|  
            Kalibangan - the largest 
            prehistoric site in Rajasthan
           | 
         
         
         
         
         
         
         
         
 
  
         
         | 
          
            |  |  
            | 
                 |  
            | 
                 |  
            | 
                 |  
            |  |  
            |  |  
            |  |  
            | 
              
                | 
                Kalibangan - the largest prehistoric site in Rajasthan |  
                | 
  Just as scholars and laymen all 
  over India, have heard of Mohenjodaro, so scholars and laymen, particularly in 
  Rajasthan, should have heard or read about Kalibangan. 
 Kalibangan is perhaps the largest prehistoric site in Northern Rajasthan. It 
  is situated in the Ganganagar District (now Hanumangarh), which was formerly a 
  part of the state of Bikaner. It is easily approachable from Hanumangarh 
  Railway Station on the Delhi- Bikaner line.
 
 People of Rajasthan and particularly the residents of Jodhpur, Jaipur and 
  Bikaner know that this region is practically a desert, with occasional thorny 
  bushes of babul and thor. In this surroundings, a village or a town with 
  whitewashed mud houses, or timbas (tillas) – small mounds or hillock, strike 
  the eye immediately. While many of these tillas are nothing but sand-dunes, 
  some of these, as Rangmahal on Kalibangan are littered with innumerable 
  potsherds. Because these potsherds were originally red or even bright red, 
  these mounds even now after hundreds or thousands of years appear reddish. 
  Particularly this is so at Rangmahal.
 
 Normally no one cares for or looks at these potsherds, certainly not the 
  caravan driver who passes by these tillas while carrying goods to and fro, 
  from distant towns and cities. But to an archaeologist these potsherds are 
  like open books. All these potsherds speak. Perhaps each potsherd has some 
  story to relate. You may wonder “how”. The reason is simple. Though the shapes 
  of the pots to which these potsherds belong are rarely intact, still the 
  potsherds are not dead. It is one of the wonders of nature that once a clay 
  pot is given some color, or naturally painted with some designs and then 
  fired, neither the color nor the design goes, though the pot or its pieces may 
  be exposed to the sun and rain and even used for hundreds of years. It is this 
  Nature’s secret that archaeologists discovered some 150 years ago. For by 
  carefully studying potsherds and intact pots if available, an archaeologist 
  can gradually tell how old the pot is, he can also say by further study and by 
  piecing together broken parts of the one and the same pot, what the original 
  shape was, and what part it played in the life of the person who possessed it.
 
 It is the peculiar feature – almost total indestructibility – of pottery – 
  that is one of the main clues which an archaeologist looks for whole searching 
  for bygone cultures and civilizations. Hence Shri Amalananda Ghosh during his 
  exploration of the valleys of the Ghaggar, the ancient Saraswati and the 
  Drishadvati primarily looked for collected potsherds. Of course he was not the 
  first scholar to do so. Before him Shri Aurel Stein had done so for that part 
  of the Ghaggar which flows into the Bahawalpur District of Pakistan. Stein 
  thus had discovered numerous ancient sites. In some of these, he had 
  discovered the kind of pottery which had been discovered previously at 
  MohenJodaro and numerous other sites in Sind. Ghosh did exactly what Stein had 
  done, but being more experienced and well acquainted with the Harappa or the 
  Indus Civilization he noticed that three or four different kinds of potteries 
  were found littered over these tillas in the Bikaner State. While those 
  similar to or identical with that of the Indus Civilization can be easily 
  assigned to the Indus Civilization, others belonged to different cultures. The 
  pottery found at Sothi and other sites in and around the present town was 
  designated as “Sothi”. While another – found at Rang Mahal was called “Rang 
  Mahal Culture.” Because of its bright red color and painting, Rang Mahal 
  appeared promising. It was certainly new. But when it was excavated by a 
  Swedish Expedition, it was found to belong to the early Historical period, to 
  the period of the Kushan ruler of Northern India, including Rajasthan. So when 
  the Archaeological survey of India though of examining of pursuing Ghosh’s 
  discoveries they took up the mounds at or near Kalibanga. For here had been 
  found potsherds and chert knife blades indicative of the existence of the 
  Indus Civilization and also another culture or civilization called the Sothi 
  Culture by Ghosh.
 
