This Content Downloaded From 130.63.63.214 On Sat, 14 Jan 2023 19:29:03 UTC

A Tamil Lineage for Saiva Siddhānta Philosophy
Author(s): Karen Pechilis Prentiss

Source: History of Religions , Feb., 1996, Vol. 35, No. 3 (Feb., 1996), pp. 231-257
Published by: The University of Chicago Press
Stable URL: https://www.jstor.org/stable/1062814
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
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend
access to History of Religions
This content downloaded from

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
All use subject to https://about.jstor.org/terms
Karen Pechilis Prentiss A TAMIL LINEAGE FOR
SAIVA SIDDHANTA
PHILOSOPHY
"Saiva siddhanta" is the name that certain medieval philo

gave to their philosophical system. Who these philosophe
understanding of their method and the integrity of the
system, has recently become a focus of Western acad
scholars involved in this study are overwhelmingly conc
skrit writings of the Saiva siddhantins. That Saiva siddh
wrote authoritative texts in Tamil has been long acknow
relationship between the two authoritative traditions ha
1 Aspects of Saiva siddhanta philosophy received brief mention in

influential during the last century, especially the Indian Antiquary and
Royal Asiatic Society. Greater understanding of this philosophical sy
identification and editing of the Sanskrit texts, called Saivfgamas, whic
some extent in the early part of this century, has been an intensive focu
times. Contemporary efforts have been spearheaded by scholars affiliat
Francais d'Indologie in Pondicherry, including H. Brunner, B. Dagens,
by scholars affiliated with the Kuppuswami Sastri Research Institute
S. S. Janaki and N. R. Bhatt (formerly head of the Pondicherry Institut
scholars besides myself studying the Saiva siddhanta include R. H. Da
2 The Tiruvarutpayan, one such authoritative text, had been first tran
in 1896, under the supervision of the Dharmapuram mutt (J. M. Na
varutpayan of Umapathi Sivacharya [1896; reprint, Dharmapuram
1945]). This text had been translated again by G. U. Pope, and he used
to his translation of the bhakti hymns of Manikkavacakar, in The Tiruv
Utterances" of the Tamil Poet, Saint, and Sage Manikkavacagar (Oxfor
The authoritative moder work in English on the subject of the Saiva sid
in Tamil is M. Dhavamony, Love of God According to Saiva Siddhanta
1971).
? 1996 by The University of Chicago. All rights reserved.

0018-2710/96/3503-0002$01.00

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
232 A Tamil Lineage
understood.3 In this article, I consider issues of difference between the

two traditions of Saiva siddhanta, the school that composed its canonical
texts in Sanskrit and the school that composed its canonical texts in
Tamil, from the perspective of my study of the latter's philosophy. For the
purposes of this article, I call these philosophers the "Sanskrit Saiva sid-
dhantins" and the "Tamil Saiva siddhantins," respectively, emphasizing
the language of their texts rather than their ethnicity. Through my discus-
sion of general characteristics of the two groups, followed by a detailed
consideration of one of the Tamil philosophers, Umapati Civacaryar, I
argue that the Tamil Saiva siddhantins sought to distinguish themselves
from the Sanskrit Saiva siddhantins. They accomplished this primarily
through their dissociation of Saiva siddhanta philosophy from a specific
temple ritual tradition and by their creation of a Tamil lineage with re-
spect to the Tamil bhakti poets, known as the nayanmar.
ON UNDERSTANDING SAIVA SIDDHANTA
The name Saiva siddhanta translates to "the end of knowledge o

in the sense of "culmination." Like the Vedanta, which purports
authoritative explanation of the Vedas, Saiva siddhanta makes an
mous philosophical claim. However, in the case of the Saiva s
the knowledge is centered around the divine figure Siva, rather
corpus of divine hymns. To call oneself Saiva siddhanta is to ass
one is within a philosophical lineage that extends back to Siva h
and that this philosophical knowledge supersedes any other con
the realities of Siva and salvation.
The dominant Western scholarly perspective on Saiva siddhanta is to
understand it as a regional school. While it is acknowledged that there
was a school in Kashmir, the preeminent locus of Saiva siddhanta is in
Tamil Nadu. A recent encyclopedic definition, authored by the influen-
tial interpreter of Saiva siddhanta M. Dhavamony, illustrates the wide-
spread characterization of this philosophical school: "Saiva siddhanta is
an important medieval system of Saiva thought. The term technically
refers to a set of Saiva theologies written in Sanskrit and Tamil in South
3 Richard H. Davis gives a brief discussion of this topic in Ritual in an Oscillating Uni-
verse: Worshipping Siva in Medieval India (Princeton, N.J.: Princeton University Press,
1991), pp. 18-19. He cites two pages of an article by H. Brunner ("Importance de la lit-
terature agamique pour l'6tude des religions vivants de l'Inde," Indologica Taurinensia 3-4
[1975-76]: 107-24) as "the only critical work devoted to this, as far as I know" (p. 169).
But see also K. Sivaraman, "The Role of Saivagama in the Emergence of Saivasiddhanta:
a Philosophical Interpretation," in Traditions in Contact and Change: Selected Proceedings
of the XIVth Congress of the International Association for the History of Religions, ed.
Peter Slater and Donald Wiebe (Waterloo: Wilifrid Laurier University, 1983), pp. 53-64.
He views the Saiva siddhanta as the sum of both Tamil and Sanskrit streams of thought:
"Saivasiddhanta holds together the Tamil tradition of the Agamas and the Hymns, and the
Sanskritic tradition of Upanisadic thought" (p. 64).

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
History of Religions 233
India, although this classification need not be considered a rigid one."4

While this description acknowledges the difference in textual languages,
it still portrays the sina qua non of Saiva siddhanta as its location in
south India, in spite of the author's modest disclaimer.
Although it is the case that the Tamil lands became an important locus
for the development of Saiva siddhanta, as I will discuss below, Richard
H. Davis has recently argued that the Sanskrit Saiva siddhantins did not
view their philosophical school as a regional school, and thus the aca-
demic definition of Saiva siddhanta as localized primarily in Tamil Nadu
is seriously flawed and misleading.5 Davis notes that there were regional
centers of Saiva siddhanta, including Kashmir and Tamil Nadu, but he
understands these as part of a pan-Indian network for the Saiva siddhanta
school, involving groups of wandering scholars who used Sanskrit as a
lingua franca to communicate with each other. The image of Saiva sid-
dhanta is thus one of a widespread network of scholars. Davis argues
that we should understand medieval Saiva siddhanta as comprising vari-
ous lineages of scholars who accomplished the work of writing and
teaching philosophy at various scholastic "bases" around India, includ-
ing but not limited to Tamil Nadu.
Davis cites as an example the acarya (leader, teacher) Aghorasiva, who
lived and taught in twelfth-century Tamil Nadu and wrote Sanskrit texts
that are today widely influential in ritual practice in Tamil Nadu. Yet
Aghorasiva did not construct his lineage with respect to things Tamil, but
rather to north Indian figures and sites, and the texts on which he wrote
commentaries were from Kashmir. Importantly, the flow of information
worked the other way as well, since a text believed to have been written
in Kashmir appears to mention the singers of the Tamil bhakti hymns,
suggesting familiarity with distinctly southern temple formulations.6 For
Davis, the discourse that Aghorasiva both created and participated in is evi-
dence of a pan-Indian network. Supported by additional detailed study of
inscriptions and lineages, Davis concludes that "the peripatetic monastics
documented in medieval inscriptions and the authors of Sanskrit Saiva Sid-
dhanta paddhati texts saw themselves as members of a single pan-Indian
Saiva order composed of several interrelated lineages or monastic orders."7
4 M. Dhavamony, "Saivism: Siddhanta," in Encyclopedia of Religion, ed. Mircea Eliade

(New York: Harper, 1987), 13:11. Richard Davis notes that this type of definition turns up
in Pope's translation (The Tiruvacagam, p. lxxiv) and in Thomas J. Hopkins, The Hindu
Religious Tradition (Encino, Calif.: Dickenson, 1971), p. 118 (Davis, Ritual, pp. 19, 169).
5 Richard H. Davis, "Aghorasiva's Background," Journal of Oriental Research ("Dr.
S. S. Janaki Felicitation Volume") 56-62 (1986-92): 367-78. He in turn acknowledges a
debt to members of the French school of Indology, especially H6elne Brunner.
6 This is the JatinirnayapurvakalayapraveSavidhi, attributed to Ramakantha, a dualist
Saiva who probably lived in Kashmir in the eleventh century. The text is translated and
discussed by Pierre-Sylvain Filliozat, "Le Droit d'Entrer dans les Temples de Siva au XIe
Siecle," Journal Asiatique 263, facs. 1 and 2 (1975): 103-17.
7 Davis, "AghoraSiva's Background," p. 377.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
234 A Tamil Lineage
So the Sanskrit Saiva siddhantins do not appear to have constituted

their identity as localized exclusively in one place. Why then has schol-
arly consensus fixed on Tamil Nadu as the source for Saiva siddhanta
philosophy and praxis? The tendency of this consensus is to read the
widespread influence of Saiva siddhanta on contemporary philosophy
and religious praxis in Tamil Nadu into the distant past, without paus-
ing to consider either the larger historical picture or the possibility of
change within Saiva siddhanta itself. Instead of harboring a picture of
Tamil origins, we need to imagine that the Saiva siddhanta formulations
became prominent in Tamil Nadu because of the actions of agents who
made it so.
The origins of the Sanskrit Saiva siddhanta school is something of a

mystery. Detailed evidence of this ritual and philosophical tradition is
not currently available prior to the period of the tenth century or so,
when attempts were made to systematize the Sanskrit agamas, the au-
thoritative texts of the Saiva siddhanta.8 However, there are suggestions
that early on in Tamil Nadu, Saiva siddhanta became connected with
royal agents who identified themselves with the tradition. For example,
one of the Pallava kings, Mahendravarman I (ca. A.D. 610-30), says in
an inscription at Kfancipuram that he is a "follower of Saiva siddhanta,"
and he also uses the five-syllabled pancaksara mantra (nama Sivaya),
from the Vedic Satarudriya, which became identified with Saiva sid-
dhanta ritual practice, to begin some of his inscriptions.9 More solid
evidence comes from the late tenth and early eleventh centuries, when
the Chola ruler Rajaraja appears to have used the agamas both to create
and maintain his imperial temple system, beginning with the central
capital temple at Tanjavur.
The significance of the Chola ruler's use of the agamas in constructing
his temple has to do with the history of the site of Tanjavur. Early on,
Tanjavur was invested with political meaning; the first medieval Chola
ruler, Vijayalaya, is said in an inscription to have laid hold of the city
"as if it were his lawful wife," and to have built a temple to the goddess
Nisumbhasidini (Durga) there.10 In order to invest this essentially mili-
tary site with the sacrality essential to the foundation of a Siva temple,
Rajaraja I turned to the Sanskrit Saiva siddhanta agamas, which have
such ritual acts as their focus: "The primary focus of the texts ... is
clearly on religious practice. The agamas spell out in detail the organi-
zation of the temple cult, from the ritual procedures and architectural
8 On this point see Davis, Ritual, pp. 12-14.

