Since the Ganga-Yamuna Doab does not begin till Delhi, we have here for consideration Pakistani and Indian Punjab, Haryana, the part of Rajasthan which borders Haryana, and the Cholistan desert of Bahawalpur in Pakistan. This is a vast alluvial plain, and what we know of its history tells us that during the protohistoric period this plain was the home of two river systems: one based on the Indus and the other based on the now-dried course of what is called Ghaggar in Haryana and Hakra in Cholistan. In the upper portion we have the Punjab plains proper, i.e. the area between the Sutlej on the east and the Salt Range on the west, and the Indo-Gangetic divide between the Sutlej and the Yamuna. Its lower section includes Sind and Bahawalpur.
The Ghaggar-Hakra course has been traced on the ground and its history speculated on for a long time beginning with C.E. Oldham in 1874. The network of miscellaneous rivers and drainage channels coming from the Himalayan foothills through modern Haryana and forming the Ghaggar system has also been reconstructed. In the present context we are not concerned with the details of this reconstruction; all we have to note is that till the protohistoric period the Indus flow on the west was matched by a parallel river-flow along the now-dried channel of the Ghaggar-Hakra on the east. Why and when the latter dried up cannot be determined; at some point its headwaters must have been subject to serious loss of water. It is not definitely known whether this was due to denudation and erosion or to the assumed deviation of the Sutlej and the Yamuna to their present channels from their previous flows into the Ghaggar. The major point is that the Ghaggar-Hakra flow existed and this point is important for understanding protohistoric India.
Downstream, the course of the Ghaggar-Hakra has been traced up to the Rann of Kutch. When we come to the Indus it becomes clear too that the present Indus flow through the Sukkur-Rorhi gorge is a late development and that its upper course in Sind was perhaps more to the east and eventually joined the Ghaggar-Hakra course before flowing as a combined river into the Rann of Kutch. Thus, the protohistoric picture of this part of the subcontinent is that of two majestic rivers flowing along the two edges of a vast alluvial plain and possibly joining together before forming a single delta at the mouth of the present Rann of Kutch.
The Hakra Ware Sites in Cholistan14
All that can be pointed out in this sector is the density of site distribution and the inferences drawn from the surface observation of these sites. The site survey in Cholistan undertaken by R. Mughal is among the most significant surveys in subcontinental archaeology. The Hakra ware (thus named because Hakra ware, a thick and underfired handmade pottery whose main form is a globular jar with a thick application of a slightly liquid mud paste mixed with broken bits of sherds, was first found located in this Hakra area) sites are 99 in number with the following size hierarchy: from 0.1 to 5 ha in size—21 sites; 5.1 to 10 ha—5 sites; 10.1 to 20 ha—7 sites; 20.1 to 30 ha—4 sites. A grading of the relative importance of these sites is clear and suggests a long background of adaptation to this area, of which we know nothing at present. A distinction has also been made between camp sites (52), settlement sites (45) and industrial sites with kilns (2). The sites are located mostly on mud-flats but some are on top of the fossilized sand dunes as well. Further, they stand out as distinct sites, with only six of them being occupied in later periods (two in the Kot Diji phase and four in the Indus civilization phase). The non-pottery finds include terracotta cattle figurines, terracotta and shell bangles, grinding stones, bits of copper and a microlithic industry. At present, the Hakra ware sites mark the first defined phase of village occupation in the Indus-Hakra plain. We have noted that the jar-type pottery with a slurry of mud applied to its body, which is typical of the Hakra ware complex, occurs at Sheri Khan Tarakai in Bannu and on this basis the beginning of the Hakra ware phase may be put, according to Mughal, at the very beginning of the fourth millennium BC. Mughal’s more recent explorations have also added some Hakra ware sites in Pakistani Punjab beyond Cholistan, extending up to the Faisalabad area in the central section of the Ravi-Chenab doab.
The Kot Diji Sites in Cholistan and Kot Diji in Sind
The densest distribution of the Kot Diji phase sites is not in Sind, where this phase was identified as marking the immediately antecedent level of the Indus civilization at the site of Kot Diji in the 1950s, but in Cholistan. Here 40-odd sites share a slightly different but overlapping area of distribution with the sites of the Hakra ware phase. A four-tiered site hierarchy is clear: 0.1 to 5 ha—19 sites; 5.1 to 10 ha—8 sites; 10.1 to 20 ha—3 sites; 20.1 to 30 ha—2 sites. Kilns are found at 14 sites. At Kot Diji, the Indus civilization levels overlie the Kot Diji culture levels after a deposit with traces of burning. The Kot Diji culture settlement comprised a fortified (mud wall on stone-block-built supporting wall, with bastions and revetments of mud-brick in places on the outside) ‘citadel’ complex and a habitation area at the lower levels of the site. Terracotta toys, plain and painted bangles, cakes, cones, beads, cattle figurines, miscellaneous objects of shell and bone and a microlithic industry occur alongside a rich pottery assemblage. A short-necked ovoid jar which is painted at the neck is considered almost a type-fossil, and among designs one notices ‘horned deities’, pipal leaves and fish-scales. The earliest calibrated point of dates from Kot Diji is c. 3300 BC, which tallies with the earliest point from Harappa.
Fig. 16 Kot Diji pottery (Khan 1965)
The Hakra Ware and Kot Diji Phase at Jalilpur
The site is near Multan in Pakistani Punjab and here gold, coral and semi-precious stone beads, plastered floors, chert blades, terracotta net-sinkers, bone points, etc. have been found in association with the Hakra ware in Period I. The Kot Diji culture is seen in Period II, but not much is reported of this culture here except pottery.
