably quite intensively recycled, so it is not surprisingthat these materials are not often recovered in the courseof excavation. A rare discovery of preserved seeds of theCoix plant comprises the only evidence of organic beadsfrom Harappa. These various materials, regardless oftheir abundance or scarcity reveal the importance ofbeads to Harappan culture and the dynamic nature of thebead industry over time.

The largest proportion of beads comes from theHarappa Phase (Period 3, 2600-1900 BC) that has beenexcavated most extensively, but there are important col-lections from all of the other periods as well. At this timeit is not possible to separate all of these beads and man-

Bead Technologies at Harappa, 3300-1900 BC: A Comparative SummaryJONATHAN MARK KENOYER

Harappa Chronology

Early HarappanPeriod 1A/1B Ravi aspect of Hakra Phase* > 3700 BC - 2800 BCPeriod 2 Kot Diji Phase c. 2800 BC - 2600 BCHarappan (Indus)Period 3A Harappa Phase A c. 2600 BC - 2450 BCPeriod 3B Harappa Phase B c. 2450 BC - 2200 BCPeriod 3C Harappa Phase C c. 2200 BC - 1900 BCLate HarappanPeriod 4 Late Harappa Transition Phase c. 1900 BC - 1800 BCPeriod 5 Late Harappa/Cemetery H Phase c. 1800 BC - 1300 BC

* called the «Ravi Phase» in this article

Table 1.

1 5 8

Bead Technologies at Harappa ❘ k e n o y e r ❘

ufacturing debris into different chronological periodsbecause we are still in the process of analyzing the beadstyles and the contexts in which some of them were dis-covered. However, as will be illustrated below, discretesamples of beads and manufacturing debris from well-stratified contexts in different periods, make it possible tocompare specific aspects of bead technology, beadshapes and styles, as well as the organization of bead pro-duction over time. Scanning electron microscopy allowsfor the comparison of sawing and drilling techniques,while basic sequences of manufacture (chaîne opératoire)can be defined through the analysis of manufacturingdebitage, tools and finished commodities. Preliminaryresults indicate that some important similarities and con-tinuities link one period to the next. However, distinctivepatterns can be identified from each period indicating

Fig. 1 – Bead Making Areas of Harappa: 3300-1700 BC.

important changes in the technology as well as the organ-ization of production. Some of the technological featurescan be used to reliably date many beads found in sec-ondary contexts.

Organization of Bead Technology

The organization of bead technology is defined tosome extent by the nature of the raw materials being usedand the complexity of the technology and tools used toprocess the raw materials. Generally speaking we candivide the production of beads discovered so far atHarappa into the following eight categories (Table 3).Terracotta, soft stone and hard stone bead making isfound in all periods at Harappa. It is not unlikely that

1 5 9

organic beads were used throughout the history of the siteas well, but we have only found them preserved in theHarappa phase. The absence of shell from the LateHarappan phase is significant, as it may reflect the break-down of trade networks to the coast (Kenoyer 1995).Given the abundance of gold and silver ornaments foundfrom Late Harappan sites in India (Khatri & Acharya1995), the absence of such beads at Harappa during the

Late Harappan Phase may be due to the small area exca-vated. It is also possible that gold objects found on thesurface have generally been lumped with the Harappaphase. Bone, antler and ivory beads have not been found,but can be expected in the future.

Terracotta Bead Production

The manufacture of terracotta beads involves the col-lection of fine clay and preparation of a bead by hand orwith the aid of some tools. Most terracotta beads weremade simply by rolling or pinching the form and perfo-rating the bead using a long thin tool, such as a reed oreven a porcupine quill. Incising, impressing, or pinchingthe surface of the bead was often done to create addi-tional decoration. During the Harappa phase some terra-cotta beads were made using different colors of clay tocreate the effect of banded sandstone or jasper. Finally,the firing of terracotta beads was a relatively simpleprocess and could be done using a small fire or in thecourse of firing pottery.

Faience Bead Production

In contrast with terracotta beads, the production offaience beads is extremely complex (Kenoyer 1994;McCarthy & Vandiver 1990) and requires materials thatwere not locally available. Rock quartz was brought tothe site; heated and crushed to make a powder that wasthen melted using a flux, probably from plant ash

Harappa 1986 to 2001: Beads and Manufacturing Debris from all Periods*

Artifact Type NumberSteatite / Paste Bead 17037Terracotta Bead / Whorl 3993Faience Bead 2224Lapis, Serpentine, Garnet, Amazonite, etc Beads 227Agate, Carnelian, Jasper Beads 907Copper/Bronze Bead 143Gold Bead 63Marine Shell Bead 136Organic – Coix seed 3

