The Contribution of Tamil Literature To The Krishna of Sanskrit Cilappatikaram Charlotte Schmid

books.openedition.
org
The contribution of Tamil literature to the

Kṛṣṇa figure of the Sanskrit texts: the case
of the kaṉṟu in Cilappatikāram 17
Schmid, Charlotte
97-124 minutes
Charlotte Schmid
Editor's note
This paper is intended as a tribute to the work of the late Friedhelm Hardy.
My heartfelt thanks are due to Vincenzo Vergiani and Whitney Cox for having
organised a workshop on a challenging topic, to R. Varada Desikan, G.
Vijayavenugopal and Dominic Goodall for having worked on this paper with
me and to Suganya Anandakichenin who was kind enough to read it and to
revise its English.
Full text
• 1 See Edholm and Suneson (1972), the magnum opus of Friedhelm Hardy
(1983) and several papers of the (...)
• 2 See Hardy (1983: 179); see also Hudson (2002: 136–138).
• 3 F. Hardy (1983) has developed such a point in his book, but it would be
impossible to summarise hi (...)
1Though strongly marked by the Śrīvaiṣṇava movement, the history of

Kṛṣṇaism in South India is still very little known. According to recent research,
there is however a rather broad consensus according to which the
Śrīvaiṣṇava Bhāgavatapurāṇa (BhP) — which is said to have been
composed in Sanskrit in the Tamil land between the 9th and the 11th centuries
of our era — would testify to the introduction of motifs born from an
autonomous South Indian Kṛṣṇaite development into Sanskrit literature.1 The
Tamil Tivviyappirapantam (Tiv.), considered today as having appeared
between the 7th and the 9th centuries, would be the central work in which
such Kṛṣṇaism was expressed. Dated to the 7th century at the latest, the
canto 17 of Cilappatikāram (Cil.) introduced the character of Kṛṣṇa into Tamil
literature. Thus, it would constitute the first known testimony of some of the
“southern”, Tamil Tiv. motifs and was considered the very first known
acknowledgment of a legend developed in a southern milieu devoted to
Kṛṣṇa distinct from the northern ones.2 The folk–which is supposedly oral —
character of such a milieu is then held responsible for the almost complete
absence of Kṛṣṇa in the South Indian, Tamil texts considered more ancient
than Cil. 17.3 The specificities of the ancient Tamil tradition of the thoroughly
literary Caṅkam, to which these texts belong, would have prevented a folk
Kṛṣṇa-bhakti movement from appearing on the literary scene.
• 4 I have not taken into account the Bālacarita here, a work that is sometimes
considered one of the (...)
• 5 I cannot give all the features of the Kṛṣṇa of Cil. 17 with the corresponding
archaeological mater (...)
2We can still wonder if it is not possible to have a different point of view on
Cil. 17. Instead of considering this text as the first known literary attestation of
a southern folk Kṛṣṇa who would have predated the poem in popular oral
circles, I would rather regard this canto 17 as a crucible of a distinctive type.
The southern identity of a deity of northern Indian origin, whose legend is first
attested by Sanskrit sources older than the Cil., must have been forged in this
Tamil poem for the first time. The comparison between the old Sanskrit
sources — essentially, the Mahābhārata (Mbh), the Harivaṃśa (HV) and the
Viṣṇupurāṇa (VP) — and canto 17 of Cil. indeed reveals a complex literary
reality.4 It emphasises the originality of the Kṛṣṇa of the Cil., and
iconographic data validate the South Indian profile of the divinity who
dominates Cil. 17. All the features of this deity (such as playing the flute and
dancing with pots) are indeed initially represented only in South Indian
temples.5 However, is the originality of this Kṛṣṇa necessarily due to a
religious reality of a different nature than the one expressed in the ancient
northern Sanskrit texts? The hypothesis of a pre-Cil. folkloric origin does not
appear to me as the only possible one. Some of the features of Cil. 17 indeed
suggest that a more central role should be assigned to the known Tamil
literary tradition — already old by several centuries when this canto was
composed — than to any folk movement.
3Thus, I would like to offer a reassessment of the fundamental question

asked by Friedhelm Hardy while writing about the deity of Cil. 17: “How can
the almost total absence of references to such a genuine Tamil Kṛṣṇa
tradition in earlier sources be explained?” (Hardy 1983: 181) In order to
consider different answers to such an issue from the ones provided until now,
I have chosen to focus on a specific motif of the legend of Kṛṣṇa, that of the
animal thrown onto a tree, the kaṉṟu of Cil. 17.19. After Cil. 17, this kaṉṟu is
often mentioned in the Tiv. The motif is mysterious in these Tamil texts and
has been mostly understood in the light of the Sanskrit Kṛṣṇaite tradition,
which indeed knows the story of an animal thrown to the top of a tree. As a
minor but well-defined element shared by Sanskrit and Tamil sources, the
kaṉṟu motif could well throw an unexpected light on the relations between the
southern and northern literary traditions developed in Kṛṣṇa-bhakti
environments.
1. Kṛṣṇa elements in Cilappatikāram
• 6 This text can be considered from various other literary and religious angles
and I am aware how po (...)
4As canto 17 of the Cil. is the pivot of this paper, I will begin with a short
summary of this text. Commonly labelled an epic, the Cil. belongs to what
could be considered as one of the latest layers of a literature which still
participates in the Caṅkam corpus and as one of the earliest strata of the
Tamil bhakti corpus.6 Kaṇṇaki, a model of the virtuous wife and a
personification of the powers of the Indian female deities, is the heroine of the
Cil. The hero is Kaṇṇaki’s husband, Kōvalaṉ. The Pāṇṭiya king puts Kōvalaṉ
to death, mistaking him for a robber. Kaṇṇaki’s revenge is terrible as she
burns down Maturai by throwing one of her breasts in the streets of the
Pāṇṭiyaṉ capital city.
5There is an important concentration of Kṛṣṇa-related elements in the story.

Two of the main characters have a rather noticeably close link with Kṛṣṇa: the
Pāṇṭiya king is shown as an incarnation of Kṛṣṇa and Kōvalaṉ is the
Tamilised name of one of the most common Sanskrit designations for Kṛṣṇa,
gopāla, the “herdsman”. One of the major settings of the Cil. is the finally
destroyed Pāṇṭiyan capital, Maturai, which is none other than the southern
equivalent of the birth place of Kṛṣṇa, the northern Mathurā.
6Moreover, Cil. 17 is one of the few passages of the epic that present
strikingly devotional traits. Cil. 12 is indeed devoted to a warlike goddess,
while Cil. 24 is dedicated to Murukaṉ-Skanda, the god of love and war in
Caṅkam literature and the son of Śiva in Sanskrit texts from at least the 5th
century onwards. Presenting a form of Viṣṇu called Māyavaṉ — a name
which could be a quasi-literal translation of Kṛṣṇa, “the Black one” (among
other meanings) — Cil. 17 is an important part of the devotional corpus of the
epic: each of these three cantos, 12, 17 and 24, introduces different types of
praise in honour of the deities whose praise they sing. Songs and dances are
indeed offered to Māyavaṉ in Cil. 17.
• 7 See Edholm and Suneson (1972), Hudson (1982: 239–240) and Gros
(1968: li).
• 8 This human wife is unknown from the older Sanskrit texts. The identity of
the two other characters (...)
7Canto 17 is nonetheless a key moment from a narrative point of view. Canto

16 features two stages. One of these is the herdsmen’s camp where Kaṇṇaki
stays for a while. The second one is Maturai, where the killing of Kōvalaṉ
takes place. Cil. 17 opens with bad omens occurring in the camp of the
āycciyar (“cowherdesses”; cf. “cowherds”, kōvalar or āyar). These are linked
to Kōvalaṉ’s death, which Kaṇṇaki and the cowherds are unaware of. To allay
such omens, the herdsmen decide to celebrate a rite to venerate their god,
Kṛṣṇa. Songs that recount the exploits of the young god are sung. Girls
perform a particular dance, a kuravai, described with details that are so
difficult to interpret that they have given rise to fierce debates.7 In the dancing
circle which is then formed, three girls incarnate three characters from the
Kṛṣṇa-legend: Māyavaṉ himself, his elder brother “Muṉṉai” — Balarāma or
Saṃkarṣaṇa in Sanskrit texts — and “Piṉṉai”, the human wife of Kṛṣṇa in the
later South Indian, Tamil corpus.8
8This devotional Cil. 17 is of a puzzling nature that has long been recognised
as such. I would first of all emphasise that such strangeness might be
essentially of literary origin. In fact, Cil. 17 displays strong similarities with the
literary modus operandi later developed in the Tiv. We see here a primary
sample of a literary technique used with great virtuosity in the devotional
Vaiṣṇava Tiv.: mythology and daily practices of the herdsmen, presented as
ideal figures of devotees, are superimposed so as to merge — here during
the celebration of rituals. The motif of churning, occurring several times in Cil.
17, is one clear instance of the literary structure. Churning underlines the ties
between Viṣṇu, the god who churned the ocean of milk in a mythological fight
against demons, and this particular form of him that is Kṛṣṇa, whose
childhood takes place in a herdsmen’s camp where churning is a daily
practice. Then the churning is also the typical activity of the cowherds
amongst whom Kaṇṇaki has found shelter when Cil. 17 begins. Finally,
churning is part of the manufacturing of dairy products for the Pāṇṭiya king,
enthroned in his palace of Maturai. Cowherds have to prepare butter for the
Pāṇṭiyaṉ, i.e. to churn, the very day the bad omens appear (Cil. 17.1).
However, the milk does not curdle in the pot below, and the butter does not
melt in the other pot suspended to the ceiling (uṟi). What could possibly have
happened?
9Thus, churning constitutes a kind of thread stitching together different

locations (Maturai and the cowherds’ camp), mythological levels (Viṣṇu and
his incarnation Kṛṣṇa), characters (Kaṇṇaki and the cowherds, Kōvalaṉ and
the Pāṇṭiya king) and also literary traditions, as the motif of churning appears
in a northern Kṛṣṇa-bhakti tradition, while Cil. made its own peculiar use of
the theme. But in fact, the light under which the whole poem gets Kṛṣṇa to
make his first appearance on the literary stage seems indeed very different
from the one presented in the northern tradition.
• 9 We rarely come across deities in the Caṅkam corpus, except in the

Paripāṭal and the Tirumurukāṟṟup (...)
10We can begin with this tragic irony, which constitutes a major twist of canto
17. Kōvalaṉ’s death has already occurred. The rites performed to woo
Kṛṣṇa’s grace are doomed to fail in bringing the desired object even before
they are started. What is the use of such rites? What is the place held by a
god who does not protect his devotees, and lets incarnations of his, Kōvalaṉ
and the Pāṇṭiya king, die? Moreover, Cil. 17 is not the only passage of the
epic in which Kṛṣṇa’s role calls for re-examination. Kōvalaṉ gets involved with
a courtesan and his death is miserable. The Pāṇṭiya king makes a fatal
mistake and is punished by a violent supernatural death. However, Cil. 17
starts and ends with the eulogy of this king, composed with formulas and
myths used in Tamil Pāṇṭiya inscriptions of the 8th and 9th century CE, which
attest to a certain “reality” of such praise. “The Pāṇṭiya kingdom is here...”
Could this be another expression of the kind of derision found in some of the
Caṅkam poems in which the priest of Murukaṉ is made fun of, as a character
whose ritual communication with the god appears as a risible action?9
11But one of the most mysterious elements of the entire poem is probably the
legendary deeds of Kṛṣṇa presented in Cil. 17. Secondary literature has
indeed pointed out particular motifs in Cil. 17 that are considered unknown in
the ancient northern Sanskrit texts. Such themes are the ones that are
supposed to have emerged from a southern folk milieu, of which they would
thus be the first attestation (Edholm and Suneson 1972; Ate 1978; Hardy
1983; Hudson 1982, 2003).
2. The legend of Kṛṣṇa in Cilappatikāram 17
• 10 Before the dance involving Piṉṉai, Cil. 17 associates seven bulls of

different colours with girls (...)
• 11 It should be said, however, that while this legend is often mentioned in the
Tiv., as far as Cil. (...)
12To demonstrate the actual existence of a typical southern folk Kṛṣṇaism,

researchers have mainly concentrated on three elements of Cil. 17. One of
them is the young lady Piṉṉai who dances the kuravai with Kṛṣṇa in Cil. 17
and who is a human spouse of Kṛṣṇa’s in the Tiv. The second one is the
reference to a fight against a herd of seven bulls which are associated with
seven girls in Cil. 17; this becomes a fight against seven bulls through which
Kṛṣṇa gains Piṉṉai’s hand in the Tiv.10 The theft of the herdswomen’s clothes
by a playful Kṛṣṇa, who hides them in a kuruntu tree when the gopīs go
bathing in the river, an event evoked in the Tiv., would be a third element
originating from a Tamil folk tradition.11
• 12 In the southern recensions of the HV, Kṛṣṇa marries a young human girl
called Nīlā, the Black one, (...)
13The analysis of Sanskrit data indeed tends to present these motifs as

