The Indus Civilization Page 9
The Early Harappan (c. 3200—2600 B.C.) is made up of four regional phases that are thought to be generally contemporary: the Amri-Nal, Kot Diji, Damb Sadaat, and Sothi-Siswal (figure 2.11). There is pronounced geographical expansion during the Early Harappan on to the Potwar Plateau and into the Indian Punjab, Haryana, northern Rajasthan, and western Uttar Pradesh as well as Gujarat. The chronology of the Early Harappan has been determined from a long series of radiocarbon dates. I have presented a discussion of this complex chronology elsewhere.50
The Amri-Nal Phase of Sindh, Baluchistan, and North Gujarat
The Amri-Nal Phase was first defined by Majumdar at the type site of Amri following his excavations there in December 1929-January 1930;51 however, the definitive excavation at this site was conducted by J.-M. Casal between 1959 and 1962.52 Exploration by Stein and B. de Cardi in southern Baluchistan has shown that assemblages sharing some features of period I at Amri are also found in the mountains there. While Nal material is found in both southern Baluchistan and Sindh, it is perhaps more at home in the highlands than the riverine plains. The two assemblages have been merged into a single regional phase; in Sindh, the Amri side tends to predominate, and in Baluchistan, the assemblage is more Nal-like (figure 2.12).
The Amri ceramic assemblage is made up of extremely well made fine wares, generally fired light red or buff. Red and buff slips are also found, often with black paint. At the beginning of the phase the designs are exclusively geometric, developing into more curvilinear motifs toward the end. Typical of Amri ware are open bowls and jars and tall vases with simple, featureless rims.
Nal ceramics are among the best-made and most attractive wares of prehistoric times in South Asia. They, too, are fine wares, and tend to have been fired buff to pink. The slips have a tendency to be very light, buff or weak red, giving a tint to the surface, rather than a dense overall color. The characteristic vessel forms are canisters and straight-sided bowls, with simple, knife-edge rims. Polychrome infilling of these designs includes the use of red, pink, blue, and yellow. Painting in white over a black slip is also known and is one of the features shared with Amri. In fact, the use of white paint is a hallmark of the Early Harappan (figure 2.13).
The Expansion of Food-Producing Peoples into Gujarat
Evidence for the early expansion of farmers and herders into Gujarat appears at a number of sites, the most prominent of which are Dholavira, Surkotada in Kutch and Moti Pipli, the Santhli sites, and Nagwada in North Gujarat (figure 2.14).
The sites in North Gujarat have ceramics suggesting a combination of Amri-Nal and Kot Diji stylistic features. Padri, Somnath, and Loteshwar have ceramics that are quite different from those of the Early Harappan and suggest an even earlier presence of pottery-making peoples in the region—a good example of one of the problems on which important, productive archaeology can be done in India.
Period I at Dholavira has ceramics that have typological similarities with Amri Period II (Transitional Stage),53 Surkotada is on this list since a reanalysis of the ceramics from the small cemetery, 300 meters north of the habitation site, has shown that this was a place of interment for Amri-Nal peoples.54
Figure 2.11 All Early Indus sites
The General Model of Expansion
Diffusion and the spread of peoples and their cultures is one of the oldest and most intractable topics in archaeology. The general hypothesis used here is that prehistoric pastoral peoples were usually the lead element in exploring new territory in the Greater Indus region. The reasoning behind this has to do with ecology and the mobility of these people and their need for new pasturelands, plus the economic rewards and prestige that would come with the discovery of new resources. But we cannot discount other motivations in ancient exploration: the pleasure and adventure of travel and exploration of new territory, the opportunity to meet new people even if possible conflict might be involved.
