PHOTOGRAPHS OF HARAPPAN SITES AND ARTEFACTS

The Diversity of the Harappan Subsistence Base

The Harappan civilization covered an enormous area within which there was great ecological variety—alluvial plains, mountains, plateaux, and sea-coasts. The resource potential of this area was rich enough to generate the food surpluses that are an important aspect of urbanization. The diversity of the subsistence base may also have been an important sustaining factor—if one food resource failed, people could turn to others. Agriculture was the mainstay, supplemented by animal husbandry and hunting. Riverine and marine food resources were tapped, where available. The sources of information on the subsistence patterns of the Harappans consist of plant remains, animal bones, artefacts, motifs on seals and pottery, and analogies with modern practices.

Subsistence is closely related to environment, and the nature of the Harappan environment is the subject of continuing debate. Archaeologists such as Mortimer Wheeler and Stuart Piggott suggested a wetter climate in Harappan times on the basis of the following arguments: (a) the large number of burnt bricks found at Harappan sites would have required large quantities of fuel, which would only have been possible with a heavy forest cover, supported by heavier rainfall; (b) the gabarbands (embankments) constructed in the Baluchistan area suggest heavier rain; (c) the depiction of animals such as the tiger, elephant, and rhinoceros on seals indicates a forest and grassland vegetation that could only have been supported by heavier rainfall; (d) the elaborate drainage system of the cities was geared towards carrying off rain water. The first and last points can be refuted most easily. It is not easy to estimate just how much wood (and forest) would have been required to make the burnt bricks, and the Harappan drains were largely part of a system of sewage disposal.

Many scholars hold that climatic conditions in the greater Indus valley have remained more or less constant since Harappan times. However, some studies suggest otherwise. Plant palynologist Gurdip

Singh (1971) analysed pollen from the three salt lakes of Sambhar, Didwana, and Lunkaransar, and the freshwater Pushkar lake, and constructed a profile of rainfall in this part of Rajasthan from c. 8000 BCE to 1500 BCE. He concluded that there was an increase in rainfall in c. 3000 BCE and a decrease in 1800 BCE. However, a recent study of the Lunkaransar lake (Enzel et al., 1999) suggests that it had dried up by 3500 BCE and that the climate had become drier long before the emergence of the Harappan civilization. The issue of the nature of climatic conditions in Harappan times thus remains unresolved.

Given the area covered by the civilization, naturally there were regional variations in the plants grown by farmers. Wheat has been found at Mohenjodaro and Harappa; barley at Mohenjodaro, Harappa, and Kalibangan; and sesamum at Harappa. Harappa has also given evidence of watermelon seeds, peas, and dates. Rice occurs at Harappa, Kalibangan, Lothal, and Rangpur. Millets have been identified at Harappa, Surkotada, and Shortughai. Grapes were known, so was henna (mehendi). Cotton may also have been grown. Detailed evidence of the plant economy of the early and mature Harappan phase is available from Balu (in Haryana) (Saraswat and Pokharia, 2001–02). The crop remains identified here included various types of barley, wheat, rice, horse gram, green gram, chickpea, field pea, grass pea, sesamum, melon, watermelon, date, grapes, and the earliest evidence of garlic. Apart from the wide range of cereals, pulses, vegetables, and fruits grown by the Harappans, another striking point is the similarity of the past and present plant economies in the various regions.

TERRACOTTA PLOUGH FOUND AT BANAWALI

Modern cropping practices provide some clues to protohistoric patterns. Today, in Sindh, rainfall levels are low, but the Indus brings down flood waters and silt. The fertile land requires no deep ploughing, irrigation or manuring. Sesamum and cotton were probably sown in June/July and reaped in September/October, as kharif (summer) crops. Crops such as wheat and barley would have been sown in November and reaped in March/April as rabi (winter) crops. In Gujarat, rice is a kharif crop, and it must have been so in Harappan times as well.

Reference has already been made to the discovery of a ploughed field at early Harappan levels at Kalibangan. The continuing use of the plough into the mature Harappan phase can be inferred. Terracotta models of ploughs at Bahawalpur and Banawali give further evidence of the use of this implement. The fact that no actual ploughs have survived is no doubt because they were made of wood.

Farmers must have built bunds (embankments) of mud or stone to divert river water, as they do today in areas like Baluchistan. Irrigation canals have been found at Shortughai. Fairservis suggested that a well and associated drains at Allahdino may represent an irrigation system, but the evidence is far from conclusive. Similarly, Leshnik’s hypothesis that the dockyard at Lothal is actually an

irrigation reservoir is not convincing. Even if the Harappans did dig canals in the alluvial plains, it would be very difficult to identify them. However, H. P. Francfort (1992) has identified remains of a small-scale canal network in the Haryana area, and some of the ancient canals traced in the Ghaggar-Hakra plain may belong to the Harappan phase.

Bones of wild animals have been found at Harappan sites. These include many varieties of deer, pig, boar, sheep, goat, ass (?), and pig. Bones of tortoise and fish have also been found. Rhinoceros bones occur only at Amri, although this animal is depicted on numerous seals and in terracotta figurines. Elephant and camel bones occur in very small quantities, although the elephant appears on seals. Tigers are represented often in figurines, leopards more rarely. Rabbits, peacocks, pigeons, ducks, monkeys, and wild fowl are represented in figurines and paintings on pottery. The Harappans exploited riverine and marine resources where these were available. At coastal sites in Gujarat, molluscs provided an important protein-rich element in people’s diet. The discovery of marine catfish bones at Harappa suggests that coastal communities may have traded in dried fish in inland cities.

Harappan sites have also yielded remains of domesticated animals such as humped and humpless cattle, buffalo, sheep, and goat. Cattle and buffaloes were the most important domesticated animals. They would have been used for meat, milk, and also as draught animals. Goats and sheep could have been used for meat, wool, milk, and as pack animals (they are still used to carry loads of salt and grain in some of the Himalayan stretches). Dog figurines suggest the domestication of this animal.

The issue of the horse is controversial and hinges on the stratigraphic context in which the remains have been found and the identification of the species they belong to. For instance, it is not easy to ascertain whether the bones in question belong to the half-ass (Equus hemionus khur) or domesticated horse (Equus caballus). Horse remains have been reported at Harappa, Lothal, Surkotada, Kuntasi, and Kalibangan, and at superficial levels at Mohenjodaro. Sàndor Bökönyi (1997) examined the equid bone samples from Surkotada and concluded that at least six of them probably belonged to the true horse. His conclusions were challenged by Meadow and Patel (1997). Brigadier Ross (1946) reported horse teeth at pre-Harappan levels at Rana Ghundai, but this identification was questioned by Zeuner (1963). While horse bones may not be completely absent at Harappan sites, they are not prolific either.

NEW DIRECTIONS IN RESEARCH

Animal bones at Shikarpur

Shikarpur is a Harappan site in Kutch district in Gujarat, excavated by the Gujarat State Department of Archaeology in 1987–90. The excavation was a small one. It revealed an over 3 m thick deposit, of which the lower layers (layers 10–19) represent an early Harappan phase and the upper layers (layers 1–9) the mature Harappan phase. The animal remains found at the site were sent to the Archaeozoology Laboratory at Deccan College, Pune. The preliminary results of the detailed investigations by P. K. Thomas, P. P. Joglekar, Arati Deshpande-Mukherjee, and S. J. Pawankar have given important information about the subsistence patterns of the Harappans in Gujarat:

A total of 15,483 pieces of bone were unearthed in the excavations. It was possible to identify 53.46 per cent of them, i.e., 8,267 fragments. There were cut marks and signs of charring on some of the bones, indicating slaughtering and cooking. The faunal assemblage consisted of 47 species—23 mammals, 3 birds, 2 reptiles, 5 fish, 13 molluscs, and 1 crustacea. The wild animals included wild buffalo, nilgai, chowsingha, blackbuck, gazelle, various kinds of deer, wild pig, wild ass, jackal, hare, and rhinoceros. The domesticated animals included cattle, buffalo, sheep/goat, horse, pig, and dog.

The bones of domesticated animals comprised over 85 per cent of the total faunal assemblage in both the early and mature Harappan phases. Cattle bones were most numerous. In the early Harappan phase, 77.48 per cent of the bones were of cattle, while in the mature Harappan phase, their percentage was 77.84 per cent. Sheep/goat bones (it is difficult to distinguish the two) amounted to 11.26 per cent of the early Harappan phase, and were reduced to 4.63 per cent in the mature Harappan phase. Buffalo bones were 4.28 per cent and 4.61 per cent in the early and mature Harappan phases respectively. Dog bones were only found in the mature Harappan phase, and that too in very small quantities (0.116 per cent). Very few horse bones were found (0.13 per cent), and these occur only in the mature Harappan phase.

The evidence shows that the consumption of meat of domesticated animals was an important part of the diet of the people of Shikarpur. The contribution of wild and aquatic animals varied considerably in different layers.

The analysis of bones and teeth showed that domesticated animals were killed at different ages. Most of the cattle and buffaloes lived up to the age of maturity—about 3 years—and were killed at various ages up to the age of about 8 years. The fact that some were older than 8 years suggests that they were also valued for secondary products and used for draught purposes. Sheep/ goats were killed at relatively younger ages—between 6 months to their respective ages of maturity, suggesting they were primarily reared for meat.

Towards the end of the mature Harappan phase at Shikarpur, there seems to have been an increase in the exploitation of wild animals. It is not clear whether this was the result of a decline in agricultural production, failure of rains, population pressure, or a combination of several such factors.

SOURCE Thomas et al., 1995

Harappan Crafts and Techniques

Earlier writings tended to contrast the plainness of Harappan artefacts with the opulence of their Egyptian and Mesopotamian counterparts. Nowadays, the technological sophistication and beauty of some of the Harappan artefacts are recognized. There is a great variety of standardized, mass-produced craft items at Harappan sites. The artefacts are far greater in quantity and range, and show greater technical finesse than those found in earlier cultural phases. While some sites specialized in

the production of a single or a few items, others such as Harappa manufactured a wide range of goods. Craft activity was often localized in a certain part of the settlement.

Ceramics include all items involving the heating of clay such as bricks, terracotta, and faience. The Harappan pottery reflects efficient mass-production. Pottery kilns were found at Mohenjodaro, Harappa, Nausharo, and Chanhudaro. The pots were fired in funnel-shaped up-draft closed kilns, although open-firing kilns may also have been used. There is a great variety of pottery, including black-on-red, grey, buff, and black-and-red wares. Most pots were wheel turned. Both fine and coarse fabrics occur and their thickness varies. The typical Harappan pottery is a fine, sturdy, wheel-made ware with a bright red slip, decorated with painted black designs. Polychrome painting is rare. The red colour for the slip was made from red ochre (iron oxide, known as geru), while black was made by combining dark reddish-brown iron oxide with black manganese. Distinctive shapes include the dish-on-stand, vase with s-profile, small vessel with knobbed decoration, large slender-footed bowl, cylindrical perforated jar, and goblet with pointed foot. The decorative patterns range from simple horizontal lines to geometric patterns and pictorial motifs. Some of the designs such as fish scales, pipal leaves, and intersecting circles have their roots in the early Harappan phase. Human figures are rare and crude. At the earliest levels of Mohenjodaro, a burnished grey ware with a dark purplish slip and vitreous glaze may represent one of the earliest examples of glazing in the world. Although there is a certain level of uniformity in pottery styles and techniques across the Harappan culture zone, there are also differences between regions.