 And as rightly anticipated by Ghosh, several years of excavations at 
  Kalibangan by Prof. B.B. Lal and Shri B.K. Thapar have brought to light the 
  existence of a fairly extensive town of the Indus Civilization Harappan 
  Culture, and also the earlier existence of a town to the pre- Indus or Sothi 
  culture. However, as it is the practice with archaeologists, these Sothi or 
  pre - Indus culture have been designate respectively Harappan cultures 
  respectively.
 
 The ancient habitations was spread over an area of a quarter of a square 
  kilometer, and from the beginning consisted of two closely-knit but distinct 
  mounds, an eastern and a western mound. These form a prominent feature of the 
  landscape with their slopes strewn with dark brown nodules, mud-bricks, and 
  numerous potsherds. No traveler in this desert, whether he be an archaeologist 
  or not, could but be struck by this feature of the landscape with their slopes 
  strewn with dark brown nodules, mud – brick, and numerous potsherds. No 
  traveler in this desert, whether he be an archaeologist or not, could but be 
  struck by this feature for these are so conspicuous among the masses of sand 
  dunes on the west, east and south and the green fields on the north; the 
  latter as a result of irrigation.
 |  |  
            |  |  
            | 
              
                | 
                Fortification |  
                | 
                This pre – Harappan settlement was protected 
                by a mudbrick fortification. When first built it was about 6 
                feet (1.90m) wide, but later the width of the wall was almost 
                doubled. It varies between 3.70 and 4.10m. the brick size 
                however remained the same. No corner angles of these walls have 
                been found. The north – south distance of the fortified area 
                measures approximately 250 m. The necessity of such an increase 
                indicates that the inhabitants felt insecure with a wall that 
                was only 6 feet wide and hence made it up to nearly 12 feet. 
                This is certainly a good thickness for a fortification wall at 
                this period, for it had to withstand only such missiles as stone 
                or copper-tipped arrows and clay or stone sling balls. Whether 
                this wall could be easily scaled or not cannot be said, there is 
                no means of knowing its height, since the later people – the 
                Harappans in our present knowledge had to break it or remake it 
                to sit their requirements. 
 What is important is that the traces of a fortification wall 
                have survived. We were told by Marshall some 40 years ago that 
                the non-violent people. Then came Sir Mortimer Wheeler who was 
                the first to identify a defense wall at Harappa and then later 
                at Mohenjodaro. This discovery made him propound his famous 
                theory that the Aryans destroyed the Indus Civilization, for he 
                saw in indra, the Purandara, one who destroyed “walled” 
                “fortified cities.”
 
 Now with the discovery of fortification at Kalibangan, and also 
                at Kot Dijji in Sind, where the mud-brick wall has a plinth of 
                stone rubbles, the whole problem of fortification takes a 
                different turn.
 
 The least we can say is that the Harappans were not the first to 
                have fortified cities in Sind and Rajasthan. And hence the 
                question of Aryans along being “the Purandaras” does not arise. 
                These might as well be the Harappans, who at Gumla destroyed the 
                pre-Harappan habitation.
 
 Again, they were not the first to introduce wheeled conveyance 
                and metal tools/weapons, in these regions, for these were also 
                known to the pre-Harappans. But what the latter did not have was 
                the first access to the flint quarries of Sukkur and Rohri so 
                that their tools for daily use in the house for cutting, 
                slicing, and piercing had to be made from (presumably local) 
                material such as agate, chalcedony and carnelian. These tools 
                are in now way different from the microliths made by the Bagor 
                and Tilwara people, except that at Kalibangan we have mostly 
                straight-sided blades including serrated and banked and fewer 
                lunates, trapezes and such geometric shapes. This small 
                difference is significant, indicating that man no longer needed 
                and made compound tools like the sickle and harpoon and the 
                arrow-head with stone tips, but utilized (probably) copper tools 
                instead.
 |  |  
            |   |  
            | 
              
