The Indus tradition and the Indo-Aryans
Subhash C. Kak,
Louisiana State University,
Baton Rouge
Vol. 32, Mankind Quarterly, 04-01-1992, pp 195.
The earliest Indian literature has astronomical allusions to events of the 3rd millennium B.C. and earlier (Tilak 1893, Sengupta 1947, Kak 1987b). The setting for the hymns of the Rigveda is the area of Sapta Saindhava, the region of Punjab bounded by the Indus and the Ganga rivers. There are also geographical references in the Rigveda that are of chronological value. In the past 5000 years north India has undergone major tectonic and hydrological upheavals, and so it becomes possible to correlate certain geographical references to different time epochs. Other significant references are to sea-going vessels and to settlements on the Sarasvati (Ghaggar-Hakra river, which is now dry) that also provide significant clues. The archaeological record suggests that this river turned dry around 1900 B.C.
The genealogies of the Puranas and the later Vedic literature also reach back at least into the third millennium B.C. (Pargiter 1922, Roy 1983). This later literature, starting with the Satapatha Brahmana, describes an expansion of the civilization outside of the original area of the Indus and the Sarasvati valleys. There are also references to geophysical changes in the later Vedic literature.
Archaeological investigations have also shown that the Indus cultural tradition represents the beginnings of the Indian civilization. This tradition has been traced back to about 6500 B.Cin remains that have been uncovered in Mehrgarh and other sites (Jarrige and Meadow 1980, Possehl 1982, Kenoyer 1991, Shaffer 1991). It reached its flowering during the period 2600-1900 B.C., called the Harappan age, when several cities and towns were established and writing, based on the so-called Indus script, was used. In the 1970's it was found by M.R. Mughal (1982) that most of these towns and settlements were on the banks of the Sarasvati river. One of the most striking insights of the continuing studies is that "there were several ethnic groups in the Indus Valley throughout the third and second millennium B.C., of which the Harappan was only one. Rather than an area dominated by Harappan culture, the Indus Valley saw the development of a complex cultural mosaic of related but distinct ethnic groups" (Shaffer and Lichtenstein 1989). Hydrological changes and the socio-economic evolution of the groups led to an abandonment of large areas of the Indus valley. The tradition continued in a state of decline to 1000 B.Cand perhaps later. A second urbanization began in the Ganga-Yamuna valley around 900 B.C. The earliest surviving records of this culture are in Brahmi script. This second urbanization is generally seen at the end of the Painted Gray Ware (PGW) phase (1200-800 B.C.) and with the use of the Northern Black Polished Ware (NBP) pottery. Late Harappan was partially contemporary with the PGW phase. In other words, a continuous series of cultural developments link the two early urbanizations of India.
The restructuring of society that occurred between the two urbanizations is partially mirrored in the restructuring of the Indus script. The Brahmi script, which was in use throughout India during the reign of Asoka [reigned 269-232 B.C.], was highly systematic, reflecting clearly the theories of Indian grammarians. Literary evidence as well as signs on early punch-marked coins suggests that writing in India during the second urbanization goes back much before the middle of the first millennium B.C. The punch-marked coins that go far back to at least the seventh century B.C. (Mitchiner 1973) use a Harappan weight standard. It appears that the coins were originally issued as silver blanks by traders and the weights checked by traders. The checking was represented by marks that are strikingly similar to the Harappan signs. By the sixth century B.Cthe kings began putting their own issuing marks on the coins. These pictorial marks were generally representative of the meaning of the king's name. The Indus script uses many more signs than Brahmi and it is generally written from right to left, in a direction opposite to that of Brahmi. On the other hand, there are instances of both the scripts being written in the boustrophedon style, that is, written in opposite directions in alternate lines. But the change in the normal direction of writing does indicate a fundamental shift. Whether this shift took place between the two urbanizations or just prior to the Mauryan empire is not dear.
The spread of the Indian civilization, as indicated by the literary and the archaeological records, can further be checked by an analysis of the processes that might have supported this spread. Ammerman and Cavalli-Sforza (1984) argue for the parallel situation in Europe that the spread there of farming after 7000 B.Ccould be seen as a diffusionary process brought about by a combination of cultural diffusion, and population growth and displacement, which they call demic diffusion. Sokal, Oden, and Wilson (1991) have published genetic evidence supporting this model. Genetic variation across the Indian sub-continent suggests that a diffusion model might be at the basis of the spread of the Indian people as well.
