Religions of South Asia 14.1–2 (2020) 87–116
https://doi.org/10.1558/rosa.19324
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
ISSN (print) 1751-2689
ISSN (online) 1751-2697
The Roots of the Two Sides of Kāmākhyā:
The Blending of Sex and Death in Tantra
PAOLO E. ROSATI1
Independent scholar
Rome
Italy
paoloe.rosati@gmail.com
ABSTRACT: The shrine of Kāmākhyā (Assam) was supposed to be the eminent
yoginī-pīṭha. Inside the sanctum of Kāmākhyā a yoni (vulva) stone is concealed as
the main cultic image of the Goddess, which Kaulism identified as the ‘mouth of
the yoginīs’. This article analyses the symbolism related to the Tantric cult of the
yoni and its historical evolution through the combined lens of History of Religions
and Cognitive Science of Religions. Kāmākhyā thus emerges as a hyper-blended
space, whose origin should be tracked down to the intersection of death symbolism
related to the non-Brahmanic cult of Heruka and his retinue of yoginīs and the
Kaula erotic reformation of the cult of the yoginīs. Therefore, the yoginīs played
a fundamental role in the construction of Kāmākhyā—as either a caring mother
or as a dreadful mother—conveying other blended spaces to the yoni metonymic
symbol. Hence, this article aims, through the analysis and interrelation of textual,
epigraphic and material evidence, to debate the dialectic between eros and thanatos
in order to shed light on the overlap, superimposition and blend of trans- and
cross-cultural elements in the multifarious goddess Kāmākhyā.
KEYWORDS: conceptual blend, CSR, Heruka/Hevajra, hyper-blend, Kaula, yoginī,
yoni.
We can investigate and analyze some of the most interesting domains of human
experience using insights and methods from cognitive science.
(Hayes and Timalsina 2016: 4)
1.
Paolo E. Rosati obtained a PhD in Asian and African Studies (South Asia section) from
Sapienza University of Rome, defending a thesis on the yoni cult of Kāmākhyā. Paolo’s
main field of research is the medieval and pre-colonial development of Assamese Tantra,
applying modern social theories. He is co-editing with Andrea Acri a volume provisionally
entitled Tantra, Folk Religion, and Magic: Experiences, Practices, and Practitioners at the ‘Margins’
of Monsoon Asia.
© Equinox Publishing Ltd 2021, Office 415, The Workstation, 15 Paternoster Row, Sheffield S1 2BX
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
88
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
INTRODUCTION
Today, Kāmākhyā is worshipped as the Mother Goddess of Assam by Hindus
and tribespeople.2 During the Ambuvācī melā, when her menstrual cycle is
celebrated for three days every year in the monsoon season (June or July), a
crowd of śākta, Tantric, non-Tantric and tribal devotees reach her shrine on
top of Nīlācala, in the Brahmaputra valley, to pray to Kāmākhyā and obtain
her favours (Fig. 1).
Figure 1: Possibly, the menstruating Earth goddess, Kāmākhyā temple (outer northern wall).
Photograph by Paolo E. Rosati, 2016.
2.
During the Middle Ages, Assam was part of the kingdom of Kāmarūpa (having the form of
desire), which geographically covered most of what is today known as northeastern India,
and part of Bhutan and of Bangladesh.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
89
Her earthly abode is Nīlācala (Nīlakūṭa), ‘the blue mountain’ (KP 62.1), also
renowned as Kāmagiri, ‘the mountain of love-making’; there, Śiva and Satī
secretly make love (KP 62.73b–85a), after the Goddess was dismembered by
the gods, and her yoni (vulva) had fallen onto its peak, turning the mountain
blue (KP 62.57).
The sacred zone of the Goddess’s womb at Kāmākhyā is the site of the auspicious
Tripurabhairavī [KP 74.93b–98a]. It is the best of hallowed places in this earthy
realm, and here Mahāmāyā dwells. There is no better place on earth than here,
where every month the Goddess herself resides during her menses. All the deities
of that place have assumed the form of mountains. Even the great deities dwell
within those mountains. The whole earth there is the very essence of the Goddess,
so think the wise. There is no better dwelling place than this sacred zone of the
womb at Kāmākhyā.
(Devībhāgavatapurāṇa 7.38.15–18, trans. Brown 2002: 105, diacritics added)
It can be observed that a deep symbolism is enclosed in the textual sources
devoted to the worship of the goddess Kāmākhyā, whose cult is closely
related both to Śiva and to Kāma. Śiva not only is the inseparable lover of
Kāmākhyā, but was the one who burned Kāma, from whose ashes Nīlācala
was made (KP 62.89).3
The yoni-stone is concealed inside the garbhagṛha (‘womb-house’, sanctum) of the shrine of Kāmākhyā, and is worshipped as the source of the
whole universe (Yoginītantra 1.15.52). This chamber is an obscure and claustrophobic place, located a few meters below ground level. The grotto was
variously described in medieval Sanskrit literature,4 such as the ‘manobhavaguhā’ (cave of Kāma) (KP 63.8–9a), the ‘kāmākhye girikandare’ (the cave of the
mountain of Kāmākhyā) (Devīpurāṇa 39.5cd), and the ‘yonigarta’ (recess of the
yoni) (Yonitantra 4.2). The Umāchal Rock inscription (470–94 ce)5 confirms the
presence of a sacred ‘guhā’ on Nīlācala (l. 4), which was constructed by king
Surendravarman (who was identified with Mahendravarman [450–85 ce])
(ll. 1–2) to worship ‘valabhadrasvāmināya’ (Lord Valabhadra) (l. 3), probably a
3.
4.
5.
The madana-bhasma (incineration of Kāma) myth is a Purāṇic adaptation of some Vedic
mythologems. More specifically, the Kālikāpurāṇa underlines sexual and death symbols in
connection with Nīlācala and with the Goddess’s yoni. Kāma, however, played a different
role from Vedas to Purāṇas (Doniger 1973: 141–73). In the Vedic time, he provoked the
desire of Brahmā/Prajāpati and, consequently, his incest with his daughter, an action that
caused his curse and incineration by Śiva (Rosati 2016).
In this study the period between the fourth and the twelfth century is described as the early
medieval period, while from the thirteenth to the sixteenth century is described as the late
medieval period. This division was pointed out by the historian Jae-Eun Shin (2010).
‘mahārājādhirāja śṛī | surendravarmanā kṛtam | bhagavataḥ valabhadra | svāmināya idaṃ guhaṃ’
(Sharma 1978: 2).
The name Valabhadra (i.e. Balabhadra) is an evidence of the link between the early
Varman kings and the Vaiṣṇava cult in the early phase of Varman kingdom.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
90
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
form of Viṣṇu or a hero associated with the cave during the Varman period
(Sharma 1978: 2–3).
To access this womb-like chamber, the devotees have to climb down more
than a hundred steps as a metaphoric descent into the underworld (Rosati
2017a: 9–10). There, from a cleft in the yoni-stone, a water stream oozes out,
completely submerging the sacred rock, as a metaphoric replica of its yonitattva.6 This symbolic representation of a yoni, covered and surrounded by
her sexual fluids, is reinforced during the Ambuvācī melā. In that period,
the water acquires a reddish colour, thus representing the menstrual blood
of Kāmākhyā. This phenomenon is explained through the interaction of the
rainfall during the monsoon season and the hematite in the rocks; nevertheless, the worshippers believe it is the menstrual blood of the Goddess, who
gives back her śakti (energy) to earth once every year.
According to the medieval texts, the yoni is inhabited by Kāmākhyā
(KP 62.75; Kāmākhyātantra 1.4), but she also pervades the water. She exists,
indeed, in the form of red water (‘raktapānīyarūpiṇī’ [Yoginītantra 1.11.37a]).7
The goddess Kāmākhyā is thus not only indissolubly linked to the fluidity of
water and blood, but combines in her identity fluid and dry elements:
[w]e may say that fluid essence is one of the most archaic and universal features
of the sacred, closely linked to Mother Earth, whose very nature is arid and dry
because through her monthly menstruation she loses her generative power and so
needs to reintegrate it constantly by absorbing liquids … The earth is the Mother,
the source that produces and sustains life in all its forms, but her initial sacrifice to
generate life needs in turn to be nourished in order to grant vitality to all.
(Lussana 2015: 76)
Kāmākhyā is fluid like water/blood, but at the same time she is arid like
stone/earth. She is thus transcendent and immanent at the same time. She
maintains her fluidity, after losing it through her menstrual blood, by being
nourished by her devotees through the blood of male sacrificial victims.
Today, blood offerings are replaced by offerings of fruit and vegetables in a
right-hand Tantra exotericization of the ritual praxis (Urban 2010: 167–72)
inside the garbhagṛha.8 Among the offerings, coconuts and their milk are very
common, as metaphoric substitutes for the ritual victims’ heads and their
6.
7.
8.
The yonitattva is both the sexual fluids produced in the vulva through sexual stimulation
and the menstrual blood (Aktor 2016: 97).
In the Sanskrit tradition, the association between motherhood and water (ap, a feminine
noun) is corroborated since the Vedic period. Indeed, the rivers—who are usually considered
female—are the ones that fertilize the earth as well as transform everything (Lussana 2015:
73–74). Nevertheless, the greatest river of Assam, the Brahmaputra (son of Brahmā) is one
of the few examples of a male river.
Today, blood sacrifices are confined outside the main temple and may be considered a
left-hand Tantra exotericization of the ancient ritual tradition, when animals and human
beings were slaughtered in cremation grounds as well as in secret places.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
91
bone marrow—which the head-hunting tribes of northeastern India closely
related to ‘heroism’, ‘manhood’ and power (Zou 2005: 89–90). Therefore, the
non-bloody offerings performed inside the garbhagṛha maintain a memory of
the past, a feature that has been observed in many other traditions:
Traces of the past, however, survive in diverse linguistic and ideological forms, and
local people themselves add a new dimension to the custom when they creatively
appropriate and translate their historical heritage of headhunting into an evocative
metaphor of modern practices.
