India's Parthian Colony - On the origin of the Pallava Empire of Dravidia

India's Parthian Colony
On the origin of the Pallava Empire of Dravidia
By: Dr. Samar Abbas, May 14, 2003

Note: This article uses UniCode Diacritics and due to some technical reasons Internet browsers may not be able to show them properly.

Page: 1 2     Page 1

Revealing the ancient Pallava Dynasty of Dravidia to be of the Iranic race, and as constituting a branch of the Pahlavas, Parthavas or Parthians of Persia. Uncovering the consequent Iranic foundations of Classical Dravidian architecture. Describing A Short History of the Pallavas of Tamil Nadu, including the cataclysmic 100-Years' Maratha-Tamil War. The modern descendants of Pallavas discovered amongst the Chola Vellalas of northern Tamil Nadu and Reddis of Andhra.

1- Pallavas, Pahlavas, Parthavas, Parthians and Persians

Fig.1: Map of Pallava Empire at Greatest Extent c.550-600 AD

1.1. Introduction
The Pallava Empire was the largest and most powerful South Asian state in its time, ranking as one of the glorious empires of world history. At its height it covered an area larger than France, England and Germany combined. It encompassed all the present-day Dravidian nations, including the Tamil, Telugu, Malayali and Kannada tracts within its far-flung borders (see map). The foundations of classical Dravidian architecture were established by these powerful rulers, who left behind fantastic sculptures and magnificent temples which survive to this very day. Initially, the similarity of the words "Pallava" and "Pahlava" had led 19th-century researchers to surmise an Iranic origin for the Pallavas. Since then, a mountain of historical, anthropological, and linguistic evidence has accumulated to conclusively establish that the Pallavas were of Parthian origin.

1.2. Occurrence of Parsas across the world
The wide occurrence of the Iranic root-word Par in various place-names proves the dispersion of the Pars or Persians across much of Asia in ancient times. Thus, Persia, Persepolis, Pasargadae ("Gates of Parsa") and "Parthaunisa (ancient city, Parthia)" or Nisa (Enc.Brit., vol.9, p.173) are all constructed from the ancient Iranic root-word Pars. In this regard, the learned Prof. Waddell notes in his masterpiece "The Makers of Civilization": "Barahsi or Parahsi [of Akkadian inscriptions] now transpires to be the original of the ancient Persis province of the Greeks, with its old capital at Anshan or Persepolis, the central province of Persia to the East of Elam and the source of our modern names of "Persia" and "Parsi". And it is another instance of the remarkable persistence of old territorial names" (Waddell 1929, p.216). The Parsumas mentioned in Assyrian annals are also generally identified with the Persians, and the Zoroastrian Parsis of Maharashtra are clearly of Persic descent. Moreover, the word Parthian is itself derived from Parsa, as the Encyclopedia Britannica notes: "The first certain occurrence of the name is as Parthava in the Bisitun inscription (c.520 BC) of the Achaemenian king Darius I, but Parthava may be only a dialectal variation of the name Parsa (Persian)." (Enc.Brit. Vol.9, p.173)

Further, Euphrates means `Persian river' in Greek, while the biblical Perizzites of Palestine have been linked with the Iranians (Derakhshani 1999, Kap.8). The word "Pastho" in turn has also been connected with the Persians, as the eminent Iranologist Prof. P.Oktor Skjærvø notes:

"It is not impossible that Old Persian Pārsa is descended from an earlier form *Pārswa, but that cannot be proved. On the other hand, modern Paštō (the language of the Afghans) must be descended from Old Iranian *Pārsawa, which is very close to the Assyrian forms. Finally, Old Persian Parθava is also not unlike these forms." (Skjærvø 1995, p.156)

The distinguished Professor Michael Witzel of Harvard University has further identified the Parnoi as the Pani mentioned in the Vedas:

"Another North Iranian tribe were the (Grk.) Parnoi, Ir. *Parna. They have for long been connected with another traditional enemy of the Aryans, the Paṇi (RV+). Their Vara-like forts with their sturdy cow stables have been compared with the impressive forts of the Bactria-Margiana (BMAC) and the eastern Ural Sintashta cultures (Parpola 1988, Witzel 2000), while similar ones are still found today in the Hindukush." (Witzel 2001, p.16)

Thus, the Persians, Parthians, Pashtos, Panis and Perizzites are all offhoots of the ancient proto-Persians. This testifies to the supreme achievements of the glorious Persian branch of the Iranic race in civilizing and colonizing Southern Asia. All this, of course, is well known and the subject of numerous books (cf, eg. Derakhshani 1999). Less famous is the fact that the magnificent Pallava Dynasty of Southern India was also of Iranic descent.

1.3. Pahlava History in Iran
The Pahlavas made important contributions to Iranian civilization. The modern Farsi tongue is derived from the Old Parthian language, as noted by the Encyclopedia Britannica: "Of the modern Iranian languages, by far the most widely spoken is Persian, which, as already indicated, developed from Middle Persian and Parthian, with elements from other Iranian languages such as Sogdian, as early as the 9th century AD." (Enc.Brit.vol.22, p.627) Furthermore, "Middle Persian [Sassanian Pahlava] and Parthian were doubtlessly similar enough to be mutually intelligible." (Enc.Brit.22.624); a statement which further confirms the identity of the Pahlavas and the Parthians.

Moreover, the Pahlava alphabet is the ancestor of the Sasanian Persian alphabet: "The Pahlava alphabet developed from the Aramaic alphabet and occurs in at least three local varieties: northwestern, called Pahlavik or Arsacid; southwestern, called Parsik or Sasanian, and eastern" (Enc.Brit.vol.9, p.62). Some authorities seem to insist that it was the Semitic Aramaic alphabet which gave birth to the Parthian alphabet. This is not so; it was actually the Assyrian variant which developed into the Pahlava characters, just as it was Assyrian art, not Aramaean, which inspired later Achaemenid culture. The Achaemenid empire was in many ways the successor-state of the Assyrian empire.

1.4. Pallavas of Dravidia as Pahlavis
The Pallavas are first attested in the northern part of Tamil Nadu, precisely the geographical region expected for an invading group. This, together with the evident phonetic similarity between the words "Pallava" and "Pahlava", has long led researchers to advocate a Parthian origin of the Pallavas:

"Theory of Parthian origin: The exponents of this theory supported the Parthian origin of the Pallavas. According to this school, the Pallavas were a northern tribe of Parthian origin constituting a clan of the nomads having come to India from Persia. Unable to settle down in northern India they continued their movements southward until they reached Kanchipuram⁵. The late Venkayya supported this view ⁶ and even attempted to determine the date of their migration to the South. A crown resembling an elephant's head was issued by the early Pallava kings and is referred to in the Vaikunthaperumal temple sculptures at the time of Nandivarman Pallavamalla's ascent to the throne. A similiar crown was in use by the early Bactrian kings in the 2nd century BC and figures on the coins of Demetrius. It is presumed on this basis that there is some connection between the Pallavas of Kanchi and Bactrian kings. [ 5. Mysore Gazetteer, I. p.303-304; 6. ASR {Ann.Rep.ASI), 1906-1907, p.221 ]." (Minakshi 1977, p.4)

As Venkayya notes,

"[T]he Pallavas of Kāñcīpuram must have come originally from Persia, though the interval of time which must have elapsed since they left Persia must be several centuries. As the Persians are generally known to (p.220) Indian poets under the name Pārasīka, the term Pahlava or Pallava must denote the Arsacidan Parthians, as stated by Professor Weber." (Venkayya 1907, p.219-220)

Philogists concur in connecting the names Pahlava, Parthava, Parthian and Pallava:

"The word Pahlava, from which the name Pallava appears to be derived, is believed to be a corruption of Pārthava, Pārthiva or Pārthia, and Dr. Bhandarkar calls the Indo-Parthians Pahlavas. The territories of the Indo-Parthians lay in Kandahar and Seistan, but extended during the reign of Gondophares (about AD 20 to 60) into the Western Punjab and the valley of the lower Indus. The Āndhra king Gotamiputra, whose dominions lay in the Dakhan, claims to have defeated about AD 130 the Palhavas along with the Śakas and Yavanas. In the Junāgaḍh inscription of the Kṣatrapa king Rudradāman belonging to about AD 150, mention is made of a Pallava minister of his named Suviśākha." (Venkayya 1907, p.218)

2- Evidence for Parthian Descent of Pallavas

A whole mountain of evidence from various fields of science support the Parthian, and hence Iranic, origin of the Pallavas. It would be of interest to summarise the evidence here.