We also call attention to the dating of 4th millennium B.Cfor a copper-based head found in Delhi (Hicks and Anderson 1990). It has been argued by Hicks and Anderson that the head represents Vasishtha, who is one of the authors of the Rigvedic hymns, or his king Sudas. This dating supports the traditional chronology of the Vedic literature.
The evolution and restructuring of the ethnic groups (Shaffer 1991) can be seen in the archaeological record. No evidence for any break in the Indian tradition due to any invasions has been found. Jim Shaffer (1984) who sums up the archaeological evidence summarizes:
Current archaeological data do not support the existence of an IndoAryan or European invasion into South Asia at any time in the pre- or proto-historic periods. Instead, it is possible to document archaeologically a series of cultural changes reflecting indigenous cultural development from pre-historic to historic periods. The early Vedic literature describes not a human invasion into the area, but a fundamental restructuring of indigenous society that saw the rise of hereditary social elites . . . The Indo-Aryan invasion(s) as an academic concept of the 18th and 19th century Europe reflected the cultural milieu of that period. Linguistic data were used to validate the concept that in turn was used to interpret archaeological and anthropological data. [page 88]
Since the Indo-Aryans were the dominant group in northern India by the late 2nd millennium B.C., one is compelled to the conclusion that they must have been one of the ethnic groups in the Indus and the Sarasvati valley areas as early as 6500 B.C.
In this essay we show that there is enough convergence in the archaeological evidence about the Indus tradition and the literary and the geographical evidence of the Vedic literature to conclude that the Indus tradition was essentially Indo-Aryan. Nevertheless, since the Indus tradition was characterized by several ethnic groups (Shaffer 1991, Kenoyer 1991), it is very likely that several languages, some of them non- Indo-Aryan, were also present in the area. Our conclusion that the Indo-Aryans represented the Harappan ethnic group is based on our analysis of the Indus script which seems to have been used for an Indo-Aryan language (Kak 1987a, 1988, 1989, 1990) and on the geographical evidence from the Vedic literature. A considerable part of our paper is, therefore, devoted to a review of our analysis of the Indus script. Astronomical evidence already summarized elsewhere (Kak 1987b, 1991) will not be reviewed. But it should be noted that such evidence also supports the conclusions of this paper.
The Indus Tradition
We begin with a brief review of the Indus tradition. In its earliest phase this tradition was characterized by cultivation and animal husbandry. Cattle pastoralism was an extremely important component of the economy. It is estimated that as early as 5500 B.Cdomesticated cattle were already central to food production. In this respect the Indus tradition is different from the tradition of Mesopotamia which emphasized domesticated sheep and goats. Jim Shaffer (Shaffer and Lichtenstein 1989, Shaffer 1991) views the evolution of the culture in the Indus region in four broad eras. The first is the early food producing era (c. 6500-5000 B.C.) that is characterized by an absence of ceramics. The next is the regionalization era (5000-2600 B.C.) where distinct artifact styles (including ceramics) develop regionally. The third is the integration era (2600-1900 B.C.) where we see pronounced cultural homogeneity and the development of urban centers. The fourth era is that of localization (1900-1300 B.C.) where characteristic patterns from the integration era are seen to be blended with regional ceramic styles. This era indicates decentralization and restructuring of the interaction networks.
Buildings were made out of mud bricks and fired bricks and stone. There was public architecture as in plazas, streets, public buildings, wells, drains, and tanks. Pottery was mass produced by using wheels and sometimes by molds. Painted decorations used a variety of geometric, animal, and floral motifs which are still popular in India. A network of long distance trade existed. Turquoise from central Asia, lapis lazuli from northern Afghanistan, and shells from the coast of the Arabian sea have been found at Mehrgarh.
The Indus tradition consists of several overlapping cultures and styles, that probably represent different ethnic groups. Figure 1 presents a chart that shows the relationships between some of these groups. The richest period of this tradition is named Harappan after the site where the first excavations were made. While the ruins of Harappa had been known for a long time, their importance was appreciated only after Daya Ram Sahni in 1921 conjectured that they may belong to the pre-Mauryan era. A year later R.D. Banerji opened a trench in Mohenjo- daro 400 miles away and he suggested that the two ruins were related. But it was an article by John Marshall in 1924 in the Illustrated London News on the Indus Valley discoveries which spread the story far and wide. Experts working on Mesopotamian antiquities pointed out connections with finds from that area which went back to the 3rd or 4th millennium B.C. The ethnic and the regional differences amongst the Harappans themselves are being revealed by the more recent work. Thus Possehl and Raval (1989) have found two different styles amongst the Harappans in Gujarat.