being typical of a southern tradition. None of them appears in the MBh. This
is not surprising given that this text is rather silent about the exploits of Kṛṣṇa
as a child in Mathurā. But the fact that they do not appear either in the critical
text of the HV, the sequel of the MBh that recounts the childhood of the god
Kṛṣṇa, or in the VP in which a whole section is devoted to Kṛṣṇa’s birth,
infancy and youth, is remarkable. Both these texts are considered today as
having been composed in northern India, the HV between the 2nd and the 4th
century of our era and the VP during the 5th (Couture 1991: 72–77, for the
HV; Rocher 1983: 248–249, for the VP). Then, we do come across a specific
version of the Piṉṉai myth in the versions of the HV which have been written
later on in South India. They include a fight against seven bulls. Regarding
the theft of the clothes, it is told extensively in the BhP. This Purāṇa also
mentions fights of Kṛṣṇa against young bulls, on the one hand, and the
existence of a young girl more specifically linked to the god, on the other.12
We shall keep in mind that the BhP was composed in Sanskrit in the Tamil
land, from which it has drawn a great number of specific legends, known
mainly from the Vaiṣṇava corpus of the Tiv.
14To recapitulate, the legend of Kṛṣṇa in southern India presents specific

traits coming from a particular religious tradition, appearing first in the Tamil
Cil. 17, and then developed in the Tiv. This tradition is supposed to have been
originally of a popular nature as it does not appear in any text, whether in
Sanskrit or in Tamil before the Cil., but it proves its appeal afterwards, notably
by spreading to the whole of India. The legendary themes such as they
appear in the Sanskrit BhP were indeed broadly diffused in a pan-Indian
context. To such a picture, we have to add that the southern Sanskrit versions
of the HV have also sometimes developed legendary motifs first encountered
in the Tamil Tiv. Such a literary development tends to confirm the southern
specificities of a Kṛṣṇa-oriented devotional movement of some kind.
15This synopsis might nevertheless ignore the complexity of the bond

between the Sanskrit tradition and the Tamil literature, and that doubly so.
First, if we consider that Piṉṉai, the combat against the bulls and the theft of
clothes represent the first known literary expressions of a southern folk
Kṛṣṇaism, the whole of canto 17 of the Cil. might have to be considered the
same way. These three elements are not the only ones mentioned in Cil. that
are rather difficult to trace back to ancient Sanskrit texts: Cil. 17 states four
other activities of Kṛṣṇa’s that are also not so easy to find there. The young
god of Cil. 17 plays the flute, steals butter that is kept in a suspended pot,
uses a kaṉṟu to get fruits and tears down a kuruntu tree. None of these
seems to be recounted within Sanskrit texts before the BhP. The butter thief
and the flautist are rather famous as they have become some kind of
emblems of Kṛṣṇa in sculpted and painted images as well as in texts. Still,
they do not make any appearance in Sanskrit literature before the BhP
(Schmid 2002). Kṛṣṇa is not represented as a flautist in North India before the
13th century. The breaking of the kuruntu tree is considered by Hardy (1983:
193–197, 515–516) as an allusion to theft of the gopīs’ clothes, which would
be a variant of an old Caṅkam motif. We shall engage below in a more
specific discussion of the kaṉṟu motif.
16On the other hand, which of the motifs from the northern Indian legends of
Kṛṣṇa (i.e. as told in the Mbh, HV and VP) appear in Cil. 17? In these ancient
Sanskrit texts, the young Kṛṣṇa breaks a chariot, fights against the demoness
Pūtanā of poisonous breasts, breaks a pair of Arjuna trees by passing
between them with a mortar, fights against a giant snake in a lake, and kills a
donkey, a bull, a horse and an elephant; he also raises the Govardhana
mountain and defeats wrestlers in the arena in Mathurā before killing his
demonic uncle there. A good number of these exploits have been carved
since at least the 5th century of our era in North India, while none of the
occupations of the god of Cil. 17 has been illustrated in stone before the 10th
century, or perhaps the end of the 9th in the case of the flautist in South India.
These depictions are first restricted to the Tamil land only. Before these,
some of the “northern” exploits of Kṛṣṇa, such as the lifting up of the
Govardhana hill, started to be represented in the Tamil land with the plastic
productions of the Pallava dynasty (see Francis 2009 and forthcoming).
17Thus, there is no apparent reason not to consider the whole of Cil. 17 in

the same way as the episodes of Piṉṉai, the seven bulls and the theft of
clothes. Cil. 17 in its entirety could attest a specific folk Kṛṣṇaite milieu. This
folk devotion has supposedly given rise to legendary themes of its own: the
breaking (ocittaṉ) of the kuruntu tree, the use of a kaṉṟu to get fruits, the
playing of the flute and the stealing of butter could be some of them, although
they have been largely ignored in secondary literature. This southern
movement would be better attested in Tamil literature by the Tiv., which
features these Tamil folk elements side by side with the legendary motifs
found in the northern Indian legends. A little later, the BhP may have provided
a harmonious integration of this blend of northern and southern traditions into
Sanskrit Puraṇic literature.
18The apparent innovation seen in Cil. 17 leads me towards another

interpretation of the data. The scrutiny of what I have called the kaṉṟu motif of
Cil. 17 makes me wonder if the deeds of Kṛṣṇa in Cil. 17 would not point to a
first attempt to adapt the northern Indian legend to a Tamil literary context.
3. The kaṉṟu motif
• 13 I am grateful to Eva Wilden who was kind enough to look for references to
flautists in Caṅkam lite (...)
• 14 On the vastraharaṇa motif, see Hardy (1983: 193–196; 497–501;

512–516) and infra: 45–46.
19In each motif of Cil. 17 the inspiration from Caṅkam literature may be more
or less easy to spot, but it is nevertheless always possible to detect it. The
case of the flautist is particularly interesting, though complicated, and I refer
the reader to my previous and forthcoming studies (Schmid 2002: 41–45;
Schmid forthcoming).13 The stealing of the butter that is kept in the typically
South Indian uṟi, mentioned in Cil. 17 and in the Tiv., is another complicated
case, to which I also intend to devote a further study. Suffice it to say here
that if this does not appear in Sanskrit texts before the BhP, it is a recurrent
motif of the sculpted legend of Kṛṣṇa in the northern regions from around the
5th century. The breaking of the kuruntu tree is another fascinating issue, but
both the quantity of sources and the thorough study of the Sanskrit
vastraharaṇa presented by Hardy based on the links between the two themes
prevent me from presently venturing into such a complicated theme.14
• 15 See the commentaries given in the edition of U. Vē. Cāminātaiyar

(451.18).
20The choice to concentrate on the kaṉṟu motif is based on the relative

simplicity of the case. Mentioned only in one line of Cil. 17, the kaṉṟu motif is
attested at least twenty times in the Tiv., while its equivalent is found in the
oldest Sanskrit texts. In my opinion, its study brings to light the mechanism of
literary transposition from Sanskrit to Tamil and then back to Sanskrit in the
legend of Kṛṣṇa. If the kaṉṟu motif appears indeed for the first time in Tamil
literature in stanza 19.1 of Cil. 17, nevertheless it is not understandable by
itself. It presupposes a legendary background known to the authors and their
audience. The old commentaries explain it with what looks like a summary of
a legend featured in the later Tiv., in which Kṛṣṇa throws a calf (kaṉṟu)
towards the top of a tree.15 On the other hand, Mbh, HV and VP establish a
motif showing similarities with this kaṉṟu theme in the Sanskrit Purāṇic
tradition and involving an animal thrown at a tree. Sanskrit texts refer to a
donkey called Dhenuka. There are strong reasons to believe that the
Dhenuka legend was the source of the kaṉṟu motif in the Tamil texts.
• 16 See Hawley (1989: 311–375, appendix A), for a list of sites where a fight
against a calf and a fig (...)
• 17 The Gwalior museum exhibits a representation of a distinctive tāla tree,

with an animal stuck on i (...)
21Among the various occurrences of the kaṉṟu motif in the Tiv., in fact, some
do correspond to Cil. 17, while others are closer to the Dhenuka tale of the
Sanskrit tradition and others are designed according to a new pattern. This
evolution finds its culmination in the southern Indian Sanskrit accounts: from
BhP 10.11.4–44 — which is the first Sanskrit text to recount Kṛṣṇa’s fight
against a strange calf, vatsa — to a whole chapter of the southern versions of
the HV, the kaṉṟu motif has indeed been expanded in an unexpected way. In
another medium, a fight against an animal thrown at a tree is found in
carvings from the 5th century onwards, if not before. Such a combat is
represented later on in the miniatures illustrating the BhP.16 The first known
representations are from northern India and they show a donkey (Fig. 1).17 In
the later miniatures, two different fights have been represented. In one
instance, a donkey is thrown at a palm tree; in the other, a calf is flung at a
large kapittha tree.
22The passages from Sanskrit to Tamil and back to Sanskrit of the myth of
hurling an animal-demon at a tree account for the major transformations that
this fight scene undergoes — until the combat splits into two distinct legends
in the Sanskrit tradition from which it first emerged through the way it was
presented in Tamil texts. I suggest that this constant process of exchange
between Sanskrit and Tamil is representative for a large part of the invention
of a Kṛṣṇa legend in South India. Given below are the details of this little
survey.
4. The Cilappatikāram 17 stanza and the oldest Sanskrit texts
• 18 The text is given as it appears in U. Vē. Cāminātaiyar’s edition that I have

used; see this editio (...)
23In Cil. 17.19.1–3, one cowherdess sings about the soft koṉṟai flute of
Māyavaṉ, who comes surrounded by his herd, after felling (utirtta) ripe fruits,
kaṉi, using a kaṉṟu as a kuṇil, a stick:18
kaṉṟu kuṇilāk kaṉiy utirtta māyavaṉ

iṉṟu nammāṉuḷ varumēl avaṉvāyiṟ
koṉṟaiyantīṅkuḻal kēḷāmō tōḻī
24R. Parthasarathy (1993: 173) has given the following translation of this
passage:
“O friend! Māyavaṉ struck down the fruit
With a calf as his stick. Today if he came
Among our herd of cows, won’t we hear
the sweet laburnum flute at his mouth?”
• 19 I refer here to the translations of the Cil. into English by Ramachandra

Dikshitar (1939), Daniélo (...)
• 20 Hardy (1983: 175) translates the Cil. passage in the following way:
“Māyavaṉ who with the calf as (...)
25In most of the English translations of this passage I had access to, kaṉṟu is
translated as calf and all the authors of secondary literature identify this
episode as that of a calf, something which recurs many times in the Tiv.19
According to Dikshitar (1939: 268, n. 4), this refers to an episode occurring in
the BhP. In two of the numerous footnotes of his significant work Viraha-
bhakti, Hardy associates this “calf” with “Dhenuka”, the demonic donkey as it
appears in the oldest versions of the HV. Unfortunately he does this without
giving any further explanations.20 We can try to surmise what they could
have been.
26MBh, HV and VP do not speak of any calf with which Kṛṣṇa would fell
fruits. However, the HV as well as the VP recount a combat against an animal
with which Kṛṣṇa’s elder brother, Saṃkarṣaṇa-Balarāma, fells fruits from a
tree. Chapter 57 of the HV is the first known text to tell this story in a detailed
manner. The herdsmen seek the ripe fruits of a forest of tālas and they have
to fight against the ass-demon who guards it, Dhenuka. That Dhenuka is a
gardabha, an ass, which rules on a large herd of asses:
27“A terrible demon of the name of Dhenuka had taken the shape of an ass;
accompanied by a large troop of asses, it walked [in this place]. The ass
protected this terrible forest of tāla (palm tree, borassus); it frightened the
men, the birds and the large animals; it was evil-minded. [...]
• 21 dāruṇo dhenuko nāma daityo gardabharūpavān / kharayūthena mahatā

vṛtaḥ samupasevate // sa tat tala (...)
28“Then he [Saṃkarṣaṇa] seizes the ass-demon by its two hind legs and,
after having whirled round its head and torso, hurled it at the top of a tāla
tree. The malicious ass, its thighs, sides, neck and back broken, fell on the
ground among the fruits of the tāla tree” (HV 57.12–13; 19–20).21
29VP 5.8 follows HV 57 almost word for word. Thus it presents Balarāma
fighting a gardabha, a donkey, which is sent flying onto the top of a tree from
which the fruits fall to the ground. In HV and VP, the episode ends with a
general clash between Kṛṣṇa, accompanied by his brother, and a troop of
donkeys, which they dispatch to the tops of tāla trees. The ground is strewn
with fruits.
• 22 ariṣṭo dhenukaś caiva cāṇūraś ca mahābalaḥ / aśvarājaś ca nihataḥ

kaṁsaś cāriṣṭam ācaran, Mbh 5.12 (...)
• 23 dhenukaḥ sa mahākāyo dānavaḥ sumahābalaḥ / nihato vāsudevena