                | 
                Pottery |  
                | 
                However, the most striking difference between the pre – Harappan 
                and the Harappan, which is of utmost importance to an 
                archaeologist, is pottery. The Harappan pottery is bright or 
                dark red and uniformly sturdy, and so well baked that no part of 
                the core remains yellowish or blackish showing imperfect firing. 
                This is not the case with the pre-Harappan pottery. The latter 
                is pinkish, comparatively thinner, and not so well baked as the 
                former. Some of it is distinctly carelessly made. One of its 
                varieties, though well – potted, has its outer surface, 
                particularly the lower part; roughened or rusticated (this is 
                also seen at Ahar). Still another variety, represented mainly by 
                basins, is decorated all over by obtusely incised patterns on 
                the outside. 
 Not only the fabric and most of the decorative patterns, but the 
                forms of the pre-Harappan pottery are strikingly different from 
                the Harappan. While the graceful painted Harappan vase, the 
                goblet and the cylindrical perforated vessel, and the variety of 
                footed dishes or foot-stands are conspicuous by their absence, 
                present are some six to eight types of small and medium-sized 
                vessels. And amongst these, the most noteworthy is a small footed 
                cup. This and its likes remind us on the one hand of the earlier 
                Iranian goblets from Sialk and Hissar, and on the other the 
                goblets or footed cups from Navdatoli on the Narmada.
 |  |  
		|  |  
		| 
              
                | 
                New Features |  
                | 
                Though in a general 
                way all this conforms to what we know of the Indus Civilization, 
                Kalibangan has revealed certain new features. 
 First there are the usual two habitations. One is the so-called 
                “Citadel” on the western side located on the earlier pre – 
                Harappan settlement overlooking the ancient Saraswati. It is a 
                coincidence that in all the three sites – Harappa, Mohenjodaro 
                and Kalibangan, the citadel is located on the western side and 
                that on a previous habitation? The other is situating towards 
                the east, at a little distance from the first, rich on the sandy 
                plain. It appears now that both these – the “citadel” as well as 
                the “lower city” was enclosed by a separate mud-brick 
                fortification wall. Of the city fortification only the east west 
                wall running for nearly 230 feet (over 80 m.) has so far been 
                exposed. The north – south wall is not yet fully laid bare. 
                Within the city, so far five-north south and three east – west 
                roads, and a number of east – west running lanes have been 
                explored, showing how well – planned the city was. The roads and 
                streets were found to be clear of any intrusions from the 
                house-owners and squatters – a civic feature which is becoming 
                rare all over India today.
 
 Whether there were too many carts moving in the streets or not 
                we do not know. But to avoid damage to the houses at street 
                comers, by the sudden turning of the cart, wooden fender posts 
                were provided, a few of which survive.
 
 Rectangular platforms outside some of the houses seems to have 
                been made for two purposes. Either as outdoor rest-places, or 
                contrivances specially made for mounting over an animal’s back, 
                or rests for laborers carrying heavy load over their heads.
 |  |  
		|  |  
		| 
              
                | 
                Houses |  
                | 
                Such well laid out 
                streets were not metalled, except in the late phase of the city, 
                not were they provided with regular drains, as in Harappan 
                cities. However, the houses had drains made of either wood 
                scooped out in the shape of ‘U’, or more often with baked 
                bricks. These drains emptied themselves in the soakage jars 
                embedded in the street floor. It is observed that each house 
                opened or had a frontage on at least two or three streets, as in 
                Chandigarh, for instance. Normally, only one, the corner house, 
                can have such a frontage, but others at the most two, a front 
                and a back one, that too if there is only a single row of houses 
                in a street or a lane. In the Harappan phase at Kalibangan there 
                was only a single row of houses in each street, and this again, 
                divided into several small blocks, so that many open to so much 
                light and air in a region like N. Rajashtan? Or was it after the 
                current fashion, as today in Chandigarh? Even internally the 
                houses were well provided with light and air, for they were 
                built on the Chatussala principle, that is, there was a central 
                courtyard, at times provided with a well and six or seven rooms 
                on its three sides. There is some evidence to say that these 
                earliest houses in Kalibangan were storeyed, for in one house 
                were found traces of a preserved stairway. 
 The roofs of these houses were probably flat. As today in 
                Kalibangan and many villages in Rajasthan, the houses were built 
                of mud-bricks. The size of which was 20x15x7½. cm. That is the 
                length was twice the thickness, the proportion being 4:2:1. 
                However, the Harappans of Rajasthan were judicious, for they 
                have consistently used baked bricks in doorsills, wells and 
                drains, all places where the wear and the tear was much, and the 
                structures liable to be damaged if baked bricks were not used.
 
