THE MAKING OF THE ANCIENT TAMIL LITERARY CANON

Author(s): V. Rajesh
Source: Proceedings of the Indian History Congress , 2006-2007, Vol. 67 (2006-2007), pp.
153-161
Published by: Indian History Congress

Stable URL: https://www.jstor.org/stable/44147932

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Indian History Congress is collaborating with JSTOR to digitize, preserve and extend access to
Proceedings of the Indian History Congress

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms
 THE MAKING OF THE ANCIENT TAMIL
LITERARY CANON
V. Rajesh

In a multi-cultural and pluralist society like India, when litera

production is associated with power, the notion of canon is histo
contingent. It has been argued that the wide array of overlapp
institutional practices like writing commentaries, citing sourc
reference confers a literary text a canonical status.1 In other words
levels of objectification that a literary text undergoes in tradition c
it a canonical status. The reasons for objectification may differ
scholarly concern to that of an assertion of power by an emerging
elite. It was the Orientalists who took up the canonization of cl
Sanskrit texts during the colonial period. A normative value was
subsequently established and it was widely believed that the canons thus
produced represent the total literary ethos and the essentials of Indian
culture. More than any other factor, it was the print culture that
established the fixity and closure for literary texts, which in pre-modern
period was more fluid. This does not however, mean that it was left
uncontested. Here, we seek to demonstrate that the notion of 'Sanganv
literary canon is historically contingent or to state more clearly, shaped
by the contemporary needs. At the first instance, it will be argued how
the Tamil literary tradition itself was historically shaped, and secondly
it will be suggested how this process effected the modern understanding
of the Tamil literary canon. The aim here is not to question in any way
the chronology or the integrity of the text but aspects like the
commissions and omissions, interpolations and marginalisation in the
making of a canon will be highlighted.
The 'Sangam' literature constitutes Ettutokai (Eight anthologies)
and Pattupattu (Ten songs) dated to first three centuries of the Christian
era. Some scholars are averse to include the latter as a part of 'Sangam'
literature since the tradition itself did not sanction.2 'Sangam' means an
assembly. The first detailed account of the existence of the assembly of
poets and their works in Tamil tradition emerges in the 8lh century A.D.
commentary to the grammatical work Iraiyanar Akapporul. The
commentary states that the ancestors of the Pandya kings established
three literary academies where poets, gods and sages participated and
contributed poems. The claims made for the number of years for each
academy was enormous which goes beyond the comprehension of
modern historical understanding.3 The works of the first two academies
were lost and for the third the commentary outline the following,

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms
154 IHC: Proceedings , 67th Session , 2006-07

Netuntokainanuru, Kuruntokainanuru, Narrinainanuru, Purananuru,

Ainkurunuru, Patirrupattu, Nurraimpatu Kali, Elupalu Paripatal and
Tolkappiyam.
As we have seen above, the commentary outlined only the Ettutokai
and the grammatical text Tolkappiyam as third 'Sangalli' works but did
not include Pattupattu. Some of the poets who composed poems in
Ettutokai were the authors of the Pattupattu. In that sense, it is enigmatic
as to why the 8th century commentary must include the former as works
of third Sangam while leaving out the latter. Was Pattupattu not available
to the 8lh century commentator? According to K.N. Sivaraja Pillai, the
commentary to Iraiyanar Akapporul was orally transmitted for ten
generations and that it was one Nilakantan of Muciri who had actually
put it down in writing in 8lh century A.D.4 Though, the names of the
individual anthologies were mentioned as part of 'Sangam' literature,
no where in the poems of the anthologies we find any evidence of the
assembly of poets and the concomitant production of literature. The term
'Sangam' , as Zvelebil argues was appropriated from the Jainas who
founded the Dravida Sangam.5
'Tokai' means anthology, which presupposes a collection and
compilation. Thus Ettutokai means eight anthologies. Each of the
collections in Ettutokai can be understood as an anthology. Tokai in
Kuruntokai means an anthology of short four hundred. Thus, in
Kuruntokai, we have four hundred unique lyrical love poems in aciriyam
meter composed by 203 poets. The tradition has preserved, apart from
the names of the poets who composed the poems, the names of the
compilers and patrons were also preserved. Thus, Purikko compiled
Kuruntokai and the tradition also informs us that the poems were
compiled based on the number of lines. For example, the poems of
Kuruntokai were compiled based on the length of four to eight lines.
Similarly, Narrinai comprises the poems of length in between Kuruntokai
and Akananuru i.e. 9-12 lines. The name of the compiler of this anthology
is not known but the tradition informs us that it was commissioned by
Pannatu tanta Pantiyan Maran Valuti. That the poems of Narrinai were
composed in aciriyam meter is commonly known. Akananuru means
four hundred poems on 'Akam'. Urutirra Canmar compiled it under the
patronage of Pantiyan Ukkirap Peru valuti. The length of the poems ranges
from 1 3 to 3 1 lines. The four hundred poems were composed in aciriyam
meter.

