Professional Documents
Culture Documents
Canon of The Saivagama and The Kubjika Tantras of The Western Kaula Tradition (Mark S. G. Dyczkowsk)
Canon of The Saivagama and The Kubjika Tantras of The Western Kaula Tradition (Mark S. G. Dyczkowsk)
Editorial Board
Edward Dimock, Wilhelm Halbfass, Gerald J, Larson,
Wendy D. O’Flaherty, Andre Padoux, Navjivan Rastogi,
Ludo Rocher, Alexis Sanderson.
The
Canon of the Saivägama
and the
Kubjikä Tantras
of the
Western Kaula Tradition
Dyczkowski, M ark S. G.
The canon of the Saivagam a and the K ubjika
Tantras of the western K aula tradition.
10 9 8 7 6 5 4 3 2 I
Acknowledgem ents v ii
P art O ne
P art Two
P art T hree
Appendices
Appendix A 95
A History of the Study of the Kubjika Cult
A ppendix B 97
The M anthanabhairavatantra
A ppendix C 101
The Canon of the Jayadrathayam ala
The Pitha Division
The M antrapl(ha
The VidyapHha
The M udrapitha
The M aijdalapltha
The ‘Eight Times Eight* Bhairavatantras
The Srotas Division
Conclusion
A ppendix D 127
M anuscripts of the K ubjikatantras
Abbreviations 135
Notes 137
Bibliography 203
Index 209
A cknow ledgem ents
Although it is not possible to say exactly when the first Agamas were
written, there is no concrete evidence to suggest that any existed much
before the sixth century. The earliest reference to Tantric manuscripts
cannot be dated before the first half of the seventh century. It occurs in
Bana’s Sanskrit novel, Kadam barl in which the author describes a Saiva
ascetic from South India who “had made a collection of manuscripts of
jugglery, Tantras and M antras [which were written] in letters of red lac on
palm leaves [tinged with] sm oke.” Bana also says that “he had written
down the doctrine of M ahakafa, which is the ancient teaching of the
M ahapasupatas,”7 thus confirming th at oral traditions were in fact being
committed to writing.
If our dates are correct, it seems that the Saivagamas proliferated to
an astonishing degree at an extremely rapid rate so that by the titne we
reach A bhinavagupta and his immediate predecessors who lived in ninth-
century Kashmir we discover in their works references drawn from a vast
corpus of Saivagamic literature. It is this corpus which constituted the
source and substance of the Saivism of Kashmir, which Kashmiri Saivjtes,
both monists and dualists, commented on, systematized and extended in
their writings and oral transmissions. The dualist Siddhantins, supported
by the authority of the Siddhantagam as, initiated this process by
developing the philosophical theology of the Siddhanta. Subsequently,
from about the middle of the ninth century, parallel developments took
place in monistic Saivism which drew inspiration largely, but not
exclusively, from the Bhairava and other Agamic groups which
constituted the remaining part of the Saiva corpus.
It was to this part of the Saivagam a that the sacred texts of the
Pascimamnaya belonged. These Tantras, unlike those of the Siddhanta,
advocated in places extreme forms of Tantricism that actively enjoined
such practices as the consum ption of meat and wine as well as sex in the
course of their rituals.
Agamic Saivites who accepted these practices as valid forms of
worship constituted a notable feature of religious life throughout India.
Although many of these Saivites were householders, the mainstays of these
traditions were largely single ascetics, many of whom travelled widely and
in so doing spread their cults from one part of India to another.
An interesting example of this phenom enon is Trika, nowadays
virtually identified with monistic Kashmiri Saivism. Abhinavagupta, who
was largely responsible for developing Trika Saivism into the elevated,
sophisticated form in which we find it in his works, was initiated into Trika
by Sam bhunatha, who came to Kashmir from neighbouring Jalandhara,
Sumati, his teacher and an itinerent ascetic like himself, was said to have
travelled to the N orth of India from some “sacred place in the South.”8
6 T he S a iv a A gam as
The Jain Somadeva confirms that Trika was known in South India during
the tenth century. Somadeva identifies the followers of the Trikamata as
Kauias who worship Siva in the company of their Tantric consorts by
offering him meat and wine.9 It is worth noting incidently that Somadeva
was very critical of the Trika Kauias. “If liberation,” he says, “were the
result of a loose, undisciplined life, then thugs and butchers would surely
sooner attain to it than these K auias!” 10 Although Somadeva was a Jaina
monk and so would naturally disapprove o f such practices and tend to take
extreme views, it appears nonetheless that Trika was not always as elevated
as it now seems to us to be.
Although the Pascimamnaya is entirely confined to Nepal at present,
it was, according to one of its most im portant Tantras, the K ubjikam ata,
spread by the goddess to every corner of India, right up to Kanyakumari
in the South, identified, by allusion, with Kubjika the goddess of the
Pascimamnaya. 11 A long list of initiates into the Pascimamnaya and their
places of residence is recorded in the K ubjikanityahnikatilaka, a work
written before the twelfth century. It is clear from this list that the cult had
spread throughout India although it was certainly more popular in the
N o rth .12 That the PaScimamnaya was known in South India in the
thirteenth century is proved by references in M ahesvarananda’s
M aharthamanjarV3 to the K ubjikam ata14 as well as a work called
i4pascim am ”15 which may or may not be the same work but most probably
belongs to the same tradition. An old, incomplete manuscript of the
Kubjikamata is still preserved in the m anuscript library of the University
of Kerala in T rivandrum .16
Despite relatively early references to the existence of Agamic Saivism
in the South, it seems that the Saivagamas originally flourished in northern
India, spreading to the South only later. Madhyadesa (an area covering
eastern Uttar Pradesh and west Bihar) was, according to A bhinavagupta,
considered to be the “repository of ail scripture” 17—hence also of the
Saivagama and the K ulasastra.58 The im portance of this part of India is
indirectly confirmed by the fact that Benares, in the centre of this area, is to
be visualized as a sacred place (pifha) located in the heart of the body in the
course of the Kaula ritual described in Tantraloka, 19 Similarly, Prayaga
and Varapasi are projected in the same way onto the centre of the body
during the ritual described in the Yoginihrdaya,20 a Kaula Tantra of the
original Saivagama. The sacred circle (mandala) shown to the neophyte in
the course of his initiation into the cult of the Brahmayamala is to be drawn
in a cremation ground with the ashes of a cremated hum an corpse. In it are
worshipped Yak^as, Pisacas and other demonic beings, including Rak?asas
led by; Ravana, who surround Bhairava to whom wine is offered with
oblations o f beef and hum an flesh prepared in a funeral pyre. The name of
The Saivagamas 7
Thus the term “T an tra,” which has anyway a wide range of connotations,26
is used at times to denote a Sakta Tantric scripture to distinguish it from a
Saiva Agama, whereas we find both words used in the early corpus without
distinction.27
The reason for this loss has certainly much to do with the ethos of
Hinduism itself and its history as a whole. The secrecy that these types of
Tantras have always imposed on themsdlves is indicative of the uneasiness
which these Tantric cultural elements must have aroused in m any.28 Thus
the Puranas, which are from m any points of view the bastions and
guardians of Hindu orthodoxy, initially tended to reject the authority of
the Tantras and so largely avoided quoting from them. However, insofar as
the Puranas aimed to be complete compendiums of Hindu spirituality and
practice, they later included long sections from Tantric sources, especially
when dealing with ritual, the building and consecration of temples, yoga
and related m atters.29 Thus there is much Tantric m aterial to be found
everywhere in the Purapas. The Brahmavaivartapurana contains, a
brilliant theological exposition of the Supreme as the Goddess Nature
(Prakrti), which is a theme dear to T antra in its later phases. The D evi
Devlbhagavata, the Kalika and large portions of the Narada Puranas are
extensively Tantric. The hymns eulogizing the names of the goddess,
accounts of her actions, lists of female attendants of male gods, their
M antras, Yantras and much more show how strong the influence of
Tantric ideas was on the Puranas. They also dem onstrate that such trends
were not only clearly apparent in the history of the development of Tantra
but applied to them also. These developments, in other words, concerned
the whole of the literate tradition, and so the P uranas could, without
difficulty or self-contradiction, incorporate relevant material from
the Tantras.
These incorporations were drawn from the entire range of Tantric
sources available at the time in which they were made. It seems likely, in
fact, that a possible way in which we can gain some idea of when these
Tantric passages were added to the Puranic text is to establish the type of
Tantric source from which they were drawn. No one has yet attem pted to
apply this method in an extensive or systematic way. However, the validity
of this approach finds the support of R. C. Hazra, who in his work does
attem pt to date some Puranic passages on this basis.30 The A gni and
Garutfa Purapas, for example, deal extensively with Tantric topics. Their
treatm ent is based largely on the Saivagamas and Pancaratrasam hitas
which belong to the early Tantric period, i.e., prior to the tenth century. As
an example relevant to the study of the Pascimamnaya, we may cite
chapters 143-147 o f the Agnipurana where the goddess Kubjika is extolled
and the m anner in which she is to be worshipped is described.30 As Kubjika
The Saivägamas 9
The wise man should not elect as his authority the word of the Vedas,
which is full of im purity, produces but scanty and transitory fruits and is
limited. [He should instead sustain the authority] of the Saiva scriptures,
Abhinava remarks:
therefore seek a cause which was immediate and directly effective. Nor do
we have to look far. The eleventh century, which marks the beginning of
this change in the Tantric tradition, coincides with a sudden reversal in the
course of India’s history, namely, the advent of Muslim rule. In the
beginning of the eleventh century the brief incursions of Muslim raiders
into Indian territories that had been going on for centuries turned, under
the lead of M aham ud of Ghazni, into a full-scale invasion. The onslaught
of Islam forced Hinduism to retreat, challenging its resistance and
stability as a whole. The Muslim scholar, Al-BirunI, who came to India
with the invading armies noted that “the Hindu Sciences have retired far
away from those parts of the country conquered by us, and have fled to
places which our hand cannot yet reach, to Kashmir, Benares and other
places,“43 M aham ud of Ghazni did not manage to conquer Kashmir,
although he plundered the Valley in 10I4A .D . and again attacked it, this
time without success, in the following year. But even Kashmir, although
outside the M uslim ’s reach for the time being, felt the intense impact of the
Muslim presence in India. Ksemendra, the Kashmiri polymath, describes
in his Acts o f the Incarnations o f Visnu (Dasdvataracarita) written in
1066 A.D., the dire conditions that will prevail in the world on the eve of
the coming of Kalki, Vispu’s last incarnation who was to finally herald the
dawn of a new age of freedom. He says:
The D ards, T urks, Afghans and Sakas will cause the earth to wither
as do the leper his open, oozing sores. Every quarter overrun by the
heathen (mleccha)ythe earth will resound with the sound of swords drawn
in com bat and her soil will be drenched with blood.44
After the twelfth century the Siddhanta seems to have been losing
popularity over most of India, giving way to more syncretic forms of
Hinduism. The Siddhantins also seem to have lost their posts as Royal
Preceptors with the downfall of the dynasties that patronized them, as did
the Kalacurls in the early thirteenth century. W hat remained of the
Siddhanta was apparently annihilated by the ever-increasing Muslim
incursions into Central India from the more northern regions of the
country already under Muslim dom inion. The Hindu will to resist the
Islamic invaders, never very firm, was demoralized by Prthvlraja I l l ’s
12 T he S a iv a A gam as
defeat at T arian in 1192. M alwa, the S iddhanta’s hom eland, was raided
by the M am luk Delhi S ultan Iltutmish in 1234-35, by his successor
Balban in 1250, and by the Khaljl Sultan Jalaiuddln in 1292. The
Muslims finally trium phed in 1305, when A lauddin Khalji overran
Malwa, destroying, am ong other places, the sacred city of Ujjayini and
Bhoja’s capital D hara, The Siddhanta had to take refuge in the Tamil
country to survive.45
though charged with meaning, had lost all significance for them beyond
the ritual act itself. Thus the Nepalese, unlike the Kashmiris, did preserve
the original Tantras they made use of in their rituals, but largely failed to
see anything beyond their immediate content.
But apart from these extrinsic factors, a highly significant intrinsic
factor contributed to the loss o f these scriptures, namely, the internal
development of the Saivagama itself. The Saivagamas, even those most
Saiva-oriented, accomm odated within themselves the concept of Sakti.48
This trend developed within some of the Saivagamas towards such a
female-oriented view that at a certain stage, they simply ceased to be Saiva.
The Sdktatantras took over and, permeated with the earlier Kaula doctrine
and ritual forms, preserved, along with those Tantras which continued to
consider themselves to be Kaula (although not directly connected to their
predecessors), the presence of this antinom ian element within Hinduism.
The old was transform ed into something new, which replaced what had
gone before so completely that all that remained was a dim memory of a
glorious past in the form o f the names o f the ancient Agamas now given to
new works.49 The most hardy survivor of these far-reaching changes was
the mild Kaula cult of Srividya. The Pascimdmnaya is another Kaula cult
that has managed to survive almost to the present day, although only
within the narrow confines of Nepal.
vijayottara were singled out as the ultimate authorities for the Trika
Tantricism he sponsored. He also sustained the Siddhayogesvarimatd’s
claim to being the most im portant of all the scriptures.
The prim ary textual tradition orders itself into canonical corpuses of
sacred literature, which reflect upon themselves as belonging to a single
group and so strive to concretely supplement and extend each other within
the param eters chosen for themselves. We can observe this happening
more commonly with primary texts belonging to the early formative
period of Tantra. This tendency is more noticeable the more restricted the
group becomes. Thus the scriptures of the Pascimamnaya, which
constitute a subdivision of a much larger category o f scriptures, form a
relatively coherent group. The rituals o f the individual Tantras of the
Pascimamnaya do, in fact, share many common details. The basic
M antric system, for example, is fairly uniform in the majority of the
Pascima Tantras, and m atters which one T antra deals with cursorily are
taken up and elaborated in another.
In this way the huge body of primary texts consists of a manifold in
which each member is independent although connected, more or less
directly, through the mediation of com m on affiliations variously
established, to others. It is not surprising, therefore, if we come across a
certain am ount of incoherence (not to say, at times, outright contra
diction) even between texts belonging to the same group. Indeed, one of
the functions o f the Tantric m aster in the ordering and development of the
Tantric tradition is to sort out these textual problems. Thus Abhinava-
gupta explains in his Tantraloka that the Tantric m aster who teaches his
disciples the meaning of the scriptures m ust, if necessary, proceed through
them as a frog does, leaping from place to place within them. He must have
an eye for every detail and observe it in its broader context, viewing the text
as a whole, as does the lion strolling through the jungle who looks in all
directions as he walks. Paraphrasing the Devyayamala, Abhinava says:
The m aster should explain the statem ents [of the T antra,
dem onstrating and] corroborating their consistency by means o f their
interconnection and apply this principle to the various sections of the
text, its chapters, sentences, words and root meaning (sutra). He should
ensure that the preceding and subsequent sections of the scripture do not
appear to contradict each other and so apply, as required, the principles
[that one must move from one section to the next] as a frog leaps
[om itting what is irrelevant], or as the lion who looks around [in all
directions as he walks]. In this way he should coalesce the meaning of the
scripture into a single coherent expression unconfounded with other
teachings. Knowing well the meaning o f each phrase as he expounds it,
he avails himself of sound associations (tantra), repetition (avartana),
M ajor and Secondary Divisions 15
Vámasiva
Dak§inasiva
Siva
Miáraáiva or Yamalaáiva
Síddhantaáiva
Paáupata
Saiva Som asiddhanta
Lakula
the same four declaring, mistakenly, that they are Pasupata sects because
their followers agree that PaSupati is the instrum ental cause of the
universe.59 Kesava Kasmlrin agrees with this view,60 while Srikantha
correctly explains that they are all believers in the Agamas revealed by
Siva, rather than just Pasupatas.61
We find the same set of four in the Purapas and other independent
sources. Lorenzen has collected a num ber of references to these groups
and lists them in his book. Although Lorenzen’s chart is somewhat
lengthy, the material it contains is sufficiently relevant to our present
discussion for us to quote most of it here.62 See Table I.
coincides with the addition of the Bhairava and Vam atantras, which are
major components of the Saivagamic canon. Thirdly, the entry
“Kdpalika” (variants: Kapala, Soma, Saumya and Kankala) appears
regularly in almost all the lists and so must be considered to be an
independent group. This is true also of the “Pasupata” and the “Lakula”
(variants: Nakula, Vakula, Lanjana, Laguda and Larigala). Finally, we
notice that “K alam ukha” (variants: Kalam ukha, Kalanana and Kalasya)
sometimes appears to take the place of the Lakulisa PaSupata entry. The
Kalamukhas studied Lakulisa’s religion (Lakulasamaya)80 and doctrines
(Lakulasiddhanta)M Kalam ukha teachers are regularly praised in
inscriptions by identifying them with Lakulisa.82 There can be no doubt,
therefore, that the K alam ukhas were Lakulisa P aiupatas. Even so, not all
Lakulisa Paiupatas were Kalam ukhas; it would therefore be wrong to
simply identify the two.
Let us now turn to a more detailed discussion of these groups
individually. We shall deal first with the Pasupata sects, then discuss the
Kapalikas, and then finally turn to the Saivagamas and their major
divisions.
I, Rudra, for the first time created the mysterious religion o f the
Pasupata, beneficent to all, facing in all directions, one th at can take
years or only ten days to master, one which although censored by the
foolish because it is in places opposed to the order of the rules of caste and
stages of life (varnasramadharma), is nevertheless appreciated by those of
perfected wisdom (gatanta) and is in fact superior to it.84
20 T he S a iv a A gamas
historical basis there is for these names, there appears at least to have been
a tradition which admits the existence of Pasupata teachers prior to
Lakullsa. But, w hether Lakullsa was the first Pasupata or not, he is
without doubt an im portant founder figure whose contribution was so
substantial that he came to represent Pasupata Saivism as a whole.
Presumably this is why A bhinavagupta divides Saivism (Saivasasana) into
two main currents (pravdha): one associated with Lakullsa and the other
with Srikanfha, whose teachings (sasana) consist of the five major streams
(srotas) of the Saivagamas we shall discuss later.93
Unfortunately, no original Pasupata scriptures have been recovered.
Moreover we are hard pressed to find evidence to prove that such
scriptures ever existed. We do come across expressions like “Pasupata-
sastra’’ and even hear of its fabulous size94 but we have managed to trace
only one concrete reference to a possible Pasupata scripture. This occurs
in Bhaftotpala’s tenth century comm entary on V araham ihira’s Brhat-
samhita95 where he says that the Pasupatas worship Siva according to the
procedures enjoined by the Vatulatantra, All of the few works so far
recovered belong to Lakulisa’s school. The oldest is the Pasupatasutra
attributed to Lakullsa himself. We also have a commentary called
“pancdrthabhdsya"by Kaundinya, whose date, although far from certain,
is generally thought to be sometime between the fourth and sixth centuries
A .D .96 A lthough we cannot be sure that the sutras are, as the commentator
says, by Lakullsa himself, they do, in fact, appear to be quite old and bear
many archaic traits. These are apparent particularly in the figure of
Pasupati himself who is identified with Prajapati and associated with the
Vedic Rudra with whom he shares a num ber of Vedic names such as
Aghora, Ghora, Sarva and Sarva. A part from the Pasupatasutra the only
other extant Lakullsa Pasupata work is the Gatjakarika by Hara-
dattacarya and a tenth century com m entary, the Ratnaftka, by
Bhasarvajna. These works, along with summaries of Lakullsa Pasupata
philosophy found in medieval treatises on the philosophical systems, are
the sole sources we possess.97 It is possible that Lakull^a’s disciple,
Musalendra, wrote a work called the Hrdayapramana98 and there are
numerous quotes from lost works both in the Ratnaiika and Kauijtjinya.
Unfortunately, the sources are never named, and it is hard to say whether
they are original Agamas or not, or if they are specifically Pasupata
scripture or secondary works.
The spiritual discipline these works prescribe does not involve
complex rites or require extensive intellectual development. It is, however,
largely intended for the renunciate, rather than the householder. Thus lay
worshipers have only to recite obeisance to £iva (namah sivaya) with
folded hands while the celibate ascetic is given much more to do. He can be
22 T h e S a iv a A g a m a s
either fully naked or wear a single strip of cloth to cover his privities. He
should practice austerities, such as the penance of sitting amidst five fires.
After his morning ablutions he smears his body with ashes and does the
same at noon and in the evening. After his bath he goes to the temple where
he sits to meditate on Siva. As he does so, he should sometimes laugh
loudly, sing and dance. Before leaving and saluting the deity, he repeats
the seed-syliable “huduk " three times and recites his m antra. When he
bathes alone, he should pay homage to the lineage of Pasupata teachers
{tlrthesa) headed by Lakulisa. After his bath he should select a clean place
for m editation and stay there to practice it through the day. In the evening
the site is again cleared and purified with ashes. When he feels sleepy, he
again spreads ashes on the ground and lies down to sleep.
When the Pasupata yogi has developed a degree of spiritual insight
(jnana), his teacher permits him to practice antinom ian behaviour." At
this stage of his spiritual discipline, he should act like a m adm an ignorant
of right and wrong. Pretending to sleep, he snores loudly or rolls on the
ground and talks nonsense. When he sees a beautiful woman, he should
make lewd gestures at her. In this way he courts abuse in the belief that his
disgrace will gain for him the double benefit of purifying him of his sins
and gaining the merit of those who abuse him.
The Lakulisa ascetic is, however, basically a disciplined, continent
man. M oreover, although he is told to behave in a m anner contrary to
accepted norm s, his conduct falls short of the total abandon extremist
Tantrics allow themselves. He can laugh and sing in the temple but he is
not allowed to offer Siva anything else if it is not prescribed.500 In fact, his
behaviour is regulated by injunctions (vidhi) down to the smallest detail.
For instance, he must offer garlands to Siva, but they must not be made of
fresh unconsecrated flow ers.101 He cannot simply abandon himseif on his
own initiative; he must wait for his teacher’s permission to do so. His lewd
gestures are just play-acting: in reality he must avoid wom an’s company
whenever he can. He is specifically prohibited from even talking to
w om en102 and must be strictly celibate.103 Women are a particularly
dangerous source of tem ptation; they are not em bodiments of the goddess
and as such potential Tantric consorts through whom communion with
Siva could be attained. As Kaundinya says:
mere sight, even at a distance, deludes. The world is bitten by the snake
whose form is w om an’s sexual organ who, with m outh cast downwards,
moves between [her] thighs, beyond all control [even that] of the
scriptures.504
Tranquil, with the mind under one’s control, the body covered with
ashes, devoted to celibacy and naked, one should observe the Pasupata
vow. In form er days I created the supreme Pasupata vow, more secret
than secret, subtle and the essence of the Veda, [for m an’s] liberation.