Figure 2.12 Amri-Nal ceramics (Amri 1-18; Nal 19-27) (after Hargreaves 1929)
Settlement Pattern Data on the Amri-Nal Phase
The settlement pattern data for the Amri-Nal Phase (shown in table 2.11) document the continuation of a pattern of village farming communities and pastoralism. The presence of Nal ceramics in Sindh and Amri Wares in Baluchistan seems to indicate that groups of people who kept sheep, goats, and cattle spent summers in the highlands, had winter abodes in the Indus Valley, and ventured into the seasonally quiescent floodplains of the Indus Valley to gain access to resources there. This pattern of archaeological evidence may not signal the earliest pioneering of this environment, but it does seem to mark a time when it would have been a significant part of the subsistence pattern.
Table 2.11 Estimates of settled area for the Amri-Nal Phase
Total sites known 164
Sites with size estimate 88
Settled area of sites with known sizes 322.64 hectares
Sites with size unknown 76
Average site size 3.67 hectares
Estimated settled area of sites without size 278.92 hectares
Estimated total settled area 610.56 hectares
The penetration of the deeper riverine zones of Sindh by pastoralists during the dry season might also mark the beginning of the process of burning, clearing, and perhaps draining this landscape. The search for grass and browse, cutting wood for temporary shelters, campfires and firing ceramics, along with the possibility of large-scale burning to create a more efficient, productive landscape for humans and animals would constitute a kind of “softening up” of this difficult terrain, preparing it for farming in Mature Harappan times. The use of fire for clearing land is not well documented in the archaeological record. But regional, manmade landscapes almost certainly began to emerge from the millennia of intensive use by domesticated animals as well as clearing land for cultivation.
Figure 2.13 Ceramics from Nal (after Hargreaves 1929)
The villages along the piedmont of Baluchistan were placed to take advantage of the hill torrents or nais and the natural springs that dot the outer face of the mountain front. Lake Manchar, the natural inundation basin of the Indus, was extensively utilized as an environment to be directly exploited for food (fish and shellfish, seasonally for birds) and as a huge, naturally irrigated farming tract, resulting from the seasonal expansion and contraction of the lake waters, as at the site of Lohri.55
Amri-Nal Site Size
Some Amri-Nal sites were very small camps, such as Jebri Damb One or Kuki Damb in Kalat, at one-tenth of a hectare, or Santhli Four in Gujarat at one-hundredth of a hectare. The largest Amri-Nal site in the “Gazetteer of Settlements of the Indus Age”56 might be Dholavira, on Kadir Island in Kutch if it comes close to the size of the Mature Harappan settlement at 60 hectares. There is also some evidence that the first occupation there took place during the Transitional Stage, not the Early Harappan.
Sites of the Amri-Nal Phase cover the southern tier of the Indus Valley and Baluchistan. During this phase, the continued maturation of the subsistence system and a spread of farming and herding peoples to the seacoast and southeast into Gujarat is seen. Modest attempts at fortification may also characterize some Amri-Nal settlements.
The Kot Diji Phase of Northern Sindh and Elsewhere
Kot Diji is a splendid small site on the national highway linking Karachi and Hyderabad to Sukkur. It is situated on the old alluvium of the Indus Valley, below a huge Talpur Dynasty fortress of the nineteenth century A.D., which looms over it from the escarpment of the Rohri Hills.57
The Kot Diji archaeological assemblage is distinct from the Amri-Nal, although there is some overlap in ceramics, with some common vessel forms and decorative motifs especially among the simpler pots (figure 2.15). Both assemblages present us with extremely fine examples of the potter’s art with well-fired red and buff wares. The tall jars and vases with featureless rims of the Amri-Nal assemblage are not a part of the Kot Diji ceramic corpus and neither are the Nal canisters and fat-bodied pots. If anything like the distinctive
Nal painting was found at a Kot Diji site, it would be dubbed an import. Settlement data for the Kot Diji Phase are given in table 2.12.