9 See his inscriptions at Kaficipuram, in E. Hultzsch, "The Pallava Inscriptions on the
Kailasanatha Temple at Kanchipuram," South Indian Inscriptions (1890), 1:8-24.
10 K. A. Nilakanta Sastri, The Colas (Madras: Madras University Press, 1951), p. 110,
citing the Tiruvalangatu plates of Rajendra Chola.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
guidelines needed to construct and animate Saiva temples, through the

regular program of daily worship and subsidiary rites, to the much lar-
ger occasional festivals."1l Specifically, "The temple of Rajaraja reflects
the teachings of one particular dgama-based school of Saivism, the Saiva
siddhanta."12 The architectural portions of the Kamikagama may have
been developed in dialectic relationship to Chola architecture of the elev-
enth century.13 Thus, during the reign of the Cholas, it appears that the
system of the Saiva siddhanta agamas and the system of the imperial
temple were becoming mutually constitutive. Royal agent and philoso-
pher actively and purposefully joined together to establish a new reli-
gious, social, and political order that centered around the temple, and that
had a continuing dominant presence in the Tamil lands for many centu-
ries to come.
The period of Rajaraja I represents a break with earlier Chola imperial

formations in that he took greater steps to systematize the social order.
Beyond the exigencies of the sanctity of temple site and architecture, the
Sanskrit Saiva siddhanta was appealing because it, too, was moving to-
ward system.14 Traditionally, the Sanskrit-language Saiva siddhanta has
a canon of twenty-eight Saivagamas15 that were given by Siva to human
rishis, then subsequently transmitted and commented on by members of
various lineages who were in contact with each other through their jour-
neys to regional centers and their use of the Sanskrit language. Material
within the agamas and interpretations of them by philosophers suggest
that Saiva siddhanta was originally constituted as eristical to Vedic for-
mulations, although later the two competing systems became mutually
constitutive in temple worship.16 The Saiva Siddhanta agamas articulate
a world of transformative embodiment with the central praxis of nyasa,
in which the priest places a body of Siva mantras on his own body or that
of an image (especially the lingam) to prepare himself to approach the
Lord and then to invoke the Lord into the image. It is a dualist system
whose central domain is the temple and its homologue of the human
body, where the worshiper pursues his goal of "becoming a Siva."17
1 Davis, Ritual, p. 10.

12 Ibid., p. 14.
13 Possibly via the Mayamata, see ibid., p. 13.
14 Ibid., pp. 19-21, on "system."
15 Twenty-eight Muilgamas ("root" agamas); there are also 207 Upfgamas, which are
supplements to the root agamas.
16 On these developments, see N. R. Bhatt, "Development of Temple Rituals in India"
in Siva Temple and Temple Rituals, ed. S. S. Janaki (Madras: Kuppuswami Sastri Research
Institute, 1988), pp. 24-45; and Wayne Surdam, "The Vedicization of Saiva Ritual," in
Janaki, ed., Siva Temple and Temple Rituals, pp. 52-60.
17 Richard Davis identifies the Sanskrit Saiva siddhanta daily ritual activity of piij as
the practice of a "modified dualistic metaphysics" and contrasts it to the "monist schools
such as Advaita Vedfnta and more bhakti-oriented dualist schools" (Ritual, pp. 135-36).

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
236 A Tamil Lineage
The system of Sanskrit Saiva siddhanta developed and flourished in

Tamil Nadu because of the interest of agents there, not only the Saiva
siddhantins themselves, but also the royal Cholas, who ruled the region
and provided patronage for the Sanskrit Saiva siddhanta formulations,
beginning with Rajaraja I. In addition, two broad historical changes may
have solidified the connection between Tamil Nadu and Saiva siddhanta
so prominent in academic discourse on the philosophical school. One
was Sankaracarya's emphasis on the Advaita (monism) school of Saivism
(ca. eighth century) and its enormous development throughout the sub-
continent in the ensuing centuries by some of India's most renowned
philosophers.18 It is possible that the developing pan-Indian influence
of Sankaracarya's Advaita Vedanta philosophy may have cut into, or
usurped, the support base of other groups, including the Saiva siddhanta,
resulting in the latter's regionalization in Tamil Nadu and Kashmir. The
other was the Muslim incursions in north India during the medieval per-
iod and beyond, which would have at least disrupted the Saiva siddhanta
network's lines of communication. As dualists, the Saiva siddhantins had
an interest in and commitment to temples that the Advaita Vedantins did
not share; in addition, temples in north India were at times a locus for
attack by Muslim conquerors. It is possible to hypothesize that the Saiva
siddhanta, while surviving in text and community through the late me-
dieval period, became localized in response to these cultural events in
the relatively safe if increasingly isolated (from the north) region of
Tamil Nadu.
There was nothing, then, that was particularly Tamil about the early
medieval Saiva siddhanta formulations; they were not Tamil in origin
language, or outlook. However, the Tamil region came to represent a
place where Sanskrit Saiva siddhanta could be developed, in part because
of the interest of powerful Tamil agents, and in part because communities
there were stable.
Importantly, at the time the Sanskrit Saiva siddhanta became localized
in Tamil Nadu, Saiva siddhanta philosophy began to be written in the
Tamil language, and therefore the school in some sense became Tamil.
Or, perhaps more accurately, both a Tamil canon and a Tamil lineage for
18 Abhinavagupta, the famous tenth-century Kashmiri exponent of nondualism, de-

scribes the seer Durvasas's division of the Saivagamas into three classes: monism, dualism,
and monism-cum-dualism (imparted to his sons, respectively, Tryambaka, Amardaka, and
Srinatha). Although in his formulation the three schools can be traced back to Siva, the
groups were contradictory in philosophical premises, and it is this rather than their unity
that seems to have become emphasized over time. The summary of the full passage of
Abhinavagupta's Tantrdloka (verse 36) is in Davis, "Aghorasiva's Background" p. 373.
Some very interesting criticisms of Sankara's formulations occur in biographies of the
great sage himself, discussed in David N. Lorenzen, The Kapalikas and Kalamukhas: Two
Lost Saivite Sects (1972; reprint, Delhi: Motilal Banarsidass, 1991).

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
the philosophical school of Saiva siddhanta were created by Tamil phi-

losophers, many of whom were in fact fluent in Sanskrit. The writing of
Tamil Saiva siddhanta-for this was very much a philosophical, written
formulation-is traditionally considered to have begun in the mid-
twelfth century, with the Naindmirdam by Vakisar or Vakisa Munivar,
and was continued by philosophers in the Meykantar lineage for the next
200 years.19 Their corpus of fourteen texts, which was in the main (al-
though not exclusively) written by philosophers who traced their lineage to
Meykantar, himself one of the authors, became constituted as a canon,
known as the Meykanta Cattirankal.20 Though it remains an open ques-
tion when this corpus was constituted as a canon, we can suggest from the
Tamil texts that the authors wrote with a heightened sense of intertex-
tuality, which related to their ideology of the Meykanta guru lineage,21
and that they considered their texts to be authoritative. That is, the con-
stitutive texts display a collective coherence not often found in a canonical
corpus, but always insisted on when a group designates a corpus as canon.
19 See T. B. Siddalingaiah, Origin and Development of Saiva Siddhanta Up to the 14th

Century (Madurai: Madurai Kamaraj University, 1979), p. 73.
20 The title of the canon is called by one of the author's names, Meykanta(r), who wrote
one text, the Civafiana Potam, that is considered to be foundational, though not chrono-
logically first. The canonical sastras (Tamil cdttirankal) are fourteen in number. They are
(in the traditional order, which scholars take as chronological): Tiru Untiyar by Tiru-
viyalur Uyyavanta Teva Nayanar (ca. A.D. 1147), Tirukkalirruppatiyar by Tirukkatavur
Uyyavanta Teva Nayanar (ca. A.D. 1177), Civaiina Potam by Tiruvenneynallur Mey-
kanta Tevar (1221), Civafdna Cittiydr by Tirutturaiyur Arunanti Civacariyar (1253), who
also wrote Irupd Irupatu (1254), Unmai Vilakkam by Tiruvatikai Manavacakamkatantar
(1255), and eight by Umapati Civacariyar (of Korravankuti): Civappirakdcam (1306),
Tiruvarut Payan (1307), Vind Venpd (1308), Porrip Parotai (1309), Kotikkavi (1309),
Neincuvitu Titu (1311), Unmaineri Vilakkam (1312; a minority opinion by some scholars
attributes the authorship of this text to Cirkali Tattuva Natar-see below), and Cankarpa
Nirakaranam (1313; the Saka equivalent of this date is given in the text itself). These
generally widely accepted dates are taken from Dhavamony, Love of God, (n. 2 above),
pt. 2. We do not know when this corpus was constituted as a canon, or by whom. An
anonymous poem strings a short version of each title of the fourteen sastras together:
"Undikaliru uyar bfdam siddiyar, pindirupa unmai pirakasamvanda arutpanbu vina porri
kodi pasamila neficuvidu, unmaineri sankarpa murru" (quoted in Siddalingaiah, p. 72,
n. 7). Janaki states that the earliest place this poem is found to date is in the Madurai Na-
yakan's 1866 edition of these fourteen works in a single collection (S. S. Janaki, "Uma-
pathi Sivacarya, His Life, Works, and Contribution to Saivism," Saiva Siddhanta 24, no. 2
[April-June 1989]: 53-62, esp. p. 56). The issue of closure may not have been settled
early on; for example, some scholars argue that the Tukalaru Potam by Cirkali Cirram-
balanatikal replaces Unmai Vilakkam as one of the fourteen sastras. However, it is defi-
nitely a carryover from the study of the Christian New Testament that we consider closure
to occur only when the authors are dead; establishment of a canon in this sense is a second-
order act of selection performed by those other than authors. I feel that this may or may
not be the case where the Tamil Saiva siddhanta canon is concerned. Although it is beyond
the scope of my argument to fully account for this here, I will note that owing to the lin-
eage of the authors there is more internal coherence in this canon than in many others, and
that Umapati Civacaryar, the last author, was concerned with the issue of canon, as I discuss
below with respect to the Tirumurai.
21 See the discussion in n. 23.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
238 A Tamil Lineage
The writing of authoritative treatises on Saiva siddhanta philosophy in