2.1. Archaeology
Archaeologists note the occurrence of oblong earthenware coffins in sites coinciding with the region of Pallava hegemony:

"Oblong earthenware sarcophagi, both mounted and unmounted, have been reported from several sites in S.India from Maski in the North to Puduhotta in the South. Their distribution in what was during historical period the region of Pallava hegemony is not without significance in the light of a Parthian origin of the Pallavas suggested by Heras (Heras, H.J.: Origin of the Pallavas, J. of the Univ. of Bombay, Vol. IV, Pt IV, 1936) and afterwards by Venkatasubba Iyer ("A new link between the Indo-Parthians and Pallavas of Kanchi", J. of Indian History, Vol. XXIV, Pts 1 & 2, 1945). A possible link between the Parthians and the Pallavas is the mode of tying the waist-band as evidenced by their statuary (compare the knot in Pallava waist-band with knot in Parthian waist-band ...)" (Nair 1977, p.85)

2.2. Administration
Pallava administration was based on the Maurya pattern, which was in turn based on that of the Achaemenid Empire.

"[T]he early Pallava kings issued their charters in Prakrit and Sanskrit and not in Tamil and their early administration was based on the Mauryan-Satavahana pattern, essentially northern in character. Their gotra (Bharadvaja) also stands in the way of their identification with the Kurumbar who had no gotra claims." (Minakshi 1977, p.5)

2.3. Dress
The dress of the Pallavas is cleary Parthian. Thus, Nair notes,

A possible link between the Parthians and the Pallavas is the mode of tying the waist-band as evidenced by their statuary (compare the knot in Pallava waist-band with knot in Parthian waist-band ...)" (Nair 1977, p.85)

The entire city of Mamallapuram or Mahamallapuram in Tamil Nadu is named after the Pallava King Mahamalla who is celebrated as the founder of this city. This original Prakrit name "Mahamallapuram" was later corrupted in the Sanskrit into "Mahabalipuram". In this regard, Venkayya notes the origin of the name "Mamallapuram":

"[I]n ancient Cōḍa inscriptions found at the Seven Pagodas, the name of the place is Māmallapuram which is evidently a corruption of Mahāmallapuram, meaning `the city or town (p.234) of Mahāmalla.' I have already mentioned the fact that Mahāmalla occurs as a surname of the Pallava king Narasiṁhavarman I in a mutilated record at Bādāmi in the Bombay Presidency. It is thus not unlikely that Mahāmallapuram or Māvalavaram was founded by the Pallava king Narasiṁhavarman, the contemporary and opponent of the Calukya Pulikēśin II., whose accession took place about AD. 609. Professor Hultzsch is of opinion that the earliest inscriptions on the rathas are birudas of a king named Narasiṁha. It may, therefore, be concluded that the village was originally called Mahāmallapuram or Māmallapuram, after the Pallava king Narasiṁhavarman I., and that the earliest rathas were cut out by him." (Venkayya 1907, p.233-234)

Fig.2: Pallava King Mamalla or Narasimhavarman I (Dharmaraja Ratha, Mamallapuram, Tamil Nadu). Note the cylindrical Persian hat, long thin nose and long-headedness. (Image by Michael D. Gunther)

Surviving contemporary sculptures of this celebrated King Mamalla depict him wearing a typical cylindrical Iranian head-dress. Furthermore, the elephant-head crown used by Pallava kings resembled those worn by Bactrian kings (Fig.2).

2.4. Prakrit Language
The Pallavas initially propagated Prakrit, a language containing a much higher percentage of Indo-European words compared to Sanskrit as it represented a later, and hence purer, heliolatric Indo-European invasion. "These three Prākṛt grants prove that there was a time when the court language was Prākṛt even in Southern India." (Venkayya 1907, p.223) That they initially did not propagate Sanskrit or Tamil is significant as it rules out a Vedic or Dravidian origin for the Pallavas.

2.5. Toponyms and Personal Names
Evidence from toponyms (place-names) corroborates the Iranic origin of Pallavas. For instance, the Pallavas named a city in Tamil Nadu as Menmatura or Men-Matura, after Mithra, the ancient Iranic Sun-God, formed from tbe consonantal root MTR. The large town in southern Tamil Nadu, Madurai, is named after the Sun-temple city of Mathura in Oudh, which is also based on "Mithra". Further, the Pallavas had a fondness for Iranic Prakrit personal names such as Ashoka:

"In the Kāśākuḍi plates, Aśōkavarman is referred to as the son of king Pallava. Here Aśōkavarman is evidently a reminiscence of the Maurya emperor Aśōka who lived long before the Pallavas." (Venkayya 1907,p.240, footnote 8).

The Pallavas thus sought to emulate the Maurya kings, who were of Iranic origin (Spooner 1915, p.406ff). It is important to note that the Iranic root-word "Mor" occurs all across the Iranian world: consider the "Mardian" tribe of Persians mentioned by Herodotus; "the Avestan name Mourva, the Marga of the Achaemenian inscriptions" (Spooner 1915, p.406), and the city of Merv, also known as "Merw, Meru or Maur", whose inhabitants are known as "Marga and Mourva" (ibid.), the legendary "Meru" mountain, the "Amorites" or "Amurru" of Syria and Palestine who possessed an Iranic ruling caste, the "Amu-Darya" river, "Amol" town just south of the Caspian, "Marwar" in Rajputana, the Oudh towns of "Mor-adabad" and "Meerut", the "Maurya" dynasty of Ashoka, and the "Marut" warriors in India. 2.6. Official Symbolism
To this evidence we may add that the Pallavas had as their crest the lion, just as the Achaemenids carved lions at Persepolis. Describing the cave at Siyamangalam, Venkayya notes:

"This was excavated by king Laḷitāṅkura, ie. Mahēndravarman I. and was called Avanibhājana-Pallavēśvara, Ep.Ind., Vol.VI, p.320. I recently inspected the cave and the two inscriptions found in it. The two outer pillars of the cave on which they are engraved also bear at the top a well-executed lion (one on each of the two pillars) with the tail folded over its back. The tail resembles that of the lion figured in No.54, Plate II. of Sir Walter Elliot's Coins of Southern India, which has been attributed to the Pallavas. It has therefore to be concluded that the lion was the Pallava crest at some period or other of their history." (Venkayya 1907, p.232, ftn.6)

2.7. Anthropology
The depictions of Pallava nobles on sculptures further confirms their Iranic origin, for they are depicted as tall and dolichocephalic (long-headed) along with clearly Iranic features.

Fig.3: Court Scene, Mamallapuram 7th century AD (Pallava). Note the long-headedness and leptorrhine (long and thin) nose of the surrounding Iranic courtiers. Contrast this with the platyrrhine (flat) nose, thick lips and Negroid features of the Dravidian God Shiva standing with his bull in the centre. Note clear Persepolitan influence on the pillars. (Image by Michael D. Gunther)

The long-headedness of these sculptures rules out an Outer Indo-Aryan origin for the Pallavas, while their leptorrhine noses rule out a Dravidian origin.

2.8. Architecture
The architecture of the Pallavas was clearly based on Iranian forms, down to the last detail. Pillars especially were copies of Persepolitan originals (see Fig.4 and Fig.3).