FIGURE 4.8 HARAPPAN POTTERY

Inferences can be made about the functions of some of the Harappan pots. The large jars may have been used to store grain or water. The more elaborately painted pots may have had a ceremonial use or may have belonged to rich people. Small vessels may have been used as glasses to drink water or other beverages. The function of the perforated jars is not clear. One suggestion is that they may have been wrapped in cloth and used for brewing fermented alcoholic beverages. Another possibility is that they may have had a ceremonial or ritualistic use. Shallow bowls probably held cooked food; flattish dishes were used as plates. Cooking pots of various sizes have been found. Most of them

have a red- or black-slipped rim and a rounded bottom; the lower part of the pot is often strengthened by a thick slurry or clay mixed with ground pottery or chaff. The rims of the cooking pots are strong and project outwards to help pick them up or move them around. Some of the forms and features of the pots used by the Harappans can be seen in traditional kitchens even today. Apart from ceramic vessels, the Harappans also made and used metal ones.

MINIATURE PERFORATED POT; BEAKER; POT WITH POINTED BASE (WITH SEAL IMPRESSION); RING STAND

Harappan sites have yielded a profusion of terracottas. There are figurines of animals such as bulls, buffaloes, monkeys, and dogs. There are toy carts with solid wheels. Human figurines include male figurines and more numerous female figurines of various types. The Harappan craftspersons also made terracotta bangles. Terracotta masks have been found at Mohenjodaro and Harappa. Faience is a paste made out of crushed quartz and coloured with various minerals. The Harappans made faience bangles, rings, pendants, miniature vessels, and figurines (including those of monkeys and squirrels). Another distinctive Harappan craft was the making of hard, high-fired bangles known as stone ware bangles. These were highly burnished red or grey-black, with a standard inner diameter of 5.5–6 cm, and usually had tiny letters written on them.

POTTERY DESIGNS

TERRACOTTAS: HUMAN AND ANIMAL FIGURINES; MASK; CIRCULAR AND TRIANGULAR CAKES

Stone work was another important craft. Reference was made earlier to the stone masonry and fine polished pillars at Dholavira. More visible at all Harappan sites were the mass-produced chert blades made by the crested guided ridge technique. Some of these may have been used as knives for domestic use, others as sickles. Harappan stone quarries have been identified in the Rohri hills of Sindh. Some of the stone blades may have been obtained from contemporary hunter-gatherer communities. The fact that stone flakes and cores occur in many houses at Mohenjodaro suggests that at least some of the tools were made by people in their homes.

The Harappan civilization is marked by a large number of copper objects. Apart from making artefacts out of pure copper, Harappan craftspersons alloyed copper with arsenic, tin, or nickel. Copper and bronze artefacts included vessels, spears, knives, short swords, arrowheads, axes, fishhooks, needles, mirrors, rings, and bangles. The axes were flat, without a shaft hole, and were probably hafted in a split and bound handle. The number of pure copper artefacts was far greater than alloyed bronze ones. Usually, tools like knives, axes, and chisels, which needed hardened edges, were alloyed. Alloys increased over time—for instance, at Mohenjodaro, bronze tools increased from 6 per cent to 23 per cent from the lower to the higher levels. The small proportion of alloyed objects compared to those of pure copper may suggest cultural preference rather than technological backwardness.

Sixteen copper furnaces were found at Harappa, and copper workshops were found at Lothal. A large amount of copper oxide was discovered in a brick-lined pit at Mohenjodaro. That metal objects were considered precious is clear from the fact that they were buried in hoards for safekeeping by their owners. One hoard found at Harappa consisted of a large cooking pot with a bronze cover. Inside were several types of copper tools and weapons, including various types of axes, daggers, spearheads, arrowheads, chisels, and a bowl. Some of the objects were unused, others used and worn.

Beautifully worked gold and silver jewellery including necklaces, bracelets, brooches, pendants, and earrings have been found at Harappan sites. A hoard of jewellery made of gold, silver, and semi-precious stones was found at the small village site of Allahdino. The Harappans used silver to emboss conch shells and to make vessels. Lead was used to make plumb bobs and in copper casting. It may be noted that two metal objects found at Lothal contain 39.1 per cent and 66.1 per cent iron. The latter can be called an iron object. What this suggests is that the Harappans (at least those of Gujarat) may have had some familiarity with iron smelting.

Seal making was another important Harappan craft. Most of the seals are square or rectangular. The average size of the square seals is about 2.54 cm, but there are larger ones, a little over 6.35 cm. Some have a perforated boss at the back for handling and suspension. A few cylindrical and round seals have also been found. Most of the seals are made of steatite, but there are a few silver, faience, and calcite ones as well. Two fine silver seals with the unicorn motif were discovered at Mohenjodaro, and some copper and soapstone ones were found at Lothal. To make the stone seals, the stone was sawed and shaped with knives, and then carved, using fine chisels and drills. The seal was coated with an alkali and heated, giving it a white lustrous surface. The carving is in intaglio— i.e., it is a sunken engraving, with the impression appearing in relief. Motifs include the elephant, tiger, antelope, crocodile, hare, humped bull, buffalo, rhinoceros, and the one-horned mythical

animal referred to as a unicorn. There is often a small feeding trough or stand in front of the animal. There are also composite animals, human figures, and plants. Most of the seals have a short inscription. Some rectangular seals have writing, but no motif.

CHERT BLADES; STONE GAMESMEN

COPPER ARROWHEAD AND CELT

STONE SEALING; SEAL

Bead making was a craft known in earlier cultures, but in the Harappan civilization new materials, styles, and techniques came into vogue. A new type of cylindrical stone drill was devised and used to perforate beads of semi-precious stones. Such drills have been found at sites such as Mohenjodaro, Harappa, Chanhudaro, and Dholavira. The Harappan craftspeople made beads out of steatite, agate, carnelian, lapis lazuli, shell, terracotta, gold, silver, and copper. The Harappan long barrel cylinder beads made out of carnelian were so beautiful and valued that they found their way into royal burials in Mesopotamia. Tiny micro-beads were made of steatite paste and hardened by heating. Beads were also made of faience.

FURTHER DISCUSSION

Sculpture in stone and metal

Apart from utilitarian items made of stone and metal, a few pieces of stone and metal sculpture have been found at Harappan sites. Most of them are small, but they display fine artistic skills and sensibilities. They include the stone bust (17.78 cm high) of a male figure found at Mohenjodaro, which has been labelled the ‘priest-king’. Two fine stone torsos of a male figure (about 10 cm high) were found at Harappa, a seated stone ibex or ram (49 × 27 × 21 cm) at Mohenjodaro, and a stone lizard at Dholavira. The only large piece of sculpture is that of a broken, seated male figure from Dholavira.

Two bronze female figurines were found at Mohenjodaro. One of them has become famous as the ‘dancing girl’. This figurine was found in a small house in the southwestern quarter of the city (in the HR area) during the 1926–27 excavations. The figure is 10.8 cm high and was made by the lost-wax method.

The lost-wax method involves first making a wax model and then covering it with a clay coating, leaving some holes as passageways. When the clay-covered moulds are heated in ovens, the wax melts out. Molten bronze is then poured in, and takes the place of the wax. When the mould has cooled, the outer clay envelope is chipped off and the craftsperson can then put the finishing touches to the solid bronze statue. This technique is still used in certain parts of India.

But to get back to the ‘dancing girl’: She represents a very thin woman standing with her right hand on the back of her hip and left hand resting on her left thigh, just above the knee. She may have once held some object in this hand. She is naked. She wears a necklace and has 24–25 of bangles on her left arm and just 4 on her right arm. Her arms are unnaturally long. Her head is tilted back, and she has a defiant, nonchalant air about her. Her hair is swept back in a low, loose bun at the nape of her neck. John Marshall named her the ‘dancing girl’ because he thought she had the air of a semi-impudent ‘nautch girl’, hand on hip, beating time to the music with her feet. The name has stuck. But the ‘dancing girl’ may not have been dancing at all, and even if she was, she may not represent a professional dancer.

THE ‘DANCING GIRL’

Bead making factories with tools, furnaces, and beads in various stages of preparation have been found at Chanhudaro and Lothal. At Bagasra in Gujarat, there is evidence of the production of artefacts of shell, faience, and beads of semi-precious stones (agate, carnelian, amazonite, lapis lazuli, and steatite). Clay-lined silos, varying from 0.30 to 1 m in diameter and 0.15 to 0.30 m in depth, were used to store semi-precious stones. The bead-making tradition in Gujarat today gives us clues on how the Harappan craftspeople may have made their beads.

Beads, bracelets, and decorative inlay work of shell show the existence of craftspersons skilled in shell working. Bangles were often made from conch shell. Chanhudaro and Balakot were important centres of shell work. Further evidence of site specialization comes from Gujarat. An intensive surface survey and excavations at Nageshwar (in Jamnagar district) have shown that this site was exclusively devoted to shell-working and specialized in making bangles. Evidence of shell working also comes from Kuntasi, Dholavira, Rangpur, Lothal, Nagwada, and Bagasra. This craft was clearly very important in the Gujarat region of the Harappan culture zone. Bone working was another specialized craft. Beads, awls, and pins were made out of bone. There are a few examples of ivory carving in the form of combs, carved cylinders, small sticks, pins, gamesmen, and a carved plaque.

SHELL LADLE, LOTHAL

JEWELLERY: NECKLACES OF CARNELIAN BEADS, GOLD; BANGLES OF TERRACOTTA, COPPER, STONEWARE, LAPIS LAZULI BEADS; GOLD SPIRAL PIN; GOLD AND TERRACOTTA BEADS

It can be inferred from the available evidence that the Harappans made cotton and woollen textiles. The terracotta figurines wearing clothes (shawls, skirts, etc.) reflect the kinds of clothes people wore. Mesopotamian texts mention cotton as one of the imports from Meluhha (an area which included the Indus valley). Traces of cotton cloth were found at Mohenjodaro, preserved over the centuries due to their being in contact with a corroding silver jar. Several examples of cotton thread and cloth were identified on copper tools. At Harappa, cotton threads were found wrapped around the handle of a small copper mirror in a burial and also around the handle of a curved copper razor. Recent excavations at Harappa have given evidence of woven textile impressions on the inside of faience vessels. The uniform thickness and uniformity of the weave suggest the use of spinning wheels. Various kinds of spindle whorls for spinning thread have been found at Harappan sites.