TOTAL 24,733

Unfinished Bead – Steatite 655Unfinished Bead – Agate, etc 92Unfinished Bead – Lapis etc 24

TOTAL 771

Sawn Steatite 1542Steatite Fragments 654Sawn Agate, Lapis etc. 19Blocklet – Agate, Lapis etc. 197Flake – Agate, Lapis etc. 3141Nodule – Agate, Lapis etc. 121Gold Fragments 29Shell Bead Manufacturing Debris 153

TOTAL 5,856

Chert Drill/Agate Drill 2050Constricted Ernestite Drill 28Ernestite Drill Manufacturing Debris 37

TOTAL 2115

Glass Bead – Pre-Modern 1Glass Bead – Modern 434Plastic Mead – Modern 44Shell Bead – Modern 1Agate Bead – Modern 1

TOTAL 481

GRAND TOTAL 35,590

* This is a preliminary tabulation and will be updated after microdebitage samples have been sorted.

Bead Production at Harappa by Phase

Bead Material / Ravi Kot Diji Harappa Late HarappaProduction Phase Phase Phase Phase

Terracotta present present present presentFaience present present presentSoft Stone (steatite) present present present present

Hard Stone (agate, lapis, etc.) present present present present

Shell ? present present ?Copper/Bronze present present presentGold/Silver present present ?Organic present

Table 2.

Table 3.

1 6 0

Bead Technologies at Harappa ❘ k e n o y e r ❘

(potash). Colorants were added to the resulting friteither during the initial processing or in the course ofsubsequent grinding and melting. The frit was regroundand mixed with additional fluxing materials and col-orants prior to being shaped into beads and ornaments.After careful drying to allow the fluxing agent to efflo-resce to the surface, the beads were fired in speciallyprepared ceramic canisters and glazed. Recent experi-mental studies of firing techniques suggest that the can-isters may have been fired in a bonfire or pit kiln and notin an updraft kiln. Large quantities of misfired beadsand faience slag, burnt bone and vitrified canisters arecharacteristic of debris from faience bead making activ-ities.

Soft Stone Bead Production

Soft stone beads at Harappa are made from varietiesof steatite (talc) that have different colours and qualities.The most common varieties are greyish-white to bluishgrey, black to grey-black, yellowish green to greenishwhit, and brown to tan. Some rare examples of reddishsteatite and mottled variations of the more commonforms have been found. Detailed sourcing of the steatiteat Harappa is being done by Randall Law and more pre-cise classification of steatite types will be possible whenhis study is completed.

Unlike clay, which is locally available, all steatite atHarappa was brought to the site from distant resourceareas more than 300 km from the site. Unworked lumpsof steatite range in size from 1 to around 7 cm maximumlength, but most of the fragments have been sawn orground to prepare bead blanks. The saws used to cutsteatite were primarily fine toothed metal blades of cop-per or bronze. Stone blades were occasionally used tocreate a groove for splitting a blocklet of steatite, butnever to make the thin wafer-like slabs used to make discbeads. The manufacture of steatite disc beads at Harappais similar to that described for Mehrgarh and Mohenjo-daro (Vidale 1989, 1995). Flat slabs were sawn fromblocklets and then either chipped or ground to the basicshape and size of the desired bead. The bead blank wasperforated using a fine pointed object and then all of theperforated beads were strung on a cord. The strand ofrough beads was rounded by rolling diagonally across a

flat grinding stone and sometimes the exterior edge wassmoothed or faceted. During the Ravi Phase, the exteriorof the bead was left rough to better hold a glaze.

After the string of beads was ground to the desiredshape and size the beads were either fired with no furthertreatment, or bleached using some form of alkali and thenfired. Some steatite beads during the Ravi Period werecoated with a powdered frit to create a blue green glaze.During the Harappan and Late Harappan periods steatitebeads were sometimes incised with decorative designsthat were filled with colors, such as vermilion, or just leftplain. The firing techniques used for steatite beads isprobably identical to that used for faience describedabove.

Hard Stone Beads

A wide variety of stone beads were produced atHarappa during all of the major periods. The manufac-turing process involved various stages of shaping,grinding, perforation and polishing that was defined bythe nature of the rock itself (Bhan et al. 1994). Eachvariety of stone was acquired from some distant resourcearea that was more than 300 to 800 km from the site.Microcrystalline rocks such as agate, carnelian and jasperappear to have been heated prior to flaking. Other vari-eties of rock that would not benefit from heating, such aslimestone, sandstone, rock quartz, amazonite, serpentine,lapis lazuli and grossular garnet, were processed directly.Chipping, grooving, sawing, splitting, pecking andgrinding were the major methods used to shape the beadblanks. The drilling techniques used on hard stone variedover time and will be addressed in the appropriatesections below. Major drilling techniques includedpecking with a stone tipped percussor, drilling with atapered cylindrical stone drill or with a constricted cylin-drical Ernestite drill (Kenoyer & Vidale 1992), or acopper drill with abrasive. The copper drills were eithersolid rods or hollow tubular in form (Kenoyer 1997).Polishing, heating to enhance the color or bleaching tocreate white patterns were the final stages of productionbefore a bead was threaded onto an ornament.

