i. The Origins of the Aparaśikhā Brahmans, Their Śrauta Traditions and Their Arrival in the Tamil Country
The outwardly distinguishing feature of the Aparaśikhā Brahmans, corresponding to the pūrvaśikhā of the Pūrvaśikhā Brahmans, is their aparaśikhā, “pin kuṭumi,” or back tuft in Tamil, as opposed to the “mun kuṭumi” or front tuft of the Pūrvaśikhās (Illustration 2; the illustration is a painting in the Panjab Hills school of the 16th century, precisely the area to which we will trace the Aparaśikhā group below.) Indeed, the aparaśikhā style is the ubiquitous mode now, in all of India, so much so that kuṭumi neutrally signifies the aparaśikhā mode, although in Sangam period, it did the pūrvaśikhā. As we saw, the ‘poetic code’ surrounding the representation of the kuṭumi in the Sangam poems clearly excludes the aparaśikhā mode.
Unlike the case with the Pūrvaśikhā Brahmans, everything about the Aparaśikhā Brahmans is grounded in recorded history: their arrival in the Tamil country is one of the most meticulously recorded movements of human groups in history, especially considering its time span, ca. from 4th to 14th centuries CE. Like the Pūrvaśikhā Brahmans, they too brought with them live, if semi-literate, śrauta traditions to peninsular India. Their Veda śākhās fall into the following groups:78
- i. Ṛgveda: Only the Śākhala śākhā of the Ṛgveda and its Āśvalāyana tradition are known among the Aparaśikhā Brahmans. The Kauṣītaki tradition of the Ṛgveda, the mainstay of the Pūrvaśikhā śruatism, is entirely unknown among them.
- ii Yajurveda: Both the Kṛṣṇa and the Śukla Yajurveda śākhās, the latter both its Kāṇviya and Mādhyandina recensions, are attested among the Aparaśikhā Brahmans, although a śrauta tradition has not survived along the Śukla Yajurveda matrix. The Kṛṣṇa Yajurveda is entirely of the Taittirīya śākhā, attested in four schools, a minority Baudhāyana school and the prepossessingly dominant Āpastaṃba school and its two sister traditions, the Bhāradvāja and Hiraṇyakeśi (aka Śāṭyāṣāḍha) schools.
- iii. Sāmaveda: Only the Kauthuma śākhā of the Sāmaveda is attested among the Aparaśikhā Brahmans, in its Drāhyāyana school.
Of the above, the Śukla Yajurveda occurs only among the Aparaśikhā Brahmans in the Tamil country.79 Likewise, the Āpastaṃba (along with the nearly identical Bhāradvāja and Hiraṇyakeśi) and the Drāhyāyana traditions also occur only among the Aparaśikhā 50 Brahmans: these signify thus positive control with respect to the Aparaśikhā Brahmans in epigraphy and fieldwork, just as the Kauṣītaki Ṛgveda, Vādhūla/Āgniveśya Yajurveda and Jaiminīya Sāmaveda do for the Pūrvaśikhās. And as with the Pūrvaśikhā term “paviḻiya” for the bahuvṛca tradition and ś/jāṃbavya for a branch of the Kauṣītaki tradition, the term “pravacana” for the Baudhāyana tradition seems to be an exclusive Aparaśikhā usage, in epigraphy (see below).
As a Śukla Yajurveda Śrauta tradition is not extant among the Aparaśikhā Brahmans, the following four Śrauta matrices are possible among them:
- i. Āśvalāyana Ṛgveda-Baudhāyana Yajurveda-Drāhyāyana Sāmaveda
- ii. Āśvalāyana Ṛgveda-Āpastaṃba Yajurveda-Drāhyāyana Sāmaveda
- iii. Āśvalāyana Ṛgveda-Hiraṇyakeśi Yajurveda-Drāhyāyana Sāmaveda
- iv. Āśvalāyana Ṛgveda-Bhāradvāja Yajurveda-Drāhyāyana Sāmaveda
The second axis seems to be the near universal tradition extant among the Aparaśikhā Brahmans, mostly in the agrahārams along the Kaveri river from Tiruchirapalli to Tanjavur and onward to Kumbakonam.80 Key epigraphic records, as we will see below, show that at least 70% of the Aparaśikhā Brahmans belong to the Āpastaṃba tradition, the Yajurvedis as a whole forming possibly upto 90 per cent, and they are the subjects of the Pallava-Cōḷa land grants in the villages along the Kaveri river. This striking statistic helps us trace the Aparaśikhā group to the Mathurā regions of the Yamunā River, to which the Āpastaṃba tradition has been localized.81 The region would extend to the Hariyana area in the northwest (Map V) to the old Kuru area in the north with its Kauthuma Sāmaveda, the Malva territory in the south and southeast. A name that appears frequently in the Aparaśikhā epigraphic records is Daśapuriyan, after the Malva city Daśapuri (also known as Mandasor). The Aparaśikhā emigration seems to coincide in the main with the fall of the entire region first to the Huns (5th-6th centuries CE) and the Muslims later, with widespread dispersion of the Brahmans of the area, including the Daśapuri Brahmans.