The poems, written and recited from lsl century A.D. onwards were
compiled around 3rd century A.D. According to Kamil Zvelebil, 'the
first compilation was not much removed in time from the stage of actual
composition of some of the poems: it may even have been

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms
 Ancient India 155

contemporaneous with it.'6 At the e

information about names of compilers
was compiled by Madurai Uppurikuti
the commission of Pantiyan Ukkira
Purananuru 21, 367. Commenting on
J.R. Marr observes that, 'If such tra
conclude that these anthologies were co
when these kings lived, and possibly
of the personages who figure in the po
and puram'.7 It is pertinent to underst
poems into an anthology form aroun
attributes this compilation to the di
consequently, the consolidation of the
he argues that since 'Sangam' poems
'literacy', the oral poems, which wer
selected, compiled and put down int
might have been a factor for compilin
of compilation i.e., based on number
there was a consciousness of the exis
compilation reflect the systematizat
we are not sure whether the compile
compiler had selected the poems fro
have to suppose that the sources we a
was considered to be essential by th
society. We do not have any evidence
Purananuru. Patirruppattu is exclusivel
in all probability it must have been com
Chera chiefs. Kalitokai and Paripatal,
is a part of eight anthologies, it is n
scholarly circles that the two antholog
of post-Sangam period (5lh-7lh A.D
treating these two anthologies as lat
Indo- Aryan loanwords with thought c
of the anthologies.10 Thus, the six o
been compiled around 3rd century A
content and thematic unity of these
fact it represents simultaneity of pr
As already stated, it was the comm
grammatical text Iraiyanar Akapporu
the existence of łSangam' According
the grammar is the lord Siva himself.
time when there was fervor of Saiva and Vaishnava bhakti in the Tamil

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms
156 IHC: Proceedings , 67th Session , 2006-67

land. Moreover, contestation against the sramanic religion was endemic

to the period under discussion. The author of the commentary was very
much conscious of the existence of 'Dravida Sanganť of Jainas, though
there is no direct allusion to it. The commentary, according to K.
Sivathamby, 'attempts to take over an obviously Jaina and Buddhist
institution (sanga) and give it a Hindu form and content'.11 The references
to the existence of only the individual anthologies in the commentary
i.e, Netuntokainanuru, Kuruntokainanuru, Narrinainanuru, Purananuru,
Ainkurunuru, Patirrupattu, Nurraimpattukali and Elupatu Paripatal
indicate a fact that these individual anthologies were not yet collected
together into super anthology - the Ettutokai. The process of
anthologisation was intimately connected to the politics of identity
assertion and hence it is pertinent to address the issues of legitimization
by the newly emerging ruling elite.
The invocatory poems to the five out of eight anthologies require
considerable attention as it may relate to the process of anthologisation.
Akananuru, Kuruntokai, Narrinai, Ainkurunuru and Puranauru contain
an invocatory poem written by Paratam Patiya Peruntevanar. The subject
matter of these invocatory verses is completely alien to the rest of the
poems in the anthologies. Akananuru has an invocatory poem in praise
of Siva. So is also the case with Purananuru. Narrinai and Ainkurunuru
which have an invocatory poem in praise of Vishnu. Kuruntokai alone
contains an invocatory poem in praise of Murukan. John Ralston Marr
has shown that these invocatory poems are alien to the subject matter of
the rest of the poems in the respective five anthologies.12 The preface or
the invocatory poem in the Patirupattu is missing and it can only be
speculated that it might have been written for this anthology too. The
rest two anthologies Kalitokai and Paripatal did not contain any
invocatory poems written by Peruntevenar. Whether Kalitokai and
Paripatal were composed or not at a time when Peruntevanar wrote
invocatory poems for the rest of the anthologies are open to doubt. But
there is a general agreement, as already discussed among scholars that
these two works are later than the rest of the anthologies. Marr has
convincingly argued that Peruntevenar must have been contemporaneous
with or later than the 9lh century A.D. Pallava Nandivarman.13 It is rather
anachronistic that M.G.S. Narayanan treats these invocatory verses as a
part of the original anthology in outlining the Vedic, Sastric and Puranic
elements in the Cankam period.14 Even the secular nature of the Sangam
anthologies were attributed a religious fervour when invocatory verses
were written during early medieval Tamilakam. These attempts reflect
the case that bhakti was a dominant ideology during early medieval
Tamilakam. The historiography on Tamil bhakti religion in one way or