Tamil was the act of creating a distinctive lineage of the Saiva siddhanta
school. Instead of maintaining the balance between regional centers with
a pan-Indian outlook, as had the Sanskrit Saiva siddhantins, the Tamil
Saiva siddhantins were unabashedly regional, both in terms of location,
since their primary canonical authors were all from the Chidambaram
area in Tamil Nadu,22 and in terms of the language of their philosophy,
Tamil, which is a region-specific language. As I noted above, various
cultural events may have resulted in the localization of Sanskrit Saiva
siddhanta in Tamil Nadu, which is important for understanding the de-
velopment of this school. However, they are of only marginal importance
for understanding the genesis of the Tamil Saiva siddhanta. In the first
place, the canonical texts in Tamil are not verbatim translations of the
Sanskrit texts; thus, it is not the case that Sanskrit Saiva siddhanta simply
came to be "expressed" in Tamil. The Tamil canon consists of original
texts that are not translations of the canonical texts in Sanskrit.23 In the
second place, although the Tamil texts engage many of the features of
the Sanskrit school, including philosophical concepts and the importance
of a guru lineage,24 they are also critical of the Sanskrit school, which
I explain more fully below. A more accurate explanation for the forma-
tion of a Tamil Saiva siddhanta school involves the premise that Tamils
actively made the decision to create a Tamil school for the glorification
of Siva and the works to Siva in their mother tongue.
22 According to tradition, Meykantar was born in Tiruvenkatu, near Cirkali, some

twenty kilometers from Chidambaram; Arunanti was born in Tirutturaiyur, near Panrutti,
approximately twenty-five kilometers west of Cuddalore and fifty kilometers north of
Chidambaram; and Marainana Campantar was born in Tiruppennakatam, near Viruttac-
calam, also near Chidambaram (Tiruppennakatam is connected with the story of Meykan-
tar as well; his father was born here and the acarya's samadhi took place here). Umapati
was from Chidambaram itself.
23 There was a scholarly debate on whether Meykantar's Civafina P6tam was a trans-
lation of the pasamocana pathala of the Rauravdgama, but no editions of the latter,
including the most recent, have these sutras in them. See Siddalingaiah, pp. 94-95.
24 The lineage of Tamil philosopher-seers begins with Mekantar, who tradition says
had 49 disciples. Except for the first two authors of the canon, who came before Meykan-
tar, the other authors are claimed to be among these 49 disciples. That the first two
authors wrote their treatises before Meykantar wrote his appears to have bothered later
interpreters of Tamil Saiva siddhanta. The Cantanavaralaru (eighteenth century?), a prose
work on the history of the cantandcdrydrs (leaders of the lineage: the canonical authors),
explains that the two prior authors, both named Uyyavanta Tevanayanar, belonged to the
Vifnnantevar Cantanam, which was "like" the Meykantatevar Cantanam. One day they
placed their works before him and prostrated themselves. Meykantar approved of their
work, thus legitimating them for inclusion in the sastras. See Siddalingaiah, p. 86, n. 65.
Otherwise, Meykantar was the guru, Arunanti was his preceptor-cum-disciple, Manava-
cakam was a disciple of Meykantar, and Umapati was the disciple of Marainanacampan-
tar, who had received instruction from Arunanti. A special status is given to two pairs of
guru-disciples: Meykantlr, Arunanti, Marainanacampantar, and Umapati are known as
the four cantandiicrydrs. See ibid., pp. 97-98, 108, 110.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
One of the main differences between the Tamil school and the San-
skrit school of Saiva siddhanta is that the former did not connect its phi-
losophy with a specific temple ritual tradition.25 This could suggest that
the Tamil Saiva siddhantins accepted several of the premises of the
Advaita Vedantins, a philosophy that became widely influential through-
out India. In fact many of the contemporary philosophers of Tamil Saiva
siddhanta I spoke with in Madras on a recent trip to India indicated
that this school "leans toward" Advaita philosophy.26 Although the exact
relationship between Tamil Saiva siddhanta and Advaita Vedanta can be
understood only through continuing research, it appears clear that the
Tamil school's rejection of the close connection between philosophy and
temple ritual practice as advocated by the Sanskrit school should be un-
derstood as a criticism. This is not explicitly stated in the Tamil school's
canonical texts; however, the texts' concern with pure knowledge (suddha
jidna) is clear, and may suggest that the Sanskrit school's participation
in temple ritual compromised the "purity" of the knowledge of Siva
because of the politics of both priestly practice and patronage.27
In addition to these two factors, there was also the relationship
between the Tamil Saiva siddhanta philosophers and the Tamil bhakti
hymns of the nayanmar, especially Campantar, Appar, and Cuntarar. The
authors of the Tamil Saiva siddhanta canon, especially Umapati Civa-
caryar, were interested in these bhakti hymns, an interest that I interpret
in two ways. On the one hand, the bhakti poems provided a precedent of
inspired writings on the nature of Siva in Tamil. Thus, the Tamil bhakti
tradition provided an important heritage for the Tamil Saiva siddhantins,
and it became constituted as part of their collective lineage. The Tamil
bhakti hymns proved that questions of universal import (e.g., Who is
God? How do we worship Him?) could be asked and answered in the re-
gional language of Tamil. On the other hand, the bhakti poems projected
a path of religiosity that the Tamil school found more compelling than
the complicated temple ritual worship of the Sanskrit Saiva siddhantins.
The Tamil bhakti tradition may have become institutionalized as part of
ritual worship in the Chola temples, but the bhakti hymns themselves
did not propagate a specific form of temple ritual worship. The Tamil phi-
losophers simultaneously linked yet distinguished themselves from these
practitioners of religion; the philosophers approved of their religious
25 This has been briefly noted by Davis (Ritual [n. 3 above], p. 18).
26 I thank the Fulbright organization for making this yearlong trip for the purposes of
completing my dissertation possible. On this point see also J. M. Nallasvami Pillai,
Studies in Saiva-siddhanta (Madras: Meykandan, 1911), esp. pp. 244-72; and Rohan A.
Dunuwila, Saiva Siddhanta Theology. A Context for Hindu-Christian Dialogue (Delhi:
Motilal Banarsidass, 1985), esp. pp. 66-75.
27 For another statement on the purity of the Tamil school, see the anonymous poem I
cite later in this article.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
240 A Tamil Lineage
content, but they strove to identify themselves as philosophers, not re-

ligionists. This tension between religion and philosophy was ultimately
crystallized into the Tamil titles28 given to each group, reflecting their
distinctive domains. The philosophical authors are known as cantan-
acaryas, where cantanam means "progeny" or "succession," and acarya
means "leader"; and the bhakti poets are known as camayacaryas, where
camayam means "religion." Yet the titles also connected the two groups
as Tamil acaryas.
The Tamil school thus not only had its own canon, language, and guru
lineage; it also had its own religion, all of which distinguished it from
the Sanskrit Saiva siddhanta. In order to discuss these features in more
detail, I turn to the figure of Umapati Civacaryar.
UMAPATI CIVACARYAR AS AUTHOR AND GURU
Umapati Civacaryar was the most prolific of the authors of Saiva

dhanta canonical texts in Tamil; he is credited with eight of the four
sastras.29 In several of these texts, he locates himself within a lin
that extends back into cosmic time and forward into his own histori
time. Calling his various predecessors "the ones who are by nat
leaders of the cantdnam [lineage]," he lists Tirunanti, Cattiyafinn, Ta
canikal, Parafic6ti, Meykanta, Arunanti, and Maraiiiana Campanta
Tradition has divided these eight acaryas (including Umapati) into
groups of four, according to the division of interior (akam) and exte
(puram) known from classical Tamil grammar and poetry. The first f
are called akaccantdnam, suggesting that they are symbolically in
the center of tradition. Mythically, these preceptors are not of
world; they reside at the abode of Siva, Mount Kailasa. The latter
are called puraccantdnam, the lineage of gurus that carries the inspir
teachings out into the world and radiates them among the people. Tra
dition says that Meykantar and Arunanti lived just prior to Umap
while Maraiinana Campantar was Umapati's living guru. The abod
these mortal gurus was the region of the famous temple-town, Chida
baram, in Tamil Nadu.31
According to the hagiographies about Umapati, he was a temple pri
at Chidambaram, among the group of hereditary Tiksitars (ones who
initiated),32 who alone had and have the right to perform worship in
28 These titles are actually from Sanskrit, but they are transliterated into Tamil.
29 For a list of these fourteen texts, see n. 19.
30 Civappirakacam, stanza 5, in Vai. Irattinacapapati, Meykanta Cattirankal (Ce
[Madras]: Cennaip Palkalaik Kalakam, Caiva Cittantat Turai [Madras University, T
Department], [1988]), p. 190. Unless otherwise specified, the Tamil canonical tex
consider in this article are from Ratnasabapathy's (Irattinacapapati's) volume.
31 See n. 22.
32 The term Tiksitar derived from Sanskrit, is often Anglicized to Deekshithar.