Figure 2.14 Early Indus and early third millennium sites in Gujarat
Table 2.12 Estimates of settled area for the Kot Diji Phase
Total sites known 111
Sites settled area of sites with known 83
Sizes with size estimate 523.38 hectares
Sites with size unknown 28
Average site size 6.31 hectares
Estimated settled area of sites without size 176.68 hectares
Estimated total settled area 700.06 hectares
The Damb Sadaat Phase of Central Baluchistan
Contemporary with the Kot Diji and Amri-Nal Phases is a smaller, more localized cultural phase of the Early Harappan, centered on the Quetta Valley. It rests on a long history of occupation in this fertile, well-watered valley. Quetta-Pishin is blessed with substantial subsurface water resources, available even to relatively primitive cultivators in the form of artesian wells. This valley is also the center of a natural corridor linking southern Afghanistan to the Indus Valley via the Bolan and Khojak Passes. Historically, these factors have made it a regional hub of settlement, trade, travel, and administration. In prehistoric times a distinctive set of archaeological assemblages developed in Quetta-Pishin and the valleys to the immediate north, south, and beyond into Afghanistan, as at the famous site of Mundigak, and even reaching Shahr-i Sokhta in Seistan.
Damb Sadaat Phase Sites
There are thirty-seven Damb Sadaat sites, twenty-nine of which have data on size. They average 2.64 hectares. The largest site is the Quetta Miri (23 hectares), located at the one spot in the Quetta Valley that has been occupied continuously from prehistoric to modern times. The next largest site, Mundigak, is 18.75 hectares and is in the Kushk-i Nakhud Valley of the Helmand River drainage, over 200 kilometers to the northwest of the Miri. Mundigak was a town during Early Harappan times. It is entirely possible that the settlement was smaller than the 18-plus hectares estimated here. Settlement figures are given in table 2.13.
Pottery of the Damb Sadaat Phase was often slipped to create a uniform surface, even if it was not painted (figure 2.16). Paring was often done to thin the walls of vessels and the stems of goblets. A painted ceramic called Quetta Ware is the most characteristic type of pottery.58 Short-neck, globular jars of Kot Dijian type also occur. A ceramic known as Faiz Mohammad Grey Ware was also manufactured. It is a fine ware ceramic, generally a plate, decorated with Quetta Ware designs. The technology used in its manufacture is somewhat complicated since it took place in two stages of firing: first baked as a red-buff ware in an oxidizing atmosphere, then refired under reducing conditions to obtain the finished gray color.59 The designs in black paint (sometimes red) are closely paralleled in the Quetta Ware repertory of motifs.
Figure 2.15 Kot Diji ceramics from various sites
Table 2.13 Estimates of settled area for the Damb Sadaat Phase
Total sites known 37
Sites with size estimate 29
Settled area of sites with known sizes 76.42 hectares
Sites with size unknown 8
Average site size 2.64 hectares
Estimated settled area of sites without size 21.12 hectares
Estimated total settled area 97.54 hectares
The Sothi-Siswal Phase of the Eastern Region
To the east, on the other side of the Greater Indus region, there were also important developments taking place during the Early Harappan Stage. These began along the natural route of the Sarasvati River and its Indian tributaries, especially the Drishadvati or Chautung, but it spread into Punjab, Haryana, with one lone site (Nawanbans) in the Ganga-Yamuna doab of western Uttar Pradesh. The name of this phase, “Sothi-Siswal,” is derived from the names of two excavated sites (figure 2.17).
Figure 2.16 Ceramics of the Damb Sadaat Phase from sites in the Quetta Valley (after Fairservis 1956)
Sothi-Siswal Settlement Patterns
There are 165 Sothi-Siswal sites, 91 of which have information on site size. The average settlement size is 4.28 hectares. There are two Sothi-Siswal settlements larger than 20 hectares. Settlement data are given in table 2.14.