gavāṃ trāṇāya durmatiḥ, HV 96.39.
• 24 On the date of the Bālacarita, see above, note 5.
30This legend and that which Cil. 17.19.1–3 alludes to present similarities. A
god uses an animal to cause fruits to fall from trees. If, according to the
common Puraṇic tradition I have just referred to, it is not Kṛṣṇa himself who
hurls an animal at the tree, the objection to bringing the Cil. 17 kaṉṟu motif
closer to Dhenuka’s legend fails when one examines the whole of the
Sanskrit data closely. First of all, in the HV itself, Kṛṣṇa’s brother
Saṃkarṣaṇa-Balarāma is the hero of only two episodes: he fights against this
demon-donkey; he fights against a giant called Pralamba. These two exploits
are modelled on the episodes in which Kṛṣṇa is the hero — as the fight
against Dhenuka echoes the fight scene with Keśin the demon-horse (Fig. 2).
Balarāma’s character here is evidently derivative of Kṛṣṇa’s, and Utz Podzeit
(1992) has even suggested that Kṛṣṇa was the original opponent of the giant
Pralamba. Besides, there is a rather well-attested textual tradition which
makes Kṛṣṇa Dhenuka’s adversary, rather than Balarāma. Indeed, in a list
given in MBh 5, in which the names of several of Kṛṣṇa’s opponents are listed
without indication as to their shapes, Kṛṣṇa fights against a certain
Dhenuka.22 In similar lists, the HV (65.2; 28–29) also assigns a fight against
Dhenuka to Kṛṣṇa. The species of the demon is not specified either, but the
setting of the combat is the same tāla forest as in chapter 57. In another
passage, HV 96.39, it is said that Dhenuka is a giant that Kṛṣṇa fought.23 In
about the 9th century, in South India, a similar tradition is followed in the third
act of the Bālacarita, in which, right before fighting against Keśin, Kṛṣṇa
struggles against the donkey Dhenuka in the forest of tālas, which is the
backdrop given by the whole of the Puraṇic tradition.24 On the other hand, it
should be noted that from the first known data, in the iconographic texts as
well as in the narrative textual tradition and the sculpted representations, the
tāla is Balarāma’s emblem. The species of tree in the HV might have been
chosen to emphasise the “Balarāmic” nature of the episode — perhaps it was
needed to attract Kṛṣṇa’s brother to the middle of a stage in which he was not
originally present.
31Whatever the interpretation, the legend narrated in HV 57 appears to be

one of the bases of a tradition which travels intact to South India, where the
BhP devotes a whole chapter to the fight of Balarāma opposing an ass
(khara) in a tāla forest (10.15).
• 25 The demonic donkey is then placed as the guardian of the city of

Dvāravatī.
32It is worth noting that in Sanskrit Dhenukā, with a final long ‘a’, is one of the
designations for a cow and Dhenuka with a final short ‘a’ could be interpreted
rather naturally by a reader/listener as a member of the ox genus. We can
also suppose that this demon called Dhenuka gained the nature of a donkey
within the HV itself. This text was indeed written in the area of Mathurā,
where a demonic ass may have been a specific trait of the region, since the
Buddhist account of the Kṛṣṇalegend linked to Mathurā (the Ghata jātaka)
also mentions a fight against a donkey-shaped yakṣa.25 That a malevolent
cow was given the identity of a demonic ass haunting the surroundings of
Mathurā may account for the strange name of the donkey in the HV.
• 26 For a critical presentation of these posts, see Williams (1982: 55–58).
33The first known representations could correspond to the oldest known

textual versions of the myth. On what is conventionally called the Maṇḍor
pillars found in Rajasthan, where several episodes of the legend of Kṛṣṇa
have been represented on two posts dated approximately to the 5th century
CE, a fight against a horse and a fight against a donkey have been sculpted,
among other exploits of Kṛṣṇa (Fig. 1 and Fig. 2).26
34The two animals are quite different from one another, while the human
figures of the fighters are similar. They can be the two brothers, introduced
like the two faces of the same character in the Mbh and the HV; or they could
be two representations of Kṛṣṇa himself. But the mode of combat
corresponds to the account of the HV in both instances. The donkey-demon
is taken by his hind legs and he does resemble a donkey. The fragment of a
tree that is still visible is that of a palm tree. The combat against the horse is
different. A young lad fights against an aggressive steed (Schmid 1999). The
size of the ass-demon is also strikingly smaller than the one of the horse.
Fig. 1 One of the two “pillars” from Maṇḍor, Rajasthan: the fight against an
ass-demon (top of the pillar; photo by courtesy of the American Institute of
Indian Studies).
Fig. 2 One of the two “pillars” from Maṇḍor, Rajasthan: the fight against a
horse-demon (photo, author).
5. kaṉṟu, Dhenuka and the Tiv
35While in Tamil the term kaṉṟu usually designates a calf, according to the
Tamil Lexicon it can also mean the young one of an impressive list of
animals, including a kaḻutai, a donkey. Thus, the taxonomic situation
encountered in Cil. 17 is close to the one expounded in the HV in which the
donkey bears a Sanskrit name connected to that of a cow. Moreover, as my
colleague Dominic Goodall has pointed out to me, the final -ka of the name
Dhenuka could also be interpreted in Sanskrit as a diminutive indicating the
small size or the youth of a cow, dhenu. In other words, Dhenuka could
definitely be a calf.
36The four thousand poems of the Tiv. constitute one of the axes of the
development of a literary tradition focusing on Kṛṣṇa-bhakti that is proper to
South India. Mentions of the narrative involving the kaṉṟu motif are numerous
in this Vaiṣṇava devotional corpus. They reveal the complexity of a legendary
evolution which, in my opinion, they show to be of a fundamentally literary
nature. It is impossible, in fact, to draw a general picture of any kaṉṟu myth in
the Tiv. In this anthology of poems in which the works of twelve Vaiṣṇava
saints have been gathered, each author might have had his own vision of the
motif. Moreover, most passages involving a kaṉṟu and its contact with a tree
are difficult to understand. Commentators as well as translators present
various interpretations. Those texts are indeed not narratives and the stanzas
come up with one or two lines evoking the association of a kaṉṟu with the
fruits of a tree, most of the time a viḷa, as well as a particular modus operandi,
indicated by the verbal root eṟital, a transitive verb meaning “to throw, cast,
fling, discharge, hurl”.
37These elements are often regrouped in concise formulas which do not

easily allow us to identify the detail of the action. Besides, other elements can
also be incorporated within these formulas. They sometimes seem to give the
same episode a meaning differing from one poem to the next.
38On the other hand, the Tiv. also refers to a fight against a demon called
tēṉukaṉ, the Tamilised form of Dhenuka. These references are rare: I have
found only four of them in the whole of the Tiv. The methods of the combat
against tēṉukaṉ are the same as the ones seen with the kaṉṟu motif: the
demon is thrown, eṟital, in the air at a tree which then loses some of its fruits.
The shape of this “tēṉukaṉ” is never specified; the term kaṉṟu does not seem
to be associated with it either. However, kaṉṟu and tēṉukaṉ are found in one
and the same stanza in Periyāḻvār 1.5.4, which is the first mention of both
fights found in the editions of the Tiv.
• 27 For references to the text, see above, n. 20; see Periyāḻvār 1.5.4; 2.3.10;
2.4.8; 2.5.5; 3.1.6; 3 (...)
39Here is the part of the stanza which mentions both a kaṉṟu and
Tēṉukaṉ:27
• 28 Here is the whole stanza:

vāṉavartām makiḻa vaṉ cakaṭam uruḷa
vañca mulaippēyiṉ nañcam atu uṇṭa vaṉē (...)
[...] kāṉaka vīl viḷaviṉ kāy utirak karutik

kaṉṟu atu koṇṭu eṟiyum karuniṟa eṉkaṉṟē
tēṉukaṉum muraṉum tiṇtiṟal vennarakaṉ [...].28
• 29 See infra, n. 28; in the old commentaries of the edition of Kiruṣṇasvāmi

Āyyaṅkār, it is said that (...)
• 30 atu shows that the kaṉṟu should be considered a second case and koṇṭu
could then be seen as a kind (...)
40An independent syntagma is reserved for the episode of the kaṉṟu (kāṉaka
vīl viḷaviṉ kāy utirak karutik kaṉṟ’atu koṇṭ(u) eṟiyum); then we see an address
to the god (karuniṟa eṉ kaṉṟē), followed immediately by the name of Dhenuka
(tēṉukaṉ) associated with other demons’ names (tēṉukaṉum muraṉum tiṇtiṟal
ven narakaṉ). Thus, syntax does not necessarily establish a link between the
kaṉṟu and the tēṉukaṉ. But it does not rule it out either, and without the other
occurrences of the kaṉṟu episode in the Tiv., the Tamil term kaṉṟu and the
Sanskrit name Dhenuka might not have been dissociated, as they are in the
translations and the commentaries I have examined as well as in the minds of
the 21st-century Tamil devotees I have questioned about the subject.29 In
fact, only an address to the god as a black-coloured kaṉṟu separates the
“calf” from the “ass” in the three lines first mentioning the kaṉṟu motif: the
word order allows for several interpretations30 and, as in the Cil., one cannot
understand such a mythical allusion without the help of other texts.
41In the rest of Periyāḻvār’s corpus, the fight with the kaṉṟu appears five
times in the first three books, which speak of Kṛṣṇa’s early childhood. The
narrative scheme of the combat with the kaṉṟu is always as follows: a kaṉṟu
is whirled in the air and hurled at a tree, causing the fruits to fall. It is often
specified that the kaṉṟu is young, iḷan. In Periyāḻvār 2.3.10 this fight follows
the one against a bull and the kaṉṟu is said to be iḷaṉ, young, which seems to
suggest that the kaṉṟu is indeed a calf, i.e. a young bull (see above, note 27).
• 31 In the opinion of the old commentators and their present heir R. Varada
Desikan, this is one of th (...)
42In the whole of the Tiv., kaṉṟu is often used in contexts other than the one
we have just studied and it is clear that it usually refers to calves (of a herd, to
which cows attend, etc. and as a sweet name for Kṛṣṇa himself, like in
Periyāḻvār 1.5.4 for instance). When the species of the tree is specified, it is
always a viḷa or viḷavu, wood-apple or eagle-wood. In the scheme used in
Periyāḻvār 2.10.4 to evoke Dhenuka’s tale, the fruits represent the only
element which is different from the kaṉṟu theme as they are said to be those
of the paṉam, the borassus palm tree. This detail is in conformity with the HV
version of the story, as the Sanskrit tāla is a borassus palm tree. The mention
made in 2.4.8 of the kaṉṟu motif is noticeably puzzling. It associates the kaṉṟu
with the palm tree through the term ōḻai which means a palm-tree leaf, but
what is done with these two is far from clear. kaṉṟiṉai vāl ōḻai kaṭṭi kaṉikaḷ
utira eṟintu has been translated in Dr. S. Jagathratchagan’s edition as “You
tied a Palm leaf to a calf’s tail, then threw a calf against a tree felling its
fruits”, and by Lynn Marie Ate (1978: 190) as “It seems you tied palm leaves
to a calf’s tail, and threw it to knock down ripe fruit”. Such translations agree
with the ancient commentaries. In the printed edition of Kiruṣṇasvāmi
Āyaṅkār, it is explained that the boys made fun of a calf by tying a palm-tree
leaf to its tail so that it ran helter-skelter. Then Kṛṣṇa whirls the kaṉṟu and
throws it towards the top of a tree.31
43The mention of a palm tree and of its leaves does not seem
understandable if we do not keep in mind that in the myth of Dhenuka the
animal is thrown away to the top of a palm tree after having been caught by
its hind legs and whirled around by its tail. The kaṭṭi of Periyāḻvār might have
been a way to imply that the animal (kaṉṟu) has been strongly linked (kaṭṭi) to
a palm leaf (ōḻai). But we may also be dealing here with a spurious passage.
Ōḻai is a hapax legomenon in the Tiv. and might not be understandable in the
present state of our knowledge.
• 32 Aṇṭāḻ, Tiruppāvai 24: kaṉṟu kuṇilāv eṟintāy! kaḻal pōṟṟi! (Tiv. 497).
• 33 See Nammāḻvār, Periya tiruvantāti 54: [...] āṉ īṉṟa / kaṉṟu uyara tām eṟintu
kāy utirttār tāḷ paṇi (...)
44The only occurrence we come across in Āṇṭāḷ’s poems is particularly close

to Cil. 17 as it also refers to the kaṉṟu being used as a kuṇil, a cudgel, as we
have already seen. The god has literally thrown the kaṉṟu as a stick to get
fruits.32 The myth is mentioned only once in Nammāḻvār, in whose work the
god fells the fruits by flinging what is undoubtedly a calf since it was born
from a cow.33 Poykaiyāḻvār (87) mentions the kaṉṟu legend once, without
making any significant annotation.
45Conversely, the eight allusions to the kaṉṟu legend in Tirumaṅkaiyāḻvār’s