(Zou 2005: 76–77)
Kāmākhyā lives on Nīlācala with her retinue of sixty-four yoginīs, a band of
dreadful goddesses, who appeared in the Hevajratantra (1.7.10–19), a Buddhist
text belonging to the middle of the ninth century (Szántó 2012: 14), as dreadful deities connected to blood sacrifices and graveyards.9 They are well
described by David White (2006: 3–4) as hungry deities, angry and thirsty
for human flesh and blood. Their origin is traced back to the Purāṇic mātṛs
(mothers) who appeared in the sixth to eighth centuries ce Devīmāhātmya
(‘mātṛgaṇa’ [8.39], ‘mātṛ’ [8.49], ‘mātṛgaṇa’ [8.62]), or to the sixth to seventh
centuries Bṛhatsaṃhitā (‘viśvasya mātarah’ [47.68], ‘mātṛgaṇa’ [57.56],
‘mātṝṇām’ [59.19]).10 The Kālikāpurāṇa (54.35–45; 63.37–43) preserves two lists
of sixty-four yoginīs, who are described as ‘attendants’ of Kāmākhyā, who
reside in her ‘maṇḍala’ (KP 54.34; 63.44), which is identified with the yoni. The
superimposition and blending between Kāmākhyā and the concept of yoginī
is reinforced through the inclusion of the name of Dikkaravāsinī among the
yoginīs, who is identified as the dreadful form of Kāmākhyā (KP 54.36a).11
The Kālikāpurāṇa merged the cult of these dangerous and powerful deities
with the yoni cult. The text explains that eight of the sixty-four yoginīs—
Guptakāmā, Śrīkāmā, Vindhyavāsinī, Koṭīśvarī, Vanasthā, Pādadurgā,
Dīrgheśvarī and Bhuvaneśvarī—surround the goddess Kāmākhyā. They live
in pīṭhas (seats) that are ‘all in the form of water’ (jalarūpāṇi KP 62.91–95,
trans. B. Shastri 459–60). When the yoni of Satī fell on earth it broke into
pieces, each of which originated a different yoginī (KP 62.74–75; particularly
see Shin 2010: 16). This group of dreadful deities is thus linked either to the
(aquatic) water’s fluidity or to (earthy) stone’s aridity—in fact their origin
is connected to the yoni which fell on Nīlācala; furthermore, according to
9. Among these places—which were presided by the yoginīs—were pīṭhas, upapīṭhas and
sandohas. This terminology is also preserved in the KjN (8.4–6, 16–17, 24–25). This fact could
be considered to better understand the connection between Buddhist yoginīs and their
‘seats’ and the probably later Hindu-Tantric tradition.
10. Regarding the date of the Devīmāhātmya, see Yokochi (1999: 71), and particularly 91n1.
11. The goddess Dikkaravāsinī has two different identifications: one of her forms is tīkṣṇakāntā
(‘fond of cruelty’; (Monier-Williams 1960: 448) (KP 80.36b–39), the other is lalitakāntā
(‘playful beloved’) (80.52–54a).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
92
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
popular belief the yoni-stone is a meteoritic one, although this cannot be
confirmed by any scientific analysis.
Kāmākhyā is therefore an ambivalent goddess; she is described in texts
as either a lovely and caring mother or a dreadful and dangerous mother.
What elements stand at the origin of the cult of Kāmākhyā? Why and how
were they integrated within her religious universe? Combining textual and
epigraphic evidence with Cognitive Science of Religions (CSR), this paper
aims to shed light on the heterogeneous and contrasting factors which were
blended in the construction of the Tantric Mother Goddess of Assam and her
metonymic symbol par excellence—the yoni.
A LONG-STANDING QUESTION: ‘WHO IS KĀMĀKHYĀ?’
The first comprehensive study on the goddess Kāmākhyā focused on the
‘fusion of Aryan and Primitive Beliefs of Assam’ (Kakati 1989). Since then,
both the concept of Indo-Aryan as an ethnic homogeneous group (Bryant
2001) and that of a dichotomic division and interaction between the
Brahmanic people and the local cultures have become outdated (Rosati 2016,
2017a, 2017b).12 Furthermore, in the last decades, local, tribal and oral-based
cultures of South Asia are no longer described as primitive, illiterate, minor,
low, or little societies (as they were by Redfield 1955, 1956: 60; Singer 1955).
Also the sociological model termed ‘Sanskritization’ (Srinivas 1952), which
describes the attempt of the local tradition ‘to acquire the traditional symbols
of high status’ (p. 30), has been dismissed as ‘a concept too vague and ambiguous for analytic or descriptive utility’, that ‘leads to inaccurate and misleading assumptions concerning the social history of modern India’ (Carroll
1977: 366). Ronald Davidson (2002) supposed that there was an ‘aggressive
Hinduisation’13 of the tribespeople during ‘the early medieval time’ (p. 225),
a process, however, that has no evidence in early medieval Kāmarūpa (Rosati
2017a: 4), where universalization processes have collaborated and clashed
with the process of parochialization (see Marriott 1955: 181–200).
Therefore, describing the cult of Kāmākhyā as the result of an interaction
between Brahmanic and local or tribal traditions is an oversimplification.
12. Indeed, Indo-Aryan is not an ethnic but a linguistic group. At the same time, medieval
northeastern India was a cradle of various linguistic groups belonging to the Indo-Aryan,
Mon-Khmer, Tibeto-Burmese and Austroasiatic linguistic families. Each of these groups
had its own sub-groups and its own socio-cultural and religious traditions. However,
intermingling and exogamic links between these groups cannot be denied, and ethnic
purity can be labelled as an absurdity.
13. The term ‘Hinduization’ emphasises that kṣatriyas (i.e. warrior and/or royal class) as well
as brāhmaṇas (sacerdotal class), influenced the Sanskritization process (Eschmann 1986:
79–80).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
93
However, most scholars who are involved in the study of Kāmākhyā have
identified her as a multicultural deity, who was constructed through the
interrelation and dialectic of pan-Indian, Vedic-Brahmanic, Hindu-Purāṇic
and non-Brahmanic elements (Borkataky-Varma 2019: 181–82; Dold 2004:
90; Shin 2010; Urban 2001, 2010; Rosati 2016, 2017a, 2017b, 2018a, 2018b).14
Nevertheless, all the evidence inferred from texts, inscriptions and sculptural remains pinpoints Kāmākhyā as a Tantric goddess, who is part of the
heterodox folds of Brahmanism.
Most of the information regarding the goddess Kāmākhyā and her yoni cult
is preserved (1) in a group of śākta Purāṇas and Tantras compiled in northeastern India between the ninth and sixteenth centuries ce;15 (2) in a series
of inscriptions belonging to the medieval period;16 and (3) in some sculptural
remains which are identified as representations of the yoginīs.
NARAKA AND THE CULT OF KĀMĀKHYĀ
The mytho-history of Kāmākhyā began when Naraka defeated the Kirātas
and founded the kingdom of Kāmarūpa (KP 38.104–124),17 which lay in the
Brahmaputra plain up to the forests of Nagaland (Shin 2018).18 After that, he
14. In contrast to this trans- and cross-cultural analysis of the Tantric cult of Kāmākhyā,
Gavin Flood (2006: 14) and Christian Wedemeyer (2012: 20–30) consider Tantra, in all its
manifestations, as a primary Brahmanic religious development, finding themselves unable
to trace any tribal or non-Brahmanic element in Tantra.
15. The sources can be divided into early medieval texts (Hevajratantra, Kulacūḍāmaṇitantra, KjN,
KP) and late medieval ones (Devībhāgavatapurāṇa, Yoginītantra, Yonitantra).
16. See the edition of Assamese ancient and medieval inscriptions edited, translated and
annotated by Mukunda M. Sharma (1978).
17. According to the Epic and Purāṇic tradition, Naraka was the founder of the Bhauma (‘son of
the Earth’) dynasty, a mytho-historical line of kings of Kāmarūpa. He was succeeded by his
son Bhagadatta, followed by his grandson (or younger son) Vajradatta. Only in the middle of
the fourth century ce do we have evidence of a historical king, Puṣyavarman, who ascended
the throne of Kāmarūpa after Naraka’s family ruled for three thousand years (Shin 2018: 27).
18. J. E. Shin (2018: 26–30; 41–46) enquired whether Kāmarūpa was part of ‘Āryāvarta’ or not.
Indeed, Kāmarūpa was categorized in the Allahabad pillar of Samudragupta (middle of the
fourth century ce) as ‘the land of frontier kings’, who had negotiations with the Gupta
empire. According to Shin, it cannot be certain that Kāmarūpa belonged to Āryāvarta,
considering that many socio-cultural, political and religious traditions cooperated in the
formation of this regional entity. The association of Kāmarūpa with Āryāvarta is nationalist
contestation of the colonial description of Assam as a peripheral land related with religious
esoterism. However, the discourse is far more complex; as Shin (pp. 40–41) pointed out,
following the KP and the Yoginītantra accounts, within Kāmarūpa there was an area more
influenced by Brahmanism, and an eastern offshoot far more subjected to indigenous
traditions.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
94
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
adopted the Kirāta cult of Kāmākhyā into the Brahmanic fold, placing the
local cult of Śambhu (Śiva) in the background (KP 38.158).19
[Viṣṇu:] O my son [Naraka]! You shall not worship any other goddess except
the great goddess Kāmākhyā, the mother of the world, who is none else than
Mahāmāyā, Ambikā. Acting otherwise you shall die, therefore, O Naraka! Adhere to
the promise with great care.
(KP 38.149–50, trans. B. Shastri 2008: 247)
The link between Kāmākhyā and Naraka underlines a first aspect of the
Goddess, as far more related to the Earth goddess and the Vaiṣṇava tradition
than to the Śaiva.20 Kāmākhyā, indeed, is Yoganidrā,21 who is no other than
Mahāmāyā, ‘the Goddess of Great Illusion’ (Biernacki 2007: 3).
Yoganidrā, the great goddess, the source of the world, who has merged at the feet
of Satī at Devīkūṭa, is known as Mahābhāgā. The goddess Yoganidrā is known as
Kātyāyanī at Uddīyāna, as Kāmākhyā, the protean (who assumes shape at will) in
Kāmarūpa, as Pūrṇeśvarī in Pūrṇagiri, as Caṇḍī on the mountain of Jālandhara and
as Dikkarāvasinī at the east end of Kāmarūpa, who is also called Lalita-Kāntā.
(KP 18.48–50, trans. B. Shastri 2008: 98–99)
Therefore, Kāmākhyā is ‘kāmarūpiṇī’,22 ‘the one who can assume any shape at
will’ (39.17–20, trans. B. Shastri 2008: 250). This shape-shifting power is also
associated with Bhūmi (Earth goddess), who is described as māyāmanuṣarūpiṇī,
a goddess who can assume ‘the human form with her illusory power’ (KP 38.38,
trans. B. Shastri 2008: 239; see also KP 18.49). This power of shape-shifting,
therefore, emerges as a fundamental one, and is related to the Goddesses
presiding over the śakti pīṭhas (seats of power).23
Naraka was the son of Bhūmi and Varāha (boar),24 who had had intercourse when the Earth was menstruating. The story is summarised as a
rape in the Kālikāpurāṇa, narrating that Varāha ‘raped the licentious Pṛthvī
19. However, the cult of Śiva was prominent among the Kirātas (KP 38.101), leading us to suppose
an extremely variegated non-Brahmanic tradition in the South Asian autochthonous
religious substratum.