 This common sense is again witnessed in the way the flooring of 
                houses are made. Unlike Mohen-jo-daro and Harappa the floors 
                were made firm by ramming (called Koba), and sometimes capped 
                additionally with mud bricks or terracotta nodules.
 
 However, in one case, the floor is found paved with tiles, 
                bearing the typical intersecting design of circles. Exactly 
                similar design occurs at Kot Diji in what is called a “bath 
                tub”. While there is no doubt about the existence of this design 
                in the tub-like large vessel at Kot Diji, it should be 
                ascertained, if not already done, whether at Kalibangan it is 
                real flooring or too is a part of a tub. Anyway, this is a most 
                interesting feature, which does not seem to be merely 
                ornamental, but perhaps of some religious significance, or else 
                some other design would have been preferred. For we know this 
                was a favorite design with the Harappans, and occurs on the 
                graceful vase.
 
 The Bikaner Harappans thus show considerable originality even in 
                the make up or construction of their house. This is further 
                illustrated by three other features. All these are seen in what 
                is called the “Citadel Mound”.
 |  |  
		|  |  
		| 
              
                | 
                Fortification |  
                | 
                The exposed fortification in this mound makes it look roughly 
                like a parallelogram on plan, exactly as at Harappa, that at 
                Mohenjodaro is not fully exposed, but would probably be of the 
                same shape. This was divided into two almost equal halve. Each 
                half may be described as a rhomb. Again each of this rhomb was 
                enclosed by a fortification wall. The width of th8is wall was 
                quite large, as much as 7 m. (about 20 feet) at places, the 
                minimum being 3 m. (10 feet). This was further strengthened at 
                intervals with rectangular salient (projections) and towers. The 
                wall, it would appear was built in two phases or twice, for 
                initially very large bricks measuring 40x20x10 cm. were used in 
                its construction. Later the normal sized bricks (30x5x7½ cm.), 
                used in civic houses were preferred. |  |  
		|  |  
		| 
              
                | 
                Platform |  
                | 
                The southern rhomb is found to contain five or six platforms of 
                mud or mud bricks each separate from the other, and different in 
                size, so that the space (passage) between the two platforms is 
                never uniform. Now here are these platforms connected with the 
                fortification wall. Access to these platforms had to be by a 
                flight of steps, which rise from the passage between the 
                platforms. Further the passage fronting the steps was paved. 
 These mud or mud-brick platforms seem to be quite different from 
                the platform at Lothal, Harappa and Mohen-jo-daro, for instance, 
                the latter were largely built for protecting the superstructures 
                from recurring floods. But at Kalibangan they seem to have a 
                religious function, though this cannot be ascertained, for 
                except in one case the superstructures have disappeared. Or is 
                (was) it because it had by this time become a custom, convention 
                or fashion to build the citadel on artificial mud or mud-brick 
                hillock?
 |  |  
		|  |  
		| 
              
                | 
                Sacrificial Pits (?) |  
                | 
                In the one surviving example was found a rectangular pit (1 x 
                1.25 m.) lined with baked bricks. This Kunda contained bones of 
                a bovine and antlers, perhaps a sacrifice was performed. This 
                suggestion is strengthened by the fact that adjoining the Kunda 
                was found a well and a “fire-altar”. |  |  
		|  |  
		| 
              
