Skip Navigation Links





"The Coming of the Aryans to Iran and India and the

Cultural and Ethnic Identity of the Dāsas"


      This is the title of an article covering pp. 195-265 of Studia Orientalia, vol. 64, Helsinki, 1988. As soon as I heard of the thesis I wrote to its celebrated author, some of whose views expressed elsewhere I had already discussed. I requested an offprint. He was kind enough to post it at once. It was graciously inscribed "With best regards" and signed with his name. I thanked him for the personal touch as well as for the prompt dispatch, but while greatly appreciating his paper 1-hinted that with a different attitude to the same materials one might come to conclusions not quite the same as his. The article laid before me fascinating information of various kinds. Here is a remarkable piece of original research, a wide-sweeping scholarly synthesis of all the data appearing to bear at present on the theme. The most interesting, unexpected and significant of them are the latest archaeological findings in what Parpola terms 'Greater Iran' because the area concerned includes parts of Iran and most of the Iranian plateau.


The claim for Aryan entry in Baluchistan and Sind


      Let us first, without passing any judgment, look at the picture presented by Parpola. French work and extensive Soviet excavations have brought to light in Greater Iran "a long continuous belt of many sites sharing a fairly uniform


      * In finalizing this Supplement I have received very valuable help from my friend Richard Hartz, an expert Sanskritist and a keen as well as meticulous mind, from the Archives and Research Department of the Sri Aurobindo Ashram, Pondicherry.

culture at the end of the third millennium B.C."1 Parpola2 calls it "the bronze Age culture of Greater Iran" or simply "Namazga V culture" after an important site which is representative of all the others. The people of the Namazga V phase have been conclusively proved by French archaeologists under the lead of Jean-François Jarrige to have colonized around 2000-1900 B.C. the Bolan Pass leading from Baluchistan to the Kachi plain in the southern Indus valley,3 Among the diverse traits of the Namazga V phase "the number of weapons is conspicuous and there is evidence for horse and chariots, for transport of the entire cultural complex including intrusive necropoles, and for richly furnished aristocratic burials."4 As a result, "there is fair unanimity," says Parpola,5 "that 'Greater Iran' was in the Namazga V period controlled by a seminomadic military elite." Five golden and two silvery trumpets among the finds further confirm that the ruling class was engaged in chariot warfare.6 "The trumpet with its far-reaching sound was indispensable in directing horse-drawn chariots during battles. It was used also in training horses."7 As horse-drawn chariotry is the typical sign of the Aryans, the people who from Greater Iran colonized the Bolan Pass were Aryans.


      So we have the arrival of Aryans in Baluchistan around 2000-1900 B.C. This means that the Namazga V phase "flourished, in part, simultaneously with the Indus civilization, and there is evidence of some contact between the two even during the third millennium".8 Through the Bolan Pass these people moved further east. The site of Pirak in the Kachi plain in Pakistan "from c. 1800 B.C. testifies to the rapid diffusion of the horse and the two-humped Bactrian camel in northwest India during the first quarter of the second millennium B.C. These animals brought about a major change in the economy of the area. It is obvious that


      1. P. 203.                  5. Ibid.

      2. Ibid.                      6. P. 206.

      3. P. 202.                  7. Ibid.

      4. P. 204.                  8. P. 204.

Sind served as a channel through which immigrants representing the Namazga V and shortly thereafter also the Namazga VI culture continued to other parts of the Indian subcontinent."9


Who were the Dāsas of the Rigveda?


      In characterising these immigrants comes the startling originality of Parpola's picture. According to him, they are the people we know of through the Rigveda as the enemies of the Aryans: the Dāsa-Dasyus of whom the Panis are one sect. These enemies, in Parpola's exposition, are neither the pastoral Dravidian aborigines they were once thought to be with their small palisades termed 'forts' in the Rigveda, nor the urban inhabitants of the Indus Civilization whom Wheeler and some others took to be the targets of the Rigvedic Aryans who repeatedly speak of attacking mighty forts - actually, in this perspective, the fortified Harappan cities like Mohenjo-daro. Parpola10 explains:


      In Old Iranian, Proto-Aryan s has become h. In old Persian an ethnic name Doha- is attested, also as a proper noun in the administrative tablets found at Persepolis; the masculine plural is used as the name of a province of the Persian empire, placed before the similarly used name of the Sakas in a Persepolis inscription of Xerxes (h 26). In the Greek sources Herodotus (1,125) is the first to mention the people called Dáoi, as a nomadic tribe of the Persians. More accurate information on them, however, is delivered by Alexander's historians. According to Q. Curtius Rufus (8,3) and Ptolemy's Geography (6,10,2), the Dahas lived on the lower course of the river Margos (modern Murghab) or in the northern steppe area of Margiana. Pomponius Mela (3,42), based on Eratosthenes, tells that the great bend of the river Oxus towards the


      9. P. 206.

      10. Pp. 220-21.

northwest begins near the Dahas (iuxta Dahas), Tacitus (Ann. 11,10) places the Dahae on the northern border of Areia, mentioning the river Sindes (modern Tejend) as the border. These placements agree neatly with that of Namazga V culture of Margiana and Bactria [in greater Iran].


      The Dāsas of Rigvedic nomenclature are for Parpola the Dahas, and "Sanskrit dasyu- corresponds to Old Iranian dahyu- iand, (administrative) province, district (of a province)'",11 obviously turned by the Rigveda into a tribal designation.


      The Panis are to Parpola12 the Párnoi said by Strabo (11,9,2), again one of Alexander's historians, to have belonged to the Da(h)as. They are reported to have lived previously in Margiana, from where they founded the Arsacid empire of Parthia. Parpola13 elaborates:


      The Greek form of the name, Párnos< (from Iranian *Parna-), corresponds to Sanskrit Pani-, if it is assumed to be a "Prakritic" development of the reduced grade form *Prni-. The full grade seems to be found in the name Parṇáya- attested as an enemy of the king (Divodāsa) Atithigva in R[gveda] S[aṁhitā] 1,53,8 and 10,48,8. These names may go back to the same Aryan verbal root as the name of the Dāsa king Pipru, namely pr- (present piparti, prnāti) 'to bring over, rescue, protect, excel, be able'. The ar:r variation reflects a dialectal difference within Indo-Iranian.


 Parpola adds:


      Some other proper names of the Dāsa chiefs are also clearly of Aryan origin, for example Varcin- 'possessed of


      11. P. 222.

      12. P. 223.

      13. P. 224.

(vital) power' (cf. RS varcás = Avestan varǝčah 'vital power').


      Parpola concludes:


      The etymologies of the names used by the Rgvedic Aryans of their enemies thus speak for their above suggested identification with the carriers of the Bronze Age culture of Greater Iran, and for the proposal that these were speakers of an Aryan language.


      Hence, in Parpola's eyes, the ethnic identity of the Rigvedics' enemies is completely different from what it was formerly believed to be. Nor is this the only novelty he offers. But at the moment we shall not dwell on the subject. We shall just touch on his identification of the Rigvedics themselves.


The Rigvedics in the new perspective


      Parpola writes: "As far as the Vedic Aryans are concerned, Sind is definitely a peripheral area, though the Vedic texts do refer to Sindhu as producing excellent horses. This fully agrees with the archaeological evidence, which is important in confirming the arrival of horsemen from the northern steppes c. 1800 B.C. ... The horsemen of Pirak constitute the earliest evidence for the use of the horse in the Indian subcontinent."14


      Parpola's historical reconstruction implies "two separate early waves of Aryan speakers in Greater Iran and in India.... The Aryans of the earlier wave including the Dāsas could be called 'proto-South Aryans'. Since the Rgveda clearly states that the Dāsas did not offer Soma (<*Sauma), the main cultic drink of the Vedic and (as Haoma <*Sauma) the Zarathustrian ritual, the Aryans of the second wave


      14. Pp. 238-39.

which brought the Soma religion to Iran can be called 'Sauma Aryans'."15


      Here comes the role of the Andronovo culture which "came into being in the southern Urals" and "in the course of the second millennium spread over vast areas of the Eurasian steppes". How it connects with the Sauma Aryans we can gather from two subsequent passages:16


      An important hint to the origin of the Sauma Aryans and their route of advance is supplied by the fact that [in the words of Gershevitch] "there was an Iranian people, additional to the Avestan, whom the Persians knew to be devoted to Hauma. These were the Saka nomads whose name is given as Haumawarga in inscriptions of Darius and Xerxes. There is at present virtual agreement among scholars ... that the territories of the Haumawarga Sakas extended from Tashkent to the Alei valley, including Ferghana as centre-piece." This is well in agreement with the hypothesis that the Sauma Aryans were Andronovo nomads.


      The old hypothesis that the carriers of the Andronovo culture were ancestors of the later Iranians and Indo-Aryans is endorsed by many Soviet archaeologists. In recent years they have been arguing that the immigrants from the northern steppes were a partial cause of the collapse of the Bronze Age civilization of Greater Iran, and that they represented the arrival of the Aryans associated with the Rgveda and the A vest a. Deriving the Andronovo culture from the early Timber Grave culture [which evolved around 2000 B.C. in the south Russian steppes], they stress that these two cultures cover an area full of toponyms of Aryan etymology.