20. It is corroborated by the fact that in the Devīpurāṇa (39.5cd), the son of the earth (bhauma)
is said to worship a goddess (devīṃ) on the mountain of Kāmākhyā (kāmākhye girikandare).
21. The sleep of Viṣṇu, when he was lying on the coils of the cosmic snake Ananta (endless), is
called yoganidrā (waking-sleep).
22. A power that is associated also with Satī, the former wife of Śiva and with the yoginīs (KP
9.47).
23. Previously, the power of shape-shifting was associated to the mātṛgaṇa (band of mothers)
(Mahābhārata 9.45.29–40a), a type of deities associated with caves, mountains, trees and
graveyards. According to Jae-Eun Shin (2010–11: 49), the mātṛgaṇas mirrored ‘a myriad of
autochthonous and non-Vedic goddesses’ who were categorized in the early Sanskrit Epic
without any agreement as to their number.
24. Varāha, an avatāra (descent) that Viṣṇu assumed to save the Earth goddess from the abyss of
the cosmic ocean (Viṣṇupurāṇa 1.4, see pp. 57–61). On the other hand, the Devībhāgavatapurāṇa
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
95
[i.e. Bhūmi] in the water in the past, and as she was in her menstruation
period she conceived a terrible embryo’ (KP 29.13, trans. B. Shastri 2008: 187).
Therefore, this dangerous physical condition of the Earth goddess contaminated Varāha, ‘[t] he body of mine had been contaminated for having sexual
intercourse with a woman in her period of menstruation’ (KP 30.34, trans. B.
Shastri 2008: 193) as well as their son Naraka, who was thus perceived as an
asura (anti-god).
The ambivalent and dangerous identity of Naraka is also summarized in
the Dubi Copper Plates of Bhāskaravarman (v. 2):
One who, in the days of yore, in the form of a boar out of compassion recovered
the lost Earth for the stability of the people and put the same in his mouth, begot
a superior son, Naraka by name, who was powerful enough even to torment the
ambrosia-drinking gods and who was all powerful on earth being the king of kings.
(Trans. Sharma 1978: 20)
Menstrual blood is, indeed, either a polluting substance or an auspicious
one. It is powerful and sacred, encompassing both auspicious and inauspicious aspects but exclusively neither one nor the other (Appfel-Marglin 1985:
40–44). Hence, menstrual blood well fits the kingship’s need for a type of
power that could not be manipulated by anyone without risk—an element that
influenced the need of a Tantric specialist or the need of the king to become
well-versed in the Tantric rituals. Naraka, being the first mytho-historical
king of Assam, well-synthesized the image of the Tantric king. Naraka (hell)
was found by his stepfather, Janaka, lying and crying inside a human skull
(KP 37.48–51)—a replica of the skull-cup used by Tantric adepts for drinking
blood or other prohibited liquids, such as alcohol. This symbolism also traces
a connection between the Assamese medieval kingship and the world of dead
spirits.
The syncretic amalgam of Vaiṣṇava ideology with local and tribal female
deities has been observed in many regional contexts of South Asia (Tiwari
2002: 128); in fact, a stratification of local, non-Brahmanic, Vaiṣṇava, Śaiva
and śākta strands is a shared element among the Hindu deities (Erndl 1993:
43–44). However, for Śaivism it was far easier to blend with tribal and local
traditions because of ‘its popular character’ (Bhattacharyya 1974: 73).
The Cult of the Yoginīs and Its Intersection with the Goddess Kāmākhyā
If we are to look for the origins of the yoginīs, it appears that we must turn to the
simple village cults and to the grāma devatās ... These village goddesses seem to have
been gradually transformed and consolidated into the … sixty-four [yoginīs] … It
was tantrism that elevated these local deities … The philosophy, rituals and cults of
(9.9.42–45) narrates that from the union of Varāha and the Earth goddess was born the
planet Mars.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
96
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
these deities along with others that were originally non-Brahmanical, were brought
together under the rubric Tantra and thus given legitimacy in later Hinduism.
(Dehejia 1986: 1–2, italics added)
The other and more dangerous aspect of the goddess Kāmākhyā derived from
her association with the dreadful band of yoginīs. Their roots are often traced
back to outside the mainstream Indian religions, more specifically to both
the mātṛgaṇas who were crystallised in the group of the saptamātṛs (seven
mothers) around the middle or the end of the sixth century ce (Shin 2010–11:
59–62), and the grāmadevatās (village goddesses) (Dehejia 1986: 68).
Shaman Hatley (2014) questions the existence of a relevant process of
absorption of local goddesses within Kaulism. Nevertheless, he considers
possible a link between local cults and yoginīs in an earlier stage:
It does appear possible that Yoginī temples incorporated local deities into their
iconic programs,25 but in no case has this been demonstrated to be a significant
process. Some Yoginīs who have been singled out as ‘local’ deities on the basis
of non-Sanskrit names … are in fact also attested in the pan-regional tantric
literature in Sanskrit. Tribal and local traditions might have been significant to the
Yoginī cult, but if these were formative influences, they should be looked for in
the early Śiva and Buddhist esoteric traditions, given the apparent chronology of
evidence, for tantric worship of Yoginīs appears to predate the temples by at least
two centuries … Without more evidence, it is therefore difficult to concur in the
view that the Yoginīs enshrined in temples ‘represent’, as Donaldson [2002: 2.658;
cf. Dehejia 1986: 93–94] suggests, ‘localized cult traditions of village deities that
eventually were transformed into potent groups of sixty-four yoginīs’.
(Hatley 2014: 209–10)
However, it seems that Hatley does not totally deny an origin of the yoginīs
outside the mainstream religions of South Asia, but it has to be looked for
into the intersection of local and tribal cultures with Tantric Buddhism and
the Vidyāpīṭha tradition. The association of the yoginīs with Bhairava (Śiva)
as well as with the sexual practices of Kaulism is a cultic development of
the Vidyāpīṭha, in which ‘the Kāpālika culture of the cremation grounds’ was
‘pervasive’ (Sanderson 1988: 670).
The Yoginī cult, like the main cults of entry into the Vidyāpīṭha, was the speciality
of skull-bearing ascetics removed from conventional society. It might reasonably
have been expected to remain so but for Kaulism. This movement within esoteric
Śaivism decontaminated the mysticism of the Kāpālikas so that it flowed into the
wider community of married householders … Kaulism developed from within these
Yoginī cults … In the cults of Vidyāpīṭha the propitiation of the deities involved
sexual intercourse with dūtī [consort]. This practice continued in Kaulism. Indeed
it moves to the very centre of the cult. However while its principal purpose in the
Vidyāpīṭha was to produce the power-substances needed to gratify the deities, here
25. It is necessary to point out here that the temples of the yoginīs in South Asia are dated no
earlier than the ninth or tenth century.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
97
the ritual of copulation is aestheticised … The Kāpālika of the Vidyāpīṭha sought
the convergence of the Yoginīs and his fusion with them (yoginīmelaka, -melāpa)
through a process of visionary invocation in which he would attract them out
of the sky, gratify them with an offering of blood drawn from his own body, and
ascend with them into the sky as the leader of their band. The Kaulas translated
this visionary fantasy into the aesthetic terms of mystical experience. The Yoginīs
became the deities of his senses (karaṇeśvarīs), revelling in his sensations.
(Sanderson 1988: 679–80)
Hence, in the tenth–eleventh century, with the Kaula reformation, sexual
intercourse in relation to the worship of the yoginīs assumed a central
position in the ritual praxis, while the cremation ground was a subsidiary
element.
What we do know regarding the cult of the yoginīs is that when it was
integrated within the Kaula tradition, through its systematization in the
Kaulajñānanirṇaya (ninth to eleventh century), the yoni pūjā (worship of
the vulva) was upraised to ‘the “primal sacrifice” (ādiyāga)—one of the six
types of “clan sacrifice” (kulayāga)—and Tantric initiation’ in Kaulism (White
2006: 106). In fact, the transmission of the gnosis could take place through
the vaktrādvaktram, ‘mouth to mouth’ transmission (KjN 18.22b).26 Thus, the
yoginī transmitted the gnosis through their sexual fluids from their nether
mouth to the practitioner’s mouth in a ritual called rajapāna (White 2006:
106).27 This mystic-erotic practice was founded on the consumption of female
sexual fluids from ‘the mouth of the yoginī’ (Tantrāloka 29.122a–26a, 127b–
28b; Kāmākhyātantra 4.22), the vulva, source of both gnosis and inestimable
powers (siddhis).
According to a northeastern Sanskrit tradition, the master who
founded the Kaula sect, around the beginning of the tenth century, was
Matsyendranātha,28 who learnt the doctrine from Bhairavī in a circle of
yoginīs in Kāmarūpa (KjN 16.21–22). Hugh Urban (2010: 41) identified some
sculptural representations of female goddesses, belonging to the early medieval period, scattered around Nīlācala, with a retinue of yoginīs (ibid.: 205 n.
43). The fact that a considerable number of sculptures of yoginīs (Fig. 2) was
found on Nīlācala reinforces the connection of their cult with the cult of the
yoni of Kāmākhyā.
26. Furthermore, the ‘Kaulika Knowledge’ is said to reside ‘in the navel, the heart, the throat,
the mouth and the nostrils’ (KjN 6.7–8).
27. A ritual that is represented across South Asia in a number of sculptural representations (e.g.
White 2006: 107 fig. 4c, 110 fig. 4d, 111 fig. 4e).
28. Matsyendranātha is identified with Bhairava (KjN 16.11).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
98
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
Figure 2: Cāmundā on a pretāsana, Śrī Kedareśvara Śivālaya temple (Nīlācala). Photograph by
Paolo E. Rosati, 2016.
METHODOLOGY: TANTRA AND CSR
A person may be having a vision of a divine being, but that experience is embedded
not only in the sociocultural and religious contexts of the individuals, but also in
the neural and cognitive systems of the human body.
(Hayes and Timalsina 2016: 3)
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
99
The Mother Goddess of Assam emerges as a very syncretic and multicultural
deity, who blended and stratified a substantial number of different traditions, thus absorbing, after a centuries-long cross- and trans-cultural dialectic, many heterogeneous elements. Among those elements, menstrual blood
and sexual fluids, blood sacrifices, and death imaginary will be pointed out,
to shed light on the ambivalence and nuance in the cult of Kāmākhyā.