Weaving may have been a cottage industry practised in villages, and also to some extent in the cities. Impressions on clay floors and fired clay lumps suggest traditions of making baskets and mats out of reeds and grasses.

The Harappan crafts display an impressive level of standardization. Kenoyer (1998: 149–50) has suggested that state control may have been responsible for the high level of standardization in crafts that were considered to have a value in maintaining the socioeconomic or ritual order and which used non-local raw materials and highly complex technologies (e.g., the making of seals, stoneware bangles, and stone weights). Leaving aside pottery and bricks, crafts using local materials and simple technologies tend to show greater variation.

Standardization extended to units of weights and measure. Cubical weights made of chert, chalcedony, black stone, etc. have been found at all excavated sites, and their accuracy all over the Harappan culture zone is remarkable. The system is binary in the smaller weights (1:2:8:16:32:64) and decimal in the higher weights (with a ratio of 160, 200, 320, and 640). The largest weight found at Mohenjodaro weighs 10.865 g. A shell scale was found at Mohenjodaro and an ivory scale at Lothal; a shell object found in Saurashtra was probably used to measure angles.

STONE WEIGHTS, DHOLAVIRA

What is the explanation of the high level of standardization in crafts such as pottery-making and brick making? Does it imply centralized control by merchants or rulers? Some element of central direction is suggested, but its nature and degree are far from certain. If not direct, it may have taken the indirect form of facilitating or controlling the flow of at least some of the raw materials and finished goods. On the other hand, the level of standardization could also indicate the fanning out of hereditary craft specialists over large areas, or a well-developed network of internal trade. It is possible that craftsmen and traders may have been organized in corporate groups similar to guilds, but there is no proof of this.

NEW DIRECTIONS IN RESEARCH

The making of long carnelian beads

The city of Khambhat (Cambay) in Gujarat is one of the largest centres of stone bead- making in the world today. Mark Kenoyer, Massimo Vidale, and Kuldeep K. Bhan conducted an ethnoarchaeological study, examining the techniques used by modern bead makers of this place.

They supplemented this with experimentation and an analysis of the remains of bead manufacture at the site of Chanhudaro in south Pakistan. The results throw light on how the Harappan craftspersons may have made their beautiful long barrel cylinder beads. The process must have been something like this:

Long nodules of carnelian (a reddish orange variety of agate) were brought from Gujarat to Chanhudaro. The best were chosen and separated. These were dried in the sun for many months and then heated in shallow ovens to make the stone easier to work. The heating also deepened the red colour. The bead roughouts were made using a copper-tipped stake and an antler or horn hammer, using indirect percussion or pressure flaking techniques. Larger nodules were cut lengthwise and chipped to make bead roughouts. These roughouts were then partially ground on grooved sandstone or on quartzite grinding stones.

Then came the drilling of holes through the beads. This was done using a special cylindrical drill made out of a rare metamorphic rock which was heated to make an extremely hard and durable tool. This material has been given the name of ‘Ernestite’, after the archaeologist Ernest J. H. Mackay, who was the first to discover the drills and understand their significance. It could have taken a craftsperson a whole day of work—heating, chipping, and grinding—to make a drill. The Harappan bead makers used many different sizes of drills (at least six sizes) to make a single bead. The drilling was probably done with a hand-held bow drill. The friction would have produced intense heat, so the work may have been done under water, or at least by dripping water continuously on the drill hole.

The study conducted by Kenoyer and his team showed that even with these superior drills, it would have taken over 24 hours or three 8-hour days of steady drilling to perforate a single 6 cm long bead. The beads on the belts found at Mohenjodaro and Allahdino vary from 6 to 13 cm in length. It would have taken 3–8 days to make one of the longer beads, probably more, considering that the bead makers of Khambhat take long breaks after a couple of hours of work, as it is a very strenuous and tiring process. Once the beads were perforated, there was a laborious polishing process.

Taking the process from start to finish, it would have taken over 480 work days to make a belt of 36 beads of the kind found at Allahdino. Even if more than one worker was put on the job, it would still have taken up to a year. These beads must have been highly valued and worn only by the rich. For people who could not afford the expensive long carnelian beads, Harappan craftspeople made imitations in terracotta and painted them red.

Kenoyer, Vidal, and Bhan also analysed the archaeological patterns of manufacturing waste and finished artefacts, the structural evidence, and settlement layout in order to make inferences about the way in which bead manufacture was organized and controlled. Why did the Harappans transport carnelian nodules from Gujarat to Chanhudaro, instead of getting at least some of the preliminary work, such as discarding poor quality nodules, done near the source of the raw materials? The evidence suggests that all stages of carnelian bead manufacture at Chanhudaro were centralized and controlled by a powerful and wealthy group of merchants. This also

explains the uniformly good quality of the raw materials used and the high level of standardization. This is in contrast to evidence from the Moneer area at Mohenjodaro, which is suggestive of short-term production by several entrepreneurs.

SOURCE Kenoyer et al., 1995

Networks of Trade

The discovery of the Harappan civilization generated a great deal of interest in Harappan– Mesopotamian trade links. This is because before the advent of radiocarbon dating, these links gave vital clues for dating the Harappan culture, and also due to the prevailing interest in cross-cultural comparisons. Over the years, however, many scholars have come to the conclusion that Harappan– Mesopotamian trade may not have been as substantial as earlier held. Other areas such as the Persian Gulf have been identified as important zones of interaction as far as the long-distance trade of the Harappans is concerned. However, it is clear that trade networks within the Harappan culture zone and those linking the culture with other areas in the subcontinent were extremely significant; they are crucial for understanding the structure of the Harappan civilization as well as its striking level of cultural homogeneity. The importance of such trade is clear from the very wide range of raw materials and finished goods that found their way to different parts of the vast Harappan culture zone. This was an age before the advent of coinage, and the vibrant trade of the Harappans was based on barter.

One of the important aspects of Harappan trade is the identification of the sources of major raw materials used by the Harappans. The best way of doing this is to scientifically analyse the artefacts and to compare the results with raw materials from various possible sources. Unfortunately, there are not enough studies of this kind so far. Another method is to plot the location of the known resources of various raw materials, especially those closest to the Harappan culture zone. Proof that these were being used in protohistoric times would, of course, give clinching evidence. Unfortunately, this is not usually available, and the earliest evidence of the exploitation of these resources is often contained in 18th/19th century textual references. In spite of its limitations, this kind of exercise is useful in helping identify probable sources of raw materials used by the Harappans.

The discovery of factory sites in the limestone hills of Sukkur and Rohri indicates that chert blades were mass produced here and sent to various Harappan settlements in Sindh. The Khetri (< />)deposits of Rajasthan must have been an important source of copper. Reference was made in Chapter 3 to the links between the copper-manufacturing Ganeshwar–Jodhpura culture and the Harappan civilization. Lead and zinc probably also came from Rajasthan. Tin is available in the Tosam area of modern Haryana, but other possible sources are Afghanistan and central Asia. Gold may have come from the Kolar fields of Karnataka, where it may have been obtained via trade from the neolithic people who lived there. These neolithic herders may also have been exporters of cattle. (Fine disc beads, probably of steatite paste, found at Piklihal may have been obtained from the Harappans.) Gold could also have been panned from the sands of the upper Indus. Most varieties of semiprecious stone used for bead manufacture came from Gujarat. The exception is lapis lazuli, which was probably obtained from Afghanistan, although it also occurs in the Chagai hills in Baluchistan.

Traders must also have been engaged in a brisk trade in grains and other food products, transporting these between villages and cities.

Two-wheeled carts were an important mode of transport for people and goods. Bronze and terracotta models of carts have been found at various sites. No carts survive, but their tracks have been found at several sites, indicating spans roughly similar to those used today. Traders must also have transported their merchandise across long distances in caravans of pack animals such as oxen, sheep, goats, and donkeys. Towards the end of the mature Harappan phase, there is evidence of the use of the camel. The use of the horse seems to have been very minimal. Boats are depicted on seals and moulded tablets, and clay models have been found at Harappa and Lothal. River boats had cabins, ladders leading to the roof, and a high seated platform on the stern for navigation. Seafaring boats had a sharp keel, pointed prow, high flat stern, and mast and ropes for sails.

TERRACOTTA CART, HARAPPA

Several routes of trade and communication connected the various parts of the Harappan culture zone—Baluchistan, Sindh, Rajasthan, Cholistan, Punjab, Gujarat, and the upper doab. These routes can be reconstructed by studying the geographical landscape, settlement patterns, and the distribution of raw materials and finished products. Lahiri (1992: 112–43) points out that major trade routes connected the following areas: Sindh and south Baluchistan; coastal Sindh, upper Sindh, and the central Indus plains; the Indus plains and Rajasthan; the regions lying to the north of the Indus and Harappa; Sindh and east Punjab; east Punjab and Rajasthan; and Sindh and Gujarat. Some of the routes were already well defined in the early Harappan phase—e.g., the Baluchistan–Sindh route via the Kirthar mountains, and the route from east Punjab and Rajasthan via the Cholistan tract. The route connecting north Afghanistan, the Gomal plain, and Multan with a feeder route going to the Taxila valley also continued to be important. Certain routes that were being used in the earlier period became more important in the mature Harappan phase—e.g., the routes within Sindh, between Sindh and the central Indus plains, and between Sindh and Baluchistan via Kutch and Kathiawar. It is likely that the Indus saw a certain amount of riverine traffic. There was also a coastal route linking the Gujarat sites such as Lothal and Dholavira to sites such as Sutkagen-dor on the Makran coast. The location of some of the important sites can in fact be explained in relation to the trade routes of the time. For instance, Mohenjodaro lay at the intersection of the water-route of the Indus and the east– west land route that linked the Quetta valley and the Bolan river to Kot Diji and the western Nara.

MAP 4.3 HARAPPAN ROUTES OF INTERNAL TRADE (AFTER LAHIRI, 1992)

The main sources of information on long-distance trade include a number of Harappan or Harappan-related (i.e., similar to Harappan types) artefacts found at sites outside the subcontinent, and foreign objects found at Harappan sites. These are supplemented by textual sources in the case of Indus–Mesopotamian trade (see Chakrabarti, 1990).

A number of Harappan and Harappan-related objects have been found in south Turkmenistan at sites such as Altyn Depe, Namazga, and Khapuz. These include ivory dice, two types of metal objects (a spearhead and ladle), an ithyphallic terracotta, perforated ware, a segmented bead, and a silver seal. The most definite evidence comes from Altyn Depe, in the form of a rectangular Harappan seal bearing the Harappan script. The sites in Iran which have yielded Harappan and Harappan-related artefacts are Hissar, Shah Tepe, Kalleh Nisar, Susa, Tepe Yahya, Jalalabad, and Marlik. The main evidence consists of seals and carnelian beads (both the etched and long barrel cylinder types). The most important evidence of trade with Afghanistan comes from an isolated Harappan trading outpost at Shortughai.