      We may quote a few other remarks of Parpola's to bring more precision to his chronology:17


      15. P. 230. 

      16. P. 232. 

      17. P. 236.

The Rgvedic language is connected with Old Iranian by some philological and morphological innovations, and the Rgveda also shares with the Avesta a number of identical phrases. Moreover, the Rgvedic Aryans called themselves "Aryas", as did the Avestan, Median and Old Persian speakers and at least a part of the "Iranian" speaking steppe nomads (the Ossetes of Transcaucasia). The Rgvedic Aryans, the pre-Zarathustrian Aryans and the Mitanni Aryans, therefore, should all belong to the same hypothetical first wave of Proto-Andronovo immigrants that are supposed to have submerged the late Namazga V culture; in their language the Iranian change s > h had not yet taken place.


      Very recent archaeological discoveries from Margiana now enable us to view the situation from a new perspective. A huge rectangular building complex 130 x 100 m. excavated at Togolok-21 has been identified, undoubtedly correctly, as a temple "used by proto-Zoroastrians whose religious beliefs and rites became (in changed form) part of official Zoroastrianism" [in the words of Sarianidi]. The most spectacular discovery at Togolok-21 is the earliest evidence of Haoma cult. The old problem concerning the original identity of the plant called in Avesta Haoma and in the Rgveda Soma was ably reviewed in 1987 by Harry Falk, who convincingly opted for the identification with Ephedra [from the Moscow State University's Prof. N. Meir-Melikyan's examination of microscopic twigs contained in a row of vessels placed inside special brick platforms].


      Parpola refers to the oath in the Mitanni document, dated to c. 1380 B.C., which mentions Vedic deities and he18 says that it "suggests that the decisive thrust of the Sauma Aryans took place in the 16th century at the latest". Such a date bringing "the Mitanni Aryans" to Mesopotamia leaves


      18. P. 232.

room for their companion tribes to reach from the direction of the Russian steppes first the Namazga V culture in Greater Iran and then the Swat valley in Afghanistan and finally India in approximate consonance with the upper limit of the time-bracket generally favoured: "Most authorities ... place the Rgveda between 1500 and 1000 B.C."19


      Further precision to Parpola's vision may be brought from some other words of his:


      The temple of Togolok-21 provides a most precious temporal and cultural indicator for the coming of the Sauma Aryans by testifying that their fusion with the Dāsas took place between the late Namazga V and the late Namazga VI periods. This means that their arrival more or less coincided with the beginning of the Namazga VI period around 1800 B.C. This agrees very well with the fact that the relations of Margiana and Bactria with Syria developed in the 18th century B.C., while the Proto-Indo-Aryan' dynasty of Mitanni dates at least from the 16th century B.C. The Rgvedic hymns in their turn suggest that part of the Sauma Aryans did not stop in Margiana and Bactria, but continued immediately to northwest India. Such a short stay would well account for why the cultural assemblage of the Ghalegay IV period in Swat (c. 18th to 15th centuries B.C.) resembles that of Dashly in Afghanistan, but is not identical with it.


      The valley of Swat occupies a strategic position in the archaeological identification of the early Rgvedic Aryans, because they must have passed through this area. This is clearly implied by the occurrence of the name of the Kabul river and its tributaries in the Rgveda....20


      The Ghalegay IV-V periods in Swat are chalcolothic, except for a little iron towards the end. This tallies with the textual evidence, for references to iron are hard to find in the Rgveda, while the black metal was known to the


      19. P. 198.

      20. Pp. 240-41.

Atharvaveda (11,3,7). Inhumation and cremation occur side by side, as in the Rgveda. The Vedic texts of the later period speak of an earthen vessel, into which the bones of the dead were collected after the cremation. A link from the Ghalegay V culture to the Painted Grey Ware (PGW) is supplied by the urns with perforations near the neck (resembling the eyes and the mouth of the Ghalegay V 'face-urns') in the PGW layers of Ahicchatra and of Ghalegay V type terracotta human figurines in the PGW layer of Jakheran, U.P.


      Thus the archaeological evidence allows the hypothesis that Rgvedic Aryans started moving from Swat to the plains of Punjab during the latter half of the Ghalegay IV period, c. 1600-1400 B.C., and continued during the following Ghalegay V period. After this, the northwest developed in relative isolation, losing its contacts with the late Vedic culture of the plains, associated with the early PGW.21


A caution against Parpola's time-gauge for the Rigvedics


      Now we have - minus several interesting but subordinate details, linguistic and archaeological, with which Parpola enriches his case - his broad picture of the Aryans in general and the Rigvedics in particular invading India. To make an assessment of it we shall have to bring in some further points that he makes. But before doing so let me essay one short cautionary remark.


      Parpola has spoken of 'the Mitanni Aryans' along with those whom he enumerates as having called themselves 'Aryas'. But the fact is that neither in the document which is a treaty, nor in another document known as Kikkuli's manual of training chariot-horses, both of which have been found to have an affinity to the Rigvedic language and


      21. P. 248.

culture, is there, as B.B. Lal22 noted long ago, "any reference to the name of the concerned people". The Mitan-nians do not call themselves 'Aryas'.


      A second fact is a strange imbalance in the very formula of the Mitannians which introduces in Rigvedic language the Rigvedic religion. The formula runs: "Mitra-Varuṇa, Indra, Nāsatyā." Compare it with the Rigvedic phrase to which Parpola,23 without quoting it, directs us, saying that these deities "are all mentioned together in Rgveda 10,125,1". The phrase is "Mitra-Varuṇa, Indra-Agni, Aśvinā", in which "Aśvinā" is equivalent to "Nasatya". Parpola fails to note that the dvandva, the dual form, in which Agni accompanies Indra just as Varuṇa accompanies Mitra, is missing in the Mitanni formula. Evidently, Agni is not a Mitanni god whereas in the Rigveda he is the one most frequently hymned after Indra. Nor is that particular dvan-dva a freakish occurrence: it is a very prominent expression in the Rigveda. Whole hymns are devoted to Indra-Agni: e.g., 1,21 and 108; 5,86; 6,59 and 60; 7,93 and 94; 8,38 -besides the expression coming in hymns otherwise dedicated, as it does in 10,125. In view of the dissimilarity between the Mitanni formula and the Rigvedic, as well as because of the absence of the name 'Arya' for the people in both the treaty and Kikkuli's book, it may not at all be safe to take c. 1380 B.C. as a time-gauge for the Rigveda's epoch.


      There is also the fact mentioned by Parpola:24 "The Indo-Aryan deities ... are invoked after 104 other deities at the end of a Mitannian treaty." Obviously, the Mitannian ruler gave prime importance to those numerous non-Aryan gods and the Agni-lacking Aryan ones formed just the fag-end of


      22. "The IndoAryan hypothesis vis-à-vis Indian archaeology", cyclo-styled copy of the paper read at the Seminar on "Ethnic problems of the early history of the peoples of Central Asia and India in the second millennium B.C.", held at Dushanbe (USSR) from 17 to 22 October, 1977, p. 7.

      23. P. 198.

      24. Ibid.

his religious commitments. This is indeed a far cry from the mentality of the Rigveda. It is as if two different widely removed epochs were involved.


      Another fact of high significance is a linguistic one. Dr. Satya Swarup Mishra, Head, Department of Linguistics, Faculty of Arts, at Banaras Hindu University, writes to me about the words in the Mitanni documents: "These words were first of all taken as Indo-Iranian. Then the western scholars themselves decided that they were Old Indo-Aryan and not Indo-Iranian. But I have shown that these words show the linguistic change of a very early Middle Indo-Aryan type. The assimilation of pt [of Sanskrit sapta] to tt in satta-vartana [in Kikkuli], the change of v to b in several words are some of the important Middle-Indic features in these loan words." Surely such features set a big gap between the epoch of the Rigveda and that of the Mitanni documents. Middle-Indic is substantially distant in time from the Rigveda's Sanskrit.


An inconsistency about the horse


      Before we arrive at some idea of the Rigvedic epoch I should like to dwell a little more on Parpola's account of the Namazga V people in India. Their "arrival... seems to have disrupted the political and cultural unity of the Indus valley soon after 2000 B.C. The urban system of the Harappans and the processes of city life, such as centralized government with the collection of taxes and organization of trade, ceased to function. The thousands of countryside villages, however, persisted. In peripheral regions, especially in Gujerat, mature Harappan traits, mixed with new elements, lingered longer, until 1750 B.C. The newcomers did not stop in the Harappan area, however, but pushed on further into the Deccan and towards the Gangetic valley."25


      Then Parpola refers to - among some other cultures - the Rajasthani chalcolithic culture of the Banas valley, c. 1800


      25. P. 206.

B.C. and the Malwa culture of Navdatoli I-II in the Deccan, dated to c. 1700-1400 B.C. These cultures "have produced bowls ('wine-cups'), channel-spouted cups and other ceramics as well as copper objects resembling those of the Bronze Age culture of Greater Iran. The Malwa culture evolved into the 'Jorwe culture' (c. 1400-1100 B.C.). From a Jorwe stratum at Daimabad in Maharashtra comes a cylinder seal with a horse motif. I am now inclined to think that in Rajasthan, Gujerat and the Deccan the originally Aryan-speaking nomads of Namazga V-VI derivation fairly soon adopted the local language, namely, the Proto-Dravidian, derived from the Harappan language spoken in this southern extension of the Indus civilization."26


      Here is Parpola's second allusion in his new article to his own theory that the Harappans spoke a proto-Dravidian language. His first allusion comes near the start of the article:27


      A major reason against assuming that the Harappans spoke an Indo-European language is that the horse is not represented among the many realistically depicted animals of the Harappan seals and figurines. Comprehensive bone analyses by one of the best experts, Richard Meadow, have yielded the conclusion that there is no clear osteological evidence of the horse (Equus caballus) in the Indian subcontinent prior to c. 2000 B.C. Obviously the Aryans are not likely to have been present in India in large numbers before about 2000 B.C., if the horse played a central role in their life.