1 6 1

Shell Bead Production

Shell beads were produced using two differentprocesses (Kenoyer 1984, 1995). The most expedientprocess involved making a hole in a natural shell to cre-ate a bead or pendant. Perforation was done by grinding,chipping, or using a drill made from chert or copper. Themore complex process of shell bead making involved firstbreaking or sawing off a segment of the shell body or thewhorl and then shaping the bead blank by sawing orgrinding. Perforation was done by drilling with a stone orcopper drill. After final polishing the bead was ready foruse. During the Harappa phase shell and stone weresometimes combined to created layered beads with bandsof white shell alternating with different colors of stone.

Metal Beads

The production of beads from copper, bronze, goldor silver would have involved all of the basic processesused for other aspects of metallurgy (Kenoyer & Miller1999). There is no evidence for smelting of copper ore atHarappa, so we assume that all of the copper/bronzebeads were made using imported copper ingots or scrap

metal. Some copper beads from Harappa have beenalloyed with tin, but it is not possible to determine if thesebeads were made from recycled tools or if they wereintentionally alloyed to make a golden colored bead. Goldbeads at Harappa are made using a wide range of tech-niques that include drilling solid discs of gold, hammer-ing thin sheets of gold to make tubular forms, drawingand hammering to make wire beads, or by making thinsheets of gold to cover copper beads. Similar techniqueswould have been used for silver beads, but we only havea few examples that are heavily corroded. No primaryworkshop for metal working has been excavated atHarappa, but numerous small crucibles have been foundthat may have been used for gold working, and largercopper crucibles have also been discovered.

Organic Beads

As mentioned above, we have only one example ofthree preserved Coix seeds (Job’s Tear) that were used asbeads (Meadow et al. 1998). These seeds were foundinside a tiny pot along with carnelian, faience and steatitebeads. The association of the natural bead with othermore valuable stones and artificial materials suggests that

Fig. 2 – Selected Ravi Phase Beads at Harappa: 3300-2800 BC.

1 6 2

Bead Technologies at Harappa ❘ k e n o y e r ❘

they may have had some specific symbolic meaning. Wecan assume that other perishable materials were also usedfor ornamentation and must keep looking for preservedexamples.

This brief summary of the different types of bead pro-duction shows how each category of raw materialrequires distinct processing techniques and stands aloneas a production technology. It is not unlikely that crafts-men were capable of making beads from different mate-rials and some workshops may have been involved withthe production of beads from several related materials.However, the specialization required for some of thesetechnologies, such as metal, faience and hard stone beadworking would suggest that they were produced in dis-tinct workshops. In the following section, the major evi-dence for the different bead technologies during eachmajor occupational period at Harappa will be presentedto demonstrate how the technologies are interconnectedand how they changed over time. The differencesbetween the specific bead technologies, from raw mate-rial acquisition to final distribution of finished beads, arecritical to understanding the relative “value” of beads andfor defining the different ways in which production couldhave been controlled by elite groups (Vidale 2000).

Ravi Phase Bead Making

During the Ravi Phase occupation we have evidencefor terracotta bead making as well as both soft stone andhard stone bead production (Fig. 2). Terracotta beadnecklaces and isolated beads provide a wide range ofshapes and styles of beads. Necklaces were made withmultiple bead shapes, including long barrel, short bicone,long bicone and triangular section bicones as well as bothshort and long cylindrical beads. Most of the terracottabeads were undecorated, but a few have been impressedwith simple weave coarse fabric to create a textured sur-face. The production of these beads may have takenplace in the household or in conjunction with potterymaking, but so far we have no direct evidence for terra-cotta bead production.

On the other hand we have been fortunate to find twodistinct areas with evidence for soft and hard stone beadmanufacture even though we have excavated only twosmall areas of the site (Trench 39S; 4 µ 4 m and Trench

39N; 6 µ 11 m). Both types of beads were being made inopen areas associated with household debris, pits, hearthsand domestic pottery. The distance between the twoareas is only 25 to 30 m, with the hard stone beads beingmade in the northern area and the soft stone (steatite) inthe south.