Many years ago, a round seal with a short-horned bull motif and Harappan writing was found at Failaka in the Persian Gulf. In recent years, there has been a substantial increase in the evidence of Harappan trade contacts with the Persian Gulf area. Harappan and Harappan-related artefacts (including a piece of ivory, a linga-shaped object, a circular mirror, and seals with Harappan motifs and/or writing) have been found at Rasal-Qala on the island of Bahrain. Excavations near Hamad in Bahrain yielded a typical Harappan seal and carnelian beads in burials. A seal with the bull motif and Harappan script was found at the site of Hajjar. From Failaka, apart from the ‘Persian Gulf seal’ mentioned above, there was a flat, round seal with the Harappan script. Jar fragments with Harappan writing have been found at many sites in the Persian Gulf. These were probably containers used to transport perishable goods from the Harappan culture zone to this region.

The Harappans were also trading with the Oman peninsula. An etched carnelian bead of the Harappan type was found at Umm-an-Nar. There are similarities between certain other types of objects found at this site (a square steatite seal, fragments of pottery, carnelian beads, a cubical stone weight, etc.) and Harappan artefacts. Maysar, an excavated copper-smelting site, has yielded evidence (e.g., pottery decorations and motifs on a seal) that suggests Harappan influence. The major imports from Oman may have included chlorite vessels, shell, and perhaps mother-of-pearl. Copper has been mentioned as another Omani export to the Harappans, but this is unlikely, as the metal was available closer, in Rajasthan. As for Harappan exports to Oman, the items that survive in the archaeological record include beads, chert weights, and ivory objects.

There is literary as well as archaeological evidence for Harappan trade with Mesopotamia. Mesopotamian records of the time of king Sargon (2334–2279 BCE) refer to ships from the lands of Dilmun, Magan, and Meluhha tied along the quay of the capital city, Akkad. Dilmun can be identified with Bahrain, and Magan with the Makran coast and Oman. Meluhha may have been a generic term for areas lying to the east of Mesopotamia, including the Indus valley, or it may refer specifically to the Indus valley. The archaeological evidence for Harappan–Mesopotamian trade consists mainly of a few Harappan or Harappan-related seals and carnelian beads at Mesopotamian sites such as Kish, Lagash, Nippur, and Ur. Carnelian beads (both the etched type and the long barrel-cylinder type) were also found in the royal graves at Ur. Certain motifs such as the bull on Mesopotamian seals have been cited as reflecting Harappan influence. Cylinder seals (which are common in West Asia) with Harappan-type motifs suggest interaction between merchants of these two areas. The absence of Mesopotamian seals and sealings in the Harappan context suggests that Mesopotamian traders were not directly involved in the Harappan–Mesopotamian trade interactions.

MAP 4.4 LONG-DISTANCE TRADE ROUTES

Carnelian beads were clearly an important Harappan export to West Asia. Textiles and conch shell objects were other possible exports. Ivory and ivory objects may have been exported by the Harappans to Afghanistan, Turkmenistan, and perhaps the Persian Gulf. Mesopotamian texts mention the following items as imports from Meluhha: lap1/22/2016is lazuli, carnelian, gold, silver, copper, ebony, ivory, tortoiseshell, a chicken-like bird, dog, cat, and monkey. Mesopotamia’s general exports included fish, grain, raw wool, woollen garments, and silver. It is possible that wool and silver found their way to Meluhha, but there is no archaeological proof of this.

There are two very different assessments of Harappan–Mesopotamian trade. Ratnagar (1981) highlights the importance of this trade, especially the trade in lapis lazuli, and even argues that its decline was a reason for the decline of the Harappan civilization. Notwithstanding the long list of items mentioned in texts, the fact remains that there are very few Harappan artefacts found in Mesopotamia and even fewer Mesopotamian artefacts found at Harappan sites. A few Mesopotamian-type stone weights have been reported from Mohenjodaro and Harappa. Three motifs found on some Harappan seals are seen by some scholars as reflecting Mesopotamian influence—the whorl design, a man grappling with two animals, and the gatepost motif. The evidence as a whole is not very substantial. Chakrabarti (1990) and Shaffer (1982b) argue that Harappan trade with Mesopotamia was not direct, extensive or intensive. This trade does not seem to have been particularly important for the development or sustenance of the Harappan civilization.

Among the Harappan imports via long-distance trade, lapis lazuli was probably an import from Afghanistan (or it could have been obtained closer from the Chagai hills of Baluchistan). Jade must

have come from Turkmenistan. Tin may have been obtained from Ferghana and eastern Kazakistan in central Asia. Carved chlorite and green schist vessels were a popular item of trade in West Asia and the Persian Gulf, and a few fragments have been found at Mohenjodaro. These may have been imported from southern Iran or from Baluchistan. Very few West Asian artefacts have been found in Harappan contexts. A seal of the Persian Gulf type was found at Lothal as a surface find. A lapis lazuli bead from Mohenjodaro and a pendant with lapis lazuli inlay found at Cemetery-H levels at Harappa were possibly imports from West Asia. A cylinder seal (as mentioned earlier, cylinder seals were common in West Asia) with Indian motifs was found at Kalibangan.

Harappan objects in Mesopotamia can be dated from the Early Dynastic IIIA period (c. 2600/2500 BCE) to the Isin–Larsa period (c. 2000/1900 BCE) in the Mesopotamian sequence, which corresponds to the entire span of the mature Harappan phase. The finds from other parts of West Asia also belong roughly to this period. However, the discovery of a Harappan seal at the site of Nippur in a 14th century BCE context suggests that Harappan contact with Mesopotamia may have continued, although in a diminished form, into the late Harappan phase. The continuation of some amount of trade with the Persian Gulf region is suggested by two Harappan seals found at Failaka in a 14th century BCE context, and a late Harappan seal found at Bet Dwarka. The latter has Harappan writing and a three-headed animal motif similar to that found on certain Persian Gulf seals.

The importance of overland routes from the Harappan civilization through Afghanistan is evident from the location of Harappan sites near each of the passes and routes that lead through Baluchistan into Afghanistan. Pathani Damb is near the Mula pass, Nausharo near the Bolan pass, Dabarkot in the Gomal valley, and Gumla and Hathala in the Derajat, along the route via the Gomal pass. The Gomal route seems to have been the most important.

Two main overland routes connected the Harappan civilization with West Asia. The northern one passed through northern Afghanistan, north Iran, Turkmenistan, and Mesopotamia, crossing sites such as Shortughai, Tepe Hissar, Shah Tepe, and Kish. A southern route passed through Tepe Yahya, Jalalabad, Kalleh Nisar, Susa, and Ur. The maritime route to Mesopotamia may also have been used. It is likely that sites such as Sutkagen-dor, Balakot, and Dabarkot (the latter two may at that time have been located at the coast instead of some distance away) were important points along this route. Lo-thal (10 km away from the Gulf of Cambay) and Kuntasi (on the Phulki river, 4 km from the coast), Dholavira (in the Rann of Kutch), and the sites along the coast of Kutch no doubt played an important role in maritime trade.

FURTHER DISCUSSION

Shortughai—a Harappan trading post in Afghanistan

Shortughai is located near the confluence of the Oxus and its tributary, the Kokcha, in north-east Afghanistan. It is a small site, only about 2 ha. The cultural deposit is 2.5–3 m thick, within which four periods of occupation have been identified. Period I (50 cm thick) was dated by radiocarbon to the end of the 3rd millennium BCE.

The discoveries of Period I included the following: pottery with Harappan designs, terracotta cakes, fragments of toy carts, copper and bronze objects, pieces of gold and lead, a discoidal

gold bead; lapis lazuli, agate, carnelian, steatite, small barrel-shaped agate beads; long tubular and etched carnelian beads; flint micro-blades and drill heads; shell bangles; and mud-bricks of the typical Harappan size. Harappan graffiti occurred on the rims of jars and on beakers. There was a square Harappan seal with the motif of a rhinoceros and the Harappan script. The discovery of so many typical Harappan artefacts and manufacturing techniques proves that this was not a site which had mere contact with the Harappan civilization, but a site belonging to the Harappan civilization.

Shortughai also has some unique features. A ploughed field covered with flax seeds was found in an area unsuitable for irrigation, showing the practice of dry farming. Small irrigation canals drawing on the water of the Kokcha, located about 25 km away, were found in other parts of the site.

What were the Harappans doing at Shortughai? This site seems to have been connected with the lapis lazuli mines nearby. However, lapis lazuli objects are not particularly numerous at Harappan sites. A second possibility is that Shortughai owed its importance to its proximity to the tin mines of Afghanistan and Ferghana. A third possibility is that it had a role to play in camel trade.

SOURCE Chakrabarti, 1990: 1–2, 86–89

The argument that the quantum of Harappan long-distance trade was not great is persuasive. Unlike the resource-poor area of Mesopotamia, the Harappan culture zone was rich in a variety of natural resources. Food requirements and most of the raw materials required by Harappan craftspersons could have been met by resources available within the Harappan culture zone. The diverse, well-developed craft traditions meant that most of the finished goods required by the Harappans were likewise available from within this area. A few raw materials and products were obtained from other parts of the subcontinent and from areas such as Afghanistan and central Asia. Very few essential items had to be imported from distant places. Harappan trade must have involved highly organized merchant groups as well as nomadic peddlers in the mountainous stretches. The extent of state control over this activity is a matter of debate.

The Nature and Uses of Writing

Among the biggest mysteries about the Harappan civilization are the language (or languages) the Harappans spoke and their writing system. It is likely that people living in various parts of the Harappan culture zone spoke different languages and dialects. The writing on the seals was probably in the language of the ruling elite. Some scholars have suggested that this language belonged to the Dravidian family of languages, while others have argued in favour of the Indo-Aryan family. However, there is so far no consensus on the affiliation of the Harappan language or on the decipherment of the script.

A total of about 3,700 inscribed objects have been found at Harappan sites (for details, see Mahadevan, 1977, Parpola, 1994). Most of the writing appears on seals and sealings (seal

impressions), some on copper tablets, copper/bronze implements, pottery, and other miscellaneous objects. About 50 per cent of the inscribed objects have been found at Mohenjodaro, and the two sites of Mohenjodaro and Harappa together account for about 87 per cent of all inscribed material. Most of the inscriptions are very short, with an average of five signs. The longest one has 26 signs. The script seems to have emerged in a fully evolved state and does not show any significant changes over time. This conclusion may, however, be the result of the inadequacies of earlier excavations, which did not record the stratigraphic context of all objects, making it difficult to sort out earlier and later samples of writing.

THE DHOLAVIRA 'SIGNBOARD'

There are 400–450 basic signs and the script is logo-syllabic—i.e., each symbol stood for a word or syllable. It was generally written and meant to be read from right to left (this is reversed on the seals). This is evident from that fact that in inscriptions, the letters are cramped on the left side, where space had clearly run out, and from overlapping letters scratched onto pottery. There are a few instances, however, of writing from left to right. Longer inscriptions that consisted of more than one line were sometimes written in the boustrophedon style—with consecutive lines starting in opposite directions.