      In an earlier Supplement I have dealt in great detail with the bearing of Harappan evidence on the horse-question and shown that Parpola's negative conclusion omits to take into account the complexity of the case. Now I shall draw attention to some other facts.


      26. Pp. 206-07.

      27. P. 196.

Lal,28 after examining the full material in his search for the earliest culture in India which could qualify for Aryanism, rules out Aryanism at both the Malwa and the Banas sites. He ends his comment on the Malwa culture thus: "Lastly, the Aryan animal par excellence, viz. the horse, is conspicuous by its absence from all the Malwa sites excavated so far." On the Banas culture he has a similar remark: "the most significant animal associated with the Aryans, viz. the horse, is conspicuously absent from all the sites of the Banas Culture, either by way of its skeletal remains or even terracotta representations." Thus Parpola is inconsistent in Aryanizing these two cultures while Dravidianizing the Indus Valley Civilization for its lack of direct archaeological signs of Equus caballus during its most characteristic phase -namely, before c. 2000 B.C. The cylinder seal at Daimabad of the 'Jorwe Culture' (c. 1400-1100 B.C.) which evolved from the Malwa culture makes no odds to the observed absence of the horse from the latter and to the inconsistency Parpola has committed.


Horse-evidence from both outside and inside the Indus Valley


      Richard Meadow seems to have overshot the mark in the matter of equine evidence. Lai, though unwilling to believe that the Harappa Culture knew the horse, was not so dogmatic. He29 refers to an area outside the Punjab as being "known for having had its own indigenous variety of the horse." Dr. K.R. Alur, a veterinary surgeon, has some pertinent information detailing a faunal report on the excavation at Hallur, a border village in Mirekerur taluka of Dharwad district in Karnataka. His paper of 16.6.1990, Aryans and Indian History: an archaeo-zoological approach, says: 


This site was excavated by Dr. M.S. Nagaraja Rao during


      28. Op. cit., pp. 18 & 22.

      29. Op. cit., p. 29.

February-March 1965. Excavation of two trenches showed that the occupation of the site was during the neolithic period circa 1800 B.C. The excavator has distinguished two cultural phases.


      Period I, which is designated as neolithic, has been subdivided into two phases.... The earlier phase is neolithic characterised by the hand-made pottery and a few ground stone-tools.


      Phase 2 has been called neolithic-chalcolithic. It is distinguished by the occurrence of hand-made pottery, a large number of stone-tools and a new stone-blade industry with tools of copper.


      Period II ... is called the early iron-age although some of the earlier elements continue. The new elements ... are the typical highly burnished black and red ware pottery with white painted variety, and iron implements.


      Carbon-14 determination for the latter period showed that the iron age could be ascribed to circa 1000 B.C. and, according to the excavator, the earlier phase of the neolothic chronologically falls to circa 1800 B.C. and the second to about 1500 B.C.


      After this introduction Dr. Alur reports on the faunal collection, evidently covering the dates just mentioned:


      From this collection I identified the following bones of Horse:


      S no. 212. Small metacarpal (splint bone).

      S no. 467. Proximal extremity of small metacarpal.

      S no. 497. Molar (from the middle series).

      S no. 517. Second phalanx.


      When I wrote this report, I least expected that it might spark off a controversy and land me in the witness box before the Indian historians' jury.... I was apprised of the gravity of the situation when I began to get letters asking me for clarification of the situation against the prevalent belief that the horse is a non-indigenous species and was

introduced into India only by [invading] Aryans....


      To make my position clear, I wrote in my article "Archaeological remains of Animals" that "whatever may be the opinion expressed by archaeologists, it cannot either deny or alter the find of a scientific fact that the horse was present at Hallur before the [presumed] period of Aryan invasion...."


      The find of this fact put the Indian archaeologists and historians in a predicament in which they could not deny a scientific fact, yet could not accept it. So those on whom the responsibility lay made a reasonable approach and ordered a second excavation near the original site to avoid a probable introduction of an artifact. I examined the faunal collection of this excavation also and found the presence of some more bones of the Horse.


      After some reflections on how "foreign scholars, who came to India with the advent of British rule, built up the theory of Aryan invasion on the findings of excavations conducted on the so-called migratory route, where remnants of horse and chariots were traced", Dr. Alur touches on how the Indian tradition, which knew nothing of an invasion and took the horse's presence in India to be natural from the beginning, got flouted further by "the report written by S. Sewell and B. PRasād on the faunal study from Mohenjo-daro and Harappa". This report declared "that there is no evidence of the presence of the horse in the Indus valley" though "they declared that they had recovered a few metacarpals of the domestic Ass".


      Then Dr. Alur brings to light a little-known riposte to that report: "Dr. J.C. George of the M.S. University of Baroda stated that the study of the above table of comparative measurements shows beyond doubt that the metacarpals recorded by Prasad are definitely not of the domestic Ass and it is therefore possible to conclude that the smaller size horse did exist in Harappa. He further states: 'It is rather incredible that in a great civilisation like India, the horse

alone should be conspicuous by its absence, while allied species like that of the Ass have been identified. It is equally unbelievable that the domestication of the prehistoric horse has been established in all the neighbouring countries such as Turkestan (Durest 1908) and Palestine (Garrod and Bate 1937) but not so in India.' "


      A little later, Dr. Alur refers appreciatively to the opinion expressed by R.S. Panchmukhi, chairman and editor of the Diamond Jubilee Volume of the Karnataka Historical Society, to which Dr. Alur contributed an article on "Horse in the Prehistoric period in India and its Aryan Affinities". Panchmukhi, after taking Dr. Alur to have proved the horse indigenous in India, suggests that whatever remnants of horse and chariot are claimed to be pointers to an Aryan immigration into India may really be signs of an Aryan emigration from India. "India," says Panchmukhi, "has a history of migration to all its neighbouring countries, both for trade and spread of religion." Towards the end of his paper, Dr. Alur agrees that the Aryans were the original inhabitants of India, some of whom migrated out of their country "to popularise their faith".


      Dr. Alur has certainly provided evidence that the 'Aryans' whom Parpola brings into India in c. 1600-1400 B.C. from the Swat valley could not have introduced the horse into Hallur between c. 1800 and 1500 B.C. Even as a location, Hallur would be too far. Can we conceive as a likely candidate the first wave posited by Parpola in c. 1800 B.C. into Sind through the Bolan Pass in Central Baluchistan? Sind, again, is too distant from Hallur. The closer cultures - those of Rajasthani Banas and of Deccan Malwa (c. 1800 and c 1700-1400 B.C.) as well as others adjacent to them - which Parpola is inclined to trace to the advance and spread of this wave - are themselves not close enough to Karnataka. Besides, as we have shown on the authority of Lal, they had no equus. The Jorwe culture (c. 1400-1100 B.C.) which has a stratum at Daimabad in Maharashtra evincing a cylinder seal with a horse motif is not only

sufficiently removed from Hallur in space but also too late in time to account for Hallur's horse-bones dating between about 1800 and 1500 B.C.


      From every point of view Parpola-cwm-Meadow stand faulted by Dr. Alur's information.


      Still more devastating is the report published in 1980 by the Department of Ancient History, Culture and Archaeology, University of Allahabad: History to Prehistory: Archaeology of the Vindhyas and the Ganga Valley by G.R. Sharma. Co-workers with Sharma were not only Indian archaeologists but also Dr. M.A.J. Williams and Keith Royce, who were members of the team led by Professor J. Desmond Clark of Allahabad University. The following passages from Sharma are well worth study:


      The explorations in the valley of the Belan and Son have resulted in discoveries of thousands of animal fossils. From the Belan section these fossils have been obtained from four Gravels as well as from the red silt overlying Gravel II. Most of the fossils, however, have been obtained from Gravels I & II. The species include bos-nomadicus, bos-bubalis, gavialis, sus, elephas, antelope, bos-elephas, stag, deer, equus, chelonia (tortoise) and unio....30


      The excavations of neolithic sites of Koldihwa and Mahagara have brought to light evidence of domestication of animals and cultivation of plants. The domesticated animals include cattle, sheep, goat and horse....31


      Mahagara and Koldihwa have yielded evidence of both wild and domesticated cattle, thus presenting an interesting picture of transition from wild variety to domesticated ones. The change in size and bone structure attest to nature's law of selection. Evidence of wild sheep/goat and equus has also been found from Cemented Gravels III and IV in the Belan valley. They are still wild at Mahadaha


      30. P. 98.

      31. P. 110.

and Sarai-Nahar-Rai, the Mesolithic sites of the Ganga valley. The Neolithic Mahagara offers evidence of their domestication, suggesting a natural selection and domestication of these animals almost parallel to that of cattle. Swine is present in wild condition both at the Mesolithic lake settlements in the Ganga valley and in the Neolithic Mahagara in the Belan valley.