In this article, CSR will be used in order to explain the hidden process
that stands at the origin of the mental perception of the divine at Kāmākhyā.
The cognitive theory of conceptual blend, which was introduced in the
early 1990s by Gilles Fauconnier and Mark Turner, will be integrated within
the field of religious studies, in order to explain the figure of Kāmākhyā as
a hyper-blend (or megablend).29 The conceptual blend theory is a further
development of the conceptual metaphor theory of George Lakoff and Mark
Johnson (1980), and both theories can combine to better explain religious
processes not being antithetic.30
The blend (Fig. 3) is a mental tool for compression (of time, space, identity,
etc.) leading to new meanings. In blending, partial structures of input (mental)
spaces are projected and compose the blended space. The input spaces share a
generic space, which connects them. The blended space is a selective projection
Figure 3: The conceptual blend and the four space-model. Scheme adaptation from Fauconnier
and Turner (2003: figs. 3.2–6, see pp. 41–43, 45–46).
29. A hyper-blend is an integration network of two blended spaces or of a blended space and a
simple network.
30. More information on their application in the South Asian religious studies, and more
specifically on Tantra, may be found in Hayes (2012), Hayes and Timalsina (2016) and
Timalsina (2016).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
100
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
of input structures, which can be brought into the blend as mixed in one
or as separate entities—this process is composition. Two other processes participate in mapping a blended space: completion, which integrates a familiar
background frame, and elaboration, which develops the blend through imaginative mental processes. ‘Crucially, the blended space remains connected to
the inputs by the mappings, so that real inferences can be computed in the
inputs from the imaginary situation in the blended space.’ In this way, blending firstly ‘provides relations that do not exist in separate inputs’. Secondly, it
unconsciously integrates ‘background models’. Thirdly, it is treated ‘as a simulation’ which is run ‘imaginatively’ (Fauconnier and Turner 2003: 58–60).
Hence, this article will try to explain the yoni, a metonymic symbol of
Kāmākhyā, as a hyper-blend, a mental space that is ‘at the heart of language,
art, religion and much daily life’ (Hayes 2012: 197). More specifically, this
hyper-blend seems to correspond with the ‘double-scope network’, which
‘produce[s] the most imaginative and creative blend’ (ibid.: 199). In this
typology of blend space, the two inputs have two distinct ‘organizing frames’
that ‘clash with each other, so that only parts of each frame are projected into the
blend, leading to an entirely new “blended space” which has its own “emergent structure” and frame’ (ibid.).
THE MOUNTAIN OF SEX AND CORPSES
The bipolar symbolism linked to Nīlācala as a Tantric mountain of death and
life has its roots in the intersection and mutual transformation of Brahmanic
with local and tribal traditions. Both northeastern Tantric and Purāṇic
sources identify this mountain as the most sacred among the pīṭhas (‘evaṃ
pupyatame pīṭhe kubjikāpīṭhasaṃjñake || nīlakūte mayā sārdhaṃ devī rahasi saṃsthitā’ [KP 62.73b–74a]), the greatest pīṭha (‘pīṭhānaṃ paraṃ pīṭhaṃ kāmarūpaṃ
mahākalam’ [Kulacūḍāmaṇitantra 5.36b]), the original pīṭha (‘prathaṃ pīṭhamutpatraṃ kāmākhyānam suvrate’ [KjN 8.20]).31 The yoni of Satī has conferred on the
site an exceptional power (‘yuvābhyāṃ tatra paśya tvaṃ jātaṃ me yonimaṇḍalam
| mama tejaḥ samudbhūtaṃ jānīhi yonimaṇḍalam’ [Yoginītantra 1.15.51]). Not
only is the yoni intrinsically associated with the mountain (‘kāmarūpe kāmagirau nya patadyonimaṇḍalam’ [KP 18.42a]; ‘tasmimstu kubjikāpīṭe satyāstadyoni31. According to the more ancient list of pīṭhas preserved in the Hevajratantra (1.7.12), the four
original pīṭhas are Jālandhara, Oḍḍiyāna, Paurṇagiri and Kāmarūpa—which are not yet
associated with the Goddess’s limbs. In the KP (18.41–43), instead, there are seven pīṭhas,
each related to a limb of Satī: Devīkūṭa (feet), Uddīyāna (thighs), Kāmagiri (i.e. Nīlācala)
(yoni), Kāmagiri’s foot (navel), Pūrṇagiri (arms and neck), Jālandhara (breasts) and a
non-specified pīṭha supposed to be in eastern Kāmarūpa (head, which ‘fell beyond the
region of Kāmarūpa’). Due to their connection to Satī’s body they can be called śakti pīṭhas
(see Sircar 1998).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
101
maṇḍalam | patitaṃ tatra sā devī mahāmāyā vyalīyata’ [62.56]), but also Śiva
was attracted to the mountain and to any other pīṭhas of the Goddess,
where Śiva assumed both ‘the shape of the liṅga (phallus)’ (‘yatra yatrāpatan
satyāstadāpādādayo dvijāḥ | tatra tatra mahādevaḥ svayaṃ liṇgasvarupadrik |
tasthau mohasamāyuktaḥ satīsnehavaśānugaḥ’ [KP 18.46], see also KP 18.53),
and ‘the form of a mountain’ (‘sīnāyāṃ yoganidrāyāṃ mayi parvatarūpiniḥ | sa
nīlavarṇaḥ śailo’bhūtpatite yonimaṇḍale [KP 62.57]), which ‘turned its colour
into blue’ (ibid.). According to the Sanskrit tradition, Śiva was indissolubly
associated with Kāmākhyā, either in his liṅga form or in his mountain form;
in fact, the sacred mountain crosses over any sex-gender differentiation,
being the symbol of both Śiva and Satī (e.g. ‘mahāmāyā jagaddhātrī śailarūpapradhāriṇī | āgṅeyyāṃ ca tathā viṣṇurekarūpeṇa saṃsthitaṃ’ [KP 62.42]).
Therefore, the traits blended in Nīlācala are related both to manhood and
to womanhood; Nīlācala is also well-known as Kāmagiri, the love-making
mountain where Śiva and Satī have been intimately united in an endless
intercourse—the primary symbol of Kaulism (Dyczkowski 1988: 60–61).
Nīlaśaila (the mount Nīla) is triangular, being low in the middle, is Sadāśiva himself.
In the midst of it there is a maṇḍala which is beautiful drawn and provided with
thirty śaktis; there is the guhā of Manobhava (Kāma).
(KP 62.87–88a, trans. B. Shastri 2008: 459)
On the one hand, the mountain is the fruit of a sexual fusion, incorporating
the śakti, as its pulsing heart, which is preserved inside the garbhagṛha in the
form of the yoni stone, inside the body of Śiva. However, another Purāṇic
story links the mountain of love-making to the death and resurrection of
the god Manobhava (‘born from mind’), also known as Kāma (‘desire’). In
fact, ‘in each narration Kāma ultimately dies in the wrathful fire of Śiva’s
third eye’ (Benton 2005: 39), but he is reborn as anaṅga (bodiless) after the
wedding between Śiva and his second wife Parvatī.32 On the other hand, the
Kālikāpurāṇa (18.3–6; 42.148–173a) fixed a connection between Kāma and
Kāmarūpa, the place where he was reborn. In fact, this śākta Purāṇa outlines
a close link between Kāma and the goddess Kāmākhyā in a passage belonging
to one of its relatively later and far more Tantric sections:
(Inside the cave) on that stone there is a very lovely yoni in the form of stone which
is twelve aṅgulas in width and twenty aṅgulas in length, gradually narrowing and
sloping, and lying along the Bhasma-śaila (the hill of ashes of Kāma). It is reddish
in colour like vermilion or saffron, who fulfils all desires. On that yoni the amorous
goddess Kāmākhyā, who is the supportress and the source of the world and the
Primordial Force, the eternal one, always amuses herself in five different forms.
(KP 62.88b–90, trans. B. Shastri 2008: 459)
32. The madanabhasma (incineration of Kāma) is preserved in Brahmandapurāṇa (4.11–12;
4.29.141–48; 4.30; see also Benton 2005: 59–64), Brahmapurāṇa (38.1–13), Brahmavaivartapurāṇa
(4.39.40–59), KP (44), Matsyapurāṇa (154.212–86), Śivapurāṇa (2.3.17–20), Skandapurāṇa
(1.1.21.48–99; 1.2.24.17–20), Vamanapurāṇa (6.1–107).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
102
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
Among the powers of Kāma are ‘manmatha’, the ability to agitate, and ‘madana’,
the power ‘to excite the mind of all the living beings’ (KP 2.4–6; Śivapurāṇa
2.2.3.6). While his weapons are five flower-arrows, which are called harṣaṇa
‘delighting’, rocana ‘appealing’, mohana ‘deluding’, śoṣaṇa ‘withering’ and
māraṇa ‘killing’ (Śivapurāṇa 2.2.3.11–12), the tools through which ‘Śiva gave
up self-meditation’ (‘parityajyātmacintanam’ [KP 10.55, trans. B. Shastri 2008:
54]), and was enchanted to love his first wife Satī, an act which Śiva punished
by incinerating Kāma with his fire (KP 42.42).33
Nīlācala, therefore, was not only equated to the intersection of two triangles, symbols of male and female in their limitless sexual union as Śiva-Śakti
(Rosati 2016: 291), but also to the death of Kāma, who was restored in his corporal form by Kāmākhyā. The Goddess is described as the one who ‘gives love,
is a loving female, is embodiment of love, the beloved, she restores the limbs
of Kāma and also destroys the limbs of Kāma’ (KP 62.2, trans. B. Shastri 2008:
453). She absorbed the destructive powers of Śiva, being the one who dismembered Kāma, who, in his turn, was equated to Satī, the Goddess who was
dismembered by the gods. Fusion, intersection and superimposition emerge
as a common trait of the Purāṇic tradition, which often completely subverted
an earlier tradition or, at least, reinterpreted and completely transformed it
(Coburn 1991: 13–17; Brown 1990: 4–5).