What was the connection between the motifs on the seals and the writing? What was the extent of literacy among the Harappans? What was writing used for? In order to understand the uses of writing in the Harappan civilization, it is necessary to try to interpret the functions of the inscribed objects. Writing appears very frequently on the seals. Some of these were impressed onto small moist clay tablets known as sealings, probably by merchants to authenticate their bales of merchandise. The evidence of textile impressions on some sealings supports this interpretation. However, more seals than sealings have been found, and the seals are generally worn at the edges and not inside. This suggests that some of the so-called seals may have had other functions. They may have been tokens used in the buying and selling of goods. They may also have been worn as amulets or used as identification markers (like modern identity cards) by well-to-do people like landowners, merchants, priests, artisans, and rulers. Those no longer in use must have been intentionally broken so that they could not be misused by anybody. Tablets with narrative scenes may have had a religious or ritualistic function. The so-called ‘seals’ were thus used for multiple purposes.

Writing also appears on miniature tablets made of steatite, terracotta, and faience. Since these objects were not used to make impressions, unlike the seals, the writing on them was not reversed. Many of the objects were discovered at Harappa and other large cities. Rectangular copper tablets with writing and animal motifs were found at Mohenjodaro, while a few tablets with raised writing were found at Harappa. The limited number of places where they occur suggests a restricted use. Interestingly, there are many duplicates of both the miniature and copper tablets.

The evidence of writing on pottery suggests a wider use in craft production and economic transactions. Harappan potters sometimes inscribed letters onto pots before firing. At other times, inscriptions were made on pots after they were fired (this is termed ‘graffiti’). Even if the potters who made the marks on their pots were themselves illiterate, they must have been able to recognize

the symbols. Pointed goblets sometimes have seal impressions, which may have indicated the name or status of the person for whom the pot was made.

Items like copper and bronze tools, stoneware bangles, bone pins, and gold jewellery were sometimes inscribed. A copper vessel found at Mohenjodaro contained a large number of gold objects. These included four ornaments with tiny inscriptions, all apparently written by the same hand, probably giving the name of the owner. Some of the writing inscribed or painted on personal possessions such as bangles, tools, beads, and bone rods may have had some sort of magico-religious or ritualistic significance.

The Dholavira ‘signboard’ may or may not indicate a high level of urban literacy, but it does indicate a civic use of writing. It is likely that a very small proportion of Harappan written material survives, and that people wrote on perishable material as well. The evidence of a common script all over the vast Harappan culture zone shows a high level of cultural integration. The virtual disappearance of the script by c. 1700 BCE suggests both a close connection of writing with city life and the lack of sufficient downward percolation of writing.

HARAPPAN SEALS

THE SCRIPT–THE 'JAR' SIGN AND THE 'MARKER' SIGN ARE THE MOST FREQUENTLY OCCURRING SYMBOLS

Religious and Funerary Practices

The basic elements of what can be loosely described as ‘Harappan religion’ were outlined by John Marshall in 1931. Although some aspects of Marshall’s interpretation can be criticized—especially his tendency to read elements of later Hinduism into the evidence—he did succeed in identifying several important features of Harappan religion. Hypotheses about this issue are bound to be subjective, especially in view of the fact that the script is undeciphered.

The worship of female goddesses associated with fertility has long been held as one of the major features of Harappan religion. This conclusion is based on the following factors: (a) the concerns that agricultural societies are invariably known to have with fertility; (b) cross-cultural parallels with other ancient civilizations; (c) the importance of goddess worship in later Hinduism; and (d) the discovery of a large number of terracotta female figurines that were labelled ‘Mother Goddesses’. Certain representations on seals are also relevant. For instance, a seal showing a nude woman, head downwards, with her legs apart and a plant issuing from her vagina is often interpreted as a prototype of Shakambhari, the Earth Mother.

Describing all female figurines as representations of a single great ‘Mother Goddess’ associated with fertility and maternity clearly over-simplifies the situation. The attributes of the figurines and the contexts in which they were found have to be considered carefully before assigning them a religious or cultic significance. As pointed out in an earlier chapter, not all female figurines necessarily represented goddesses (let alone a single goddess), and not all goddesses necessarily

had maternal associations. Some of the Harappan female figurines may have had a cultic significance and may have been part of household rituals. Others may have been toys or decorative items.

A study of the Harappan terracottas by Alexandra Ardeleanu-Jansen (2002) has underlined the great variety in the form of female figurines. The type which is frequently interpreted as having a religious significance is a slim female figure with a distinctive fan-shaped headdress, wearing a short skirt. She is heavily ornamented with necklaces, armlets, bangles, anklets, and earrings. Some of the figurines have cup-like attachments and flowers on either side of the head. In certain cases, the cup-like attachments have traces of black residue, suggesting that they were used to burn oil or some sort of essence. Such figurines may have been religious images worshipped in households, votive offerings made to a deity, or part of the paraphernalia of domestic rituals. It is interesting to note that such figures do not appear on Harappan seals and tablets or in stone or metal sculpture.

There is also a matronly, pot-bellied type of female figurine who may represent either a pregnant woman or a prosperous woman. She is naked and sometimes wears some jewellery and a turban or head-dress. Both the ‘matronly type’ and the ‘slim type’ of female figurines may hold a baby in their arms. The ‘matronly type’ can stand without support, while the youthful, ‘slim type’ needs support. It is interesting to note that female figurines—including those with possible religious significance—are found in large numbers at sites such as Mohenjodaro, Harappa, and Banawali, but not at sites such as Kalibangan, Lothal, Surkotada or Mitathal.

Most of the terracotta figurines (including the female ones) were found broken and discarded in secondary locations. None were found in a context that could be interpreted as a temple. This was one of the reasons why Marshall suggested that they were votive offerings rather than cult images. The fact that so many of them were broken suggests that they may have been part of a ritual cycle and were made for short-term use for certain specific occasions. The relationship between the female figurines and the male and animal figurines with which they are associated needs to be explored.

FEMALE FIGURINE WITH FAN-SHAPED HEADDRESS

Marshall suggested that the Harappans also worshipped a male god represented on a steatite seal discovered at Mohenjodaro, usually referred to as the Pashupati seal. This shows a male figure with a buffalo horn head-dress seated on a dais with his legs bent double under him, heels together, toes pointed down. His outstretched arms are adorned with bangles, his hands rest lightly on his knees. He is flanked by four animals—an elephant, rhinoceros, water buffalo, and tiger. Beneath the dais are two antelopes or ibexes. Marshall thought the male figure was three-headed and ithyphallic (with erect penis). He saw a striking resemblance between this deity and the Shiva of later Hindu mythology, who is also known as Mahayogi (the great yogi) and Pashupati (lord of the animals).

FEMALE FIGURINE, BANAWALI

Another aspect of the fertility-related beliefs of the Harappans was the worship of male and female creative energy in the form of stone icons of lingas and yonis (representing the male and female sexual organs respectively). A number of such stones were identified by John Marshall. Many years later, George Dales argued that the contexts in which these stones were found do not suggest cultic significance. Some of the ring stones had lines on them and may have had architectural use, either to guide masons in pillar building or to measure angles. Alternatively, they may have been used to make astronomical calculations. Marshall himself had suggested that some of the linga-shaped objects may have been grinders or unfinished weights. Dales made his arguments forcefully; however, a terracotta piece which closely resembles a linga with a yoni-pitha (yoni base) has recently been found at Kalibangan.

The Harappan seals, sealings, amulets, and copper tablets depict a number of trees, plants, and animals, some of which may have had cultic significance. The pipal (Ficus religiosa) tree appears often and may have been venerated. Sometimes, there is a figure peering out from between its branches, possibly a tree-spirit. A seal found at Mohenjo-daro shows a row of seven figures with long braids standing in front of a pipal tree which has a horned figure standing in it. It is not clear whether the figures are male or female, but because they are seven in number, scholars have speculated that there may be a connection with the later traditions of the seven rishis or the seven mothers.

Some of the animals depicted on seals and sealings—for instance, the humped and humpless bull, snake, elephant, rhinoceros, antelope, gharial, and tiger—may have had cultic significance. The bull, a symbol of male virility in many ancient cultures, seems to have been particularly important. We can note the steatite bull statuettes discovered at certain sites, including a very sophisticated terracotta bull found at Mohenjodaro. It is possible that some of the terracotta animals on wheels may have been cult images rather than toys. Two Harappan sealings appear to represent animals being carried in processions; one of them resembles a bull or cow. The composite animals (tiger– human, bull–elephant, ram–bull–elephant, etc.) and the ‘unicorn’ depicted on some seals and sealings may also have had some sort of religious or mythological significance. Some of the terracotta, shell,

faience, and metal tablets may have been amulets. Their motifs, such as the svastika, may have been associated with a protective function or auspiciousness. Terracotta masks and puppets found at Mohenjodaro and Harappa include those in the form of real and mythical animals, and these may have been used in religious, political, or politico-religious rituals.

Man, god, or goddess?

THE 'PASHUPATI SEAL'

Marshall concluded that this seal showed that the Harappans worshipped a god who seems to have been a proto-Shiva. This conclusion has not gone unchallenged. The questions that have been asked include the following:

Is1.the figure really sitting in a yogic posture of ritual discipline?

Is2.he really three-headed?

Is3.he ithyphallic?

Is4.the figure a male?

Shiva5. as Pashupati in later Hindu mythology protects domesticated cattle, while the figure on the seal is associated with wild animals. In view of this difference, can the two really be connected?

The figure has been variously identified as a chieftain, a divine bull-man, Indra, or the demon Mahisha of the Puranas. M. K. Dhavalikar and Shubhangana Atre (see Atre, 1985–86) have suggested that it represents a goddess—a ‘lady of the beasts’. Notwithstanding all these alternative interpretations, the basics of Marshall’s interpretation are still persuasive. The figure can be accepted as that of a male seated in a yogic posture, although it is not certain that he was three-headed. The similarities between the deity—for he seems to be no ordinary man—and certain attributes of the later-day Shiva remain striking. Of course, we do not know what name the Harappans gave him.

We can recall here the ‘horned deity’ that appears on a Kot Diji pot, Kalibangan terracotta cake, and the Padri jar. This indicates that the worship of a horned deity goes back to the early

Harappan phase.