      With the help of a number of radiocarbon dates obtained from the Belan and the Ganga valley, Stone Age Cultures from Upper Palaeolithic to Mesolithic have been dated. The Cemented Gravel III which has yielded the Upper Palaeolithic tools has also yielded the C-14 dates -23840 B.C. and 17765 B.C. As the earliest date is not from the lowest horizon, the Upper Palaeolithic in this area had possibly still an earlier antiquity.


      For the pre-pottery Geometric Mesolithic we have two dates, one from the Belan valley and the other from the Ganga valley. The date obtained from Shari-Nahar-Rai is 8395±110 B.C., while that of Mahagara reads 8080±115 B.C. We have two dates from the Neolithic levels of Koldihwa reading 5440±240 B.C. and 4530±185 B.C.


      Within the chronological framework provided by C-14 dates for terminal Upper Palaeolithic reading 17765 ±340 and for the pre-Neolithic 8080±115 and the early Neolithic levels reading 6570±210 and 5540±240 B.C., the totality of evidence furnished by these excavations and explorations ... presents a continuous story of human achievements....32


      In the face of Sharma's report, how shall we judge Parpola's contention, on the basis of Meadow, that the horse was introduced in 2000 B.C. by his 'Aryans' from outside India and therefore could not have existed in the Indus Valley Civilization?


      Surely the horse of this report can never be connected


      32. Pp. 111-12.

with those argued entrants from abroad? It is far too ancient for them. Even apart from its much earlier date and its location outside the Indus Valley we can say: "Whatever may have been brought from Parpola's Greater Iran was a domesticated and not a wild animal. How shall we account for Sharma's wild equus no less than his domesticated one? Prior to the stage of domestication, there was the wild stage which particularly stamps the creature as having been native to the Indian soil. Meadow's findings are very limited and cannot suffice to rule out the theoretical possibility of equine presence in the Indus Valley Civilization."


      Furthermore, if the domesticated horse specially distinguishes the Aryan, we have the Aryan in India long before Parpola's intruders from outside India and far earlier than even the Indus Valley Civilization. But such antiquity of the Aryans in an area sufficiently close to the Indus Valley would render not at all fantastic the notion of Aryanism at least colouring substantially enough the civilization flourishing in that locality in c. 2500-1500 B.C.*


Parpola's insufficient appraisal of Harappan Indo-Aryanism


      Actually a valid ground for this notion - quite independently of Sharma - we may underline from an observation by Parpola himself. He33 writes about two sites of the Mature Harappan:


      The fire-altars of Kalibangan and Lothal are so far without parallels at Mohenjo-daro and Harappa. Indeed, it has been asked [by Raymond and Bridget Allchin]: "Fire-worship being considered a distinctly Indo-Aryan trait, do these [ritual hearths of Kalibangan] carry with them an indication of an Indo-Aryan presence even from so early a date?" This hypothesis now seems quite plausible to me, if "Indo-Aryan" here is understood to refer to carriers of


      * For further horse-evidence see Appendix 2, pp. 419-420.


        33. P. 238.

the Bronze Age culture of Greater Iran, who had become quickly absorbed into the Indus Civilization, culturally and linguistically. It is supported further by the cylinder shape of the famous Kalibangan seal showing a Durga-like goddess of war, who is associated with the tiger. The goddess on the Kalibangan cylinder seal is said to be similar in style, especially the headdress, to one depicted on a cylinder seal from Shahdad [in Kerman on the desert of Lut in Iran, a major centre of the Bronze Age cultural tradition]. Seated lions attend to a goddess of fertility on a metal flag found at Shahdad.


      While the Indo-Aryan presence in the Indus Civilization cannot be doubted, Parpola appears to play down its basic significance, as if in its cultural and linguistic milieu it hardly counted for much. Putting aside the assumption that the Harappan language was Proto-Dravidian, is there any reason to talk of this presence as having been "quickly absorbed" into that milieu? The milieu itself might have been sufficiently in tune with Indo-Aryan speech. As for cultural absorption, can we say that the presence of so fundamental, so typical a trait of Indo-Aryanism should not be regarded as a natural expression of the Harappa Culture?


      H.D. Sankalia34 has some words which might mitigate the positive assertion by Parpola that the fire-altars of Kalibangan and Lothal are so far without parallel at Mohenjo-daro and Harappa. After writing that "such a 'fire-altar' has also been noticed by Casal at Amri", he adds: "Perhaps such fire-altars also existed at Harappa and Mohenjo-daro, but were missed in mass diggings, and have only been revealed in a slow, careful excavation."


      Sankalia's words strike us as quite pertinent when we realise the importance of Kalibangan. He35 has observed that


      34. Prehistory and Protohistory of India and Pakistan (Deccan College, Poona, 1947), p. 350.

      35. Ibid. p. 361.

this site was "perhaps a third capital in Rajasthan". Furthermore, not only Kalibangan but also Rakhigarhi, a site 190 kms east of Kalibangan, has revealed fire-altars. And about it O.P. Bharadwaj36, on the authority of Suraj Bhan's Excavation at Mitathal and Other Explorations in the Sutlej Yamuna Divide,37 writes: "Rakhigarhi ... is supposed to be the most extensive of the known Harappan sites in India and deemed worthy of being considered as a possible easternmost capital of the Harappans."


      Along with the apparent parity of these sites with Harappa and Mohenjo-daro, there is the question of their date. In the query Parpola quotes from the Allchins, the general phrase - "at so early a date" - occurs in relation to Kalibangan. This would suggest a substantial antiquity on a par with that of those two sites. Bharadwaj38 supplies a chronological table. Lai gives the span of Harappan Kalibangan as 2200-1700 B.C., while Thapar's figure is 2300-1750 B.C. George F. Dales39 corrects the former to 2700-1900 and the latter to 2850-1950 B.C. E.K. Ralph, H.N. Michael and M. Han40 have the corrections: 2630/2670-2060 and 2850/2870-2110 B.C. So the central Indo-Aryan nature of the feature concerned goes to the very root of the Indus Civilization. And surely such a radical element should carry us to a deeper sense than of merely 'some contact' - as Parpola posits - between the Indus Civilization and the Bronze Age culture of Greater Iran. Here is not just 'contact', but a degree of intrinsicality of Indo-Aryanism. And the sense of intrinsicality is greatly deepened when we


      36."Identification of Vinasana and some Consequential Observations", Svasti Sri (Felicitation Volume in honour of Dr. B. S. H. Chhabra, Delhi 1984), p. 217.

      37. Kurukshetra, 1975, p. 95.

      38. Op. cit., loc. cit.

      39. In South Asian Archaeology, ed. Hammond Norman (London, 1973), p. 162 ff.

      40. In Ancient Cities of the Indus, ed. Gregory Possehl (Delhi, 1979), pp. 339-42, Table 4.

ponder a fact pointed out by Shashi Asthana41 in connection with Kalibangan:


      ... the so-called 'citadels' at Indus cities were taken to be the seats of government but B.B. Lai (1981) has now conclusively proved that at least at Kalibangan it was not at all so; it was possibly the place where collective religious ceremonies were held around the 'fire altars'. In other words, underlying the mature Indus Civilization or Harappan Culture was a great deal of social change, all of which is not easy to comprehend but without which the cities would not have emerged on the Indus plains. Social changes and cultural changes keep on interplaying variously at various levels (Gupta 1974).


      Lal's proof suggests something like centrality for the 'fire altars' and imparts to the socio-cultural history of the ancient Indus cities a basic colour of Indo-Aryanism. And when such is the case, can we ignore the implication that the Indus Civilization was not unaware of what played, in Parpola's phrase, "a central role" in the life of the Aryans: the horse? Indeed ill-founded is his belief that "the Aryans are not likely to have been present in India in large numbers before about 2000 B.C."


      How, then, face to face with his long discourse, stands my thesis that there is no proof archaeological or documentary of a Rigvedic invasion of India in the period usually allotted - c. 1500-1000 B.C., the upper limit of which is part of Parpola's own time-bracket c. 1600-1400 B.C.? The question basically is not whether an incursion into India, which .may be called Aryan, took place or notat that time as well as somewhat earlier, as Parpola's


      41. Pre-Harappan Cultures of India and the Borderlands (Books & Books, New Delhi, 1985), p. 240. The references to Lai and Gupta are based on the following two papers respectively: (1) "Some Reflections on Structural Remains at Kalibangan", Indus Civilisation - New Perspectives (ed. A.H. Dani, Islamabad), pp. 47-54. (2) "Two Urbanizations in India: A Side Study in their Social Structure", Puratattva, no. 7), pp. 53-60.

discourse about two waves would have it. The basic question is whether any entry in that period can be dubbed Rigvedic and whether, if the Rigvedics are anterior to the Harappa Culture, as I have tried to show, they can be termed outsiders rather than autochthones for all practical purposes? But before we examine it in some detail we may do well to ponder first Parpola's claim for the two waves flowing into India from the Bolan area and the Swat valley respectively.