The sage, devoted to the practice of the Vedas, wearing nothing but a
loincloth or single piece of clothing, should meditate upon Siva, the Lord
of Beasts (P asupati).127
But even though Siva enjoins the observance of the Pasupata vow, he
goes on to say that scriptures of the followers of Lakulisa and the
Pasupatas are am ongst those that he has created which run counter to the
ordinances of the Veda and so should not be follow ed.128 We seem to be
faced with a contradiction. The Lakulisa Pasupata path, as outlined in the
Pa&upatasutra and other extant works of this school, basically falls in line
with the Pasupata path described in the Kurmapuraria. Vet both the
Pasupatas and the Lakulisa Pasupatas are repeatedly censored in this
Purana as heretics and outside the Vedic fold (vedabahya). Similarly, the
Devibhagavata stresses that knowledge of the Veda bears fruit only by
applying ashes to the body as a sign of devotion to Siva. It warns, however,
that the ashes must not be prepared in the m anner described in the Tantras
nor should they be accepted from the hands of a Sudra, Kapalika or other
heretics including, presumably, non-Vedic P asu p atas.129
The Sivapurdpa distinguishes between two types of Saivagama,
namely, Vedic (srauta) and non-Vedic (asrauta). The form er consists of
the essential purport of the Vedas, and is that in which the supreme
Pasupata vow is explained. The latter is independent and consists of the
twenty-eight Siddhantagam as.130 Why then does the Kurma reject some
Pasupatas and not others? Again, w hat should we make of Abhinava’s
analysis of the Saiva teachings (Saivasasana) into two currents—one
associated with Lakulisa and the other with Srikanfha?131 Does he mean
that Lakullsa’s current flows through the Puranas and Smrtis? Probably
not, otherwise he would not distinguish between the “Saiva teachings1’and
those of the Vedic tradition, which he says are its very opposite.132 The
Saivagamas do, in fact, frequently refer with approval to the Pasupatas
and make room for them in their world view.
Thus the Svacchandatantra, as Dvivedi indicates in his article,133 has
homologized the places associated with Siva’s incarnations prior to
Lakulisa with the worlds located in the metaphysical principles (;tattva)
which constitute the cosmic order. The Agamas in general, as we have
already noted, accept both the Pasupata and Lakulisa Pasupata as
branches of the Saiva teachings. M oreover, Abhinava associates them
particularly with the monistic Tantric traditions which have contributed to
the form ation of Kashmiri Saivism and sees them as being intimately
26 T he S a iv a A gam as
The Kapalikas
“My charm ing ornam ents are m ade from garlands o f hum an skulls,”
says the K apalika, “I dwell in the crem ation ground and eat my food from
a hum an skull, I view the world alternately as separate from God (tsvara)
and one with Him, through the eyes th at are m ade clear with the ointm ent
of yoga. . . We (Kapalikas) offer oblations of hum an flesh mixed with
brains, entrails and m arrow . We break our fast by drinking liquor (sura)
from the skull of a Brahmin. At that time the god M ahabhairava should
be worshipped with offerings of awe-inspiring hum an sacrifices from
whose severed throats blood flows in currents.” 133
“W hy,” he then asks, “do you not worship Kapalin? If he does not
receive your worship as Bhairava with liquor (m adhu) and the blood-
smeared lotuses which are hum an heads, how can he be blissful when
embraced by Uma, his equal?”
The LaukikaM stras are the secular arts and sciences, including
politics, medicine, dram aturgy and horticulture. The adhyatmikasastras
are the Samkhya and Yoga while, according to K§emaraja, the knowledge
taught on the Higher Path is “beyond the well-known path of the Veda,
Samkhya and Yoga, etc.. It refers here to the means [to liberation and the
acquisition of power] explained throughout the Supreme Lord’s scripture
in general, namely, the many rituals, mapdalas and gestures (mudra). It is
not independent of the [other] four.” 162 It is quite clear that what
K$emaraja is saying here is that the Saiva groups that belong to the Higher
Path draw from the sacred texts and treatises belonging to the other
categories and do not possess a corpus of scripture of their own. We are
reminded here of the Pasupata K alam ukha teachers who are praised in
32 T h e S a iv a A gam as
This list is im portant for several reasons. Firstly, we know that the
author of the PLSS, Vairocana, was the son of the Bengali Pala king
Dharm apala who reigned from 794 to 814 A .D ..m Thus this list furnishes
an ante quem for these works which can be ascertained with reasonable
accuracy. M oreover, it is the only list we know of at present in which the
Daksiriatantras are related to those of the Siddhanta in the basic five-fold
scheme and in which all the Agamas are clearly enumerated. Thirdly, it
appears that this was an ea rly standard list (or one of them) and that it does
present a faithful picture of these currents of scripture (srotas) at an early
stage of their development, as the following discussion will hopefully
establish.
In the Mrgendragama we find that the Saivagama is divided into five
major currents and eight secondary ones. Unfortunately, the reference is
very concise; even so it supplies us with a num ber of facts. Here it is:
belong to the South, those starting with Sam m ohana are in the North.
To the West is the extensive [group] starting with Trotala. Eastern are
those starting with Cantfasidhara o f C an d an ath a.172
Id a ’s track, he is nectar and gives the power to perform the pure acts
(sucikarman). Between these two nerves lies Susum na, which is the way of
release (mok$amarga).m The Tantra, however, alm ost exclusively deals
with the first two paths. Anyway, the rituals and M antras taught in the
VST lead to both worldly success and release.195 Once the adept has fully
enjoyed all the worldly pleasures the T antra can bestow, he ultimately
reaches Siva’s ab o d e.196
C ontem plate your own nature that creates and destroys [ail things],
beautiful and of the form of passion, as pervading [literally ‘flooding’] the
entire universe and encircled with garlands of flames burning radiantly
from [the abode of] Brahma up to this world. To be successful in all
Garuda and Bhüta Tantras 41
[your] undertakings, rem ember [that you are] Bhairava who is G aruda.2*8
[Contem plate yourself as Bhairava] who has ten arms and four faces. His
eyes are a tawny-brown and he carries a lance. His terrible teeth are
exposed in a fearful grimace. Extremely terrible, he has three eyes and
bears the [crescent] moon on his head. To destroy snakes contemplate
[yourself as] G aruda, awesome and frightening. At his feet lie the hells,
the quarters are his wings, he bears on his chest the seven heavens and the
universe {<brahmantfa) on his throat while his head contains all space.
G aruda, the Lord of the World, is Siva himself who bears Sadariva with
[his] three powers in his topknot. In all your undertakings think of
G aruda brilliant like the Fire of Time, his body M antra, his face
frightening, devouring, three-eyed and his form terrible, the destroyer of
snakes and poison. Having perform ed the projection in this way,
whatever one thinks becomes [easily] attainable and man becomes in
truth G aruda. Seeing him, ghosts, spirits, yaksas, snakes, gandharvas,
rak$asas and all the fevers are destroyed.2*9
As we can see from this example, rites designed to counter the effects
of poison can also be effective against malevolent spirits. It is not
surprising, therefore, if the Garuda and Bhuta Tantras, which specialized
in these matters, had, like the Vama and Daksina Tantras, much in
common. This supposition is supported by the only known extant Tantra
which associates itself with these two groups, namely, the Kriyakdla-
gunottara, manuscripts of which are deposited in the National Archives
in Nepal. K^emaraja quotes this work extensively in the course of his
commentary on chapter 19 of the Netratantra ,220 which deals with the
various types of possession by ghosts and spirits including Pisacas, Mdtr,
Daityas, Yaksas and Rak$asas. It is clear from the introductory verses of
this work quoted in the Nepalese catalogue that this Tantra subsumes
under a single category of concerns the matters treated in both the Garuda
and Bhuta Tantras and deals with them all equally.221 That there was
much common content in the T antras of these two groups is further
confirmed by the citations from the Totulagama K,?emaraja quotes, along
with the Kriyakalaguijiottara, in his com m entary on the Netratantra.222
Although the Totalatantra is listed in the PLSS in the Garutfa group, these
passages refer to possession (bhutaveia and yaksagraha), rather than
magical antidotes for poison.
The Totala and Totalottara are the only Gdrudatantras, and the
Candasidhara the only Bhutatantra, to which we find references.223 It
seems, therefore, that most of these Tantras were lost at quite an early
date. The reason for this is possibly because the matters they dealt with
could be accom m odated into the wider perspective of other less specialised
Tantras. Thus the Netratantra, as we have seen, contains a long chapter
42 T he S a iv a A gam as
The Daksinatantras
The Tantras that issued from Sadasiva’s Southern (daksina) Face are
the Tantras of the Right Current of scriptures (daksinasrotas), while those
that issued from the Northern (vama) Face are those of the Left {vama-
srotas). The Tantras of the Right are called “Bhairavatan.tra s”11*1because
Bhairava is their supreme god and is, in most cases, the one who teaches
them to the goddess, his consort. The Siddhantagamas belonging to the
Upper C urrent and the Bhairavatantras of the Right became the most
im portant of all the Agamic groups.225 As we have already noted, the
Garuda and Bhutatantras were largely lost at a relatively early date. The
Vamatantras must have been valued and studied, as their presence and
influence in distant Cam bodia testifies. Even so their corpus did not grow
as did that o f the Daksinatantras which, on the contrary, developed
extensively. The Siddhantagamas largely superseded all the other Agamas
in the South of India. In the North, in Kashmir and Nepal—the only
regions about which we have sufficient source material to make relatively
detailed assessments—the Vamatantras were mostly ignored (in Nepal) or
relegated to a secondary place (in Kashmir).
The Bhairavatantras neither dwindled in importance nor acquired the
stability of the Siddhantagamas but kept on growing both in terms of their
number and internal categories. We cannot be absolutely sure that similar
extensive developments did not take place in the other currents of the
Saivagama without examining their Tantras or discovering further notices
of them in other sources; even so this possibility seems remote. The fact of
the m atter is that, in the regions in which the Agamas have been preserved,
we are left with two basic categories of Agamic text. One includes the
Siddhantagamas and their numerous subsidiary Agamas (updgama)
which are preserved largely in South India.226 The other, preserved in
Nepal, includes the Bhairavatantras and num erous groups closely
affiliated to them, the most im portant of which are the Kaulatantras we
shall discuss in P art Two of this study. Let us see now how these
developments are reflected in the Agamic accounts of the Saiva canon.
Before the ninth century, the division into five currents of scripture,
gave way to a new basic three-fold division into Left, Right and
S iddhanta.227 This scheme is found in the Netratantra, which presents the
Daksinatantras 43
Listed below are the names of the eight groups of the sixty-four
Bhairavatantras according to the SKS (see Table 6). In the first column
(AI) are listed the names of each group of eight Tantras in the order in
which they are enumerated in the SKS prior to their detailed exposition.
Their corresponding Bhairavas are listed in the second column (Bl). In the
third column (A2) these same groups are listed in the order in which they
appear when the Tantras of each group are named individually in the SKS.
Their corresponding Bhairavas make up the fourth column (B2).
AI Bl A2 B2
Original Order o f the Bhairava Order o f Detailed Bhairava
Groups Enumerated Presentation
in the $ K S
L Bhairava Svacchanda Svacchandarüpa Bahurupa
2. Yamala Bhairava Yamala
3. M ata Canda M ata Canda
46 T h e S a iv a A gam as
What we want to establish is that the eight Bhairavas who are said to
preside over the eight groups of Tantras are in fact the eight Tantras that
belong to the first of these groups, namely, the Bhairavasiaka. Once we
have done this, we can go on to compare this group with the eight Tantras
that head the list of Daksipatantras in the PJLSS. First of all, we can
assume that the order of these groups is as we have it in column A 1. This is
a reasonable assum ption insofar as this order coincides exactly with the
one we find in the JY. Now if we compare these lists, we find that entries
3) A l+B l and 3) A2+B2 as well as 4) A l+ B l and 4) A2+B2 correspond
exactly. To 8) A I+ B 1 corresponds 7) A2+B2. It is clear that Kapalisa has
been displaced from his position as no. 8. Again, the empty space created
by the absence of a Bhairava for the Yamala group238 seems to have moved
Cantfa and Unmatta of list BI down one place. If this is so, the order of the
first five Bhairavatantras in the Bhairavasiaka of the SKS and JY
corresponds exactly to those of the Daksinasrotas according to the PLSS.
Again 6) and 7) B2 are Ruru and Kapalisa who follow one another as no. 6
and 7 in the list of the Daksinatantras. The original order given in the SKS
(column B 1) places Kapalisa in the eighth place with M ahocchusma in the
seventh. It seems, however, that the seventh and eighth were originally
Ruru and Kapalisa respectively because their corresponding divisions,
Bahurupa and Vagina, are the seventh and eighth in list A I. If this is so,
then M ahocchusma is dislodged from its position as no. 7 in column B1
and moved up to the place of no. 6. In this way this T antra preserves its
place next to Asitahga. The resultant order then is: Svacchanda, Canda,
Krodha, Unmatta, Asitahga, M ahocchusma, Ruru and Kapalisa. If this
order is correct, then all that needs to be done to the list of Dak$inatantras
is to eliminate Sam uccayam —which is not the name of a Bhairava—and
M ahocchusma can then be accommodated in the gap left in position 6. It
is clear, therefore, that the Bhairavasiaka and the first eight Tantras of the
Daksinasrotas were originally the same. In other words, what came to be
known as the Bhairavasiaka was a standard group in the Bhairavatantras
of the Daksinasrotas which, possibly because it headed the list of these
Tantras, came to be considered as the basis of the sixty-four Bhairava
tantras. The Kamikagama says: “The Bhairavatantru originated two-fold
from the Southern M outh.”339
Daksinatantras 47
Can it be that the two types mentioned here were the first eight
Bhairavatantras as one group and the remaining Daksinatantras as the
other? That the Bhairavdstaka existed as a group in its own right is
confirmed by the Nityasodasikaryava which refers to it as constituting
eight of the sixty-four Tantras that it enumerates (see below). The list of
eight Bhairavatantras in the BY is further proof that this is so. Thus, if we
identify Ugra with Carina and restore him to his place as no. 2 in the list,
then the first four correspond exactly, while of the three not found in the
Bhairavdstaka only one is not found in the list of Tantras belonging to the
Daksinasrotas.24° Finally, it is worth noting that, apart from these eight,
not a single T antra in the SK S’s list corresponds to any of the Daksina
tantras noted in the PLSS. The reason for this seems to be that the first
eight Tantras of the Daksinasrotas have been extracted from it to serve as
the basic model for the a$tastakabheda, which although originally just a
part of the Daksinasrotas assumed an independent status in a different
sphere from the original Daksinasrotas. This appears to be clearly the case
when we consider that the SKS retains the older classification as subsidiary
to its own Trika-based exegesis of the Saivagamic corpus in which the
Daksinasrotas now figures as consisting of twenty-four Tantras and not
thirty-two. Is this not because the Bhairavdstaka has been removed
from it?241
The Bhairavdstaka is not the only group which has acquired an
identity of its own. A nother im portant group is that of the Yamalas, In the
Kdmikdgama, the Yamalas (without specifying their number) figure as a
separate category which was not even specifically connected with the
Saivdgama although the possibility that Saivas could practice according to
them was allowed fo r.242 In the BY they form a group of eight along with
the eight Bhairavatantras and other Tantras in the Vidyapifha to which the
BY itself belongs.243 The Yamalas are represented as forming a group of
their own also in the JY; so too in the NSA 244 which is probably older than
the BY.245 There can be no doubt, however, that there were a good deal
more than eight,246 and judging from the original Yamalas still preserved,
many were probably of considerable length. Finally, Bahurupa and Mata
are two divisions of eight found both in the SKS and the NSA indicating
that they were also considered to be groups in their own right.
Let us consider next the sixty-four Tantras as a whole. A comparison
of the lists of sixty-four Tantras found in the NSA and in the SKS proves to
be highly instructive from many points of view, both because of their
similarities as well as differences. Firstly, it is a striking fact that there are
hardly two titles com m on to both lists. This could perhaps be justified by
saying that the SKS lists the sixty-four Bhairavatantras while the NSA lists
what it calls the sixty-four Matrtantras. In this case, however, the
48 T h e S a iv a A gam as
least two centuries before Abhinavagupta, that is, in the eighth or ninth
century and is probably older. This is the most we can say at present.
Before we proceed to the next section of our exposition, a few remarks
remain to be made about some other Tantras listed by the PLSS as
belonging to the Dak$iijasrotas. There are three Tantras we notice herein
this list that we know to be Trikatantras, namely, the Tristram (called
“Trisirobhairava"or “Triiirom ata”in Kashmir), the Nisisameara and the
Siddhayogesvaram (or Siddhayogesvarimata).251 Their presence in this
list establishes that these Tantras are old members of the Saivagama. If we
accept that these are amongst the Dak$i#atantras which existed at the time
of the compilation of the Siddhdntagamas, there are good grounds to
argue that they are older than some of them, at least. Thus, it transpires
that Tantras teaching Trika doctrine and ritual already existed at the time
of the form ation of the Siddhanta as a coherent group of Saivagamas.
Moreover, it may also be argued, in broader terms, that the Siddhanta’s
notion of itself as a group presupposes the existence of an older
classification into four divisions to which it has added itself as an upper
fifth. This is a standard pattern of development of the canon as we shall
have occasion to observe again when we come to deal with the dmndya
division of the K ulatantras and the place of the Pascimdmndya in it.
The pifhas are also generally linked together in pairs. The BY treats
the Vidya and M antra pifhas together and similarly considers the M udrd
and M a q fa la p ifh a to be a pair.271 This agrees with A bhinava’s exposition
of the division of the pifhas found in the Anandasastra which states:
East
M udrapifha
M aydalapifha
West
The M udrd and M andala pifhas seem to have been the least well
defined of the four pifhas. The BY enumerates the Tantras of the Vidya—
and M antra—pifhas as belonging to the current of the Right but then
simply states that the other two pifhas include all the Mudras and
Mandalas of the Tantras in all the currents of scripture.273 Similarly, the
JY refers to only one root T antra in the Mandalapifha and then simply
states that this pifha is part o f the contents of all the pifhas. The
M udrapifha contains only three root Tantras of which one is the
K ubjikdm ata,274 The M antra and Vidya pifhas are thus generally
considered to be the most im portant pair and we shall therefore limit
our discussion to them.
Mantrapifha 53
The Mantrapifha
Svacchanda — Asitdhga
Canda — Ruru
Krodha — Jhahkdra
Unmatta — KapaMa
If we place the right-hand column below the left, we have the eight
Bhairavatantras in the order in which the JY enumerates them. It seems,
therefore, that we can trace a line of development here from the BY
through the Sarvavira to the JY. These Tantras must, therefore, also
succeed each other chronologically.
The Vidyapifha
currents of scripture, viz., Left, Right and Middle, The JY has, however,
rearranged their contents. The Siddhant agamas have been entirely
excluded from the pifha classification.280 Therefore, the Middle Current is
now vacant and a new category is created to take its place, namely, the
Sakthantras. However, this is just a new name for old familiar Tantras
amongst which are the Siddhayogesvarimata, the Sarvavira and the JY
itself. It is worth noting that the SYM is regularly assigned to this pitha.
The BY does so and Abhinava tells us that it is this pitha which dominates
in this, the root Trika T antra, and hence also in the Mdlinivijaya, which
presents the essentials of the form er.281 In fact, the SYM itself tells us that
it belongs to the Vidydpifha,282
The BY locates the Vidya and Mantra p ith a in the current of the
Right283 while the JY extends the Vidydpifha to include the Tantras of the
Left amongst which are the M ahasammohana and Nayottara. Although
the Vinasikha is not amongst the major Tantras, it may be the Sikhatantra
listed as one of the secondary Tantras associated with the Satnm ohana2U
The Right Current of the Vidyapitha consists of the Yamalas amongst
which the BY is considered to be the most im portant. The JY thus allots a
major category to the Yamalas and they are, as we have already had
occasion to rem ark, treated at times as a category on their own. The Tamil
poem, the Takkaydgapparani by O ftakkuttar written in the twelfth
century, frequently refers to the u Yamalaiastra. ” According to this work
there are ninety-one secondary Yamalas and Tantras associated with the
main Yamalas of which one of the most im portant is the BY.285 It is indeed
an extensive and interesting work which deserves to be edited and care
fully studied.
The Mantra and Vidya p it has are closely related, so much so that
Jayaratha says that they stand for Siva and S akti.28fi Similarly, the JY
states that the Mantrapifha is associated with masculine words and the
Vidyapitha with feminine ones.287 Perhaps we can understand this to mean
that the Tantras in the form er group are more Siva-oriented than those
belonging to the latter. The Svacchandatantra, which is said to belong to
the M antrapifha, is indeed markedly more ‘Saiva’ than the Siddha-
yogeivarim ata of the Vidydpifha which is more ‘Sakta’. Moreover,
A bhinava’s statem ent that the Vidydpifha sustains and strengthens the
M antrapitham is exemplified in the context of the Trika exegesis of Saiva
scripture by the secondary and yet vitally im portant place given to the
Svacchandatantra which supplies, amongst other things, along with the
Mdlinivijaya, the cosmology of the Trika.
In Kashmiri circles the Vidydpifha was considered to be the most
im portant of the pifhas. A bhinava quotes the Kularatnamaldtantra to say
that Trika, as a Kaula school which embodies the essence of the doctrines
Vidyapitha 55
of the Tantras of the Left and the Right currents, is superior to them ail.289
He does this immediately after he has extolled the superiority of the
Vidydpifha, implying perhaps in this way that Trika as a whole belongs to
this pifha. Thus Abhinava exalts the Vidyapitha as the ultimate essence of
the other pifhas by stating, on the authority of the Anandasdstra, that all
the pifhas ultimately derive from the Vidyapitha in such a way that, as
Jayaratha puts it: “there is only one pifha which is of the nature of
them all.1’290
The Vidyapitha is also im portant in Nepal. Most of the Tantras
preserved there, which affiliate themselves to a pifha, belong to this one.
Amongst them are two texts which represent themselves as elucidating the
essentials of the doctrines of this pifha. One is called " Vidyapitha”and is
quite short291 while the other, the Srividyapifhamatasara, claims to be
12,000 verses long.292 The Vidyapifha and its im portance in Nepal is
particularly relevant to our present study because m ajor Tantras of the
Kubjika cult affiliate themselves to it. The Manthanabhairavatantra,
which is amongst the most im portant Tantras of this school, belongs to this
pitha293 and tells us that the goddess of this tradition resides in it.294
Certain manuscripts of the KM T bear long colophons that are very similar
in form and content to those of the M BT and include a reference to the
affiliation of the KM T to the Vidyapitha. As these colophons are not
uniform in all the manuscripts, it is hard to say on the basis of this evidence
alone whether the KM T did, in fact, originally affiliate itself to this pitha.
Although, as we have noted above, the KM T does considerits doctrines to
be the essential teachings of all these pifh a s ,293it does not expressly say that
it belongs to any pifha. Possibly the JY is right to assign it to the Mudra-
pitha. If this is so, it appears that later tradition shifted the KM T’s
affiliation to the Vidyapitha. Anyhow, many later Tantras of the Kubjika
cult most certainly do belong to this pifha. Thus the Srimatottaratantra
which is considered to be a direct successor of the KM T (which is also
called Srimata) is a Vidydp it hat antra,196 and so is a Tantra closely
associated with it, namely, the Kadibheda of the Gorak$asamhita.m
To conclude the first part of this m onograph, let us recall what
K. C. Pandey wrote more th an three decades ago concerning Saivagamic
studies: “How can any correct conclusion be possible unless all of (the
Agamas) or at least a respectable num ber of them be carefully read?”298
Indeed, we cannot say much about the structure, history and form of the
Saiva canon without having access to, and carefully studying, the extant
material in m anuscripts which, although a tiny fraction of this vast corpus
of sacred literature, is vast in itself. This is a major area of Indology which
has, sadly, not even gone past the stage of preliminary assessment.