hymns present variations that stem from a formula which appears rather well
fixed and allows for some interesting remarks. It associates iḷaṅkaṉṟu, the
young kaṉṟu, and the fruits of the wood-apple tree, viḷa, which are ripe or
unripe (see viḷaṅkāy, 3.8.5; 7.8.10, the green fruits of the viḷa, or viḷaṅkaṉi,
3.9.7; 7.4.2; 9.8.6; 9.10.7, the ripe fruits of the viḷa). The absolutive eṟintu
ends the formula. The rather common absence of case markers allows for
different interpretations of the phrase, which I detail here.
• 34 These lines have been translated in Dr. S. Jagathratchagan’s edition as

follows: “the Lord who thr (...)
• 35 Such a destructive power may be already mentioned in one of the hymns

to Tirumāl in the Paripāṭal;(...)
46The god could whirl a calf and cast it at the tree in order to destroy it by
means of the calf, or destroy both the fruits and the kaṉṟu by this act, as for
example in 7.10.8 and 9.8.6. The addition of koṇṭu to the kaṉṟu that we see in
Periyāḻvār 5.1.4 recurs in Tirumaṅkaiyāḻvār 3.9.7 and 7.4.2. Here, it is an
equivalent to the exceptional instrumental case marker of 8.6.9 (kaṉṟāl
viḷaṅkāy eṟintāṉūr). In all these passages, the kaṉṟu has been characterised
as an instrument of the action, and therefore, not as its main object. In other
words, the motif shifts towards a slightly different one, which can be defined
as the one of a fight against a tree. In 7.4.2, when the calf is used to get the
fruits, the tree is presented as a strong one: [...] mallal / komp’uruva viḷaṅkaṉi
mēl iḷaṅkaṉṟu / koṇṭu eṟinta kūttar pōlām,34 “Like a dancer, they say, who
felled the ripe fruits of the strong wood-apple tree with the young kaṉṟu you
whirled on top of [them]”. Such a scheme is still in conformity with the legend
of Dhenuka, in which the cowherds have to fight with the donkey because
they want the fruits. But in some of the stanzas of the Tiv., the fruits do not
appear any more as the aim of the action but as a side effect of the combat.
Their fall indicates the destructive power of the god Kṛṣṇa.35 Contrary to the
Sanskrit legend of Dhenuka, the fruits are unripe in several stanzas while it is
not said that the kaṉṟu has had any kind of demonic behaviour. Is the cudgel
with which fruits are felled a real opponent to Kṛṣṇa?
• 36 See Periyālvār Tirumoḻi patuvurai, p. 48.
47Considering such details, I would suggest that the Tiv. indicates a

progressive shift towards another type of struggle than the usual ones familiar
from the oldest Sanskrit texts. In the Tamil corpus, Kṛṣṇa’s fight implying a
kaṉṟu would have gradually become a fight against a demon-tree. Such a
mythological act is quite suitable for its integration into a Tamil literary context
nourished by the Caṅkam literature, in which the combat against the mango
tree-shaped Cūr is the most famous deed of the young god Murukaṉ, whom I
assume to be one of the models of the Cil. Kṛṣṇa. The role of trees is indeed
very important in the Tiv. The tree called kuruntu is another instance of the
prominent place of trees in the text, under different avatars. When this
kuruntu tree is mentioned in Cil., it is said that Kṛṣṇa breaks it; the
commentaries on Tiv. consider it a demonic figure. In the old commentaries
on the kaṉṟu theme in Periyāḻvār’s corpus, it is clearly explained that Kṛṣṇa
was then fighting against two adversaries. One is the kaṉṟu, while the other is
the tree, designated by the Sanskrit term asura in the maṇipravalam
commentaries.36
• 37 This tī could also come from tīmai, sweetness. Contrary to the hypothesis
developed here, it would (...)
• 38 The line has been translated in Dr. S. Jagathratchagan’s edition as

follows: “you swirled a demon- (...)
• 39 Two of Pūtattāḻvār’s stanzas present formulas very close to those of

Tirumaṅkaiyāḻvār’s (see Tiv. (...)
48To conclude on the kaṉṟu motif in the Tiv., the transformation of the viḷa
tree into an adversary of the god seems to be clearly acknowledged in the
original corpus itself, in one of the three occurrences found in Pūtattāḻvār.
There the viḷa tree is preceded by tī, “fire, poison”.37 The viḷa tree is
henceforth fiery, poisoned, “bad”: [...] kuḻāk kaṉṟu / tī viḷaviṉ kāykku eṟinta,38
“he has swirled a kaṉṟu as a stick at the unripe fruits of the poisonous wood-
apple tree” (Tiv. 2200).39 The circle closes here. The tree bearing delicious
fruits has become a poisonous enemy.
• 40 In the Periyāḻvār corpus alone, for example, the demoness Pūtanā is

encountered 13 times.
• 41 It is not possible to broach such a complicated issue as the internal and

external chronology of t (...)
49The internal chronology of the Tiv. is much too debatable for this corpus to
be the basis of a chronological ordering of the evolution of the myth of the
kaṉṟu. But it is noteworthy that the Tiv. presents a rather confused image,
formed out of several versions of what cannot be seen as one and the same
legend but has already started to split into two different combats in this
corpus. A tree is associated with an animal called kaṉṟu without, most of the
time, a clear distinction of who the opponent of the god is. It is also significant
that, with only four occurrences, the myth of Dhenuka is very rarely found in
this Vaiṣṇava corpus, while each of the other episodes of the northern Indian
legend is frequently mentioned.40 On the other hand, the kaṉṟu episode is
most often evoked by Periyāḻvār and Tirumaṅkaiyāḻvār. If in Periyāḻvār it
appears rather close to the myth of the Sanskrit Dhenuka, in
Tirumaṅkaiyāḻvār the kaṉṟu is more like the instrument of a combat that is
carried out against a tree, while the viḷa is presented as being poisonous in
Pūtattāḻvār. Most probably the kaṉṟu was understood as a calf by all the
Āḻvārs, even if we can think that in the specific case of Periyāḻvār there might
have been a certain hesitation about the species of the animal. We shall,
therefore keep in mind that Periyāḻvār’s is the first corpus to appear in the Tiv.
and that its relative simplicity of style could be accounted for by the early date
of his work.41
6. Sanskrit texts of South India
50This latter fact accounts for the mention of the vatsa in the Sanskrit texts
composed later in southern India. BhP 10.11, 41–43 narrates a brief version
of the vatsa-episode:
51kadācid yamunā-tīre vatsāṃś cārayatoḥ svakaiḥ / vayasyaiḥ kṛṣṇabalayor

jighāṃsur daitya āgamat // taṃ vatsa-rūpiṇaṃ vīkṣya vatsa-yūthagataṃ hariḥ
/ darśayan baladevāya śanair mugdha ivāsadat // gṛhītvāparapādābhyāṃ
saha-lāṅgūlam acyutaḥ / bhrāmayitvā kapitthāgre prāhiṇod gata-jīvitam / sa
kapitthair mahā-kāyaḥ pātyamānaiḥ papāta ha //
52“One day, on the banks of the Yamunā, a demon came, wishing to kill
Kṛṣṇa and his brother, who were driving their calves along with their friends.
Hari (Kṛṣṇa), having perceived this one who had taken the shape of calf
(vatsa) and had mixed in with the herd of calves, approached, as if somewhat
confused, showing it to Baladeva. Having seized it by his two hind legs
together with his tail, Acyuta whirled it and threw it on the top of a Kapittha
tree, bereft of his life. The giant body fell, bringing down the Kapittha fruits”.
53The kapittha is defined as the wood-apple tree, Feronia limonia, in Sanskrit
dictionaries and thus corresponds to the viḷa of the Tamil Tiv.
• 42 On this fight, cf. infra: 41–43. Note that here again Murukaṉ has most
probably played an importan (...)
54This account is short. It is preceded by a scene in which Kṛṣṇa and

Balarāma fight several bulls, and is followed by the fight against a demon-
bird, Baka, two elements that are unknown to Sanskrit texts written before the
BhP. The combat against a bird is indeed first attested in the Tiv., and its
oldest known representations in stone are found in the Tamil land itself.42 I
have already mentioned the bull contest several times. In other words, the
narrative context in which the vatsa episode made its first known appearance
in Sanskrit is strongly marked by a southern, Tamil, tradition.
• 43 See Couture (1991: 363–374) for a presentation and a French translation

of this passage. The text (...)
55The southern versions of the HV develop the episode of the vatsa at length
in what is presented now as appendix I. 11 of the critical edition of this text.43
This is an account of 150 verses. After performing a dance in honour of Śiva,
a Brahmin haunted by a Śaiva bhūta advises the herdsmen to destroy a
poisonous kapittha under the shadow of which calf-demons have found
shelter. Kṛṣṇa and Balarāma go to the tree, which is said to be covered with
large fruits, some ripe, some unripe. The herdsmen cause those fruits to fall
with sticks; the calf-demons attack them; Kṛṣṇa seizes them by their legs,
whirls them and hurls them at the tree. The kapittha, its fruits and the calves
die. Kṛṣṇa and the herdsmen celebrate their victory by taking a bath in the
river.
56The influence of the South Indian tradition is obvious in such an account,

from the haunting of the Śaiva Brahmin up to the demonic kapittha, which is
said to flower after a gap of 12 years, which corresponds to the gap of two
blossoms of a kuṟiñci, a tree belonging to the South Indian soil and its
literature. The ripe and unripe fruits found on the same tree remind us of the
kāy and kaṉi versions of the Tiv. viḷa. To attribute a precise date to these
southern versions of the HV would be unrealistic, but in the current state of
our knowledge we can surmise the BhP was known to their authors in a
version that must have been rather similar to the one we know today.
57The data expounded above do not suggest that the story of the combat
against a calf, a kaṉṟu, was first developed through a southern Indian oral folk
tradition. There are enough textual clues to suggest that the myth of the
combat against Dhenuka — a donkey named after a cow in Sanskrit texts, in
which it fights against one of the Vṛṣṇi brothers, Kṛṣṇa or Saṃkarṣaṇa — was
the source of the kaṉṟu motif in the Tamil texts. The fight against the calf that
appears in the Tamil Tiv. has been composed with allusions from the Sanskrit
Dhenuka myth found in the Tamil texts themselves. From Cil. 17 to the Tiv.
we observe a shift of the accent from the struggle against an animal to the
one against a tree in a structure that is more familiar to a Tamil literary
tradition, while a donkey called “little cow” in Sanskrit is converted into a calf.
The translation of the Sanskrit name Dhenuka with the term kaṉṟu — which in
Tamil could mean either a young donkey or a little cow, i. e. an animal that is
very close to a calf — has finally sanctioned the birth of a new legend, which
has spread throughout India.
7. The shadow-motif
58The kaṉṟu motif is only one of the numerous deeds ascribed to Kṛṣṇa in
Cil. 17, then in the Tiv., to be reshaped in the Sanskrit BhP. This feat is also a
rather minor one, but there are strong arguments to think that other themes of
the Kṛṣṇa legend, some appearing in many more lines of Cil. 17, would have
followed a similar literary path. I would broach two other examples of what I
shall call shadow-motifs. I suggest that in the process of translation-
adaptation to a Tamil literary context, an item from a northern Indian legend
has been transformed to such an extent that it would have given birth to two
motifs, the original one and its “shadow”. The shadow-motif is apparently
unknown to the oldest Sanskrit sources and often makes mysterious
appearances in the Cil. and later in the brief stanzas of the Tiv., in which it is
particularly difficult to reconstitute the narrative nodes of the legends; it is
then developed in later Sanskrit texts of southern India, i. e. in most cases
probably in the BhP for the first time. Shadow-motifs feed on a southern
Indian tradition that we know mainly through the Caṅkam corpus, and which
has originally nothing to do with the god Kṛṣṇa.
Fig. 3 Deogaḍh, 5th century (Madhya Pradesh; National museum of New
Delhi; photo, author).
59We will succinctly survey the Baka episode appearing in the BhP, which
scholars frequently consider as being referred to in the Tiv. under the form of
a combat against a bird, and subsequently the uprooting of the kuruntu tree,
mentioned twice in Cil. 17 and then regularly in the Tiv.
• 44 L. M. Ate (1978: 136, n. 11) stresses the originality of this fight, speaking
of an “indigenous Ta (...)
• 45 This bird of the oldest known lore of the legend represents a variant of an
ancient theme, featuri (...)
60The fight against a demon-bird called Baka in the BhP constitutes a clear
instance of a shadow-motif process.44 It represents the southern Indian
shadow of the struggle carried out against the demoness Pūtanā, who took
the shape of a bird, a śakuni/śakunī, in the HV, in the oldest known
representations of this fight and in the Jaina versions of Kṛṣṇa’s childhood,
amongst other data (Fig. 3).45
61The detail of the bird-like appearance of this demoness disappears from

the Tamil tales of the Tiv., which mentions the killing of Pūtanā and a fight
against a bird (puḷ), sometimes in the same stanza, while the South Indian
carvings start representing the combat against the bird from the 8th century
and Pūtanā as an ogress from the 8th or the 9th onwards (see Figs. 4 and 5).
62In South Indian sculptures, Pūtanā is never given the ornithological

features that she is at times provided with in North-Indian sites. In her case,
carvings outline the contours of the shadow-motif. Sculptures also put us on
the right track when we look at the carved representations of the kuruntu
myth, which I will very briefly refer to, as it is a fascinating case which calls for
the writing of a separate paper in its own right.
• 46 The uprooting of the Arjuna tree is recounted in HV 51 and VP 5.6. For

clear illustrations in the (...)
63It seems to me indeed that the breaking or bending (ocittaṉ) of the kuruntu
tree alluded to in Cil. 17.21 and 17.29 is the first known occurrence in Tamil of
the uprooting of the two arjuna trees, which is first told, to my knowledge, in
the Sanskrit HV and VP as one of the exploits of the child Kṛṣṇa. It also
appears as such in the Tiv.46 The stone carvings help us to see what has
occurred in this case. The shadow-motif of the kuruntu tree appears in the
Tiv. in the scene that is usually called the vastraharaṇa after its Sanskrit title,
in which Kṛṣṇa is seated on the top of a kuruntu to hand over the clothes to
the gopīs from whom he had stolen them.
• 47 This carving adorns the base of the early Cōḻa period temple of
Tirucceṉṉampūṇṭi. It can be safely (...)
64In the oldest known representations of an uprooting of trees by Kṛṣṇa in

the Tamil land, we see a young man standing between two trees that he
grasps very firmly (Fig. 6).47 This is different from the carvings in which the
HV version of the uprooting of the arjuna trees is illustrated with a crawling
baby trailing a mortar behind him, an act that uproots two trees that are
standing closely together as it passes between them (Fig. 7).
Fig. 4 Tiruveḷḷaṟai, 8th or 9th century (Tamil Nadu; photo, author).
Fig. 5 Brahmatēcam, 8th or 9th century (Tamil Nadu; photo, author).