FURTHER DISCUSSION

The ‘fire altars’

The citadel complex at Kalibangan consists of a northern and southern unit, separated from each other by a wall. In the southern sector, archaeologists found five or more mud-brick platforms, separated from each other and from the back of the fortification wall by streets. Steps or ramps led up to the platforms. On one of these platforms, there was a row of seven clay-lined pits, each about 75 × 55 cm. These have been identified as ‘fire altars’, i.e., pits in which offerings were made into the fire as part of sacrificial rituals. Ash, charcoal, the remains of a rectangular clay piece, and terracotta cakes were found in them. To the west of this row of pits, within easy reach of whoever sat in front of them, was the lower half of a jar containing ash and charcoal, embedded into the ground. Nearby was a well and the remains of bath pavements with attached drains, all made of burnt bricks. A ‘fire altar’ and a well were discovered on another platform in the southern sector of the citadel complex. There was also a 1.25 × 1 m brick-lined rectangular pit, containing cattle bones and antlers. This suggests the practice of animal sacrifice. The southern sector of the Kalibangan citadel complex seems to have been a place where sacrificial rituals of a congregational character were performed. The northern part of the citadel complex contained houses. B. B. Lal suggests this may have been where the priests who performed the rituals lived.

‘Fire altars’ have also been reported at Banawali, Lothal, Amri, Nageshwar, and Vagad in Gujarat and at Rakhigarhi in Haryana. But it is only at Kalibangan and Banawali that they may have signified some community event; in the other cases, they seem to have been associated with domestic rituals. Again, as in the case with female figurines, the fact that the ‘fire altars’ have been found at a few sites but are absent at most, indicates variations in religious practice within the vast area of the Harappan culture.

SOURCE Lal, 1984

The Great Bath was probably the scene of an elite ritual activity involving ceremonial bathing. A triangular terracotta cake found at Kalibangan has a carving of a horned deity on one side and an animal being dragged by a rope by a human on the other. The latter has been tentatively interpreted as suggesting the practice of animal sacrifice. A Kalibangan cylinder seal shows a woman flanked by two men who hold her with one hand and raise swords over her head with the other; this may represent a scene of human sacrifice. The most striking evidence suggesting ritualistic practices comes from the ‘fire altars’ found on the citadel mound at Kalibangan.

Harappan cemeteries have been located at sites such as Harappa, Kalibangan, Lo-thal,

Rakhigarhi, and Surkotada. The most common method of burial was to place the body of the

deceased in an extended position, with the head towards the north, in a simple pit or brick chamber. Grave goods including food, pottery, tools, and ornaments were placed along with the body, but they were never too many or lavish. Clearly, the Harappans preferred to use wealth in life rather than bury it with their dead. At Harappa, there was a coffin with a shroud made of reeds. Symbolic burials with grave goods but no skeletons were found at Kalibangan. Fractional burials (where the body was exposed to the elements and the bones then gathered and buried) were found at Mohenjodaro and Harappa. These two sites also gave evidence of urn burials suggestive of cremation. Multiple burials of men and women were discovered at Lothal.

The religious and funerary beliefs and practices of the Harappans show great variety. While there are dangers in viewing these through the lens of later-day Hinduism, it is interesting to note that the Harappan civilization does display a few features reminiscent of later traditions, except, however, the important element of temple worship. Not a single structure found at any Harappan site can conclusively be identified as a temple.

HARAPPAN SEALS WITH DEPICTIONS OF TIGER AND ELEPHANT

The Harappan People

What did the Harappan people look like? What sorts of clothes and ornaments did they wear? How did they relax and have fun? Terracotta, stone, and bronze sculptures (some of which have been described in earlier sections) help answer such questions. The form of human terracotta figurines was connected to their function, stylistic conventions, and audience, and they may not be realistic representations of what all or even most Harappans looked like. Nevertheless, they do help insert three-dimensional people into our picture of the Harappan civilization.

The human terracottas can be divided into female and male figurines, those whose sex is not clear, a few that have both female and male attributes (e.g., a figurine from Harappa which has breasts and

a beard), and a few males in feminine dress. Going by the figurines, Harappan women wore a short skirt made of cotton or wool. They wore their hair variously in braids, rolled into a bun at the back or side of the head, arranged in separate locks or ringlets, and wrapped around the head like a turban, or left loose. What looks like a fan-shaped headdress could actually represent hair stretched over a frame made of bamboo or some other material. At Harappa, it is supplemented by flowers or flower-shaped ornaments. Such hairstyles or headdresses could indicate women of distinction or deities. Female figurines wear ornaments such as necklaces, chokers, hair ornaments, bangles, and belts. We can recall the beautiful jewellery found at many Harappan sites.

Male figurines are usually bare headed, though some are turbaned. Most of them are nude, so it is difficult to say what sort of clothes men wore. Certain stone sculptures suggest the use of a dhoti-like lower garment and an upper garment consisting of a shawl or cloak worn over one shoulder and under the other. There are various hairstyles—braids, buns, and hair hanging loose. Most of the male figurines have beards, in styles ranging from the ‘goatee’ to the more common combed and spread-out style as in the case of the ‘priest-king’. There is some degree of overlap in male and female hairstyles and ornaments, but also some differences. For instance, men and women both wear bangles and necklaces, but men rarely wear multi-strand necklaces made of graduated beads.

Children of all cultures and all times play with toys, and Harappan children were no exception. Terracotta toys of various kinds have been found at Harappan sites. They include balls, rattles, whistles, gamesmen, carts with moveable parts, and animals on wheels. There are spinning tops made of terracotta and shell. Some have a shallow depression, while others have a copper tip to make them spin around a long time. Clay marbles have been found in courtyards of houses. Miniature terracotta cooking vessels, beds, and other toy furniture have been found, with which children must have played house. There are figurines of children playing with toys. One of them holds what seems to be a clay disc. Many clay discs have in fact been found at Harappan sites, and it is possible that these are remnants of a pithu-like game played with a ball and piled-up pieces of clay or stone. Lots of terracotta figurines of dogs have been found at Harappan sites, some with collars, suggesting that people kept dogs as pets. Some of the terracotta figurines of people and animals have a comic appearance, reflecting a sense of humour.

The social implications of the worship of female deities are complex. Although such worship reflects the ability to visualize divinity in feminine form, it does not necessarily translate into power or a high social position for ordinary women. While some of the female figurines found at Harappan sites may represent goddesses, many seem to represent ordinary, mortal women. Terracotta figurines of women at work are few. Figurines depicting women grinding or kneading something (food/clay?) have been found at Nausharo, Harappa, and Mohenjodaro, suggesting the association of women with food-processing activities. In ancient societies, childbirth was a process fraught with danger. Some of the fat female terracotta figurines may represent pregnant women. Recent excavations at Harappa have yielded a burial with a woman and baby, perhaps a case of death in childbirth. Some female figurines found at Harappan sites carry a suckling infant on the left hip; others show women carrying infants close to their breast. An unusual terracotta figurine found at Nausharo (Period ID) shows a male with feminine headdress holding an infant. Tiny terracotta figurines of small children have been found at most sites. Were all of them toys or could they be votive objects? Can a statistical analysis of the child figurines help us identify whether there was a cultural bias in favour of male or female children? This is a very interesting question, but answers can only be speculative.

TERRACOTTAS: FIGURINE; GAMES AND DICE

PERFORATED BIRD-SHAPED RATTLE; BULL WITH MOVEABLE HEAD; CART

NEW DIRECTIONS IN RESEARCH

How healthy were the Harappans?

The early excavations at Harappa focused on architecture and artefacts. The more recent excavations carried out during the 1980s and 1990s reflect the advances in the field of archaeology and included a careful collection and scientific analysis of bone remains. The results give us important information about the health and nutrition of the Harappans.

Cemetery R-37 is located in the southern part of the site. Excavations were carried out under the supervision of J. M. Kenoyer. A team of four physical anthropologists—K. A. R. Kennedy, John R. Lucacs, Nancy Lovell, and Brian Hemphill—had the special job of carefully excavating the skeletons and removing them to the laboratory for analysis. Ninety skeletons were recovered from the cemetery. Most of them represented females. The number of skeletons in different age ranges were as follows:

Children (< 16="" yrs)="" />

:

15

Young adults (17–34 yrs)

:

35

Middle-aged adults (35–55 yrs)

:

27

Older adults (> 55 yrs)

:

13

The general health of this sample of the Harappan population was quite good. The skeletons showed a low incidence of traumatic injury, chronic infectious diseases, and neoplastic diseases (tumours). There were no traces of nutritional inadequacy such as rickets, scurvy, or anaemia. There were, however, three cases of arrested growth lines, suggesting that growth during childhood was halted temporarily. This could have been due to malnutrition or some serious illness. The most common ailment suffered by the people buried in this cemetery was arthritis. Signs of this appeared in the spine and in the joints of knees, hands, and feet. There were also several instances of severe arthritis in the neck, which may have been the result of unusual stress on the neck vertebrae, perhaps due to carrying heavy loads on the head over a long period of time.

The teeth of the people were analysed and the dental pathology profile was what would be expected in a community of agriculturists. The most common dental problem was gross enamel hypoplasia (pitted or missing enamel) and the least common was hypercementosis (excessive deposit of cementum, a calcified hard tissue covering the root surface). Dental caries (cavities) were present in 43.6 per cent of the individuals examined. The dental caries rate was worked out as 6.8 per cent, which is a high rate typical of agricultural groups. Tooth loss, calculus (hardened plaque or tartar), and alveolar resorption (wasting away of the bony socket) occurred with moderate frequency. There were differences between males and females in the incidence of tooth loss and enamel hypoplasia. But the frequency of dental abscesses, calculus, and alveolar resorption were more or less the same for men as for women.

The study showed that the Harappans buried in Cemetery R-37 were relatively healthy agriculturists. A statistical analysis of the crania of the skeletons shows biological similarity among the people buried in the cemetery, a similarity between them and the skeletons found in the late Harappan Cemetery-H, and with the modern populations inhabiting this area today. This shows a broad biological continuity between the inhabitants of the area from mature Harappan to late Harappan into more recent times.

SOURCE Dales and Kenoyer, 1991: 191–99, 210–12

Early studies of Harappan skeletons focused on classifying the Harappans into racial types. More recent studies have abandoned the old, rather arbitrary racial classifications. They have asked different questions and given an interesting set of conclusions. Kenneth A. R. Kennedy’s study (1997) of skeletons found at Harappan sites shows biological heterogeneity between the different regions, and similarity with the people who live in these areas today. This means that the Harappans of Punjab resembled the present-day Punjabis in appearance, while the Harappans of Sindh resembled the modern inhabitants of Sindh. Kennedy also identified the incidence of malaria among the Harappans.

There is the larger question of the analysis and assessment of the structure of Harappan society. The absence of deciphered written evidence is a major handicap, and inferences have to be made very carefully on the basis of archaeological data. The people who lived within the Harappan culture zone comprised villagers and city folk. Harappan society included occupational groups such

as farmers, herders, hunter-gatherers, craftspeople, fisherfolk, merchants, sailors, rulers, administrative officials, ritual specialists, architects, carpenters, brick masons, well diggers, boat makers, sailors, sculptors, shopkeepers, sweepers, garbage collectors, and so on. Some farmers may have lived in the cities and tilled their fields nearby. Terracotta net sinkers and arrow points found at Mohenjodaro and Harappa suggest that the city population included hunters and fisher-folk. The level of social differentiation may not have been as great as in Mesopotamia and Egypt, but differences in house sizes and the hoards of jewellery do indicate a concentration of wealth and differences in social and economic status. The affluent social groups would have comprised rulers, land owners, and merchants. Class and rank differences based on occupation, wealth, and status must have existed. However, claims that the caste system existed in Harappan society are highly speculative.