Fig.4: Varaha Cave Temple, Mamallapuram, Tamil Nadu, late 7th century. Note the clear Achaemenid influence on the pillars, and the Persepolitan capitals. Flanking lions are reminiscent of Persepolis and Assyria. (Image by Michael D. Gunther)

2.9. Legendary Descent
The traditional genealogy of the Pallavas also points to their Parthian origins:

"One point which might be taken as proof of the foreign origin of the Pallavas has to be noted here. The indigenous Kṣatriya tribes (or at least those which were looked upon as such) belonged either to the solar or to the lunar race. For instance, the Cōḷas belonged to the solar race and the Pāṇḍyas to the lunar. The Cēras seem to have belonged to the solar race. The Calukyas - both the Eastern and Western - were of the lunar race. The Rāṣṭrakūṭas were also of the same race. On the other hand, the Pallavas trace their descent from the god Brahma but not from the Sun or the Moon, though they are admitted to have been Kṣatriyas. Besides, none of the ancient kings mentioned in the Purāṇas figures in the ancestry of the Pallavas. The indigenous tribes, however, always traced their ancestry from some of the famous kings known from the Purāṇas. The Cōḷas, for instance claimed Manu, Ikṣvāku, Māndhātr, Mucukunda and Śibi; the Pāṇḍyas were descended from the emperor Purūravas; the C ēras had Sagara, Bhagiratha, Raghu, Daśaratha and Rāma for their ancestors. The Calukyas had a long list of Purāṇic sovereigns in their ancestry. The Rāṣṭrakūṭas were descendants of Yadu and belonged to the Sātyaki branch or clan. The Gaṅga kings of Kaliṅganagara were descended from the Moon and claimed Purūravas, Āyus, Nahuṣa, Yayāti and Turvasu for their ancestors. (Ind. Ant. Vol. XVIII, p.170). The Western Gaṅgas of Taḷakāḍ were apparently of the solar race and had Ikṣvāku for their ancestor (Mr Rice's Mysore Gazetteer, Vol.I, p.308). The only king mentioned in the mythical genealogy of the Pallavas is Aśōkavarman, son of king Pallava , who, as Prof. Hultzsch rightly suspects, is probably "a modification of the Maurya emperor Aśōka" (South Ind. Inscrs. Vol.II, p.342). No doubt the earliest Pallava records were found in the Kistna delta. But this cannot be taken to point to an indigenous origin of the family. All these facts together raise the presumption that the Pallavas of Southern India were not an indigenous tribe in the sense that the Cōḷas, Pāṇḍyas and Cēras were." (Venkayya 1907, p.219, footnote 5)

The above evidences, taken together rather than singly, provide almost conclusive proof of the Parthian origin of Pallavas.


3- History of the Pallavas

3.1. Early History: Adoption of Dravidian Culture
After immigrating from Parthia, the Pallavas settled down in the Andhra region. From here they entered northern Tamil Nadu. Initially, the Pallava Empire was restricted to Toṇḍai-maṇḍalam, the northern part of Tamil Nadu: "It thus appears that the Pallava dominions included at the time [Sivaskandavarman, beg. 4th century AD] not only Kāñcipuram and the surrounding province but also the Telugu country as far north as the river Kṛṣṇā." (Venkayya 1907, p.222) Subsequently, the Pallavas expanded to conquer large parts of Andhra:

"The Pallava dominions probably comprised at the time [5th-6th centuries AD] the modern districts of (p.225) Nellore, Guntur, Kistna, Kurnool and perhaps also Anantapur, Cuddapah, and Bellary. The Kadambas of Banavāsi, who were originally Brāhmaṇas, threatened to defy the Pallavas." (Venkayya 1907, p.224-225)

Tamil poets described the boundaries of Toṇḍai-maṇḍalam as follows:

"According to the Toṇḍamaṇḍala-śatakam, Toṇḍamaṇḍalam (ie. the Pallava territory) was bounded on the north by the Tirupati and Kālahasti mountains; on the south by the river Pālār; and on the west by the Ghauts (Taylor's Catalogue, Vol.III, p.29). A verse attributed to the poetess Auvaiyār describes Toṇḍai-maṇḍalam as the country bounded by the Pavaḷamalai, ie. the Eastern Ghauts in the west; Vēṅgaḍam, ie. Tirupati in the north; the sea to the east; and Piṇāgai, ie. the Southern Pennar in the south. The greatest length of the province is said to be full 20 kādam or nearly 200 miles.... A variant of the name Toṇḍai-maṇḍalam is Daṇḍaka-nāḍu, which is apparently derived from the Sanskrit Daṇḍakāraṇya, ie the forest of Daṇḍaka mentioned in the Rāmāyaṇa and the Purāṇas." (Venkayya 1907, p.222, footnote 2)

After settling in Tondai-mandalam, the Pallavas rapidly adopted the Dravidian culture, religion and language of their subjects. This case was not unique in history; there are many examples of ruling classes adopting the culture of those they ruled: consider the Hellenic Ptolemies in Egypt, the Paleo-Siberian Manchus in China, the Germanic Lombards in Italy, the Nordic Visigoths in Spain, the Mongol Il-Khans in Persia, the French-speaking Normans in England, and the Germanic Carolingians, Merovingians, Burgundians and Franks of France. Thus, the Pallavas adopted the Old Tamil language and the Dravidian religion of Shaivism and became vigorous promoters of Dravidian culture.

3.2. Expansion of the Pallava Empire
From its nucleus in Tondaimandalam, the Pallava Empire expanded in all directions. The Pan-Dravidian nature of the Pallava empire is manifested through the extent of their dominions. Thus, the Pallavas vanquished the Cholas, Cheras and Pandyas and conquered their territories, uniting Tamil Nadu, Malabar, Karnadu and Telingana into one giant empire:

"The earliest king of this series is Siṁhaviṣṇu, who claims to have vanquished the Malaya, Kalabhra, Mālava, Cōḷa and Pāṇḍya kings, the Siṁhala king proud of the strength of his arms and the Kēralas." (Venkayya 1907, p.227)

This was the first pan-Dravidian empire in history. Perhaps they were able to unite the Dravidian nations precisely because they were outsiders, and hence did not possess any history of feuding with local clans. Thus, we find the Pallavas conquering all the three mutually warring Pandya, Cola and Cera kingdoms:

"The Cēra, Cōḷa and Pāṇḍya kingdoms of the south are mentioned already in the edicts of the Maurya emperor Aśōka. Of their subsequent history, almost nothing is known from the epigraphical records, until we get to the period of Pallava rule, when all the three figure among the tribes conquered by the Pallavas." (Venkayya 1907, p.237)

After consolidating their rule over the Dravidian nations, the Pallavas extended their empire to South-East Asia:

"The Pallavas were the emperors of the Dravidian country and rapidly adopted Tamil ways. Their rule was marked by commercial enterprise and a limited amount of colonization in South-East Asia, but they inherited rather than initiated Tamil interference with Ceylon." (Enc.Brit. Vol.9, p.89)

However, the exact extent of Pallava colonization in South-East Asia is not clear due to paucity of sources. Even so, the Pallava Empire was the largest South Asian state of its age, and served as the model for future pan-Dravidian empires such as that built by the Cholas.

3.3. The 100-Years' Maratha-Tamil War (AD 634-747) & Decline
The Indian equivalent of Europe's Anglo-French 100-Years' War was the prolonged conflict between Marathas and Tamils under the Chalukyas and Pallava dynasties which lasted well over a century.