130.63.63.214 on Sat, 14 Jan 20hu, 01 Jan 1976 12:34:56 UTC
temple, although we do not know his Tiksitar name.33 The identity of the
Tiksitars is a matter of some scholarly uncertainty. Although Chidam-
baram has been the focus of many scholarly books and articles that
describe the mythology of the Dancing Siva, the temple's art and
architecture, and the lore surrounding this renowned temple town, only
recently has there been a detailed attempt to reconstruct its very com-
plicated history.34 A cornerstone of the temple's history is the origins
and identity of the family of priests that apparently has been in contin-
uous control of the ritual activities therein. It is clear from both mytho-
historical legends and discussion with present-day priests there that the
Chidambaram priests distinguish themselves from other priests in Tamil
Nadu temples. They seek to differentiate themselves in several ways: by
their ethnicity (as all members of a select group of interrelated families
that are descendants of the original priests at the temple); by the unique-
ness of the Dancing Siva as their temple's main image; by their distinc-
tive traditions of worship, as opposed to the traditions of the majority of
other Saiva temples, which follow the Sanskrit Saiva siddhanta tradition;
and by the fact that their temple, unlike all others in Tamil Nadu, is not
under government control. Thus, the priests present Chidambaram as a
locus for traditions independent of the agamas.
Significantly, both Chidambaram and the priests there are praised in
Tamil bhakti literature. For example, eleven hymns of the nayanmar,
eight of which were sung by Appar,35 celebrate Chidambaram under
the names of Tillai36 or simply Koyil (temple). It is one of the three
dozen or so towns in the Tamil lands about which the nayanmar sang
five or more hymns each.37 The nayanmar's attention to Chidambaram
33 Janaki has compiled a list of the sources for his biography, which are in Tamil and
Sanskrit: Parthavana mahatmyam (Korravangudi Puranam), in 240 Sanskrit verses, also
found in the Chidambarasara as a dialogue between Brahmananda yati and Sankaracarya;
Rajendrapura (Tillai) mahatmya or Umapati-vijaya, 108 Sanskrit verses by Tillai Siva-
nanda Diksita (both of these Sanskrit sources are found in the introduction to the
Pauskara bhdsya, one of Umapati's Sanskrit works); a fifteen-verse biography by Chidam-
bara brahma yati, found in Umapati's Kuicitiighristava; and the Cantanacarya Purana by
Svaminatha Dikcitar. This last work, in Tamil, is from the eighteenth century. See Janaki,
"Umapathi Sivacarya" (n. 20 above), p. 54. Siddalingaiah (n. 19 above) adds a few Tamil
sources, the Pulavarpuranam, and the invocatory poems by various later poets, e.g., the
Endtkanni of Tayumanavar (p. 124).
34 This is Paul Younger's book-length manuscript, "The Home of the Dancing Sivan:
The Traditions of the Hindu Temple in Citamparam." I am very grateful to Dr. Younger
for sharing his fascinating and important study with me.
35 Hymns 1.80, 3.1 (Campantar); 4.22, 4.23, 4.80. 4.81, 5.1, 5.2, 6.1, 6.2 (Appar); 7.90
(Cuntarar, who refers to the place as puliyur, "tiger-town").
36 Called as such perhaps because there was a forest of Tillai trees there at one time;
in general, temples in south India are associated with sacred trees.
37 For a list of such important places in Tevaram, see Indira Peterson, Poems to Siva:
The Hymns of the Tamil Saints (Princeton, N.J.: Princeton University Press, 1989), 339-40.
The total number of places sung about by the nayanmar is 274.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
242 A Tamil Lineage
was shared by other authors of various viewpoints whose Tamil works

were eventually incorporated into the canon of Tamil bhakti literature
to Siva, known as the Tirumurai.38
The hymns of Campantar, Appar, and Cuntarar also praise the tillai
muvdyiravar, the 3,000 Brahmins of Tillai. For example, the poet Cun-
tarar began his list of bhaktas, the "Tiruttontattokai," with mention of
the antanar (Brahmins) of Tillai: "I am the servant to the servants of
the Brahmins who live at Tillai."39 A short verse by Nampi Antar Nampi
elaborates on this line of poetry mainly by giving a description of Siva.40
The Periya Puranam gives us some more information on the priests, em-
phasizing that they serve the Dancing Siva by their knowledge of Vedic
scripture and practice. For example, they maintain the three fires in the
established way so that souls will be filled with grace; they are very
learned in the four eternal Vedas as well as the six angas; and they per-
form the six deeds (learning and teaching scripture, understanding and
performing sacrifice, and giving and receiving gifts).41 The author of the
Periya Puranam, Cekkilar, even reflects that there is nothing greater that
he can say about the Tillai Brahmins beyond the fact that Cuntarar
placed them first in his poem.
However, the impression one gets of the Brahmins of Tillai from
Saiva bhakti literature is not only of praise. For example, in the Periya
Puranam story of Nantanar (Tirunalaipp6var), the Brahmins of Tillai
obstruct bhakti worship at the famous temple to the Dancing Siva.42 In
that story, the Brahmins of Tillai were priests who inspired fear in the
untouchable saint, Nantanar, so intimidating him that he dared not enter
the temple although his desire to see Siva was overwhelming. Eventually
38 Including Manikkavacakar, in his Tiruvacakam, esp. 1, 2, 31, 40 (the eighth book of

the Tirumurai); Gandaraditya, the husband of Sembiyan Mahadevi, in his Tiruvicaippd
(included in the ninth book of the Tirumurai); and Tirumular, who discusses Chidam-
baram for some eighty verses (2723 ff.) in his Tirumantiram (tenth book of the Tiru-
murai). These passages are selectively included in paraphrase in Younger's "Home of the
Dancing Sivan," story 4.
39 Tillaivalantanartam atiyarkku matiyen, 7.39. Tradition, beginning with Nampi Ant.ar
Nampi, understands the brahmins themselves to be the servants of Siva; thus, the meaning
of the phrase is in effect, "I am the servant of the servants who are the Brahmins who live
at Tillai." See the following note.
40 I translate: "I will tell about those who perform hereditary service to the One of
coral hue who bears the crescent moon, to the One of the ambrosial sacred dance, to the
Father who burned the cities of the heavenly asuras and the three worlds, who is enjoyed
by those who know the excellent Vedas at the place of Tillai, whose fame is fit to
declare" Tiruttontar Tiruvantati (1963; reprint, Tarumai: Tarumai Atinam, 1986).
41 Cekkilar's Tillaivalantanar Puranam, verses 5 and 6, in Periya Puranam ena valankum
Tiruttontar Pura.nam, ed. Civappirakacatecika Paramacariya Cuvamikal (Tiruvavatu-
turai: Tiruvavatuturai Atinam, 1988), p. 46.
42 On this issue see my article, "The Story of Nantanar as a Clue to Changes in the Order
of Things," in "Untouchable Saints: An Indian Phenomenon," ed. Eleanor Zelliot (unpub-
lished manuscript).

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
the Brahmins were compelled by the order of Siva himself to circumvent

the tradition of prohibiting an untouchable from entering the temple.
Even though they did follow Siva's command, the Tillai Brahmins in this
story represent an obstacle to, not representatives of, bhakti.43
The ambivalence towards the Brahmin priests at Chidambaram char-
acteristic of the Saiva bhakti literature is also found in some of the texts
Umapati wrote in Tamil.44
Umapati's history of the Chidambaram temple, the K6yil Puranam, is
one of the most important sources for the history of the Chidambaram
temple in general, and the Tiksitar priests there in particular.45 His Tamil
text has four of its six sections in common with the Sanskrit Citambara-
mahatmya,46 perhaps of the thirteenth century, but even within these
sections there are differences that reflect Umapati's perspective and ren-
der the text as his own composition. For example, although the texts
have in common many of the stories about Viyakirapata, a legendary as-
cetic who settled in Tillai, and about Pataficali, a snake devotee who
watched Siva's dance,47 they differ in their stories of Hiranyavarman, a
legendary king who settled in Tillai, especially in the second half of this
story, which focuses on the origins of the priests.
43 1 understand that there is also an episode in Cekki!lr's story of Campantar in which