In the southern area, evidence for steatite bead pro-duction is seen in the large quantities of unfinished beads,some sawn fragments of manufacturing waste, andnumerous finished beads that had been hardened andwhitened by glazing and firing at high temperatures.Some of the fired beads had been decorated with a blue-green glaze which was probably made with powdered fritand copper oxide combined with a flux from plant ash(sajji), a process that is well documented for glazedfaience (Kenoyer 1994). Steatite bead shapes are less var-ied than terracotta beads, and short cylindrical or discbeads are the most common. Based on SEM studies ofthe drill holes, it appears that the longer steatite beadswere drilled by slender copper drills (Kenoyer 1997),while the shorter beads could have been perforated withsharp thorns or tiny slivers of chert.

The manufacturing debris and finished hard stonebeads found in the Ravi levels reveal the use of three dif-ferent perforation techniques, but the range of shapes iseven more limited than that of the steatite beads. Themain types are short and long biconical beads, and shortand long cylindrical beads. Lapis beads were drilled withwhat appears to have been stone (chert or jasper) drillswhile carnelian beads were perforated by pecking. Beadsmade from amazonite, which has a fragile crystallinestructure were perforated with a tapered cylindrical stonedrill (Kenoyer & Vidale 1992). The presence of differentdrilling techniques in the same area of the site indicatesthe virtuosity of craftsmen who had developed special-ized ways to effectively perforate different types of rawmaterials (Fig. 3).

Evidence from bead production and other artifactsreveal the establishment of extensive exchange networksduring the Ravi Period. Beginning around 3300 BC,Harappa was connected with the agate and amazonitesources in Gujarat and Rajasthan, the lapis lazuli sourcesto the west and north in Baluchistan and Afghanistan,and copper resource areas in Baluchistan and/orRajasthan (Kenoyer 1997). It was also connected to thecoastal regions to the south for shell used in bangle mak-

1 6 3

ing. The household production of beads from importedraw materials suggests that some individuals or commu-nities were engaged in entrepreneurial trade that wouldprofit from the local production of specialized ornaments.This pattern of long distance trade and local productionbeginning in the Ravi Phase continued through the sub-sequent expansion of the site during the Kot Diji Phase(2800-2600 BC) (Kenoyer 1997).

Kot Diji Period

Although the total area of the different excavations forthe Kot Diji Phase, Period 2 is much greater that that ofthe Ravi period, no significant primary concentrations ofbead manufacturing debris have been discovered. InTrench 39N on Mound AB (13 µ 15 m) it has been pos-sible to expose a long north south street and open drainflanked by houses made of mud brick. Along the north-western corner of Mound E (c. 2 µ 15 m) the fragmen-tary remains of mud brick houses and a small kiln werediscovered. The other areas where this period wasexposed consist of small sondages along the city wall tothe south and east of Mound E, but no significant archi-tecture or domestic features could be defined due to thesmall areas of exposure (usually less than 2 µ 2 m). Nev-ertheless, it is possible to determine the general featuresof the bead industry during this period based on the typesof finished beads and manufacturing debris that havebeen recovered and the contexts in which they werefound.

Finished beads of terracotta and steatite beads revealless variation than in the earlier Ravi Phase (Fig. 4). Forterracotta, the main forms are short cylindrical beadsalong with some short and long bicones. Fired steatite

beads were primarily made in the form of short cylindri-cal forms or what are commonly referred to as disc beadsthat range in size from very small microbeads (2 mmdiameter) to large discs (+ 12 mm diameter). While in theRavi Phase, the steatite beads were relatively thick, theKot Diji phase beads are often only 1 to 2 mm in thick-ness. Some of the steatite beads have been found ingroups that appear to be necklace fragments of graduatedbead sizes. The original complete necklace appears tohave been made with hundreds of thin beads that wouldhave formed an elegant flexible ornament. The apparentdecrease in variation of terracotta and steatite beads mayrelate to the presence of faience technology during theKot Diji Phase.

Although relatively few faience beads have beenfound, they occur in several different shapes, and a widerange of sizes. A segmented faience bead found onMound E, indicates the production of microbeads offaience, while a large blue-green lenticular barrel fromMound AB is an example of highly refined compactfaience production (Fig. 4). Hard stone beads continuedto be made in a variety of materials such as agate, jasper,carnelian, lapis, as well as some new forms of stone thatare totally unique and unidentified. The shapes of thestone bead have more variation than in the Ravi Phase,with long barrel, long bicone as well as lenticular and tab-ular shapes being added to the previous range of forms.Longer beads of hard stone had to be drilled with moreeffective drills, and it is possible that the development ofthe Ernestite drill began in the Kot Diji Phase, but so farno Ernestite drills have been recovered from these levels.