Hence, a close association emerges of Nīlācala-Kāmākhyā both with the
ashes of Kāma (KP 62.89a) and with the śakti of Kāma (KP 63.8–9a), being her
sacred abode, the manobhava-guhā (‘cave of Kāma’ [KP 62.88a]) born from the
primordial desire (ṚV 10.129.3–4). This wide association between the two
deities is explicit in the etymology of the name Kāmākhyā, ‘she whose name
is love or desire’, which perhaps was an adaptation of Kāmākṣī, ‘she whose
eyes are full of love’; both names may be speculated to be Sankritizations of
unknown tribal Goddess(es). On the other hand, her etymological association
with an Austroasiatic word related to death imaginary through the association of Kāmākhyā with the proto-Mon-Khmer *kmuuc (dead person) (Khmer,
Katuic, South Bahnaric, Palaungic, Aslian) (Bhattacharyya 1995: 88–89),34 was
dismissed by Paul Sidwell,35 who pointed out that *kmuuc is related to *bruuʔ
(hill).
Although the association of Kāmākhyā with death imaginary seems an
incongruity, and there is no linguistic evidence to link her name to that
symbolism, her mythology suggests a double identification of the Mother
Goddess of Assam. She is either a lovely and creative mother who cares of
her devotees’ necessities and desires, or a terrifying mother who needs her
33. However, in the Kālikāpurāṇa (10.20–21), Kāma was unable to strike Śiva without the
intervention of Satī with her illusory power (māyā).
34. It was erroneously hypothesized that words for ghost (ke-moyd/ke-moyt; khmoch), grave
(khmoch; kamui), corpse (komuoch; kamet), demon (kemoi), devil (kamoit) derived from *kmuuc
(Kakati 1989: 42).
35. Personal communication (email correspondence from 25 March to 4 May 2017).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
103
devotees’ blood to strengthen her śakti prior to releasing it. When Kāmākhyā
begins to bleed because of her menstrual cycle, during the Ambuvācī melā in
the monsoon season, she loses her ability to vivify the earth.
The absorption of human blood is closely related to the identification of
Nīlācala as an ancient cremation ground (śmaśāna), or graveyard. The śakti
pīṭhas in the Middle Ages were religious centres influenced by trans- and
cross-cultural dialectic; they were closely linked to Tantric exoteric and esoteric practices, although they were absorbed within the fold of Brahmanic
society. However, underpinning blood sacrifices to the Goddess, the śakti
pīṭhas maintained a continuity with their non-Brahmanic past, when
they were probably tribal graveyards (Sircar 1998: 33; Urban 2010: 75–78).
Hence Brahmanism (in its larger sense), through what can be defined as
inter-cultural dialogue and cross-cultural exchange with the tribal traditions
of the hills and forests of the Brahmaputra Valley, was able to incorporate the
graveyards within its theology (Davidson 2002: 224–25; Rosati 2017b: 147–50).
Nīlācala’s ‘primeval function as cremation ground’ was maintained
through the identification of the terrifying Heruka (blood drinker) as the
presiding deity of the cremation ground (śmaśāna-bhairava) at Nīlācala
(Rosati 2017a: 9). Heruka, who is a male deity, is described as ‘being of red
colour’, with a ‘knife’ as his weapon, ‘extremely terrible’ in aspect, greedy
of ‘human flesh’, and of the blood bleeding from the severed heads of ritual
victims. He is naked, ‘stands on a corpse’, is attended by ‘ghosts’, and is worshipped through ‘meditation’. Heruka himself is the cremation ground (KP
63.135–37a, trans. B. Shastri 2008: 474).
The Lord [Hevajra] plays in the cemetery surrounded by his eight yoginīs. ‘In the
cemetery’, we say, because here we have a play on words, for śvasiti means ‘he
breathes’ and śvavasati means ‘resting place of corpses’.
(Hevajratantra 1.3.16, trans. Snellgrove 1980, I: 59)
In the Hevajratantra (1.3.4–5), Heruka is identified with Hevajra. They
were both experienced as states or deities by their devotees during ecstatic
rituals.36
‘Heruka’ is both a wrathful supreme deity with a distinctive but rarely found iconic
form and a generic name for wrathful manifestations of the supreme Buddha that
include Hevajra (with eight heads), Saṃvara (with elephant hide), and Mahākāla
(black, with tongue extended).
(Sharrock 2006–07: 56)
Claudine Bautze-Picron (2014: 5-6) noticed that in a Heruka temple
in Bahal II (Sumatra) there is an icon of Heruka as a double-armed deity,
36. Rob Linrothe (1999: 250) pointed out that according to the literature, Heruka is associated
to ‘skull-cups, ornaments of bones, animal skins, ash-smeared skin’, all elements that
‘reinforce a connection’ with the yoginīs and link Heruka’s iconography to both ‘deities and
human practitioners’.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
104
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
‘trampling or dancing on a corpse’ and surrounded by eight yoginīs (see also
Linrothe 1999: 256). The icon is stylistically close to Bengal iconography,
something that ‘must be seen as part of a network which extends across the
Bay of Bengal in the eleventh and twelfth century’. In the same temple, an
icon which was confused with a ‘demoniac woman standing on a corpse’ has
the same place usually reserved to Hevajra (Schnitger 1937: 27). ‘She wears
large ear rings, puffed hair and a string of skulls. The left hand holds a skull
in front of her breast, the right hand rests on the hip and holds a Vajra and a
sacrificial knife’ (p. 27).37 However, it is far more possible that the icon is not
a female but an image of Mahākāla (Bautze-Picron 2014: 7).
Is it possible that a sex-gender confusion afflicted the image of Heruka’s
manifestation as Mahākāla, in a region such as Assam (a region close to Bengal
and perhaps preserving similar stylistic patterns)? Although we are in the
realm of speculation, in a recent article, I point out the fluidity of Kāmākhyā’s
sex-gender identity, which switched from a female-marked to implicitly
pansexual-marked (Rosati 2019).38 This sex-gender fluidity could reflect the
influences that Kāmākhyā received from contact with Mahākāla-Heruka,
which resulted in a complex process of blending between male and female
deities that originated the Tantric goddess named Kāmākhyā in her medieval
form.39
FROM HERUKA TO THE BAND OF THE YOGINĪS
Thence, Nīlācala and its presiding Mother Goddess seem to incorporate,
through intersections, mutual transformation and fusion, traits of non37. My greatest thanks are addressed to C. Bautze-Picron, who sent me this bibliographical
reference (email correspondence, 22 November 2018).
38. In this context the adjective pansexual is not describing the mixing between female and
male traits, but it describes the sex–gender fluidity as well as the power to transcend any
sex–gender differentiations.
39. R. Davidson (2002: 213) formulated this theory of switching from Heruka to Kāmākhyā,
although he did not enter into any speculation regarding how this process could have
worked out. Thus, Heruka emerges as an intriguing deity, who switched from the pantheon
of Tantric Buddhism to the ranks of Hindu-Tantric deities. According to Davidson (ibid.),
Kāmākhyā appropriated that quality of Heruka as blood-absorber of the deceased, which
is indissolubly implicit in the cremation ground. Today, there are four cremation grounds
on Nīlācala, each placed on one of the four main directions (north, east, south and west),
although the one mentioned in the Kālikāpurāṇa, which was described as Heruka himself,
has been supposed to be the one on the eastern side of the mountain, near the car park and
the cricket field. Nowadays, its presiding deity is a form of Kālī who is inside the small shrine
attached to the cremation field, while outside this small shrine Mahākālī is represented in a
semi-anthropomorphic form, with red skin and clothed with a black dress (see Rosati 2017b:
figs. 1–2).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
105
Brahmanic traditions and their close connection with the cult of the dead.
More specifically, although Kāmākhyā explicitly recalls the universe of sexual
desire, in a far more implicit way she reflects the imaginary of death, through
elements such as ashes, corpses and blood. These elements, however, were
mediated by Purāṇic Hinduism before they burst into the peculiar Tantric
cult of the yoni of Kāmākhyā.
The yoni symbol is a metonymic representation of the Goddess and of her
retinue of yoginīs. According to the Kālikāpurāṇa, indeed, Kāmākhyā and the
yoginīs both originated from the splitting of the yoni of Satī (Rosati 2018a:
176–78) and they both inhabited the yoni and its inner and outer maṇḍala, as a
microcosm of the śākta universe. The text established that ‘all the sixty-four
yoginīs are to be worshipped in the middle of the maṇḍala’ of Kāmākhyā
(KP 54.34–40, trans. Shastri 2008: 384). This maṇḍala is equated to the lotus
flower, and each of its eight petals is associated with a different yoginī. Hence,
‘Śailaputri, Caṇḍaghaṇṭā, Skandamātṛ, and Kālarātri’ are on four petals associated with the four directions ‘beginning with the east’, while ‘Caṇḍikā,
Kuṣmāṇḍī, Katyāyanī, and Mahāgaurī’ are on the petals ‘in the direction following the order from the south-east (agni)’. Mahāmāyā is ‘on the middle
of the lotus’, and ‘the adept should worship’ her ‘by repeating eight times
mūlamantra and pray to the goddess “pardon me”. This is to be followed by
the offering sacrifice (balidāna)’ (KP 54.42–45, trans. B. Shastri 2008: 384).
In the Hevajratantra (1.7), the yoginīs were associated with the cremation
ground and blood sacrifices. However, they also emerged as a circle of dreadful goddesses connected with meditation, such as Heruka, and to the bhaga
(i.e. a ‘triangle white as the moon’ [Snellgrove 1980, I: 73n1]),40 which is the
‘origin of the elements (dharmas)’ (ibid.).
Now I shall explain the circle of the yoginīs. Concentrate upon the triangle of
origination in the midst of space, and then perform the meditation at its centre,
first, the figurative representations of the four elements in their right order—in the
due order of appearance of the divinities. First earth and water, then fire and wind,
which correspond with the appearance of the goddesses, and with the meditator
himself. The maṇḍala which now arises pure and unblemished from the triangle,
consists of two concentric parts, one formed by the eight central petals of the lotus,
and the other by the triangle. At their centre one should imagine a corpse, which
is in effect that seat of the fifteen yoginīs. Resting on that there should be a lunar
disk, upon that the seed-syllable and upon that a solar disk. The conjunction of
these two, lunar disk and solar disk, is great bliss. āli has become moon, and the
sun has revolved into kāli, and from this mingling of sun and moon Gaurī and her
companions are proclaimed to be … Gaurī and her companions arise each from a
separate letter. Now in the inner circle there are five yoginīs, whom the knowing
yogin always regards as representing the five skandhas. In the east is Vajrā, and Gaurī
to the south, Vāriyoginī in the west, Vajraḍākinī to the north, and Nairātmyā at the
40. Bhaga also means ‘vulva’ (Monier-Williams 1960: 743); however, any association to sexuality
in this context can be only speculated.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
106
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
centre. In the outer circle there are Gaurī II, Caurī, Vatālī, Ghasmarī and Pukkasī,
Śavarī and Caṇḍālī, and Ḍombinī as eighth. At the zenith is Khecarī and at the nadir
Bhūcarī, O thou of great compassion, and these two stand to represent saṃsāra and
nirvāṇa. All these goddesses are black in colour and exceeding fearful … They have
one face and inflamed eyes and clasp in their hand the knife and the skull.