"The history of this period consists mainly of the events of the war with the Calukyas which lasted almost a century ⁷ (footnote 7: The war apparently began with the Eastern campaign of Pulikēśin II. which must have taken place some time before AD 634-5 (Ep.Ind., Vol.VI, p.3). The last important event of the war is the invasion of Kāñci by the Calukya king Vikramāditya II, who reigned from AD 733-4 to 746-7. Kirtivarman II, son of Vikramaditya II, also claims to have led an expedition in his youth against the Pallavas. ... ) and which seems to have been the ultimate cause of the decline and downfall of both the Pallavas and Calukyas about the middle of the 8th century." (Venkayya 1907, p.226)

At this point, we may note Mr. Rice's hypothesis that the Calukyas were Seleucids:

"Mr. Rice says: `The name Calukya bears a suggestive resemblance to the Greek name Seleukeia, and if the Pallavas were really of Parthian connection, as their name would imply, we have a plausible explanation of the inveterate hatred which inscriptions admit to have existed between the two, and their prolonged struggles may have been but a sequel of the contests between the Seleucidæ and the Arsacidæ on the banks of the Tigris and Euphrates.' (Mysore, Vol.I, p.320)" (Venkayya 1907, p.226, footnote 6)

However, Mr. Rice's suggestion has not been accepted by other historians, and is merely a phonetic coincidence, for there is no other evidence of any connection whatsoever between the Calukyas and Seleucids.

Historians have found several reasons for explaining the bitterness of the Maratha-Dravidian wars. Venkayya notes the religious aspect of the conflict, with the Vaishnava Marathas on one side and the Dravidian Shaivites on the other:

"No satisfactory explanation has, so far, been offered for this natural enmity between the Pallavas and Calukyas. It is possible that the hatred had a religious basis. The Pallavas were Śaivas and had the bull for their crest, while the Calukyas were devotees of the god Viṣṇu and had the bear for their crest." (Venkayya 1907, p.226, footnote 6)

Shaivism and Vaishnavism are poles apart in all details of theology. Vaishnavites revere the cow, Shaivites slaughter the cow but worship the bull; Vaishnavites uphold the four-fold caste system, Shaivites oppose the caste system tooth and nail; later Vaishnavism upholds the authority of the Vedas and the Brahmans, Shaivism rejects the Vedas and is anti-Brahmin. Thus, observers have noted that Vaishnavism and Shaivism are like cat and mongoose, theologically destined to be locked in an eternal war of opposites. Hence, religion played an important role in exacerbating the hatred on both sides.

However, a far deeper reason contributed to the conflict, namely that of ethnicity. Abstract theological formulae, on account of their nebulous definition and easily modified nature, no doubt hardly mattered to the great majority of inhabitants. Rather, it is race and ethnicity which combined to make the Pallava-Chalukya conflict especially bitter. Thus, the so-called Calukya-Pallava dynastic conflict was in actual fact a racial Maratha-Dravidian war.

On the one hand were the Marathas speaking Outer Indo-Aryan languages, of brachycephalic (round-headed) Turanoid race. The survival of Burushaski - a language isolate linked with the Transcaucasian and Finno-Ugric languages - in the Himalayas testifies to the immigration of brachycephalic Turanian peoples into India. The Turanoid Maratha is thus fair-skinned, short-statured and round-headed. On the other hand were the long-headed and taller, black-skinned Dravidians of Sudanic Negroid origin. The Dravidians, however, had a long-headed Iranic Pallava ruling class. The Iranoid longheads are fairer and taller than the Dravidoid longheads, who are in turn taller but darker than the Turanoid Outer Indo-Aryan roundheads. Thus, racial differences no doubt played, along with language and religion, a prominent role in the conflict.

At the outset of the 100-year Maratha-Tamil War, it is the Marathas who gained the upper hand, defeating the Pallavas and driving them from the Vengi delta area of Andhra. However, the Pallavas later defeated the Maharashtrians and sacked their capital Vatapi, annexing it to the Dravidian Empire:

"The son of Mahēndravarman I. was Narasiṁhavarman I., who retrieved the fortunes of the family by repeatedly defeating the Cōḷas, Kēralas, Kalabhras and Pāṇḍyas. He also claims to have written the word `victory' as on a plate, on Pulikēśin's back, which was caused to be visible (ie. which was turned in flight after defeat) at several battles. Narasiṁhavarman carried the war into Calukya territory and actually captured Vātāpi, their capital. This claim of his is established by an inscription found at Bādāmi in the Bombay Presidency - the modern name of Vātāpi - from which it appears that Narasiṁhavarman bore the title Mahāmalla. In later times, too, this Pallava king was known as Vātāpi-koṇḍa-Naraśiṅgappōttaraiyan. Dr. Fleet assigns the capture of the Calukya capital to about AD 642. ⁷ The war of Narasiṁhavarman with Pulkēśin II is mentioned in the Singhalese chronicle Mahāvaṁsa. It is also hinted in the Tamil Periyapurāṇam. The well-known saint Śiṛuttoṇḍa, who had his only son cut up and cooked in order to satisfy the appetite of god Śiva disguised as a devotee, is said to have reduced to dust the city of Vātāpi for his royal master, who could be no other than the Pallava king Narasiṁhavarman.⁹ [footnote 9: Ep.Ind. Vol.III, p.277. . Paramēśvaravarman I. also claims to have destroyed the Calukya capital. A still later conquest of Vātāpi is also known. It was effected by a Koḍumbāḷūr chief, apparently during the second half of the 9th century. (Ann.Rep. on Epi. for 1907-8, Part II, para.85)] The Śaiva saint Tiruñānasambandar visited Śiṛuttoṇḍa at this native village fo Tirucceṅgāṭṭaṅguḍi, and the Dēvāra hymn dedicated to the Śiva temple of the village mentions the latter and thus helps to fix the date of the former as well as of the Śaiva revival of which he was the central figure." (Venkayya 1907, p.228)

Unsung and forgotten are the countless heroes on both sides, their deeds and brave acts lost in the mist of time, yet heroes they were nevertheless. Like the knights of the 100-Years' Anglo-French War, the glorious warriors of the 100-Years' Maratha-Tamil War fought and died for their homelands, strengthening these nations' foundations with their blood and bones.

This 100-year Maratha-Tamil war had far-reaching consequences, leading to the exhaustion of both the Maratha and Dravidian states and sapping their vitality. These states started to decline after the war. Ultimately, both the Calukya and Pallava states disappeared from history.

3.4. Modern-Day Pallavas
After the Pallava Empire was annexed by the Chola Empire, the Pallavas merged into the Tamil population:

"The Pallavas of the Tamil country seem to have taken service under the Cōḷas after the Gaṅga-Pallavas were conquered by Āditya about the end of the 9th century AD. Karuṇākara Toṇḍaimāṇ, who, according to the Tamil poem Kaliṅgattu-Paraṇi led the expedition against Kaliṅga during the reign of Kulōttuṅga I. (AD 1070 to about AD 1118), was a Pallava and was the lord of Vaṇḍai, ie. Vaṇḍalur in the Chingleput District. Among the vassals of Vikrama-Cōḷa mentioned in the Vikkirama-Śōḷaṇ-ulā, the Toṇḍaimāṇ figures first." (Venkayya 1907, p.241)

The Pudukkottai royal family is apparently descended from the ancient Pallavas:

"In a Tanjore inscription belonging to a later period, the name Toṇḍaimāṇ is applied to a local chief named Sāmantanārāyana, who granted to Brāhmaṇas a portion of the village of Karundiṭṭaiguḍi, the modern Karattaṭṭāṅguḍi. Thus the name Toṇḍaimāṇ actually travelled from the Pallava into the Cōḷa country. There is therefore reason to suppose that the Toṇḍaiman of Pudukkōṭṭai, who bears the title Pallava Raja, is descended from the Pallavas, who form the subject of this paper." (Venkayya 1907, p.242)

In addition to the royal family of Pudukkottai, other groups are also probably descended from the Pallavas, such as the Reddis of Andhra and some of the Kshatriya and Vaishya castes of northern Tamil Nadu:

"We now have to examine if there are any Pallavas in our midst beyond the royal family of Pudukkōṭṭai. The Pallavas are believed to be identical with the Kurumbas, of whom the Kurumbar of the Tamil country and the Kurubas of the Kanarese districts and of the Mysore State may be taken as the living representatives. The (p.243) kings of the Vijayanagara dynasty are also supposed to have been Kurubas. In one of the inscriptions of the Tanjore temple belonging to the 11th century, a certain Vēlāṇ Ādittaṇ is called Pirāntaka-Pallavaraiyan, meaning "the chief of the Pallavas of Parāntaka." Śēkkiḷār, the author of the Tamil Periyapurāṇam, was a Veḷḷāḷa by caste and got from his patron, the Cōḷa king Anapāya, the title Uttamaśōḷa-Pallavarāyaṇ, meaning "the chief of the Pallavas of Uttamaśōḷa." Uttamaśōḷa and Parāntaka are titles of Cōḷa kings and the word Pallava seems to be used in both of the titles as an equivalent of Veḷḷāḷa, or the caste of agriculturalists to which both of them belonged. In the Telugu country, too, some of the Reḍḍis who belonged to the fourth or cultivating caste, called themselves Pallava-Triṇētra and Pallavāditya. Sir Walter Elliot has told us that Pallavarāja is one of the thirty gōtras of the true Tamil-speaking Veḷḷāḷas of Madura, Tanjore and Arcot. It is borne by the Cōḷa Veḷḷāḷas inhabiting the valley of the Kāvēri, in Tanjore, who lay claim to the first rank. All these facts taken together seem to show that there was some sort of connection between the cultivating caste and the Pallavas in the Tamil as well as in the Telugu country. The available evidence is, however, not sufficient to formulate the nature of this connection. But it may tentatively be supposed that some of the Pallavas settled down as cultivators soon after all traces of their sovereignty disappeared. The other sections of the agricultural class were probably proud of their association and considered it an honour to be looked uon as Pallavas." (Venkayya 1907, p.242-243)

Page 2

4- Iranian Origin of Dravidian Architecture and Contribution to Dravidian Civilization

4.1. Iranic Origin of Dravidian Architecture The Pallava foundations for Dravidian architecture is universally accepted by scholars. For instance, a standard textbook on World Architecture states, "Mahabalipuram, the five temples (rathas), Pallava (7th century AD), are embryonic models of later Dravidian, or Southern, temple styles." (Holberton, p.55). Confirming this view, the Encyclopedia Britannica notes:

"The home of the South Indian style, sometimes called the Dravida style, appears to be the modern state of Tamil Nadu ... The early phase, which, broadly speaking, coincided with the political supremacy fo the Pallava dynasty (c.650-893), is best represented by the important monuments at Mahabalipuram." (Enc.Brit., Vol.27, p.767)

Suthanthiran summarises the views of various eminent scholars:

"The prototypes of later developed Kopurams are found in the Pallava period. There are different views regarding the proto-types. Heinrich Zimmer was of the view that the Pimaratam is the earliest prototype of the Kopurams. Raghavendra Rao says that the finished oblong plan and the two storeyed waggon roof of Kanesaratam is the prototype of all South Indian Kopurams ... A.H. Longhurst says that the Kailasanatha temple entrance Tavaracalai is the proto-type of all later Kopurams." (Suthanthiran 1989, p.30)

Venkayya agrees with the Pallavite origin of Dravidian architecture:

"We now enter into a period of Pallava history for which the records are more numerous. The facts available for this period are definite and the chronology is not altogether a field of conjecture and doubt. The earliest stone monuments of Southern India belong to this period. In fact, the foundations of Dravidian architecture were laid by the earlier kings of this series.⁵ (footnote 5: The monolithic caves of the Tamil country were excavated by the Pallava king Mahēndravarman I. The rathas at the Sevan Pagodas probably come next. The temples of Kaliēsanētha and Vaikuṇṭha-Perumal at Kañcīpuram and the Shore temple at the Sevan Pagodas have probably to be taken as later developments of Pallava architecture.)" (Venkayya 1907, p.226)

Fig.5: Stupendous Granite Kailasanatha temple (formerly Rājasiṁhēśvara), Tamil Nadu, view from NW, c.695-722 AD. Central shrine built by Rājasiṁha (Venkayya 1907, p.230). Note the Iranic vaulted-barrel cupola similar to Sassanian arch at Ctesiphon and the Babylonian-style step-pyramid tower or "Shikara". Longhurst holds that the Kailasanatha temple entrance is the proto-type of all later Gopurams. (Image courtesy Dr. Vandana Sinha, American Institute of Indian Studies, Gurgaon)

One of the gems of Pallava architecture is the Kailashanatha temple, which was also known as Rajasimha-Pallavesvara in ancient times (Venkayya 1907, p.234, footnote 3).

The pyramid-shaped tower or Shikara of the Kailashanatha temple is strangely similar to Babylonian step-pyramids. Babylonia was an integral part of the Parthian empire. While such innovations could have been due to independant innovation, it is more likely that the Pallavas were emulating Babylonian prototypes during the construction of Kailasanatha.

Fig.6: Pancha-ratha Pallava Temple at Mamallapuram, Tamil Nadu. Note the Saka-Buddhist vaulted-barrell cupola on central building. (Image by Stewart Lane Ellington)

The Pancha-ratha Pallava temple at Mamallapuram consists of five temples, one having a Saka-Buddhist cupola, one an Egyptian-style pyramid, and three having ziggurat-shaped roofs reminiscent of Sumer and Babylon (cf. Fig.6). This combination of designs is unlikely to have been independantly invented without external stimulus. These influences could only have come via Iran and the Pallavas, for the Parthians ruled over Assyria and Babylonia.

4.2. Spread of Buddhism
The Pallavas played a major role in propagating the religion of Buddhism. Buddha was known as Sakya-muni, Prakrit for "Lord of the Scythians", and was an Iranian. Thus, there is little surprise when we find Pallavas being the most ardent propagators of Buddhism: "The sect of Buddhism preached in China by Buddha Varman, a Pallava Prince of Kanchi came to be known as Zen Buddhism and it spread later to Japan and other places." (Damodaran 1980, p.70). In other words, Zen Buddhism, like its parent faith of Buddhism, was founded by an Iranian, Buddha Varman.

4.3. Dravidian Shaivism
As noted above, the Pallavas rapidly adopted the indigenous Dravidian religion of Shaivism, and became staunch propagators of the faith. Scores of Shiva temples constructed by the Pallavas remain. While the Pallavas, like the Achaemenids and Parthians, were religiously tolerant, the devotion of some Pallava kings to Shaivism went so far that they went to the extent of demolishing Jain temples:

"According to the Periyapurāṇam, the saint Tirunāvukkaraśar (also called Appar), and elder contemporary of Tiruñānasambandar, was first persecuted and subsequently patronised by a Pallava king who is said to have demolished the Jaina monastery at Pāṭaliputtiram and built a temple of Śiva called Guṇadaravīccaram." (Venkayya 1907, p.235)

By and large, however, the primordial tolerance of Dravidian Shaivism manifested itself, absorbing the other faiths in due course of time.


5- Refutation of Rival Theories on Origin of Parthians

Ayyar has summed up the various non-Parthian theories as follows:

"Thus some scholars considered the Pallavas as of Chōḷa-Nāga origin ², [2. Ind.Ant. Vol. LII, pp.75-80.] indigenous to the southern part of the Peninsula and Ceylon and having nothing to do with Western Indian and Persia, while others placed their original home in the Andhra country between the rivers Kṛishṇā and Gōdāvarī; yet others connected them with the Mahārāshṭra Āryans ³ [3. C.V.Vaidya: History of Mediaeval India, Vol.1, p.281.] and the Imperial Vākāṭakas ⁴ [4. J.B.O.R.S., 1933, p.180ff.]" (Ayyar 1945, p.11)

We now turn to the three theories, namely Chola-Naga, Andhra and Maharashtra Aryan origins.