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms
 Ancient India 157

the other emphasized the connectio

poetry with the Sangam poems in term
employed for the composition of the p
their songs based on Akam and Puram
though the varying thought content
poets must have mastered the classic
back to the question of the availability
period of existence.
It has been argued in the nationali
period in the Tamil history is consid
century A.D.). The Kalabhras patron
country there was a disruption of text
conventions. Ethical texts were pro
stressed on peace, virtue and morali
witnessed a break under the Jain pat
and songs were condemned by Jaina
disruption of bardic tradition during
country.16 There is a story in the comm
which states that the grammar was wr
because of the absence of men spe
(poetics) of the classical 'Sangam' liter
from outside the Pandya country wh
men specialized in Col (orthography)
rationale for the legend is to legitim
patronized poetics associated with
Iraiyanar Kalaviyal. It must be under
discovered the grammar written by lo
according to the story. In other wor
newly emerging elite groups were a
invocatory verses written to the clas
the deliberate stripping of some of th
the attempts of the newly emerg
Tamilakam. The contestation against th
from the attempts that were made as
made in the verses of the Saiva and Vaishnava saints. Buddhists and
Jains countered these attacks and it can be gleaned from the text
production from their religious viewpoint. Thus, in Viracoliyam, the
author who was a Mahayanist stated that it was Avalokiteshwara who
gave Tamil to Agasthya.17
In the epigraphical textual tradition of Pandyas too, we find the
allusion to 'Sangam*. In the Tamil portion of the Dalavaypuram copper
plates of the early Pandyas, we find allusion to the establishment of

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms
158 IHC: Proceedings , 67th Session, 2006-07
'Sanganť by the ancestors of the Pandyas.1R In the larger Cinnamanur
plates, we find the allusion to the establishment of 'Sangam' by Pandyas
and also the translation of Mahabharata into Tamil.19 It has been argued
that an inscription from Ramnad district published in Madras
Epigraphisťs Report [No. 334] alludes to the ancestor of Etticatan, who
was a poet who sat on the 'Sangam' bench.20 Thus, both the literary and
the epigraphical texts produced during early medieval Tamilakam
represent a sense of belonging to the classical Tamil literature. In thought
content and ideology, both literature and inscriptions of early medieval
Tamilakam represent an alienating distanciation from the classical Tamil
literature.21