P art T w o
Trika, prides itself in being higher than other Kaula schools because it
includes them all within itself and does not make unnecesssary distinctions
between them, although, of course, it maintains its own independent
existence as a sastra to be followed without resorting to others.35
equal in every way.46 She can also be the unattached yogini encountered by
the adept (.siddha) who, in search of yogic accomplishment, wanders on
pilgrimage to the sacred places of the Kaulas where meetings take place.47
It is from her that the wisdom of the tradition is learned and how the rituals
should be performed. The Manthanabhairava of the Pascimamnaya
insists that there is no difference between the teacher and the yogini.48 The
secret of all the scripture, the supreme essence o f the oral tradition, is on
the lips of the y o g in i49 Thus she is venerated as the Supreme Power which
bestows the bliss of the innate nature of all things (sahajdnanda) and is the
em bodim ent o f Bhairava’s will.50 The yogini is the womb from which the
enlightened yogi is born and her mouth, from which issues the tradition, is
the sacred m atrix {yoni), the triangle consisting of the powers of
consciousness to will, know and act.51 As the womb (yoni) o f creation, it
is the Lower M outh (adhovaktrd) which is the essence of Kaula doctrine.52
This lower face is that o f the yogini equated in the Trika with the Primary
Wheel (m ukhyacakra), namely, that of consciousness into which all the
Secondary Wheels (anucakra) of consciousness—those of the senses, both
physical and mental—-dissolve away and from which they emerge.55 It is
the Circle of Bliss (anandacakra) from which the energy of emission
(■visargasakti) flows forth as Kundalini, that is, as KaulikiSakti, who in the
Pascimamnaya is represented as Kubjika, the presiding goddess.
The ‘Lower M outh1, which is the M outh of the Yogini, is generally
considered by the Kaula tradition as a whole to be the source of Kaula
doctrine. From it flows the sixth current below the five currents of the
Saivagama. The Lower Current is hidden there, below the faces of
Sadasiva, symbolizing its esoteric character.54 By virtue of the monism of
its doctrines, it is said to rise through, and permeate, the other Saiva
traditions, leading them ultimately to the undivided bliss of consciousness
which is the experience of Siva in His highest state (parasiva).55 The
Pascimamnaya, like other Kaula traditions, calls this face “Picuvaktra, "
i.e., the face of the yogini called Picu.56 It is the Face of the Nether Region
(pdtalavaktra) from which creation stream s forth.37 According to a system
of classification outlined in the $atsahasrasamhita of the Pascimamnaya,
the Agamas are divided into seven groups corresponding to the seven
psychic centres in the body (cakra). The lowest centre is the Wheel of the
Foundation (adharacakra), which is that of the N ether Region, followed
by the five currents of the Saivagama spoken by the five faces of Sadasiva.
Above these is the Wheel of the Uncreated (ajacakra). Schoterman
explains that in the first—the highest—m outh resides Siva as Adideva
together with the Adi^akti, while in the seventh—the lowest—mouth
resides the goddess as Guhyasakti: the union of these two mouths is the
M outh o f the Yogini 65
goal,58 One of the points, it seems, that is being made here is that the
Pascimamnaya is the highest of the Agamic schools and contains them ail
by combining within itself both the highest Kaula and the highest Saiva
doctrines.
The tradition which emerges from the yogini’s m outh is called in the
Pascimamnaya, a SaivasrotasP9 The Siddhdntdgamas also consider the
Kaula tradition to be represented by two of eight subsidiary currents
(anusrotas) associated with the five principal currents of the Saivagama.
These two are called Yoginikaula and Siddhakaula. The Yoginikaula is so
called because the yoginfs heard it from Siva’s m outh and kept it within
their own line of transm ission. The Siddhakaula is similarly originally
derived from Saiva doctrine but is transm itted by Siddhas, the male
counterparts of the yoginfs.60 These two categories are well known in the
Kaulatantras and are vitally connected with each other. In the Kaula-
jhananirnaya, M atsyendranatha figures as the founder of the Yoginikula
tradition which is especially associated with the fabulous land of
Kdmarupa ,60 although he himself seems to have belonged to the Siddha,
or Siddham rtakaula.62 T hat Kaulatantras did, in fact, reflect on
themselves as belonging to one or other of these two broad categories
transpires from the characterization of the Urmikaulatantra as belonging
to the Siddhasantana transm itted through one of its branches.63 The
Yoginikula is mentioned in a work quoted by Jayarath a.6‘i The typifying
characterization of these two classes by the KM T of the Pascimamnaya is
essentially the same as that found in the Siddhantagama referred to
above.65 The Pascimamnaya, consistent with its characterization of Kula
doctrine as the tradition which expounds the essence of the teachings of the
Yogini,66 considered itself to be the tradition of the yoginfs (yoginikrama)67
and the secret of their oral transm ission.68 Even so, Pascima doctrine is
considered to be that of Siddhas of the Pascimamnaya69 and is not to be
revealed to those who do not belong to the Siddhakaula school.70 There
are places, however, where the Yoginikula is made to appear to be a part of
the Pascimamnaya.11 Again, Kubjika, the presiding deity, is Kundalini,
which is the essential teaching of the Yoginikula. Thus amongst the Kaula
traditions originating from the sacred places (pithas), the Pascimamnaya
presents the most vital doctrine of all the Kaula tradition72—including the
Yoginikaula. At its highest level, however, the Pahcimamndya agrees with
the Yoginikula that the ultimate object of devotion is Siva (here called
Sambhu). He is the abode of the Sdm bhava state and as such the Supreme
Place that, although beyond all characterization, bestows infinite
qualities. It is where all practice ceases and all things appear immediately
present directly before the yogi.73
66 T he K a u la T a n tr a s
The division into amnayas seems to have been originally into four,
with each amnaya symbolically set in one of the four directions. We have
seen that the KM T knows only of four. The Yogakhaytfa of the MBT also
refers to only four amnayas where they have a m ore tangible identity.83
They are represented as corresponding to the four Ages (yuga), with the
Pascimamnaya as that which is most fit for the present Kali Age.84 In the
$atSS the amnayas are said in various places to be either five, six or even
seven. The five-amnaya division is equated with the five vital breaths in
such a way that the Pascimamnaya corresponds to the Pervasive Breath
(■vyana), the experience of which is the universal pervasion of conscious
ness to which the teaching leads.85 The division into five amnayas (formed
by adding an upper one to the original four) is at times represented as
spoken by the five faces of Sadasiva, following the basic Siddhanta
pattern. The six-fold scheme can be formed by adding a sixth upper
current “beyond the upper“ (urdhvordhva), although a division into six is
also possible by adding a lower current, an example of which we have
already noted in relation to the five-fold Siddhanta pattern with Kula as
the sixth.86 The form er alternative is found in Trikatantras like the
Bhargasikha where Trika is located above the Upper Face which is that
of IsanaA1
The four-fold division appears to be the oldest. This supposition is
confirmed by the Kularnavatantra which characterizes the secret of the
“secrets more secret than secret” ,(rahasyatirahasya) of its own Kaula
doctrines as an upp&i~amnayau situated above the four amnayas to which
the many Kaula traditions belong that are “known to m any.”89 These five
are here said to be spoken by Siva. The Sam ketapaddhati, an early Kaula
text,90 refers to just four amnayas.91 A four-fold division which, as in the
Samketapaddhati, is equated with four metaphysical moments in the
dynamics of ultimate reality, represented as aspects of the power Speech
(bhdratisakti) which issues from the four faces of the “beginningless
M other”— M ahavidya, is found in A m rtananda’s Saubhagyasudhodaya
which he quotes in his com m entary on the Yoginihrdaya91
An account of the spirituality and history of the four amnayas, from
the Pascimamnaya point of view, is recorded in a short but interesting
work called the Cihcinimatasdrasamuccaya. All the manuscripts of this
text located up to now are found in N epal.93 The CM SS claims that it
belongs to the Divyaugha and is a compendium or essence of the Supreme
Kaula doctrine of the Siddha tradition.94 It also implicitly identifies itself
with a type of Tantric work common in the earlier period (i.e., prior to the
68 T he K a u l a T antras
Pürvâmnâya
K hagendranätha
Fourth Yuga
* Oünàtha travelled to Kamaru, to the south of which was a place called Trikhandinl. There he performed austerities
according to the instructions of Candra bimba muni.
Four A m nayas 71
consort was the princess Kum kum a from whom were born twelve princes;
six of these were: Bhadra, A m arapada, M ahendra, Khagendra,
M ahidhara and G undikanatha. These princes were said to have no
authority to teach, while the six others, listed below, did teach and were
the founders of six traditions (ovalli). They are each associated with a
pitha, a town and a forest where they practiced austerities for a varying
number of years according to the instructions of a master. This data is
listed schematically opposite page:
Dak§iQamnaya
Karnesvarl descends into this amnaya. She arises from the three
ptthas and resides in their centre, pleasing to behold as the early morning
sun and yet brilliant like a hundred million lightning flashes. She is the
Passionate One, full of the passion (Kama) which devours Kumarikula
desirous of herself. KameSvarl descends into the world in the form of a
young virgin (kum arl). She melts the Circle of Birth of her own nature by
her energy, and by the intent of her own vitality fills it. She is Kulayogini
of divine form, peaceful and pure as translucent crystal. She has two arms,
one face and three eyes and her waist is thin. She resides on the northern
side of the MalinI M ountain behind which is a bower (gahvara) called the
Place of the Nightingale. It is filled with wild ganders, ducks and other
birds of all sorts. Khecari, Bhucari, Siddha and SakinI reside there
absorbed in meditation.
There, in the Divine Circle of the Triangle, is located the cave called
the Face of the M oon in which resides the goddess, the virgin who is the
flow of vitality (sukravahini). Siddhas, munis (including Krodhamuni)
and ascetics practiced austerities there for thousands of years, until they
became aged and emaciated. Their gaze was averted upwards, to the Inner
Face until they saw the goddess Sukra and thus attained the state of divine
inebriation (ghurmyavastha) by virtue of her divine splendour. Once the
goddess had transm itted this divine knowledge to them, Kamadeva
appeared before her in divine form and ‘melted’ by the power of
Kamesvari. Thus the two became one and gave rise to the Rudra Couple.
The son born of this union was Kaulesa who taught this divine knowledge.
In this amnaya, Kamesvari is described as the twelve-lettered Vidya,
surrounded by twelve goddesses. Then come Vdgeivari, Tripura,
Vagabhava and Bhagamalini who are Kamesvari’s powers {prabhava).
The Dak?i$amnaya is where all the Nityas come from.
72 T he K a u l a T antras
Uttaramnaya
Analysis
This account of the amnayas is striking both for the richness of its
expression and the heights of the yogic experiences it conveys through the
imposing visionary symbolism of the Kaula traditions it presents. The
CM SS is later than the KM T and differs from its doctrinal position in
many respects due largely to the development of Pascima doctrine (see
below). Even so, this account is of value to the historian of Kaula Tantra
and the Pascimamnaya because of what it tells us about the character of
these traditions and their relationship to the Pascimamnaya. Particularly
interesting from the latter point of view is the assignment of Trika to the
Pürvamnáya because of the close relationship that the Trika has with the
Pascimamnaya—a point we shall deal with later when discussing how
Pascima doctrine is built up and its historical antecedents. We shall
therefore refer to it last after discussing the Dak^irya- and Uttar a amnayas.
Dak§ÍQámnaya
Uttarâmnâya
this life in which freedom and enjoyment (m oksa and bhoga) are united.
Although, as we have already had occasion to rem ark, neither Abhinava
nor the Kashmiri authors before him refer to the Krama system as
Uttaramnaya, there is evidence to suggest that it was known as such to
some, at least, of the earlier Tantras, although this may not have been its
original identity in the earliest sources.145 Thus the colophons of the
Yonigahvara by Jñánanetranátha state that this T antra belongs to the
Oñkárapifha of the Uttaramnaya146and says of itself in the body of the text
that it is “the tradition of the Great Teaching,” and “the essence of the
Northern K ula.” 147 Similarly, the colophons of the Devipañcaáatika
declare that this T antra, which deals with the Kálikakrama, belongs to the
Northern T radition.148
It transpires from this evidence that, although we can talk of a “Kula
system” as a doctrinal standpoint in the context of Kashmiri Saivism as
well as Hindu Tantricism in general, the generic meaning of the term
“Kula,” when it is used to refer to the entire Kaula tradition with its many
schools, is not to be confused with the form er sense. Similarly, it appears
that the term “K ram a,” like “K ula,” also conveys a broad generic meaning.
It refers, in one sense, to the sequence of actions in Kaula ritual, the order
of recitation of M antras, deposition (nyasa) of letters or the seed-syllables
of M antras on the body or on a manéala, image or otherrepresentation of
the deity and its surrounding entourage such as a pitcher or the sacrificial
firepit.149 “K ram a” can also mean the liturgy or ritual itself and so is
virtually synonymous with the term “prakriya. ”150 Again the term
“K ram a,” variously qualified, can serve as the appellation of a Kaula
school. Thus the Kashmiri Krajma system as a whole is at times called
“K allkram a” although the term also refers to the order of the sequence of
Kalis worshipped in the course of certain rituals or as a series of states of
consciousness. Similarly, the K ubjiká school or Pascimamnaya is also
sometimes called “Srlkram a.” 151 “K ram a” and “K ula” are in this sense, to
all intents and purposes, virtual synonyms: the expressions “Kalikula” and
“K allkram a” are interchangeable, as are the terms “Srikram a” and
“Srikula.” The term “K ram a” lays emphasis on the typical ritual form a
particular Kaula school exhibits, while the term “Kula” stresses its
doctrinal affiliations and individual identity as a specific Kaula tradition.
Thus the com bination of the two terms, as in the expressions “Kallkula-
kram a” or “Srikulakram a”, although hardly different from “KalIkula” or
“K allkram a”, etc., focus prim arily on the character o f these schools as
possessing distinct liturgies of their own. Again, there appears to be a
distinction between Kaula schools which were “K ram akulas” (or equally
one could say “K ulakram as”) and those that were not, in the sense that the
Tantras of these schools do not align themselves with any Krama. This is
78 T he K a ula T antras
The Purvamnaya
He does so in this case by projecting the /Vzrcfand M dlini M antras onto his
body in the prescribed m anner and, once filled with their cosmic power,
then identifying himself with Bhairava. He now offers libations to
Bhairava and the circles of his energies that surround him, which are
identified with the officiant’s own sensory and mental powers. This is done
by drinking a mixture of male and female sexual fluid (kundagolaka) from
a sacrificial ja r previously filled for this purpose. According to Abhinava,
the officiant attains in this way a vision of the fullness of his universal
nature which has been rendered brilliantly manifest by the energy of the
sacrificial offering. He has, therefore, no need to do anything else unless he
wishes to see this same fullness manifest also in the outer world through the
outpouring of his sensory energies, in which case he proceeds to perform
the outer ritual.164 This begins, as usual, with the worship of the Kaula
teachers in a sacred circle {mandala) drawn on the ground with coloured
powders. The form of this mandala, called “siddhacakra’\ is basically as
we have illustrated it here. See Figure 3. The Yugandthas with their
consorts and disciples are worshipped in the inner square. Their names are
as shown in Table 7.
Figure 3. Siddhacakra.
find groups of three related to one anoth er.181 The goddess Kubjika herself
is three-fold in the form of a young girl, maiden and old w om an.182
Moreover, she is explicitly said to be three-fold as the union of the
goddesses Para, Pardpara and Apara. An im portant triad is here, just as it
is for the Kashmiri Trika, that of Sdmbhava, Sakta and Anava, which are
three basic ritual patterns at one level, and at the inner level of conscious
ness correspond—-as in the Trika—to will, knowledge and action. Cosmic
counterparts are attributed to them in the form of the Three W orlds into
which the universe is divided.183 The whole of the teaching is similarly
divided into these three types,184 which is why there are three types of
initiation {devidiksa) through which the Srikram a becomes m anifest.185
Even more vital and fundam ental than these similarities are the basic
forms of the M antric codes adopted by the Kubjika cult, namely, those of
Sabdardsi and M dlint.m They are the very backbone of the entire M antric
system of this tradition, ju st as they are of that of Trika, so much so that the
god tells the goddess in the KRU that:
Although the use of the future here implies that the god is going to
make Trikatantra after the revelation of the Kubjika cult, there can be little
doubt that Trika precedes the Kubjikatantra and it is the latter which has
borrowed from the form er, not the other way around. Indeed, in places
these Tantras themselves inform us that they have drawn elements from
Trikatantras.m M oreover, the oldest known Pascimatantra, the
Kubjikam ata, must be later than the first Trikatantra, the Siddha-
yogesvarimata, because it refers to it. Again, the M BT and KRU both
know Trika as Trika. This means that they were redacted when Agamic
Trika had reached an advanced stage of development because the earliest
Tantras that taught Trika doctrine and ritual, such as the SYM, did not
consider themselves to be Trikatantrasas such.189 Thus the KMT, which is
earlier than the MBT and KRU, does not refer to Trika as a school,
possibly because it precedes this phase of T rik a’s development. Whether
this is the case or not, it it a significant fact that the later Paicimatantras
know of T rika’s existence especially because reference to Trika is rare in
the primary sources. Moreover, that the CM SS knows the Trikasara, an
extensive Trikatantra frequently referred to by the Kashmiris, shows that
the followers of the Kubjika cult continued to consult Trika sources
throughout its developm ent.190 It is significant, from this point of view,
that it is in Nepal, where the Kubjika cult flourished, that the Siddha-
Kulatantras and Saivagama 85
The Tantras of the Kubjika cult were, however, it seems, Kaula right
from the start and thought of themselves as belonging specifically to the
Pascimamnaya. We know that the KM T is older than the JY —a proto-
Krama T antra—which refers to it (see appendix C) and that it is also older
than the NSA. The JY is well aware of an independent current of Kaula
scripture although it does not say specifically that the KMT belongs to it.
Are we therefore to assume that the Kalikula already existed at the time
and that it represented an Uttaramndya in relation to the K ubjikatantras?
Or is the KM T the oldest extant type of amndya-oriented Tantra? We have
already noted that it nowhere clearly defines the dmndyas of the other
directions, although it refers to them. Could this be because they were
simply empty categories? In other words, did they have no more than an
ideal existence as mere logical complements to an existent Paicimdm-
nayal If we accept this hypothesis, we are led to consider the possibility
that the Kallkrama accommodated itself later to this pattern, as did the
Srividya tradition in a less certain manner. Perhaps, on the other hand, it
would be better to think of them as developing together with their roots
firmly embedded in the Saivagama, drawing life from it and growing out
of it, as well as alongside it.
was informed during a recent visit to Nepal that Kubjika is still worshipped
on certain occasions in the Kathm andu valley although her cult is now
hardly known to anybody.
Scholars have pointed out that references to Kubjika and her school
are rare,196 nor are images of her com m on whether drawn or sculptured.197
This is true of another goddess associated with her worship and whose cult
has similarly been popular in Nepal since the inception of the cult of
Kubjika, namely, G u h y ak all198 The reason for this seems to be that such
cults either disappeared along with countless others or else managed to
survive only at the regional level. A prime example o f this phenomena is
Trika Saivism which, but for its following in Kashmir and the genius of
those who applied themselves to it there, would probably be unknown. It
is not quite right, therefore, to think of the Kubjika cult as a school which
‘‘remained very independent and stood aloof from other Tantrik
s c h o o l s .99 In fact, one of the aims of future research into this school
could be to identify, as far as the available sources perm it, the various
elements of other Tantric traditions which have contributed to its
form ation and to distinguish them from its own original contributions.
becomes bent over with ‘sham e’.205 Again, Kubjika’s crooked form relates
to her nature as Kuridalini who is the m atrix (yoni) or Triangle (srgdta)
from which creation pours forth and in which it resides. As such, she is
bent over, not when awakening but when she is dorm ant, and the pow erof
consciousness (cicchakti) is contracted, which is as one would expect, and
is in fact usually the case in representations of Kuridalini. From this point
of view, consciousness is ‘straight’, i.e., unconditioned when it is free of
‘crooked’ obscuring thought-constructs.206 From another point of view,
Kubjika is said to be crooked because she must contract her limbs to reside
in the body of Kulesvara just as someone whose body is large must stoop
down when moving about in a small hut.207 While according to the
Paratantra, she is bent over because she initially ‘churns’ her navel with
her tongue to give birth to the universe within her womb.208 This is
because, as the CM SS explains, the navel is the Great M atrix (mahayoni)
from which Kuridalini rises. Thus in this account Kubjika is bent over in
order to stimulate her cosmic power which, rising through the body, leads
the yogi to liberation.200
Kubjika is the Supreme Goddess (Para devt) where form is the divine
light of consciousness that shines in the centre of the brilliant radiance
enlightened yogis perceive.210 As such, she is the Great M other they
experience within themselves.211 As K uydalini she is pure bliss, the power
of the Light which resides in all the six centres in the body (cakra) and so is
of six form s.212 As the power of consciousness she is the source of all
M antras and as such has three aspects: Supreme (Para), Middling
(.Parapara) and Inferior (Apara). In this three-fold form she is, just as in
Trika doctrine, M dlinPli consisting of the fifty letters of the alphabet in a
state of disorder, symbolizing the rising of K w jtfaliniand the disruption of
the cosmic order that takes place when it is reabsorbed into the Supreme
M atrix (parayoni)-—the goddess K ubjika—from whence it was originally
emitted. At this level the goddess resides in the state which is Beyond Mind
(unmandbhavatita) as Siva’s divine power (sarribhavasakti) to will, know
and act. She has thus three forms (trirupa) and travels along the three
paths (tripathaga)214 of the sides of the triangle of the organ of generation
(bhaga) which is both the source and ultim ate end of all creation.213 At the
corners of this triangle are located the Mahdpifhas: Purnagiri, Jalandhara
and K am arupa. In the centre is Odiyanapltha where the goddess resides in
union with the divine linga whose nature is bliss itself and whose seed
(bindu) is the Sky of Consciousness.216 In the centre Akula and Kula unite,
while the goddess, as R udra’s power (rudrasakti), pervades each part of the
triangle and lords over it as the Mistress of Kula (Kuleivari) who is also
called M ahakaulika and BhairavL217
According to the CM SS, the form in which Kubjika resides in the
90 T he K a u la T a n tr a s
centre of this triangle is as a tam arind tree: Cine ini. Thus it refers to the
doctrine of the Kubjikatantras as Cincinisastra.2>8 The abode of this
doctrine is the Circle of the Divine Transmission (<divyaughamandala),
higher than that of the Transmission of the Perfected Ones and of Men
(Siddha and Manavaugha).219 It is under this Tamarind Tree that
Ctncininatha (alias Vrksanathd) attained, by divine command, the highest
state.220 This tree is the tree of the light of consciousness (cincinicit-
prakasa) brought into this world by Siddhanatha22' who planted it on the
Island of the M oon (candradvipa) which is in the Great Ocean of Kula.