5.1. Refutation of the Maharashtrian and Vakataka Origin
The surviving sculptures in Tamil Nadu depict Pallavas as tall and dolichocephalic (long-headed) (Fig.3), while the Marathas are short-statured and brachycephalic (round-headed). Moreover, the Pallavas were Shaivites, as opposed to the Maharastrians, who were adherents of the Vaishnavite religion. Further, the Pallavas waged the brutal 100-year Maratha-Tamil war against the Maratha Chalukyas. Had the Pallavas been Maharashtrians, it is unlikely the conflict would have been so prolonged and of such intensity. Thus, the Pallavas were almost certainly not of Maharastrian origin. The slight Maharastrian influence amongst Pallavas is to be attributed to their migration through Maharashtra on their way from Persia to Tamil Nadu.

5.2. Refutation of alleged Vedic Origin
It is sometimes asserted that the Pallavas were of Vedic origin. However, the Vedic and Puranic evidence itself contradicts this view:

"The word Pallava is apparently the Sanskrit form of the tribal name Pahlava or Pahṇava of the Purāṇas. The Pahlavas are described as a northern or north-western tribe¹ (footnote 1: In chapter 9 of the Bhīṣmaparvan of the Mahābhārata, the Pahlavas are mentioned among the barbarians (mlēccha-jātayaḥ)) whose territory lay somewhere between the river Indus and Persia." (Venkayya 1907, p.217)

Furthermore,

"In the Harivaṃśa ⁴ (footnote 4: XIV. verses 15 to 19) the Pahnavas⁵ (footnote 5: In the Rāmāyana (I.55, verse 18) the Pahlavas are said to have emanated from the bellowing of the miraculous cow Nandini, which belonged to the sage Vasiṣṭha.) are said to have been Kṣatriyas originally, but become degraded in later times. They are mentioned here along with the Śakas, Yavanas and Kāmbōjas and their chief characteristic was the beard ⁶ (footnote 6: The beards of the Westerns (ie. the Yavanas), are also mentioned by Kālidāsa in his Raghuvaṁśa, IV, 63) which Sagara permitted them to wear. In the Viṣṇu Purāṇa , the Yavanas, Pahlavas and Kāmbhōjas are said to have been originally Kṣatriya tribes who became degraded by their separation from Brāhmaṇa and their institutions.⁷ (footnote 7: Muir's Sanskrit Texts, Vol.II, p.259, and Ind.Ant. Vol.IV, p.166). In Manu, the Pahlavas are mentioned along with the Puṇḍrakas, Draviḍas, Kāmbōjas, Yavanas, Śakas and other allied tribes. These were all Kṣatriyas originally, but gradually became degraded by their omission of the sacred rites and transgressing the authority of the Brāhmaṇas." (Venkayya 1907, p.217)

Had the Pallavas been of Vedic origin, they would not be cursed in this manner in the Brahmanic scripture. Moreover, the Pallavas did not practice the custom of Vedic human sacrifice (purushamedha or naramedha) and horse sacrifice (asvamedha). Nor did they permit sati (widow-burning) or bride-burning. The Vedic and Brahmanic caste system was also not supported. Also, the Pallavas in their earliest times promoted Prakrit and not Sanskrit. Thus Venkayya notes, "The earliest known records of the Pallavas are three Prākṛt copper-plate charters, viz. (1) the Mayidavōlu plates of Śivaskandavarman, (2) the Hirehaḍagalli plates of the same king and (3) the British Museum plates of Cārudēvi." (Venkayya 1907, p.222) These facts disprove the Vedic origin of the Pallavas.

5.3. Refutation of the Dravidian Origin
That the Pallavas were not Dravidians is evidenced from the fact that their migration can be clearly traced via copper-plate grants as being from the Telugu to the Tamil country. The Pallavas initially promoted Prakrit, which also goes against the proposed Andhra origin of Pallavas. Had they been Andhras, they would no doubt have propagated the proto-Telugu Dravidian dialect.

In further opposition to the Dravidian origin of Pallavas, Venkayya has fittingly asked why the Andhras should have adopted a name which would lead to them being confused with the Pahlavas of Persia.

"Why the indigenous tribe which was formed in the Gōdāvari delta called itself Pallava, a name which would lead to their being mistaken for being Palhavas of Western India is a question which, to my mind, must be satisfactorily answered before the theory of indigenous origin can be accepted." (Venkayya 1907, p.219, footnote 5)

However, the Pallavas rapidly adopted the indigenous Dravidian religion of Shaivism and propagated it, just as the Germanist Lombards accepted the Roman Catholicism of their Latin Italian subjects. That the Pallavas were able to flourish in Dravidia is a testimony to Dravidian tolerance and open-mindedness, a rare characteristic in those days.

The remaining rival theories on the origins of the Pallavas having been undermined, the Parthian origin of the Pallavas remains as the sole logical alternative.


6- Consequences and Conclusion

The Parthian origin of the Pallavas was eagerly adopted by virtually all schools of Dravidologists from the very beginning, Formerly, Indo-European influence in Dravidian had been attributed solely to Sanskrit. Anti-Sanskrit Dravidianists welcomed the Iranic origin of Pallavas as it decreased the Sanskrit proportion in the Indo-European component of Dravidian civilization. Indeed, certain votaries of this school believe that Iranic influence in Dravidian is more important than that of Sanskrit, a view which would no doubt make Iranists proud. Dravidianist evangelists have in their turn used the Pallava example to demand that the Tamil Brahmins adopt Dravidian culture. Their chief argument is that, if the Pallavas from distant Persia could so eagerly adopt Dravidian civilization, then why couldn't the local Tamil Brahmins? Multiculturalist Dravidianists, meanwhile, upheld the Pallavas as an example of ancient Dravidian tolerance and multi-culturalism. The South Indian Brahminist school, which is also largely multiculturalist (often miscalled `secularist') in character, has largely followed this path as well. The political use - and abuse - of history goes on.

The Parthian origin of Pallavas also provides an explanation for the presence of tall, fair-skinned members of non-Brahmin castes in Tamil Nadu and other Dravidian states. Formerly attacked as mixed-caste, part-Brahmin, offspring, it is observed that such persons are at present claiming a Pallava-Parthian origin instead. This is certainly true of certain Cholas, Vellalas and Reddis. Especially in case of those fair individuals who are long-headed, a Pallavite origin is more plausible than a mixed-Brahmin one, for the South Indian Brahmins are generally round-heads. The Parthian theory of the origin of Pallavas has thus helped a large number of people to be rehabilitated in Dravidian society.

It is hoped that Iranists will be inspired by this work to carry out further research on the achievements of the enterprising Pallavas in Dravidia, and bring to light the full scale of Iranic influence in Dravidian civilization.

Appendix: Extracts from "A New Link between the Indo-Parthians and the Pallavas of Kanchi"

As Ayyar's 1945 paper is difficult to obtain both in India and abroad, I am reproducing extracts below for reference purposes.