By the end of early medieval period (around 1 2th century A.D.), Tamil
society became complex. From 12lh century A.D., there was a
proliferation of the institutions of pedagogical learning located at the
sacred sectarian centres. Both Saivite and Vaishnavite hagiographical
works were produced by then and there was a need to comment on the
text. Commentary (Urai) presupposes the existence of text (Mulam).
The cultural milieu of these commentators was different from that of
the original text. In the process of the comprehending of the meaning
for certain verses, the commentators disagreed. But there was an attempt
at systematizing the knowledge during the medieval period.
Commentaries often carry the subjective bias of the commentator, since
the commentator may be associated with a particular sect.22 While
commenting the commentator had to cite references and sources. U. V.
Swaminatha Iyer while delivering the lecture on Arachi (research) in
early 20lh century stated that in his life as a researcher, he understood
the existence of verses (mulam) from commentary (Urai) and
commentary (Urai) from verses (mulam).23 Thanks to the medieval
commentators whose systematization of knowledge helped modern
scholars like U. V, Saminatha Iyer to re-discover the texts from oblivion.
In Indian tradition, even the commentaries were oral in origin. It was
formulaic suited for memorization and recollection. Zvelebil has shown
a verse from Iraiyanar Akapporul's commentary, as to how this
commentary was orally transmitted ( ini urai natantu vantavaru
collutum ).24
The individual anthologies that we noticed in Iraiyanar Akapporul's
commentary, in the course of time around 10th - 1 1th centuries A.D, were
collected together and made into a super anthology the Ettutokai. It was
first mentioned, along with Pattupattu (Ten songs) in Peraciriyar's
commentary on Tolkappiyam's Porulatikaram (Poetics). Zvelebil
attributes the chronology of the commentary to around 1 3lh century A.D.25
Around the same period, Mayilainathar's commentary on the

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms
 Ancient India 1 59

grammatical text Nannul also alludes

anthologies Ettutokai and Pattupattu.
scholars of final redaction and codification of classical Tamil literature.
The anthologies must have attained the present shape at the point in
time under discussion. But having said that, it is extremely difficult to
locate the interpolations in the original text. The internal uniformity in
terms of meter, language and theme is remarkable and there is no question
of interpolations. The redactors faithfully copied the text and they only
super anthologized. U.V. Swaminatha Iyer stated that in those days, the
commentators showed great respect to the verses (Mulam) and they were
trained in copying it exactly without mistake, though there was a freedom
of commenting on the verses.26
There were number of commentators whose commentary on the
grammatical text Tolkappiyam deserves discussion. It was felt necessary
by the medieval linguists to master the language for pedagogical
purposes. Series of commentaries were written, so that pedagogical
exposition could be made without much difficulty. It was around 11th
century A.D, that the first commentary to Tolkappiyam was written by
llampuranar. In his commentary there are no allusions to the existence
of super anthologies, the Ettutokai and Pattupattu. The individual
anthologies must have been redacted and codified after his period. If it
were in existence, we would not expect great commentators like
llampuranar for not mentioning it. Naccinarkkiniyar commented apart
from Pattupattu on the first two full books and five chapters to the third
book of Tolkappiyam. He must have been a contemporary of
Parimelalakar and thus can be dated to 14th century A.D. Apart from
llampuranar and Naccinarkkiniyar, Cenavaraiyar (1 3th c.), Peraciyar ( 1 2th-
13th), Teyvaccilaiyar and Kallatar (16lh -17th c.) also commented on a
few portions of Tolkappiyam. Though old anonymous commentaries exist
for the individual anthologies of Ettutokai, it is enigmatic as to why it
was not commented in detail like Tolkappiyam. Moreover, except for
Kuruntokai, the names of the commentators who wrote commentaries
for the rest of the anthologies are not available. In medieval period,
primacy was given to religious texts and grammatical works whereas
earlier secular literatures like 'Sangam' literature was sidelined and not
recommended for pedagogical learning. For Ainkurunuru, Patirrupattu,
Akananuru and Purananuru, we have old anonymous commentaries in
fragments. No conclusive dates have been established for these
anonymous commentaries so far. Only for Kuruntokai, for the first 380
poems, we have evidence that Peraciriyar wrote commentary, who lived
around 12th -13th centuries A.D. For the rest 20 poems in Kuruntokai, U.
V. Swaminatha Iyer stated that Naccinarkinniyar of the 14lh century A.D

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms
 160 IHC: Proceedings , 67th Session , 2006-07
wrote a commentary.27 But unfortunately, both these commentaries are
not available for us today. When U. V. Swaminatha Iyer and Damodaram
Pillai re-discovered the classical Tamil texts, they retained the old
anonymous commentaries and also v/rote full detailed commentaries
wherever they felt it necessary. Thus, the making of the ancient Tamil
literary canon was intimately tied to the needs of the time.