Like the Pūrvaśikhā group, the Aparaśikhās also fall into several internal divisions, not endogamous with one another till recent times and even today not fully so. We know that this division goes back to the time—and place—of migration. Its first attestation comes to us from the famous family history of Rāmānuja. His family was of the “vaṭama” division, his preceptor’s that of “bṛhatcaraṇam” (as it happens, the two principal and largest groups of the Aparaśikhā Brahmans) forcing, as we noted above, Rāmānuja’s wife into a conduct unbecoming toward his guru and embarrassing personally to him.82 That is, these divisions existed among the Aparaśikhā Brahmans before their arrival in the Tamil country and they arrived as strangers, despite adherence to common Veda śākhās. We will see that the Vaiṣṇava group, when it begins to emerge as a separate group within Ramanuja’s life time, is made up almost entirely from the Aparaśikhā group, all of the vaṭakalai group and 85% of the tenkalai, the balance of 15% made up of the Pūrvaśikhā group, the Brahman element of the founders of Āḻvār- Vaiṣṇavism.
I list here from Thurston (1909) the names of these divisions, from the most numerous to the least as determined in my field work:83 i. the vaṭama; ii. the bṛhatcaraṇam; iii. the aṣṭasahasram; iv. the vāttima; v. the prathamasākī. The first four are all Taittirīya adherents, mostly its Āpastaṃba Sūtra; the last is made up entirely of adherents of the Śukla Yajurveda in both its recensions, the Kāṇva and the Mādhyandina.
C. ii. The Pallava Period Epigraphy and the Aparaśikhā Brahmans
As I noted above, the first Aparaśikhā Brahman we can positively identify as one may well be Jyeṣṭa Śarman of the Gautama gotra and group-specific Āpastaṃba Sūtra of the Vēsantha (Jalapuram) Copper Plates of the Pallava King Simhavarman II, issued in his 19th Regnal Year, in the 5th century CE, granting the village of Vēsantha to Jyeṣṭa Śarman (Mahalingam 1983: 52-54; Item 7).84 The royal order is issued from Kanchipuram (not perhaps the extant city of that name in the Tonṭaimaṇṭalam area of the Tamil country85) to the “villagers of Vēsantha in Nādattapādi and to the Mahāmātras, Adhyakṣas, Rājapuruṣas, and Cancarantas,” the oral order recorded by “Kulippoṭṭar, a Rahasyādhikṛt”. The village lay still in the present Guntur district of southern Andhra Pradesh, the northern reaches of Toṇṭaimaṇṭalam, in the east coast area between the Pennār River in the north and the Pennaiyār river: this will include as Frasca (Map VI after Frasca 1990: 3; Map 2) shows well-known centers like the state capital, the city of Madras (also known as Chennai), Kanchipuram in the south and Tirupati and Nellur in the north, the whole area containing islands of both Tamil and Telugu communities even today.86 We already face here the Vēnkata hills, the northern boundary of the Tamil country as recognized in the Sangam poems. The Kaveri delta lies still to the south, the eventual destination of many of the Aparaśikhā Brahmans. Nāthamuni, the redactor of the Āḻvār hymns, was born in Viṣṇunārāyaṇapuram ca. 11th century CE and Rāmānuja, in the 13th century CE in Śrīperubendūr, three generations later in the same family lines, both with signature Aparaśikhā Vedic affiliations, both still in the Toṇṭaimaṇṭalam region.