Fig. 6 Tirucceṉṉampūṇṭi, 9–10th century A.D. (Tamil Nadu; photo, author).
• 48 For the breaking of the marutam, see Periyāḻvār’s decade, Tiv. 525–531,
Tirumaṅkaiyāḻvār, Tiv. 788 (...)
65Does the Kṛṣṇa from the Tamil land uproot the two trees? The doubling of
the trees and the position of the god in the middle of them correspond to the
northern Indian legend, whereas the Tiv. refers to different actions and
different tree species when it portrays the god as a destroyer of trees. I have
studied in detail the fight against the viḷa tree, which takes the place of a
paṉam (still mentioned sometimes in the Tiv.). The case of the breaking of the
Arjuna trees is rather similar. In the Tiv., Kṛṣṇa sometimes uproots the
marutam trees. As marutam is another name for the Arjuna tree, such an
uprooting is a clear transposition into Tamil of the uprooting of the Arjuna tree.
But the study of the motif featuring the bending or breaking of the kuruntu in
the Tiv. has led me to the conclusion that the breaking of the kuruntu in Cil.
should be understood as another transposition of the breaking of the Arjuna
trees.48 The marutam does not appear in the Cil. Then in the Tiv., where the
marutam has been recorded as the equivalent of the Sanskrit Arjuna tree, the
breaking of the kuruntu that had appeared in Cil. 17 has participated in the
creation of the vastraharaṇa story, narrated later at length in the Sanskrit
BhP.
Fig. 7 Somnāthpur, 12th century (Karnataka; photo, author).
66The picture in the Tiv. is rather blurred as it superimposes two different

images, one of them keeping some of its original characteristics, the other
constituting a first attempt to design a “shadow” of the same motif, the profile
of which became eventually rather different from its source.
8. Conclusions
67The prospective conclusions of this survey would plead for a reassessment

of the elements of northern Indian origin in Cil. 17 and, more broadly, for a
reconsideration of the beginnings of Kṛṣṇaite movements in South India.
68The elements from northern India, most probably Sanskrit and Prakrit
texts, as well as other, more concrete materials, such as images, were
adapted to a southern Indian cultural context, in which a young and beautiful
god, the Muruku-Murukaṉ of the Caṅkam literature, fights against demon-
trees in a literary universe where the herdsmen play the flute while bringing
back their herds. The question of a folk Kṛṣṇaism remains debatable. It may
have gradually spread in the sophisticated texts of the Caṅkam literature,
then flourished in the corpus of the Āḻvārs, in which a synthesis between such
popular Kṛṣṇaism and the northern Kṛṣṇa tradition may have been made. To
my mind, however, it is difficult to link Cil. 17 to such a suggestion. Each of
the supposed folk-motifs of this text has a very literary and more ancient
equivalent in the northern Sanskrit texts and a kind of counterpart in the
oldest known southern Indian literature. There is enough grounds to make us
think that Cil. 17 has been composed in a milieu that was as much obsessed
with literary issues as with devotional ones. Moreover, the devotional
character of the poem can itself be regarded as linked to a courtly sphere,
which would not only be a literary stage but also a place for the expression of
the rivalry between kings in South India. Cil. is divided into three cantos
representing the three legendary Tamil kingdoms of the Cōḻas, the Pāṇṭiyas
and the Cēras, and the absence of the Pallavas, who sponsored Sanskrit, is
noticeable. The dynastic Pāṇṭiya mythology appears infused with Kṛṣṇaite
themes. Could this have been a kind of response to the rather marked
Śaivism of some of the Pallavas, as the latter were the Pāṇṭiyas’rivals,
involved in a world of marked Sanskritic leanings in the northern part of the
Tamil land?
• 49 The Pāṇṭiyas have been the focus of recent discoveries and research
work. See Y. Subbarayalu and V (...)
• 50 On several occasions, the mullai series of the Kalittokai evokes the

Pāṇṭiya king. In mysterious a (...)
• 51 For the role the Cēras have been traditionally invested with, see Nilakanta
Sastri (1955: 115–45), (...)
69For all these reasons I would reconsider the relations linking northern and
southern Indian legends involving Kṛṣṇa. When Cil. 17 was composed,
devotion to Kṛṣṇa may have appeared quite recently in South India. The first
representations of Hindu gods are found within the Pallava territory, whose
kings are praised with beautiful Sanskrit eulogies. Moreover, the Pāṇṭiya
kingdom, the evocation of which begins and ends Cil. 17, and which is one of
the main backdrops of an epic narrating the destruction of Maturai, is held to
have played a particular part in the emergence of Caṅkam anthologies.49 Its
exact contribution to the Cil. is still to be investigated.50 But in the structure I
have outlined, the ancient tradition of a specific role attributed to the Cēras
could make sense.51 Kṛṣṇa is known for being clever and deceitful. In the Cil.
these characteristics are emphasised even more than usual and he finally
appears to be overshadowed by the goddess, who is praised in Cil. 12, and
to whom, under the form of Kaṇṇaki, the whole epic is eventually
consecrated. This is the story of the establishment of the cult of a goddess by
the Cēras. There might be more than literary truth contained in this scheme.
Bibliography
DOI are automaticaly added to references by Bilbo, OpenEdition's

Bibliographic Annotation Tool.
Users of institutions which have subscribed to one of OpenEdition freemium
programs can download references for which Bilbo found a DOI in standard
formats using the buttons available on the right.
The Bibliographic Export Service is accessible via institutions subscribing to

one OpenEdition freemium programs. If you wish your institution to become a
subscriber to one OpenEdition freemium programs and thus benefit from our
services, please write to:[email protected]
BIBLIOGRAPHY
I. Primary Sources
Harivaṃśa. Critical edition by Vishnu S. Sukthankar, S.K. Belvakar and

Parashuram Lakshman Vaidya. Vol. I, Introduction, critical text and notes.
Poona: Bhandarkar Oriental Research Institute, 1969. Vol. II, Appendices.
Poona: Bhandarkar Oriental Research Institute, 1971.
Iḷaṅkōvaṭikaḷaruḷiceyta cilappatikāramūlamum arumpatavuraiyum aṭiyārkku-

nallāruraiyum. Edited by U. Vē. Cāminātaiyar. 9th edition. Ceṉṉai: U. Vē.
Cāminātaiyar nūl nilaiyam, 2001 (1st edition 1892).
Jātaka. Edited and translated by Viggo Faussbøll, 6 volumes. London: Luzac

and Company, 1962–64 (1st edition 1877–97).
Mahābhārata. Critical edition by Vishnu Sitaram Sukthankar. Poona:

Bhandarkar Oriental Research Institute, 1933–1963.
Le Paripāṭal. Texte tamoul. Introduction, traduction et notes par François

Gros. Pondichéry: Institut Français d’Indologie, (Publications de l’Institut
Français d’Indologie no 35), 1968.
Periyāḻvār Tirumoḻi patuvurai, Mutalāyirattil mutal pirapantanakiya Periyāḻvār

aruḷicceytu, Periyāḻvār Tirumoḻi patuvurai. Edited by Śrī S. Kiruṣṇasvāmi
Āyyaṅkār. Ceṉṉai, 1995.
Tivviyappirapantam Śrī nālāyira tivya pirapantam aintām vētam. The Sacred

Book of Four Thousand (Four Thousand Hymns of Twelve Alwars and
Commentary). By Dr. S. Jagathratchagan. English rendered from the Sacred
Book by Sri Rama Bharati. Chennai: Āḻvārkaḷ āyvumaiyam, 2002 (1st edition
1993).
Viṣṇumahāpurāṇaṃ Viṣṇucittyātmaprakāśākhya Śrīdharīya

Vyākhyādvayopetaṃ. (Viṣṇupurāṇa with Śrīdhara’s commentary). [No editor].
Bombay: Veṅkateśvara Press, 1967.
II. Secondary Sources
Ate, Lynn Marie (1978). Periyāḻvār’s Tirumoḻi — a Bāla Kṛṣṇa Text from the
Devotional Period in Tamil Literature. PhD thesis. The University of
Wisconsin-Madison.
Bansat-Boudon, Lyne (2006). Bhāsa, (entry) in Lyne Bansat-Boudon (Editor),

Théâtre de l’Inde ancienne, pp. 1181–1192. Paris: Gallimard (Bibliothèque de
la Pléïade).
Burnouf, Eugène (1840–1898). Le Bhâgavata purâṇa ou histoire poétique de

Krishna. 5 volumes. (Vol. 4 translated by Eugène-Louis Hauvette-Besnault,
vol. 5 by Eugène-Louis Hauvette-Besnault and Alfred Roussel). Paris:
Imprimerie Royale.
Couture, André (1991). L’enfance de Krishna. Paris: Les Editions du Cerf et

Les Presses de l’Université Laval.
Couture, André (1992). “Le Bālacarita attribué à Bhāsa et les enfances
hindoues et jaina de Kṛṣṇa”. Bulletin d’Etudes Indiennes, 10, pp. 113–144.
Couture, André (2006). La Geste du jeune Kṛṣṇa. In Lyne Bansat-Boudon

(Editor), Théâtre de l’Inde ancienne. Paris: Gallimard (Bibliothèque de la
Pléïade).

Couture, André and Charlotte Schmid (2001). “The Harivaṃśa, the Goddess
Ekānaṃśā and the Iconography of the Vṛṣṇi Triads”. Journal of the American
Oriental Society, 121, 2, pp. 173–192.
DOI : 10.2307/606559
Daniélou, Alain (1967). Shilappadikaram: The Ankle Bracelet. London: Allen

& Unwin.
Daniélou, Alain and R.S. Desikan (1961). Le Roman de l’anneau. Paris:

Gallimard-Unesco (Connaissance de l’Orient).
Dehejia, Vidya (1995a). Slaves of the Lord. The Path of the Tamil Saints.
Delhi: Munshiram Manoharlal Publishers (1st edition 1988).
Dehejia, Vidya (1995b). “Iconographic Transference between Kṛṣṇa and three

Śaiva Saints.” In John Guy (Editor), Indian Art and Connoisseurship, Essays
in Honour of Douglas Barrett, pp. 141–149. New Delhi-Ahmedabad: Grantha
Corporation and Indira Gandhi National Centre for the Arts in association with
Mapin Publishing.
Dikshitar, V.R. Ramachandra (1939). The Śilappadikāram. Madras: Oxford

University Press.

Dubianski, Alexander M. (2000). Ritual and Mythological Sources of Early

Tamil Poetry. Groningen: Forsten (Gonda Indological Series no 8).
DOI : 10.1163/9789004486089
Francis, Emmanuel (2009). Le discours royal. Monuments et inscriptions

pallava (IVeme–IXeme siecles). PhD Dissertation. Louvain-la-Neuve: Institut
Orientaliste, Université Catholique de Louvain.
Francis, Emmanuel (forthcoming). Le discours royal. Monuments et

inscriptions pallava (IVème–IXème siècles). Louvain: Peeters (Publications de
l’Institut Orientaliste de Louvain).