NOTES AND REFERENCES

1 . A.R. Venkatachalapathy, 'In Print, On the Net: Tamil Literary Canon in the Colo
and Post-Colonial Worlds' in Suman Gupta, Tapan Basu and Subarno Banerji (eds
India in the Age of Globalisation : Contemporary Discourses and Texts , Neh
Memorial Museum and Library, 2003, p. 135.
2. John Ralston Marr, The Eight Anthologies: A Study in Early Tamil Literatu
Institute of Asian Studies, Madras, 1985, p. 7.
3. Thus the first Cankam lasted for 4440 years and the second for 3700 years wh
third Cankam for 1850 years. See K.R. Govindaraja Mudaliyar and M.V.
Venugopala Pillai (ed.), Kalaviyal ennum Iariyanar Akapporul mulamum,
Nakkiranar Uraiyum , 1939, pp. 5-7.
4. K.N. Sivaraja Pillai, Chronology of Early Tamils , University of Madras, 1932, p.
19.

5. Kamil V. Zvelebil, The Smile of Murugan: On Tamil Literature of South India ,

Leiden, E.J. Brill, 1973, p. 48.
6. Quoted in K. Sivathamby's, Literary History in Tamil. A Historiographical Analysis ,
Tamil University, Tanjavur, 1986, p. 32.
7. John Ralston Marr, op. cit., 1985, p. 331.
8. K. Sivathamby, op. cit., 1986, p. 33.
9. Ibid., p. 32.
10. Kamil Zvelebil, op. cit., 1973, pp. 121-130.
11. K. Sivathamby, op. cit., 1986, p. 35.
12. John Ralston Marr, op. cit., 1985, p. 71.
13. Ibid. p. 72.
14. M.G.S. Narayanan, 'The Vedic-Puranic-Sastric Element in Sangam Society and
Culture' , Proceedings of the Indian History Congress , 36th Session, Aligarh, 1975,
p.78.
15. A.K. Ramanujan and Norman Cutler, 'From Classicism to Bhakti* in Vinay
Dharwadkar (ed.), The Collected Essays of A.K. Ramnujan , Oxford University Press,
2001, pp. 232-259; Indira V. Peterson, Poems to Siva: The Hymns of the Tamil
Saints, Princeton University Press, Princeton, 1989, p. 84; Friedhelm Hardy, Viraha-
Bhakti : The Early History of Krsna Devotion in South India, Oxford University
Press, Delhi, 1983, p. 276.
16. M. Arunachalam, The Kalabhras in the Pandya country and Their Impact on the
Life and Letters There, University of Madras, 1979, pp. 84-90.
17. K. Sivathamby, op. cit., 1986, p. 36.

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms
 Ancient India 161

18. Dalavaypuram Copper Plates, Tamil Portio

of the Early Pandyas c 300 B. C to 984 A.Dt
2002, pp. 74-75.
19. Cinnamanur Plates V. 102-104, Ibid. p. 1
20. Kamil V. Zvelebil, Tamil Literature , Leid
21 . 'Sense of Belonging' and 'Alienating Dist
employed by Paul Ricoeur in understanding t
see Paul Ricoeur, Hermeneutics and the Hum
J. Thompson, Cambridge University Press
22. For an excellent analysis of medieval and
Tirukkural text see, Norman Cutler, 'Int
Commentary in the Creation of a Text 'JA
23. U. V. Swaminathier, Canka Tamizhum Pirka
Library, 1929, p. 144.
24. Kamil V. Zvelebil, op. cit., 1973, p.249.
25. Ibid., p. 25.
26. U. V. Swaminathier, op. cit., 1929, p. 14
27. See preface in U. V. Swaminathier's (ed.),
first published 1973.

This content downloaded from

205.253.37.196 on Wed, 13 Mar 2024 09:56:14 +00:00
All use subject to https://about.jstor.org/terms