(Hevajratantra 1.8.1–16, trans. Snellgrove 1980, I: 73–74)
Thus the band of yoginīs, not yet crystallized into the Kaula group of
sixty-four, was fully recognised in connection with the directions of space
and the maṇḍalic representation of the Universe. This trait was probably
absorbed by the yoginīs because they were the retinue of Heruka/Hevajra,
a deity widely linked to mental speculation. Then, in a continuous process
of transmission and transformation, they channelled this trait to Kāmākhyā
who was their leader in the Hindu-Tantric tradition of medieval Kāmarūpa.
THE YONI OF KĀMĀKHYĀ AS A HYPER-BLEND SPACE
The goddess Kāmākhyā, through a historical-religious analysis, emerges as
the result of the dialectic of a number of traditions and their related symbols.
However, the process that stands at her origin is far more complex. In previous studies (Rosati 2016, 2017a, 2017b, 2018a, 2018b), the intersection of the
processes of universalization and parochialization was variously pointed out.
In this article, the primary purpose is to shed light on the intertwining of
distinct traits and on their selective projection and superimposition, through
CSR.
The goddess Kāmākhyā and her earthly abode, Nīlācala, blended a number
of dichotomic symbols, which may be simplified and summarized in a
primary division between sex/life and violence/death. The Mother Goddess
of Assam, indeed, is a dangerous and terrifying mother, who is in need of
blood to increase her power (śakti); but at the same time she cares about her
devotees, giving them back her power through her yonitattva (menstrual
blood), which vivifies the earth and is the key ingredient to obtain any siddhi
in the Kaula sexual rites.
The yoni as the metonymic symbol of the Mother Goddess Kāmākhyā
blended the two symbolic universes of blood and sexual fluids (Fig. 4). A key
role in this transmission and superimposition of selected elements from distinct traditions into the yoni of Kāmākhyā seems to be played by the band of
yoginīs. On the one hand, they mediated the Kaula symbolism of Śiva-Śakti
into the yoni, which indeed reflects both these principles.41 On the other
hand, they channelled to Kāmākhyā the death imaginary related to the
41. Some notes on the yoni as a pansexual symbol were presented during the 44th Spalding
Symposium on Indian Religions, which was held at Storey Institute, Lancaster (12–14 April
2019), and they are most of the material used for Rosati (2019).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
107
Figure 4: The CSR reading of the yoni metonymic symbol.
cremation ground and to Heruka, whose retinue they were in the Buddhist
esoteric tradition.
The association of blood sacrifices and sexual rites in the Kaula tradition
and at Kāmākhyā is not part of the Hevajratantra, but it can be supposed to be
Figure 5: The CSR reading of the goddess Kāmākhyā (Yoni-Goddess).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
108
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
a later development. The Kaula school perceived sexual rites as the counterpart of blood sacrifices (White 2006: 17). Both kinds of ritual were (and still
are) necessary to release the śakti, which is the main purpose of the Tantric
practitioners at Kāmākhyā (Urban 2001: 783). Hence, the Hindu-Tantra medieval yoni intersected with the memory of the endless Śiva-Śakti union which
followed the dismemberment of Satī’s corpse (Fig. 5). This intersection was
given textual form in the Kālikāpurāṇa, the most ancient text where the splitting of the Goddess and of the yoni is narrated.
Śiva-Śakti’s union and Satī’s dismemberment were already blended
spaces that collaborated in the generation of the hyper-blend—i.e. the yoni.
A further clarification is needed. The yoni symbol is neither a substitute for
Satī’s death nor an equation of Śiva-Śakti. On the contrary, it is a new structure that incorporates both those inputs in a new network. This metonymic
structure did not exist in the two original inputs, but emerges from their
blending. The yoni compresses together death and sex, originating a larger
symbol that is a ‘mental space’, which is unconsciously shared by the devotees at Kāmākhyā. In this way, what Glen Hayes (2012: 201) said about Tantra,
that it ‘may be regarded as a blend which has been repeatedly “reblended”’,
is corroborated by the yoni symbol of Kāmākhyā.
Input 1: Satī’s death
The first input is the dismemberment of Satī’s corpse, which is replicated in
every blood sacrifice performed at Kāmākhyā and in every śakti pīṭha. The
animal slaughter and dismemberment palpably recalls this episode, which
originated the Tantric network of śakti pīṭhas. The Goddess’s death was not
only the first step of Tantra in the Brahmanic world, but also a conceptual
blend,42 which cross-mapped the violence of blood spreading to nourish the
yoginīs—in a religious context centred on the cult of the dead—with the multiplication of the Goddess’s cults across South Asia.
Blood sacrifices, through a selective projection, switched from being
related to the cult of the dead to a praxis related to the female body.
Beheading a sacrificial victim became one of the powerful sources of power
for the goddess Kāmākhyā. Furthermore, this extremely dangerous ritual act
traces its roots back to the head-hunter tribes of northeastern India.
42. Although the Kālikāpurāṇa’s extant recension belongs to ninth to eleventh centuries,
Hazra (1963: 194–259) speculated on the existence of an earlier version of this Purāṇa
in the sixth century. There is no evidence regarding the existence of this earlier source;
nevertheless, it is possible that it existed, basing on the very different traditions preserved
within the Kālikāpurāṇa (Wendt 1996: 180). Hence, it is possible that an earlier version of the
Kālikāpurāṇa, or an earlier source that influenced its extant recension, preserved the full
story of Satī (Rosati 2017a: 12–13)—which, instead, was overlooked in the Hindu-Purāṇic
literature belonging to the northern and northwestern regions of the sub-continent.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
109
Previous studies associated the head-hunting raids with the death of a
tribal chief, whose grave was adorned with the skulls of the killed enemies
in order to lead him to the other world. However, head-hunting practices
represent ‘heroism and manhood’ (Zou 2005: 89). Coming back from a raid
without any head trophy was a sign of shame for the tribal warriors, for
whom head-taking was essential for marriage. In fact, a warrior could appropriate the enemy’s soul to himself by cutting off their head, a necessary act
for the generation of offspring. The human head, indeed, being linked to
bone marrow and its sexual symbolism, was inter-culturally described as the
‘seat of life’ or the ‘seat of soul’ (Onians 1994: 94–98; Zou 2005: 94–97).
What was not projected from input 1 to the blended space (i.e. yoni) is the
cremation ground, which was the primary location of the yoginīs before the
Kaula reformation.
Input 2: Śiva-Śakti
The second input is the endless union of Śiva-Śakti in every śakti pīṭha, which
was replicated through Kaula sexual rites when the male Tantrika drank the
female fluids from what the Kaula texts call the ‘mouth of the yoginī’ (i.e.
the yoni). In medieval Kāmarūpa, every woman was considered a yoginī (KjN
22.8–12), although their agency in ritual is ambivalent. According to Loriliai
Biernacki’s (2007) study of a group of northeastern medieval Tantras, the
woman is described either as a divine being or as a mere ritual object which
was used by Kaula siddhas (perfect beings) in order to obtain their sacred
fluids. However this may be, the woman is fundamental in Kaula theology.
The sexual symbolism of Śiva-Śakti union is projected and enhanced at
Kāmākhyā, where the main deity is a Yoni-Goddess—indeed, according to the
Purāṇic sources the gods make love secretly at Kāmākhyā. There is no other
Purāṇic pīṭha in South Asia that claims to house the yoni of Satī.43
Hence, Śiva-Śakti is another conceptual blend, which fused puruṣa (spirit)
and prakṛti (nature) in a pansexual union—which underlined the necessary
crossing of sexual binarism to initiate cosmogenesis. This pansexual unity
was mediated through the yoni of Kāmākhyā in medieval Kaulism, where any
Brahmanic sexual taboo was violated in order to obtain supernatural powers
and religious gnosis. However, apparently this input projected only the feminine aspect, the yoni, while the masculine one is completely hidden, as no
liṅga is preserved inside the garbhagṛha.
43. In the Mahābhārata (3.80.100–05; 3.82.80–85; 3.82.130–35) three tīrthas related to both the
breast and the womb of the goddess are described. The yoni tīrtha is placed at Udyantaparvata
(Mahābhārata 3.82.130–35), which was supposed to be on eastern Himalayan range (Sircar
1998: 9). However, there is no mention of any yoni stone in the text, but the site is described
as a yonidvāra (‘gate-of-the-womb’, trans. Van Buitenen 1975: 390).
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
110
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
The Two Sides of the Goddess Kāmākhyā
The hyper-blend thus manifested itself as a non-coherent space which integrates eros (desire) and thanatos (death)—which today are both unconsciously
perceived by the devotees. Although there is no analogy between the binomials death/violence and life/desire, after a centuries-long cross-mapping
they have become the two faces of the same coin. Kāmākhyā emerged as a
polymorphic deity as well as a hyper-blended space. Her connection to the
universe of kāma is evident in her name. Entering inside her garbhagṛha, the
practitioners feel that they have penetrated inside the maternal womb—the
moisture mixed with the darkness of the chamber and the final contact with
the sacred water that oozes out from the sacred recess of the yoni-stone are
indisputable evidence of a connection between Kāmākhyā and motherhood
(Fig. 6).
Hence, this symbolism is related to the peculiar power of the yoni of
Kāmākhyā—which is not the śakti per se, but a śakti blended with an ancestral
energy that has its roots in the universe of death. Climbing down the steps
towards the garbagṛha and touching the yoni allows the devotees to transcend
our worldly universe. This aspect of Kāmākhyā is the most hidden part of her
Tantric cult. Her devotees do not consciously perceive this ancestral symbolism, and the medieval texts have not explicitly preserved it.
Figure 6: Mother feeding her child, Kāmākhyā temple complex (inner northern enclosure-wall).
Photograph by Paolo E. Rosati, 2016.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
111
This energy, although dangerous, is extremely powerful, and it is intertwined with the ‘holy water of Kāmākhyā’ and its ‘intoxicating power’ (Van
Kooij 1972: 25–26). Thereby bathing in the sacred pond of Kāmākhyā has
allowed the practitioners to access ‘heaven’ (KP 80.31b–32a), while drinking
its water was both the source of siddhis (KP 62.88b–90), and the vehicle to
reach the mahānirvāṇa (highest extinction) (KP 80.87). This ritual description is a clear exotericization of the Kaula sexual rituals that have been kept
secret—being reserved to a restricted circle of initiates who were part of
the kula (clan). Thereby, the dangerous power of the yoni was opened to and
accessed by a larger number of practitioners, Tantric as well as non-Tantric.