For the Pallava period, we have data for some 467 Brahmans arriving into the Tamil country, in 20 Copper Plate deeds that have survived, ranging from single families as with Jyeṣṭa Śarman above, to 308 families of the Taṇṭamtōṭṭam Plates of Nandivarman II, dated to his Regnal Year 33, 765-6 CE, with 108 families becoming interim the standard complement in a grāmadeya. The happenchance discovery of the original Copper plates,87 mostly unearthed by farmers tilling the land, suggests that the discovered deeds constitute only a fraction of the total, as suggested by Burton Stein for the later Cōḷa period.88 Of the 467 families, the Veda śākhās of 442 families are recorded in the plates. The Veda śākhā breakdown of these immigrants is given in Table I:
- Āpastaṃba 274
- Hiraṇyakeśi 18
- Bhāradvāja 1
- Pravacana 101
- Āśvalāyana 7
- Candogā 23
- Kātyāyana 8
- *Agniveśya 2
- *Paviḷiya 4
- *Jaiminīya 1
- Kalarakha 2
- Kaṭu 1
- Goduma 2
Table I: The Veda Śākhās of Pallava Aparaśikhās
The followers of the Āpastaṃba tradition constitute 62% of the total, the number increasing to 66% if we include the closely related Hiraṇyakeśi and Bhāradvāja Sūtra adherents, reaching 90% (including the prarvacana adherents) for the Yajurvedis as a whole, giving rise eventually to the adage that every “house cat” in South India, as Witzel notes (1995:335), can recite the Taittirīya Saṃhita. The backbone of the Aparaśikhā Brahman group takes shape in this period, constituting close to 95% by the modern period of the Tamil country, eventually coming to define the rubric “Tamil Brahman” for the area.89 We do not know what Veda Sūtras the Kaḷarśa (also Kaḷarakha), Goduma (also Godu), and Kaṭu signified.
The three starred items in Table 1 belong to the Pūrvaśikhā group, represented by five families, identifiably so from their Veda śākhās (Jaiminīya, Āgniveśya and “paviḻiya,” a corrupt form of Bahuvṛca but part of the Pūrvaśikha argot) although there may have been some Pūrvaśikhā families in the Āśvalāyana group, the Baudhāyana group excluding itself out, however, being all “pravacana,” the Aparaśikhā term for the Baudhāyana tradition.90
The “paviḻiya” term for the bahuvṛca appellation is of exceptional interest: today,
as noted above, it occurs only among the Nambudiri Pūrvaśikhās, yet the four paviḻiya
families, at least two of them, are shown coming from the villages in southern Andhra
Pradesh (#23, Vaḍuga Śarma of Kāśyapa gotra from Nimbēi and #134, Dāmodarabhaṭṭa
of Garga gotra from Vaṅgippāṛu), both in the Toṇṭaimanṭalam area extending northward
into southern Andhra Pradesh, suggesting that the Pūrvaśikhās were present in areas
beyond the traditional boundaries of the Tamil country during the early era of the Pallava
regime. It is possible as well that the families were Śōḻiya Pūrvaśikhās, who regularly
share with the Nambudiris several rare Veda śākhās. The term occurs, designating a
Veda śākhā at NDP: 1611-12:
Candōgā! Pauḻiyā! Taittirīyā! Cāmavēdiyinē! neṭumālē
Anto! ninnaṭiyanṛimaṟṟiyēn aḻuntūrmēlticainin ammānē
It is of interest in the above that there are two terms for the Sāmaveda: one Candogā, the Aparaśikhā śākhā, beginning to be known in the Tamil county among the Brahmans coming under the Pallava grāmadeya system and the other neutral Sāmaveda, possibly designating the Jaiminīya śākhā of the Pūrvaśikhā Brahmans, being the Veda śākhā of Maturakavi, one of the four Brahman Āḻvārs and the figure supplying the corpus Nālāyiradivyaprabhadam to the Aparaśikhā Nāthamuni in the Vaiṣṇava tradition.