the saint has some reluctance to go to Chidambaram, but then he sees a band of Siva's
ganas (dwarflike bhaktas) and is convinced to go.
44 It is fairly certain that Umapati authored the Koyil Puranam, the Tirumuraikanta
Puranam, and the Cekkilar Puranam, all of which I discuss in some detail in this section
and the next. In addition to the latter two, there are several other texts attributed to Uma-
pati that focus on Saiva bhakti, including: the Civatcetra civandmak kalivenpd, a text that
tells in 300 couplets of the 274 pilgrimage sites visited by the three most famous nayan-
mar (Campantar, Appar, and Cuntarar), and names the presiding form of Siva found in
each; similarly the small texts of Tiruppatikkovai and Tiruppatikakovai enumerate the
pilgrimage sites; and the Tirunatcattirakovai that tells of the piranta natcattirankal (birth
star days, which are occasions for festivals) of the nayanmar and those of the cantanac-
drydrs (including Umapati's). Umapati is also credited with a distillation of Cekkilar's
Periya Puranam, the Tirttontarpuranacdram (Essence of the stories of the holy servants).
Scholars dispute Umapati's authorship for these five texts, but it remains significant that
tradition closely associates Umapati with descriptions of nayanmar bhakti.
45 Umapati's Koyil Puranam, verse 381. For this parallel and the following discussion
of this text I rely on Younger's excellent analysis in "The Home of the Dancing Sivan,"
story 3. Younger notes that festivals are more conservative than priestly worship; the fes-
tival section of the text describes many festivals that are performed today, in contrast to
the section on daily worship, which is quite different from that practiced today. However,
the festival involving the procession of Pataficali is described in the text but not per-
formed today. Further, this procession is not mentioned in the Sanskrit sthalapurana of
Chidambaram, the Citambaramdhitmya.
46 A comparison of the two texts by section theme is as follows: introduction-Citam-
baramahatmya (C) sections 1-4, K6yil Purana (KP) verses 1-29; on Viyakirapata-C
6-10, KP 30-59; on Patanjali-C 11-18, KP 60-158; on Natarajan-KP 159-229; on
Hiranyavarman-C 19-26, KP 229-362; on festivals-KP 363-415. See Younger, "The
Home of the Dancing Sivan," story 3.
47 These names are of course familiar from north Indian myths, but in the stories of
Chidambaram their identities are given a different content. See ibid.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
244 A Tamil Lineage
There are at least three stories of the priests' origins within the
Hiranyavarman sections of the respective texts. Paul Younger has sug-
gested that the priests themselves may have edited these sections and
were concerned about how they would read, since the stories concerned
their own origins and thus legitimacy. The Tamil and the Sanskrit texts
have two of the stories in common. One story has the priestly group as
the Muvayiravar (three thousand) accompanying King Hiranyavarman
on his journey from his hereditary kingdom in northeast India down to
Tillai in the south to establish his new kingdom. The other story locates
the priests as hailing from the Antarvedi region in north central India,
where they performed elaborate Vedic sacrifices. The king convinced
them to accompany him as he journeyed south. As Younger notes, the
Antarvedi region of India (between the Ganges and Yamuna Rivers) is
known from many stories as the place from which the highest-status
Brahmins hail. This story thus emphasizes the high status of the priests
and their autonomous existence as tenders of the sacred sacrifice.48
The third story is found only in Umapati's text. This story places the
other two stories in a new chronological sequence. Here, the priests were
originally at Tillai, where they witnessed Siva's sacred dance. Then they
were requested to come to Antarvedi to perform one of the greatest
sacrifices of all time. Afterwards, they were brought back to Chidam-
baram by king Hiranyavarman, who was traveling south to establish his
new kingdom. When they reached Chidambaram, a count was taken of
the priests and it was discovered that there were only 2,999 of them. At
this point, Siva's voice was heard from the heavens announcing that he
was one among them, thus making the total of 3,000. The effect of Uma-
pati's telling is to place the other stories in the frame of direct contact
with Siva, which determines the priests' origins and identity. The sac-
rifice at Antarvedi suggests the priests' competence in their performance
of Vedic sacrifices, an aspect of the story that confers a high value on
them as practitioners in the world, and presents this competence as a
complement to their divine darshan of Siva's dance. The inclusion of
the story in which they accompany Hiranyavarman gets them back to
Chidambaram, and sets the scene for another example of the priests' con-
tact with Siva. In Umapati's text, the priests participate in Siva in the
known bhakti ways of darshan and proximity to God. They are the bhak-
tas as claimed by earlier bhakti literature.
However, later in the text, in the section on Natarajan, Umapati draws
a distinction between the priests as bhaktas and the significance of Siva's
dance. At the dance, the priests are observers, not practitioners. They,
along with the sun and the moon, sages, Viyakirapata and Pataicali, and
48 Younger suggests that the claims in this story were made in the thirteenth-century
context of many north Indian pilgrims coming to Chidambaram and asking questions
concerning the legitimacy and authority of the priests; ibid.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
the Goddess, watch the dance, which Siva performs as a boon to Pataiicali.
They do not play the role of interpreting the dance. This task Umapati
takes on himself, describing Siva's dance at length using terminology
from Saiva siddhanta philosophy, such as arul (grace), pati (lord), pasu/
uyir (soul), and pdsa/mala (bond, fetter). In several verses Umapati puts
the philosophy of Saiva siddhanta into the mouth of Siva as teachings to
the assembled multitude.49 The distinction is that the priests are praised
as bhaktas, having the right participation in Siva; but it is Saiva sid-
dhanta that provides the right knowledge, and in Umapati's text the
priests are not proponents of this right knowledge.50
The suggestion of distance between the views of Umapati and the
Tiksitar priests found in the author's work is given concrete expression
in the hagiographical literature about him, for this literature describes a
break in his association with them. Umapati, who was a Tiksitar, left
the group of his fellow priests in order to follow the Saiva siddhanta
guru, Maraiinanacampantar. Information about this guru is obtained from
Umapati's texts; apparently his guru left no written philosophical works.51
We have already noted that Maraiinanacampantar was a disciple of Arun-
anti, and is part of the puraccantanam line of gurus in Tamil Saiva sid-
dhanta. He was a Brahmin, and well versed in the Vedas (marai). After
his initiation from Arunanti at Katantai (Tiruppennakatam), he went to
Chidambaram, worshiped Lord Nataraja, and then settled in Tiruk-
kalaniceri, where tradition believes he eventually attained release. One
day Umapati, who had finished worship at the Chidambaram temple and
was riding home in a palanquin, passed by Marainanacampantar on the
road. The latter remarked, "The blind by day is riding on dead wood,"
referring to the palanquin and the torch customarily on the front of the
vehicle.52 Recognizing the stranger as his guru, Umapati immediately
got down from the palanquin and began to follow him.
Maraifanacampantar is said to have wanted to test the "ripeness" of
Umapati, that is, the readiness of his soul to be granted initiation into
49 For example, Koyil Puranam 138 ff.; discussed in Younger, "The Home of the Danc-
ing Sivan."
50 The distinction between priestly activity and Saiva siddhanta knowledge is also
found in tradition surrounding Umapati's canonical text, the Kotikkavi (The flag-song).
Tradition says that on one occasion, the temple flag, which is raised to commence impor-
tant festivals, would not ascend. It was only when Umapati was brought to the temple to
deliver a minidiscourse on Saiva siddhanta that the flag went up. The discourse was the
Kotikkavi, in four verses: the first verse describes the darkness of ignorance and the light
of knowledge; the second describes the realities of God and the soul; the third "explains
the 'advaita' relationship of God with soul"; and the fourth is on the recitation of the pan-
caksara mantra in the three ways of five-syllabled, six-syllabled, and eight-syllabled. See
Siddalingaiah (n. 19 above), pp. 125, 146-48.
51 Some scholars believe he is the author of Satamanikkovai, but Siddalingaiah (pp. 120-
24) disagrees.
52 Patta kattaiyir pakarkurutu ekutu parir; see ibid., p. 124.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
246 A Tamil Lineage
Saiva siddhanta. Therefore, he went down a lane populated by the low-

caste weavers and drank from the rice water that was used to starch the
warp. In a traditional show of subservience, Umapati drank the water that
ran down the guru's forearm. However, although Maraiinanacampantar
was pleased, the Tiksitar community was not, because it perceived
Maraiinanacampantar, though a Brahmin, to be of a lower-caste status
than themselves since he was not of their group.53 As a result of his in-
gestion of water from the hand of a lower-caste Brahmin, Umapati was
ostracized by the Tiksitars, and he took up residence at Korravankuti on
the outskirts of Chidambaram.54
His break with the priests suggests that Umapati was willing to aban-
don the idea that temple worship was the most efficacious way to reach
Siva. His rejection is supported by formulations in his non-sastric texts,
including the omission of the priestly role in his Tamil Koyil Puranam,
discussed above, and his famous Sataratnasahgraha in Sanskrit, which
is a compilation of slokas from well-known Sanskrit Saiva siddhanta
agamas, such as the Svayambhu, Mrgendra, and Kirana, that carefully
avoids passages that would contextualize the knowledge in temple praxis.
His sastras are in a number of literary genres, two of which underscore
the point that knowledge comes through the relationship between guru
and disciple, for example, the Vina venpa (vind is "question" in Tamil;
Umapati asks questions directed to his guru, Maraifiina), and the Nencu
vitu tutu (tutu is "message"; Umapati imagines himself and his teacher
as beloved and lover, and sends a heartfelt message to Marainana). Other
sastras in the larger corpus of Tamil Saiva siddhanta also highlight the
guru-disciple relationship, including the link between the texts of Mey-
kantar, Arunanti, and Umapati,55 and other examples of the question-and-
answer format (prasna-uttara).56
53 Many have written about the various classes of Brahmins and their relationship to
the non-Brahmin Vellala caste; see the discussion in Peterson (n. 36 above), pp. 44-47,
esp. n. 81.
54 A matam (some people referred to it as a temple) is there today, a small building
housing an image of Umapati that is washed and decorated each day by a Viragaiva care-
taker. According to the caretaker, the Virasaivas have "been in control" of the matam for
many years.
55 Meykantar's text, the Civahdna Potam, is the source; Arunanti's text, Civandna Cit-
tiyir, says that it is related to Meykantar's text (verse 3), and tradition considers it a
valinul "derived text" (val in general meaning "way" in this context, "son"); Umapati's
Civappirakdcam is possibly organized in accordance with the twelve sutras of Meykan-
tar's text (see Siddalingaiah, p. 135, n. 2), and tradition considers it cdrpunul, a "supple-
mentary treatise" (cdrpu meaning "reliance" or "connection").
56 For example, Arunanti's Irupa Irupatu is in this form, as is Tiruvatikai Manavaca-
kamkatantar's Unmai Vilakkam. Another noted genre in the Tamil sastras is pirva-paksa
siddhanta, in which rival schools are refuted, e.g., Arunanti's Civanidna Cittiydr, of which
the first half, the parapakkam (Sanskrit parapaksam), refutes the tenets of other schools,
while the second half, the cupakkam (Sanskrit svapaksam), expounds the Saiva siddhanta
perspective; see also Umapati's Cahkarpa Nirakdranam.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
Although Umapati expressed a reluctance to link the Tamil Saiva sid-

dhanta idea of knowledge with a specific temple tradition, this does not
mean that he rejected temple praxis. In all three categories of his works,
Tamil sastras, Tamil non-sastras, and Sanskrit works (non-sastras), Uma-
pati can be seen to have a continuing concern for the Chidambaram
temple as a locus of the Dancing Siva; one of his sastras (the Kotikkavi)
is linked to a "miracle story" of his raising the flag at this temple, his
Koyil Puranam gives the history of this temple, and one of his Sanskrit
texts (Kuncitanghristava) gives a philosophical interpretation of the im-
age and temple, and may additionally be evidence for donations he made
at the temple.57 However, the idea of philosophical knowledge predomi-
nated in Umapati's sastraic texts and in the larger Tamil Saiva siddhanta
corpus of sastras as the most efficacious path to Siva, in the process hier-
archizing other ways of reaching the Lord. In this context, Umapati re-
jected membership in a family of priests in favor of membership in a
"family" of teachers.58
Instead of the temple, tradition associates Umapati with another type
of institutional organization. Monastic educational centers that were un-
der the leadership of a guru (known as matam; often translated as mutt)
were established in the Chidambaram area probably around the fifteenth
century.59 At least two of these centers trace their lineage to Maraiiinna
Campantar and to Umapati; for this reason, later tradition considered
Umapati to be a link between the puraccantanam and the institutional
phase of Saiva siddhanta, called the abhiseka paramparai (lineage of the
initiated).60
UMAPATI CIVACARYA AND TAMIL SAIVA BHAKTI
Earlier in this article I noted that the Tamil Saiva siddhantins dis
themselves from the Tamil bhakti saints by the categories of
57 Umapati is credited with several "miracles" at Chidambaram besides the ra

the festival flag, including the presence of Siva Nataraja at Korravankuti inste
temple at Chidambaram; the initiation of a low-caste man, Perran Campan, into
est initiation, the Nirvana Diksa; granting liberation to a plant; and controlling
at the Kalpaka Vinayaka temple at the Chidambaram temple complex's west
On the possibility that Umfpati made a donation to the temple, see David Sm
Dance of Siva as Portrayed in the Kuficitanghristava of Umapati," Journal o
Research (Madras) 56-62 (1986-92): 154-61.
58 Umapati praises Siva Nataraja as a teacher in his sastra Porrip Parotai (Song of
praises).
59 For information on three Tamil Saiva siddh8nta matams, Tiruvavatuturai matam,
which traces its lineage to Umapati, the Tarumapuram matam, and Tiruppanantal matam,
see K. Nambi Arooran, "The Changing Role of Three Saiva Maths in Tanjore District
from the Beginning of the 20th Century," in Changing South Asia: Religion and Society,
ed. Kenneth Ballhatchet and David Taylor (London: Asian Research Service for the Cen-
tre of South Asian Studies at SOAS, 1984), pp. 51-58.
60 The lineage of the Tiruvavatuturai Atinam is traced to Umapati; that of the Dhar-
mapuram Atinam to his guru. See Siddalingaiah, p. 119. See also Mutt and Temples
(Dharmapuram: Dharmapuram Adhinam, 1981).