Harappa was in a poor state when archaeologists started examining it because engineers had mined the ancient bricks from the site decades earlier to lay the roadbed of about 100 miles of the Lahore-Multan railway line. How much of priceless evidence was lost there one will never know. Since then hundreds of other sites have been discovered. These include major sites at Ganweriwala, Rakhigarhi, Kalibangan and Lothal. The Harappan world covered an area of 650,000 to 800,000 square kilometers that stretches from the Himalayas in the north to the Tapti river in the south, and from the Indus river valleys in the west to the plains of the Ganga and Yamuna rivers in the east (Kenoyer 1991). Most of the sites have been found on the greater Indus plain, which was formerly watered by two major river systems, that of the Indus and Sarasvati. A recent count has identified close to a thousand settlements. Figure 2 gives a map of the geographical features of northwest India and adjacent regions in 3rd millennium B.C. Note that in this reconstruction the Sarasvati flows independently into the sea.
While the Harappan city seemed to evolve out of an irregular net plan, it had two distinctive elements. In many cases to the west lay a "citadel' on a high platform that based the public and ceremonial buildings. To the east was the lower city with straight, wide main streets that divided the city into large blocks. The blocks in turn were served by narrow curving lanes. The houses were built of generally standardized burnt bricks. They were planned as several rooms around a square courtyard, and were often of two or more stories. Some houses had bathrooms which were connected by bricklined drains to sewers under the main streets. Many windows were screened with grilles of terra cotta or alabaster. Houses presented blank walls to the streets and, in many cases, the doors opened on the side lanes. All this presents a picture reminiscent to the architecture found in a traditional city in contemporary Punjab.
One of the striking buildings in the citadel at Mohenjo-Daro is the great bath. The oblong bath is 39 feet long, 23 feet wide, and 8 feet deep. It was sunk into the paving of a courtyard and it was approached from the north and south by brick steps with possible wooden stairtreads. The floor of the bath sloped to an outlet that led in turn to an arched drain deep enough for a man to stand upright. Just north of the pool were eight small bathrooms drained by little runnels in the floor. Each bathroom had its own staircase leading to the second storey which may have housed cells for the priests, if the whole complex was like a tank of a Hindu temple.
The largest building uncovered at Mohenjo-Daro was probably a palace of size 230 feet long and 78 feet wide. At Harappa a building twice this size, that may have served as a granary, has been discovered. The iconography includes figures of a Mother Goddess and that of bearded **** men in a posture the Jains call kayotsarga, where the teacher's posture is upright with legs slightly apart and the arms are parallel to the side of the body but do not touch it. But the most striking Harappan deity is a horned god in a yogic posture, who is strongly reminiscent of Shiva in his Pashupati form. Amongst the ornaments of the Harappans beads of different materials and sizes have attracted scholarly attention.
There is a distinct difference in the brickmaking of the two urbanizations. The Harappans are known to have used several brick sizes: 30 x 15 x 7.5 centimeters and 40 x 20 x 10 centimeters being the most common, where in these cases and others the length, breadth, thickness ratio is about 4:2:l. The bricks used during the second urbanization of the Ganga valley are not according to these ratios, one type encountered there has the ratios 7:4:1. But this departure may merely be a result of the popularization of one of the existing styles in a multi-ethnic and multi-cultural society. We do not observe a complete break with the earlier tradition, but only reorganization and re-adjustment. Thus corbelled drains, characteristic of the Harappans, persist in Ganga valley.
The Harappan settlements along the Sarasvati were located on the desert scarp above the entrenched river. On the other hand, the settlements during the age of Painted Grey Ware (1200-800 B.C.) are found within the entrenchment on the bed of the river (Possehl and Raval 1989). This establishes two important chronological markers. First, the river had dried up before the PGW era. Second, the Rigvedic age which describes the Sarasvati river as one of the largest of its times lies prior to the drying up of Sarasvati. There are reasons to assume that the abandonment of the Harappan sites in the Sarasvati valley region was caused by this drying up. This suggests the epoch of 1900 B.Cfor this change.