The roots of this Tree and its branches are extensive and it bears the fruits
of the Divine Transmission. It is Kaula rooted in Kula and flourishes in
Siva’s Circle. The juice of this tree is one’s own nature (svasvabhava); its
young sprouts are the Vedas and Vedahgas; its flowers are the senses, and
the divine fount which waters it is the delight of consciousness, its inner
glory.222 To lie in its shade is to experience the highest state of rest
(visranti) In which one is free of the opposites of pleasure and pain.
body.243 Her body is pure consciousness and bestows the bliss of the
‘churning’ or arousal of the power of enlightened consciousness.244 As the
power of consciousness, she is also Speech and as such is adorned with the
fifty letters of the alphabet and resides in this form within the divine
triangle of the Three Peaks once she descends into it along the Path of
M eru.245 The MBT, according to the colophons, has also emerged along
the Path o f Meru in the Prim ordial Sacred Abode (adyapifha).246 The
KRU explains that the Path of Meru means, according to Kula doctrine,
the Tradition (santana). The Lord of the Tradition is the Great M eru, the
Supreme Sky of Consciousness which knows its own nature completely. It
is from here that the beginningless sequence of the progressive unfolding of
consciousness and the transmission of the doctrine originates and so is
called the Prim ordial Abode (adipiiha) located in the Centre between
Kailasa and Malaya. The Lord resides here with his power that pervades
all things.247
Although it is not possible on the basis of the evidence so far available
to say anything definite about where the Kubjika cult originated, if we
assume that the Tantras of these schools invest with symbolic meaning the
environment and localities in which the Kubjika cult originally developed,
it seems likely that we would be right to seek its origins somewhere in the
western Himalayas. That this cult was known in the mountains of the
North of India during the earliest period of its development seems to find
support by a reference in the KM T to birch bark as the material on which
a M antra is to be w ritten.248 The MBT also refers to it as a writing
material249 and the GS prescribes that a sacred diagram be drawn on it
with sexual fluids mixed with poison.250
The study of the cult o f Kubjika and the Tantras of the Western Kaula
Tradition is barely in its infancy. We hope that some of the more
im portant of these texts will be edited and studied in the near future.
There can be no doubt that this is an early Kaula tradition which is of
great interest not only for the richness of its doctrines and the beauty of
its symbolism but also because it can tell us something im portant about
a part, at least, of the history of Hindu Tantra.
P art T hree
Appendices
A p p e n d ix A
The Manthanabhairavatantra
cannot be older than this nor is it younger than 1180 A.D. which is the date
of the oldest Nepalese manuscript.
To conclude, we shall say a few words about the traditional accounts
of the revelation of the MBT. Up to now, I have managed to trace two.
One is found in the Yogakharnja10 of the MBT, and the other in the
concluding portion of the Kumarikakha#<ja.[i According to the latter
account, the original form of thisT antra (here called “Paramesvaramata ,r)
was a fabulous million, million verses long (lak$ako{i). The essence of the
doctrine of this T antra was transm itted in another seventy million verses
from which another version was supposed to have been derived consisting
of one hundred and twenty-five thousand verses. The most essential
doctrines of this T antra were then recorded in a twenty-four thousand
verse version which is the present MBT. This was transmitted by Kujesi
to M itranatha in three sections.
The Yogakhancld's account is more complex than that of the
Kumarikakhancia and differs from it in certain details. Here the Divine
Transmission of the Khahjinimata is said to have passed originally from
Adinatha to M itranatha, who then transm itted it to $a§tinatha, who gave
one half to C aryanatha. C aryanatha in his turn passed on half of what he
had learned. Unfortunately, a break occurs in the text here so we do not
know whom C aryanatha taught; possibly this was Oijiyapanatha who is
usually associated with the three other teachers. Whoever this teacher was,
he transmitted half to Brhatkantha, who gave one half to Ga^avek§a. The
remaining portion of the Khanjinlmata was transm itted by Akulanatha.
The text now goes on to explain that the version consisting of one hundred
twenty-five thousand verses was brought to earth by Cinca(natha) and that
all the rest of the teaching was transm itted by the community of Siddhas on
the Island Free of Thought-Constructs (nirvikalpadvipa) from whence it
spread to India. We are also told that the Tantra of twenty-four thousand
verses was brought to earth by Kalartkanatha, These two accounts are
illustrated graphically in Table 8.
Adinatha
M itranatha
$a§tinatha - Caryanatha - ? - Brhatkaijtha -
Gapavek§a - A kulanatha
C incanatha (also called Cincinmatha) revealed
the 125,000 verse version.
K alaiikanatha revealed the 24,000 verse version.
A pp e n d ix C
Once Bhairava has enum erated these categories, he tells the goddess
that this progressive development from less to more specialized knowledge
through these four stages is inevitable because it is impossible to teach
everything at once. Even so, however broad or focalized the terms of
reference may be, all hum an and divinely revealed knowledge refers to the
same reality. Bhairava also implies here that there is a temporal as well as
a logical progression between these categories and, indeed, the last three
do follow each other in roughly chronological order. Thus these are:
102 A p p e n d ix C
which are the sources of the five currents {srotas) of Saiva scripture.9 In
his lower (apara) form, SadaSiva is the lord of the M antrapifha a.nd is here
said to have four faces. Now, although the JY itself belongs to the
Vidyapifha, it also considers the M antrapifha to be very im portant and
deals with this class extensively. Accordingly, the M antrapifha is said to
be the divine p ifh a (divyapifha) of the M udra- and Man^ala- pifhas of
which Sadasiva is also the Lord. From Sadasiva’s five faces issues the pure
knowledge (suddhavijhana) through which the divine Sound (ndda) and
Drop (bindu) emerge. When these combine with the metres, they generate
through their rhythm all the ¿dstras. The gods first revealed reality (artha)
through the Vedic metres (chandas); even so all the scriptures are forms in
which Sadasiva embodies himself. Consciousness has no form in itself but
is said in this way to possess a body made of M antras. As such, Sadasiva
is full of divine powers and has a form (vigraha). Through the four
instruments (kara$a) of mind, ego, intellect and speech, he assumes the
identity o f the teacher who instructs; while when the fourth instrum ent—
speech—is substituted by the power of hearing, he becomes the disciple
who listens. He who knows the four instrum ents is said to be a “M aha-
karunika”.10 W ithin these gross bodies is the subtle body (ativahikalihga)
governed by the vital breath which moves through the right and left
currents of vitality. In this way the right and left currents of scripture
descend into the world along with a ‘m ixed’ current formed by their
conjunction to which the Yamalas belong.
The JY deals with Agamic Saivism from two points of view. Firstly
it enumerates the Tantras which belong to the pifhas; then it enumerates
the lines of transmission of the scriptures from master to disciple (santana)
and the schools (m afhika) belonging to the currents (.srotas) of the
scriptures.11 Thus the JY seeks to present a picture o f Agamic Saivism and
its many different branches from the point of view of both systems of
classification. It reserves, however, pride of place for the p ifh a system as
the most fundam ental classification of scripture while the srotas system is,
in a sense, devolved to a classification of groups of Saiva traditions and
lines of transmission of doctrines, each peculiar to its own parent current.
The ideal Saiva master (here the term “g u ru "is generally used rather than
“dcdrya,r) is one who is conversant with the teaching of all the currents.12
The relationship between the p ifh a and srotas classifications is
peculiar and somewhat strained. It appears, as we have already had
Canon o f Jayadrathayamala 105
occasion to rem ark, that it is a system which developed after the classifi
cation into five currents, at a time when they were reduced to three (see
above p. 42 ff), namely into ‘right’, ‘left’ and ‘m ixed’ currents and takes
over from it as a system of classification of scriptures as a whole (and as a
set of groups to which the T antras affiliated themselves), cutting across the
distinctions made within it. We have already noted that this is happening
in the BY’s account of the Saiva canon while the following rem ark in the
J Y seems to echo the view of the BY which, as we shall see, is considered by
the JY to be a particularly im portant T antra: “The fo u rpïfhas are together
said to reside in the right, left and mixed currents as if they were the
principal branch in a subsidiary branch (of the teachings).“ 13
Before dealing with the p ïfhas extensively, Bhairava talks about them
in general and deals with their contents summarially. Bhairava starts by
listing a num ber of synonyms for the word "p ïfh a ”, all of which mean a
“group” or “collection”. 14 The M antrapïfha is associated with masculine
words while the Vidyâpïfha is associated with feminine ones. Perhaps by
this is implied that the Tantras of the M antrapïfha deal more extensively
with male deities while those of the Vidyâpïfha with female ones, as the
Svacchandabhairavatantra and the SYM, which belong to these two
pïfhas respectively, exemplify. The M udrâpïfha is of three forms related
to mind, speech and body. It is called “terrible” (ghora). The Ma#<$ala~
pïfha is where all three unite and is said to be undivided. This does not
mean that the Tantras of this p ïfh a are monistic but rather that it has,
according to the JY ’s account, no subdivisions. Again, the preference the
JY has for the M antrapïfha is evidenced by the way it characterizes it as
the repository of the supreme brilliance {paramarfi tejas) which symbolizes
the state of the enjoying subject (b h o k tf), while the Vidyâpïfha is the object
of his enjoyment (bhogya). M udrâ includes both of these while M anda la
is present in them a ll.15
To the M antrapïfha (“beyond which there is nothing”) are said to
belong twelve thousand M antras. It consists essentially of eight Bhairava-
tantras, namely, the Svacchanda, Unmat ta, Asitânga, Ruru, Canfa,
Krodha, Jhankâra and Kapâiïsalb with which secondary Tantras are
associated. The JY appears to quote a verse from the Sarvavïratantra
recorded by Kçemaràja in his com m entary on the SvT which states that the
Tantras of this p ïfh a are four, namely, the Svacchanda, Canfa, Krodha
and Unmat ta- bhairavat ant rasJ7 The apparent contradiction is recon
ciled if we understand that what is m eant here is that these are the
four major Tantras of the M antrapïfha, each of which is linked with one
of the four remaining Bhairavatantras. Consequently the M antrapïfha is
said to be tw o-fold.18
The Vidyâpïfha is also extensive.19 The m ajor Tantras listed as
106 A p p e n d ix C
2. Mantraplfha
This is basically divided into the Tantras associated with four root 169b
Tantras or m ulasutras, namely, Svacchanda-, Canda-, K r o d h a and
Vnmatta-bhairavasuiras.
ramifications are but a drop from the infinite ocean of SadaSiva. The
Unmattapaddhati is divided into two lines, each of which is of eight
divisions corresponding to eight Bhairavas. U nm attabhairava taught the
Unmattatantra to which belong twenty-one Saiphitas, while Kapallsa
presides over the second line of which there are eight principal Saiphitas,
namely: Anantabhairava, Anantavijaya, Bhairava, Para, VinaSikha,
Acaloddisa, M artandasamhitd and Canfograbhairava. These Saiphitas
were brought to earth by eight sages am ongst which only ViSalalocana,
Tarapati and Sutara are named, the latter being the revealer of the
Martandasarfihita. These Saiphitas are linked to each other and to other
Tantras as follows:
3. The Vidyaptyha
The Tantras belonging to the R ight C urrent These are basically the
Yâmalas of which five are the root texts. Of these, four are in a group of
174a their own, namely Raudra-, Aum a-, Vaiçnava- and Skanda-yàmalas.
The fifth is the Brahmayâmala which is said to expound “the conduct
which involves the use of rotten flesh11(kravyacara) whereas the others do
not. It is also called “P icum ata,” “D vayakfara” or ju st “M ata. ” Seven
other root T antras (sütra) are associated with it, namely Ucchufma,
Nirâcâra, M ahâcârat Sarvatobhadra, 28 D vika, Sarvâtm aka and
Mahâdakçina. The last T antra is divided into two sections. Thus there
are eight Tantras altogether and as these are all M atatantras they are
called the “Eight M atas11 (matd?faka).
After concluding that there are fifteen root T antras (mülasütra) in
the Vidyapifha, the JY moves on to a more detailed treatm ent of each
of them.
transmitted it to men.
The basic form of this T antra (mulasutra) is said to be of three
hundred and fifty (chapters?), its concise (sarpgraha) version is called
“sahasra" which presumably means that it is one thousand verses long.
Other versions consist of sixteen thousand, twelve thousand and twenty-
four thousand verses. Altogether this Tantra is said to be over one
hundred thousand verses long and has been heard, spoken and transmitted
many times.
Teacher Tantra
M&ladhara M dlini (?)
Karalin _ (?)
A jakarpa Khecarlvijaya
Saiikhadharin M ahaiahkhatantra
M arm aphetkara Phetkaraikak$ara
Vidyadhipati Vidydlaksa
Vidyolka Vidyaprasuti
Aghora Sit aghora
G horeivara Raktaghorefvari
Vidya *■K a n a k a
Kl|acak$us Kj-ynaghoresvari
B hairava’s will Pitaghori
Padm am udra M aham aya (?)
N iranjana Bhadra
Iccha Kali
Jnatr Ka rankini
M antra. Nddicakra (also called Sutara) and the Guhyacakra dea\ with the
nature of Sakti. There are two Kalacakratantras; both are said to deal with
the duration of m an ’s life (ayus). One of the two is Saura and said to be
auspicious (subha); the other is Buddhist and is considered to be
inauspicious (asubha). Vdhneyacakra is said to be ‘enflamed’ (dipta) by
M antra while the teachings of Somajacakra (also called " Candatantra")
destroy death.
The Pahcamftatanira. The only inform ation supplied about this root
Tantra is the line of its transmission which is: Sakti —yoginis, ganas, gods
and snakes -* Prahlada — Ausanasena — Bhargava — Devala —
Kr?iiatreya (“who knew the Vedas”) — K usadhvaja — Sitoda —
Patadra (?) — Dantya.
The Visvadyatantra. This Tantra is named " Viivddya"because it teaches
the nature of the group o f yoginis (yoginijdla) starting with the one named
175b Visva. The line of transmission is as follows: Sada&va — Visvatman —
V im aiadda — K anticchatra — Saundin — Bhairava — Sakti who hid it. A
few of the M antras of this T antra were known to Kilaheli who learned
them from Sakti and transm itted them to the world of men.
Yoginijala&ambara. There are twelve Tantras associated with this root
Tantra: Mahasamvarttaka, Bhima, Tilaka, Nakha, Bimba, Candralekha,
Candasitaka, Sildda, Bhagamala, Bhogini, Sukesin and Sudhama™
Associated with these Tantras are twenty Upasamhitas: Tar aka,
176a Ak$i, Pdsaghni, Nandini, Gamint, Bhfguni, Satya, Ddk$dyani, Umd,
Maya, Mahakali, Candali, Acalasri, Bhadrakali, Sumedha, Tara,
Arkamani, Tdrakabharana, Raudrl and Jvalamdlantika,
The line o f transm ission o f the Yoginijalasambara is as follows:
M ahasam vartanatha — Bhima — Tailaka — D irghanakhin — Bimba —
Candavrddha — Canda&lataka — Silada — Bhagamalin — Bhaginl —
Vira — Sukesin — Sudham a — Jvalam ukhantika — Bhairava — Sakti —
Urdhvaretas — A nanta.
The Yoginijalasambaratantra is said to the glory of the Anantasrotas.
Teacher Tantra
Vidyesana Siddhartha
Pralam ba Vidydlayd
P adadru Vidyarajhi
Kara la Vidyamani
A jatm an Vidyaraii
Canon o f Jayadrathayämala 113
Vidyesäna Vidyaprasuti
A m saphala T ridanfi
Six of these Samhitäs have not yet been revealed.31 Bhairava will give 176b
them to the world in the last age (yuga).
The Siraichedatantra. This T antra is taught by Bhairava himself in the
last aeon. It is the last of the seven root Tantras of this Current to come to
earth and also belongs equally to the Right and Left Currents. It is
expressly said to be a Yämalatantra and is also called “Jayadratha"or
"§a}$dhasra. ” It was transm itted from Mlna(?) — Narasimha —*Yogini —*
Daivavatsyaikacak?us — Sukra. A num ber of Tantras are associated
with it:
Teacher Tantra
Vajrakäya Vajramäiinisamhitä
Jvaiantaksa Jvälatantra
Yam äntika Yamäntakatantra
K älänta Kälän takatantra
Plavana Plavanatantra
Candavega Candavegasamhitä
Nidhlsa Täränidhitanira
378a The Nayottaratantra. The sages fulcra and Angiras received this Tantra
and brought it down into the world, transm itting it to Candrasekhara. A
number of related Tantras are listed: Trailokyamohana, Bimba,
D anavdrivim ohana, Tdrdkdbhyudaya, Sam grdm avijaya, Nayasara,
Tilaka, Koldhala, Amburasi, Capamali, Mrgadhvaja, Rdr ay ana and
Vasafkara.
The Maharaudratantra. The deity of this Tantra is called M aharaudra or
Ruru (bhairava). He taught this Tantra to Sarva — R udra — Samvarta —
Canda and Saundin A rdhanarisa who taught a short version to Bhrnga.
182b The Yamalas. The term “yam ala” which literally means “a couple in
union” and commonly refers to the coupling of the god with his consort,
is here given a special meaning as the union of M antra and Vidya. The
implication here is that these works deal with both the ritual formulas
corresponding to and embodying the gods (as M antras) as well as the
goddesses (as Vidyas),35 Again these operate in the dom ain of another
couple, namely, knowledge ijndna}, which is both insight into ultimate
principles and an ,understanding of the methods described in the Tantras,
together with action (kriya) which is the application of this knowledge.
They depend on one another.36 The goal of spiritual endeavour is achieved
by following the teachings of the Yamalas, dealing as they do with these
four topics.
179b The Brahmaydmala. The root of all the Yamalas is considered to be the
Brahmayamala. it has three principal divisions called Raurava, Andhaka
and Kanaka. The Visnu, Skanda and Rudraydmalas belong to these three
respectively. The four Vedas are said to originate from them along with
Canon o f Jayadrathayamala 115
Associated with the eight anutantras are eightparifisfatantras. How these }goa
are linked is not always made clear. They are Pancdlika (linked with the
Utpalatantra), Mdnava (associated with the Vagavalitantra), Kardlimata
and its isdrasamgraha> , Laksm i (associated with the Kalakhya), Vimala
(associated with the Lampafatantra), Kacchapi (associated with the
Kampakuhadrutantra) and Garbhaprakarana (associated with the
Nayo day at an tra).
From the parisistatantras originate the upasutras. These are
Asvapluta, Sarahga, Gojika (?), Bhedavipluta, Vaibhahga, Mdtanga,
Kusumali and Savitraka.
All these Tantras are the root ancilliaries (mulapari&ista) of the BY.
Along with them are seven m isra su tra s which are said to belong to all five
currents of scripture. These are A iu sfi, K usa. L ava, K alasdra, A m b ik a ,
Tilaka and A vadhuta,
116 A p p e n d ix C
Mülasütra: This is the ro o t Tantra. There are different root Tantras for
each p ifh a * ' Each one is supposed to deal extensively with the teachings
of a given tradition as a whole. It is the text which indicates (sücaka) the
entire extent of the doctrines and ritual procedures of its tradition.
Gühyasüíra: The T antras of this type discuss the hidden esoteric meaning
o f the teachings,
Nayasütra: These Tantras deal with how the adept is likely to fall from
the path and lose the attainm ents he has already acquired (siddhi) and
how he can get them back,
Validation (gam aka}: The teacher now gradually unfolds the meaning of
the sütra. In order to m ake it clear, he makes use of examples and refutes
possible objections.
Clarification (prakasaka): This follows from the previous correlate.
Canon o f Jayadrathay amala 117
Other Agamas also deal with these correlations in their own way.
Abhinava devotes a section of his Tantraloka to an explanation of how
the teacher should transm it the meaning of the scriptures to his disciples
(vyakhyavidhi).42 Abhinava bases himself mostly on the Devydyamala
which Jayaratha accordingly quotes extensively in his commentary.
Paraphrasing this T antra, Abhinava starts by saying that there are five
kinds of teachers:
Those who know ju st on e Kalpa, that is, a version of a given root T antra
or a T antra associated with it.
Those who know the entire range of Kalpas associated with a given
T antra.
Those who know all scriptures belonging to a given class (sastra).
The best teacher is, of course, one that belongs to the fifth group.
However, if a student cannot find a teacher who knows the scriptures in
their entirety, he should seek a teacher who is well conversant with the
particular T antra he wishes to learn and whose teachings he seeks to put
into practice.
The Devydyamala, like the JY, says that there are five correlates
through which the teacher can explain the meaning of a Tantra with
reasoned argum ent in order to preserve the doctrines of his own tradition
(svamndya). These are:
The M eaning o f the Words (padartha): The teacher now explains each
statem ent in the Tantra individually.
The Meaning o f the Sütra: The teacher can now deal with the entire root
T antra. He reflects on its overall content, distinguishing it from other
sütras, in order to elucidate its unique character in relation to other
sü tra s44
4. The Mudrapi|ha
measured (m apita) and saved (tranita) and of whom there are many
diverse forms (bheda).46 There are five Sam hitas associated with the
M dtrbheda, namely Apardjita, Vastuvidyd, Sarvabimba, Ka rot ini,
Siddhasdrartha, Citrartha (?).47
i 84a The Kubjikamata. The K M T is said to have originated in the m ountains.48
It is considered to be the most im portant T antra of the MudrdpUha
although it is linked with all the pifhas. This is the T antra of the goddess
Kubjika. She is the power Beyond Mind (;m anonm ani) who, in the bliss of
loving union (samplava), withdraws her limbs to form the divine triangle
of the Seed (blja), Drop (bindu) and Sound (nada) through which the
universe is created. The KM T is said to deal with the Seal of Speech
(vdcika mudra). This is of countless form s although it consists essentially
of the fifty letters of the alphabet. It denotes the meaning of countless
words and indicates the true nature of things leading thus to its realization.
The sections (or associated texts) of this T antra are Ekak$ara, Dvydk$ara,
Padamala, Sab dam aid, Sabdamalartha, Mdlini, Vakyamdld and
Vidyamdld.
5. Man^alapltha
This p itha is said to be equally present (sdm dnya) in the other three
pithas. A lthough it has only one m ulasutra, which is not named here, it is
divided in many parts (kalpa). The line of transmission is as follows:
Pancasekhara who brought it to earth transm itted it to Ratisekhara —
Sam harabhairava — Dip^i — Dantfesvara — Ghoreya ™* Gomati -*
Trikala — Srlkantha — Bhairava — Brahm a who heard it in each pitha.
After the goddess has been told about the contents of the four pithas
associated with Bhairava, she then wants to know about the sixty-four
Bhairavatantras said to constitute the Wheel governed by Sadasiva. These
Tantras are listed below along with the names of their teachers who are
considered to be incarnations of Bhairava. The ordering of the eight
groups is just as we find it in the S K S (seeabovep.45ff.). M oreover, more
than half of the Tantras listed in the SKS are the same as those listed here.
We have therefore recorded their names in a separate column as shown in
Table 9.
Tabie 9. The Sixty-Four Bhairavatantras.
with the founder of the monastic centre from which it was propagated.