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
248 A Tamil Lineage
(cantanam) and religion (camayam), respectively. I suggested that the

Tamil Saiva siddhanta's break with the Sanskrit Saiva siddhanta was in-
formed by this division, such that the Tamil school did not associate
itself with any specific religious praxis in the tradition of the imperial
temples. In this section, I consider the window of possibility for overlap
between Saiva siddhanta philosophy and the bhakti hymns of the nayan-
mar, as formulated by Umapati Civacarya.
Part of Umapati's interest in Tamil Saiva bhakti was to chronicle the
development of the hymns and the stories of the lives of the nayanmar.
Two texts attributed to him, the Tirumuraikanta Puranam61 and the Cek-
kilar Puranam, are devoted to these ends, respectively. Significantly, both
of these texts set the development of Saiva bhakti in Chidambaram, the
locus for the creation of the Tamil Saiva siddhanta. The nayanmar may
have sung their hymns throughout the Tamil lands, but according to
Umapati the hymns were subsequently "discovered" and set to music at
Chidambaram. Cekkilar is said to have written his Periya Puranam
within the Dancing Siva temple complex; legend remembers this as tak-
ing place within the thousand-pillared hall. So the first feature to note
about Umapati's chronicles is that they suggest there was a precedent for
developing Tamil religious literature at Chidambaram.
Otherwise, the stories describe two important events in the evolution
of Tamil bhakti: the discovery of manuscripts of the nayanmar's hymns,
and the completion of the Periya Puranam. According to the Tirumurai-
kanta Puranam, a king, named Rajarajamannan apayakulacekaran in the
text,62 was emotionally moved ("melting" and "his hairs standing upon
end") when he heard the hymns being sung to Lord Siva at Tiruvarur.63
The text says that since the texts of the hymns could not be found
61 This text can be found printed in some editions of the Periya Puranam. I have used
the text in Arumukat Tampiran Cuvamikal, ed. and commentator, Cekkilarcuvdmikalaru-
!icceyta Tiruttontarpuranamennum Periyapurdnam (Cennai [Madras]: Pa. Comaiyaravar-
kalatu, 1888).
62 Rajaraja is "king of kings," mannan is "king," and apayakulacekaran is "fearless
leader of the clan." All scholars believe that this was a Chola king, but they differ in their
identification of which Chola king. Kamil Zvelebil identifies this king as Kulottunka I
(A.D. 1070-1122), and places Nampi Antar Nampi in the period of 1080-1100 in Tamil
Literature, in Handbuch der Orientalistik, Zweite Abteilung, Indien, ed. Jan Gonda, Band
2, Literatur und Biihne, Abschnitt 1 (Leiden and Cologne: Brill, 1975), p. 134. Compare
Dorai Rangaswamy, who speaks of the possibility that the king was either Rajaraja I or
Kulottunka I, in The Religion and Philosophy of Tevdram, with Special Reference to
Nampi Arurar (Sundarar), 2d ed. (Madras: University of Madras, 1990), pp. 22-23.
63 Verse 1 of the Tirumuraikanta Puranam, where both Arur and the name of the form
of Siva there, Tiyakecar, are mentioned. Note that Tiruvarur is especially sacred to the
bhakti poet-saint, Cuntarar: "This town becomes Cuntarar's home and a fixed point in
all the stories about him; his restless wanderings always, eventually, bring him back to
Tiruvarur (note that he also bears the name of the god of this town, as Nampi Aruran)"
(David Dean Shulman, Songs of the Harsh Devotee: The Tevdram of Cuntaramurttinay-
anar [Philadelphia: Department of South Asian Regional Studies, University of Pennsyl-
vania, 1990], p. xxviii). It was at Tiruvarir that Cuntarar composed his list of saints, the

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
anywhere, at the moment of the king's joy an Aticaiva (Adisaiva)

Brahmin, as precious and unique as a ruby, descended from heaven (and
was born on earth). This boy was Nampi, known subsequently as
Nampi Antar Nampi, and he was eventually commissioned by the king
to retrieve the lost hymns.64 Nampi prayed to his istadevatai (chosen
deity), Ganapati, in the form of Pollappillaiyar of Tirunaraiyur. This
Lord revealed to him that the hymns were locked in a room at the
Chidambaram temple.65 The story imagines a reluctance on the part of
the priests, who did not immediately open up the sealed chamber, but
instead required that the hymnists be present before they would do so.
Nampi again prayed to Ganapati, who gave him the idea to bring the
images of the mivar (the three most famous poets, Campantar, Appar,
and Cuntarar) before the room. Then the priests opened the room, in the
presence of the miivr, Nampi, and the king, and "manuscripts" of the
hymns were revealed. Many of the pages had been destroyed by white
ants, but a divine voice told the king not to worry, what was available
was all that was necessary, and Nampi began his organizational tasks.66
The Cekkilar Puranam describes a ceremony marking the completion
of the Periya Puranam. This ceremony is a festival scene, in which many
agents participate, to celebrate Cekkilar's completion of the Periya
Puranam. During the ceremony, the text was placed on the back of an
elephant and paraded around the temple environs. It has been sug-
gested that this was a canonization ceremony, but the event may also be
conceptually linked to a festival described in Umapati's Koyil Puranam.
In Umapati's K6yil Puranam, Patancali's grammar is carried in pro-
cession at Chidambaram in the context of a coronation ceremony. As
the story goes, the legendary king Hiranyavarman was married and
crowned in the Chidambaram temple. Subsequently he provided for the
ritual needs of the temple and built homes for the 3,000 priests. Then
he was taken in procession on the back of an elephant, he anointed an
image of the god, and last, he had the grammar of Patancali (Patanijali)
taken in procession on the back of an elephant. This procession of Patafi-
cali's grammar, mentioned only in Umapati's text, may be intended as a
classical precedent to symbolize the authority conferred on the Periya
Puranam. According to the Cekkilar Puranam, upon its completion the
"Tiruttontattokai." At that time, Cuntarar had just returned from a pilgrimage to Chi-
dambaram. The first line of the poem, which was provided by Siva for the tongue-tied
Cuntarar, is in praise of the priests of Chidambaram, as I discussed above. See Shulman,
p. xxix.
64 For a summary of this story in English, see ibid., pp. xix-xx.
65 There is a room today at the temple that the priests say has held the hymns since
medieval times.
66 This text is also viewed as a classical authority on the number of patikams composed
by the three hymnists, although I do not take up this issue here; see Dorai Rangaswamy,
pp. 36-41.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
250 A Tamil Lineage
Periya Puranam was officially read out to an assembly of important

persons, including the king, at Chidambaram;67 the king, priests, and
festival-goers celebrated the Periya Puranam as though it were one of
the most important texts in Indian culture.
Umapati's attention to these critical events in the "life" of nayanmar
bhakti suggests to me more than a passing interest in Saiva bhakti. His
was, I would argue, the successful attempt to create a canon of Tamil
devotional literature to Siva, which we now know as the Tirumurai.68
The Tirumurai has twelve books; it begins with the hymns of Campan-
tar and ends with the Periya Puranam. This format appears to follow an
early trajectory toward canonization; for example, the hagiographical
texts on the nayanmar by Nampi Antar Nampi and Cekkilar follow the
order of saints listed in one of Cuntarar's hymns, thus "canonizing" the
earlier list. Yet the coherence of the beginning and the end of the Tiru-
murai can be misleading; books 8-11, coming after the hymns of the
nayanmar and before the Periya Puranam, are actually a collection of
quite obviously diverse texts, written by different agents, probably in
different times, and in different literary genres. How is it that these texts
became bound together, both conceptually and materially, into a canon?
The term tirumurai may have originally referred to the collection of
nayanmar hymns that we now know as Tevaram (i.e., the first seven
books of the Tirumurai). It is a term that was used early on by two
influential agents who developed the nayanmar hymns: Rajaraja I, in
an inscription, and Cekkilar, in his Periya Puranam.69 Later, Umapati
Civacaryar wrote the Tirumuraikanta Puranam, which purports to de-
scribe, as its name suggests, the compilation of the Tirumurai. Uma-
pati's purana begins with a section of twenty-four verses in lines of
eight feet.70 In this section, Umapati attributes the compilation of seven
books (i.e., the first seven of the Tirumurai, which is the Tevaram) to
67 Cekkilar Puranam, verse 96, cited in ibid., p. 20.