Indus Writing
The surviving records of the writing of the Harappans are mainly carvings on seals, small pieces of soft stone, and copper tablets. The total number of inscribed objects is around 4200, but many of these are duplicates (Mahadevan 1977, Joshi and Parpola 1987). The number of different signs used is close to 400, but these include the various numeral signs as well as the conjuncts of the more basic signs. Most texts are very brief, the average length being 5 signs, and the longest text, on a three-sided 'amulet', is 26 signs long. The longest inscription on a single side is 17 signs, in three lines, on a seal. The primary purpose of the seals was perhaps to mark ownership and the copper tablets may have served as amulets. A large number of seal impressions on clay have also survived. These are likely to have served as tags which were attached to bales of goods, for the reverse sides often show traces of packing materials. The impressions of the seals are likely to have served as signatures. The pictorial motifs that accompany the writing include the humped bull, buffalo, elephant, tiger, rhino, crocodile, antelope, fish, tortoise, and so on. Geometric designs include the swastika, spoked wheel, and a circle with a dot. These pictures are similar to the ones that show up in the seals of two thousand years later.
The Harappans did maritime trading, and their seals have been recovered in Mesopotamia from the 24th century B.Conwards. On the other hand Persian Gulf seals have been found in the Harappan port of Lothal. Inlands the Harappans moved their goods using wheeled carts, camels, and boats. They used strikingly accurate weights in a series that is preserved in later Indian weights. The same unique series is also found on the island of Bahrain in the Persian Gulf suggesting this might have been their colony. Some of the weights are so tiny that they could have been used by jewelers to measure gold, others are so big that they must have been hoisted by ropes. Their products would have included fine pottery wares, jewelry, copper and bronze vessels, and woven cotton goods. The variety and extent of this trade indicates that credit-keeping and calculations were very important to the Harappans.
The seals of the historical period, from the time of Asoka and later, also carry brief texts. In almost all cases the legends end in the genitive case, representing ownership. The exceptions are where no case-ending is used, or where the ending is nominative as in religious formulae. The impressions from these seals, like the earlier seals of the Harappan period, were used to authenticate records, or to serve as signatures.
Indian scripts have undergone continual change for centuries and by the medieval times the earliest scripts had been quite forgotten. James Prinsep deciphered Brahmi in 1837 and this allowed the edicts of the emperor Asoka to be read. In the late 19th century G. Buhler argued that many Brahmi letters were derived from a North Semitic script. G. Ojha soon demonstrated flaws in these derivations.
In Brahmi, as in later Indian scripts, each letter represents a consonant combined with a. Combinations with other vowels are represented by the use of distinctive marks which modify the basic sign. Two consonants together were expressed by placing the signs for the two, one on top of another. This process of combinations makes the total number of distinctive Brahmi signs to be 330 for the 33 consonants alone, without considering the conjuncts. It is not surprising, therefore, that Indus has about 400 signs, and many of these signs are modified in exactly the same regular manner as in Brahmi.
C.S. Upasak in 1957 divided Brahmi into two groups: the basic signs, and the secondary or the derived signs. This division was based on morphological considerations. It is significant that it is these basic Brahmi signs that look closest to the Indus signs. Many of the Brahmi signs are the first syllables of familiar objects: thus g, ch, m, s, h appear to have been derived from the representations of girl (hill), chatra (umbrella), matsya (fish), sara (arrow), and hasta (hand).
It is tempting to assume that a systematically designed script such as Brahmi would represent a simplification of an earlier script. And with the parallel of ancient Egyptian writing in mind, one may further assume that the ancestor of Brahmi may have included logograms as well as determinatives in addition to the phonetic signs. These determinatives in Egyptian indicated grammatical categories; thus an egg-shape following a name indicated a feminine name, and a name in a 'cartouche' represented royalty. Assuming Indus to be this ancestor, one would expect a core syllabary together with many logograms and other signs. The decipherment would best proceed by concentrating on this syllabary first.
How difficult might this decipherment be? Historically, the easiest writings to decipher were the ones with the least signs, like the Phoenician and Ugaritic writings, that consist of only 22 to 30 signs or Brahmi, where the shapes were preserved greatly in the successor Nagari script. The syallabaries of Old Persian, Cypriot, and Linear B were more difficult because the number of signs was greater. The decipherment of logo-syallabic writings, such as Egyptian hieroglyphic, Akkadian cuneiform, Hittite hieroglyphic, and Indus have posed the greatest problems since each writing here involves hundreds of signs.