Rare examples of copper and gold beads have beenfound and these metal ornaments were made as shortspheres or sequins. While copper and gold ornamentsmay have been present earlier, it is only during the Kot

Fig. 3 – Ravi Phase Drills and Perforation Types.

1 6 4

Bead Technologies at Harappa ❘ k e n o y e r ❘

Diji Phase that they make it into the archaeologicalrecord. They have been found in drain deposits anddomestic debris, which suggests that people who worethese valuable metals were quite wealthy and did notspend a lot of time trying to find lost ornaments.

The significance of the changing patterns of beadstyles and variations in raw material during the Kot DijiPhase is consistent with the overall trend of site expansionand larger populations in the settlement. As the settle-ment grew from a small village to a larger town, there wasincreased standardization and specialization in someforms of bead production, such as terracotta and steatite.On the other hand we see evidence for more variation intypes of bead raw materials and styles in the faience, hardstone and metal bead technologies. This variation can beassociated with increased economic complexity and theemergence of social or political hierarchies within thesociety. It is significant that the exotic hard stone beads,as well as gold and faience beads from the wide drain onMound AB, were found in association with seal makingdebris and pottery with Early Indus script. Seals and writ-ing are indicators of economic and political power andthe people with this power may have been the ones usingand losing their ornaments in this part of the site.

Harappan Period

Most of the evidence for bead making at Harappacomes from the Harappa Phase, Period 3 (2600-1900BC). Beads appear to have been made from every avail-able material and in a wide range of styles. Since Harappaphase bead production has been extensively discussed inother articles (Bhan et al. 1994; Kenoyer 1986, 1994,1997; Vidale 2000) only the major patterns will be dis-cussed here. Terracotta beads were once again made inmany different shapes and sizes (Fig. 5), with even morevariation than seen during the Ravi Phase. This same pat-tern is seen in the manufacture of steatite beads. What ismost significant, is that both terracotta and steatite rawmaterials were used to replicate shapes and styles ofbeads made from other materials. Long biconical beadsof terracotta were painted red to imitate the valuable longcarnelian bicones. Steatite was painted red and white orincised and inlaid with different colors to imitate stonebeads. This same pattern is seen in faience beads, wherethe colors of faience glazes appear to have been made toimitate turquoise, lapis, shell or white steatite, and evencarnelian. Even in the production of hard stone beads, wesee carnelian being painted to imitate natural orbicular

Fig. 4 – Selected Kot Diji Phase Beads at Harappa: 2800-2600 BC.

1 6 5

jasper with eye designs. Copper was alloyed with tin tocreate a golden color and many copper beads with hightin content may have been made to imitate more preciousgold beads.

The production of beads that look similar but aremade from different qualities of materials represents anextremely important new aspect of Harappan society.Bead ornaments were clearly important for ritual or eco-nomic status and they appear to have been worn as pub-lic displays of ideology, status, wealth and/or power.However, people who were unable to afford original rawmaterials were still able to wear ornaments that lookedsimilar, even if they were made from less valuable mate-rials. This aspect of the bead industry suggests that beadswere being used in two different ways within the Harap-pan society. On the one hand, elites would have been dis-tinguishing their status and wealth by wearing beads

made from exotic materials or produced through exqui-site manufacturing processes. Examples would be theexotic patterned jasper beads with eye designs, long car-nelian beads perforated with slender Ernestite drills(Fig. 6), or tiny microbeads (1 mm diameter) made fromsteatite (Fig. 5). At the same time, other communitieswould have been wearing imitations of these more valu-able ornaments either for their ritual value or in anattempt to emulate the elites. The shared styles of orna-ments reflects the vertical integration of hierarchicalIndus society, while the variations in raw materials andtechnology can be seen to reflect ritual, economic orpolitical status.

Another important aspect of stone bead makingrelates to the people who actually acquired the raw mate-rials and produced the beads, as well the communitieswho controlled the production and distribution of fin-

Fig. 5 – Selected Harappa Phase Beads at Harappa: 2600-1900 BC.

1 6 6

Bead Technologies at Harappa ❘ k e n o y e r ❘

ished beads. For example, agate and steatite bead makingrepresent two very different craft traditions and both ofthese technologies developed side by side at Harappa,beginning in the Ravi Phase. However, as outlined above,the organization of production stages and the varyinguses of pyrotechnology, drilling techniques and finishingare quite distinct.

Limited excavations of the early levels of the HarappaPhase (Period 3A, c . 2600-2450 BC) on both Mound AB(Trench 42) and Mound E (Trench 54), have revealeddistinct concentrations of steatite bead manufacturingdebris. This indicates that specialized workshops forsteatite disc bead production were located in both of themajor mounds and the discovery of manufacturing debrisin other areas suggests that there may have been severalcompeting workshops within each walled sector of thecity. The absence of agate manufacturing debris in direct

association with the steatite debris would indicate thathard stone bead production was located in other parts ofthe mound or in adjacent areas that were not excavated.