Francis, Emmanuel and Charlotte Schmid (2010). “Preface”. In Pondicherry

Inscriptions. Part II. Translation, Appendices, Glossary and Phrases by G.
Vijayavenugopal, pp. v–vxlvii. Pondichéry: Institut Français de
Pondichéry/École française d’Extrême-Orient (Collection Indologie no 83.2).
DOI : 10.1016/S0378-2166(99)89001-8
Edholm, Erik af, and Carl Suneson (1972). “The Seven Bulls and Kṛṣṇa’s
Marriage of Nīlā/Nappiṉṉai in Sanskrit and Tamil Literature”. Temenos:
Studies in Comparative Religion, 8, pp. 29–53.
Hardy, Friedhelm H. (1983). Viraha-Bhakti, The Early History of Kṛṣṇa

Devotion in South India. Delhi: Oxford University Press.

Hawley, John Stratton (1983). Krishna, The Butter Thief. Princeton: Princeton
University Press (Indian edition, Delhi: Oxford University Press, 1989).
DOI : 10.1515/9781400855407
Hudson, D. Dennis (1982). “Piṇṇai, Krishna’s Cowherd Wife”. In John Hawley

and Donna Wulff (Editors), The Divine Consort. Rādhā and the Goddesses of
India. Berkeley: Religious Studies Series — Delhi: Motilal Banarsidass.
Hudson, D. Dennis (1994). “Vraja among the Tamils. A Study of Bhāgavatas

in Early South India”. Journal of Vaishnava Studies, 3, 1, pp. 113–140.
Hudson, D. Dennis (2002). “Rādhā and Piṉṉai: Diverse Manifestations of the

Same Goddess”. Journal of Vaishnava Studies, 10, 1, pp. 115–153.
Nilakantha Sastri, K.A. (1955). The Cōḷas. Madras: University of Madras

Press (2nd edition).
Obeyesekere, Gananath (1984). The Cult of the Goddess Pattini. Chicago-

London: University of Chicago Press.
Orr, Leslie (2007). “Cholas, Pandyas, and ‘Imperial Temple Culture’ in

Medieval Tamilnadu”. In Adam Hardy (Editor), The Temple in South Asia, pp.
83–104. London: British Academy.
Parthasarathy, R. (1993). The Cilappatikāram of Iḷaṅkō Aṭikaḷ. An Epic of

South India, Translated, with an Introduction and Postscript. New York:
Columbia University Press. (Translations from the Asian Classics) (Indian
edition: Penguin Books India, 2004).
Podzeit, Utz (1992). “A Philological Reconstruction of the Oldest Kṛṣṇa-Epic”.

Wiener Zeitschriftfür die Kunde Südasiens und Archiv für indische
Philosophie, 36, pp. 55–59.
Ramanujan, A.K. (1967). Interior Landscapes, Love Poems from a Classical

Tamil Anthology. Bloomington: Indiana University Press (Indian edition:
Oxford University Press, 1994).
Rocher, Ludo (1986). The Purāṇas. Wiesbaden: Otto Harrassowitz (A History

of Indian Literature, vol. II, fasc. 3).
Schmid, Charlotte (1999). “L’avatāra de Viṣṇu et le dieu du Mahābhārata:

représentations anciennes de Kṛṣṇa sur des haltères”. Bulletin de l’École
française d’Extrême-Orient, 86, pp. 65–104.

Schmid, Charlotte (2002). “Aventures divines de Kṛṣṇa: la līlā et les traditions
narratives des temples cōḻa”. Arts Asiatiques, 57, pp. 33–50.
DOI : 10.3406/arasi.2002.1479
Schmid, Charlotte (forthcoming). “Sur le chemin de Kṛṣṇa, la flûte et ses

voies”.
Subbarayalu, Y. and V. Vedachalam (2007). “Pāṇṭiyaṉ neṭumāṟaṉiṉ

ilayāṉputtūrc ceppēṭu (Copper-plates of Pandya king Nedumaran)”. Āvaṇam,
18 (July), pp. 1–16.
Subrahmanian, S.V. (1965). Descriptive Grammar of Cilappatikāram.

Trivandrum: University of Kerala.
Subrahmanyam, Ka. Nā. (1977). The Anklet Story: Silappadhikāram of Ilango

adigal. Delhi: Agam Prakashan.
Tamil Lexicon. Published under the Authority of the University of Madras. 6

volumes, 1924–1939.
Tieken, Herman (1993). “The So-called Trivandrum Plays Attributed to

Bhasa”. Wiener Zeitschriftfür die Kunde Südasiens, 37, pp. 5–44.

Tieken, Herman (2001). Kāvya in South India: Old Tamil Caṅkam Poetry.
Groningen: Egbert Forsten (Gonda Indological Studies no 10).
DOI : 10.1163/9789004486096
Trautman, Thomas R. (1981). Dravidian Kinship. Cambridge: Cambridge

University Press.
Vaiyapuri Pillai, S. (1988). History of Tamil Language and Literature. Madras:

New Century Book House (2nd revised edition).
Williams, Joanna G. (1982). The Art of the Gupta India, Empire and Province.
Princeton: Princeton University Press.

Zvelebil, Kamil Veith (1975). Tamil Literature. Leiden-Köln: Brill (Handbuch

der Orientalistik no 2.2.1).
DOI : 10.1163/9789004492981

Zvelebil, Kamil Veith (1995). Lexicon of Tamil Literature. Leiden-New York-

Köln: Brill (Handbuch der Orientalistik no 2.9).
DOI : 10.1163/9789004491731
Notes
1 See Edholm and Suneson (1972), the magnum opus of Friedhelm Hardy
(1983) and several papers of the late D. Dennis Hudson (1982, 1994, 2002),
among others. The latter focuses usually on Piṉṉai, the consort of Kṛṣṇa in
Cil. 17. His analysis of this character leads him, however, to broaden his
approach to speak of the courtly telling of the legend of Kṛṣṇa that has
adapted the Tamil literary theme of cowherds conquering bulls to win brides.
2 See Hardy (1983: 179); see also Hudson (2002: 136–138).
3 F. Hardy (1983) has developed such a point in his book, but it would be
impossible to summarise his views in a footnote: pp. 172–183 are exclusively
devoted to Cil. 17.
4 I have not taken into account the Bālacarita here, a work that is sometimes
considered one of the first mentions of Kṛṣṇa in the southern texts (see, for
instance, Hardy 1983: 80–82), as it can now safely be assumed that this play
was composed later than was initially thought. For the recent debates on the
set of plays the Bālacarita is part of, see Lyne Bansat-Boudon (2006), who
examines the hypothesis made by Herman Tieken (1993). Regarding more
particularly the story of Kṛṣṇa, André Couture presents convincing arguments
that allow us to consider the Bālacarita as a post-7th-century work (see also
Couture 1992).
5 I cannot give all the features of the Kṛṣṇa of Cil. 17 with the corresponding
archaeological material in detail here. A few of them have been studied by
Schmid (2002 and forthcoming; see also infra: 42–46).
6 This text can be considered from various other literary and religious angles
and I am aware how potentially provocative any general statement bearing on
the Cil. might be. Ancient Tamil poetological works do not consider this epic
as part of the Caṅkam corpus itself. I use the term Caṅkam here implying that
the latest chronological limit for this corpus can be considered the 7th century,
to which I am inclined to locate at least an important part of the composition
of the Cil. Zvelebil gives various definitions and dates for this text in his works
(see Zvelebil 1975: 110–115; 1995: 144–148, for a compendium of most of
the discussions). In Alexander Dubianski’s opinion (2000), Cil. is a post-
Caṅkam text; Tieken highlights Pāṇṭiyan elements in the poem and
concludes it cannot be earlier than Caṅkam poetry (Tieken 2001: 206);
according to R. Parthasarathy (1993: 7), it is a 5th-century poem, while it
should be noted that S. Vayapuri Pillai (1956: 152) considered it a post-8th-
century work.
7 See Edholm and Suneson (1972), Hudson (1982: 239–240) and Gros
(1968: li).
8 This human wife is unknown from the older Sanskrit texts. The identity of
the two other characters linked to Kṛṣṇa noticeably rests on that of Kṛṣṇa
himself: they are indicated by terms of kinship, the “elder one” for Muṉṉai, the
“junior one” for Piṉṉai. Hudson (2002) is the sole author who seems to have
seriously considered the parallel which suggests that there is an elder and a
younger family member. On the possibility of the junior female character
being a sister, see infra: 21–24.
9 We rarely come across deities in the Caṅkam corpus, except in the

Paripāṭal and the Tirumurukāṟṟupaṭai. The poems in which gods appear have
been recognised as belonging to the latest layers of the corpus. To such late
strata belongs the motif of the vēlaṉ ordering the celebration of a ritual in
honour of Murukaṉ to cure the heroine of a sickness which is nothing other
than love. It displays similarities in the way relations with deities are
presented in Cil. 17. Deities — or rather rituals celebrated in their honour —
are of no use; these rituals are composed of different kinds of dancing
performed by characters who are presented as belonging to a rural
background.
10 Before the dance involving Piṉṉai, Cil. 17 associates seven bulls of

different colours with girls who participate in this dance, in what is usually
interpreted as a proto-version of Piṉṉai’s legend as it appears in the later Tiv.
The motif of Piṉṉai and the one of the fight of a young lad against bulls would
be represented in Cil. 17 for the first time as Kṛṣṇa-related themes. However,
these legendary elements have probably followed a literary pattern that is
somewhat older than the Cil., provided by the mullai series of the Kalittokai
(100–116), one of the latest works of the Caṅkam corpus. Kalittokai features
the Lover as a cowherd fighting against bulls to gain the hand of the heroine.
The scene is set in a herdsmen’s camp, which is of a highly mullai flavour as
it follows the conventions of the mullaittiṇai, the jasmine “interior landscape”,
in a rather scrupulous way. Tiṇais are psychological lands ruled by a certain
love emotion and associated with a precise deity. I borrow the expression
“interior landscape” from the works of A.K. Ramanujan whose poetic
translations are largely accountable for the fame that the concept of tiṇai has
acquired in our days. The whole history of this concept stirs up debates that I
cannot focus on here and I refer the readers to the works of A.K. Ramanujan
(especially Ramanujan 1967) and Hardy (1983: 119–238) for more specific
information on the mullai or jasmine tiṇai. Kṛṣṇa is the god of the mullaittiṇai,
the jasmine (mullai) landscape, which is the land linked with the longing born
out of the separation of the lovers. The cowherds and their various activities
are some of the characteristics of this tiṇai. In Kalittokai 100–116, the
elements of the mullai theme that can be found in older poems seem to be
gathered around a new narrative thread that gives great coherence to this
“interior landscape”. It describes how, in order to gain the hand of a girl, the
young herdsman fights against bulls. Kalittokai 101 and 102 describe a dance
called kuravai, which is danced by the herdsmen and is rather close to the
kuravai of Cil. 17. Kalittokai 102, 103 and 104 begin and end with eulogies or
wishes addressed to the Pāṇṭiya king, just like Cil. 17, which opens and ends
with such a eulogy. Kalittokai 102 quotes Māyōṉ’s (Mayavāṉ) fight against a
horse, the oldest known exploit of Kṛṣṇa’s childhood in Sanskrit texts (Schmid
1999). The different colours given to the bulls there remind us of the seven
colours ascribed to the bulls of Cil. 17.6–12. Finally, Kalittokai 105 mentions a
koṉṟai flute, a musical instrument played by the Kṛṣṇa of Cil. 17.19 in the very
stanza of the kaṉṟu motif I will analyse below (infra: 27).
11 It should be said, however, that while this legend is often mentioned in the
Tiv., as far as Cil. 17 is concerned, in order to make such a claim one has to
gather bits of allusions and references scattered in the whole canto. Kṛṣṇa’s
breaking of a kuruntu tree on the one hand and the shame of a girl whose
clothes had been hidden away by her lover while she was bathing in the river,
on the other hand, are not directly connected in Cil. 17. This theft of the
herdswomen’s clothes has been thoroughly analysed by Hardy who has
shown it was introduced into the Sanskrit narrative tradition through the BhP,
in which the northern Indian narrative scheme probably underwent an
important reworking. Regarding the constitution of the motif, I would
sometimes differ from Hardy’s theory, see infra: 47–48.
12 In the southern recensions of the HV, Kṛṣṇa marries a young human girl
called Nīlā, the Black one, as if she was a kind of female double of Kṛṣṇa.
Nīlā becomes Kṛṣṇa’spouse after a contest with seven bulls (appendix I. 12
of the HV critical edition). A rather exact link with the Kalittokai mullaittiṇai
series is seen here. To underline the South Indian identity of Nīlā, I stress that
the young black girl can represent the figure of the younger cross-cousin on
the maternal side, with whom one should marry in a southern context (cf.
Trautman 1981). The Piṉṉai-Nīlā motif is, however, of such an intricate nature
that I cannot do it justice here. I would just point out that we do come across
the story of a younger one linked to the family of Kṛṣṇa in the text of the
critical edition of the HV. In this northern recension Kṛṣṇa has indeed a
younger sister, Ekānaṃśā, as she is most often referred to in the HV, see
Couture and Schmid (2001). Piṉṉai, “the young one”, the term that
designates the girl dancing with the Elder (Muṉṉai) and the Black one
(Māyavaṉ) in Cil. 17, suits both Nīlā, Kṛṣṇa’spouse in the South, and
Ekānaṃśā, his sister in the North — and in HV 64, Kṛṣṇa fights against a
bull...
13 I am grateful to Eva Wilden who was kind enough to look for references to
flautists in Caṅkam literature. She has found more than thirty mentions of the
herdsmen’s flute, most of them occurring in a context fitting the mullai theme
(Wilden, personal communication, April 2009).
14 On the vastraharaṇa motif, see Hardy (1983: 193–196; 497–501;