G.R. Hunter in 1934 and J.E. Mitchiner in 1978 published significant studies relating Indus and Brahmi. However these studies did not provide conclusive arguments establishing the relationship. By the mid-1980' s no one had yet tried to scientifically relate the Indus signs to the later Brahmi letters. This was despite the fact that many inscriptions in Brahmi looked almost identical to Indus inscriptions. The absence of a scientific study of the two scripts was due to the influential view, fostered by many scholars in the 19th century, that Aryans could not have been present in India before the arrival of the Indo-Europeansin Greece. Furthermore, it was taken as axiomatic that the Indus script could not have been used for an Indo-European language. But no archaeological evidence of the assumed invasion of India by the Aryans has been found.
My analysis of Indus and Brahmi based on computer created concordances revealed obvious connections between the two scripts that could not be explained as arising out of chance (Kak 1987a, 1988, 1989, 1990). Such an analysis is possible since letters in a script occur with different probabilities. These probabilities may also be characteristic of the language used. Thus for English e occurs about 12 percent of the time followed by t at 10 percent, a and o at 8 percent, and so on; at the other end of the spectrum q and x are at about 0.2 percent, and z is at about 0.1 percent. In fact such differences in frequencies make it possible to break substitution ciphers. My analysis showed that the most frequent letters of Indus and Brahmi looked almost identical and besides they were in the same order of frequency. A probabilistic analysis confirmed the connection between the two scripts (Kak 1988). One does encounter changes in the orientations of the signs. For example the fish sign has been flipped over. But such modification can also be seen in the evolution of Brahmi to the later Nagari, where many signs have been turned sideways or upside-down.
Briefly the connections between Indus and Brahmi scripts are as follows. Both scripts use conjuncts where signs are combined to represent compound vowels. The core set of most frequent Indus signs seems to have survived without much change in shape into Brahmi where it corresponds to the most frequent sounds of Sanskrit. The writing of numerals in Indus, especially the signs for 5 and 10, appears to have carried over to Brahmi.
Next I turned to the structure of the inscriptions. These are generally only about less than 10 letters long and there is other evidence that suggests they are often proper names indicating possession. Such indications are made by grammatical case-endings. Thus in English adding an s to a name shows ownership. For example David's means ownership by David. The genitive case-ending in Sanskrit is often sya or sa and in Prakrit the ending is generally sa or ssa [like s in English] and this is what we frequently see in these inscriptions. This suggests that the Indus language is likely to have been Prakritic. Note that the sign value for the case-ending was obtained independently through frequency considerations. Such independent cross checking is absolutely essential in decipherments.
Now I turned to specific inscriptions for further checks (Kak 1988). Here the attested contacts between Sumer and Harappa turn out to be invaluable. Sumerian documents mention the regions of Magan, Meluhha, and Dilmun as lying to the east of their land. Dilmun is identified by most scholars to be the island of Bahrain in the Persian Gulf, Magan is taken to be the coast of Makran in Baluchistan, and Meluhha is considered to refer to the region of the Indus valley. The Sumeriologist S.N. Kramer in 1952 in a translation of a Sumerian epical story 'Enmerkar and the Lord of Aratta' found that a fourth region to the east is described as being Bad Imin, which if freely translated represents 'the land of seven cities' or 'the land of seven high places'. (This is from bad meaning 'city', ia meaning 'five', and min meaning 'two' .) Now the Vedic Indians called their land Sapta Saindhava, which Harold Bailey suggested originally meant 'the land of seven high places' . J.V. Kinnier Wilson (1974) identified a commonly occurring combination of Indus characters as representing Bad Imin or Sapta Saindhava on the basis of parallels with Sumerian writing. I found that these very signs are read just the same using my Indus-Brahmi theory (Kak 1987a). This provides evidence of commonality between the Harappan and the Vedic worlds. Unfortunately, the phonetic values for the most frequent Indus signs do not help us in reading most of the seals and other texts. The reason for this is that the short lengths of these texts disallows unambiguous readings. Further, progress in the decipherment of the logograms has been slow.
To return to the question of the evolution of writing it needs to be added that the pottery marks in late second millennium B.Care reminiscent of the Indus signs. It is reasonable to assume that this was the period when the logosyllabic Indus was being reorganized into a proto-Brahmi script.