68 I will make this argument with some caution because I realize it is at variance with
the two most influential studies of Tevaram (and Tirumurai) history, Dorai Rangaswamy,
Religion and Philosophy, esp. pp. 19-35; and K. Vellaivaranan, Pannirutirumurai Varaldru,
2 vols. (Annamalai: Annamalaip Palkalaikkalakam, 1962), esp. 1:1-48. The basic differ-
ence in our positions is that they view Umapati Civacaryar's Tirumuraikanta Puranam as
a description of earlier times, especially the time of Nampi Antar Nampi, whereas I view
the text as a history of its own period, the early fourteenth century.
69 Rajaraja uses the term in an inscription (158/1911) that "records a gift of land to the
'Tirumurait tevarac celvan matam' on the northern side of the Tiruttontisvara mutaiya
Nayanar temple at Tirukkalumalam by the residents of Muniyur." Dorai Rangaswamy offers
several interpretations of the passage, all of which involve the idea of "private worship,"
which is his interpretation of the term tevaram. To me it means the donation was given to
the "mutt belonging to the God of the sacred collection of tevdram (i.e., nayanmar
hymns)." It is a way of describing Siva as the god celebrated by the hymns, in the con-
text of the larger temple complex that celebrates the actions of the holy devotees (tirut-
tontar). See Dorai Rangaswamy, p. 29. Cekkilar uses the term in the Kananatar Puranam,
verse 3 (ibid., p. 27).
70 This is the same meter Umapati uses in his Cekkilar Puranam.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
Nampi, under the commission of the king; he explains the significance

of the number seven as representing the seven distinctive ways that
mantras can end. The next section of the text is in twenty-one verses in
a different meter. It is this latter section that mentions other works by
Saivas and describes their compilation by many (although unspecified)
persons. However, Nampi Antar Nampi is said to have compiled the
eleventh book (in which his own works are found) at the request of the
king. In addition, the text specifies that the hymns were set to music by
a female descendant of Tirunilakanta Yalppanar, who had been a con-
temporary of the poet, Campantar, and had set his hymns to music.
For some scholars, the fact that the text comprises two sections in two
distinctive meters signifies authorship by more than one person, with the
second section later than the first. I quote the influential scholar M. A.
Dorai Rangaswamy on this opinion: "The twenty-sixth verse, at once
abruptly starting to mention in the most summary way the other hymns
and poems of other Saivite Saints and poets without any explanation
about them, comes as a surprise.... One may not be wrong in believing
that this part was a later addition and that the original Tirumuraikanta
Puranam must have closed with the first twenty-four verses."71 But an-
other possibility exists. I agree with Dorai Rangaswamy that tirumurai
probably initially represented the hymns of the nayanmar;72 however, I
believe that this changed with Umapati. It may be that the different
meters were employed in the text with a purpose, mirroring the nature
of two distinctive halves that became the canon.73 The first half of the
canon is musical, and the first half of Umapati's text reflects the origin
of Tevaram as song, and its development as a liturgy in Pallava and
Chola temples. By contrast, the second half of the canon comprises sun-
dry texts that were neither sung nor composed by the three most famous
nayanmar, and so the second half of Umapati's text seems to acknowl-
edge the diversity, while yet suggesting that all these texts belong to-
gether. If it is Umapati himself who is extending the idea of canon to
incorporate these sundry texts, building upon the possibly earlier notion
that the hymns of the nayanmar are the tirumurai, then he signifies this
dichotomy by his use of different meters, but otherwise, why need he
provide an explanation, as Dorai Rangaswamy suggests? Umapati ex-
plains the material compilation of the canon; he leaves the conceptual
justification open, because this originated with himself.
71 Dorai Rangaswamy, p. 20.

72 Some scholars believe that, more specifically, the term referred to the hymns of Appar.
73 In fact, the Tirumurai canon is traditionally understood to have two halves: the Te-
varam, constituting the first seven books, is understood to have been traditionally set to
music; while the second half of the canon is understood to have originally been nonmu-
sical. Today, works in the second half, including the hymns of Mainikkavacakar, Tirumular's
Tirumantiram, and Cekkilar's Periya Puranam, are sung.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
252 A Tamil Lineage
In general, scholars have overlooked Umapati's likely role in concep-

tualizing the canon. Although there is no reason to doubt that Nampi
Antiar Nampi does play a role, along with other agents, in the material
compilation of the Tirumurai, insofar as he organizes the first seven
books according to the Tirumuraikanta Puranam, he is not the only
figure involved in the creation of the Tamil bhakti canon. It is more
likely Umapati Civacaryar who assembled the entire canon, which is
based on the lives of the nayanmar (their hymns begin the canon; their
hagiographies end it). His interest in the hymnists and hymns informed
his creation of the Tirumurai canon as we know it today.
Umapati's discussion of the canon continues into the related text, the
Cekkilar Puranam. This text is about completion. I have noted above the
ceremonial and mythological cues that suggest that for Umapati, Cek-
kilar's Periya Puranam represented the culmination of medieval Saiva
bhakti literature in Tamil, and thus it stands as an important text for
Indian culture. Tradition remembers the Periya Puranam as the last
"voice" of nayanmar Siva bhakti, a perspective that Umapati's descrip-
tion certainly encourages. This is to say that the way Umapati repre-
sented the history of the Saiva bhakti literature is largely the way it has
been understood in the present day.
There is reason to believe, then, that Umapati's agency was important
in defining the Saiva bhakti canon, and in its closure. The authoritative
literature concerning the nayanmar, namely, the hymns in several vol-
umes and the hagiographies by Cekkilar, also in several volumes, were
used to frame the new and encompassing Saiva bhakti canon. It was
quite possibly an expansion of the old sense of tirumurai, originally
signifying the hymns alone, to a larger whole. In the process, texts that
originally may have had nothing to do with nayanmar bhakti were
framed by its texts; for example, Manikkavacakar's work constitutes the
eighth book of the Tirumurai, and Tirumular's Tirumantiram is the
tenth book. The strategy may be summarized as making wholes into
parts of another whole. Each autonomous text now became part of a
whole-the canon-which suggested that there was an intrinsic like-
ness or similarity to them, that they were in some way of a piece.
The creation of a unified, and closed, image of Saiva bhakti would
have been important to Umipati for several reasons. First, he was inter-
ested in creating a coherent Tamil lineage for Saiva siddhanta. Second,
the Tamil bhakti poet-saints had emphasized the "purity" of Tamil
(centamil).74 Last, Umapati sought to interpret bhakti hymns according
to Saiva siddhanta philosophical premises. For all of these reasons, he
74 For example, Campantar 2.183.11, Cuntarar 7.65.2; see Peterson (n. 37 above), pp. 40,
189, 329.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
appropriated bhakti in both form and content. On the last issue, we can
understand from Umapati's work that the bhakti hymns, especially the
Tevaram, were specifically relevant to his expression of Tamil Saiva
siddhanta philosophical ideas. How? First we must note that Umapati
was a writer of essences. Most of his texts are rather short, and thus
have the character (if they are not in fact named as such) of "essential"
Saiva siddhanta philosophical ideas. They are the "essence" of pure
philosophy. We can therefore anticipate that Umapati would have been
concerned to relate the hymns to the essentials of Saiva siddhanta phi-
losophy as he saw them.
Umapati represents the essentials of Saiva siddhanta in his sastraic
Tiruvarutpayan (The fruit of divine grace), one of his eight canonical
texts in Tamil Saiva siddhanta, in which he provides ten category head-
ings that signify the essence of Saiva siddhanta philosophy. In brief,
the categories are: God, soul, bond, grace, guru, methodology, enlight-
enment, bliss, mantra, and liberation.75 Under each of these category
headings he included ten original couplets that illuminate the meaning
of the category heading. It may be the case that Umapati used these ten
categories as a method for organizing the Sanskrit slokas of agamic
philosophy he collected in his Sataratnasatigraha (Compendium of one
hundred gems).76 But he clearly used these categories in interpreting
the Tamil bhakti hymns, as evidenced by his Tevdra Arulmuraittirattu
(Summary of the collection of graceful Tevaram), a text that can be
considered to be the first anthology of the nayanmars' hymns.77
The notion that the Tevara Arulmuraittirattu is a summary should
not be taken uncritically. The text reveals an intensive selection; each
category has a different number of hymns, so there is no impression of
a serialization or an extrapolation of the selection-the hymns under
the heading are all that there are in the corpus to illustrate the category
heading. In addition, Umapati did not use sets of ten or eleven hymns
(patikam), as found in Tevaram; he used parts from them, often splitting
75 These are my one-word summaries of the chapters, not translations of the headings
as found in the text; e.g., what I summarize as "Liberation" is actually "The State of
Those Who Have Attained" in the text.
76 N. R. Bhatt suggests that there is some overlap in the content matter of the Tiru-
varutpayan (TP) and the Sataratnasatigraha (S)-verses 1-10 TP = verses 7-17 S; 11-
20 TP = 18 S; 21-30 TP = 19-33 S; 31-70 TP = 34-70 S; 71-90 TP = 71-78 S; 91-100
TP = 79-91 S-but the categories from the Tiruvarutpayan are not provided as titles in
the Sataratnasahgraha, nor is the similarity convincing in all sections. See N. R. Bhatt,
"Pasu and Pasa in Sataratna-Sangraha," in Umapati Sivacarya: His Life, Works, and Con-
tribution to Saivism, ed. S. S. Janaki (forthcoming). See also P. Thirugnanasambandhan,
Sataratnasangraha of Sri Umapati Sivacdrya (Madras: University of Madras, 1973).
77 Subsequent anthologies are numerous, including contemporary pamphlets from which
pilgrims can sing Tevaram at sites of which the nayanmar sang. Many of these are pub-
lished by the Tirumai Atinam.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
254 A Tamil Lineage
one patikam into several verses that appear nonserially within a head-
ing. Umapati's intensive selection suggests that there is a small degree
of overlap between the religious formulations of the nayanmar and the
philosophical formulations of the Saiva siddhantins. His thesis appears
to have been not only that specific nayanmar hymns are examples of
the philosophy expressed in the Tiruvarutpayan, but that the Saiva
siddhanta philosophy is embedded within certain hymns. He used the
categories of the Tiruvarutpayan as a focusing lens to extract and call
attention to these kernels of philosophy in specific stanzas culled from
the hymns of the nayanmar.
To illustrate the relationship between Tevdram and Saiva siddhanta in
Umapati's formulation, I will cite as example three verses from the nayan-
mar, included in his Tevara Arulmuraittirattu.
Appar 6.31.3/6566 Tiruvarur

Oh my mind
if you think to achieve the ultimate state, then come!
Enter the temple of our Lord daily
sweep it before dawn
smear the floor with cow dung
decorate Him with flower garlands and worship
sing songs of praise
bow in worship with your hands above your head
and dance,
crying "Victory to Sankara, Hail! Hail!"
"My Highest Lord, whose red locks bear the waves of water [Ganga]!"
"Lord of Arur!"
over and over and louder and louder.
Campantar 1.21.4/220 Tiruccivapuram

The servants who always tread on the path
worship Him thoughtfully,
with sandal paste, flower buds and blossoms, incense,
bright lights, and copious water.
They think of Him at Civapuram
where water is contained in tanks;
the powerful place where the Lord who has the gunam of Grace
grants them the proper state without deficiency.
Those who think of Him here
will become the husband of the goddess of Victory.
Cuntarar 7.51.4/7756 Tiruvarir