Was there at all an entry into India?


      Parpola42 prefaces his own thesis by citing C.C. Lamberg Karlovsky43 who has recently pointed out the distinction between two types of archaeological evidence suggestive of culture contact on the one hand and on the other of expansion with or without preliminary culture contact. "If only a few types and numbers of artifacts characteristic of one culture are found within another distinctive culture, the contact was very limited. But if an entire cultural complex characteristic of a well defined archaeological culture is recovered from the area of another culture, it suggests foreign colonization, which usually leads to major cultural transformation in the colonized area." As conclusive proof Parpola44 cites the discoveries of Jarrige at the Bolan Pass in Baluchistan: "Excavations carried out since 1978 at Mehrgarh VIII and at the nearby Sibri Damb brought to light cemeteries with tombs and cenotaphs, whose burial mode and grave goods were totally different from the earlier local traditions."


      Now, does any conclusive proof of colonization hold for Pirak in Sind, where Parpola traces the first wave of the


      42. P. 202.

      43. "Third millennium structure and process: From the Euphrates to the Indus and the Oxus to the Indian Ocean", Oriens Antiquus (1986), 25:3-4, 189-219.

      44. P. 202.

Aryans whom he identifies with the Dāsas? He45 writes:


      The excavated site of Pirak in the Kachi plain of Sind comprises three occupation periods: I (c. 1800-1300), II (c. 1300-1100) and III (c. 1100-900 B.C.). From periods I and II come distinctive terracotta figurines of two-humped camels and of horse-riders. The camel figurines are quite new in the Indus valley, but have very close parallels at Namazga VI sites in Margiana, where they go back to the Namazga V traditions. The horsemen of Pirak ... have bowed legs to fit them on the back of the horse, armless torsos and heads with faces ending in a bird-like beak.... The significance of the curious beaked heads ... calls for comparison with the numerous representations of an eagle-headed anthropomorphic deity (with or without wings) in the seals and other objects of the Bronze Age culture of Greater Iran.... Also, the violin-shaped female figurines of Pirak I-II continue the Namazga V related religious traditions of the nearby site of Sibri.


Affinities with the culture of Mehrgarh VIII and Sibri are found too at various other sites - neighbouring Nausharo in Baluchistan itself and Chanhujo-daro and Amri in Sind. "At all these sites," says Parpola,46 "the traditions of the Indus Civilization continue without a break, but are transformed by intrusive traits. The new elements could now be recognised to be those associated with the cemeteries of Sibri and Mehrgarh VIII, whose entire cultural complex in its turn is practically identical to that of sites like Tepe Hissar III in northeastern Iran, Namazga V in southern Turkmenistan and Sapalli Tepe and Dashly in Afghanistan. Moreover, a related aristocratic burial was accidentally discovered at Quetta (Baluchistan) in 1985."


      Are we entitled to argue that at Pirak we have anything more than "culture contacts"? The new elements are 'asso-


      45. P. 239.

      46. Pp. 202-03.

ciated' with the cemeteries that have proved colonization at the Bolan Pass. But there are no such cemeteries at Pirak. We may remember also another phrase from Parpola47 quoted earlier: "the entire cultural complex including intrusive necropoles". Without 'intrusive necropoles' - that is, cemeteries - can we posit a substantial intrusive population colonizing Pirak? Again, the figurines of horsemen and the two-humped camel as well as of violin-shaped females show intrusive elements; but if we can indicate old contacts with the cultures of Greater Iran, they may be seen as a further phase of an already existing relationship which involves no mass immigration of a novel cultural life. Here we may draw on Parpola48 himself: "The simple terracotta seals of Pirak mostly continue earlier local traditions, but some have close parallels at Shahr-i-Sokhta (18th century B.C.) in Seistan and at Namazga VI sites in Margiana and Bactria. The pottery of Pirak is supposed to go back to the local third millennium traditions of Baluchistan and Afghanistan; close parallels are so far known only from Ispelanji and Dabar Kot in southern Baluchistan, but affinities are seen also in Mundigak IV-V in Afghanistan and now in Sarazm in Sogdiana." What is even more suggestive of a natural new introduction from regions already in contact, rather than a result of invasion, are a couple of remarks by Parpola49 which indicate a two-way movement.


      Terracotta "fire-dogs" are a novelty of the Pirak culture.... Very similar "fire-dogs" have been excavated around fireplaces at very early Iron Age sites in Fergana, such as Shurabashat.... Traffic with Central Asia was ... not in one direction only. Besides the "fire-dogs", convex copper buttons with a loop and sickle blades with deep serrations found on sites of the Yaz complex in Margiana, resemble [as Jarrige and Santoni point out] similar objects


      47. P. 204.

      48. P. 239.

      49. P. 240.

from Pirak "in a way that could not be fortuitous,... the examples from Pirak appear in earlier levels than those from the sites in the Murghab delta or in Fergana that are dated to the beginning of the Iron Age [c. 1300-1250 B.C.]. Moreover this period coincides with the appearance on these Central Asian sites of a hand-made ware with painted geometric patterns whose style recalls that of some vessels at Pirak, at a time when this type of pottery at Pirak is gradually being replaced by a wheel-made grey ware without decoration."


      The picture we derive from Parpola is of a traffic to and fro of cultural modes - continued from a fairly long past and across sufficiently wide areas - against a common religious background of various shades. It is a picture of contacts and exchanges. Unless certain specific signs are there, none of them necessarily bespeak large-scale movements of populations. Although a colonization is indicated at the Bolan Pass, nothing beyond a diversity of contacts between Baluchistan and Sind seems proved by the appearance of certain figurines at Pirak. While violin-shaped female figurines are said to hark back to Sibri, neither of the two others are said to have parallels at either Sibri or Mehrgarh VIII. The two-humped camels have correspondences in Margiana and not near the Bolan Pass, nor are the beak-headed horse-riders traced by Parpola explicitly to Mehrgarh VIII or Sibri; they are only said in general to resemble depictions on seals and other objects of greater Iran's Bronze Age culture. We may also note that their significance lies not so much in the horse-presence as in the presence of a certain type of deity which, as Parpola50 tells us "in many seals ... fights against snakes" and "is obviously related to the eagle which occupies so prominent a position in the other related seals, and which also fights against serpents". The focus is on this god; the horse is incidental. If the idea had been to set horse-


      50. P. 239.

riding as such in relief, a human figure would have been carved. So it seems too much to affirm, as does Parpola:51 "The horsemen of Pirak constitute the earliest evidence for the use of the horse in the Indian subcontinent." It may be noted that no equine bones have been unearthed at Pirak. In Sind, in the period of Pirak I-II - c. 1800-1100 B.C. - the earliest bones of equus caballus "occur at a high level at Mohenjo-daro".52 Being late in time they are suspected to be intrusive to the Indus Civilization. But they are all that Sind has to show in the period concerned.


      What distinguishes Pirak in its earliest phase is nothing that we can link with Mehrgarh VIII or Sibri: it is the local 'fire-dogs'. They precede in time comparable articles excavated in Greater Iran, as Parpola recounts. He53 has also remarked about them: "They have been found forming a support for cooking around an ash-filled cavity in the middle of a square fireplace. This type of fireplace and the habit of cooking in vessels placed directly over the fire seem to represent an innovation in the Kachi plain." What is of further interest is that this indigenous Indian novelty brings to mind, as Parpola himself observes, "the Vedic ritual". He enumerates the "three principal fireplaces" associated with this ritual: "the square āhavanīya, into which the offerings to the gods are poured, the round gārhapatya, the inherited hearth of the family head, and the halfmoon-shaped daksin-āgni 'southern fire' which is connected above all with the forefathers of the sacrificer". Then Parpola refers to an authority on the subject: "According to Hertha Krick, the first two form a pair and represent the Rgvedic tradition...." Thus, without any relation to the culture of Mehrgarh VIII and Sibri Damb, Pirak's most individual characteristic dissociates the Rigvedic tradition from non-Indian cultural traits and from origination in a foreign milieu.