Although the demonstration that Brahmi is derived from Indus does not, by itself, establish that the Harappan people were Indo-Aryan, the structural similarities in the Indus and the Brahmi texts do point to that conclusion. But there remains the possibility that the Harappan civilization was not predominantly Indo-Aryan, because a small elite could have imposed its language and script on the people. But it reinforces the other continuities between the two urbanizations that the archaeological discoveries of the past few decades have revealed. Furthermore, it agrees with the independent proposals by Gimbutas (1973, 1985), T.V. Gamkrelidze and V.V. Ivanov (1985) [see also D' iakonov 1985], and Colin Renfrew (1987) all of which, though with radically differing interpretations, posit a dispersal of the IndoEuropean languages much earlier than previously thought. We will not go into the relative merits of these proposals here.
The relationship between Indus and Brahmi is one more piece of evidence that interlocks with other similar evidence from archaeology and literature linking the Harappan and Ganga civilizations. It opens up a new direction for a further study of the Indus script. It could also stimulate interest in the need for further studies between the rise of writing in Mesopotamia and the Harappan world.
Geographical Evidence from the Rigveda
The geography of Rigvedic India presents several clues that are significant for chronology. The main home of the Vedic Indians was the region between the Sarasvati and the Drishadvati rivers. But this was just the heartland of the wider region of Sapta Saindhava that extended from the Indus to the Ganga. This area is described as a region that is bounded by four seas (lakes). Bhargava (1964) theorizes that these lakes were in the courses of Indus, Ganga, Sarasvati to the three sides of Punjab and the lake in the valley of Kashmir. There is also several descriptions of the rivers of this region draining into a sea and of ships, some of which were equipped with sails. In the Vajasaneyi Samhita it is mentioned that five rivers feed into the Sarasvati. This suggests that trade via rivers and the sea must have been important then. Sarasvati is praised the most out of all the rivers which indicates that many of the Rigvedic hymns must have been composed at its banks. By the Brihmana period, this river had disappeared. Also after the period of the Samhitas, the river Indus is rarely mentioned, suggesting that the focus of the civilization had shifted a good distance (Macdonell and Keith 1907). In the Rigveda itself Ganga is mentioned only once (10.75.5) in contrast to the frequent references to Sarasvati.
Although one may not agree with all of Bhargava's interpretations, he does stress several important issues that have often been glossed over. Bhargava sees reference to major tectonic changes in the literature that follows Rigveda. This leads to extensive flooding, draining of many lakes, and change in the courses of rivers. It is interesting that Robert Raikes and George Dales also hypothesize that flooding may have contributed to the decline of Mohenjo-Daro (Raikes and Dales 1986). They suggest that the waters of the Indus river were impounded by a natural dam some miles downstream from Mohenjo-Daro leading to its abandonment. This scenario may have been a part of a larger process of tectonic changes that caused Sarasvati to dry up.
We have already mentioned some reasons in favour of the view that the Sarasvati may have dried up around 1900 B.C. This would then represent a terminus ad quem for the Rigvedic age.
There are references to Sarasvati reaching the sea at Prabhas in Gujarat in the Mahabharata (see Bhargava 1964) and the Puranas (for example Brahmapurana 77.2). This suggests that the original core of the epic and the Puranas must also date back into the Rigvedic era.
Further Parallels and Convergence
At this point it is not possible to clearly know why Indus writing disappeared for centuries. The answer to this puzzle is bound to be a complex one, and related to the decline of the Harappan cities. The decline of Mohenjo-Daro was possibly associated with rising water levels due to the forcing upwards of a large section of the Arabian Sea coastline by a tectonic shift or by the natural damming of the Indus river. Furthermore, sedimentation and tectonic movement led to the capture of the Sarasvati system by the Sutlej river of the Indus system and the Yamuna river of the Ganga system. This led to the drying up of the Sarasvati and the Drishadvati rivers, in whose valleys most of the Harappan settlements were situated. This disruption to the economy of the heartland also caused cities like Harappa to decline. Now a fundamental shift was taking place in the economic life of the Harappans. Theirs had been a dual economy based on agriculture and pastoralism. The emerging society was more intensely focused on agriculture. There was therefore an emigration away from the earlier towns to new communities in less arid regions. The new economy did not at first generate surplus wealth to lead to urban centres of the same size as in the earlier age. Eventually new powerful cities were established due east at Kaushambi, Atranjikhera, and Mathura.