The JY, as we have said, accepts the basic Five C urrent division of the
Saivagamas. However, although it does classify them in these terms, it
does not concern itself to do so in detail. The Siva and Rudra division of
the Sid dh ant agamas are said, as usual, to consist of ten and eighteen
Agamas respectively. The Left Current consists of the Tantras which
‘delude’55 which is clearly an allusion to the Satrimohanatantra that is
regularly considered to characterize the scriptures of this Current. The
C urrent of the Right contains, amongst others, the sixty-four Bhairava-
tantras and the Yamalas. To this Current also belong the SvT and the
entire M antra and M udraptfha ,56 The Garutfa and Bhiitatantras are
characterized in the usual way as concerned with remedies for poison and
with exorcism respectively,57 They are ascribed, as usual, to the Eastern
and Western C urrents.58 The Tantras of the Left Current are said to deal
with the acquisition of the eight yogic powers (siddhi) and power to control
others,59 while the Right Current is the eternal non-dual abode of all the
pleasures of this world and every type of liberation.60
The Sixth Current is above the others. It belongs particularly to
Siva’s power through which all the traditions and their teachings are
revealed in this world.61 Siva is the root source of all the Agamas; infinite
and full of the knowledge of consciousness he is both omniform and
tranquil like a waveless sea. His power is his vitality (virya) whose field is
the delight of the abode of consciousness and is beautiful with its blissful
pulsation generated through the diversity of its immanently manifest and
transcendentally unmanifest form s,62 This vitality is Speech at all levels63
and pours out of universal consciousness in countless extensions (prasara)
of its power of which the six Currents of scripture are the main stream s.64
Full of this divine power at its very source the Sixth Current leads to
realization in an instant with great force (ha(hat) and so the form of Siva
that governs it is Ha(hakesvara. He is said to know the mysteries of Kula
and Kaula and so all the secret teachings originate from here. Conse
quently this Current is present in those of the Left and Right.65 Moreover,
although it flows in the upper regions (urdhva) it is also connected with the
lower ones as well, that is, with those frequently assigned, as we have seen,
to the Kulagama. Thus Ha^hakesvara’s dom ain is said to be “the abode of
the fire of time” which burns in the hells below the world-order. The SvT
confirms that H a(aka is the name of the Lord of Sauvarna which is the
eighth and lowest hell66 while the JY tells us that all the lords of the hells
{patalanayaka) are bom from HathakeSvara’s m outh.67 According to a
passage quoted by Abhinavagupta from the Raurdgama,6B when
Srikantha resides in the hell worlds he assumes the form of H a(aka because
he bestows yogic powers suddenly and with great force (hafhatah) while
Canon o f Jayadrathayamala 125
the JY expressly states that the Upper C urrent “bestows all yogic powers.”
The Anandadhikaratantra69 declares that H ataka is the lord of the hell
where souls burn in the “fire of tim e.” This place is destined for those
initiates who have broken their pledges and, more especially, for those
who, although votaries of the Left-hand Path, censor it. Here, according
to this Tantra, go those who have been taught higher doctrine but have,
nonetheless, resorted to lower teachings or M antras of a lower order such
as those practised by the devotees of G aruda. They can, however, free
themselves of their suffering by offering themselves to Lord Hafaka and
so reach the higher divine principles and then, from there, ultimately
merge into Siva,70
8. Conclusion
Accounts such as these not only teach us a great deal about the extent
of the Agamic corpus but also about its history and, consequently, about
the history of the development of Agamic Saivism. We should, however,
treat these accounts with caution. As we have seen, all the major Agamas
have their own peculiar view of the Saiva canon and, although their
accounts are in many ways compatible and even supplement each other,
none of them are either complete or unbiased. M oreover, they give rise to
many questions which still remain unsettled. Thus, we cannot help asking
ourselves whether all these Tantras did, in fact, exist. For example,
although the eight Bhairava T antras are generally considered to form a
standard group, it is by no means certain whether they all existed.
Certainly the Svacchandatantra, at least, does and possibly a number of
others but it is hard to believe that all the Tantras said here to have been
associated with them did as well. Bhairavas are commonly worshipped in
groups of eight while the group which has given its name to these Tantras is
particularly well known. It is hard not to suspect that these lists are not
entirely genuine when we notice that groups of eight are such recurrent
features in them. May it not be the case that some of these titles are purely
fictitious additions to fill out ideal schemes? We can only be sure that a
Tantra existed when we find additional evidence to corroborate the
statements made in accounts such as these. We must, for example, find
direct quotations from it, or reference to it in independent sources such as
inscriptions. Best of all, of course, would be to find the text itself or
fragments of it.
Again, how genuine are the names listed in the lines of transmission
of these Tantras? Certainly the gods, divine beings and Upani§adic sages
who figure in these accounts cannot be considered to be historical figures
126 A p p e n d ix C
connected with them. Even so, it is not impossible that the last few
members of these lines did, in fact, live at some time and propagated these
Tantras; perhaps, indeed, they even wrote them. It is worth noting,
however, that the account of the line of transmission given here of the
SYM and its own account as presented by Abhinavagupta in the
Tantraloka do not agree. This fact should put us on our guard to deal
cautiously with what these account tell us because our knowledge of the
extent and content of even the extant Tantras is still very superficial. Even
so, there can be little doubt that future research will reveal a vast field of
inquiry and that these accounts of the Saivagamic canon will serve as
im portant guidelines by which we may orient ourselves in the course of
work to help us to locate ourselves in this vast body of literature.
A ppen d ix D
of this m onograph.
19. TA, 29/60.
20. YHr, 3/37.
21. varapasyam sm asanantu eiat sarvam sam alikhet |
BY, NA, MS no. 3/370 fl. 6b.
22. See K. N. Sukul Varanasi Vaibhava Bihar Ra$trabha§a Pari$ad, Patna
1977, p. 406.
23. T hat the Pascimamnaya was known in K ashm ir is established by a
num ber of references. Thus A bhinavagupta quotes from the Kubjikam ata'm PTv,
p. 184, U nfortunately, this reference has not been traced in the 3500 verse recension
edited, but as yet unpublished, by T. G oudriaan and J. A. Schoterm an. Jayaratha
quotes an unknow n Agama which refers to the Thohakasam ata of the
Pascimamnaya:
A nother reference is found in the following passage which Jay arath a quotes
w ithout naming his source: “The triangle is called the female organ (bhaga), the
secret circle in the Sky. Its corners are will, knowledge and action and in its centre is
the sequence (kram a) of Cincinl.1’ (TA, vol. II, p. 104).
The worship of the absolute viewed as the source of cosmic manifestation
symbolized by the female organ from whence arises the pow er of Kuntfalini is an
im portant feature o f the T antras of the Pascimamnaya. Kubjika, the Supreme
Goddess, is frequently called Cincini (see below p. 90). In a hym n to Bhaga in the
Cincinimatasarasamuccaya we find the same verse. Thus, the first line of
Ja y a ra th a ’s quote reads:
[This is] the wine which im parts boldness to the w ords of lovers and frees
them from fear while m aking love; the wine in which all the deities (of
consciousness] who reside in the shade of the tree of the Kali age and
practice mystic union (cakracarya) gladly abide; the wine which here
[in Kashmir] bestows first pleasure then liberation. (TA, 37/44).
But not all thought so highly of these practices. We know th at bans were
140 N o tes
imposed and practising Kaulas ostracized. Some, such as the satirist K$emendra,
who lived in Kashmir at the time, considered Kaula ritual and practice to be merely
a source of moral corruption and an excuse for licence. See my Doctrine o f
Vibration, vol. I SUNY Press, A lbany, 1987, chapter 1.
26. For definitions of the word “T an tra” see Tantra in Bengal: A Stu d y in Its
Origins, D evelopm ent and Influence by S. C. Banerji, Naya Prakash, Calcutta
1978, pp. 1-3.
27. Renou rem arks that the later Tantric texts W oodroffe studied that
“describe the worship o f the great G oddess” are called “T an tras” rather than
“Agam as” {preface to SSP, vol. I, p. 1). Brunner notes, however, th at in the early
Siddhantagam ic context the terms “A gam a” and “T an tra” are synonymous (ibid.,
intro, p. iv.). Thus we find the following definition of T antra, which has been
repeated frequently throughout the centuries in a wide variety of Tantric scriptures,
in the Kamikdgam a also. This must be one o f the earliest citations o f this
definition:
36. “There is no other scripture ap art from the Veda which explains Dharm a.
Brahmins should not converse with those who find delight elsewhere (in other
scriptures).” (Ibid. 1/11/271; also ibid. 1/50/23-4).
37. K urma 1/15/112-115; 2/16/15-6; 2/21/34-5; 2/37/144-6.
38. Varaha 71/58.
39. Ibid. 70/41; K urm a 1/15/104.
40. Devibhagavata 7/39/26-31 quoted by H azra p. 226. Cf. also ibid,
9/1/21-32.
41. M adhava’s ¿ahkaradigvijaya with D hanapatisuri’s pindima com
mentary. A nandasram a Sanskrit Series no. 22, AnandaSrama Press, Poona,
1915, 15/1-7.
42. RT, 6/108.
43. T ranslation by E. C. Sachau in AlberunVs India 1, 22.
44. Dasavataracarita quoted by H azra p. 89. Translation mine.
45. R ohana A. Dunuwila Saiva Siddhanta Theology, M otilal Banarsidass,
Delhi, 1985, p. 44.
46. For this reference see A ppendix B, fn. 21.
47. Sa(SS, 3 /79b.
48. Thus the Rsis in the Kamikdgama were, “intent on Siva and Sakti
desirous of understanding the Suprem e Knowledge.” They praise Siva saying
of Him:
[You] are of the nature of the flow of innate bliss whose characteristic
is one’s own consciousness, [You Who are] united with the Supreme
Power whose nature is supreme consciousness.
(P.K a., l/7a-8b).
49. There are a num ber of later T antras whose names are the same as Tantras
of the older group; examples o f these are the Kulacudamaifi, Kuldrpava,
Tantraraja, G uptatantra, Jayadrathayamala, Sarpmohanatantra and a medieval
work not at all associated with the Paicimam ndya called Kubjikatantra. Schrader
writes: “T hat occasionally the same name has been given to two or even more
different works is nothing unusual in Agamic literature. F or instance, am ong the
Sakta Tantras there a r e , . . . three Prapanca tantras, two H arigaurl tantras, three
Kubjika tantras, two Yoginl tantras and two M fdani (?) tan tras.” Introduction
to the Pahcaratra and A hirbudhnya Sarphita by F. Otto Schrader, Adyar Library
Series, Adyar, 2nd edition 1973, pp. 14-5.
In the later period Saktatantras are distinguished from Saivatantras and
called “Nigam a” and “A gam a" respectively. The characterizing distinction
between them is stated to be that the form er are spoken by the goddess and the
latter by the god. Originally the T antras were generally spoken by the male
142 N otes
partner to his com panion, however much they may have been concerned with
female divinities and their rituals. Thus the NSA, which belongs to the early
period but was in the later period considered (quite justifiably in terms of its
content) to be an exem plary Saktatantra, is taught by U vara to the goddess.
The goddess herself rem arks that the god had taught all the preceding Tantras
including those which the goddess defines as the ‘T a n tra s of the M others” although
the ones th at are listed are of many types including Siddhantagam as. The passage
reads:
The matrtantra figure as a separate group along with the Yamalas in the
Siddhantagam as (see P .K a., 1/122) while a reference from an unknow n source in
Jay arath a’s com m entary on the TA shows th at a distinction was made between the
Sankaratantra and the Devitantra (see TA, vol. X lb, p, 12). Even so, we do not find
that a clearcut distinction exists between Sakta and Saiva T an tra in the early
sources, even in the Kulatantras which were in practice Sakta in most respects.
By the middle of the ninth century, when S om ananda w rote the $ivacir$tk the
Saktas figure as a distinct group (see ibid., chapter 3) but in this case they could
arguably be said to be connected to the tradition only indirectly as its commen
tators and exegetes. The Siddhantins during this period also applied this
distinction. N arayanabhatta does so in his com m entary on the M rgendragama,
although it is not found in the Agama itself. Thus he says:
saivanam eva desikadinam iha laksanam j saktadayastu te anyadrsa
eveti vijnapayitum saivapadopadanam |
(M pT . (caryapada), 1/2).
wool. The thread indicates that the wearer is an initiate, but not necessarily a
brahmin,
142. Sacrificial Death and the Necrophagous Ascetic J. Parry, pp. 88-9.
143. Gorak$asiddhantasamgraha edited by Janardana Pandeya, Sarasvatl-
bhavanagranthamala, no. 110, Benares 1973, pp. 14-15.
144. One colophon in a manuscript of the Sabaratantra calls it the Goraksa-
siddhisopana. (Asiatic Society of Bengal MS. no. 8355). Another manuscript of
the same work ascribes it directly to Gorakhanatha (Asiatic Society of Bengal
MS no. 10542). Kasinatha Bhatta (eighteenth century) tries, in his Kdpalika-
matavyavastha, to show that there is a historical link between the Sabaratantra
and the Kapalikas. See Hindu Tantric and Sakta Literature by Teun Goudriaan
and Sanyukta Gupta, A History o f Indian Literature, edited by J. Gonda, vol. II,
fasc, 2, Otto Harrasowitz, Wiesbaden, 1981, p. 121. See also ibid., pp. 120-22 for
Goudriaan’s discussion of the Sabarataniras which he classifies amongst the
Tantras dealing with magic. For lists of the eighty-four Siddhas and nine Nathas
see S. B. Dasgupta, pp. 202-10,
145. S. B. Dasgupta, pp. 208, 377, 383 and 391.
146. The Prakrit drama, the Karpuramahjari by Rajaiekhara (ed. Sten
Konow and translated by C, R. Lanman, Harvard Oriental Series vol. IV,
Cambridge, Mass.: Harvard University, 1901)written in the tenth century features
a master magician named Bhalravananda who “follows the Kula path” (act I,
v. 22-5; IV v. 19). The author of the Rucikarafika on Kr$pami6ra’s Prabodha-
candrodaya mistakes Rajasekhara’s, Bhairavananda, for a Kapalika.
147. Vamana, 2/17, 4/1.
148. Sahkaradigvijaya, 15/12-14. The compound sa + uma - soma, implies
the Kapalika’s creed, i.e., Somasiddhanta, which affirms that the liberated
condition is analogous to the bliss (dnanda) experienced in sexual union which
reflects that of the union of Siva with his-consort Uma. The word ".soma"also
means moon and so the Kapalikas have accordingly been associated with the
moon. See Bhattotpaia’s commentary on Varahamihira’s Brhajjataka 15/1.
149. Sahkaradigvijaya, 15/24-5.
150. Ibid., 15/1-7. See also above p. 29.
151. Lorenzen, p. 20.
152. Ibid., p. 21.
153. Ibid.
154. See above p. 6.
155. See S. N. Ghosal Sastri’s Elements o f Indian Aesthetics, vol. II, part IV,
Chaukhamba Oriental Series, Benares 1983, p. 19 ff. Sastri here briefly outlines
some phases in this ritual as it is found in the BY, fragments of which he has studied
in manuscript along with the Pihgalamata.
156. TA, 27/20b-9. Abhinava’s sources are the BY and the Siddha-
yogesvarimata.
157. Lorenzen, p. 74 ff. demonstrates that the Kapalika’s vow corresponds
to the penance prescribed in Hindu law for killing a brahmin. He makes a mistake,
however, when he distinguishes a ''brahmahan'’from a “bhrutiahan"as one who
has killed an ordinary brahmin rather than a learned brahmin. “Bhrunahan"
148 N otes
actually means “one who has killed a fetus” that is, a person who has induced or
consented to an abortion, (See Panini’s Asfadhydyi, 3/2/87),
158. The Jayadraihayamala (NA, MS no. 5/4650, fl. 186a) says:
suddham samsuddha(rp) buddhasca tatha p5£upatavratam |
nagnavratarp mundadhararp parivrajvratam uttam am ||
Ksemaraja comments:
are a num ber of cases where we notice that a text travels from one group to another
or, to put it another way, is appropriated by other groups.
224. “bhairavâgamesu dakçinasrotah samuttheçu svacchandacandatriSiro-
bhairavadi$u bheditam j ’* (NTu, vol. I, p. 225). These three Tantras are No. 1, 2
and 24 in the list of the Daksinatantras. Also SvT, vol. I, p. 8.
225. Abhinavagupta exalts the Bhairavatantras which he considered to be
the highest of the five currents of scripture:
228. In certain MSs (e.g., NA, 1/285) the colophons describe this tantra as:
sarvasrotasam grahasarah [ i.e., “the essence of the summary of all the currents
of scripture.”
229. NT, 9/11.
230. Thus the goddess says:
234. It is certainly not at all clear whether we can, in fact, make such clear-cut
distinctions, although generally speaking the Stddhantagamas do tend to be
dualistic, unlike the Bhairavatantras which are predominantly monist. Even so,
it is certainly not possible to distinguish the Sivagamas from the Rudragamas on
the basis that the former are dualist while the latter conceive reality to be a unity-
in-difference.
235. Abhinavagupta goes further by making all the Saivagama ultimately a
part of Trika. The god who utters the scripture is not the five-headed Sadaiiva
of the Siddhanta. The Supreme Lord who spoke the three types of scripture (of the
Left, Right and Siddhanta) is in fact three-headed. He is Trisirobhairava, the
Trika form of Siva:
236. This state of affairs is peculiar to the Srikanthi which does anyway, in a
sense, retain the Vamatantras as a part of the same standard, older division into
sroias although it is relegated to the lower level of a secondary, subsidiary class
ification. Other Trikatantras such as the Bhargaiikha and Nisisamcara sustain
the continued existence of the Vamatantras although they integrate them into
Trika. There was, however, certainly a tendency for the Vamatantras to lose their
identity in the ambiance of Kashmiri Saiva Trika.
237. The Sikhabheda in the §KS consists of the following eight Tantras:
Bhairavi called Sikha, Vina, Vinamani, Sam m ohanam , tyamaram, Alharvakam,
Kabandha and $ira£eheda, In the Vamatantra list Vina is the seventh,
Sammohanam may correspond to the fourth called M ohanam and Vinamani to
the fourteenth called Cintamanimahodayam (although this identification is
certainly more tenuous than the previous one). Also worth noting is that the
Sirascheda, which is the Jayadrathayamala, finds itself in this, rather than in the
Yamala group, possibly indicating its closer affiliation to the Vamatantras,
238. That there was some problem here is confirmed by the fact that the
correspondent member in the Bhairavd$(aka is called simply “Bhairava. ”
239. P.Ka, I/27a.
240. Thus Jhankara is no. 31 and Sekhara no. 21,
241. dak$irie daksipo margas caturvim satibhedatab |
saivas sarvesu kurvanti ye grhastha dvijottamah |j
(TA, vol. I, p. 44).
242. yamale m atrtantre ca kapale pancaratrake [|
bauddhe carhamate caiva Iakuie vaidike’pi ca j
anyesvapi ca marge§u tat tac cha^traih svasastratal) ||
saivab kurvanti dik$adyam tal lihgasthapanadikam |
(P.Ka, 1/1 22b-4a).
243. The Yamalas listed here are Rudra0, K a n d a B r a h m a ° , Visnu°,
Yama°, Vayu°, Kuvera° and Indra°„
244. The names of these Yamalas are listed in the Rjuvimarsini (p. 43):
Notes to Part One 155
m ahabhairavatantre’smin siddhayogesvarimate |
(fl. 15a).
252. proktam bhagavata kila sam uhah pitham etac ca |
(TA, 37/18).
Notes to Part One 157
Ksemaraja explains that these four pifhas are the most im portant Vidyas,
Mantras and Mapdalas described in the SvT. The Sarvaviratantra similarly
claims that it consists of all four pithas.
258. e?a sa samaya devl atra sarvam pratisthitam |
catuh pifhe§u samaya atra sarvavinirgata ||
vamadak(finatantre§u samanya sam aya para |
kubjika nama vikhyata sam ayastha kulesvari j|
yatra visvesvaram sarvarp samayadyaqi vinirgatam |
m antram udragapo hy atra vidyam andalakadikam ||
(KMT, fl. 30b-31a).
259. yasmad bhandaram ityevam sarvasvam yoginlkule |
atha ca sarvapi^hesu m ate’yaip sam ayatm ika |j
(Ibid., 31b).
260. KRU, fl. 5a.
261. KMT, fl. 30a.
262. acaram kulacaratp m udraplfham j
(NA, MS No. 1/228, fl. la.; BSP, vol. II, p. 67).
158 N otes
263. See Jayarath a’s commentary in TA, vol. Xlb, p. 114. How these rituals,
and the texts which are their sources, are associated to theseplthas is not explained.
I am not too clear in my mind about this at present,
264. iha v id y a m a n tra m u d ra m a n b a la tm a ta y a c a tu sp ith a m tavac
chastram j
(TA, vol. X lb, p. 114).
265. NA, MS no. 1/35 is dated N.S. 307, the equivalent of 1188 A.D.
266, namämi catuhpïthapîfhadevyâ(h) kulesvaram |
(Ibid., fl. la),
267. m anträkhyam tatra vidyayä upodbalanam äpyäyah |
(TÄ, 37/ !9b-20a).
268. m antram ärgänusärena m udräbhedah sahasrasab J
(BY, fl. 199b).
269. m antrapratikrtir mudra XXXXXXXX |
(TÄ, 3 7 /20b).
270. The order in the Sarvaviratantra is different:
mudra mantjalapifham tu m anirapltham tathaiva ca j
vidyapltham tathaiveha catuspifha tu sarphita ||
(SvT, vol. I, p. 10).
K?emaraja enumerates these pifhas in the following order: Vidya, Mantra,
Mandala and Mudra (ibid.). Abhinava however says:
worshipped (TA, 28/156b). Pascima ritual also provides for the worship of
scripture. According to the MBT (Y), the Kuldgama is to be worshipped in a
specially prepared ma^data called Vidyapitha (fl. 86a). Similarly, at the beginning
of the GS the m anner in which the master and then the Agama are to be worshipped
is described (p. 5), while the SatSS explains how the Tantra of the Kubjika
tradition is to be worshipped before the master teaches the disciple its meaning
(SatSS, 3/98-102).
281. vidyapithapradhanam ca siddhayogesvarimatam Jj
tasyapi param am saram malinivijayottaram (
(TA, 37/24b-5a).
The MVT itself states its own close affiliation to the SYM. Thus the god
says to the goddess:
291. NA, MS no. 1/1693. This is a palm leaf MS only two folios long.
292. NA, MS no. 3/379 (161 folios) and NA, MS no. 1/53 (157 folios).
The colophons of the latter MS read:
PART T W O
with the Krama absolute (mahartha) and so quotes this verse (M.M., p, 170)
to suggest that M ahartha is the highest esoteric doctrine hidden, and everywhere
present, in the scriptures. Mahe$varananda is not, however, really much at
odds with Abhinava because his Krama tradition is that of Anuttara which is
the absolute Abhinava’s Trika treats as Trika, the supreme principle itself.
Accordingly, Mahesvarananda does not see much difference between Trika-
darsana and M ahartha (ibid., p, 92).
13. athatah sarppravak§yami kulam nayanidarsanam |
yagam homaip japam karyam yena sarvam avapnuyat ||
(NT, 12/1).
14. Ibid., vol. I, p. 253.
15. Ibid., p. 252.
16. Brunner, in an article on the NT, says in a footnote: “One could ask
oneself why K$emaraja deems it necessary to explain the characteristics of this
school insofar as he has said almost nothing about the others, and why he
stresses the fact that it is essentially similar to them. Is it perhaps because it
was not popular in his time? Or described? Or not yet integrated into the great
Saiva family? One defends oneself with difficulty from the impression that this
integration is achieved at the cost of a certain am ount of effort. One frequently
comes across references in the Netra to the triad Varna Daksina Siddhanta
which seems to form a solid block: if in these cases reference is made to Kula,
it is because it is not a constant fixed group; it is like an addition, in another
movement of discourse.” Un Tantra du Nord: Le “Netra Tantra” in BEFEO,
Tome LXI, Paris, 1974 pp. 154-5.
17. After the Uttarakdmikd has described the five sroias we read:
paScimad vamasamjnam tu vamad dak$ipam uttamam |
daksipat kaulikam sre?tham m ahakaulam tatah param [|
purvamnayarp tatas ¿resiharp tasmat siddhantam uttam am |
(U.Ka, 24/92-93a).