The Kot Diji Phase at Harappa15
About 70 km to the north-east of Jalilpur along the alignment of the Ravi is Harappa where the ongoing excavations have identified two periods of ‘Kot Diji occupation’. Period I—mud-brick walls, Kot Diji-related pottery, terracotta bangles and female figurines, a frequent blade industry and rare polished celts, and beads of lapis lazuli, carnelian and steatite—has been traced at the north-western edge of the mound E at the site. Period II—massive mud-brick perimeter walls with a few structural phases, terracotta toys, animal figurines, bangles, cakes, etc. and a systematic layout of habitational areas along a major north–south street—has been traced both in this section and along the southern edge of the mound. The Period II settlement covered more than 13 ha.
The Kot Diji/Sothi Phase at Kalibangan16
Sothi in the Ghaggar-Hakra belt of Rajasthan was excavated in 1952 and the similarities of its pottery with the pottery obtained at Kot Diji (excavated in 1955 and 1957) were pointed out by A. Ghosh in 1964. By then the pre-Indus civilization phase of Kalibangan was discovered, and on that basis Ghosh could speak of the ‘Sothi sub-stratum’ of the Indus civilization. Otherwise, it is the phase which is immediately earlier than the Indus civilization and which we here identify as the Kot Diji phase. It had two structural phases at Kalibangan and covered about 4 ha of fortified enclosure of uneven sides, with its entrance located at the north-western corner. The average width of the fortification wall was increased from 1.9 m in the first structural phase to 3–4 m in the second. In the south-eastern section a lane (1.5 m wide) has been identified. Ovens, lime-plastered storage pits and saddle querns have been found in mud and mud-brick houses with central courtyards. Beads of semi-precious stones, more than 100 copper objects, microlithic industry, terracotta, shell and copper bangles form the associated cultural material. To the south of the site, parts of a cultivated field (the north–south furrow marks spaced at 1.9 m and the east-west ones at 30 cm, the former possibly for mustard and the latter possibly for horse-gram) have been excavated. Close similarities with the Kot Diji pottery have been noted. Calibrated radiocarbon dates begin around 2900/2800 BC.
The Hakra Ware and Kot Diji/Sothi Phases at Kunal17
This site (c. 1 ha) in the Hissar area of Haryana has revealed three main phases of occupation antedating the Indus civilization. Before the houses of Period IA were built, the occupational area was raised by 0.71 m by dumping locally available red gravelly earth on it. The houses were built, first by creating 1.1 m deep and 2 m diameter pits with rammed floors and plastered sides. Post-holes around the pit-mouth suggest at least a 2 m high wattle-and-daub structure above it. In Period IB the pits became larger and their sides were now plastered with mud-bricks. The houses of Period IC were rectangular mud-brick (built generally in the ratio of 1:2:4 and 1:2:3) houses. The layout of this phase included a system of refuse bins and soakage jars placed on the streets. Refuse pits were also associated with the houses of the earlier phases. No radiocarbon date is yet a vailable for Kunal.
|
TABLE IV.6 |
|
|
IA: |
pottery: Hakra ware, a handmade black-and-red ware, dull chocolate burnished ware with wavy marks, a dull red with wavy incised lines on the outside, a red ware with designs in black outline and white filling (the most characteristic pottery of the period; designs including a bull with highly curved horns, and a variety of pipal leaves)—bone tools, micro blades of chalcedony, copper fish-hooks and arrowheads (of two types: hollow concave-based and inverted ‘V’ type). |
|
IB: |
pottery: continuation of all the earlier types, but also, a prolusion of Kalibangan I type pottery and ‘the first occurrence’ of sturdy red beakers and jars of the Indus civilization type. |
|
IC: |
gold and silver ornaments kept on a silver sheet and put in a red pottery jar and buried inside a house: two silver tiaras (both in the shape of a full-blown flower with pointed petals and a crown insignia at the top of the petals) and a silver armlet, in a different house: ‘a large number of gold ornaments’ including circular beads with axial tubes. Also, ‘large hoards’ of lapis lazuli micro-beads. 92 agate beads and faience and carnelian beads. Chert blades, copper coiled finger rings, coiled cones, arrowheads, fiat axes, fish-hooks, spearheads, a terracotta crucible with molten metal. Six. steatite and one shell seals with geometric designs. |
The Kot Diji/Sothi Phase at Banawali and Other Places in Indian Punjab and Haryana18
The detailed planning of this phase at this site, also near Hissar in Haryana, is not understood, but mud-brick houses (some low and squat with thatched roofs) contain hearths inside and plastered storage pits in the courtyard. The pottery is Kalibangan I type and a sherd shows the black painting of a canopied cart with spoked wheels. In addition to the presence of copper, lithic blade industry, terracotta objects, gold and semi-precious stone beads, etc. there is a cubical chert weight of 87.55 gm, supposedly 100 times of the Indus civilization weight unit of 0.857 gm. There is no radiocarbon date from this period at Banawali. Further, along the Ghaggar-Hakra drainage, the evidence of this phase has been excavated at places like Siswal and Balu in Haryana and Rohira and Mahorana in Indian Punjab. The assemblage is generally identical with the Kalibangan I assemblage, although date-wise some of these places may be later.