CONCLUSIONS
How is it possible that the Indian Mother Goddess par excellence is also a cruel
and dreadful mother? How were apparently irreconcilable elements, such as
eros and thanatos, blended in the Tantric cult of Kāmākhyā?
Today, the worship of Kāmākhyā in its public cult of her menstruating
yoni—located inside her garbhagṛha—is a peculiar and partially exotericized
left-hand Tantra praxis. On the one hand, the dreadful aspect of the Mother
Goddess is remembered through the spreading blood of the dozens of ritual
victims slaughtered and dismembered in the sacrificial open hall attached
to the main temple of Kāmākhyā. These blood sacrifices, although excluded
from the main shrine, are still performed inside the temple’s enclosure, and
are not replaced by metaphoric or metonymic substitutes. They are necessary to nourish the Mother Goddess and to feed her śakti. While their earlier
secrecy in association with the non-Brahmanic cult of the yoginīs is nowadays
lost, the animal slaughter palpably recalls the ancestral death imaginary and
its symbolic development within Satī’s story. Notwithstanding, the terrifying aspect of Kāmākhyā is the least perceptible trait of the Goddess in her
temple. Her dark and dangerous nature is not emphasized in the practitioners’ perception of the divine.
On the contrary, the practitioners’ perception emphasizes the caring
nature of Kāmākhyā as a Mother Goddess. Notwithstanding, sexual rites are
the most secret part of her worship; nowadays, only rumours regarding their
practices can be noted. Sexual rites are the counterpart of the blood sacrifices, and are necessary to unleash the śakti of the Goddess—which is the
source of extraordinary siddhis. Thus, the Kaulas through the worship (pūjā)
of a human yoni and the consumption of its secretions were able to obtain
siddhis beyond gnosis and mokṣa (liberation).
The exotericization of the yoni pūjā not only softened the Kaula sexual
praxis but was also a fundamental step in blending death imagery and sexual
symbolism. In fact, eros and thanatos had followed two distinct paths which
intersected through the absorption and reformation of the cult of the yoginīs
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
112
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
in the religious history of Kāmākhyā. From a historical religious perspective,
firstly the yoginīs conveyed the death and blood imagery, related to Heruka,
to the Hindu-Tantric cult of the yoni at Kāmākhyā. Secondly, the yoginīs were
transformed into sexual tools, becoming the main ritual agent in Kaula
sexual rites. In this way, the universe of desire accepted an agent formerly
connected to the universe of death.
In sum, the yoginīs played a fundamental double role as either active or
passive agents, contaminating Brahmanism (in its larger sense) and being
contaminated by Brahmanism. The cult of the yoginīs and its role in shaping
the Hindu-Tantra cult of Kāmākhyā corroborated the idea of a mutual osmosis
between Brahmanic and non-Brahmanic tradition in ancient and early medieval Kāmarūpa. Through the application of the hyper-blend theory to the
cult of Kāmākhyā, an apparent contradiction in the Tantric cult of the yoni
has been detected. Furthermore, the reading of Kāmākhyā as a hyper-blend
space is able to explain her far more hidden symbols related to death imagery
and to her nature as a dreadful mother.
In conclusion, the cult of the yoginīs seems to have played a fundamental
role in blending different already blended spaces into the peculiar Assamese
cult of the yoni of Kāmākhyā.
ABBREVIATIONS
CSR
KjN
KP
Cognitive Science of Religions
Kaulajñānanirṇaya
Kālikāpurāṇa
REFERENCES
Primary Sources
Brahmāndapurāṇa. Ganesh V. Tagare (ed. and trans.). The Brahmānda Purāṇa. Translated and
annotated by G. V. Tagare. Vol. 4 of 5. Delhi: Motilal Banarsidass Publisher, 2000 [1983].
Brahmapurāṇa. Jagdish L. Shastri (ed.). Brahmāpurāṇa. Translated by a Board of Scholars. Vol. 1 of
4. Delhi: Motilal Banarsidass, 1985.
Brahmavaivartapurāṇa. S.L. Nagar (trans.). Brahmavaivartapurāṇa. Text with English Translation.
Delhi: Parimal Publications, 2012.
Bṛhatsaṃhitā. M. Ramakrishna Bhat (ed. and trans.). Varāhamihira’s Bṛhatsaṃhitā. Delhi: Motilal
Banarsidass, 1981–82.
Devībhāgavatapurāṇa. Raj T. Pandey (ed.). Devībhāgavatapurāṇa. Kashi: Pandit Pustakalya. 1956.
Devīmāhātmya. (1) Vasudeva S. Agrawala (ed. and trans.), Devī-māhātmya: The Glorification of the
Great Goddess. Ramnagar/Varanasi: All IndiaKashiraj Trust, 1963. (2) See Coburn (1991).
Devīpurāṇa. Sharma Pushpendra Kumar (ed.). Devī Purāṇam. New Delhi: Sri Lal Bahadur Shastri
Kendriya Sanskrit Vidyapeeth, 1976.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
113
Hevajratantra. David L. Snellgrove (ed. and trans.). Hevajra Tantra: A Critical Study. 2 vols. London:
Oxford University Press, 1980.
Kālikāpurāṇa. Biswanarayan Shastri (ed. and trans.). The Kālikāpurāṇa: Text, Introduction and
Translation in English with Shloka Index. Delhi: Nag Publisher, 2008 [1991].
Kāmākhyātantra. Biswanarayan Shastri (ed.). Kāmākhyā-Tantra. Delhi/Varanasi: Bharatiya Vidya
Prakashana, 1990.
Kaulajñānanirṇaya. Chandra P. Bagchi, (ed.) and Michael Magee (trans.). Kaulajῆāna-nirṇaya of the
School of Matsyendranatha. Varanasi: Prachya Prakashan, 1986 [1934].
Kulacūḍāmaṇi Tantra. Louis M. Finn (ed. and trans). Kulacūḍāmaṇi Tantram and Vāmakeśvara Tantra,
with the Jayaratha Commentary. Wiesbaden: Otto Harrassowitz, 1986.
Mahābhārata. Ramachandra N. Dandeakar and Vishnu S. Sukthankar (eds.). Śalyaparvan: Being the
Ninth Book of the Mahābhārata. Bhandarkar Pune: Oriental Research Institute, 1961.
Mahābhārata. Johannes A. B. Van Buitenen (ed. and trans.). The Mahābhārata. Vol. 2: 2. The Book of
Assembly Hall; 3. The Book of the Forest. Chicago: University of Chicago Press, 1975.
Matsyapurāṇa. B. D. Basu, (ed.). The Matsya Puranam. Vol. 2 of 2. New York: AMS Press, 1917.
Śivapurāṇa. Jagdish L. Shastri (ed.). The Śiva Purāṇa. Vol. 1 of 4. Translated by a Board of Scholars.
Delhi: Motilal Banarsidass, 2000 [1969].
Skandapurāṇa. Ganesh V. Tagare (ed. and trans.). The Skandapurāṇa. Translated and Annotated by
G.V. Tagare. Vol. 1 of 23. New Delhi: Motilal Banarsidass, 1992
Tantrāloka. Raniero Gnoli (ed. and trans.). Abhinavagupta. Luce dei Tantra: Tantrāloka. Milan:
Adelphi, 1999.
Vamanapurāṇa. Anand S. Gupta (ed.). Vāmanapurāṇa. With English Translation. Varanasi: All-India
Kashiraj Trust, 1968.
Viṣṇupurāṇa. Horace H. Wilson (ed. and trans.). The Viṣṇu Purāṇa: A System of Hindu Mythology and
Tradition. Translated From the Original Sanskrit and Illustrated by Notes. Derived Chiefly from Other
Purāṇas. Vol. 1 of 5. London: Trübner & Co., 1864.
Yoginītantra. Biswanarayan Shastri (ed.). Yoginī Tantra. Delhi: Bharatiya Vidya Prakashan, 1982.
Yonitantra. Jan A. Schoterman (ed.). The Yonitantra. Critically edited with an introduction. Delhi:
Manhohar, 1980.
Secondary Sources
Aktor, Mikael. 2016. ‘Cāṇḍālī as Śakti: Untouchable Women in Some Tantric Texts.’ In Bjarne W.
Olesen (ed.), Goddess Traditions in Tantric Hinduism: History, Practice and Doctrine: 96–108. New
York: Routledge.
Appfel-Marglin, Frédérique. 1985. ‘Female Sexuality in the Hindu World.’ In Clarissa W. Atkinson
(ed.), Immaculate and Powerful: The Female in Sacred Image and Social Reality: 39–60. Boston, MA:
Beacon Press. https://doi.org/10.1177/004057368704400115
Bautze-Picron, Claudine. 2014. ‘Buddhist Images from Padang Lawas Region and the South
Asian Connection’. In Daniel Perret (ed.), History of Padang Lawas, North Sumatra. II: Societies of
Padagang Lawas (Mid-Ninth–Thirteenth Century ce): 107–128. Paris: Cahiers d’Archipel. https://
doi.org/10.2307/j.ctv1qv3nd.20
Benton, Catherine. 2005. God of Desire: Tales of Kāmadeva in Sanskrit Story Literature. Albany, NY:
State University of New York Press.
Bhattacharyya, Narendra N. 1974. History of the Śākta Religion. New Delhi: Munshiram Manoharlal
Publishers.
—— 1995. Religious Cultures of North-Eastern India. New Delhi: Manohar.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
114
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
Biernacki, Loriliai. 2007. Renowned Goddess of Desire: Women, Sex and Speech in Tantra. New York:
Oxford University Press.
Borkataky-Varma, Sravana. 2019. ‘Red: An Ethnographic Study of Cross-Pollination between
the Vedic and the Tantric.’ International Journal of Hindu Studies 23 (2): 179–94. https://doi.
org/10.1007/s11407-019-09258-z
Brown, Cheever Mackenzie. 1990. The Triumph of the Goddess: The Canonical Modela and Theological
Visions of the Devī-Bhāgavata Purāṇa. Albany, NY: State University of New York Press.
—— 2002. The Song of the Goddess. The Devi Gita: Spiritual Counsel of the Great Goddess. Albany, NY:
State University of New York Press.
Bryant, Edwin. 2001. The Quest of the Origins of the Vedic Culture: The Indo-Aryan Migration Debate.
Oxford/New York: Oxford University Press.