      51. P. 239.

      52. Sir Mortimer Wheeler, The Indus Civilization, Third Edition (Cambridge University Press, 1968), p. 82.

      53. P. 240.

    As Parpola has Jarrige's name in his footnotes, it would seem that the latter, who established immigration in the Bolan valley, is inclined to favour the idea of immigration by the same people in the Pirak area as well. But a more judicious survey might have deterred the attractive conclusion. Here Jarrige could take a leaf out of his own book. For, in his Foreword to Shashi Asthana's Pre-Harappan Culture of India and the Borderlands54 he has a wise word which might serve as a guide here also. He55 writes, apropos of the difficult question of the origin of the "Quetta" culture:


      The author assumes that the spread of the "Quetta" ware, directly related to the Geoksjur style of south Turkmenia, is to be linked to a migration of people from the Tedzen Delta. This idea, supported by several specialists with some sound arguments, is consistent with the explanation relating to such migration, the foundation of Shahr-i-Sokhta and the expansion of Mundigak. In fact, sherds in a truly "Geoksjurian" style are very few at Shahr-i-Sokhta or at Mundigak, in spite of their conspicuousness. At Damb Sadat, the "Geoksjur" motifs are painted on a fine whitish wheel-thrown pottery very different from the coarse, vegetal-tempered pottery of the Namazga III phase, mostly handmade. The situation is to some extent similar at Mehrgarh where we have a wide range of pots decorated with "Quetta" designs but mostly in the fine grey ware, a local production exported and imitated at Mundigak (period IV) and at Shahr-i-Sokhta (periods II-III). It seems to me that a systematic analysis of "Hilmand Civilization" and of the sites of the Quetta plateau would indicate that we are dealing with too complex and composite entities for interpreting them as the result of a migration. It is worth noting that a few aspects of Shahr-i-Sokhta in the field of burial practices, craft techniques,


      54. Books & Books, New Delhi, 1985.

      55. Pp. vi-vii.

material life and ideology, can be related to earlier traditions in Baluchistan or in south Afghanistan as we know them from Mehrgarh or from Mundigak (periods I and II). Exchanges, influences and contacts between Turkmenia, eastern Iran and the Greater Indus system have no doubt played an important role in the shaping of the various cultural entities of these regions at the end of the 4th millennium but these phenomena were obviously multidirectional.


      Turning to the Swat valley, we see again a situation which has no force to compel a belief in a mass immigration of whatever intrusive culture we may find in the Ghalegay IV period (c. 1600-1400 B.C.). Parpola56 has picked out from Period IV its objects and the iconographic motifs on its painted-red pottery, which "in some cases derived from the tradition of the Indus urban civilization and in others more specifically recall the culture of Cemetery H of Harappa." The fact that, as Stacul quoted by Parpola57 observes, "the Cemetery H culture is generally interpreted as a fusion of Indian traditions and new elements, probably from the west", does not join it with the later phase of Ghalegay IV by any descent of the latter from the Swat valley into the plains area. Indeed, there is no direct evidence of any immigration by such a descent. Parpola58 seems to lend a Rigvedic colour to Ghalegay IV when he writes about one of its Swat locations: "At Bir-kot-ghwandai, the painted motifs of this intrusive red ware comprise the three-branched fig, known already from Mundigak IV. 1 (c. 2600 B.C.) and the horse." The presence of the horse may indicate Aryanism in the Swat region, but cannot make the Aryanism Rigvedic on the sole strength of that presence. Perhaps it may be asked: "Would an immigration from horse-knowing Swat in c. 1600-1400 B.C. make a Rigvedic invasion?" What it can


      56. P. 242.

      57. Ibid.

      58. Ibid.

make is an Aryan entry, but to make it Rigvedic Aryan several hurdles will have to be jumped.


No Rigvedic entry possible in c. 1600-1400 B.C.


      A number of archaeological items rule out a Rigvedic entry in this period. To start with: the Togolok-21 temple is described by Parpola59 as having "two brick-faced altars dug into the earth" and as containing "a round fireplace with a central cavity." The fire-altars at Kalibangan, in Sankalia's words,60 "consist of shallow pits oval or rectangular in plan" and he adds: "around or near about were placed flat rectangular or circular terracotta pieces, known hitherto as 'terracotta cakes'." All these structures definitely indicate Aryanism. Yet they cannot be related to the Rigveda. Stuart Piggott61 correctly says about the Rigvedic Age: "There is no evidence that any temples were built, and the altar is nothing more elaborate than a pile of turf." Parpola62 himself notes in one context: "Besides the implements needed in the preparation of Soma and the sacrificial fire, the sacrificial place contained little beyond a shallow bed dug out and covered with grass for the gods to sit on." On another page he63 informs us that "the brick-built fire altar ... is never mentioned in the Rgveda." In fact, even the existence of bricks - such a marked feature of the Indus Valley Civilization - cannot be traced in the Rigveda. The Rigveda, flourishing in the same locale - the valley of the Indus - has no word for 'brick': istakā occurs only in later literature.


      Next we have Parpola's listing64 of "two silvery trumpets" and, earlier, of "goblets" not only of "gold" but also of


     59. Pp. 237 & 240.

      60. Op. cit., p. 350, col. 2.

      61. Prehistoric India (A Pelican Book, Harmondsworth, 1960), p. 283.

      62. P. 225.

      63. P. 250.

      64. P. 204.

"silver" among the "traits" of the Namazga V culture. Even the Aryanism of people who are not said to have entered India is associated with this metal. Parpola65 refers to Ghirshman's pointing out the rich civilization of Hissar III, ruled by a military aristocracy using bronze weapons, "some ornamented with silver", and adduces as a particularly important evidence for their Aryan identity a locally made cylinder seal of alabaster from Hissar IIIB level representing a horse-drawn two-wheeled war-chariot. "The Rigvedic Indo-Aryans," writes A.L. Basham,66 "were ... acquainted with ... metallurgy, although they had no knowledge of iron.... Gold was familiar and made into jewellery." He67 refers bronze and copper implements to Vedic times, but is silent about silver. At another place he68 tells us: "where the. Rigveda speaks only of gold and copper or bronze the later Vedic texts also mention tin, lead, and silver, and probably iron." A.A. Macdonell69 makes the statement:


      Among the metals, gold is most frequently mentioned in the Rigveda.... The metal which is most often referred to in the Rigveda next to gold is called ay as (Latin aes).... In most passages where it occurs the word appears to mean simply 'metal'. In the few cases where it designates a particular metal, the evidence is not very conclusive; but the inference which may be drawn as to its colour is decidedly in favour of its having been reddish, which points to bronze and not iron.... It seems quite likely that the Aryans of that period were unacquainted with silver, for its name is not mentioned in the Rigveda....


      65. P. 205.

      66. The revised part dealing with Ancient India in the Third Edition (1970) of The Oxford History of India by the late Vincent A. Smith, edited by Perceval Spear, p. 516.

      67. Ibid.

      68. Ibid.

      69. A History of Sanskrit Literature (New impression, William Heinemann Ltd.. London, 1928), p. 151.

No scholar of India's most ancient scripture breathes a word about silver. On this score the Rigveda goes out of the chronological framework within which Parpola speaks of Aryanism in India or in Greater Iran.


      Here an excursus apropos of metals would be in place. I shall lead up to it via some observations of Parpola's. As we have seen, he tallies the time of the Rigveda with the Ghalegay IV-V periods. The latter's being chalcolithic except for a little iron at the end is compared to "the textual evidence" purported to be gathered from the Rigveda. Iron is said to be hard to find in this scripture and to be present in the Atharvaveda's mention of the black metal. I believe Parpola is mistaken in his reading both of the Rigveda and of the Atharvaveda. Basham's statement which we have quoted that the Rigvedic Indo-Aryans were ignorant of iron is correct. Parpola himself in his citations from the Rigveda accepts the rendering of the term āyasī by 'copper' and not by 'iron' as done at an earlier period of scholarship. Thus he70 has "copper forts" in 1,58,8; "unattackable copper forts" in 10,101,8; and "a copper fort" in 7,95,1. According to him71 Book 7 is one of the earliest parts of the Rigveda, whereas Books 1 and 10 are the latest. 'Copper' holds for all the Rigveda in Parpola's usage. Hence this scripture in no part can be put in the Ghalegay V-VI periods, which admit "a little iron at the end." Nor can the Atharvaveda's 'black metal' be so facilely equated to 'iron'. If the Rigveda's ay as is generically 'metal' except when called 'red' to mean bronze (or copper) and if bronze is darker than copper, the Atharvaveda's śyāmāyas or 'black metal' as distinguished from its lohitāyas or 'red metal' could easily be, as D.H. Gordon72 opines, the darker-than-copper bronze, an alloy of copper and tin. In so late a Vedic text as the Śatapatha Brāhmana (V>4,1,2) we have three classes: ayas, lohāyasa,


      70. P. 212, fn. 141.

      71. P. 225.

      72. The Prehistoric Background of Indian Culture (Tripathi Ltd. Bombay. 1959). P. 153.

hiranya (='gold'). Its ay as is depicted (VI, 1,3,5) as resembling gold; so it would be 'brass', an alloy of copper and zinc; while lohāyasa or 'red metal' would be 'copper'. In the earlier Taittirīya Brāhmaṇa (111,62,6,5) krsnayasa would signify the same thing as śyāmāyas. The Chhāndogya Upanishad (VI, 1,6) has the same term as well as the derivative kārsnāyasa. In the vastly composite Mahābhārata, ay as to denote iron would depend on the period in which the section concerned of the poem was composed. Terms change their meanings in different times. Thus loha, involving redness, which denoted copper in combination with ayas or even by itself, is today applied to iron exclusively. It appears in the same form in the Manusmriti (IX,321) which contains late as well as early matter and there it could point to iron. We may be sure of this metal only with loha in literature which is not appreciably ancient.


      Therefore, not only does the Rigveda pass beyond the epoch Parpola chooses for it, but also much of subsequent literature can be taken to precede the Iron Age in India which seems to have its earliest phase at Pirak where in Period III, c. 1100 B.C. the use of iron, according to Parpola73, begins.