The evidence from the Vedic literature also speaks of a gradual relocation from the area of Sapta Saindhava which is practically identical to the Harappan domain. During the Rigvedic period Sarasvati is the most important river. This fact itself forces one to date the hymns to the latest of the beginning of 2rd millennium B.C. The earliest Vedas describe a society that is partly urban and partly agricultural and pastoral like the Harappan society. This may be seen most easily from the many occupations listed in Yajurveda. The Rigveda describes fortified towns. Rigvedic ritual requires construction af altars out of bricks. On the other hand, some scholars interpret certain structures in the lowest layers of the Harappan ruins as fire altars. The Brahmanas, which are appendices to the Vedas, describe the phase of slow expansion to the east, a region that was originally densely forested. They, in turn, are followed by Aranyakas and Upanisads that capture the cultural transformation, also parallelled in the Harappan evidence, that values living in forests and small farming communities.
That the Vedic people were literate is indicated partially by a reference to the mark of eight that occurs in the Rigveda itself (10.62.7). Chandogya Upanisad has several references to syllable counts (eg. II.10) and to analysis of words in terms of vowels and consonants. Aitareya Aranyaka, of the period of forest dwelling, also describes the distinction between vowels and consonants (II.3.1) and has reference to writing (V.3.3). Several Upanisads describe different aspects of the alphabet.
Some scholars have argued that the Indus tradition could not be IndoAryan because no representation of any horse has been found in remains prior to 2000 B.Cin India. However, horses remain rare in Indian iconography even in the later historical period when they had undoubtedly reached India. Also there is evidence of a presence of horses in Iran that goes back to 3000 B.C. And since the Harappans had trade links with Sumer, Iran and other regions they surely knew the animal. Some scholars have argued that the horse must have been relatively rare in Rigvedic culture and this appears true of later periods of Indian history as well. Rigveda mentions horses for drawing chariots but there is no mention of riding in battle. Nevertheless, in view of its importance in the Rigvedic hymns as the symbol of speed and of the sun, the fact that no representations of the horse have been found in Harappan iconography remains a puzzle.
Conclusions
The archaeological evidence about the Sarasvati river provides us a chronological framework. Literary references in the early Vedic literature compel the conclusion that before this river dried up its settlements were of the Vedic people. But the settlements before its drying up are those of the Harappans. Furthermore, as pointed out by Mughal (1982), the Sarasvati valley area was the bread-basket of the Harappans. This indicates that the large urban centers of Harappa and Mohenjo-Daro may have served as gateway cities at the borders (Ratnagar 1982). All this leads to the common identity of the Harappans and the Indo-Aryans, a conclusion that the independent study of the Indus script has corroborated. Since the Harappan phase is merely one manifestation of the Indus tradition that shows great continuity and goes back to at least 6500 B.C., we conclude that this tradition was partly Indo-Aryan. These conclusions agree broadly with the traditional chronology of India and allow us to see why the findings of Hicks and Anderson (1990) should not come as a total surprise. While our reconstruction only emphasizes that the Indo-Aryans may have reached the Indus area sometime before c. 2600 B.C., we would like to suggest that the nature of the continuity of the Indus tradition indicates that they were present as one of the ethnic groups right from the earliest times.
We also repeat that the Harappan people, in themselves, comprised of several ethnic sub-groups. It is conceivable, therefore, that the Vedic settlements in the Sarasvati valleys were different in the details of their culture and their belief system from the urban Harappans in the cities. The Vedic literature does talk of struggles between various groups and such a conflict may have occurred between the urban and the rural Indo-Aryans themselves. Such a struggle may have contributed to the eventual transformation of the Harappans and their further dispersal.
DIAGRAM: FIGURE 1: Ethnic group relationships within the Indus tradition as Sketched by Shaffer and Lichtenstein (1989).
MAP: Figure 2: Geographical map of northwest INdia and adjacent regions for the 3rd millennium B.Cas in Kenoper (1991).
References
Bhargava, M.L.
1964 The Geography of Rgvedic India. Upper India Publishing, Lucknow.
D'iakonov, I.M.
1985 On the original home of the speakers of Indo-European. Jof Indo-European Studies. 13, 92-174.
Gamkrelidze, T.Vand Ivanov, V.V.
1985 The ancient Near East and the Indo-European question: Temporal and territorial characteristics of proto-Indo-European based on linguistic and historico-cultural data. J. of Indo-European Studies. 13, 3-91.
Gimbutas, M.
1973 The beginning of the bronze age in Europe and the Indo-Europeans 3500-2500 B.C.Jof Indo-European Studies. 1, 163-214.