We must also note that the Pūrvaśikhā presence in the Pallava epigraphy is practically non-existent, seven families of the total of 467, showing that they were not part of the grāmadeya deeds, near autochthons by now in the Tamil country; it also marks the relative eclipse of the group in the Tamil country, being reduced, as noted above, to a small minority eventually. On the other hand, the epigraphy also shows that the Aparaśikhā Brahmans are the group sponsored primarily and brought in by the Pallavas. It would seem that the Pallavas adopted in return the Bhāradvāja gotra, the predominant gotra of the Aparaśikhās, regularly attested upto 30% in some gotra samples I have studied, leading to the Tamil saying, “half of Brahmans are Bhāradvājas” (“pāppānil pāti pāratvācam”).91
A line (l. 198) in the Tanṭantōṭṭam Plates noted above reads: pārataṃ vā[ci]ppānukku ppaṅgonrum (“one share for the Bhārata reader”)—in 789 CE. What recension was read by this person in the temples? We do not know. Our hypothesis is that the *Pūrvaśikhā text is in existence in the Tamil country at this time, as our Σ-text facing the Sukthankar-σ-text. Did the epic, corresponding to the Sukthankar-σ text, come with the Aparaśikhā Brahmans? I list below what would be a “learning quotient” of the Aparaśikhā Brahmans from the Pallava grāmadeya deeds:
- Dvivedi 1
- Trivedi 41
- Caturvedi 129
- Kramavittan 36
- Ṣaḍaṅgavit 40
- Somayājis 23
- Vasantayāji 1
- Sarvakratuyāji 3
- Vājapeyi 1
Table II: The Aparaśikhā Vedic Titles
It is true that titles, especially ones like caturvedi, are not always, as Louis Renou noted, to be taken literally.92 Nevertheless, we have here (as with the Pūrvaśikhā Brahmans earlier) a fairly elite group moving from one part of the country to another, and it stands to reason that a Mahābhārata traveled with them, most likely, by the 8th century CE or later, a vulgate Northern Recension text. The Aparaśikhā migration was to continue in this fashion to well-nigh pre-modern times, the last deeds of the brahmadeyas occurring in the Nāyaka period,93 giving us the veritable modern Tamil Brahman. Yet the *Pūrvaśikhā text resident with the Śōḻiya Pūrvaśikhās in the Tamil country as the Σ-text dictates the terms of reaction between it and the in-coming, Sukthankar σ-text of the Aparaśikhā Northern Recension. It seems improbable at first consideration, but as noted, it accords, on the other hand, perfectly well with the development of the texts of the emerging Vaiṣṇava movement. As we have already noted, the founding Āḻvār text, the NDP, begins its career, in part, with the Śōḻiya Pūrvaśikhā’s Bhakti compositions in Tamil, depending upon the Mahabharata Σ-text, specifically its Harivaṃśa, for its Kṛṣṇāism: it is these texts that are collected by the Aparaśikhā immigrant, Nāthamuni, with the north Indian name Miśra still common in his circles, and fashioned into the founding text of Śrīvaiṣṇavism (see below). We do not have a similarly concrete narrative as regards the interaction between northern and southern strands in the case of the Mahābhāratha. That is to say, we do not have a Nāthamuni-like figure orchestrating the formation of the Tamil (Grantha)-Telugu version of the epic. However, it would seem that the *Pūrvaśikhā text of the Śōḻiya Pūrvaśikhās functioned like the Āḻvār compositions, providing the basis for the emerging Tamil (Grantha)-Telugu versions of the Southern Recension, most likely, as I argue below, in the Tanjavur Nāyaka courts.
C. iii. The Cōḷa Period Epigraphy and the Aparaśikhā Brahmans
When the Pallava imperium comes to an end in the first decades of the 10th century CE—we need to remind ourselves (Mahalingam 1983: xxvii) that it began almost with the Guptas, in the early 4th century, outlasting them by two centuries, indeed reaching its apex with the long rule of Nandivarman II from 731 to 792, well after the decline of the Gupta period in the north—the system of the grāmadeya passes on seamlessly to the Cōḷa empire. The story that Burton Stein (1968; 1982) tells of the Brahman alliance with the land-owning Vēḷḷāḷa group under the local, segmentary control of the Cōḷa rule is essentially that of the Aparaśikhā Brahmans, and he estimates that there were some 300 grāmadeya deeds in the Cōḷa period—with the rider about this being a fraction of the original number. Champakalakshmi (2001) shows that the Cōḷa brahmadeya system builds on the Pallava practice by designating certain brhmadeya units as tankūṛu (taniyūr) as “separate unit[s] of political-economic significance from the early tenth century [CE]” (65), a total of 22 such “rural-urban continuums” attested so far in the Cōḷa realm.