Horizontal excavations in levels dating to Period 3B(c. 2450-2200 BC) provide a larger exposure of craftactivity areas along the western edge of Mound E(Trench 54). Debris found in ancient dumps along thestreet and between houses, as well as in house fill, con-tained stone beads, agate flakes and steatite manufactur-ing debris. Other crafts in the area include pottery mak-ing, gold working, and a workshop for the production ofbeads and inscribed tablets made from both faience andsteatite. The association of these crafts in one generalarea inside the ancient city walls is an important patternfor understanding the role of these crafts in the overalleconomic and political structure of the urban center. It ispossible that each walled mound at Harappa had more

Fig. 6 – Harappa Phase Drills and Perforation Types.

1 6 7

than one area for craft activities, and during the follow-ing Period 3C there are distinct locations on each mound,usually just inside a major gateway, where we see theassociation of many related crafts.

On Mound E, copper working, that may haveincluded the production of copper beads, is concentratedin the south central area of the site, just to the north of amajor gateway. Manufacturing debris from agate beadmaking, steatite bead manufacture, and shell working(possibly including shell beads) is located in the mainnorth south street leading from the gateway into the cen-ter of Mound E. It is assumed that a major craft produc-tion area is located somewhere along this major street. OnMound ET, a similar pattern of debris is found in thestreet leading down to a different gateway, and in thispart of the site a large area with workshop debris frommany different crafts was excavated. Primary bead man-ufacturing areas have been identified on the basis nodulefragments, flaking debitage, bead blanks, partly drilledbeads, several different types of spent drills (Fig. 6), anddrill manufacturing waste. Microdebitage from all stagesof steatite and agate bead manufacture has been recov-ered from some floors inside rooms and courtyard areas.Excavations in adjacent areas of the site have no evidencefor steatite or agate bead working.

Other crafts in this area include faience ornament andpossibly bead production, gold working, shell ornamentand inlay manufacture, as well as bone and ivory working.All of these crafts may also have included the manufactureof beads, but this remains to be confirmed through furtheranalysis of the manufacturing debris. In addition, this areahas evidence for chert weight manufacture, and chert toolproduction that may have been associated with woodworking or furniture inlay setting. The overall pattern inthis part of the site indicates that bead making wasrestricted to a small area inside the gateway, associatedwith numerous other specialized crafts as is seen in craftbazaars of traditional cities of the subcontinent today.Although the location of these crafts in a restricted areainside the gateway cannot be interpreted as direct controlby urban elites, it does indicate a degree of indirect con-trol through the regulation of movement into and out ofthe craft area. The presence of chert weights and weightmanufacturing debris in the area just inside the gatewaymay indicate the taxation of goods coming into or leavingthis walled sector of the city (Kenoyer 1997).

Late Harappan Period

No manufacturing areas have been discovered fromthe Late Harappan period levels at the site since most ofthem have been destroyed by brick robbing. However, acollection of assorted beads in a small pot in a LateHarappan house does provide a glimpse of technologicalinnovation, changing trade networks and socio-economichierarchies during the Late Harappan period (Kenoyer1997). A total of 133 beads and other objects were foundin the pot, but here we will focus only on the Late Harap-pan beads, which can be defined by their unique shapes,style of manufacture and raw material (Fig. 7).

Steatite beads with new styles of incised decorationindicate that this raw material continued to play animportant role in the bead industry. The Late Harappanfaience beads in the pot were made in unique shapes andwith different coloring than during the Harappan period.Azure blue faience beads may have been intended toreplicate lapis lazuli, while the bright turquoise coloredfaience of the Late Harappan period is extremely com-pact and glassy. A red-orange colored glass bead thatlooks very much like imitation carnelian was also foundin the pot and represents the earliest glass in the IndusValley. The pot was found in levels dating to before 1700BC and possibly even as early as 1900 BC. On the basisof this well dated bead and numerous other surface findsthat appear to date to the Late Harappan period, we cansay that glass production in the Indus valley is an indige-nous development and slightly earlier than glass produc-tion documented in Mesopotamia and Egypt. The earli-est sustained glass production in Mesopotamia is around1600 BC and in Egypt it appears around 1500 BC(Moorey 1994: 190 ff).