Carroll, Lucy. 1977. ‘“Sanskritisation”, “Westernisation”, and “Social Mobility”: A Reappraisal of
the Relevance of Anthropological Concept of the Social Historian of Modern India.’ Journal of
Anthropological Research 33 (4): 355–71. https://doi.org/10.1086/jar.33.4.3629747
Coburn, Thomas B. (ed. and trans.). 1991. A Translation of the Devī-Māhātmya and a Study of Its
Interpretation. Albany, NY: State University of New York Press.
Davidson, Ronald M. 2002. Indian Esoteric Buddhism: A Social History of the Tantric Movement. New
York: Columbia University Press.
Dehejia, Vidya. 1986. Yoginī Cult and Temples: A Tantric Tradition. New Delhi: National Museum.
Dold, Patricia. 2004. ‘The Mahavidyas at Kamarupa: Dynamics of Transformation in Hinduism.’
Religious Studies and Theology 23: 89–122. https://doi.org/10.1558/rsth.v23i1.89
Donaldson, Thomas E. 2002. Tantra and Śākta Art of Orissa. Vol. 2 of 3. Delhi: D. K. Printworld.
Doniger, Wendy. 1973. Asceticism and Eroticism in the Mythology of Śiva. London: Oxford University
Press.
Dyczkowski, Mark S.G. 1988. The Canon of the Saivagama and the Kubjika Tantras of the Western Kaula
Tradition. Albany, NY: State University of New York Press.
Erndl, Kathleen M. 1993. Victory to the Mother: The Hindu Goddess of Northwest India in Myth, Ritual,
and Symbol. Oxford: Oxford University Press.
Eschmann, Anncharlott. 1986 [1978]. ‘Hinduization of Tribal Deities in Orissa: The Śākta and
Śaiva Typoloy.’ In Anncharlotte Eschmann, Hermann Kulke and Gaya C. Tripathi (eds), The
Cult of Jagannath and the Regional Tradition of Orissa: 125–37. Delhi: Manhoar.
Fauconnier, Gilles and Mark Turner. 2003. ‘Conceptual Blending, Form, and Meaning.’ Recherches
en communication 19: 57–86. https://doi.org/10.14428/rec.v19i19.48413
Flood, Gavin. 2006. The Tantric Body. The Secret Tradition of Hindu Religion. New York: I. B. Tauris.
Hatley, Shaman. 2014. ‘Goddesses in Text and Stone: Temples of the Yoginīs in Light of Tantric
and Purāṇic Literature.’ In Benjamin J. Flaming and Richard D. Mann (eds), Material Culture
and Asian Religions: Text, Image, Object: 196–225. New York: Routledge.
Hayes, Glen A. 2012. ‘Conceptual Blending Theory, “Reverse Amnesia”, and the Study of Tantra.’
The Journal of Hindu Studies 5: 193–209.
Hayes, Glen and Staneshwar Timalsina. 2016. ‘Introduction to “Cognitive Science and the Study
of Yoga and Tantra”.’ Religions 8 (181): 1–11. https://doi.org/10.3390/rel8090181
Hazra, Rajendra Chandra. 1963. Studies in Upapurāṇas. Volume 2: Śākta and Non-Sectarian Upapurāṇas.
Calcutta: Sanskrit College.
Kakati, Banikanta. 1989 [1948]. The Mother Goddess Kāmākhyā, or, Studies in the Fusion of Aryan and
Primitive Beliefs of Assam. Guwahati: Punya Prasad Duara for the Assam Publishing Corporation.
Lakoff, George and Mark Johnson. 1980. Metaphors We Live By. Chicago, IL: University of Chicago
Press.
Linrothe, Robert N. 1999. Ruthless Compassion: Wrathful Deities in Early Indo-Tibetan Esoteric Buddhist
Art. Boston, MA: Shambhala.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
THE ROOTS OF THE TWO SIDES OF KĀMĀKHYĀ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
115
Lussana, Gioia. 2015. ‘Fluid Mother Goddess: Water and Blood as Flowing Sacred Essence of Mahā
Devī in the Śākta Tantrism of Kāmākhyā.’ Rivista degli Studi Orientali 88 (2): 73–83.
Marriott, McKim. 1955. ‘Little Communities in an Indigenous Civilization.’ In McKim Marriott
(ed.), Village India: Studies in the little Community: 171–222. Chicago, IL: University of Chicago
Press. https://doi.org/10.1017/s0035869x0011562x
Monier-Williams, Monier. 1960 [1899]. A Sanskrit–English dictionary: Etymologically and Philologically
Arranged with Special Reference to Cognate Indo-European Languages. New Edition, Greatly Enlarged
and Improved. Oxford: Clarendon Press.
Onians, Richard B. 1994 [1951]. The Origins of European Thought about the Body, the Mind, the Soul, the
World, Time, and Fate: New Interpretations of Greek, Roman and Kindred Evidence also of some Basic
Jewish and Christian Belief. Cambridge: Cambridge University Press.
Redfield, Robert. 1955. ‘The Social Organization of Tradition.’ The Far Eastern Quarterly 15 (1):
13–21.
—— 1956. Peasant Society and Culture. An Anthropological Approach to Civilization. Chicago, IL:
University of Chicago Press.
Rosati, Paolo E. 2016. ‘The Yoni Cult at Kāmākhyā: Its Cross-Cultural Roots.’ Religions of South Asia
10 (3): 277–99. https://doi.org/10.1558/rosa.35343
—— 2017a. ‘The Cross-Cultural Kingship in Early Medieval Kāmarūpa: Blood, Desire and Magic.’
Religions 8 (212): 1–18. https://doi.org/10.3390/rel8100212
—— 2017b. ‘The Goddess Kāmākhyā: Religio-Political Implications in the Tribalisation Process.’
History and Sociology of South Asia 11 (2): 137–55. https://doi.org/10.1177/2230807517703014
—— 2018a. ‘The Death of Satī and the Worship of Her Yoni: The Rise of the Yoginī Kaula School
in Early Medieval Assam.’ In Marina E. Miranda (ed.), Dal Medio all’Estremo Oriente: Studi del
Dottorato di ricerca in Civiltà dell’Asia e dell’Africa: 171–86. Rome: Carocci editore.
—— 2018b. ‘Nīlācala: The Mountain of Desire, Death and Rebirth.’ In David W. Kim (ed.), Colonial
Transformation and Asian Religions in Modern History: 30–53. Newcastle upon Tyne: Cambridge
Scholars Publishing.
—— 2019. ‘The Yoni of Kāmākhyā: The Intersection of Power and Gender in its Mythology.’
Religions of South Asia. 13 (3): 317–347.
Sanderson, Alexis. 1988. ‘Śaivism and Tantric Tradition.’ In S. Sutherland (ed.), The World’s
Religion: Religions of Asia: 128–72. London: Routledge and Kegan Paul.
Schnitger, F. M. 1937. The Archaeology of Hindoo Sumatra. Leiden: Brill.
Sharma, Mukunda M. 1978. Inscriptions of Ancient Assam. Guwahati: Dept. Publications Guwahati
University.
Sharrock, Peter D. 2006–07. ‘Hevajra at Bantéay Chmàr.’ The Journal of the Walters Art Museum
64/65: 49–64.
Shin, Jae–Eun. 2010. ‘Yoni, Yoginīs and Mahāvidyās: Feminine Divinities from Early Medieval
Kāmarūpa to Medieval Koch Behar.’ Studies in History 26 (1): 1–29. https://doi.org/10.1177/
025764301002600101
—— 2010–11. ‘Changing Dynasties, Enduring Genealogies: A Critical Study on the Political
Legitimation in Early Medieval Kāmarūpa.’ Journal of Ancient Indian History 27: 173–87.
—— 2018. ‘Region Formed and Imagined: Reconsidering Temporal, Spatial and Social Context of
Kāmarūpa.’ In Lipokmar Dzüvichü and Manjeet Baruah (eds), Modern Practices in North East
India: History, Culture, Representation: 23–55. New York: Routledge. https://doi.org/10.4324/
9781351271363-2
Singer, Milton. 1955. ‘The Cultural Pattern of Indian Civilization.’ The Far Eastern Quarterly 15 (1):
23–36.
Sircar, Dinesh C. 1998 [1948]. The Śākta Pīṭhas. Delhi: Motilal Banarsidass.
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
116
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
PAOLO E. ROSATI
Srinivas, Mysore N. 1952. Religion and Society among the Coorgs of South India. London: Asia
Publishing House.
Szántó, Péter-Dániel. 2012. ‘Selected Chapters from Catuṣpīṭatantra. 1/2: Introductory Study
with the Annotated Translation.’ DPhil thesis, University of Oxford.
Timalsina, Staneshwar. 2016. ‘A Cognitive Approach to Tantric Language.’ Religions 7 (139): 1–19.
https://doi.org/10.3390/rel7120139
Tiwari, Shiv Kumar. 2002. Tribal Roots of Hinduism. New Delhi: Sarup and Sons.
Urban, Hugh B. 2001. ‘The Path of Power: Impurity, Kingship and Sacrifice in Assamese Tantra.’
American Academy of Religion 69 (4): 777–816. https://doi.org/10.1093/jaarel/69.4.777
—— 2010 [2009]. The Power of Tantra: Religious, Sexuality and the Politics of South Asian Studies. New
York: I. B. Tauris.
Van Kooij, Karel R. 1972. Worship of the Goddess According to the Kālikāpurāṇā. Leiden: E. J. Brill.
Wedemeyer, Christian K. 2012. Making Sense of Tantric Buddhism: History, Semiology and Transgression
in the Indian Traditions. New York: Columbia University Press.
Wendt, Sylvia. 1996. ‘Satī – Saumyā Satī. Two Versions of one Myth in the Kālikāpurāṇa.’ In A.
Michaels, C. Vogelsanger and A. Wilke (eds), Wild Goddesses in India and Nepal: Proceedings of an
International Symposium Bern and Zurich, November 1994: 179–87. Bern: Lang.
White, David G. 2006 [2003]. Kiss of the Yoginī: ‘Tantric Sex’ in Its South Asian Contexts. Paperback ed.
Chicago, IL: University of Chicago Press.
Yokochi, Yuko. 1999. ‘The Warrior Goddess in the Devīmāhātmya.’ Senri Ethnological Studies 50:
71–113.
Zou, David V. 2005. ‘Raiding the Dread Past: Representations of Headhunting and Human
Sacrifice in North-East India.’ Contributions to Indian Sociology 39 (75): 75–105. https://doi.
org/10.1177/006996670503900103
© Equinox Publishing Ltd 2021
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44