All the same, it has not been noticed how strikingly similar the practice of the Cōḷa period (ca. 900-1350 CE) is to that of the Pallava period:94 essentially the same infrastructure supervises the same Aparaśikhā Brahmans, most, followers of the Āpastaṃba Sūtra, entering the Tamil country from an immediate domicile in southern Andhra Pradesh, and many more Daśapuriyans. Besides, the epigraphy clearly shows a gradual increase in the numbers per deed of the Aparaśikhā Brahmans arriving in the Tamil country, the earliest Pallava deeds being brahmadeya, in which the recipients of the royal bounty are single families and later ones grāmadeya, in which a number of families, eventually becoming 108, come to be settled in a village with various privileges stipulated relating to taxes, water rights, access paths and other such matters—the entire process achieving a greater level of sophistication and organization in the Cōḷa institution of tankūṛu. And because their śrauta traditions place them in the Mathurā region in north- central India, covering areas in the north in Hariyana and eastern Panjab, western Rajastan and the entire Malva region in the south and east we can say that the era of the Aparaśikhā migration begins with the arrival of the Huns in northwest South Asia (5th century CE) and continues un-interrupted with the Islamic conquests. To be sure, in the grāmadeya deeds, these Brahmans are also immediately from their domiciles in southern Andhra Pradesh, but originating eventually in the northwest, in the Malva country and its immediate northwest, the Eastern Panjab, the original Āpastaṃba home. Not many of the Cōḷa Copper Plates have come to light yet, but one spectacular find gives us three times the data of the entire Pallava epigraphy, the Karandai plates, weighing in at nearly 250 pounds of copper and miraculously unearthed in a field in the village of Puttūr in Papanāśam Taluk and Tanjavur District ca. 1920’s.95 Planned as a grand grant to 1080 families by Rajendra I, the entire process lasting almost two years, 1019-1021 CE, Tribhuvanamahādāvic-caturvēdimaṅgalam, named for the king’s mother, was made up from some 52 villages, covering a total area of 20,305 acres, almost the entire southern part of today’s Papanasam Taluk in the south and extending to the Mannargudi Taluk in the northeast of the Tanjavur district. I give below the Veda śākhā distributions of the Brahmans of the Karandai Plates in Table III below: 60
- Āpastaṃba 615
- Hiraṇyakeśi 42
- Bhāradvāja 11
- Āgastya 29
- Baudhāyana 54
- *Āgniveśya 2
- Āśvalāyana 154
- *Ś/Jāmbavya 4
- Drāhyāyana 77
- *Jaiminīya 41
- Kātyāyana 50
Table III: The Veda Śākhās of the Karandai Plates Brahmans
Essentially this is the Aparaśikhā profile of the Pallava plates above. The adherents of the Āpastaṃba Sūtra and related Sūtra traditions amount to 62% of the total, almost the same ratio as with the Pallava grant. “Pravacana,” the Aparaśikhā term for the Baudhāyana Sūtra, is absent here: the 54 Baudhāyana families could thus be from either Aparaśikhā or Pūrvaśikhā group, as is the case with the 154 Āśvalāyana families. We encounter a significant number of Vājanaseyi adherents, following the Kāṇva recension of the white Yajurveda as well, 50, many of them carrying the title kramavittan—trained to recite the birth Veda upto the krama vikṛti level. The Agastya Sūtra designates a Yajurveda tradition and seems to be confined to the Aparaśikhā Brahmans96
On the other hand, the starred items are signature Pūrvaśikhā sūtras: 47 families of the 1080, all moving from the western parts of the Tamil country to the eastern parts.
Here we note a new Pūrvaśikhā Veda sūtra, the Ś/Jāṃbavya, a close branch of the Kauṣītaki Ṛgveda tradition (Oldenberg 1884; Gonda 1977: 606)97 with four followers. Considering the date of the Karandai Plates (1029-31 CE), it is most probable that the Jāṃbavya Sūtra of the Ṛgveda would be found among the Tamil Pūrvaśikhās, almost certainly among the Śōḻiya Brahmans.98
We should note as well that the Pūrvaśikhā Brahmans of the Karandai Plates are not domiciled in the villages of the southern Andhra Pradesh: it will be recalled that the four “paviḻiya” adherents of the Pallava Plates, almost three centuries earlier, were from the Toṇṭaimaṇṭalam area. The bulk of the Aparaśikhā Brahmans of the Plates, as Krishnan notes with emphasis, are also from this area.99 On the other hand, the 41 Jaiminīya Brahmans of the Karandai Plates—the Brahmans that we can unambiguously identify as Śōḻiya Brahmans as with the Jambavya and Āgniveśya adherents–come from the following domiciles, all recognizably of the Tamil country: Kōṭṭaiyūr: 2; Emappērūr: 1; Ādanūr:7; Palurūr:1; Puḷḷamaṅgalam 10; Marudūr 2; Pulvāvūr 1; Tiṭṭakuṭi:2; Iṭaiyāṟṟukuṭi:5; Māruṇdūr:4; Anbil: 3; Nāraṇamaṅgalam: 1; Cātthamaṅgalam:1; Aruvalam: 1. Moreover, as noted already, some of the adherents of the Āśvalāyana and Baudhāyana Sūtras may also be Pūrvaśikhās, indeed cohorts of the Jaiminīyas, as several of them are from the same Tamil villages as the Jaiminīya Śōḻiyas.