The emergence of new styles and techniques is alsoseen in technological features and types of raw materialsused for stone beads from the Late Harappan period.Hard stone beads appear to have been drilled using a tinycopper tube with abrasive sand (Fig. 8). This new tech-nique appears to have replaced the more efficient use ofErnestite drills, which have not been found from thisperiod. Although we do not know the exact source ofErnestite during the Harappan period, it probably camefrom the general regions of either modern Gujarat orBaluchistan, which were also the source for carnelian andjaspers. The lack of these drills and the apparent decline

1 6 8

Bead Technologies at Harappa ❘ k e n o y e r ❘

in the use of carnelian and lapis indicates the breakdownof long distance exchange within the greater Indus valleyduring the Late Harappan period. This restructuring ofregional trade is also suggested by the absence of shellornaments and the use of new raw materials for beads,such as banded black and white agates. These uniquestones do not appear to have been available during theHarappan period and may have been obtained from newsource areas further to the east in the central Deccanplateau or the Vindhyan mountains (Kenoyer 1995).

When the evidence from stone bead making is com-bined with other artifacts and architectural features, it is

clear that major changes were happening in the eco-nomic, political and ideological spheres of the LateHarappan period. However, contrary to the model ofdecline and decay proposed by earlier scholars, the beadindustry reveals technological innovation, new styles ofornaments and the exploitation of new source areas.While the common people may have continued to wearornaments of terracotta or other more easily producedmaterials such as steatite, the faience, glass and stonebeads represent a dynamic bead industry and discerningurban consumers.

Fig. 7 – Late Harappan Beads at Harappa: 1900-1700 BC.

1 6 9

C o n c l u s i o n

Bead making techniques and styles of beads from themajor periods at Harappa provide a unique perspectiveon the complex history of this important settlement.Unlike subsistence activities or the production of utilitar-ian tools, bead makers were involved in creating orna-ments that could be used and in most cases viewed by allmembers of the society. The selection of rare and exoticmaterials and the development of more refined technolo-gies allowed early bead makers to produce objects thathad both beauty as well as power.

During the initial settlement of Harappa (Period 1,3300-2800 BC), the Ravi phase bead makers were highlyskilled in making tiny beads from material brought fromgreat distances to the site. The soft steatite and hard agatebecame the foundation of later bead making technologies.Although these two technologies remained distinctthroughout the history of the site, they were often prac-ticed side by side in larger craft areas of the city. Duringthe Harappan Phase, specifically in Period 3C (2200-1900 BC), we see bead making workshops established ineach of the major walled sectors of Harappa.

Since bead making was one of the main crafts used tocreate objects used to reinforce social and ritual statuswithin the Indus cities, it would have been important forelites and merchants to be able to control the productionof high value ornaments. While the knowledge of specificbead technologies may have been passed on within a sin-gle community, the actual production and distributioncould have been controlled, either directly or indirectly bymerchants or urban elites. One example of more directcontrol is seen in the faience and steatite workshop exca-vated in 2000-2001, where beads were being made alongwith inscribed tablets (Meadow et al. 2001) (see Meadow& Kenoyer in this volume). However, most of the beadproduction at Harappa appears to have been controlledindirectly by situating craft areas within walled sectors ofthe city. In addition to protecting the market places filledwith valuable goods, the walls and gateways of Harappawould have allowed for the taxation of both bead rawmaterials being imported into the city as well as finishedbeads and ornaments leaving the city.

As is evident from this brief overview, the general pat-terns of continuity and change in the bead industry canbe outlined on the basis of recent work at Harappa. How-

ever, additional work remains to be done on the sourcingof raw materials and specific aspects of technology as wellas the distribution of finished beads within the city ofHarappa and surrounding sites. Further excavations arealso needed to understand the nature of the Kot Dijiphase bead industry and Late Harappan bead produc-tion. It is clear however, that the study of Indus beadmaking and its legacy will continue to provide valuableinsights on the processes underlying the emergence ofIndus urban centers.

Acknowledgements: Most of the data presented in this paper derives fromrecent excavations at Harappa and I would especially like to thank theGovernment of Pakistan, Department of Archaeology for facilitating ourcontinued work at Harappa. Special thanks to all the colleagues who haveparticipated in the research at Harappa and have helped to collect and ana-lyze data, with extra special thanks to Dr. Richard Meadow, Dr. RitaWright, Mrs. Barbara Dales, Dr. William Belcher, Dr. Heather M.L.Miller, Sharri Clark, Randy Law, Brad Chase, Petr Eltsov, Asim Dogar,Nadeem Ghouri, and all the excavators at Harappa.

My ongoing research on bead technology at Harappa and the IndusValley Civilization in general has been supported by numerous organiza-tions: the National Science Foundation, the National Endowment for theHumanities, the National Geographic Society, the Smithsonian Institu-tion, the American School of Prehistoric Research (Peabody Museum ofArchaeology and Ethnology, Harvard University), the University of Wis-consin, and private donors.