                | 
                Fire Altars |  
                | A 
                row of such fire-altars was noticed on another platform and also 
                in many houses in the “ Lower city”. These “fire-altars” 
                invariably consist of shallow pits, oval or rectangular plan. 
                Fire was made and put out in situ (that is there and then), as 
                proved by lumps of charcoal in the open part of the pit. In the 
                center of the pit was found a cylindrical or rectangular 
                (sun-dried or pre-fired) brick. Around or near about were place 
                flat, triangular or circular terracotta pieces, known hitherto 
                as “terracotta cakes.” In a recent article it is said that 
                towards the end of the Harappan settlement this practice was 
                being gradually abandoned because the Saraswati was losing its 
                water to the Yamuna, the fire-altars were poorly equipped – only 
                with one centrally placed brick on edge in a small pit. 
 Such a “fire-altar” has also been noticed by Casal at Amri, and 
                something similar, but perhaps not identical, was found by Rao 
                at Lothal. Perhaps such fire-altars also existed at Harappa and 
                Mohen-jo-daro, but were missed in mass digging and have only 
                been revealed in slow, careful excavation.
 
 That here in this platformed, well fortified enclosure we have 
                first traces of a religious building with houses for its priests 
                on the site which is also borne out by the fact that no barge, 
                broad streets have been so far found within the citadel. In 
                fact, there is no room for any vehicular traffic. So we have to 
                presume that either every-body walked, or some people – like the 
                priests and the like or the ruler – were carried in palan – 
                quins. The general public could go to these platforms from the 
                southern side through a stairway which ran along the outer face 
                of the fortification wall between the two centrally located 
                salient. A similar arrangement was made for the residents in the 
                northern half of the “Citadel”.
 
 At all the three sites, these citadels are built over a little 
                higher ground, which at Harappa and Kalibangan is proved not to 
                be quite natural but due to the remains of an earlier 
                habitation. However, the elevation was further raised by mud or 
                mud – brick platforms. And this at Harappa and Mohen – jo – daro 
                (and Lothal) is explained as a precaution against recurring 
                floods. But at Kalibangan there is (so far) no evidence of a 
                flood, and again the platform are on separate block with paved 
                flooring in the passage. Further the fire-altar-like structure 
                and the sacrificial kunda on these platforms make the excavators 
                feel (and I agree with them) that these are truly religious 
                structures. Did they have a similar function at Mohen-jo-daro 
                and Harappa? Or there was the real need of a mud-brick platform 
                as a protection against floods, and this functional feature was 
                later mechanically copied at Kalibangan?
 
 The smaller, portable objects at least testify once again to the 
                rich and comfortable life which we now associate with the 
                Harappans. A varied and beautiful pottery (its manifold uses for 
                eating, drinking, storing etc. could be imagined if the numerous 
                platters, dishes and other vessels found intact in a grave are 
                drawn function-wise), ornaments, beads and bangles – in shell, 
                terracotta, semi – precious stones and faience, and some in 
                gold, weights and measures (one in graduated scale as at Lothal), 
                the undeciphered seals including one cylinder seal with half 
                human and half animal figures on it, recalling Sumerian contact 
                and features, and above all, exquisite figure sculpture in the 
                round of a charging bull.
 |  |  
		|  |  
		| 
              
                | 
                Religion |  
                | 
                There is nothing specific to tell us about 
                the Harappan religion except the so called fire-altars and the 
                Kunda and an oblong terracotta cake, incised on both sides with 
                a figures reminds once again of the figure in gold in Hissar 
                III, and a painting on a pot at Kot Diji from the junction 
                layers. The incised figure seems extent we are familiar with a 
                horn-headed deity from the famous Pasupati-like seal. But there 
                the horns are not quite clear, and hence some scholars doubt it 
                identification. But in the Kalibangan figure there is no room 
                for doubt. And this as shown here can be derived from the 
                mouflon (or wild mountain sheep) head in gold form Hissar in 
                Iran through the painting of a bullas head on a pot at Kot Diji 
                in Sind, and also at Gumla and Burzahom.  |  |  
		|  |  
		| 
              
                | Food |  
                | 
                The Kalibangan Harappans were both vegetarian 
                and non-vegetarian. Wheat and barley they must have eaten, 
                though so far only traces of barley have been fond. Among the 
                animals they knew and probably cooked for food the largest 
                percentage is that of humped cattle (cow/bull), then Indian 
                buffalo, pig, arasingha, elephant, ass (domesticated) rhinoceros 
                and camel. The camel is again important, proving its antiquity 
                in this region (sind and Rajasthan).  |  |  
		|  |  
		| 
              