Journal of Indian History	J.Ind.Hist. Vol.XXIV, Parts 1 & 2, April & August 1945, Serial Nos. 70 & 71, p.11-16; Ananda Press, Madras.
"A New Link between the Indo-Parthians and the Pallavas of Kanchi" * By: V.Venkatasubba Ayyar, Ootacamund "Though Archaeology, Numismatics and Epigraphy, each by itself, are great assets to the Historian, they rarely combine to assist him in any intricate problem of history. Sometimes the evidence adduced by these remain inexplicable when considered separately, but they gain a new meaning and assume fresh importance when collated together. Such may be said to be the case of some new evidence that is now advanced on what may be called the `Indo-Parthian Origin' of the Pallavas of Kanchi. It was my father the late Rai Bahadur Venkayya who first traced the origin of the Pallavas of Kāñchī to the Pahlavas mentioned in the Mahābhārata and the Purāṇas where they are classified as foreigners outside the pale of Aryan society 1. He thus postulated that they were a northern tribe of Parthian origin with their original home in Iran.This theory was first accepted by scholars, but discarded by later writers as resting wholly on doubtful philological resemblance of the words Pahlava and Pallava. Several arguments against the `foreign origin' were cited, such as the absence of any reference to Pallava migration in copper-plate grants, the more possible identity of the Pallavas with the indigenous Toṇḍaiyar and the Kāḍavar, and the testimony of the poet Rājaśēkhara of about the 10th century AD who refers to two distinct Pallava kingdoms, one in the south and the other in the north-west. Thus some scholars considered the Pallavas as of Chōḷa-Nāga origin 2, indigenous to the southern part of the Peninsula and Ceylon and having nothing to do with Western Indian and Persia, while others placed their original home in the Andhra country between the rivers Kṛishṇā and Gōdāvarī; yet others connected them with the Mahārāshṭra Āryans 3 and the Imperial Vākāṭakas 4 * A paper sent to the Eleventh All-India Oriental Conference, Hyderabad (1941) 1. Archl Sur. Rep. for 1906-7, pp.217ff. 2. Ind.Ant. Vol. LII, pp.75-80. 3. C.V.Vaidya: History of Mediaeval India, Vol.1, p.281. 4. J.B.O.R.S., 1933, p.180ff.
p.12 The question of the origin of the Pahlavas may, therefore be said not to have yet been satisfactorily settled. Opinion seems to have now swung back, and recently two authors 5, after examining the arguments for the `indigenous theory', were inclined to advocate the `foreign origin' first propounded by Venkayya. A very important evidence that is helpful in settling this question is now available. It is found in one of the explanatory labels to the sculptures decorating the walls of the verandah round the central shrine of the Vaikuṇṭha-Perumāḷ temple 6 at Conjeevaram. This unique and valuable epigraph narrates how on the death of Paramēśvara-Pōttaraiyar of the Pallava family without any issue, a deputation of ministers waited on Hiraṇyavarman of a collateral line and requested him to grant themn a ruler for the vacant throne. Hiraṇyavarman thereupon consulted his nobles (kullamallar) and then his four sons, and enquired who among them would accept the sovereignty. All of them declined the offer except the youngest prince, Paramēśvaravarman, aged twelve years. Thereupon the deputation offered, probably to Hiraṇyavarman as the chief of the family, the makuṭa resembling an Elephant's Head which they had brought for the sovereign-elect. The passage 7 reads: [ ... ] (This is) the scene where Hiraṇyavarmma-Mahārāja was seized with fear 8 (on) hearing Taṛaṇḍikoṇḍapōśar say `Hear (what thy) elderly servants submit. This (ie. the object brought) is not an elephant's head, but (only) they son's makuṭa'. 5. Mr. K.R.Subrahmanyam in his `Buddhist Remains in Āndhra', p.73ff and Mr. P.T.S. Iyengar in his `History of the Tamils', p.329. 6. These Historical sculptures form the subject of a Memoir (no.63) by the late Dr. Minakshi issued recently by the Arch. Sur. of India. 7. The inscription is published in S.I.I. Vol.IV, No.135 but the reading there is not quite reliable. A revised reading is given by the late Mr. A.S.Ramanatha Ayyar in Memoir No.63 noticed above. 8. Ramanatha Ayyar (Memoir, No.63) reads here `biti-[vi]du' in the sense `abandoned fear' but Paḍu is clear on the stone. My colleague Mr. G.V.Srinivasa Rao also favours the reading paḍu as the fear is at the thought of the son's impending departure and separation (pirivin santapam) mentioned later on and not really at seeing the makuṭa resembling the elephant's head. Dr. Minakshi has missed this implication and ascribes the fear either to Hiraṇyavarma's ignorance of court customs or to his old age and failure of vision.
p.13 The passage previous to the one just cited is damaged, but it also refers to `the elephant's head' brought by the deputation. These two passages clearly indicate that the makuṭa presented to the king-elect was really shaped like an elephant's scalp. Appropriately enough in the sculptural representation 9 above this label can be seen standing three persons, of whom one in the centre is carrying an object like the elephant's scalp. This object must be the crown that Nandivarman had to wear on ceremonial occasions. Excepting this single sculptural representation of the elephant's scalp and the epigraphical explanation of it just cited, no other reference to such a head-dress has so far been found in Pallava sculptures and in fact, in the whole range of Indian Art. But a study of the Greek coins of the successors of Alexander the Great throws light on this custom. The elephant's scalp 10 as a motif of head-dress is found for Alexander in the early coins of his governors, Ptolemy I of Egypt and Seleucus of Babylon. To the numismatists this headgear was a puzzle and they explained it variously, as representing the conquest by Alexander of India, the land of the elephant par excellence, as a mere symbol of power, as a mark of deification, as a mint-mark specially referring to the elephant-god of Kāpiśa 11 etc. Alexander 12 did not adopt the emblem of elephant's scalp himself on his coins, but this symbol served as the iconographical expression of the monarchical principle to some of his successors. When after the death of their master 13, his generals Ptolemy I and Seleucus, established themselves as kings, the former in Egypt and the 9. This panel is very much damaged. 10. The use of scalp as a device on coins was first started in the island of Samos belonging to the Ionian Greeks. Demoteles or his successor at Samos before the end of the 7th century caused to be struck the first official coins of Samos with the lion's scalp. This was adopted as the chief Samian device and it continued to decorate the city's coinage until she became merged in the Roman Province of Asia (Greek Coins: Seltman, p.31). But the elephant itself was used as a device in the coins of Antimachos Theos, Heliokles, Lysias, Antialkidas, Archebios, Apollodotus, Soter, Menander, Zoilos, Maues, Azes, Ayileses and Zeioinises (Ind. Hist. Quart., Vol. XIV, p.301) 11. The tutelary deity of the city of Kāpiśa is supposed to be Indra accompanied by an elephant (Ind. Hist. Quart., Vol.XIV, p.299). The kingdom of Kāpiśa formed the connecting link between Bactria and India. See Greeks in Bactria and India: W.W.Tarn, p.138. 12. In early times it was deemed sacrilegous to put the portrait of a human being on coins. But Alexander introduced his portrait on his issues in the guise of Zeus or Heracles and the figure can be recognised on coins with absolute certainty. After his death this figure came to be used as a type on coins and he was even raised to the rank of divinity. 13. Coins were issued in Alexander's name long after his death. Such types were minted by many cities in Asia Minor and they continued to be struck long after even his empire had crumbled into small states. In fact, Alexander's coinage was among the most lasting of his institutions.
p.14 latter in Babylon, they issued their early coins with the figure of Alexander wearing the elephant's scalp. Rao Bahadur K.N.Dikshit has noticed a coin of Andragoras a satrap of Parthia under Alexander the Great, bearing on the obverse the head of Alexander as on the coins of Ptolemy I of Egypt 14. Agathocles the tyrant of Syracuse who concluded an alliance with Ptolemy I issued a similar type of coin with this head-gear, in Africa 15. This symbol was also adopted by Antiochus IV 16, Alexander II 17 and Lysias 18, but the best and most artistic specimen of the elephant-scalp type of coin is that of the Bactrian king Demetrius II who re-conquered the countries of the Indus valley which had been occupied by Alexander the Great, but subsequently surrendered by his successor Seleucus I to Chandragupta. Demetrius, known as the `first king of Bactria and of India' is here represented with a helmet resembling an elephant's head complete with proboscis and tusk, and it is surmised that he was conspicuously imitating Alexander whom he regarded as his ancestor and ideal. The tradition behind the adoption of this symbol is not clear. Some of the kings like Potlemy who used it had no connection with India, and Seleucus had even bartered his Indian Province for 500 war-elephants. It is, therefore, supposed that the use of this symbol for Alexander represented the utmost extent of Power 19, for both Ptolemy I and Seleucus who first adopted it `had every object in representing themselves as the successors of the man who had reached the summit of human greatness' 20. It is however just possible that this motif is reminiscent of Alexander's connection with India. The coincidence in the use of this peculiar head-dress by the successors of Alexander in the centuries BC and by the Pallava ruler Nandivarman in the 8th century AD is of more than ordinary interest. In the Vaikuṇṭha-Perumāḷ temple inscription mentioned above, Taṛṇḍikoṇḍa-Pōśar, the vṛiddhāgāmikar (ie. aged soothsayer), is stated to have prophesied that Nandivarman would become a Chakravartti ie. the king of kings 21 (ivaṇ Chakravartti āvāṇ). The consecration (abhishēka) ceremony 14. Ind. Ant. Vol.XLVIII, pp.120-121. 15. Historical Greek Coins: G.F.Hill, No.65. 16. The Greeks in Bactria and India: Tarn, p.189. 17. The Seleucid Kings of Syria: P.Gardner, p.26, No.61. 18. Greek and Scythic Kings of Bactria and India: P.Gardner, Pl.X, 6 and Indo-Greek Coins: R.B. Whitehead, Vol.I, p.30. 19. In the royal consecration called Rājasūya, the king has to step on a tiger skin as a symbol of acquisition of power. 20. The Greeks in Bactria and India: W.W.Tarn, p.131. 21. In this connection the adoption of the imperial title `king of kings' by the Śaka and Pahlava suzerains is worthy of notice.
p.15 of this king is stated to have been immediately celebrated when he was still an unmarried 22 prince of only twelve years 23. It is not, however, clear from the inscription whether this ceremony was in the nature of mere nomination to the throne or of actual coronation (muḍi-śūṭṭudal). It is worthy of note that in the sculptures of the Vaikuṇṭha-Perumāḷ temple, the elephant's scalp is not found as a head-dress of the Pallava kings. Its use by Nandivarman alone may be said to signify his elevation to regal power, as in the case of Ptolemy I and Seleucus who first adopted this symbol on their coins when they established themselves as kings from their position as Viceroys, after the death of their master Alexander the Great. Ernest Hersfeld observes in connection with Kushano-Sasanian coins, that the helmet surmounted by an animal's head such as that of a lion, a horse, an eagle, etc. is the exclusive emblem of the members of the royal family next to the throne.24 The practice of wearing this elephant's scalp observed in common may therefore be considered as establishing a strong link between the Pahlavas and the Pallavas. This headgear has not so far been found among other Hindu rulers of India. Why, then should the Pallavas alone adopt it? The early Pallavas are not celebrated in Tamil literature, they are not classed among the Tamil speaking people and they are also not known to have had any matrimonial connection with the Tamil kings. Their culture was alien to the land of their settlement. The evidence now adduced is thus valuable as emphasising this point and indicating their original habitat beyond the borders of India. This evidence may be said to gain in importance when considered with other circumstances such as the similarity of the name Pallava to the form Pahlava 25, the reference to the rule of the Pahlava governor named Suviśāka in Ānarta (ie. the district round the modern Dvāraka) and Surāshṭra 26, the tradition of marriage alliance with the Nāgas common to both the Indo-Scythians and the Pallavas, the reference made by Ptolemy the Geographer to the Parthian princes as constantly changing their abode by driving each other 22. The Rājyābhishēka ceremony, as laid down in the Śāstras, required the presence of the chief queen. Cf. Harsha of Kanaouj; it is stated that he was not married when he succeeded to the throne and that his coronation was postponed on this account (IHQ, Vol.XII, p.142). 23. Dr. Jayaswal has tried to show that the minimum age for coronation of a king must be 25 years and he cites instances of the coronations of Asoka and Kharavela who had to pass a considerable period after the demise of their predecessors before they could ascend the throne (Hindu Polity, p.52; Ind. Hist. Quart., Vol. XII, p.142). 24. Memoir of the Arch. Sur. of India, No.38, p.21. 25. History of Indian Literature: Weber, p.188, note 201. 26. Ep.Ind., Vol. VIII, p.41.
p.16 out and the possibility of tracing the different stages of the Pahlava migration through Kathiawad, Mālwa, United Provinces 27, Dhanakada, and finally to Kāñchī on the east coast. By the time the Pahlavas settled down at Kāñchī they were so Hinduised and merged into the society of their new home that they came to be regarded as Kshatriyas 28 belonging to the Bhāradwājagotra, observing Aśvamēdha and other rituals of the Hindus. The new link now furnished thus strengthens the statement made by Venkayya that `the Pallavas of Kāñchipuram must have come originally from Persia, though the interval of time which must have elapsed since they left Persia must be several centuries' 29." 27. Dr. Fleet finds in the Pahlādpūr inscription of Śiśupala a possible reference to the Pallavas of Northern India: Gupta Ins. Fleet, p.250. 28. In the Harivaṃśa and the Vishṇupurāṇa, the Pahlavas are classed as Kshatriyas. 29. Arch. Sur. Rep. for 1906-7, p.219.