512–516) and infra: 45–46.
15 See the commentaries given in the edition of U. Vē. Cāminātaiyar

(451.18).
16 See Hawley (1989: 311–375, appendix A), for a list of sites where a fight
against a calf and a fight against a donkey (among other deeds of Kṛṣṇa)
were represented in sculptures before 1500 CE; the list focuses specifically
on northern India, however, and cannot be considered as complete regarding
South India.
17 The Gwalior museum exhibits a representation of a distinctive tāla tree,

with an animal stuck on its top. The head and the front part of the animal are
missing. Its rather noticeably big male genitals do not fit the appearance of a
calf but would suit a donkey’s, and this piece is considered as a
representation of the Dhenuka of the Sanskrit legend. The date of the pillar is
problematic as it is based on considerations of style only, while the
comparative corpus is too small to make us feel we are on safe grounds with
only these stylistic considerations to go by. In my opinion this carving can be
dated between the 2nd century and the 5th century CE.
18 The text is given as it appears in U. Vē. Cāminātaiyar’s edition that I have

used; see this edition, p. 441.
19 I refer here to the translations of the Cil. into English by Ramachandra

Dikshitar (1939), Daniélou (1967; the French translation of Daniélou-Desikan
[1961] has “veau”), and Parthasarathy (1993). Ka. Nā. Subrahmanyam
(1977) has “youngling” (“Māyavan used an youngling [sic] from the herd as a
pole to knock down all the fruits in the orchard.”). In his glossary of the Cil., S.
V. Subrahmanian (1965) gives “calf” for all the passages where kaṉṟu
appears. For a general example of what sometimes, could be an unconscious
tendency to add an asura in the scene, in the case of the Tiv. see the
translations of the Periyāḻvār corpus given by Lynn Marie Ate (1978: 136,
explaining in fn. 11 that the kaṉṟu is the demon vatsa in the Sanskrit tradition;
see also the table, p. 367) or by Dr. S. Jagathratchagan (1993). The latter
often attributes Sanskrit names to the demonic opponents of Kṛṣṇa in the
English translation. kaṉṟu is therefore translated as the “calf Vatsa” or the
“asura Vatsa”, i.e. the [calf] named Calf-demon in Sanskrit (see
Jagathratchagan 1993: 37: “O, Dark hued calf of mine! All the hard wood-
apples in the deep forest fell when you threw the disguised calf Vatsasura
against the Asura tree”, kāṉaka vīl viḷaviṉkāy utirak karuti / kaṉṟatu koṇṭ(u)
eṟiyum karuniṟa eṉ kaṉṟē; on this stanza, see also infra, n. 28). I have used
the text of the Tivviyappirapantam published by Dr. S. Jagathratchagan
(where the sandhi is deleted). In my research, I have also used the digital
texts published on the site of Project Madurai, http://www.projectmadurai.org/.
20 Hardy (1983: 175) translates the Cil. passage in the following way:
“Māyavaṉ who with the calf as his club made the fruits fall down”, adding a
footnote (note 188): “Calf~Dhenuka, the demon whom Kṛṣṇa threw up into a
tree”. Later (220), he writes that in Cil. 17, Māyavaṉ churned the milk-ocean
and then killed Dhenuka. The name Dhenuka does not appear in the whole of
Cil., and the reference given in Hardy’s appendix VII (614) does correspond
to Cil. 17.19, that is to say to the kaṉṟu motif. Even if the order in which the
mythical events are said to have happened is the reverse (the churning of the
ocean comes after the kaṉṟu motif in Cil. 17) in this chapter, I think that Hardy
refers here to the kaṉṟu motif: it seems that for this author, Dhenuka would be
a calf — something which this demon is not, as far as the known Sanskrit
texts are concerned, but which is a plausible explanation for the name itself,
infra: 30.
21 dāruṇo dhenuko nāma daityo gardabharūpavān / kharayūthena mahatā

vṛtaḥ samupasevate // sa tat talavanaṃ ghoraṃ gardabhaḥ parirakṣati /
nṛpakṣiśvāpadagaṇāṃs trāsayānaḥ sa durmatiḥ // 57.12–13 [...]; tābhyām
eva sa jagrāha padbhyāṃ taṃ daityagardabham / āvarjitamukhaskandhaṃ
prairayat tālamūrdhani // sa bhagnorukaṭigrīvo bhagnapṛṣṭho durākṛtiḥ /
kharas tālaphalaiḥ sārdhaṃ papāta dharaṇītale // 57.19–20.
22 ariṣṭo dhenukaś caiva cāṇūraś ca mahābalaḥ / aśvarājaś ca nihataḥ

kaṁsaś cāriṣṭam ācaran, Mbh 5.128.46 (“He has slain Ariṣṭa, Dhenuka, the
powerful Cāṇūra, Aśvarāja [the king of horses], and the evil-doer Kaṃsa”).
23 dhenukaḥ sa mahākāyo dānavaḥ sumahābalaḥ / nihato vāsudevena

gavāṃ trāṇāya durmatiḥ, HV 96.39.
24 On the date of the Bālacarita, see above, note 5.
25 The demonic donkey is then placed as the guardian of the city of

Dvāravatī.
26 For a critical presentation of these posts, see Williams (1982: 55–58).
27 For references to the text, see above, n. 20; see Periyāḻvār 1.5.4; 2.3.10;
2.4.8; 2.5.5; 3.1.6; 3.3.7. With, for instance, 2.3.10, ēr viṭai ceṟṟu iḷaṅ kaṉṟu
eṟintiṭṭa (“O Lord who killed the bull Arishtanemi and the calf Vatsasura,”
according to the edition of Dr. S. Jagathratchagan [a more literal translation
could be “who has thrown the young kaṉṟu [calf or donkey] after killing the
bull”]) and 2.10.4, tēṉukaṉ āvi cekuttup paṉaṅkaṉi / tāṉ eṟintiṭṭa taṭam
peruntōḷiṉāl (“He flung the asura Dhenuka against a palm tree and killed him”,
according to Dr. S. Jagathratchagan’s edition). The fact that here again this
deed has been attributed to Kṛṣṇa himself and not to his brother is noticeable.
28 Here is the whole stanza:

vāṉavartām makiḻa vaṉ cakaṭam uruḷa
vañca mulaippēyiṉ nañcam atu uṇṭa vaṉē!
kāṉaka vīlviḷaviṉ kāy utirak karutik
kaṉṟu atu koṇṭu eṟiyum karuniṟa eṉkaṉṟē
tēṉukaṉum muraṉum tiṇtiṟal vennarakaṉ
eṉpavar tām maṭiyac ceru atirac cellum
āṉai! eṉakku orukāl āṭuka ceṅkīrai
āyarkaḷ pōrēṟē! āṭukā aṭukavē.
Dr. S. Jagathratchagan’s edition gives the following translation: “The gods
rejoice when you smote the cart Sakatasura and drained the life out of the
deceitful ogress Putana. O, Dark hued calf of mine! All the hard wood-apples
in the deep forest fell when you threw the disguised calf Vatsasura against
the Asura tree, then killed the notorious Dhenuka, Mura and Narakasura in
battles that shook the Earth. O My elephant, dance! Dance the Senkirai.” L.
M. Ate (1978: 136) translates it as follows: “Deciding to knock down the
woodapples in the grove, you grabbed a calf and threw it, O my little black
calf. You fought ‘til the battlefield shook, when you slayed the demons named
Dhenuka, Mura, and cruel courageous Naraka. Little elephant, just once for
me, play the Ceṅkīrai dancing game. O Battle bull of the cowherds, dance.
Please dance!” According to R. Varada Desikan, the Vaiṣṇava specialist and
devotee who works in the Pondicherry centre of the EFEO, Dhenuka is a
donkey-demon, while kaṉṟu designates a demonic calf that has nothing to do
with this Dhenuka.
29 See infra, n. 28; in the old commentaries of the edition of Kiruṣṇasvāmi

Āyyaṅkār, it is said that Kṛṣṇa had to fight a demon who takes two forms, one
as a calf and the other as a tree. The god defeats both at the same time (p.
48). These commentaries are thought to belong to the 13th century (no critical
edition of them has so far been published) and the story they narrate is close
to the BhP and to the southern versions of the HV, see infra: 38–40. The
discussions about the kaṉṟu and the myth of Dhenuka in these commentaries
indicate that the story is not clear for their authors.
30 atu shows that the kaṉṟu should be considered a second case and koṇṭu
could then be seen as a kind of prefix to the verbal form eṟiyum; but koṇṭu
could also allow us to consider kaṉṟu as a kind of instrumental case.
31 In the opinion of the old commentators and their present heir R. Varada
Desikan, this is one of the boyish games of Kṛṣṇa who ties a bunch of palm
leaves to the tail of a calf and then makes it run away noisily (see
Kiruṣṇasvāmi Āyyaṅkār 1995: 141). L. M. Ate (1978: 190) adds a footnote to
her translation to explain that this refers to the “Vatsāsura” episode, i.e. to the
calf-demon episode told in the BhP.
32 Aṇṭāḻ, Tiruppāvai 24: kaṉṟu kuṇilāv eṟintāy! kaḻal pōṟṟi! (Tiv. 497).
33 See Nammāḻvār, Periya tiruvantāti 54: [...] āṉ īṉṟa / kaṉṟu uyara tām eṟintu
kāy utirttār tāḷ paṇintōm. See also Tirumaḻicai 107 (Tiv. 858), in which
Dhenuka is mentioned among other demons of whom only a few are named.
34 These lines have been translated in Dr. S. Jagathratchagan’s edition as

follows: “the Lord who threw the demon-calf against the wood-apple tree and
destroyed both”. The author is probably influenced by the old commentaries,
see above, n. 33.
35 Such a destructive power may be already mentioned in one of the hymns

to Tirumāl in the Paripāṭal; cf. Tirumāl II. 43–44, kuruku paṟiyā nīḷirum
paṉaimicaip / palapati ṉāyiraṅ kulaitarai utirva pōl, in which the heads of the
enemies of Māl, which fall down onto the earth, are compared to the fruits
that are on the top of a borassus tree.
36 See Periyālvār Tirumoḻi patuvurai, p. 48.
37 This tī could also come from tīmai, sweetness. Contrary to the hypothesis
developed here, it would then indicate the sweetness of the tree. However,
the poisonous nature of the tree seems to be much more probable.
Sweetness could be said of the fruits but doing so for the whole tree does not
make sense, all the more so since the fruits which are referred to here are
unripe ones. The poisonous character of the tree is clearly acknowledged in
the southern versions of the HV, infra: 39.
38 The line has been translated in Dr. S. Jagathratchagan’s edition as

follows: “you swirled a demon-calf and dashed it against a demon wood-
apple tree”.
39 Two of Pūtattāḻvār’s stanzas present formulas very close to those of