The Ruling Elite

Political organization includes a range of issues related to the exercise of power and leadership in a society. The debate on the nature of the Harappan political system has focused largely on whether or not a state existed, and if so, what sort of state it was. A great deal depends on our definition of a state and the interpretation of the archaeological evidence. Cultural uniformity does not necessarily mean political unification; therefore there is the additional question of whether the evidence suggests the existence of one state or many.

Many scholars have observed that the elements of warfare, conflict, and force in the Harappan civilization seem weak compared to contemporary Mesopotamia and Egypt. Weapons are not a dominant feature of the artefacts found at Harappan sites. There are few depictions of conflict between people in the narrative reliefs on terracotta and faience tablets. However, fortifications, especially the imposing ones at sites such as Dholavira, cannot be overlooked. It is indeed possible that the element of force in the Harappan culture has been underestimated. Force and conflict could not have been completely absent in such a large area over such a long period of time.

That the Harappan civilization lasted for some 700 years and its artefacts, traditions, and symbols seem to have continued more or less unchanged through this long period, suggests a strong element of political stability. There must have been groups of rulers in the various cities. Just who they were and how they were related to each other remains a mystery. These groups would have been responsible for the maintenance of the city facilities—walls, roads, drains, public buildings, etc. Some of the seals may bear names, titles, and symbols of these elites and could throw important light on the Harappan rulers, if the writing could be read.

One of the earliest hypotheses regarding the Harappan political structure was put forward by Stuart Piggott and was supported to some extent by Mortimer Wheeler (for details of the various theories, see Jacobson, 1986). Piggott suggested that the Harappan state was a highly centralized empire ruled by autocratic priest-kings from the twin capitals of Mohenjodaro and Harappa. This view was based on a number of features, including the level of uniformity in material traits, the use of a common script, and standardized weights and measures. Mohenjodaro and Harappa seemed to clearly stand out in the midst of the other settlements. Urban planning and monumental public works implied the mobilization of a specialized labour force. The ‘granaries’ at Mohenjodaro and Harappa fitted in with a view of the Harappan rulers as exercising a high level of control over everything,

even maintaining buffer stocks of grain to tide over times of food scarcity. The apparent lack of internecine warfare between the settlements suggested that they were united under a single rule.

This view of the Harappan state soon came in for criticism. Walter A. Fairservis (1967) argued that the Harappans did not have an empire, not even a state. He pointed to the absence of evidence of priest-kings, slaves, standing armies, or court officials. According to him, Mohenjodaro was a ceremonial centre, not an administrative one. He argued that the sort of control reflected in the Harappan civilization could have been exercised by an elaborate village administration. Later, Fairservis modified his views to some extent and agreed that there may have been some element of centralized control and a class structure. But he still maintained that force did not play a significant role and that interdependence, religion, and tradition were responsible for regulating social behaviour.

Another view of the Harappan political system came from S. C. Malik (1968), who argued that the lack of imposing monuments and supreme gods goes against the idea of a strong, centralized state. The Harappan polity, according to Malik, is an example of what Elman Service described as the chiefdom stage, transitional between a kinship society and civil state society.

KEY CONCEPTS

Defining a state

The word ‘state’ is used very often in historical and anthropological analysis; therefore, it is important to know the various meanings attached to it. Here are some of the frequently cited and used definitions:

According to Elman R. Service (1975:14), a state is characterized by the existence of civil law and formal government that are ‘institutionalized, enacted, official’, and which ‘employ, threaten, or imply the actual use of force’. For him, the essential ingredients of a state are the power of force and authority.

Ronald Cohen (1978: 69-70) identified the state as a specific type of political system characterized by a centralized bureaucracy and dominant control of the mechanisms of force by a central authority. He further emphasized that an important difference between a chieftaincy and state was the latter’s ability to counter forces of political fission (breakaway groups or splintering).

The central element in Morton H. Fried’s (1978) conception of the state is social stratification based on differential access of the members of a society to basic productive necessities. Fried makes a distinction between pristine states and secondary states. A pristine state is one which emerges from indigenous stimuli, usually with no pre-existing models. A secondary state is one which has the model of an already existing state at hand and whose origins are related to pressures from this already existing state.

Henri J. M. Claessen and Peter Skalnik (1978) define an early state in the following way: a centralized socio-political organization for the regulation of social relations in a complex, stratified society, which is divided into at least two basic strata or emergent social classes—the rulers and the ruled—and in which the relations of political dominance and tributary obligations between the rulers and the ruled are legitimized by a common ideology founded on reciprocity (mutual relations of give and take). They also suggest that early states can be divided into three types on the basis of increasing levels of complexity—the inchoate early state, the typical early state, and the transitional early state.

Since state formation is a gradual process, it is often difficult to say precisely when something that can be called a ‘state’ appeared. Elman Service suggests that the transitional period between a pre-state kinship society and a state society should be considered a distinct stage in itself called the chiefdom stage. This is characterized by ‘centralized direction, hereditary hierarchical status arrangements with an aristocratic ethos, but no formal, legal apparatus of forceful repression’. He adds that leadership in a chiefdom was exercised by an authority that possessed neither formal legal power nor a bureaucracy. There were social ranks, but no classes.

Part of the problem in defining a state is that the many different kinds of state systems that have existed in history make it difficult to formulate a universal definition. For instance, although Fried directs attention to the element of social stratification in state societies, his emphasis on centralization simply does not fit all states. Apart from the problem of definition, in the case of early states, there is also the problem of identifying levels of social and political complexity on the basis of archaeological evidence.

Recent studies of the state have questioned various aspects of the older evolutionary models and terminology. For instance, Norman Yoffee has challenged various ‘myths’ related to the evolution and nature of the earliest states. These myths include the ideas that all these states were basically similar: that they were ruled by powerful totalitarian elites who exercised a monopoly of control over goods, services, and information; that they were marked by territorial integration of large areas; and that their social structure can be understood by invoking modern ethnographic parallels.

SOURCE Claessen and Skalník, 1978; Yoffee, 2005

The two trends in recent writings are, paradoxically, a return to the idea of a Harappan empire and a complete rejection of such an idea. Ratnagar (1991) analysed the archaeological evidence and used cross-cultural parallels with other early state societies to conclude that we do seem to be looking at a Harappan empire. The strongest critique of such a view has come from Jim Shaffer (1982b). Shaffer questions the level of homogeneity in the Harappan civilization and suggests that it could have been the result of a well-developed network of internal trade rather than a strong, centralized government. He underlines the absence of huge royal tombs, palaces, and temples, and the absence of marked social differentiation of the kind visible in ancient Egypt and Mesopotamia.

At Harappan sites, artefacts of various types are distributed throughout the occupational levels rather than clustered in elite residences or structures. All the typical Harappan artefacts (including ornaments of precious metals and semi-precious stones, seals and sealings, and the script) occur in small village settlements. This suggests an equality of access to wealth or the symbols of wealth among village and city dwellers, which goes against the idea of a centralized empire.

The fact that some form of state structure did exist in the Harappan civilization cannot be denied. The absence of marked social or economic differences and tombs or palaces of the Egyptian or Mesopotamian kind does not mean that a state did not exist, rather that it was a different sort of state. The communications system, standardization in artefacts, site specialization, mobilization of labour for public works, the establishment of the trading outpost of Shortughai—all these things indicate a level of economic complexity and the existence of a state. So does the level of cultural homogeneity and the use of a common system of writing across areas in which many different languages and dialects must have been spoken. The levels of social differentiation indicate some degree of class stratification. Some of the buildings on the citadel complex seem to have had an administrative function. Centralized control is apparent in the Harappan civilization. The questions are: How much and by whom?

A priest-king?

In ancient Mesopotamia and Egypt, rulers are portrayed extensively in stone reliefs and sculptures; their palaces, tombs, and temples further proclaim their power. The Harappan case is strikingly different. The stone bust of a male figure found at Mohenjodaro has been given the label ‘priest king’. The figure is that of a man with a close-cropped beard, half-closed eyes, and a fillet with an encrusted diadem around his head. An armlet with a similar but smaller ornament is tied around his right arm. A robe decorated with a trefoil design passes over his left shoulder and under his right arm. However, whether he represents a priest or king or both is far from certain. The same is the case with a large damaged seated figure found at Dholavira. While large

houses have been found at Harappan sites, none of them matches our idea of a palace, although it is possible that certain buildings on the citadels of cities such as Mohenjo-daro were the functional equivalent of palaces.

Jacobson (1986) suggests that the Harappan state was an early state with the following characteristics: a sovereign or sovereigns closely linked to a mythical character and seen as benevolent; a military component lacking the dominance characteristic of more mature states; and weakly developed economic stratification. According to Possehl (2003: 57), Harappan society was highly disciplined and had a strong corporate element; the Harappans may have been ruled by councils rather than kings. Kenoyer (1998: 100) suggests that the Harappan state must have comprised many competing classes of urban elites, such as merchants, ritual specialists, and those who controlled resources such as land and livestock, with different levels and spheres of control.

Kenoyer also suggests that the animals on the square stamp seals represent totemic symbols standing for a specific clan, perhaps along with some additional information. At least 10 clans or communities are represented by these animals—the unicorn, humped bull, elephant, water buffalo, rhinoceros, humpless bull with short horns, goat, antelope, crocodile, and hare. The unicorn motif is found at almost all sites where the seals have been found, including in Mesopotamia. At Mohenjodaro, over 60 per cent of the seals have this motif, while it occurs on about 46 per cent of the seals at Harappa. The large number of unicorn seals at major cities led Ratnagar to suggest that the unicorn was the symbol of the Harappan ruling elite. Kenoyer, on the other hand, argues that the ‘unicorn clan’ probably represented the aristocracy or merchants who had an important executive role in the government. It is in fact the less frequent motifs such as the bull, elephant, rhinoceros, and tiger that may have been symbols of the most powerful rulers at the apex of the Harappan power structure.

While Mohenjodaro stands out in some ways (for instance, no other site has a structure comparable to the Great Bath), there are other large Harappan cities such as Rakhigarhi, Lurewala, Ganweriwala, and Dholavira. Were they provincial centres knit together through a well-worked-out system of political control? Were they the capitals of separate states? Were they city-states? In the past, scholars tended to simply presume highly centralized political structures, whereas now there is a greater acceptance of the possibility of decentralization. It is not, however, certain whether we need to think in terms of a Harappan empire or a number of separate, perhaps inter-related states. Another possibility that cannot be ruled out is that there may have been several states with different kinds of political organization.