More items against Parpola


      The next item which excludes the Rigveda is rice. "Rice cultivation on a large scale," writes Parpola,74 "is evidenced for the first time in the Indus valley in the post-Harappan period at Pirak in the Kachi plain, right from the beginning of period I dated to c. 1800 B.C. [In Jarrige and Santoni's words:] 'The Ganges valley, where numerous points of bone and ivory that are similar to the Pirak ones were carved, is also one of the earliest rice-growing centres.' The introduction of rice from the mid-Ganges valley to the borders of Baluchistan coincides with the strengthening of contacts


      73. P. 264.

      74. P. 207.

between these regions around 2000 B.C." The implications of Parpola's pronouncement for the Harappa Culture itself are not clear. Perhaps we should concentrate on the words, 'on a large scale', for rice is already present in the Indus Civilization. It has been recognised at at least three Harappan sites: not only outside the Indus valley, at Rangpur IIA (2000-1500 B.C.) and at Lothal (c. 2200 B.C.), but also in the valley at Mohenjo-daro (c. 2500 B.C.).75 The situation is quite negative for the Rigveda. Stuart Piggott76 has the clear-cut assertion: "the Rigveda knows nothing of rice." San-kalia77 has a statement of double information: "this grain was unknown to the Rigveda as well as to the Avesta" -which adds a distinct chronological aspect to the parity often underlined between these two scriptures, especially in relation to the Avestan Gāthās which are the oldest Zarathus-trian compositions and linguistically most comparable with the Rigveda. Thus the latter scripture goes beyond the Harappa Culture no less than the post-Harappan Pirak cultural phase. Later we shall calculate more precisely how far back it can go. A pre-Harappan antiquity in general applies in connection with silver as well. For, Bridget and Raymond Allchin78 admit: "Silver [in India] makes its earliest appearance, to date, in the Indus civilization."


      A few other items from the archaeological-cwm-literary angle may be taken as relevant too. Piggott's full assertion apropos of the Rigveda that it knows nothing of rice continues: "nor of the tropical animals such as the tiger"; and he adds about this animal and rice, "both of which are mentioned in the Atharvaveda", and ends with the news: "the tiger is depicted on the Harappa seals." The tiger too,


      75. D.H. Grist, Rice, Fourth edition (Longmans, 1965).

      76. Op. cit., p. 259.

      77. Indian Archaeology Today (Ajanta Publications, Delhi, 1979), p. 109.

      78. The Birth of Indian Civilization (A Pelican Original, Harmonds-worth, 1969), p. 285.

therefore, argues for the posteriority of the Harappa Culture to the Rigveda.


      Sankalia's pointer to both the Rigveda's and the Avesta's lack of knowledge of rice is in the context of the claim made by some scholars that the ceramic defined as Painted Grey Ware (PGW), in its earliest stage represented the Rigvedic Aryans. As the people of PGW were rice-eaters he cannot accept this claim. Parpola79 leaves the question open: he refuses to enter into the polemics pro and con. But he80 does say:


      There is now considerable agreement concerning the correlation of the archaeological complex characterized by the luxury ceramic called Painted Grey Ware (PGW), and the culture of the later Vedic Aryans of the Brāhmaṇa and Sutra period.... The upper temporal limit for the PGW culture is between c. 1100 and 800 B.C. and the lower limit between c. 400 and 350 B.C. It flourished in a continuous zone stretching from the Punjab and the course of the Sarasvatī and Drsadvatī rivers to the middle Ganges region. The horse was an important animal; iron was used, although it appears to have been scarce at the early sites in the Punjab; and, in the early phase, the settlements were not cities but villages with impermanent huts as ordinary dwellings. The economy was based on cattle-raising and cultivation of rice, barley and wheat. No graves or burials have been found at any PGW site. Cremation, therefore, was perhaps the usual manner of disposing of the dead as in the Vedic culture. Many of the PGW sites figure centrally in the Mahābhārata.... The epic age thus corresponds to the late, fully urban phase of the PGW. Only very few towns are mentioned in the Brāhmaṇa texts, which therefore had been completed during the oldest phase of the PGW, before about 750 B.C. It has been unclear how exactly the PGW culture is


      79. P. 198.

      80. Pp. 197-98.

linked with the cultures of Northwest India, the Iranian Plateau and Central Asia, and thus with the earliest Vedic period and its Indo-Iranian background.


      It is surprising that Parpola should speak of a "late fully urban phase" and of an early phase when there were no "cities". Lal81, who first brought the PGW into prominence, emphasises the opposite as late as 1977:


      As seen from the extensive excavations at the various sites in Panjab, Haryana, Rajasthan and Uttar Pradesh, the P.G.W. culture was essentially a rural one. Most of the settlements were small villages, there being hardly a few which could lay claim to being called 'towns'. But by no description could any of these settlements be called cities like the ones we had in the case of the Indus Civilization. There is hardly any evidence of town-planning and the houses were made of wattle-and-daub or mud, or at best mud-bricks. No house of kiln-burnt bricks has yet come to light from any of the PGW sites, though there is an indication of the knowledge of such bricks.


As late as 1979 Sankalia82 could write:


      Now the final question is posed: could this ware belong to the Mahabharata War period? Lal and all the subsequent writers, including myself and Dr. Vibha Tripathi, think that the culture represented by the Painted Grey Ware  and the things so far found with it suggest that it was at most a village culture with "advanced economy".


      This conclusion goes against our assumed view of the Mahabharata War Period, when there were several states, each with a specific name, such as Kuru, Panchala, Chedi, Kekaya, Sindhu-Sauvira, Magadha, etc., each with its own capital city. So, if we regard the Painted Grey Ware


      81. Op. cit., pp. 33-34.

      82. Op. cit., p. 95.

as a village culture we shall have to revise our opinion about the Mahabharata time.


      However, when I was preparing this work, it occurred to me that Painted Grey Ware could very well be described as a culture which flourished in the wake of Mahabharata, after the destruction of the kings of Northern India, with their armies, in Northern India - Punjab, Northern Rajasthan, Haryana, U.P. and Bihar. Thus, the Painted Grey Ware could very well be regarded as a post-Mahabharata culture.


      If PGW is 'post-Mahabharata', the war concerned might reasonably be placed around the date - c. 1400 B.C. -recently suggested by the excavations of a submerged Dwaraka which could be identified with the Dwaraka reported by tradition as having been drowned in the time of Krishna, one of the main participants in that war. Then the Rigveda, which is admitted to be quite a number of centuries earlier than this war, could never be post-Harappan, filling the period sought to be rendered plausible by Parpola's c. 1600-1400 B.C.


      To dissociate PGW from any relation to a possible entry of the Rigvedics into India or even any other type of Aryans which might have come within the time-bracket which Parpola proposes, we may take a look at what J.G. Shaffer83 had to say in 1984 after weighing the pros and cons for the kind of picture Parpola has drawn:


      If PGW represents the Indo-Aryans, then according to accepted theories, similar or antecedent types of pottery should be located west of the Ganga-Yamuna region on the Iranian plateau. B.K. Thapar (1970), has noted the absence of any PGW antecedent types of pottery any-


      83. "The Indo-Aryan Invasion: Cultural Myth and Archaeological Reality", The People of South Asia - The Biological Anthropology of India, Pakistan and Nepal, edited by John R. Lukacs (Plenum Press, New York and London, 1984), p. 85.

where along the route supposedly taken by the Aryans, and he has outlined the chronological problems associated with existing accounts. Chakrabarti, on the other hand, has proposed an eastern, rather than western, origin for the PGW, thereby negating the PGW-Aryan correlation.


      The Painted Grey Ware culture, thus, with its traits of rice cultivation and the use of domestic pig and buffalo seems to suggest a culture distinctly eastern and not a western one as its suggested Aryan authorship would indicate. (Chakrabarti, 1968, p.353)


      Recent archaeological research in Eastern Punjab (Shaffer, 1981) substantiates objections to the PGW-Aryan correlation.


      J.P. Joshi's excavations (1976,1977, 1978a, 1978b, Joshi and Madhu, 1982) at Bhagwanpura, Dadheri, Nagar and Katpalon have significantly altered the perspective of the archaeological sequence in the Punjab, particularly that regarding the PGW culture. At these sites, Joshi found PGW pottery and structures associated in the same strati-graphic unit with material belonging to the indigenous protohistoric culture of this region - Siswal. Moreover, Joshi was able, for the first time, to associate substantial mud-brick architecture units with PGW, and to define overlapping ceramic attributes between the Siswal and PGW cultures. At the same time, Chakrabarti (1974, 1977) and I (Shaffer, 1983) argue for an indigenous development of iron technology within the Indian subcontinent. At present, the archaeological record indicates no cultural discontinuities separating PGW from the indigenous protohistoric culture. That is, PGW culture represents an indigenous cultural development and does not reflect any cultural intrusion from the West, that is, an Indo-Aryan invasion.


      Two conclusions may be drawn from the archaeological data. First, there is no connection between PGW culture and that of the Aryans. Second, if the "Aryan" concept is to have any cultural meaning, then such a culture (PGW)

had an indigenous South Asian origin within the protohistoric cultures of the Gangā-Yamunā region.