All the same, the dominance of the signature Aparaśikhā Veda śākhās, already clear in the Pallava period, is even greater in the Karandai Plates: more than 800 belong to Veda śākhās recognizable as those of the Aparaśikhā group. The largest single group, at 615, is made up of the adherents of the Āpastaṃba Sūtra, with another 33, of the closely related Hiraṇyakeśi and Bhāradvāja. The 77 Drāhyāyana adherents represent a robust Aparaśikhā Sāmaveda tradition, no doubt the back bone of the Aparaśikhā Śrautism attested in the Plates, a strength that is still extant among the Aparaśikhā Brahmans in the Tanjavur-Kumbakonam area.
Some 500 families, all following the signature Aparaśikhā Veda śākhās, carry the last name daśapuriyan, derived from the city of that name in Malva, increasing from its 30 occurrences out of the Pallava total of 467, pointing to the origins of the Aparaśikhā group in north-central and northwestern regions, along the Narmadā, Chambal, and Yamunā banks.
As for the Pallava Brahmans, I give in Table 4 a breakdown of the “learning quotient” of the Karandai Brahmans:
-Trivedi 1
- Caturvedi 2
- Ṣaḍaṅgavit 4
- Kramavittan 118
- Āhitāgni 4
- Somayāji 28
- Kāṭaka- Somayāji 3
- Vasantayāji 3
- Kāṭaka- Sarvakratu Vasantayāji 1
- Sarvakratu 1
- Agnicittayāji 2
- Vājapeyi 1
- Atirātran 1
- Sahasran 151
Table IV: The Learning Quotient of the Karandai Families
There are almost 50 Śrautins in the group (of which 5 are identifiably Pūrvaśikhās, being Jaiminīyas, of a total of 41 [12.5%], indicating a robust śrauta tradition among the Śōḻiya Pūrvaśikhās in the 11th century CE; the one Atirātran may also be a Pūrvaśikhā Brahman, this being the term still in use among the Nambudiri Pūrvaśikhās to signify a ritualist who has performed the Agnicayana), with another 118 Brahmans who can recite the Vedas upto the krama vikṛti. In other words, the in-coming Aparaśikhā Brahmans continue to be drawn from the same elite levels as in the Pallava period, a trend that is to continue, further justifying the assumption that a version of Mahābhārata epic, almost certainly a Vulgate text by now, came with them.
C. iv. The Emergence of the Aparaśikhā Śrīvaiṣṇavism
One way to approach the development of the eventual Tamil (Grantha)-Telugu Mahābhārata of the Aparaśikhās is to approach it in the perspective of a precedent. Such a precedent exists in the formation of the texts and traditions of the mature Śrīvaiṣṇavism by the Aparaśikhā Brahmans from the Āḻvār songs, collected in the Nālāyiradivyaprabhandham by an Aparaśikhā Brahman. This is, as noted earlier, the famous Nāthamuni, generally thought to have been born in 11th century, in Vīranārāyaṇapuram, very much the village of the Pallava-Cōḷa epigraphy, perhaps a first generation Aparaśikhā immigrant, among, as noted above, people still with the northern name, Miśra (Carman 1973: 24).100 Once hearing a decad of the still uncollected NDP the pāsuram, Āravamudē (3194) by singers from the “west” (the traditional Cōḷa area along the Kaveri river west from Nāthamuni’s Vīranārāyaṇapuram in the relatively northern and eastern Toṇṭaimaṇṭalam-Arcot area, the region of the Śōḻiya Pūrvaśikhās who supply all the Brahman Āḻvārs), tradition has it that Nāthamuni seeks out and collects the songs into the extant text, the Nālāyiradivyaprabhandam, setting it besides to music, inaugurating the great performance tradition of the aṛaiyars in the Viṣṇu temples of Tamil Nadu. And the figure from whom he is able to collect the 4000-verse long text is Maturakavi, a Jaiminīya Sāmavedi and hence unambiguously a Pūrvaśikhā Brahman. However, and this is the other half of the reaction, tradition has him also bring to the Āḻvār Vaiṣṇavism northern texts and practices as well (the counterpart of the Sukthankar σ-text), as set forth in his Nyāya Tatva and Yoga Rahasya, laying the foundation through his grandson and disciple, Yāmuna, to the Pañcarātra-Āgama tradition (Carmen 1973: 25), the entire line of development culminating in Rāmānuja, Yāmuna’s grandson. We know that Rāmānuja belonged to the signature Aparaśikhā sūtra of Āpastaṃba, belonging in addition, as noted above, to the vaṭama group, hailing from Śriperumbedur, in the Tonṭaimaṇṭalam area, near Kanchipuram and a descendant of Nāthamuni on his mother’s side.101