In the Mrgendrdgama, the Siddha—and Yogim—-Kaula which, as we
shall see, are im portant divisions in the Kuldgama, are referred to as upasrotas
of the five srotas. (M r caryapdda, 1/37, 1/40-1).
18. antah kaulo bahih saivo lokacare tu vaidika(i |
(quoted in TA, vol. Ill, p. 27).
antah kau.lo bahi(i saivo lokacare tu vaidikah |
saram adaya tistheta narikelaphalarp yatha ||
(Ibid., p. 278).
19. ekaikakotidhabhinnam kulaiastram suvistaram |
(CMSS fl, 2b).
just “Guhyasiddhi "). A list of eight siddham âifkâ found in this work constituting
the “R ulatfaka" also occurs in a number of texts belonging to the Paécimàmnaya.
Indeed, Bagchi himself notes that he has come across it in the K.NT, a Paécimâm-
m yapaddhati. It is possible, therefore, that it is in fact a work of the Pascimâm-
naya, attributed to M atsyendranâtha. That these and other works could
wrongly be attributed to M atsyendranâtha is not difficult to explain in view of
the fact that many individuals were given this name as initiates. For example,
according to the K.NT this name was given to a Bengali brahmin initiated into
the Pahim am naya. Quoted by Bagchi, p. 68: varapâ vangadese janm a jâtîr
brâhmapo viçnuéarmà nâma . . . markka^anadyam yadà karsità tada srîmat-
syendranâthah |
24. For a table of Râjaputras and Yuganâthas, etc., see Gnoli’s Luce delle
Sacre Scritture Appendix X, pp. 876-880. Also below p. 69, 70, 80.
25. kulâni mahâkaulakaulâkulakulâkulâkhyâni \
(TÂ Xlb p. 155).
26. Bhairavakulatantra quoted in TÂ, vol. VIII, p. 182:
vâmamârgabhisikto’pi daisikah paratattvavit |
samskàryo bhairave so’pi kule kaule trike’pi sah ||
ci: vedâc chaivam tato vâmam îaîo daksam tato m atam |
tatah kularp tatah kaularp trikam sarvottam am param ||
(quoted in commentary on TÂ, !3/300b-301 also in commentary on 13/347-8),
27. kundânârp laksanam vakçye kule kaule tu pascime ||
(MBT(Y), fl. 8b).
Also: ye kecid devataS canye kule kaule tu pascime [
tiçthanti kundamadhye tu bhairavo bhairavl saha ||
(Ibid., fl. 9a).
(The indeclinable “saha" meaning “with” normally governs the instrumental.
“Bhairavo bhairavi, "which are both nominative singular, are therefore anomalous
forms. If the two words are conjoined and the compound inflected to form
“bhairavïbhairavena saha" the metre would be disturbed. For the same reason
the forms "bhairavena bhairavyâ saha," although grammatically correct, would
not fit in this verse. Moreover, it appears that both words are here in the
nominative to indicate that they carry equal weight. The gods can be either with
Bhairava or Bhairavi,as well as with both together).
28. anyathâ kulakaule ’smin sâmbhave na sprianti te |
(Ibid., fl. 29b).
Once the goddess has described the Wheel of M ahâbhairava and its sixty-four
energies she says in the same Tantra:
(Read -“su te ” for -‘'suta” and “a v a //''fo r “a v a il” “ Prabhu” should be in the
instrumental, i.e., “prabhuna." This is probably a scribal error as the present
form cuts short the metre by one syllable. “Samjna ca"should be "Samjnaya
166 N otes
in that case, however the seventh syllable would be short instead of long as the
metre demands),
in TA, 29/36 the order is 1) Bodhi, 2) Prabhu, 3) Pada, 4) Ananda, 5) Yogin,
6) A vail The order in the KM T, (fl. 3a) is the same as the KRU; the latter,
however, arranges these six into three groups of two; 4) + 6); 2) + 5) and 1) + 3), thus
allowing the Pafeimamndya to be free, in a sense, to have its own line of teachers
without losing its essential connection to the original Siddhasantam . The
KRU then links one group to the Purvamnaya and the other to the Pascimdmndya
and so aligns them through their common Kaula heritage and stresses their
intimate connection with each other. However, in the KMT, which is older
than the KRU, no such distinction is made. The six Kaula traditions are
collectively called the Siddhakrama which has been transmitted along the Siddha
path {siddhamdrgakramdydtam) (KMT, fl, 3a). The CM SS assigns the Kaula
traditions to the Purvamnaya which it integrates into the Paicimamnaya
(see below).
34. The Kaula schools display a marked tendency to exclusivity (as indeed
do all Hindu sects and cults in general). The Vrmikaulatantra, for example,
expressly prohibits contact with those who follow other scriptures (TA,
15/573b-4). A Kaula does not distinguish himself from other Kaulas on the
basis of caste but according to the Kaula group to which he belongs. The Kaula
tradition prohibits commensuality outside its circle as a whole and enjoins
measures for atoning defaults in this respect (TA, 35/272). Voicing the Kaula view,
Abhinava says: “ In order to attain a state of identity (with Siva and one another)
one must follow one’s own spiritual tradition and neither worship nor enjoy
the fruits of a Tantric gathering (cakra) along with those who belong to a different
spiritual lineage.” (ibid., 4/268b-9a).
Even members of one or other of the six schools (ovalli) affiliated to
M atsyendranatha could not worship together and went as far as to devise a
system of signs by which they could recognise members of their own group in
order to avoid others (TA, 29/37). Once within a group, however, no distinction
between its members was admissible. Caste and social status ceased to be
operational in this society; once in the fold the members become one with Siva
and so indistinguishable from one another. He who viewed the others or talked
of them in terms of former caste distinctions ran contrary to the Rule (samaya)
(ibid., 15/576). In this respect, the Kaula traditions were perfectly in agreement
with many other Tantric cults (see for example, SvT 4/414 and 4/545).
35. In an im portant passage in the Tantraloka (4/221-270) Abhinava
comments on part of chapter 18 of the MVT (18/74-81) which he presents as
typifying the Trika view and which he contrasts with that of the Saivasiddhanta,
on the one hand, and Kula on the other. Thus, whereas the former enjoins the
performance of rituals and the observance of vows and rules governing outer
conduct, the Kula position is seen to be one of denying their validity and rejection
of these outer forms in favour of inner spiritual discipline. Kula doctrine is
essentially based, from this part of view, on an exclusivist monism (advaya)
intolerant of contrasts, which thus rejects all forms of spiritual discipline that
are ‘external’, that is, ‘outside’ in the state of duality. The Trika view, however,
Notes to Part Two 167
40. “O mistress of the gods! [treasure] always in [your] heart [the teaching
transmitted] from ear to ear. That which comes from the master’s m outh and
enters the path of hearing abides [in one’s own] heart. A M antra extracted
(from a book) is like one written on water. The Vidya that bestows both enjoyment
and liberation should never be written in a book." (Parât., 8/27-8).
41. “Five are said to be [important] in the Siddhânta, four in the teachings
of the Left, three it is said in the Right, in the tradition of the East two, while in
the Western liturgical tradition (paécimakramâmnâya) of the Kutâgama it is
the master alone [who m atters]" (MBT(Y), fl. 24a; also fl. 87b).
42. “Now I will therefore tell [you] the doctrine of the master knowing
168 N otes
which the Tradition of the Master is transm itted” (Ibid., fl. 24b),
43. “This is the meditation, hard to obtain [even] by the gods, which belongs
to the tradition of the M outh of the Master; he who knows this, O Sambhu, is a
Kauia master (aearya)'* (Ibid., fl. 96b).
44. XXXXXXXX srigurum pa£cime$varam j
(Ibid., fl. 26b).
45. strimukhe nik?ipet prajna(i strimukhad grahayet punah |
ityadyukte(i kulaprakriyayaip dutlm ukhenaiva ¿isyasya jnana-
p ra tip a d a n a m n a y a t iha g u ru tad d O ty o h sa m a sk a n d h a ta y a
upadanam |
(TA, vol. 1, p. 35; see also TA, vol. Xib, p. 88).
46. See TA, 29/40, Also:
(In order to convey the intended meaning of this statem ent in a grammatically
correct form, “yoginl” should be in the genitive, i.e., “yoginyah ” and “daisi-
kendrena” should not be an instrumental but an ablative, i.e., “daisikendrat").
49. rahasyam sarvaSastranam amnayahfdayarp param |
vaktra(t) vaktragatam jnanarp yoginimukhasarpsthitam j|
(CM SS, fl. 14b).
50. yogini param a sakti(h) sahajanandadayini |
bhairaveccha tu sa vartte tena sa dvividha na hi ||
(MBT(Y), fl. 51b).
51. TA, vol. II,. p. 104.
52. TA, 28/147 states that there are four types of Saivas, namely those who
divide the knowable into: 1) Nara, Sakti and Siva (i.e., the Trika); 2) The five
faces of Sadaiiva with their five powers: VameSvari, Khecarl, Gocari, Dikcart
and Bhucari (i.e., the Krama-influenced Trika of the Nisisamcaraianlrd);
3) into the ten and eighteen divisions (i.e., the Siddhanta) and 4) six currents
(srotai)—these are the Kaulas who add an extra lower current, that of the
Kulagama, to Stva’s five faces. See TA, 28/147 and Jayaratha’s commentary.
53. vaktrarp hi nama tan mukhyam cakrarp uktam mahesina |
yoginlvaktram XXXXXXXXXXX j|
(TA, vol. Xib, p. 89. See also TA, 29/124b-5a).
54. See TA, 15/206a and commentary.
55. TA, vol. IV, p. 160.
56. Monier-Williams notes in his Sanskrit-English dictionary that “Picu” is
the name o f a Yoginl. Ghosal Sastri (vol. II, part IV, p. 40) notes other meanings of
this word. These include “cotton,” “a type of soft grain,” “a weight,” “a leper,”
Notes to Part Two 169
“the neem tree” and “thorn apple.” It is also the name of the Bhairava who
teaches in the BY and of one of his eight faces.
57. picuvaktradyaparaparyayam yogimvaktram eva mukhyacakram
uktam | patalakhyam adhovaktram sr?tyartham sarppraklrtitam |
.....................adhovaktram srstivaktraqi picuvaktram j
quoted by Schoterm an (pp. 86-7) from the $a|SS, a T antra belonging to the
Kubjika school. He also points put in his commentary to verse 2/ 37 of this Tantra
that this mouth is called Picuyoni. Picuvaktra is also equated with Patalavaktra
in TA, 15/206 and commentary. The BY (fl. 201a) also confirms that those
Tantras which deal with Kulacara are said to originate from the Lower Current
along with others. In the context of describing the content of the Lower Current,
the BY states that it is Bhairava who teaches here. The SKS’s canon is unusual
insofar as it equates the M outh of the Yoginl with the Right Current from which
arise the sixty-four Bhairavatantras: “The scripture called Bhairava consisting
of eight times eight [Tantras] entered all together into the independent Saiva
[tradition] that is, the [current of the] Right, that of the Emission of the H eart”
(TA, vol. I, p. 41).
Jayaratha comments that “the Bhairava group consisting of sixty-four
[Tantras] is predominantly monist and belongs to the Right Face which
corresponds to the [single] non-dual inner nature of the union of Siva and Sakti,
also called the M outh of the Yogini.” In other words, the Bhairavatantras are
here understood to be vitally linked with the Kaulatantras through their common
source. The BY is a good example of how this works. Its very name—Picumata
(the Doctrine of Picubhairava)—subtly implies its hidden connection with the
Kaulatantras. Accordingly, Abhinava quotes it along with other Tantras
belonging to the Bhairavasrotas in the course of his exposition of Kaula ritual in
the Tantraloka. One chapter (56) of this work is devoted to Kulacara while another
(69) deals with the Picubheda. The overall Kaula character of this work is
unmistakable, although it does not consider itself to be such. Certain passages,
however, refer directly to the attainm ent of Kaulasiddhi and union with Kula
(see quote in TA, vol. Ill, p. 64). The Kaula character of other Tantras belonging
to the Dak$irtasrotas is also evident, particularly of those Tantras which teach
Trika doctrine. Thus, according to the MVT, the SYM teaches the method by
which Kulacakra is to be worshipped (MVT, 19/48). Abhinava quotes Trisiro-
bhairava, a Trikatantra belonging to the Daksifiasrotas, when defining the
meaning of “K aula’'as a metaphysical principle (TA, vol. Xlb, p. 6). He also refers
to it several times while describing Kaula ritual in the Tantraloka, including one
passage which deals with the characteristics of the ideal Tantric consort (TA,
vol. X lb, pp. 69-70), That not ail Tantras of the Bhairava group are so markedly
Kaula-oriented becomes apparent when we consider the character of the
Svacchandabhairavatantra whose ritual programme is in many respects similar
to that of the Siddhantagama in which Kaula rituals and practice are largely
absent. It seems that the Vidyapifhatantras of the Saivagama were, in this sense,
extensively Kaula in character, and it is to thisptfha that the Trikatantras generally
belong. Moreover, it is to this pfyha that most of the Tantras of the Paicimamnaya
170 N otes
affiliate themselves and so do many other Kaulataniras. Thus the god says in the
Yoginlhrdaya; “T h e Secret Doctrine I have hidden which bestows immediate
understanding . . . resides in [those Tantras] associated with the Vidyapifha.
It bestows divine yogic accomplishments and [is attained] by those dedicated to
Kulacara " (see YHr., 2/76-80).
58. See Schoterman, p. 87, who has drawn up a table of these correspond
ences according to $a$SS, chap. 42.
59. See Schoterman, p. 87.
60. yoginyo lebhire jnanapi sadyo yogavabhasakam |
yena tad yogimkaulam nottirpam tabhya eva tat |j
(M t.l.caryapada, l/40b-Ia).
(The word "vistara"i& masculine so "vistaram "should be "vis tar a h ." Accordingly,
"kathitam " which agrees with it should be "kathitah. " The third quarter of the
verse contains an extra syllable. This is probably due to the addition of the
word "iri” made as a sign of respect for the KMT. "Vistaram kathitatp m aya”
in the second line should read “vistarafa kathitah maya. '0
The Yogakhatitfa of the MBT refers specifically to the K M T of 3500 verses
which corresponds to the length of the recovered text of this Tantra. The
Gorak$asatphitd, which postdates the MBT, to which it refers in one place (p. 278),
also knows the KMT, to which it refers in several places as Srimata (p. 4, 24, 49,
116, 158, 208, 277, 375) and once as Kulalitantra (p. 152). It is closely affiliated
to the Srimatottara (perhaps they may be the same work) which, as its name
implies, follows after the KMT. The Kularatnoddyota is also a later work as it
affiliates itself to the Kubjikdmata. The colophons are generally of the type:
ity a d in a th a v id ita rp p a n c a sa tk o tiv is tlrn a s rlm a tk u b jik a m a te srl-
kularatn o d d y o te ..........
Also the Kubjika paddhati by M uktaka, the KNT, certainly postdates the
KM T as does the Kularatnapancaka. Although all these texts are later than the
KMT, there is still room to doubt the originality of this T antra. Thus in one place
in the KM T it seems that there is a reference to an earlier T antra.
On fl. 49a the goddess says: “now explain clearly what you have alluded to
in the previous T antra but not clearly elucidated” :
This reference may, however, be construed to m ean that the m atters which are to
be discussed are found in older Tantras which do not necessarily belong to the
Pascimdmndya, We have still to study the extant literature to see whether there
is any more evidence to decide this issue one way or the other. It seems highly
probable, however, that the KM T is not only the root Tantra of the Paicimdmnaya
but also the first,
78. KMT, fl. 44b. This reference is part of a passage reproduced in the GS,
14/195-206a. O ther Tantras referred to in this passage are the Satpmohana and
Svarodaya which is a Tantric astrological work. The Svacchandabhairava is
referred to here as “the best of T antras,” Svacchanda M antras are in fact
integrated into the Paicimdmnaya (see, for example, the exposition of
Svacchandastra in GS, p. 155 ffi). C. Chakravarti notices two after works referred
to in the KMT, namely, the Labdhvitantra (emend to “Laghvitantra”) and
Aghoriqldmaratantra (RASB Tantras, cat, II, p, 874).
79. PTv, p. 184, This reference has not been traced in the 3500 verse recension
of the KM T preserved in num erous MSs in Nepal,
80. paScimam sarvam arganam tvarp tavatt anusilaya (
Notes to Part Two 173
(The fifth syllable of the fourth quarter is long whereas it should be short. This
syllable is shortened, however, if we form the compound “nirahapxkdridrdha-
vratah”.)
82. According to the following reference, the K M T presents the doctrines of
the Paicimdmndya free of contam ination from the Purvamnaya represented
in later Tantras as a close associate of the Pascimdmnaya:
are in six groups of which the TrikaSastras are the “purvardha." To understand
what is meant here we must first draw the following diagram. See Figure 4.
Figure 4. The Spatial D eploym ent o f the Tantras A ccording to the Bhargasikha.
Upper Siddhanta
N orth Vamatantra
South Bhairavatantra
The “purvardha" seems to refer not to the “half on the west side” but that
“in front” which is said to be “half of six” (safardha), th at is, the three “in front”
of Trika, i.e., the Siddhanta- Vama- and Bhairavatantras of which Trika
is the essence.
88. KA, 3/5.
89. “The five Traditions have originated from My five faces. These five are
said to be the paths to liberation, namely, [the Traditions of] the East, West,
South and North along with the Upper. O Fair One, this is indeed true nor is there
any need to question this. O Mistress of Kula, there are m any traditions
originating from subdivisions of the four Traditions I have previously explained
to You in this T antra. O Beloved, You Who are praised by the Vlras, many
are they who know the four Traditions but few those who know the nature of
the U pper One” (KA, 3/7-10).
90. F or an account of this text, see Dvivedi, intro, to the NSA, pp. 46-7.
Also intro, to LAS, vol. II, p. 73 and Ta.Sa., p. 671.
91. They are equated with the cycle of m anifestation thus; Purva-Sr$ti,
Dakpria-Sthiti, PaScima-Samhara and Uttara-Anakhya. Although we cannot
make out much here of im portance for the historian, these equations are
interesting insofar as only four amnayas are represented. Moreover, the
Vttaramnaya corresponds here to A nakhya which is significant insofar as this
is the highest principle of the Kalikrama. The association of the Kalikrama
with the Vttaramnaya is well established in other T antras (see below).
92. The YHr, pp. 100-102. The equations m ade in this passage between
the four amnayas and m oments in the cycle of manifests tion are described in detail.
Basically Paicima is equated with the precosmic seed (bija) state consisting
Notes to Part Two 175
of the couple {yamala), Rudra and His power of action. Purva is the moment of
emanation (sfsii) which includes also persistence (sthit:) and withdrawal (samhdra)
and so is three-fold. Dak$ir?a is Kamakala consisting of the union of the red
Saktibindu with the white Sivabindu and corresponds to persistence (sthiti).
The Vttardmndya corresponds to A ndkhya. Thus, these equations basically
agree with those made in the Sarpketapaddhati. Of interest here is the equation
made between the Daksitydmnaya and Kamakala, which implies that Srividyd
is associated with this dmnaya, thus agreeing with the CM SS (see below).
93. Four MSs of this work are deposited in the NA at Kathm andu; there may
be others listed in the private collections photographed by the N G M PP which
I have not had the opportunity to see. The MSs in the NA are 1/767 (38 folios),
1/199 (69 folios), 1/1560 (34 folios) and 1/245 (36 folios). References are to
MS no. 1/767.
94. CM SS, fl. 2a. Divyaugha is also referred to as Divyakaula (Ibid.) or
Divyaughkrama (ibid., fl. 2b). It is associated with the Picuvaktra (ibid., fl. 2a).
95. The T antras referred to are the Tantrasdra, Krtyasara, Kaulasara,
Trikasdra and Yogasara. The CM SS also refers to the Yoginihrdaya, Kdla-
cakrodaya and Yamala. Other scriptures and groups are the Bhairavasta and
the Astdsfabheda, the Pancasrotabheda, the Garuda- and Bhutataniras. It then
refers to a num ber of other groups (bheda) namely Mantra, Tantra, Cakra,
Hrddniara, Sara, Kald and Nitya- bheda as well as the Svacchanda which it
considers to be the highest (fl, lb). Other works referred to in the body of the text
are the Siddhayogeivarimata (fl. 14a), Rudrayamala (fl. 18a) and KM T (fl, 24a).
96. The C M SS says that grha is the inner Dwelling. It is the Wheel of
Passion (ratieakra) which is in the womb of the goddess, sanctified (bhdvita) by
Kula and Kaula. Yogis reside there in that consciousness, mentally discerning
that which is free of being and non-being and playing in the Kulamnaya, the
imperishable abode of consciousness (fl. 14b). The Pak'imagrha is;
From this point of view the Pascimdmndya is also known as the $ambhava-
sdra, i.e., the essence of Saiva doctrine. It is Siva who is the ultimate source of the
Pascimdmndya and is himself its arising and present in it:
tatrastharp sambhavam natham paicim at paicimodayam |
(MBT(Y), fl. 38a).
Indeed all four amnayas are represented as belonging to ‘Siva’s circle’ which
is identified with Kula doctrine itself:
athatali saqipravak§yami yani siddhodayam kule j
pascime dak$ine vame purve va iSmbhumapdale ||
(Ibid., fl. 40b).
Also:
tvaya mahyaqi maya tubhyam tvaya mahyam punar maya ||
kathitam tava susroni tvat sanganye$u mok$adam |
(GS, 15/299b-300a).
In the KMT, which presents the doctrine at its origins, it is indeed Bhairava
who speaks, even so, as the embodiment of the Word held within consciousness
(.sabdarasi). He is forced to acknowledge his dependence on the goddess who, as
MaiinI, is the power of his speech by virtue of which the Word can become
manifest. Again, it is by virtue of the goddess travelling and manifesting herself
in the sacred places throughout India that the doctrine is spread. She, and not he,
is the chief protagonist as, indeed, is generally the case with the non-dualist Tantras
whether they consider themselves to be specifically Kaula or not.
101. sambhavam yatra llnaqi tu vyapakaqi sarvatom ukham j
akula ca kularp jnanam vividhaqi paScimaqi grham j|
(MBT(Y), fl. 35b).
(Emend "akula>'to “akulam".)
102. vimalam pascimarp vesma sarve§am uparisthitam |
vi£e§am tena viditam dharmadharm avivarjitam |(
(Ibid.).
(The word “vi$e$a " is a masculine noun; even so it is treated here as if it were an
adjective governed by the word “ve4man.” The sixth syllable in the third
quarter is not long as it should be).
103. idam eva dak$ipanvayarupam srikamarajamadhyaste j
raudratm akastu rudrah kriyamayi saktlr asya rudrapi ||
yugalam idam tartlyarp bijam palayati paicimamnayam |
(quoted in YHr, p. 101),
See chapter 2 of the Saubhdgyasudhodaya published with the NSA, pp. 311-313,
104. For a schematic representation of their location see below, fn. 110.
105. “I praise the goddess Para Siva whose imperishable form is the letter
A which pours forth as the waves of Kula!” (NSA, 1/10).
106. “This Vidya is the Great Vidya of the Yoginls, the great arising, the
Kulavidya, O great goddess, which accomplishes the goal of all that is to be done”
(NSA, 1/103).