      Of course, all this does not mean that at no period does PGW evince contact with the West. Parpola84 writes: "A link from the Ghalegay V culture [of Swat] to the Painted Grey Ware (PGW) is supplied by the urns with perforations near the neck (resembling the eyes and the mouth of the Ghalegay V 'face-urns') in the PGW layers of Ahicchatra and of [by?] Ghalegay V type terracotta figurines in the PGW layer of Jakheran. U.P." But we cannot endorse Parpola when he goes on to say: "... Rgvedic Aryans started moving from Swat to the plains of Punjab during the latter half of the Ghalegay IV period, c. 1600-1400 B.C. and continued during the following Ghalegay V period [1400-800 B.C.85]. After this, the northwest developed in relative isolation, losing all contacts with the Late Vedic culture of the plains, associated with the early PGW." Even in Parpola's own universe of discourse everything hangs in the air so long as we remember his admitting: "It has been unclear how exactly the PGW culture is linked with the earliest Vedic period and its Indo-Iranian background."


      Nothing from archaeology appears to stand in the way of the pre-Harappan antiquity we ascribe to the Rigveda. And this antiquity seems, absolutely clinched by a couple of considerations.


Two clinching arguments


      Several of the arguments we have mustered - those relating to silver, rice, tiger - may be termed e silentio. But this characteristic should in no way diminish their force. The word 'several' counteracts the weakness one may theoretically see in them. How is it that these three things which distinguish from the preceding time the immediate post-


84. P. 248.

85. P. 244.

Harappan period in which the Rigveda is currently fitted are together absent from the Rigvedic age? Their combined non-existence prevents the argumentum e silentio from being insufficient or inconclusive.


      Furthermore, as regards silver, we can go beyond the mere though significant fact of its absence. From the linguist A.C Greppin86 we gather the following information. In the early Sanskrit texts the word rajatá which has the same root as the Greek arguros, the Latin argentum, the Armenian arcat' and the Celtic argat does not by itself denote silver as do all the other terms. It simply means 'white'. In those early texts the expression for silver is rajatám hiranyam, literally 'white gold'. The next step after Greppin is to note that the common word for 'white' in the Rigveda, the earliest Sanskrit text, is śvetá or śukrá. But rajatá does occur just once in 8,25,22. The verse concerned along with its successor reads, in Ralph T.H. Griffith:87


      From Uksanyāyana a bay, from Harāyana a white steed,

      And from Susāman we obtained a harnessed car.

      These two shall bring me further gain of troops of tawny-coloured steeds,

      The carriers shall they be of active men of war.


      In the original, we have rajatám without any noun to qualify; but the general context of the first verse and even more that of the second where steeds of tawny colour are mentioned after a reference to 'these two' make an implied white steed pair with a bay. The sense of silver is impossible with a horse, especially in the company of other horses with common colours. And if early Sanskrit knows silver only as  rajatám hiranyam, the Rigveda's rajatám - whatever it may


      86. Review of J.P. Mallory's book, In Search of the Indo-Europeans: Language, Archaeology and Myth (Thames and Hudson, London, 1989) in the Times Literary Supplement, August 11-17, 1989, p. 881, col. 4.

      87. The Hymns of the Rigveda, translated with a popular Commentary, edited by J. L. Shastri (Motilal Banarsidass, Delhi, 1973), p. 417, col. 1.

qualify - can denote nothing else than 'white'. There cannot be the slightest suspicion of silver in the Rigveda's period. Thus by the additional force of linguistics and not exclusively by the non-mention of silver can this metal be ruled out from the ken of the Rigveda and that scripture be dated as pre-Harappan.


      Here is indeed a clinching argument. Nor is it the sole one available. A clinching argument even e silentio takes shape from consideration of one particular commodity: cotton. I have elaborated the consequences of the silence here in a separate book: Karpāsa in Prehistoric India: A Chronological and Cultural Clue (Biblia Impex, Delhi, 1984). I shall not attempt to summarize its whole range but pick out two salient features which have a conclusive bearing in the present discourse.


      Dealing with the Indus Civilization Lal88 writes: "Perhaps the most remarkable agricultural achievement was the cultivation of cotton. Even Egypt did not produce it until several centuries after it was grown in the Indus valley." Mark the word 'cultivation'. At Mehrgarh on the Bolan River in Central Baluchistan Jean-François Jarrige and Richard H. Meadow89 found hundreds of cotton seeds in a hearth belonging to Period II dating back to the fifth millennium B.C. But the discoverers90 tell us: "The cotton seeds were so poorly preserved that Constantini has not yet been able to determine whether they came from a cultivated form of the plant." One may conjecture cultivation, but there is no actual ground for doing so and, even if we suppose cultivation, we cannot say whether it was for the plant's fibre or for its oil-rich seeds. In any case, there is no sequel either at Mehrgarh itself or in regions connected with it. In the nearly two thousand years between Mehrgarh's


      88. "The Indus Civilization", A Cultural History of India, edited by A. L. Basham (Clarendon Press, Oxford, 1975), p. 17.

      89. "The Antecedents of Civilization in the Indus Valley", Scientific American, August 1980, p. 124, col. 3.

      90. Ibid., p. 128, col. 3.

Period II and the mature Harappa Culture in about 2500 B.C. we do not come across the veriest trace of even wild cotton in spite of extensive excavation. So we may safely credit the statement of the Allchins91 in 1968: "The earliest evidence for the cultivation of cotton comes from the Indus Civilization."


      Otherwise too, merely the history of cotton-cultivation would be affected. The situation as between the Harappa Culture and the Rigveda would be the same. For, in the very Indus valley where the Harappa Culture flourished, the Rigvedics are said to have established themselves in the wake of the Harappans and yet they give not the least sign of knowing cotton. Centuries of handling cotton both for home use in clothing and for export to Sumer where, according to W.F. Leemans92, "an impression of it on clay has been found at Ur", had no consequence at. all for the Rigvedics. Macdonell93 informs us about their dress: "Clothes were woven of sheep's wool, were often variegated and adorned with gold." Sir John Marshall's colleague, Rao Bahadur Dayaram Sahni,94 the discoverer of the first trace of cotton at Mohenjo-daro in the Indus Civilization, brings out a startling fact. Glancing at what is broadly dubbed Vedic literature "from the Rigveda down to the Sutra period" - a large span of time even by the current chronology which, as Basham95 calculates, starts the Rigveda in c. 1500 B.C. and puts the most important Sutras between the 6th and the 2nd centuries B.C." - Sahni reports: "The Vedic literature ... contains numerous references to weavers, the art of weaving, the weaver's shuttle, wearing of clothes like turbans, shirts, etc., soiled garments and washermen. But whereas


      91. Op. cit., p. 266.

      92. Foreign Trade in the Old Babylonian Period (E. J. Brill, Leiden, 1960), p. 166.

      93. A History of Sanskrit Literature, p. 164.

      94. Archaeological Survey of India, Annual Report 1926-27, p. 55.

      95. The Wonder that was India (The Grove Press Inc., New York, 1961), p. 32.

wool (śāamulya) and silk (tārpya) are mentioned, cotton (Karpāsa) is unknown from early texts."


      Actually, cotton is first spoken of in the two oldest Sutra compositions listed by Basham,96 the Gautama and the Baudhāyana Dharma-sutras - e.g., in the former's 1.18 and the latter's 16,13,10. Between the Rigveda and these two books intervened the three other Vedas, the numerous Brāhmaṇas, Aranyakas and the early Upanishads, all of them innocent of cotton. Is it possible that, if these works came after the Indus Valley Civilization, this civilization's cotton industry would be followed by a complete blank about Karpāsa in them? Here is a colossal silence which is really thunderous! And it is not that the industry was confined to Harappan centres in the Indus valley where the Rigveda was composed. The more inland country where some of the post-Rigvedic literature was produced is also shown to have been cotton-producing or at least cotton-using. Wheeler97 refers to a "reputed example from Lothal" - that is, in Gujerat. Sankalia98 reports cotton at Nevasa (Ahmedabad District) and at Alamgirpur near Delhi - with the latest date c. 1000 B.C. He99 lists also Maharashtrian Chandoli whose C-14 dates range from c. 1330 to c. 1040 B.C. Thus, during the first 500 years or so after the alleged c. 1500 B.C. for the Rigveda's beginning, archaeology attests cotton and we may rationally presume the continuation of the use of it still later. Against the proved presence of this commodity we have the utter lack of the veriest allusion to it in Indian books until we reach the Sutras. So sustained a lack, extending over varied time and space, must carry the Rigveda and its documentary progeny short of the Sutras into an antiquity beyond the post-Harappan epoch and even past the cotton-cultivating Harappa Culture.


      96. Ibid., p. 113.

      97. Op. cit., p. 85.

      98. Letter to the author dated 16 April 1963.

      99. Op. cit., p. 487, col. 1 & p. 565, col. 2.

Horse and spoke-wheeled chariot


      Logically there should be no doubt now about such antiquity and the assumption should be justified that we could legitimately wait for conclusive future evidence against all objections that might possibly be raised against it. But out of curiosity we may try to see whether certain items which meet us either in Harappan or else in post-Harappan times can be conceived of as earlier