Tirumaṅkaiyāḻvār’s (see Tiv. 2349, 2352).
40 In the Periyāḻvār corpus alone, for example, the demoness Pūtanā is

encountered 13 times.
41 It is not possible to broach such a complicated issue as the internal and

external chronology of the Āḻvārs here. For reasons involving mythological
content but also stylistic ones, I am not convinced of the relevance of the
chronological frame presented by Hardy (1983: 267–268) who places
Periyāḻvār in the 9th century, after Tirumaṅkaiyāḻvar.
42 On this fight, cf. infra: 41–43. Note that here again Murukaṉ has most
probably played an important part in the establishment of the motif, as he is
credited with a fight against a bird since the time of the Caṅkam literature.
However, in this case, such a fight could also have been part of a broad, pan-
Indian pattern according to which birddemons were more specifically
opposed to the gods linked with infancy and youth, see infra, n. 45.
43 See Couture (1991: 363–374) for a presentation and a French translation

of this passage. The text is attested in two manuscripts in Devanāgarī and in
all the manuscripts from the South. I give two significant extracts here: tāv
ubhāv api saṁprekṣya rāmakṛṣṇau viṣadrumam / pakvair apakvair bālaiś ca
sarvataḥ saṁvṛtaḥ phalaiḥ // saṁrabdhau kāṣṭhapāṣāṇaiḥ śātayām āsatuḥ
phalam / ucchritya bāhū vipulau nirbhujya daśanacchadam // HV appendix I.
11. l. 214–217; […] evaṁ vatsair abhighnantaṁ sarve vatsā viṣadrumam /
kṛṣṇaṁ dṛṣṭvābhyadhāvanta visṛjyānyān vimarditum // tān sarvān ekahastena
saṁdhārya sa mahādyutiḥ / apareṇāhanad vatsair ucchritya balavān drumam
// ibid. 251–254.
44 L. M. Ate (1978: 136, n. 11) stresses the originality of this fight, speaking
of an “indigenous Tamil oral tradition”.
45 This bird of the oldest known lore of the legend represents a variant of an
ancient theme, featuring female demons as birds that come to kill newborn
babies, both in texts and in the oldest known carvings of the so-called
mātṛkās in North India. They personify childhood illnesses. The ogresses of
the first Mathurā school of sculpture have often been provided with a bird’s
face. These carvings belong to the period between the 1st and the 3rd
centuries CE.
46 The uprooting of the Arjuna tree is recounted in HV 51 and VP 5.6. For

clear illustrations in the Tiv. in which kuruntu trees represent a transposition of
this “Arjuna-trees” myth, see, for example, Tirumaḻicai, kāytta nīḷ viḷaṅkaṉi
utirttu / etirnta pūṅ kuruntam / cāyttu, “[you] who felled the fruits of the tall viḷa
tree once they had ripened, having broken the Kuruntu trees which were
yoked as they have grown together...” (Tiv. 788). In Periyāḻvār 4.4.7
(kuruntam oṉṟ(u) ocittāṉ) the formula is the same as in Cil. 17, thus it is also
ambiguous; see also, infra, n. 48.
47 This carving adorns the base of the early Cōḻa period temple of
Tirucceṉṉampūṇṭi. It can be safely dated (on the basis of the inscriptions) to
the first half of the 10th century at the latest, or most probably to the end of
the 9th century.
48 For the breaking of the marutam, see Periyāḻvār’s decade, Tiv. 525–531,
Tirumaṅkaiyāḻvār, Tiv. 788, 1018, 1222, etc. The kuruntu, which is one of the
markers of the mullaittiṇai in Tamil poetology, appears frequently in the Tiv., in
which it is also linked with the stealing of the clothes of the gopīs. In the Tiv.
Kṛṣṇa sits on a kuruntu, later replaced by a nīpa tree in the BhP, when the
god hides the clothes of the herdswomen who have gone bathing. Such a
legend is absent in pre-BhP Sanskrit texts. Hardy has accurately shown the
vastraharaṇa story to be a typical southern Indian motif, the introduction of
which into the Sanskrit corpus through the BhP has necessitated a complete
reworking of the sequence of time in the latter work (Hardy 1983: 499–505).
According to Hardy (1983: 161, 169, 193–196), the vastraharaṇa was a folk
variant of an older Caṅkam motif: young ladies would have been given leaf-
garments taken from a tree, which Hardy thinks is the one said to be broken
in the Tamil texts in the sense that it has had its leaves plucked off. Kṛṣṇa,
incarnating the young god of love, would be the ideal provider of such leaf-
garments, which he obtains by breaking a kuruntu tree. According to Hardy,
one should consider that the transformation of the old motif would have
produced a kind of reversal of the original situation: seated on a tree, Kṛṣṇa
steals the clothes instead of giving leaf-clothes. My conclusions are different
however, partly because the hiding of the clothes mentioned in the Tiv. and in
the Cil. is not yet linked with the kuruntu in the Cil., partly because the motif of
the leaf-clothes does not seem to be firmly established and, also, because
the list of internal contradictions of such a line of evolution of the theme
appears rather long: if you destroy the tree, you cannot be seated on it to
hand over clothes; if you are interested in the leaves, there is no need to
break or uproot the tree; the tree mentioned in the Tiv. does not correspond to
the one cited in the BhP, etc. The following lines from Akanāṉūṟu 59, quoted
and commented upon by Hardy (1983: 193–194), are greatly enlightening
from this point of view, even if my conclusions would be different from those
of this author, whose translation I borrow: “... like Māl who broke the tree by
stamping on it so that the cowherd girls might dress themselves with its cool
leaves, on the spacious sandy bank of the Toḻunai in the North, rich with
water” (vaṇ puṉal toḻunai vār maṇal akaṉ tuṟai / aṇṭar makaḷir taṇ taḻai uṭīiyar /
maram cela mititta māal pōla). The way Māl breaks the tree is significant. He
walks on it (cela). To walk on a tree in order to take its leaves is a puzzling
process, in which I would assume that two different literary universes meet,
the northern Kṛṣṇa-oriented one in which the child-god passes between two
trees, uprooting them in his way, and the Love-oriented Caṅkam universe, in
which He meets Her while she is bathing, which is a motif mentioned as such
in Cil. 17.23–25: we are here on the sandy banks of a river. I suspect the
development of the shadow-motif of the kuruntu-vastraharaṇa to be linked
with other legendary elements amongst which I would underline the
importance of the fight against the snake Kāliya that can be spotted in North
Indian legends as well as in other types of sources (old sculptures, Jaina
versions of Kṛṣṇa’s childhood, the Bālacarita). Lack of space does not allow
this theme to be developed here.
49 The Pāṇṭiyas have been the focus of recent discoveries and research
work. See Y. Subbarayalu and V. Vedachalam (2007) and Orr (2007).
50 On several occasions, the mullai series of the Kalittokai evokes the

Pāṇṭiya king. In mysterious accounts given in Cil. 17, in the Kalittokai and in
the engraved meykkirttis of the Pāṇṭiya kings, the Pāṇṭiya wins over Indra,
snatching away his garland and breaking his crown. He stops the ocean with
his spear. No trace of such legends is found anywhere else. I wonder if such
myths could not echo the northern Indian Kṛṣṇa legend, like the allusions to
episodes in Kṛṣṇa’s life in the Tiv., which seem to refer to something
everybody knows. In the Govardhana episode, Kṛṣṇa stops the ekārṇava
ocean discharged by the angry king of the gods, who eventually has to anoint
him as the new king. In this case also, the original legend would have been
adapted more than merely translated. This differs from the way Hardy has
considered the role of what he called “the Tamil renaissance in Maturai”;
however, such a hypothesis also assigns a pivotal role to Maturai in the
elaboration of the southern legend of Kṛṣṇa.
51 For the role the Cēras have been traditionally invested with, see Nilakanta
Sastri (1955: 115–45), Parthasarathy (1993: 6–8); contra, Obeyesekere
(1984: 372–75).
List of illustrations
Caption Fig. 1 One of the two “pillars” from Maṇḍor, Rajasthan: the
fight against an ass-demon (top of the pillar; photo by
courtesy of the American Institute of Indian Studies).
URL http://books.openedition.org/ifp/docannexe/image
/2861/img-1.jpg
File image/jpeg, 121k
Caption Fig. 2 One of the two “pillars” from Maṇḍor, Rajasthan: the
fight against a horse-demon (photo, author).
/2861/img-2.jpg
Caption Fig. 3 Deogaḍh, 5th century (Madhya Pradesh; National

museum of New Delhi; photo, author).
/2861/img-3.jpg
Caption Fig. 4 Tiruveḷḷaṟai, 8th or 9th century (Tamil Nadu; photo,

author).
/2861/img-4.jpg
Caption Fig. 5 Brahmatēcam, 8th or 9th century (Tamil Nadu; photo,

author).
/2861/img-5.jpg
Caption Fig. 6 Tirucceṉṉampūṇṭi, 9–10th century A.D. (Tamil Nadu;

photo, author).
/2861/img-6.jpg
Caption Fig. 7 Somnāthpur, 12th century (Karnataka; photo,

author).
/2861/img-7.jpg
Author
Historian of religions. Having studied the beginnings of the cult of Kṛṣṇa in the
area of Mathurā, she became a member of the École française d’Extrême-
Orient in 1999. Posted in Pondicherry (Tamiḻ Nāṭu) from 1999 to 2003, she is
presently based in Paris. Her work on the development of Hinduism in South
India is founded on archaeological material and texts (including epigraphy), in
Sanskrit and Tamil, produced mostly at the time of two important dynasties of
the Tamil country, the Pallavas and the Cōḻas. Her recent publications
comprise Le don de voir, premières représentations krishnaïtes de la région
de Mathurā (Paris, 2010), “Du rite au mythe, les Tueuses de buffle de l’Inde
ancienne”, Artibus Asiae 2011 and “Rite and Representation: Recent
Discoveries of Pallava Goddesses of the Tamil Land”, Marg, December 2011.
By the same author
• La Bhakti d'une reine, Institut Français de Pondichéry, 2014
• 2. Viṣṇu aux rives de la Kāvēri : combatre et séduire in La Bhakti d'une reine,

Institut Français de Pondichéry, 2014
• 3. Brahmā : au nord du sanctuaire in La Bhakti d'une reine, Institut Français

de Pondichéry, 2014
• All texts

The Contribution of Tamil Literature To The Krishna of Sanskrit Cilappatikaram Charlotte Schmid

Uploaded by

Copyright:

Available Formats

The Contribution of Tamil Literature To The Krishna of Sanskrit Cilappatikaram Charlotte Schmid

Uploaded by

Document Information

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

The Contribution of Tamil Literature To The Krishna of Sanskrit Cilappatikaram Charlotte Schmid

Uploaded by

Copyright:

Available Formats

books.openedition.

The contribution of Tamil literature to the

1Though strongly marked by the Śrīvaiṣṇava movement, the history of

3Thus, I would like to offer a reassessment of the fundamental question

1. Kṛṣṇa elements in Cilappatikāram

5There is an important concentration of Kṛṣṇa-related elements in the story.

7Canto 17 is nonetheless a key moment from a narrative point of view. Canto

9Thus, churning constitutes a kind of thread stitching together different

• 9 We rarely come across deities in the Caṅkam corpus, except in the

2. The legend of Kṛṣṇa in Cilappatikāram 17

• 10 Before the dance involving Piṉṉai, Cil. 17 associates seven bulls of

12To demonstrate the actual existence of a typical southern folk Kṛṣṇaism,

13The analysis of Sanskrit data indeed tends to present these motifs as

14To recapitulate, the legend of Kṛṣṇa in southern India presents specific

15This synopsis might nevertheless ignore the complexity of the bond

17Thus, there is no apparent reason not to consider the whole of Cil. 17 in

18The apparent innovation seen in Cil. 17 leads me towards another

3. The kaṉṟu motif

• 14 On the vastraharaṇa motif, see Hardy (1983: 193–196; 497–501;

• 15 See the commentaries given in the edition of U. Vē. Cāminātaiyar

20The choice to concentrate on the kaṉṟu motif is based on the relative

• 17 The Gwalior museum exhibits a representation of a distinctive tāla tree,

4. The Cilappatikāram 17 stanza and the oldest Sanskrit texts

• 18 The text is given as it appears in U. Vē. Cāminātaiyar’s edition that I have

kaṉṟu kuṇilāk kaṉiy utirtta māyavaṉ

• 19 I refer here to the translations of the Cil. into English by Ramachandra

• 21 dāruṇo dhenuko nāma daityo gardabharūpavān / kharayūthena mahatā

• 22 ariṣṭo dhenukaś caiva cāṇūraś ca mahābalaḥ / aśvarājaś ca nihataḥ

• 23 dhenukaḥ sa mahākāyo dānavaḥ sumahābalaḥ / nihato vāsudevena

• 24 On the date of the Bālacarita, see above, note 5.

31Whatever the interpretation, the legend narrated in HV 57 appears to be

• 25 The demonic donkey is then placed as the guardian of the city of

• 26 For a critical presentation of these posts, see Williams (1982: 55–58).

33The first known representations could correspond to the oldest known

5. kaṉṟu, Dhenuka and the Tiv

37These elements are often regrouped in concise formulas which do not

• 28 Here is the whole stanza:

[...] kāṉaka vīl viḷaviṉ kāy utirak karutik

• 29 See infra, n. 28; in the old commentaries of the edition of Kiruṣṇasvāmi

44The only occurrence we come across in Āṇṭāḷ’s poems is particularly close

45Conversely, the eight allusions to the kaṉṟu legend in Tirumaṅkaiyāḻvār’s

• 34 These lines have been translated in Dr. S. Jagathratchagan’s edition as

• 35 Such a destructive power may be already mentioned in one of the hymns

• 36 See Periyālvār Tirumoḻi patuvurai, p. 48.

47Considering such details, I would suggest that the Tiv. indicates a

• 38 The line has been translated in Dr. S. Jagathratchagan’s edition as

• 39 Two of Pūtattāḻvār’s stanzas present formulas very close to those of

• 40 In the Periyāḻvār corpus alone, for example, the demoness Pūtanā is

• 41 It is not possible to broach such a complicated issue as the internal and

6. Sanskrit texts of South India

51kadācid yamunā-tīre vatsāṃś cārayatoḥ svakaiḥ / vayasyaiḥ kṛṣṇabalayor

54This account is short. It is preceded by a scene in which Kṛṣṇa and

• 43 See Couture (1991: 363–374) for a presentation and a French translation

56The influence of the South Indian tradition is obvious in such an account,

61The detail of the bird-like appearance of this demoness disappears from

62In South Indian sculptures, Pūtanā is never given the ornithological

• 46 The uprooting of the Arjuna tree is recounted in HV 51 and VP 5.6. For

64In the oldest known representations of an uprooting of trees by Kṛṣṇa in

Fig. 5 Brahmatēcam, 8th or 9th century (Tamil Nadu; photo, author).