                | 
                Burial Methods |  
                | 
                By and large the Kalibangan Harappans buried the dead, as at 
                west south west of the citadel has been found, on the present 
                flood plain of the river. Not only this cemetery sheds some 
                light on the different burial practices current at Kalibangan, 
                but the varying provision of grave goods, and the construction 
                of the graves enlightens us about the social stratification 
                prevalent in the city. So far three types of graves have been 
                found. In the first type, which seems to be fairly frequent, we 
                have an oblong pit dug into the ground. The dead body was laid 
                in the pit in an extended position with the head towards the 
                north and the feet towards the south. Then around the head were 
                arranged pots, dishes, platters, small water vessels, cups, but 
                not large storage jars, in one case numbering over seventy. This 
                illustrates that there was no fixed number of pts which one had 
                to provide for the dead. If one could afford, and probably 
                belonged to a higher social order, he could have a large number. 
                Besides pots, at times a copper mirror was placed near the head. 
                This is further proved by the fact that this particular grave ad 
                a lining of mud bricks on all the four sides, which were then 
                plastered with mud from inside.
 In the second type the grave-pit was oval or circular on plan 
                and contained besides an urn, other pots including platters and 
                dishes-on-stand.
 
 Here again the number varied from 4 to 29, depending upon the 
                wealth (and position) of the person. Again, besides pottery, 
                ornaments such as beads, shell bangles and objects of steatite 
                were kept.
 
 In the third variety, the grave-pit was rectangular or oval on 
                plan with the larger axis oriented north south, but curiously 
                contained no skeletal remains. Usually nothing but pottery was 
                found within these simple pits, though in one case a shell 
                bangle and a string of satellite disc beads and one of carnelian 
                were found.
 
 This is the first time that burials without any human skeletal 
                remains have been found on a site of the Indus Civilization. But 
                the reason behind this non-occurrence is not easy to gauge. It 
                is because that there was the custom of cremation-cum-burial, so 
                that the body was burnt, and later only the ash and a few bones 
                were buried in the urn, or even these were not kept but thrown 
                in the river or sea, as some people do today?
 |  |  
		|  |  
		| 
              
                | 
                Trepanning |  
                | 
                Kalibangan has also provided a very 
                interesting example of ancient medical belief and surgery. In a 
                child’s skull were found six circular holes. These holes were 
                made while the child was alive, for the wounds made by these 
                holes have healed, that is the edges of the holes have nor 
                remained sharp, as when first cut. This practice of boring holes 
                in the head while alive is called trepanning, and was widely 
                current in prehistoric times in Europe, about 3,000 B.C., and 
                was still witnessed in some of the aboriginal tribes of Peru in 
                Central America. Trepanning was resorted to, it is believed, to 
                relieve headache, and alleviate inflammation of the mastoid 
                (conical) prominence in the temporal bone to which the muscles 
                are attached), and on the brain due to injury. 
 So far the only example of trepanning we had was from Langhnaj 
                in north Gujarat, Kalibangan (and Lothal) have provided two 
                more. These thus give a wide base to a belief and practice which 
                was current in Europe and Africa, some 4,000 years ago; exactly 
                the time is was prevalent in Western and Northern India, 
                including Gujarat, Sind and the Punjab.
 
 The Kalibangan has given us considerable food for thought. Again 
                the paved road and flooring. These are new features not so far 
                met with at Mohen-jo-daro. But we must also note the absence of 
                certain well-known features, such as street-drains, and among 
                the portable smaller objects the complete absence of lingas, 
                yonis, and figurines of mother – goddesses. This is also a 
                feature of the Lothal (Saurashtra) Harappan, and thus underlines 
                the importance of Mohen-jo-daro and Harappa as “religious 
                capitals” as well.
 
 In many respects then the Rajasthan Harappan has a distinct 
                individuality. It is not an exact copy of the Indus. Such a 
                regional variation is but natural, though it would be worth 
                inquiring who introduced or brought about this variation, viz. 
                the indigenous element in the population or because during 
                migration from the centre, the original features got lost or 
                changed.
 |  |  |