107. “You (O goddess) afflict with pain the one who sets out to practice
according to this doctrine not knowing Kulacara and without having worshipped
the m aster’s sandals. Knowing this, O fair one, one should always be intent on
Kaulacara ” (YHr, 3/196-7a). For other references which clearly demonstrate the
Kaula character of the SrividyS teachings, see also YHr, l/25-27a; 2/15; 2/51;
2/68; 3/139-142; 3/146-153 and 3/170-1.
108. For the dhyana of Srividya M ahatripurasundari see NSA, 1/130-149.
109. “ Kula is Sakti while she is said to be Nitya.” Quoted from the Nitya-
lantra in TA, vol. XIa, p. 51. All the presiding goddesses of the amnayas are
called “Nitya” in the Paratantra,
178 N O TES
110. This is standard practice in m any Kaula schools. The P ascim am nava
also refers to itself as divided into lines (o li) originating from m asters said to
reside in these pi/has. In the K N T they are the three listed below alo n g with
their L o rd s:
In the D ivyakram a of the Srividya the four pithas are arranged in a triangle,
the c orners and centre of w hich c orrespond to a seed-syllable (bija ) and teacher
w ho appeared in one o f the four A ges. See Figure 5 below.
113. Schoterman p. 36, In the Parátantra, where the ámnayas are six, the
goddesses presiding over the ámnáyas are fearful in form except Kubjiká and
Tripurasundarl who are both seated on Sadááiva’s lap and described as young
and beautiful.
114. The term “Kram adarsana” is not common in the original Tantras,
They generally refer to their doctrinal and ritual system as “Kramanaya,”
“A tinaya,” “ M ah arth a,” “ M a h a rth a k ram a ,” “ M ahakram a,” “K állkula,”
“Kalikulakrama,” “Kallkrama,” “Kállnaya” or “Devinaya" (see Rastogi,
pp. 16-30).
115. CMSS, fl. 23b.
116. Jayaratha refers to a number of authorities in his commentary on
TÁ, 4/148-170 where Abhinavagupta expounds the order of the twelve Kalis
constituting Sapiviccakra. Amongst them are the Devipañcasaliká, Sñsárdha-
saiika, and the Kramasadbhava which have been edited from Nepalese MSs by
Mr, G. S. Sanderson but, as yet, unfortunately not published. The Krama
sadbháva enjoins the worship of seventeen Kalis in Anákhyacakra whereas the
other sources usually speak of thirteen (apart from the Kramastotra where
they are twelve). In the Érísárdhasatika (quoted in TÁ, vol. Ill, p. 161) the thirteen
Kalis are listed in the following order:
1) Sr§ti° 2) Sthiti0 3) Saiphara0 4) Rakta° 5) Svakálí 6) Yamakáll 7) Mftyu°
8) Rudra° 9) Paramarka° 10) M artapda 11) (Kalágni) Rudra 12) Mahakáií and
13) Mahabhairavacapdograghorakall who is in the centre. The Tantrarája-
bhaffaraka (quoted in TÁ, vol. HI, p. 189) also lists thirteen Kalis, The eighth
Kali is called Bhadrakall, as it is in the CM SS, while the name of the thirteenth
Kali, Mahabhairavakali, is a variant of Mahábhairavacandograghorakáll which
is the commonly accepted form of her name. The order of the Kalis in the
Devipañcasaüká as quoted by Jayaratha is listed below:
1) Sr$|i° 2) Rakta° 3) Sthiti0 4) Yama° 5) Sam hara0 6) Mrtyu° 7) Rudra°
180 N o tes
The KRU does refer to the school to which it belongs as the Pascimamnaya in a
few places but generally prefers to identify itself as belonging to the Srikram a,
K ramanvaya or K ulakram a. In the KRU the terms Kula and K ram a are freely
interchanged. F or example, in one place the Yuvakram a of Kubjika as a young
maid {yuvati) is said to be worshipped according to the “K ulapaddhati” (3/63)
which a little further on is called “K ram apaddhati“ (3/67). This is in accord
with the “Srikulakram am arga " (3/85). We also come across cognate expressions
such as “D ivyakram apaddhati” (3/174) or “Srikulapaddhati" (fl. 58a).
152. A bhinavaTs Tantrdioka can be viewed as an extension of this same
phenomenon. One essential difference is th at this is a work of know n authorship
which makes no claim to being an original scripture. Even so, however, it is
184 N otes
Nepal, these three Devis are revered as the components of Srikula, i.e.,
PaScimamnaya:
(The last half of the third line reads: “sâmbhavasâktam ânavarp.” The copyist
has written “~akta~ "twice by mistake. In the last line read “-traye"for ' -trayâm. ")
184. See GS, 8/121. CMSS, fl. 8a-8b furnishes an example of how this works.
185. An account of these three initiations is found in the KRU, fl. 68b. ff.
186. For an extensive account of Éabdarôéi and Aialini see A. Padoux’s
standard work: Recherches sur La symboliques et L ’Énergie de la Parole dans
certains Textes Tantriques, Paris 1963. For an account of how these operate
as codes in the Tantras of the Kubjikâ school, see Schoterman, pp. 182-221.
187. Sabdaràsiéca mâlinyâ vidyànâm tritayasya ca |
sângopâhgasamàyuktâ trikatantraqi karisyati j|
(KRU, fl. 78b).
(For “sabdarâiié" read a b d a r â é e é For “-sam âyuktâ” read samâyuktarp
188. For example, the CMSS acknowledges that it has derived the
pifhasamketa from the Sid d ha y ogeèvarimata:
sarpketa eça vîkhyâtas siddhayogesvarimate {
(fl. 14a).
According to this account, there are four pifhas, namely, Ohkâra, Jalandhara,
Pürna (girt) and Kâmarü. The first is located in the mouth, which represents
the circle of the Sky of Consciousness as the state of withdrawal and suspension
of all extroverted conscious activity. Jalandhara and Purna are in the right
and left ears respectively while Kâmarü resides at the end of the flow of the
vital breath (prônante).
189. The original Trikatantras, such as the SYM and Triéirobhairava,
as we have already had occasion to remark, were Bhairavatantras. Moreover,
none of the Trikatantras I have had access to, namely, the published Malini-
vijayottara, a part of the SYM and Srïtantrasadbhôva (NA, MS no. 1/363) ever
refer to Trika as a school. The last two simply affiliate themselves to the
Vidyapifha of the Bhairavasrotas. Another unexpected feature of these Tantras
is that they are not Kaulataniras although their doctrines, metaphysical
presuppositions and rituals are of a Kaula type. Where Kula is referred to as a
ritual pattern, doctrine or tradition, it figures as an element in the broader
context of the Tantra as a whole. However, the later Trikatantras, which did
think of themselves as belonging to a Trika tradition, such as the Nisisahcara,
Kularatnamâlâ, Bhairavakula and Trikasara, define Trika in Kaula terms as the
highest Kaula school which as such is, in a sense, beyond the Kula tradition.
These facts along with the absorption of Krama doctrine in some form are
fundamental features of the history of the development of Agamic Trika before
Abhinavagupta.
190. The Paicimatantras, it seems, are so conscious of having absorbed
many Trika elements that at times it becomes necessary for a Tantra in the
course of its exposition of a topic to distinguish what it is going to say about it
from what the Trika and others have said. Thus, for example, the MBT proclaims
that it will explain the characteristics of the sacrificial hearth according to the
PaScima Kula/K aula tradition (see above p. 62) as explained in the Maharnava-
188 N otes
tantra(l) and the common ritual which concerns it, free of T rika and Dak§ina
elements:
been of the Siddhanta and Puraijic type which prescribed lihga worship and
consisted of cults such as that of Umamahesvara described in works like the
Sivadharma, Sivadharmottara and $ivapani$ad. An early tenth century
MS of these texts (NA, no. 4/531) is preserved in Nepal. Similarly early Nepalese
MSs of the Kiranagama and other Siddhantagamas also exist. Although the
earliest Saiva cults in Nepal were not Kaula-oriented, a reference to Nepal as one
of eight upapifhas occurs in the MBT(Y), fl. 15b. A nother reference has been
traced in a MS of the NUisahcaratantra (NA, MS no. 1/1606) which is written
in early Newari characters probably not younger than the twelfth century.
Here PaSupati is mentioned as residing in Nepal along with his consort Guhye£vari.
These references suggest that Nepal did not become a centre of Kaula Tantricism
much before the eleventh century.
194. See Schoterman, p. 6, 37.
195. Schoterm an, p. 6.
196. Practically the only reference to Kubjika apart from in h erT antras is a
brief description of her worship in the Agnipurdpa. See Schoterm an’s article
A link, between Pur an a and Tantra: Agnipurana 143-147 in ZD MG sup pi. IV
Wiesbaden 1980. See also Les Enseignements Iconographiques de L ’A gnipurana
by Mth. de Mailman Paris 1963 pp. 159-60, 206-207.
197. See Schoterman, p. 10, fn. 4.
198. Dr. KiSoranatha Jha in his Hind! introduction to the first part of the
Guhyakalikhan4a of the Mahdkalasarphitd (p. 18, fn. 1) informs us that the
late Parmefivara Siipha, who was a Maithili Tantric, had a statue of Guhyakall
in her ten-faced form m ade and installed in a temple in the village of M adhubani
in the Bhauragadhi district of Behar. This is the only example he knows of a
representation of this form of the goddess in India. There is, however, an old
image of Guhyakali carved in black marble preserved in the RSjputana Museum
in Ajmer (No. 193, 268). See article by P. K. M ajumdar, Sakti worship in
Rajasthan published in the Sakti Cult and Tara, Calcutta 1957, p. 68.
199. Schoterman, p. 6.
200. Kuiper, F. B. J. Proto-M unda Words in Sanskrit, Amsterdam 1948,
p. 42. ff.; referred to by Schoterm an, p. 11.
201. Kubjika is also called “ K ukaradevi” which is not only an abbreviated
form of her name but also the seed-syllable corresponding to the Earth Principle.
The word "ku ” means “earth” and so“Kuja” which is one of KubjikS’s common
names means “born of the earth” and “ Kujesvari”—“the goddess born of the
earth.”
202. tatha sa kutila vakra m adhyam olyam kulesvara |
kubjika $a|prakifra ca vrddharupena devata ||
(MBT(Y), fl. 95a).
203. The KMT, fl. 69b says: “She in whose centre the universe resides and
who resides in the centre of the universe is thus called KhanjikS; she who is
subtle and present in subtle things.” These names give rise in their turn to
various names for the Kubjika school such as "Khahjinim aia," "Khahjinikula, "
" Vakrikagama, ” “ Vakrikamata” as well as “CincinUastra”or "Cihcinimata"and
“Kulalikdmnaya“(for Kubjika as Cincinl see below, and as Kul&likathe potteress,
190 N otes
(The end of the third quarter is one syllable short. This deficiency could be
remedied by adding “tu ” or some such particle to complete the metre.)
207. KMT fl. 69b also Schoterman p. 11.
208. svanâbhimathanâd devi svakîyarasanâ purâ ||
brahmâpçlam garbhatas tasyâ jaiidivyena yoninâ
tad ârabhya maheéâni kubjâdevlti visruta ||
(Parât., 3/2b-3).
209. “The Wheel of Energy consisting of consciousness and the unconscious
resides in the wheel of the nave! which is the Great Matrix. It is supremely
divine, the illuminator of the Brahmanâdï which, by its upwards and downward
flow, pervades [all things] and faces in every direction. Piercing through the
path of the palate it causes the nectar of the power of consciousness to flow.
This, O fair one, is the door to liberation” (CM SS, fl. 5b).
210. Kubjikâ is extolled as the Divine Light of consciousness in a hymn in
MBT(Y), fl, 67a-69b.
211. MBT(Y), fl. 81a.
212. Ibid., fl. 67a.
213. ¿ftëâsane parâ devi kufilâ divyarüpinï |
parâparâparâsakti(r) yâ para param â kalâ j|
m antram âtâ parâyonir nâdiphântasvarüpipl |
(Ibid., fl. 69b).
The fifty goddesses which embody the energies of Mâiinï are extensively
described as part of Malinicakra in GS, pp. 29-45 where they are said to belong
to the Vâmamârga, In this Wheel are located the three goddesses Para, etc.
and worshipped there as aspects of Mâlinî. The dhyànas of these three are found
in verses 8/159-163 (Para), 8/113 ( Parâparâ) and 8/171-8 (Aparâ).
214. CMSS, fl. la.
215. A hymn dedicated to Kubjikâ as Bhaga is found in the CMSS, fl, 10.
216. CMSS, fl. la,
217. Ibid.
218. Kubjikâ is nowhere referred to as Cincinl in the KMT. She is called
this quite commonly, however, in later Kubjikatantras such as the MBT and
KRU as well as in the CMSS. This is clearly, therefore, a new element that has
evolved in the Pascimâmnâya in the later phases of its development.
219. CMSS, fl. 11a.
220. KRU, fl. 77b. Possibly this new development in the Pascimâmnâya
(see fn, 218 above) was initiated by this yogi. Goddesses are still quite commonly
Notes to Part Two 191
associated with sacred trees. One exorcist (ojha) I met in Benares told me that he
had gained the power to propitiate the goddess ¿ltala (who causes smallpox
and other skin diseases) when he had a vision of her sitting on her sacred neem tree.
221. CMSS, fl. la.
222. ibid., fl. la-b, 13a-b.
CincinI is also one of the ten forms of ‘unstruck sound’ which resound in the
yogi’s cosmic body, figuratively called “the belly of the machine of Maya”
imayayantrodara). These ten are in order: 1) Cini 2) CincinI 3) The sound of a
pleasing voice 4) Conch 5) Stringed instrument 6) Flute 7) Cymbals 8) Rumble
of storm clouds 9) Sound of a running stream 10) Sound of a kettle drum.
KMT fl. 50. The same verses, in a slightly variant form, are quoted from the
BY in TA, vol. HI, p. 410.
223. adav eva mahadevi adinathena nirmitam |
pascimam kram asantanam svayam ye$(am tatafl priye j|
(KRU, 2/12).
(Read “jye?tam” for “yestam”)
224. The derivative status of the goddess is variously expressed in the
KRU; thus, for example, she is called “the one who is born of Akula” (ibid, 3/32).
225. KRU, 2/30-32.
226. Ibid., 2/13-15.
227. evam uktva jagaddhata ¿rinatha adisamjnakah j
(KRU, fl. 94a).
228. MBT(Y), fl. 60a.
229. tvat prasadena srinatha asesam kulanirnayam |
(GS, 21/176).
230. See above p. 91.
231. ¿rikantharp prathame pujyam ¿amkararp ca dvitiyake |
(KMT, MS 1/229; BSP, vol. I, p. 58).
According to the KRU (fl. 73a ff.) three lines of teachers are established in
Sivatattva. The first of these starts with Srlkantha who produced twenty-four
propagators of the doctrine, the second starts with Ajesa who produced sixteen,
while the third begins with M ahakala. According to the colophons of the GS,
Srikanfha brought down to earth the Kddibheda (i.e., the Kubjika group) of
the Kulakaulamata.
232. See Schoterman, pp. 36-38.
233. In, for example, MBT(Y), 8a “OH” is a synonym of “Ovalli" which
term is defined in TA, vol. X lb, p. 28 as “the current of doctrine” ( “ovaUyo
jnanapravahah ”),
234. See SatSS, 3/90.
235. svabhave kubjikakara divyadeharp kujambikarp |
candradvlpapure kubja srikapfliasya anugrahe jj
(Ibid., fl. 94b).
(Read “-deha” for “-deharp” and “kujam bika” for “kujambikarp”. The last quarter
is defective by one syllable; “tu ” for example, may be added to complete the metre
192 N otes
APPENDIX A
APPENDIX B
(Read “sahasram” for “-sahasre” and “anitam ” for “anita.” The fourth line
is too long by two syllables. If we read “kubjikam aiam ”for “srikulalikamatam”
the metre is preserved).
7. Idrsarp cintayed rupam kaulisam srikule^varam |
agamam srlmatam haste mahayogadharam ¿ubham ||
(Ibid., fl. 86a).
194 N otes
8. For D harm aklrti’s date and life see Warder, Indian Buddhism
pp. 469-472. The reference in the MBT reads: “dharm aklrti(r) bhavet tarkam
prajnaparam itakriya [" (MBT(Y), fl. 34b). This Tantra appears to be well
aware of the existence of Buddhism and refers specifically to the Sautrantika,
Vaibha$ika, Yogacara and Madhyamika (Ibid.). It also knows that foreigners
practise Buddhism:
krtvadau uditah so hi mleccha yatra upasakah |
(Ibid.).
(The metre is defective by one syllable in the second half unless we take “mleccha “
to be "maleccha”),
9. cathurthe tu yuge svante dharmocchede samagate j
mleccha raj ye kachat re ca bhunjate prthivlmahan j|
ravanasyavatare tu simdhos tire XXXX j|
(MBT(Y), fl. 91b).
There appears to be another possible reference to an invasion in the $a|SS. Here it
is described as one of the many horrors of the Kali Age, It says:
k§atriya ahave bhagnalj kari$yanti prabhutvatam |
0119b).
10. MBT(Y), fl. 70a.
11. Colophon of the Kumdrikakhanda of the MBT NA, no. 1/241; BSP,
vol. II, p. 59.
APPENDIX C
1. The manuscript is NA, no. 5/4650 Saivatantra 431. It is 275 folios long
and written in Devanagari script. The relevant section starts on folio 165b.
It extends from chapter 35 to 45. The colophons all begin as follows: iti bhairava-
srotasi vidyapifhe siras chede srijayadrathayamale m ahatantre caturvimsati-
sahasre prathamasafke £rikalasamkar$inyam . . . Then the name of the chapter
(patala) and its numberfollows. These are chapter 35 (fl. 163b-68b) nityahnikacara
yotha (?) sam bandhavatarakhyavaroanarp; chapter 36 (fl. 168b-170a) svacchanda-
sutranirpayati; Chapter 37 (fl. 170a-17 la) bhairavasutranirnayah; Chapter 38
(fl. I71a-2a) krodhabhairavasutranirnayah; Chapter 39 (fl. 172a~3b) mantra-
plthavinirnaya; Chapter 40 (fl. 173b~I81a) brahmayamalanirnaya; Chapter 41
(fl, 181a-2a) vi^pusutranirnaya; Chapter 42 (fl. 182a-3a) umayamaladisutra-
nirnaya; Chapter 43 (fl. 183b-4a) unnamed; Chapter 44 (fl. 184a-5b) sada-
siva$fa$$akanirpaya; Chapter 45 (fl. 185b-197b) cumbakacarya.
2. See Studies in the Tantras by P. C, Bagchi, reprinted, Calcutta 1975,
See pp. 109-114 for notices of the Jayadrathayamala.
3. bhltanam sarujartanaip du$tanam capi sasanat |
bhayanam ca paritranac chastram uktam hi suribhih ||
(JY, fl. 165b).
Notes to Appendix C 195
4. See also Bagchi pp. 110-11 where this passage is quoted. The JY also
accepts the standard division of the sastras found in the Agamas into five groups,
viz., Laukika, Vaidika, Adhyatmika, Atimarga and Anava which stands for
M antratantra (see above p. 49). This division of the idstras into five corresponds
to five fruits they are supposed to yield (JY, fl. 166a).
5. The Vaimala along with the Lakula, Mausula and Karuka is one of a
standard group of four Pasupata sects mentioned in the Agamas and elsewhere.
See SvT, 11 ¡69-14 and K$emaraja1s commentary. Also Jayaratha on TA, i/33.
There are eight Pramdnas according to SvT, 10/ 1134-35. They correspond to
eight Rudras that have incarnated as Pasupata teachers who founded the following
eight Pasupata schools; Pancartha, Guhya, RudrankuSa, Hrdaya, Lak$aija,
Vyuha, Akar$a and Adaria. (Ibid., 10/1134-5; cf. also TA, 8/328-9).
6. The text (fl. 166a) simply reads Guhyadi which 1 have taken to be a
reference to the Guhyasamdjatantra. It is interesting that the JY classifies the
Vajrayana sastras as belonging to this group, thus distinguishing them from
other Buddhist scriptures which are assigned to the previous one. We should
not, however, understand this to mean that the Buddhist Tantras are aligned with
the Bhairava and other similar Saivatantras on an equal footing.
7. This account largely agrees with the way the genesis of scripture is generally
described in the Agamas. According to the Kulamulavatara, out of Siva, who
is the supreme cause, tranquil and transcendent, emerges the power of will,
followed by those of knowledge and action. Through them, the worlds are created,
as is speech in all its expressions (MM, p. 39). Similarly, according to the
Svacchandabhairava (quoted in YHf, p. 153) a pure and subtle resonance
(dhvant) emerges out of Siva, the cause of all things. This is Speech which is the
power beyond mind (unmanasakti) that goes on to assume the form of scripture
and the spoken word.
8. The Agamas regularly refer to the types of relationship that form between
the teacher and his disciple through which the meaning of the scripture and the
realization it conveys are transmitted. The basic pattern is the same although
it may vary in individual cases. Abhinava records that according to the Kula-
rainamdld, there are five relationships; great (mahat), intermediate (avdniara),
divine (divya), divine-cum-nondivine (divyddivya) and mutual (itaretara)
(TA, 1/273-4). Bhagavatotpala also refers to five; these are supreme (para),
great (mahat), divine (divya), other than divine (divyetara) and mutual (itaretara).
(Sp.Pra., p. 84). Through these relationships formed between the teacher and
disciple and taught at different levels, scripture and its meaning are transmitted
from and through the divine consciousness which is its source and basis: “The
Lord, Sadasiva, establishing himself on the plane of master and disciple, brought
the Tantra into the world through a series of questions and answers.” (SvT quoted
in VB, p. 7). There is, as Abhinava explains, an essential identity between the
disciple who inquires and the teacher who instructs, as both are embodiments of the
one consciousness. The disciple represents the aspect of consciousness which
196 N otes
questions (prasfrsarnvit) and the teacher the aspect which responds (TA, 1/252-5).
Thus as Abhinava says “relationship (saipbandha) is the identity (that is
established) between the two subjects who question and reply. Its supreme aspect
consists of the revelation of identity in all its fullness. The other relationships
mentioned in the scriptures must also be considered in the light of this principle in
order that the results one desires etc. may manifest in all their fullness" (TA, 1/275-
6)-
9. Similarly the Svacchandasatpgraha says “this is the Lord Anasrita who
has five faces each of which bears three eyes and who has one, two, four or ten
arms. He is Sadasiva, the God of the gods who utters the worldly and other
scripture. [It is He who] has spoken the countless Agamas divided into superior
and inferior” (quoted in Y.Hr., p. 271).
10. M ahakarunika is probably a Paiupata sect. The four instruments
could also possibly be Pure Knowledge, Sound, the Drop and the metres,
11. See chapter 45 of this section of the JY.
12. We are reminded of Sum ati who wasAbhinava’s grand teacher of
Trika Saivism. He was reputed to have known all fivecurrents of scripture
(see commentary on TA, 1/213).
13. vamadak$inami£re§u bhinnapRhacatustayam |
vyapadisyate mukhyavac cchakha sakhantare sthitam j|
(JY, fl. 168a).
(The fifth syllable of the third quarter is long and the sixth short. In standard
anusjubh metre of the ‘sloka1variety the fifth syllable in every quarter should be
short and the sixth long).