Fig. 8 – Late Harappan Perforation Types.

1 7 0

Bead Technologies at Harappa ❘ k e n o y e r ❘

R E F E R E N C E S

Bhan, K.K., Vidale, M. & Kenoyer, J.M. (1994)Harappan Technology: Methodological andTheoretical Issues. Man and Environment19(1-2), pp. 141-157.

Kenoyer, J.M. (1984) Shell Working Industriesof the Indus Civilization: A Summary. Paléori-ent 10(1), pp. 49-63.

Kenoyer, J.M. (1986) The Indus Bead Industry:Contributions to Bead Technology. Ornament10(1), pp. 18-23.

Kenoyer, J.M. (1994) Faience Ornaments fromHarappa and the Indus Civilization. Ornament17(3), pp. 36-39, 95.

Kenoyer, J.M. (1995) Interaction Systems, Spe-cialized Crafts and Culture Change: TheIndus Valley Tradition and the Indo-GangeticTradition in South Asia. The Indo-Aryans ofAncient South Asia: Language, Material Cultureand Ethnicity, ed. G. Erdosy. Berlin, pp. 213-257.

Kenoyer, J.M. (1995) Shell Trade and ShellWorking during the Neolithic and Early Chal-colithic at Mehrgarh. Mehrgarh Field Reports1974-1985. From Neolithic Times to the IndusCivilization, ed. C. Jarrige, J.-F. Jarrige, R.H.Meadow & G. Quivron. Karachi, pp. 566-581.

Kenoyer, J.M. (1997) Trade and technology ofthe Indus Valley: new insights from Harappa,Pakistan. World Archaeology 29(2), pp. 262-280.

Kenoyer, J.M. (2000) Wealth and Socio-Eco-nomic Hierarchies of the Indus Valley Civi-lization. Order, Legitimacy and Wealth in EarlyStates, ed. J. Richards & M. Van Buren. Cam-bridge, pp. 90-112.

Kenoyer, J.M. (2001) Early Developments ofArt, Symbol and Technology in the Indus Val-ley Tradition. Indo-Koko-Kenkyu [IndianArchaeological Studies] 22, pp. 1-18.

Kenoyer, J.M. & Miller, H. M.-L. (1999) MetalTechnologies of the Indus Valley Tradition inPakistan and Western India. The Archaeomet-allurgy of the Asian Old World, ed. V. C. Pigott.Philadelphia, pp. 107-152.

Kenoyer, J.M. & Vidale, M. (1992) A New Lookat Stone Drills of the Indus Valley Tradition.Materials Issues in Art and Archaeology, III, ed.P. Vandiver, J.R. Druzik, G. S. Wheeler & I.Freestone. Pittsburgh, pp. 495-518.

Khatri, J.S. & Acharya, M. (1995) Kunal: ANew Indus – Saraswati Site. P u r a t a t t v a 25, pp.84-86.

McCarthy, B. & Vandiver, P.B. (1990) AncientHigh-strength Ceramics: Fritted Faience Ban-gle Manufacture at Harappa (Pakistan), ca.2300-1800 B.C. Materials Issues in Art andArchaeology, II, ed. P.B. Vandiver, J. Druzik &G.S. Wheeler. Pittsburgh, pp. 495-510.

Meadow, R.H. & Kenoyer, J.M. (2001) RecentDiscoveries and Highlights from Excavationsat Harappa: 1998-2000. Indo-Koko-Kenkyu[Indian Archaeological Studies], 22, pp. 19-36.

Meadow, R.H., Kenoyer, J.M. & Wright, R.P.(1998) Harappa Archaeological Research Project:Harappa Excavations 1998, Report submittedto the Director General of Archaeology andMuseums, Government of Pakistan, Karachi.

Meadow, R.H., Kenoyer, J.M. & Wright, R.P.(2001) Harappa Archaeological Research Project:Harappa Excavations 2000-2001, Report sub-mitted to the Director General of Archaeologyand Museums, Government of Pakistan,Karachi.

Moorey, P.R.S. (1994) Ancient MesopotamianMaterials and Industries. Oxford.

Vidale, M. (1989) A Steatite-Cutting Atelier onthe Surface of Moenjodaro. A n n a l i 4 9 / 1 ,pp. 29-51.

Vidale, M. (1995) Early Beadmakers of theIndus Tradition: The ManufacturingSequence of Talc Beads at Mehrgarh in the5th Millennium BC East And West 45(1-4),pp.45-80.

Vidale, M. (2000) The Archaeology of IndusCrafts: Indus craftspeople and why we studythem. Rome.