1985 Primary and secondary homeland of the Indo-Europeans. Jof IndoEuropean Studies. 13 , 185-202.
Hicks, H.Hand Anderson, R.N.
1990 Analysis of an Indo-European Vedic Aryan head - 4th millennium B.C. Journal of Indo-European Studies. 18, 425-446.
Jarrige, J. Fand Meadow, R.M.
1980 The antecedents of the civilization in the Indus valley, Scientific American. 243, no.2, 122-133.
Joshi, J.Pand Parpola, A.
1987 Corpus of Indus Inscriptions. Memoirs of the Archaeological Survey of India, Delhi.
Kak, S.C.
1987a On the decipherment of the Indus script - a preliminary study of its connection with Brahmi. Indian Journal of History of Science. 22, 51-62.
1987b On the chronology of ancient India. Indian Journal of History of Science. 22, 222-234.
1988 A frequency analysis of the Indus script. Cryptologia. 12, 129- 143.
1989 Indus writing. Mankind Quarterly. 30, 113-118.
1990 Indus and Brahmi: further connections. Cryptologia. 14. 169-183.
1991 Astronomy in Satapatha Brahmana. Procof the Inter. Symposium on Indian and Other Asiatic Astronomies. Hyderabad, December.
Keith, A.B.
1909 The Aitreya Aranyaka. Oxford: Clarendon Press. Reprinted 1969.
Kenoyer, J.M.
1991 The Indus valley tradition of Pakistan and Western India. Journal of World Prehistory. 5. 331-385.
Macdonell, A.Aand Keith, A.B.
1912 Vedic Index of Names and Subjects. London. Indian Reprint Ed. , Delhi, 1982.
Mahadevan, I.
1977 The Indus Script. Memoirs of the Archaeological Survey of India, Delhi.
Mitchiner, J.E.
1978 Studies in the Indus Valley Inscriptions. New Delhi: Oxford & IBH.
Mitchiner, M.
1973 The Origins of Indian Coinage. London: Hawkins Publications. Mughal, M.R.
1982 Recent archaeological research in the Cholistan desert. In Possehl, G.L. (ed.), Harappan Civilization, Aris & Phillips, Warminster.
Pargiter, F.E.
1922 Ancient Indian Historical Tradition. London: Oxford University Press.
Possehl, G., (editor)
1982 Harappan Civilization, A Contemporary Perspective. Aris & Phillips, Warminster.
Possehl, G., and Raval, M.H.
1989 Harappan Civilization and Rojdi. E.J. Brill, Leiden.
Raikes, R.Land Dales, G.F.
1986 Reposte to Wasson's sedimentological basis of the Mohenjo-daro flood hypothesis. Man and Environment. 10. 33-44.
Ratnagar, S.
1982 The location of Harappa. In Possehl, G.L. (ed.), Harappan Civilization, Aris & Phillips, Warminster.
Renfrew, C.
1987 Archaeology and Language: The Puzzle of Indo-European Origins. Jonathan Cape, London.
Roy, S.B.
1983 Chronological framework of Indian protohistory - the lower limit. Jof the Baroda Oriental Institute. 32, 254-274.
Sengupta, P.C.
1947 Ancient Indian Chronology. University of Calcutta Press, Calcutta.
Shaffer, J.G.
1984 The Indo-Aryan invasions: cultural myth and archaeological reality. In J.R. Lukacs (ed.), The People of South Asia. Plenum Press, New York.
1991 The Indus valley, Baluchistan, and Helmand traditions: neolithic through bronze age. In R. Ehrich (ed.), Chronologies in Old World Archaeology (3rd Edition), University of Chicago Press, Chicago.
Shaffer, J.Gand Lichtenstein, D.A.
1989 Ethnicity and change in the Indus valley cultural tradition. In J.M. Kenoyer (ed.), Old Problems and New Perspectives in the Archaeology of South Asia, Wisconsin Arch. Reports, vol. 2, University of Wisconsin, Madison.
Sokal, R.R., Oden, N.L., and Wilson, C.
1991 Genetic evidence for the spread of agriculture in Europe by demic diffusion. Nature. 351, 143-145.
Tilak, B.G.
1893 The Orion. Poona. Reprinted 1989.
Wilson, J.V. Kinnier
1974 Indo-Sumerian. Oxford.
http://subhashkak.voiceofdharma.com/scriptures.htm