Acknowledgements
The author would like to thank Prof. Shireen Moosavi and Prof. Irfan Habib (Aligarh) for their kind assistance with references. The author is also very grateful to Prof. P. Oktor Skjærvø and Prof. Michael Witzel (Harvard) for kindly sending important research material. Many thanks to Fatema Soudavar Farmanfarmaian for fruitful discussions, and to The Iranian for publishing this paper.

The author gratefully thanks Michael D. Gunther, http://www.art-and-archaeology.com; Dr. Vandana Sinha, American Institute of Indian Studies, Gurgaon, http://www.indiastudies.org; and Stewart Lane Ellington, http://stewellington.com for permission to reproduce their wonderful images in this paper.

References:

Ayyar 1945: "A New Link between the Indo-Parthians and the Pallavas of Kanchi" by V. Venkatasubba Ayyar, Ootacamund, Journal of Indian History, Vol.XXIV, Parts 1 & 2 (April & August 1945) Serial Nos. 70 & 71, p.11-16; Ananda Press, Madras.

Damodaran 1980: "Contribution of the Tamils to World Culture", G.R.Damodaran, J.Tamil Studies, vol.18 (Dec. 1980) pp.69-76.

Derakhshani 1999: "Die Arier in den nahÎstlichen Quellen des 3. und 2. Jahrtausends v.Chr." by Jahanshah Derakhshani, International Publications of Iranian Studies, http://www.int-pub-iran.com, 2.Auflage, 1999; ISBN 964-90368-6-5.

Holberton 19??: "The World of Architecture" Paul Holberton, WHSmith, Michael Beazley Publishers, 14-15 Manette St, London W1V 5LB.

Minakshi 1977: "Administration and Social Life under the Pallavas", by Dr. G.Minakshi, University of Madras, Madras, 1977.

Nair 1977: "The Problem of Dravidian Origins - A Linguistic, Anthropological and Archaeological Approach", by T.Balakrishnan Nair, University of Madras, Madras, 1977.

Skjærvø 1995: "The Avesta as source for the early history of the Iranians", P.Oktor Skjærvø, in `The Indo-Aryans of Ancient South Asia', ed. George Erdosy, Walter de Gruyter, Berlin 1995, pp.155-176.

Spooner 1915: "The Zoroastrian Period of Indian History" by D.B.Spooner, J.of the Royal Asiatic Society of Great Britain and Ireland, 1915, p.64-89 (Pt.I); p.405-455 (Pt.II).

Suthanthiran 1989: "Evolution of Gopura in Temple Architecture of Tamil Nadu", by A. Veluswamy Suthanthiran, J.Tamil Studies, vol.35 (June 1989) pp.28-38.

Venkayya 1907: "Annual Report 1906-7", Archaeological Survey of India, "The Pallavas", by V.Venkayya, p.217-243; reprint Swati Publications, Delhi.

Waddell 1929: "The Makers of Civilization in Race and History", by L.A. Waddell, 1929, reprint S.Chand & Company, P.O.Box No. 5733, Ram Nagar, 7361, New Delhi-110055, 1986, schand@vsnl.com, http://www.schandgroup.com.

Witzel 2001: "Autochthonous Aryans? The Evidence from Old Indian and Iranian Texts," by Michael Witzel, El.J. of Vedic Studies 7-3 (2001), p.1-115 (http://www.people.fas.harvard.edu/~witzel/).