I came here
subject to the afflictions of birth

and I am becoming tired.
My Lord of Arur came there

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
as a rare medicine that rules me

so I will not lose strength.
How can I be separated from Him anywhere,
my nectar,
the Lord whose great form is the blazing fire
the Lord whose hand holds the deer?
These verses-or verses similar to these-could be, and are tradition-

ally, understood as illustrations of the fourfold hierarchy of realizing
Siva, represented in Sanskrit Saiva siddhanta agamic literature78 and in
the second half of the Tamil Saiva siddhanta canonical author Arun-
anti's Civanana Cittiyar. In this text, Arunanti sets out a hierarchy o
paths toward realizing Siva: cariya, acts of service; kriya, acts of ritual;
yoga, mental and physical discipline; and jndna, knowledge.79 In th
interpretation, the verse from Appar illustrates the path of cariya (e.g.
cleaning the temple); that of Campantar, kriya (e.g., offering flower
and incense); and that of Cuntarar, yoga (e.g., the Lord is not separat
from the poet). The path of jinna is traditionally understood to b
portrayed in the verses of Manikkavacakar, although since this author's
poetry is not part of Tevaram, it does not concern us here. Also of note
is that, in the traditional Tamil meanings of these categories, Appar is
the slave, Campantar is the son, and Cuntarar is the friend of Siva.
This hierarchy, which is often quoted in scholarship on Tamil religiou
and philosophical formulations, is not as straightforward as it seems.80
In fact it cannot be properly understood unless one takes into consider-
ation the relationship between the Tamil Saiva siddhanta and the Tamil
bhakti hymns and hymnists as formulated by Umapati Civacaryar. For
Umapati challenges the applicability of this hierarchy to the bhakti pat
of service to Siva as represented in the nayanmar poetry, by including
these three verses in the final section of his "Summary." This final sec-
tion, which is the culmination of all of the ten sections in the texts, is
78 However, understanding this fourfold schema as the structure around which the
Sanskrit agamas were written, which is widely assumed in the relevant scholarship, ha
been questioned by H6elne Brunner, "The Four Padas of Saivagamas," Journal of Orien
tal Research (Madras) 56-62 (1986-92): 260-78.
79 Civaiidna Cittiyar (cupakkam), verses 270-74. The Tamil names for these stages,
some of which use Sanskrit terms but all of which have meanings specific to Tamil, are
tatamarkkam (tatan meaning "slave" or "devotee"), carputtiramarkkam (cat + puttiran,
"good son"), cakamarkkam (caka, "combination" or "friendship"), and canmarkkam
(can, "excellence"). As I discuss below, these Tamil names and other evidence suggests
that the fourfold schema was applied to characterize, respectively, Appar, Campantar,
Cuntarar, and Manikkavacakar.
80 Some claim that the Tamil Saiva siddhantins view the nayanmar formulations simply
as examples of the first two paths in this fourfold hierarchy (e.g., Peterson, pp. 45-46),
but this does not agree with the characterization I heard from practitioners in India, nor
with the textual analysis I perform here.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
256 A Tamil Lineage
called Anaint6r Tanmai, "The State of Those who have Attained" sug-
gesting the attainment of the Grace of Lord Siva. All of the activities
described in the verses, then, are performed when the poet has already
achieved Grace; they do not represent preliminary "stages" but instead
are all grounded in Grace. Although as I noted, the Tamil Saiva sid-
dhanta did not connect itself to any specific temple tradition, Umapati
makes it clear that multiple modes of worship, including generic temple
activities, are performed within the State of Grace if one knows the path
of Saiva siddhanta expounded in the Tamil Sastras. The key to any act is
the knowledge of Siva specifically defined in the Tamil Saiva siddhanta
canon.
CONCLUDING REMARKS
This essay has taken as its focus the historical transformat

Saiva siddhanta philosophical school from a Sanskrit-based, p
network to a Tamil-based regional center. This is not to su
Saiva siddhanta philosophical formulations in Sanskrit cease
these were continued in the context of the temple for daily an
worship,81 and as I noted, some of the Tamil-based school wrot
Sanskrit, although these were not considered canonical. But
an important change in how the community of Saiva siddhanta
stituted. On the Sanskrit-based side, there was a tradition of p
philosophers, who were responsible for the writing and transm
texts, and for the discourse of commentaries. On the Tamil-ba
there were guru-disciple relationships in a shared cultural r
writing of texts which were in many cases like commentaries
canonical texts, and some commentatorial discourse.
The creation of a distinctive Tamil school highlighted the im
of Tamil heritage and identity. This was complicated, however,
the Tamil Saiva siddhanta was neither a continuous, and for
part inseparable, school within Sanskrit Saiva siddhanta, as
gested in the Encyclopedia of Religion article by M. Dhav
quoted above, nor simply a continuous development from t
bhakti hymns.82 Using aspects of both traditions, the philo
the Tamil Saiva siddhanta school sought to create their own lin
81 However, in contemporary Tamil Nadu priests use Tamil ritual manuals

believed to represent the philosophy and rites of worship according to th
agamas, in the temples. Very recently, the Tamil Nadu state government
program to reacquaint priests with the Sanskrit agamas; these are worksh
priests attend, for which they receive a small stipend.
82 That the Tamil Saiva siddhanta was a natural development from the bhak
also suggested by M. Dhavamony, in Love of God (n. 2 above).

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC
identity.83 The school that they created has had a major, long-lasting
impact in terms of structures and ideas in Tamil culture. Part of this
impact, I would suggest, has to do with the fact that the Tamil school
went beyond Saiva siddhanta philosophical literatures and into the realm
of bhakti hymns. The effect of this extension was to create a canon of
Tamil bhakti literature, a philosophical interpretation of bhakti, and a
heritage of inspired texts in Tamil. These inspired Tamil texts held, and
continue to hold, pride of place in Tamil culture.
In closing, I will quote a well-known verse that illustrates both the
status of these Tamil texts in Tamil culture and the pure knowledge of
the Tamil Saiva siddhanta. The verse imagines the relationships among
the Vedas, the agamas, the Tirumurai, and the Meykantar sastras as
that between a cow and its products: "The Vedas are the cow; the true
agamas are its milk; the Tamil sung by the four is its ghee; the essence
of the book in Tamil written by Meykantar of the famous Venney84 is
the taste of the ghee of great knowledge."85 In this simile, the essences
get progressively more subtle and pure leading up to the "ghee" that is
the Tamil hymns and the sweet "taste of the ghee" that is the Tamil
Saiva siddhanta.
Drew University
83 K. Sivaraman appears to view the Tamil Saiva siddhantins in a similar way when he
describes the Meykanta sastras as a "new phase of development... in Saiva thought"
sparked by the "revolutionary advent of the Sivajnana-Bodham" in Saivism in Philo-
sophical Perspective: A Study of the Formative Concepts, Problems, and Methods of Saiva
Siddhanta (Delhi: Motilal Banarsidass, 1973), 33-34.
84 Meykantar achieved his samadhi at Tiruvenneynallur.
85 The Tamil text of this anonymous simile (tanipdttu) is "vetam pacuvatanpal meyya
kammanalvar 6tun tamil atanin ulluruney potamiku neyyin urucuvaiyam nilvenney mey
kant.n ceyttamil nulin tiram." See Siddalingaiah, pp. 93-94.

130.63.63.214 on Sat, 14 Jan 2023 19:29:03 UTC

This Content Downloaded From 130.63.63.214 On Sat, 14 Jan 2023 19:29:03 UTC

Uploaded by

Copyright:

Available Formats

You might also like

This Content Downloaded From 130.63.63.214 On Sat, 14 Jan 2023 19:29:03 UTC

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

This Content Downloaded From 130.63.63.214 On Sat, 14 Jan 2023 19:29:03 UTC

Uploaded by

Copyright:

Available Formats

A Tamil Lineage for Saiva Siddhānta Philosophy

Author(s): Karen Pechilis Prentiss

Stable URL: https://www.jstor.org/stable/1062814

This content downloaded from

"Saiva siddhanta" is the name that certain medieval philo

1 Aspects of Saiva siddhanta philosophy received brief mention in

? 1996 by The University of Chicago. All rights reserved.

This content downloaded from

understood.3 In this article, I consider issues of difference between the

ON UNDERSTANDING SAIVA SIDDHANTA

The name Saiva siddhanta translates to "the end of knowledge o

This content downloaded from

India, although this classification need not be considered a rigid one."4

4 M. Dhavamony, "Saivism: Siddhanta," in Encyclopedia of Religion, ed. Mircea Eliade

This content downloaded from

So the Sanskrit Saiva siddhantins do not appear to have constituted

The origins of the Sanskrit Saiva siddhanta school is something of a

8 On this point see Davis, Ritual, pp. 12-14.

This content downloaded from

guidelines needed to construct and animate Saiva temples, through the

The period of Rajaraja I represents a break with earlier Chola imperial

1 Davis, Ritual, p. 10.

This content downloaded from

The system of Sanskrit Saiva siddhanta developed and flourished in

18 Abhinavagupta, the famous tenth-century Kashmiri exponent of nondualism, de-

This content downloaded from

the philosophical school of Saiva siddhanta were created by Tamil phi-

19 See T. B. Siddalingaiah, Origin and Development of Saiva Siddhanta Up to the 14th

This content downloaded from

The writing of authoritative treatises on Saiva siddhanta philosophy in

22 According to tradition, Meykantar was born in Tiruvenkatu, near Cirkali, some

This content downloaded from

This content downloaded from

content, but they strove to identify themselves as philosophers, not re-

UMAPATI CIVACARYAR AS AUTHOR AND GURU

Umapati Civacaryar was the most prolific of the authors of Saiva

This content downloaded from

This content downloaded from

was shared by other authors of various viewpoints whose Tamil works

38 Including Manikkavacakar, in his Tiruvacakam, esp. 1, 2, 31, 40 (the eighth book of

This content downloaded from

the Brahmins were compelled by the order of Siva himself to circumvent

43 1 understand that there is also an episode in Cekki!lr's story of Campantar in which

This content downloaded from

This content downloaded from

This content downloaded from

Saiva siddhanta. Therefore, he went down a lane populated by the low-

This content downloaded from

Although Umapati expressed a reluctance to link the Tamil Saiva sid-

UMAPATI CIVACARYA AND TAMIL SAIVA BHAKTI

57 Umapati is credited with several "miracles" at Chidambaram besides the ra

This content downloaded from

(cantanam) and religion (camayam), respectively. I suggested that the

This content downloaded from

anywhere, at the moment of the king's joy an Aticaiva (Adisaiva)

This content downloaded from

Periya Puranam was officially read out to an assembly of important