I noted above that the Bhāgavata Purāṇa is a literary expression of this religious synthesis. Dated to ca. 9th century CE, very much in the period of Nāthamuni, the Bhāgavata Purāṇa incorporates, as Dihejia shows,102 many elements of the Āḻvār Vaiṣṇavism, but addressing at the same time an extra-Tamil audience, in the north, still no doubt a place historical memory for many Aparaśikhās, with Nāthamuni himself going to Mathurā on a long sojourn and coming back to the peninsula only when compelled by a vision of the deity of his natal village commanding him to return (Carman: 24-25). Indeed, when the great Śrīvaiṣṇava schism into “vaṭakalai” (northern) and “tenkalai” (southern) occurs in the post-Rāmānuja period, the vaṭakalai branch is seen to be made up of entirely Aparaśikhā Brahmans, with Tirupati in the north, outside the northern boundary of the modern state of Tamil Nadu, as its center, with the tenkalai school, located in the south, in Śrī Rangam, Śrīvilliputhūr and Āḷvārtirunagarī, orienting itself to the Āḻvār Vaiṣṇavism and Tamil, its language. The tenkalai branch is made up both of the Aparaśikhā and the Śōḻiya Pūrvaśikhās, the latter less than 15% of the smaller tenkalai group and relegated to a low social status among the Śrī Vaiṣṇavas, although originally among the founders of Āḻvār-Vaiṣṇavism.103
C. v. The Tamil (Grantha) and Telugu versions of the Mahābhārata
It is in the perspective of the above precedent that we must approach the formation of the Tamil (Grantha)-Telugu versions of the Mahābhārata. In both cases, we have a resident tradition hosting an immigrant tradition, giving rise to broader and larger developments in both cases, the Bhāgavatapurāṇa in the Vaiṣṇava tradition, the summum bonum of the Śrīvaiṣṇava precedent, and the Tamil (Grantha)-Telugu versions of the Mahābhārata, its epic counterpart. However, as we noted, we have very little concrete information about the precise details of how the Tamil (Grantha)-Telugu versions of the epic developed: we do not have the equivalent of an iconic figure like Nāthamuni of the Vaiṣṇava tradition, the figure who weaves the southern Āḻvār and the northern Pañcarātra, its σ-text, into Śrīvaiṣṇavism of the Tamil country through Yāmuna first and Rāmānuja thence. It is quite possible that the Villipputhūr Mahābhārata represents a stage in the development of the Tamil (Grantha)-Telugu Mahābhārata. Śrī Villipputhur is traditionally a Pūrvaśikhā agrahāram, the birth place of Periyāḷvār and Āṇdāḷ, two of the four Pūrvaśikhā Brahman Āḻvārs. The author of the Tamil translation of the epic is named after the village and traditionally considered to be a Śrī Vaiṣṇava Brahman, and dated to the Tamil Middle Ages (12th to 13th CE) although we do not know if he was a Pūrvaśikhā or the Aparaśikhā type.104 We know that the Villipputhūr text served as the fundamental source for the Tamil kūthu repertory, in the non-Brahman circles.105 Being a center of the emerging Śrīvaiṣṇavism, second perhaps only to Śri Rangam, being in regular contact with this bigger center, Śrī Villipputhur may well have been the center of reaction between the two recensions in the Tamil country, with the final phases of it occurring in the Nāyaka period, in the 16th -17th centuries, as P.P.S.Sastri pointed out, in his Southern Recension edition of the Mahābhārata,106 made from pretty much the same Tamil(Grantha)-Telugu manuscripts of the Sarasvatī Mahāl Library of Tanjavur that went to Poona for the collation and preparation of the Critical Edition.