14. mantravidyasuharmyanam sahghastom akadam bakam |
vratavrndam ca nikara(h) samuhati samhiti(r)valam j|
vicchindo m andalam pitham paryayair upasabditam \
(Ibid., fl. 168b).
(The word “kro d h a " is a masculine noun; here it is treated as if it were neuter).
25. An Anantatantra is listed as the first upagam a of the Vimalagama.
Anantavijaya is listed as the twenty-first Vdmatantra, Para is the name of the
eighteenth Rudragam a while A m rta is the sixth upagama of the Cintyagama.
Ananda is listed as the seventeenth Vdmatantra; it is also the name of the
sixth upagama of the Dlptagama.
26. For the Sarvavtra and SYM see index. A Visvddyamata is the twenty-
fourth in the Bhairavatantra list of the §KS. Visvavikanfha is thirtieth in the
Daksipatantra list while the Yoginijalasambara ts the twenty-ninth,
27. The JY says:
We have translated this verse above (p. 36) as follows: “ Belonging to the
Current of the Left are the perfect [Agamas including] the frightening Sirascheda.
The three: Nayottara, Maharaudra and Mahdsarpmohana have, O goddess,
emerged in the Current of the Left.” A lthough this is certainly a possible
translation of these lines, the JY does not in fact reckon itself to be exclusively
amongst the Vamatantras but prefers to classify itself am ongst the Dakfinatantras,
It does, however, also say that it belongs to both Currents (see above p. 113),
although in the detailed description of the contents o f these Currents it is amongst
those of the Middle Current and so we have listed it there accordingly.
28. The Ucchu$matantra is eighteenth in the Dak$inat antra list, thirty-fourth
in the NSA list and seventh in that of the &KS. The Sarvaiobhadratantra
is mentioned in VST, v. 317 along with the M ahasam mohana, It is therefore
probably a Vdmatantra,
29. These are listed further ahead as the eight Tantras which constitute the
Cakrabheda of the sixty-four Bhairavatantras.
30. A Bhimasarphitd is listed as the fifth upagam a of the Kdranagama
and as the sixteenth Dak$ipatanlra. Tilaka may be the Tilakodyanabhairava
which is the thirty-second D akfinatantra.
31. Siddhartha, the first Tantra in this list, may be the Siddhagama which
is the sixteenth Rudragama; if so this confirms the JY ’s statem ent that the
six remaining Tantras are still ideal rather than actual.
32. These gods and goddesses are frequently portrayed as given to orgiastic
revelry o r other chaotic behaviour which threatens to disrupt the cosmic,
ethical and divine order. Goudriaan ( Vinasikhatantra, introduction p. 19)
refers us to another example found in the Yogavasitfha (the first half of the
nirvapaprakarana 18/24 ff,). Here Tum buru and Bhairava are described as
enthroned together and surrounded by eight M others (m a tfka ) said to belong
to the Left Current and to be associated with Tum buru. Their appearance and
activity is intense. Their revelry and drunkenness breaks all limits of cultured
behaviour but they are put in their place by Siva.
Noies to Appendix C 199
S ANS KR I T TEXTS
vol. VII, KSTS no. 41, 1924; vol. VIII, KSTS no. 47, 1926; vol. IX, KSTS
no. 59, 1938; vol. X, KSTS no. 52, 1933; vol. IX, KSTS no. 57, 1936; vol.
XII, KSTS no. 58, 1938.
Nitya$odasikarriavatantra. With the commentaries Rjuvimarsini by Sivananda
and Artharatndvali by Vidyananda. Edited by Vrajavallabha Dviveda.
Y ogatantragrantham ala no. I, VarapasI: 1968.
Netratamra. With commentary by K$emaraja. Edited by M adhusudan Kaul
Sastrl. Vol. I KSTS no. 46, 1926; vol. II KSTS no. 61, 1936.
Paratantra. Edited by Ghanasum Ser. Prayaga; Kalyan M andir, 1952.
Pardtrirpiikd. With commentary by A bhinavagupta. Edited by M ukunda Rama
Sastri, KSTS no. 18, 1918.
Prati$jhalak?anasdrasamuccaya, Edited by Babu Kfspa Sarm a, Nepala Ra§!riya-
bhilekhalaya, Kathm andu, Devaidcitrasaijigraha 1963; Vol. I, 1966;
vol. II, 1968.
Brhatsamhita by Varaham ihira with English translation, notes and comments
by M. R. Bhat. Delhi: Motilal Banarsidass, P art I, 1981; part II, 1982.
MANUSCRIPTS
BOOKS A N D A R T I C L E S
The names of authors are recordedjustasthey are in the book or article, if the
text is in Hindi or Sanskrit, the a u th o r’s name is transliterated.
Agrawal, V. S. The Vämana Puräna— A Study. Benares: Prithivi Prakashan, 1983,
Alphabetical Index o f Sanskrit Manuscripts in the University Manuscripts
Library, Trivandrum. Edited by Suranada Kunjan Pillai. Trivandrum
Sanskrit Series no. 186. Trivandrum: 1957,
Bagchi, P. C. Studies in the Tantras, Part I. Calcutta: University of Calcutta, 1975.
Banerjee, J. N. The Development o f Hindu Iconography. 3rd ed. Delhi: Munshi-
ram M anoharlal, 1974.
Banerji, S, C. Tantra in Bengal: A Study in its Origins, Development and Influence.
Calcutta: Naya Prakash, 1978.
Bhandarkar, R. G. Vaisnavism, Saivism and M inor Religious Systems. Benares:
Indological Book House, 1965.
Bharati, Agehananda. The Tantric Tradition. London: Rider and Co., 1969.
Bhattacarya, S. C. Principles o f Tantra. The Tantratattva with an introduction
by A rthur Avalon and Baroda Kanta M ajum dar. Edited by A rthur Avalon,
3rd ed. Madras: Ganesh and Co., I960.
Bose and Haider. Tantras: Their Philosophy and Occult Secrets. Calcutta; Firm a
K. L. M ukhopadhyay, 1973.
Brhatsücipatram. Vol. 4. Part I and II dealing with T antra m anuscripts in the
Vira (Darbar) Library. Compiled by Buddhisägara Sarmä. Kathmandu:
Vlrapustakälaya, 1964, 1969.
Bibliography 207
A Alinätha, 80
Allaia, 138n.23
Akulanätha, 99, 100 Avadya, 28
Akulendranätha, 184n. 157 Avadhüta Räma, 28
Aghora, 21, 111 Aväkhal, 23
Aghoraéivâcârya, 139n.24 Avijita, 80
Angiras, 38, 114 Asitängabhairava, 107
Ajakarria, 111
Ajaramekhalä, 80
Ajätman, 112 Ä
Ajitä, 37, 69, 80, 150n. 181
Ajeéa, I9ln.231 Ä dinätha, 28, 83, 90, 99, 100
A iiahäsa, 70 Adisakti, 64
Atlkäla, 28 Änandagiri, 18
Ananta, 107 Änandamekhalä, 80
Anantam ekhalä, 80 Äpasiam ba, 145n. 119
Anantavira, 107 Ä m ardaka, 123
Anädt, 28
Aparam ekhalä, 80
Aparäjitä, 37, 113, 150n.l81
u
Abhinavagupta, 5, 7, 13, 14, 30, 32, 38,
U ttarapitha, 182n.l34
39, 40, 50, 54, 63, 66, 76, 79, 86,
Uditäcärya, 144n.88
117, 124, 138n,23, 139n.25,
Unm attabhairava, 109-10
140n,28, 145n. 120, 155n.246,
Upamita, 144n.88
183n, 144
Umäbhagavati, 186n.l82
Amara, 70
Ulkäpuri, 23
A m arapäda, 71
A m aranätha, 80
Amalesa, 108 0
A m rtänanda, 67, 74
Amsaphala, 113 OAkârapIîha, 182n.l34
Amsumati, 108 Odiyänanätha, 99
210 I ndex
G J
Gapaveksa, 99, 100 Jadabhärata, 28
G andham alya, 91 JayantI, 39
Garuda, 40, 41, 125, 152n.214, Jayaratha, 7, 12, 51, 55, 62, 65, 76, 79,
212 Index
Y Visalalocana, 109
Visalaja, 108
Yasaskara, 10 ViSvanatha, 139n.24
Yam unäcärya, 16 Visva, 112
Vi?pu, 33, 40
V lranatha, 28
R Vrk§anatha, 83, 90
V rddhanatha, 70
R aktä, 79
Vetala, 30
Ratisekhara, 83
Vairagya, 28
Räjasekhara, 18
V airocana, 35
Ränä, D hana Saniser, Jaftgabahädur,
95
Räm änandanätha, 138n.23 L
Räm änuja, 16
Rudra, 10, 19, 23 L akulapapi, 107
Rudra couple, 71, 74 LakuliSa, 19,21,107; as founder figure,
Rüpasiva, 50, 98 19, 21, 25, 144n.87; place of birth
Rodhinabhairava, 108 and caste of, 23; as possible author
of the Pasupatasutra, 21; possible
date of, 20, 144n.88; as reformer,
V 24
Lak§m ldhara, 155n,248
Vakrä, 88 Lorenzen, D., 17, 30
Vakresvari, 88
Van Kooji, K. R., 95
Vajrakäyä, 113 s
Varadevanätha, 70, 80
Varäham ihira, 21, 26 £akty&nanda, 79
Vägbhäva, 71, 74, 178m 110 Sankaracarya, 10, 16, 29-30
Vägesvari, 71, 74, 178n.ll0 Sahkhadharin, 111
Väma, 39 Sarikhapala, 130
Väränasl, 6 Sam bhunatha, 5, 86
Vikaräla, 28, 107 Sarva, 21
ViktaçU, 80 Sasahkasekharabhairava, 108
Vijayä, 37, 113, 150n.l81 Sa&kanti, 108
Vijähutl, 69 Sasibhu$ana, 108
Vijita, 69 Sandilya, 70, 118
Vijjambä, 80 Sastrl, H., 95
Vidya, 111 Sikhadevl, 91-92
Vidyädhipati, 111 Sikhasekharabhairava, 108
Vidyänanda, 70, 72, 78, 79 SikheSa, 109
Vidyesäna, 113 Siva, 24, 25, 37, 40, 103, 113, 115; as
Vidyolka, 111 early sectarian god, 4; identified
Vindhya, 80 with G aruda, 41; incarnations of,
Vim alabhairava, 69, 80, 108 20, 25; as Kapalin, 29; as source of
Virüpàkçabhairava, 108 all Agamas, 10, 129; as supreme
Index of Titles
A Am buräsitantra, 114
Arpêumadâgama, Î97n.23
Ailreyatantra, 115 A rkam apitantra, 112
A k fit antra, 112 Arthasaratantra, 118
Agnipuräpa, 8, 40, 95 Ardhalocanatantra, 108
Aghorästratantra, 159n.283 A vadhütatantra, 115
Aghorïdàmaratantra, 172n.78 Aéritânandatantra, 115
Aghoresïtantra, 159n.283 Aàvaplutatantra, 109
A ghorefvaratantra, 159n.283 Asitângabhairavatantra, 34, 35, 36, 45
Ankaprasnatantra, 114 46, 53, 105, 107, 321, !97n.22
Acalasritantra, 112
A çaloçiçlïsQsQmhilâ, 109
A jatatantra, 122 Ä
Ajitatantra, 34
A tharvakatantra, 122, 154 Àgamarahasyastotra, 143n.59
A tharvayäm ala, 106, 121, 155n.246 Ätharvatantra, 102
Atharvaàiropaniçad, 24 Änandatantra (Änandasäsana,
A dvayatantra, 35 Anandaàâstra, Anandädhikära-
Anantatantra, 185n.l77, 198n,25 tantra, Anandâdhikâraêasana,
Anantabhäskarasarphitä, 109, 118 Änandesvara), 34, 38-39, 52, 55,
A nantam ata, 118 109, 125, 198
Anantavijayatantra, 34, 109, 118, Änandabhairavatantra, 9
198n.25
Anacalatantra, 109
Anäm akatantra, 13
I
A nityatantra, 109
Idötantra, 121
A ndhakatantra, 122
Indrayämala, 154
Aparajitatantra, 34, 120, 200n.47
A m rtatantra, 109, 198n.25
Arnftanandatantra, 109 Ï
A m oghatantra, 108
A m bikätantra, 115 Uänayämala, 106
Index 217
Kälikäpuräna, 8 Kramapüjana, 76
Kälikulatantra, 76, 79, 16ln. 12 Kramarahasya, 76
KälUantra, 11Î Kramasadbhäva, 179n.ll6, 181 n. 121
Kälottaratantra, 118, 200n.45 Kramastotra, 75, I7 9 n .ü 6 , 182n,142
Käsyapasanihitä, 152n. 216 Kriyäkälagutfottara, 41
Kiranagama, 185n. 177, 199n.34 Kriyäkram adyotikä, 139n.24
Kubjikätantra, 141n,49 Kriyäsäratantra, 175n.95
K ubjikänityähnikatilaka, 6, 78, 95, 96, Kridäghoresvaritantra, 159n.283
I72n.77 Krodhakarahkinisamhitä, 108
Kubjikäm atatantra, 6, 65, 68, 78, 83, Krodhabhairavatantra, 34,45, 53, 105,
8 7 ,9 2 ,9 5 ,9 6 ,9 8 , 106 175n.95;age 108, 121
of, 48, 84, 87; relation to Pasci- Krodhabhairavasütra, transm ission of,
mäm näya, 171n.77; relation to 108
p if has, 52, 55, 120 Krodham älinisamhitä, 108
K um kum atantra, 34 Krodhasamvarttakasarphitä, 108
Kurâhgâki'ïtantra, 109 Krodhograsarphiiä, 108
Kulakridävatära, 81 K?udrätantra, 109
Kulacüdämanitantra, 48, 141 n.49,
Î55n,248, 250
KH
Kutapahcäsikä, 163
Kulapüjana, 76
Khacakratantra, 48, 122
Kulapradipa, 185n.l78
Khahjinimata, 98, 99
Kulamülävatära, 195n,7
Khadgarävatantra, 34
Kulayogatantra, 185n.l77
Khecarivijayat antra, 111
Kuiaratnapahcaka, 172n,77
Kularatnamälä, 54, 85, 187n. 189,
195n.8 G
Kularatnamäläpancakävatära, 96
Kularatnoddyotatantra, 50, 62, 81, 82, Gayakärikä, 2 1
83, 84, 88, 90, 92, 172n.77 Ganatantra, 185n.l77
Kulasära, 48, 155n.250 Gapeiayämala, 155n.244
Kulärpavatantra, 67, 141n.49 Gamatantra, 39
Kuloddisatantra, 48 Gambhirasarphitä, 109, 119
Kuverayämala, I54n,243 Garudapuräpa, 28, 40, 41
Kusatantra, 115 Garbhaprakaranatantra, 115
Kusum äittantra, 115 Gäthäsaptaiati, 26
Kuhakatantra, 34 Gämini, 112
Kürmapuräna, 10, 17, 18, 23, 24, 25 Gudakatantra, 53
Kr ?nakat antra, 111 Gunamrtaiantra, 114
Kr$naghoresvaritantra, 111 Guptatantra, 141n.49
Kolahalatantra, 114 Gu/ywojfcrfl/fln/rfl, 48, 111, 112, 122
Kaulajnänanirn ay a, 65, 95 Guhyasamäjatantra, 102, 195n,6
Kauiasära, !75n.95 Guhyasiddhi, 163n,23
Kaulikämata, 185n. 177 Grdhrapüfisarphitä, 118
Kram akändakram ävati, 139n.24 Gojikatantra, 115
Kramakeli, 180 n. 116 Goraksasarphitä, 55, 92, 95
Index 2 Ï9
Paksirajatantra, 34 PH
Pañcabhütatantra, 34, 152
Pañcarajatantra, Í85n. 177
Phetkaraikakfaratantra, 111
P añcam ftaíantra, 35, 106, 152n.223;
transm ission of, 112
Pañearthabha$ya, 21
B
Pañcálikatantra, 115
Pañjarimata, 185n.l77
P atadrum aíantra, 34 Bahurupatantra, 46, 118, 159
Padamálátantra, 120 Bindukapalatan tra, 107
P addkfitantra, 109 B indukutiratantra, 108
Parampadaiantra, 107 Bindutantra, 48, 122
Parasarrihitá, 109 Binduvijayatantra, 108
Paratantra, 89, 97, 98, Í98n.25 Bindusdratantra, 34, 114, 199n.34
Par átriméikáviv arana, 81 Bimbatantra, 112, 114
Paramrtatantra, 114 Bim batilakatantra, 114
PaScimatanira, 6 Bijatantra, 197n.22
Pñtalavijayatanira, 108 Bijabhedatantra, 34
Páramesvaratantra, 185n. 177 B fhatsam hitd, 21
Par ame svar amata, 99 B rhadbrahm asamhitd, 152n.216
Pásaghni, 112 Brahmayamala, 30, 36, 49, 51, 85, 105,
Pasupatasütra, 21, 23, 25 106, 110, 114-15, 121, 154n.243,
P kubhairavitanira, 121 155; age of, 48; classification of
Piñgalátanira, 115, 121 Agam as in, 43; cult of, 6; eight
Picumata, 115, 169n,57 Bhairavatantras according to, 45-
Picusára, 115 47; Kaula typology of, 169n.57;
Pitaghoritantra, 111 pi}ha classification in, 51, 52, 53;
Piyu^ara tnam ahodadhi, 184n. 157 relation to Yamalas, 54; T antras
Purakalpa, 107 according to, 47
Pu$patantra, 35 Brahmavaivartapurana, 8
Püjodayatantra, 109 Brahmasutra, 16
Pürvakárapágama, 148n. 168 Brahmandavijayatantra, 185n. 177
Pecikdmaia, 115 Brahmatantra, 121
Pefikátantra, 121 B rdhm ikdtantra, 121
222 In d e x
A A tita, 62
A uttara, 182n.l40
 dhyâtm ikasâstra, 31 Auttarakram a, 182n,142
Âgamas, creation of, 195n,7; individ Auttaräm näyaf 76
uality of, 13; as members o f canon, Avali, 62
14; worship of, 158n.280
Aghorl, 27-28 B
Akula, 61, 62, 75, I61n.7
A kulam ahadarsana, 184n.l57
Âmnàyas, according to Candh 95; Bahurupatantras, 46, 47
according to M aheévarânanda, Bark, birch, 92
179n.ll î; definition of, 66; devel Behaviour, antinom ian, 22, 24, 26, 28
opm ent of, 87; goddesses of, Beef, 6
179n,l 13; num ber of, 67, 173n.86; Beyond Mind (unmana), 60
relation to the breaths, !73n.85; Bhairavacarya, 30
relation to the four ages, 67, Bhairava (sect), 17
173n.84; relation to phases of m an Bhairavasrotas, relation to Kaula-
ifestation, 174~75n.91, 92; Sàkta tantras, 59
and Saiva character of, 176- Bhairavas{aka, 45,46,53, 175n.95; con
77n. 100; system of classification, tents of, 45-46; as an independent
66 group, 47
Anakhya, 75, 78, 181n.l21 Bhairavasiatfaka, 43, !59n.283
Ananda, 62, 70 Bhairavatantras, 19, 26, 42-49, 86, 102,
Anantasrotas, 112 110, 123, 153n.225, 154n.234,
Âtjava, 84 197n.21; according to the §KS, 32;
Anuttara, 162n. 12 as authority for Kashmiri Saivites,
Anuttaram naya, 76, 182n,144 5; cults of, 30; as Daksiiyatantras,
Anuttaram rtakula, 181n. 117 46-47; development of, 33; early
Anutîaratrikakula, 181 n. i 17 loss of, 7; followed by Kapalikas,
Anuttaratrikakulakrama, 75 29; persistence of, 42; rejected the
Àrhata, 102 Vedas, 9; rejected by Puranas, 10
A$ta$iabheda, 47, 175n,95 Bhairavatantras, sixty-four, 44, 114,
228 In d e x
Dâkinîyâmalas, 119
D aksa, sacrifice of, 19
H
Dakçina, 16, 66
H eart, lotus of, 40
Dakçinâmnâya, 66, 74
H rddntarabheda, 175n.95
Dak$i$atantras> 34-35, 41-49
Dakçinasiva, 16
Dakçipasrotas, 36, 42, 46, 47, 49,
155n.246
Qëmaratantras, I97n.23 Ida, 38
230 In d ex
M N
Magic, 37
M ahakaula, 62 Nàkula (sect), 18, 197n.21
M ahakarunika (sect), 104, 196n.l0 Nameless (anâmaka), 60
M ahapasupata (sect), 5, 18 Nâtha (sect), 28
M ahartha, 179n.U l, 182n.l42 Nepal, 37, 38, 41; Kaula Tantrism in,
M ahavratadhara (sect), 17, 18 12-13; Kubjikâ cult in, 6,12,87-88;
Index 231
Trikatantras, 32, 67, 83, 154n,236, 37; rejected by Puraiias, 10; similar
156n,251, 17In.76, 187n.l89; to Daksinatantras, 41
origin of, 84, as purvdrdha, Vamasiva (sect), 16
174n.87; relation to Daksina- Vamasrotas, 42
tantras, 49 Varridsrama, 10, 19, 24
Trikaula, 72 Veda, 4, 16, 19, 23,24,25, 31,102, 107,
Trisulamaridala, 156n.251 114n.36; authority of accepted by
the Puranas and rejected by the
Agamas, 9-10
u
Vidydpitha, 49, 98, 104, 105-6, 110-19,
Udanaprana, 60 157n.254, 199n.35; according to
Udyogakramama!in/, 73 the BY, 44; as Agama and its
pedestal, 158-59n.280; contents of,
Ugra (sect), 18
53-54; lCaula character of, 169n,57;
Uncreated, Wheel of, (ajacakra), 64
relation to Manirapifha, 54, 105;
Uitardmndya, 66, 72-73, 1 7 9 n .ill;
Trika as belonging to, 54, 55
relation to Kdlikula, 75, 76-77, 78,
Vyomdrdhamafhika, 68, 69
86, 183n.l45
Utiarapifha, I84n.l62
Upagamas, 33, 148n.l69 Y
Yamata, 45, 46, 74, 106, 124, 140n.35,
w
142n,49, 155n.246, 175n.95; as
Wheel, primary and secondary belonging to mixed current, 104;
(m ukhya and anucakra), 64 definition of, ¡44, 199n.35; as an
W oman, 22, 63-64 independent group, 45,47; rejected
by Puraijas, 10
Yamalasiva, 16
V Yaugika, 62
Yoga, 8, 70
Vagisatantras, 46 Yogi, 65
Vaunala (sect), 20, 102 YoginI, 65; as teacher, 64,65; mouth of,
Vai§ijavasamhitas, 13, 40 64, 169n.57
Vakula (sect), 17, 19 Yoginikula, 65, 162n.l7, 170n.64,
Varna (sect), 16, 17, 66, 197n.2i I71n.71
Vdmdcdra, 39 Yoginimatasdra, as Purvdmndya, 68
Vdmatantras, 19, 25, 26, 34, 36, 38, 39, Yogimyamalas, 106, 119
43, 83, 86, 140n.35, 154n.236, 237; Yuganathas, 62, 79, 163n.23; their
according to the JY; according to consorts and disciples, 80; relation
the Mrgendragama, 36; according to dmndyas, 74; relation to Trika,
to SK.S, 44; concerned with siddhi, 79; worship of, 76, 79-80, 82