Conclusions for the Early Harappan
The Early Harappan has little evidence for a significant degree of social differentiation, craft and career specialization, and little evolution of the political and ideological institutions that produce public architecture.60 These sociocultural features are in marked contrast to levels of developments inferred for the Mature Harappan and must be used to shape perceptions of the nature of Mature Harappan society, its institutions, and the historical, developmental process that took it from the Early Harappan Stage to urbanization. There is, however, good evidence for a growing population from the beginnings of food production to the threshold of civilization, as seen in table 2.15.
Table 2.14 Estimates of settled area for the Sothi-Siswal Phase
Total sites known 165
Sites with size estimate 91
Settled area of sites with known sizes 389.15 hectares
Sites with size unknown 74
Average site size 4.28 hectares
Estimated settled area of sites without size 316.72 hectares
Estimated total settled area 705.87 hectares
Figure 2.17 Sothi Siswal ceramics from various sites.
The graph of Early Harappan settlement size by phase gives a sense of the size distribution of sites. It suggests that the geographical center of cultural dynamics was to the north, with the Kot Dijian and Sothi-Siswal Phases (figure 2.18). It also documents the relatively small size of even the largest Early Harappan sites, as compared to those of the Indus Civilization (figure 2.19).
Table 2.15 Settlement data on the Indus Age through the Mature Harappan
Disruptions at the End of the Early Harappan
The disruption and/or abandonment of sites toward the end of the Early Harappan is an important part of the record for this phase. Sites of central concern are Balakot, Kot Diji, Gumla and Kalibangan, Amri, and Nausharo.
Balakot Period I at Balakot, termed “Balakotian” by the excavator, appears to be a regional variant on the Amri-Nal theme. The radiocarbon dates for Periods I and II (Mature Harappan) suggest “a hiatus of several centuries between the Balakotian and the Harappan periods at Balakot.”61
Figure 2.18 Graph of the site size distribution for the four phases of the Early Indus
Kot Diji The stratigraphy of Kot Diji is complex and in some ways not yet completely understood. There are obvious signs of massive burning over the entire site, including both the lower habitation area and the high mound. As F. A. Khan has observed:
A thick deposit of burned and charred material, on top of layer (4), spreading over the entire site, completely sealed the lower levels (Kot Diji) from the upper ones (Mature Harappans). This prominent and clearly marked burnt layer strongly suggests that the last occupation level of the early settlers (that is the Kot Diji) was violently disturbed, and probably totally burnt and destroyed.62
Gumla Period III at Gumla, the Kot Diji occupation, seems to have come to a fiery end, with an ash layer separating this occupation from the succeeding Late Kot Diji. “The end of their period appears to be violent. There is a thick layer of ash, charcoal, bones, pot-sherds, etc. which all belong to period III.”63
Amri Period II at Amri, the Transitional Stage, ended with signs of a significant amount of burning. Casal has observed: “The upper levels are blackish and ashy, but they are mostly so near the surface that it is difficult to say whether this occurrence should be interpreted as evidence of some sort of violence or of fire.”64 A large fire was also evident in Period ID at the site, associated with a building that Casal believes was a godown, or storage facility.
Nausharo The excavations at Nausharo produced evidence of extensive burning associated with Period ID, a Transitional Stage occupation: “The two architectural
complexes of Period ID . . . have been heavily burnt and the walls have turned red due to heat. At Kot Diji too, the final phase of the pre-Indus period has been destroyed by fire.”65
Kalibangan There is an abandonment of Kalibangan between Periods I (Early Harappan) and II (Mature Harappan). The excavators detected signs of cleavage and displacement of strata in the Early Harappan levels, which might have been caused by an earthquake.66 According to B. K. Thapar:
This occupation continued through five structural phases, rising to a height of some 1.6 m, when it was brought to a close by a catastrophe (perhaps seismic), as evidenced by the occurrence of displaced (faulted?) deposits and subsided walls in different parts of the excavated area. There-after the site seems to have been abandoned, though only temporarily and a thin layer of sand, largely infertile, accumulated over the ruins.67
Figure 2.19 Graph of the site size distribution for the four phases of the Mature Indus