C. vi. The Grantha and Telugu Mahābhārata and the Poona Critical Edition
We have from the first half of the 16th century a kāvya work titled Viśvāguṇadarśacampū107 by a Veṅkatādhvarin, identified as “an orthodox Śrī Vaiṣṇava Tamil Brahman” (Rao et al 1992: 1) with his name ādhvarin deriving from adhvaryu, the main śrauta priest and belonging to the Yajurveda. Purporting to be an aerial journey over the Tamil country by two gāndharvas, conversing between them on the earthly sights below, the poem is an objective representation of the final Aparaśikhā ‘possession’ of the Tamil country, an aerial map literally laid over the territory of the Pallava-Cōḻa and subsequent grāmadeya epigraphy about the Aparaśikhā Brahmans. The gāndharvas begin their peninsular journey at the Karnataka Aparaśikhā centers at Udupi and Melkote and, flying due east to Tirupati, the most important, by the time of the poem, vaṭakalai, and thus all-Aparaśikhā, center of Tamil Śrīvaiṣṇavism, they turn southward and retrace the path of the Aparaśikhā immigration, covering the entire region of the Pallava and Cōḷa epigraphy, starting with Kanchipuram in the northeast, coming to the Kaveri river banks stretching from Śri Rangam through Tanjavur to Kumbakonam in the east, the Tanjavur- Kumabakonam-Mannarkode area, and south to the Tāṃṛavarṇī delta (Map V after Map 1 in Rao [1992] et al.) The gāndharvas notice the author’s village, Vīkṣāraṇya, not far from Ramanuja’s village at Sriperumbudūr, both in all likelihood villages of the Pallava-Cōḻa gramadeya system, a system the Nāyakas continued. We know that poets like Venkatādhvarin above found patronage with the Nāyaka chieftains, the latter, Telugu-speaking, coming south to the Tamil country with the dissolution of the Vijayanagara empire, and establishing themselves as rulers there, the “little kings” eventually with “hollow crowns”.108 Indeed Veṅkatādhvarin is himself linked to the Señji Nāyakas, and his poem partakes of what has been identified with the Nāyaka ethos, centering around the theme of the “unknown, unpedigreed warrior who fights his way into power and a kingdom of his own” (Rao et al. 1992: 7). Moreover, the Nāyaka courts produced “an enormous corpus of Sanskrit works, reflect[ing] the accumulated erudition of late medieval south India” (336), altogether a fitting environment for what P.P.S.Sastri has called the “Nāyaka excesses” of the Grantha- Telugu Mahābhārata.
This is particularly true in the case of Tanjavur, which by all account went through a brief renaissance—beginning thus a journey toward the eventual capital of Brahmanical culture of the Tamil country–under its three Nāyaka kings, Accutappa Nāyaka (1564-1612), his son Raghunātha Nāyaka (1600-1634) and his son Vijayarāghava Nāyaka (1631-1673). The famous Govinda Dīkṣita begins his career as the King’s Minister with the first of the three Nāyakas, providing tutelage and a splendid education for the middle Nāyaka, a Renaissance prince in every respect,109 and his son, Yagnanārāyaṇa Dīkṣita, continuing his father’s cultural and artistic leadership. The Tanjavur court was the host to many poets and musicians, with Raghunātha Nāyaka actually fashioning the extant vīṇa of the Carnatic musical tradition. As Krishnasvami Aiyangar (1941: [II] 296), a 20th century descendant from the grāmadeya village of the third Nāyaka–called at the time of the grant Raghunāthapuram in honor of his father but now Śakkoṭṭai–notes, Raghunātha Nāyaka “held a competition among the ladies of the court, several of [whom] could compose poetry in the four kinds. They were also expert in resolving curious literary puzzles. Some of them could compose hundred verses in “an hour” and write poetry in eight languages. One lady of the court by name Rāmabhadrāmba was accorded first place in this and was installed as the “empress among poets” (sāhityasāmrājya) which probably involved the honor of kanakābhiṣeka (bathing [sic] in gold)”. Thus we have every reason to think that the Tanjavur court functioned as a nursery for the Tamil (Grantha)-Telugu versions of the Southern Recension, with their inflationary excesses.
We must note that Tanjavur’s famed Sarasvatī Mahāl Library, the final source110 of the manuscripts of the Tamil (Grantha) and Telugu Mahābhārata for the Poona editors began its life as the Sarasvatī Bhandār in the early 17th century under Raghunātha Nāyaka. This tradition of scholarship and respect for the arts continued after the Maharashtrian take-over of Tanjavur in late 17th century, in 1674 CE, with Sarasvatī Bhandār metamorphosing into the Sarasvatī Mahāl Library and acquiring vast numbers of manuscripts from Benares, under Serfoji II, during his famous pilgrimage to the holy city in 1832 with a retinue exceeding 3000. True, some Maharashtrian Brahmans came to Tanjavur with the Maharashtrian conquest and rule, but there is little doubt that the city’s intellectual and cultural life was entirely the creation of the Aparaśikhā Brahmans, long prepared for their eminent role through the historical processes described by Burton Stein (1982). A roll call would include such names as Appayya Dīkṣitar (1520-1593); Govinda Dīkṣita and his son, Yagnanārāyaṇa Dīkṣita; and later, the musical trinity of composers of the Carnatic music, all from Tanjavur, all anecdotally Aparaśikhā Brahmans. In all likelihood, the final form of the Tamil (Grantha) and Telugu Mahābhārata takes shape in this period, 16th to 17th centuries, CE.