The Decline of Urban Life

At some point of time, things started going wrong in the Harappan cities. Decline had set in at Mohenjodaro by 2200 BCE and the settlement had come to an end by 2000 BCE. In some places, the civilization continued till 1800 BCE. Apart from the dates, the pace of decline also varied. Mohenjodaro and Dholavira give a picture of gradual decline, while at Kalibangan and Banawali, city life ended all of a sudden (see Lahiri, 2000 for the various theories regarding Harappan decline).

A'UNICORN' SEAL

One of the most popular explanations of the decline of the Harappan civilization is one for which there is least evidence. The idea that the civilization was destroyed by Aryan invaders was first put forward by Ramaprasad Chanda (1926)—he later changed his mind—and was elaborated on by Mortimer Wheeler (1947). Wheeler argued that references in the Rig Veda to various kinds of forts, attacks on walled cities, and the epithet puramdara (fort destroyer) given to the god Indra must have a historical basis and reflect an Aryan invasion of the Harappan cities. He identified a place called Hariyupiya in the Rig Veda with Harappa. Wheeler also pointed to certain skeletal remains found at Mohenjodaro as proof of the Aryan massacre. He subsequently modified his hypothesis, to the extent that he acknowledged that other factors such as floods, decline in trade, and over-utilization of natural resources may have had a role to play. But he insisted that the ultimate blow was given by an Aryan invasion. The Cemetery-H culture, he suggested, represented the culture of the Aryan invaders.

Many scholars such as P. V. Kane (1955), George Dales (1964), and B. B. Lal (1997) have refuted the invasion theory. The evidence from the Rig Veda, a religious text of uncertain date, is far from conclusive. Moreover, if there had been an invasion, it should have left some traces in the archaeological record. There is, in fact, no evidence of any kind of military assault or conflict at any Harappan site. The 37 groups of skeletal remains at Mohenjodaro do not belong to the same cultural phase and, therefore, cannot be connected to a single event. Not one of these skeletons was found on the citadel mound, where we would have expected a major battle to have taken place. The fact that there is a sterile layer between the mature Harappan and Cemetery-H levels goes against Wheeler’s hypothesis that the latter represents the settlement of the Aryan invaders. Moreover, K. A. R. Kennedy’s analysis (1997) of the skeletal remains does not show any discontinuity in the skeletal record in the north-west at this point of time, making it clear that there was no major influx of new settlers with a different physiognomy. The Harappan civilization was not destroyed by an Indo-Aryan invasion.

Natural disasters, not necessarily sudden or single, did have a role to play. Several layers of silt at Mohenjodaro give evidence of the city being affected by repeated episodes of Indus floods. M. R. Sahni (1956), and later Robert L. Raikes (1964) and George F. Dales (1966), argued that the floods at Mohenjodaro were the result of tectonic movements. Dales suggested that these may have

occurred at a place called Sehwan, about 90 miles downstream from Mohenjodaro, where there is evidence of rock faulting. The theory is that tectonic movements led to the creation of a gigantic natural dam that prevented the Indus from flowing towards the sea, turning the area around Mohenjodaro into a huge lake. The theory of several such episodes of flooding induced by tectonic movements is not, however, convincing. Neither is H. T. Lambrick’s hypothesis (1967), based on what he himself describes as purely circumstantial evidence, that the Indus changed its course, moving some 30 miles eastwards, starving Mohenjodaro and its inhabitants of water.

While Mohenjodaro may have got worn out due to repeated episodes of naturally occurring floods, Harappan sites in the Ghaggar-Hakra valley were affected by gradual desiccation. The Sutlej or the Yamuna once flowed into the Ghaggar. Tectonic movements led to river capture—either the Yamuna joined up with the Ganga system or (what is more likely) the Sutlej was captured by the Indus, drastically reducing the water flowing into the Ghaggar. M. R. Mughal’s (1997) study of settlements in this region shows a drastic reduction in the number of sites as the river dried up.

A sudden rise in the Arabian Sea coastline of west Pakistan could have caused floods and a rise in soil salinity. Such an uplift along the coast and in the lower Indus valley could also have seriously disrupted the coastal communications and trade of the Harappans.

Reference has already been made to the debate on the nature of the climate, especially rainfall, in protohistoric times. On the basis of his study of pollen from Rajasthan lakes, Gurdip Singh (1971) suggests a connection between the onset of a drier climate and the decline of the Harappan civilization. However, a study of the sediments of the Lunkaransar lake indicates that the onset of drier conditions in this area may have happened well before the emergence of the Harappan civilization. Whether climatic change played a role in the decline of the Harappan civilization therefore remains unclear.

The issue of environmental change can be connected to the ways in which the Harappans were treating their environment. Perhaps they were over-exploiting it through over-cultivation, overgrazing, and excessive cutting of trees for fuel and farming. This would have resulted in decreasing soil fertility, floods, and increasing soil salinity. Making estimates of population, land, food, and fodder requirements on the basis of modern data, Fairservis suggests that the civilization declined because the growing population of people and cattle could not be supported from resources within the Harappan culture zone.

Shereen Ratnagar (1981) has argued that the decline in the lapis lazuli trade with Mesopotamia was a factor in the decline of the Harappan civilization. Whether this trade was particularly important for the Harappans is, however, debatable; consequently, this could not have been a factor responsible for the decline.

Archaeological evidence does not give direct access to the possible social and political dimensions of the decline of the Harappan civilization. What it does indicate very clearly is that the Harappan culture underwent a gradual process of de-urbanization. The mature Harappan phase was followed by a post-urban phase, known as the late Harappan phase.

The Significance of the Late Harappan Phase

There are five geographical zones of the late Harappan phase: Sindh; west Punjab and the Ghaggar-Hakra valley; eastern Punjab and Haryana; the Ganga–Yamuna doab; and Kutch and Saurashtra. In

Sindh, the late Harappan phase is represented by the Jhukar culture at sites such as Jhukar, Chanhudaro, and Amri. The transition from the mature to the late Harappan phase in this region does not show any sudden discontinuity. There were gradual changes in the seals, a decrease in the frequency of cubical weights, and writing came to be confined only to pottery. The evidence of pottery suggests reciprocal contacts between the Jhukar culture of Sindh and the late Harappan culture at Lothal and Rangpur.

In the Punjab province of Pakistan and the Ghaggar-Hakra valley, the late Harappan phase is represented by the Cemetery-H culture. There is a decline in the number of settlements from 174 in the mature Harappan phase to 50 in the late Harappan phase. In east Punjab, Haryana, and north Rajasthan, the late Harappan settlements were small compared to the mature Harappan ones. In the Ganga–Yamuna doab, compared to the 31 mature Harappan sites, there are 130 late Harappan sites. The settlements were small, houses were generally made of wattle and daub, but the agricultural base was very diverse. In Kutch and Saurashtra, there is a marked increase in the number of settlements in the earlier part of the late Harappan phase, from 18 in the mature Harappan phase to 120 in the early late Harappan phase.

While there was abandonment or severe reduction in population in Sindh and Cholistan, the increase in the number of settlements in Punjab, Haryana, western Uttar Pradesh, northern Rajasthan, and Gujarat shows that this was not the case everywhere (see (< />)Chapter 5 for details). In fact, at around the time that people were abandoning Mohenjodaro, the people of Rojdi in Saurashtra were expanding and rebuilding their settlement. The data suggests an eastward and southward shift of settlements and people.

The evidence from mature and late Harappan sites shows a complex interplay of elements of continuity and change. Compared to mature Harappan pottery, the slip of late Harappan pottery is less bright. The pots tend to be thicker and sturdier. Some of the classic Harappan shapes—e.g., the beaker, goblet, perforated jar, s-shaped jar, and pyriform (pear-shaped) jar—disappear. Other shapes—e.g., jars of different shapes and the dish-on-stand—continue. Various elements of Harappan urbanism such as the cities, script, seals, specialized crafts, and long-distance trade declined in the late Harappan phase, but did not completely disappear. Some of the late Harappan sites such as Kudwala (38.1 ha) in Cholistan, Bet Dwarka in Gujarat, and Daimabad (20 ha) in the upper Godavari valley can be described as urban, but they are few and far between. Graffiti on pottery occurs in Saurashtra and northern Gujarat as well as in the eastern regions. Four potsherds with Harappan letters were found at late Harappan levels at Daimabad. Some circular seals occur at Daimabad and Jhukar; rectangular seals minus motifs were found at Dholavira. A rectangular conch shell seal with the motif of a three-headed animal, similar to that found on seals of the Persian Gulf, was found at Bet Dwarka. This suggests that contact with the Persian Gulf continued in the late Harappan phase, at least in the Gujarat region. The late Harappan phase at Bhagwanpura shows flourishing specialized craft activity; there are 2 clay tablets and 19 sherds with graffiti, which could represent a script. In Punjab and Haryana, there are faience ornaments, beads of semi-precious stones, terracotta cart frames, kilns, and fire altars.

A notable development in the late Harappan phase was the diversification of agriculture. At Pirak in Baluchistan, there was the beginning of double cropping—wheat and barley were being grown as winter crops and rice (with irrigation), millet, and sorghum as summer crops. In the Kachi plain, there were fairly large settlements, growing a variety of crops, supplemented with irrigation. In

Gujarat and Maharashtra, various kinds of millets were being grown as summer crops. Rice and millets were found at late Harappan levels at Harappa. Excavations at Hulas gave evidence of diverse plant remains. Grains included rice, barley, dwarf wheat, bread wheat, club wheat, oats, jowar, and finger millet. Pulses included lentil, field pea, grass pea (khesari), kulthi, green gram (moong), and chickpea. Almond and walnut shells were found, and a single carbonized seed of cotton was identified.

The general picture presented by the late Harappan phase is one of a breakdown of urban networks and an expansion of rural ones. There is an overlap between the late Harappan and Painted Grey Ware (PGW) culture at sites such as Bhagwanpura and Dadheri in Haryana, and Katpalon and Nagar in Punjab. Also significant is the overlap between late Harappan and Ochre Coloured Pottery (OCP) levels in western Uttar Pradesh at sites such as Bargaon and Ambakheri. The evidence from this area, Gujarat, and north Maharashtra suggests an eastward and southward migration of the Harappans due to a combination of pressures such as those discussed in the earlier section.

CONCLUSIONS

The Harappan civilization was the first urban culture in South Asia. The urban phase of the Harappan culture emerged from the proto-urban early Harappan phase. Archaeological evidence reveals a great deal about this civilization—its varied subsistence base, vibrant craft traditions, and extensive trade networks—but given the non-decipherment of the script, conclusions about many other aspects such as religion, society, and polity remain speculative. There was cultural homogeneity as well as diversity within the vast Harappan culture zone. Some of the neolithic, neolithic–chalcolithic, and chalcolithic sites mentioned in (< />)Chapter 3 were roughly contemporaneous with the Harappan civilization and interacted with it. The Harappan civilization did not come to a sudden end. The urban phase was followed by the late Harappan phase, which was marked by the decline of urban features and the diversification of agriculture.