Madan-Bhasma (Shiva Turns to Ashes) Litograph cca. 1890
Alexis G. J. S. Sanderson is an Indologist and an Emeritus Fellow of All Souls College in the University of Oxford. After taking undergraduate degrees in
Classics and Sanskrit at Balliol College from 1968 to 1971, he spent six years in Kashmir studying with the Śaiva scholar and guru Swami Lakshman Joo. From
1977 to 1992 he was University Lecturer in Sanskrit and a Fellow of Wolfson College. In 1992 he was elected to the Spalding Chair of Eastern Religions and
Ethics and became a Fellow of All Souls.
Sanderson is a scholar of Sanskrit and of Indian religions, especially of Shaivism and esoteric Śaiva Tantra. In his own words, "The understanding of
Śaivism can only aspire to objectivity if it includes a sincere effort to see how things are in the subjective perception of its practitioners. One has to
be able to enter into the spirit of their world, to be with them intimately, to see what they are saying and why they are saying it, to go beneath the
surface of their texts. There has to be empathy."
Excerpts taken from In Honoris Causa: Essays in Honour of Aveek Sarkar, edited with a foreword by John Makinson (Allen Lane, 2015), pp. 155–224
Read Previous Parts
The Śaivas’ Inclusivist View of Their Own and the Vaidikas’ Religion
As for the Śaivas, they likewise seem to undermine the unity of Hinduism by insisting not merely on the validity of their own scriptures but also on
their superiority to the Vaidika scriptures and indeed to the scriptures of all non-Śaiva systems. This was already so in the earliest known Śaiva initiatory system, that of the Pāñcārthika Pāśupatas of the Atimārga, whose ascetic initiates were to see themselves as
having severed all ties with the Vaidika religion, leaving behind their former obligations to the gods and ancestors to focus their devotion on Śiva/
and it continued to be so in the Mantramārga, even though this later form of the religion, in evidence from about 450–550 onwards, expanded the community
of the initiated beyond that of ascetics, important though they continued to be, by opening up initiation to householders, allowing them, indeed requiring
them, to remain in this status after they had received initiation.
It might appear, therefore, that Śaivism was as much distinct from, and opposed to, the religion of the Vaidikas as the latter was to the former and as
both were to Buddhism and Jainism. However, while the Śaivas thought their scriptures superior to the Vaidikas’ and the Vaidikas thought their own
superior, the two traditions’ views of each other were not symmetrical, not at least where the Śaivas of the Mantramārga were concerned, these alone having
left us adequate evidence of their views on this issue. For while most Vaidika theoreticians during this period condemned the Śaiva scriptures as false,
of the Mantramārga held unanimously that the Śruti and Smṛti of Brahmanism
are universally and uniquely valid in their own sphere, that of prescribing the conduct and religious observance obligatory for persons in their identity
and unmarried members of the caste-classes (varṇāśramadharmaḥ), and that as such they are man’s sole means of valid knowledge both
of all actions (karma) that benefit and harm the soul’s destiny in the domain of recurrent incarnation
(samṣāraḥ) and of the nature of the consequences of these actions, from the rewards of the heavens to the tortures of the hells.
Nor did they deny the reality of Brahmanism’s goal, that of liberation
(mokṣaḥ), offered to those who aspired to escape recurrent incarnation through knowledge, unmotivated obedience to ritual injunctions, or both.
They denied only that it was ultimate, holding that true, definitive liberation lay beyond it and could be reached through Śaivism alone, by undergoing
initiation in the presence of the
Maṇḍala of Śiva (śivamaṇḍaladīkṣā) and then following the Śaiva ritual and meditative disciplines, or, in the case of those prevented by
incapacity or social
responsibilites from taking up those disciplines, notably their royal patrons,
through initiation followed by fervent devotion manifest in support of the Śaiva
religion and its institutions.
Nor was the validity of the Vaidika scriptures irrelevant to the Śaivas within their own world of Śaiva rites and belief. Indeed Sadyojyotis, who is much
the earliest of the commentators on the Śaiva scriptures whose works have reached us – he flourished sometime between the second half of the seventh
century and the beginning of the ninth, probably no later than the first half of the eighth –
insisted that the defence of the validity of the Vaidika scriptures (Śruti and Smṛti)
is essential to a belief in the validity of the scriptures of the Śaivas themselves.
He offered two cogent reasons. The first is that if the Vaidika scriptures were not the source of valid knowledge in their domain, as the Buddhists and
Jains insist, then Śaivism’s central claim that it frees the initiate from the cosmic hierarchy of the levels of incarnation would be empty. Śaivas must
believe that the Vaidika scripural corpus is valid because the initiation rituals prescribed in their scriptures and performed by Śiva himself through
their officiants bring about the progressive freeing of the soul from a cosmos created and maintained for and by the enactment and consequences of
meritorious and demeritorious actions, and these actions, as we have seen, are held by the Śaivas to be good and bad on the authority of the Veda alone.
The second reason – and it is this that is more weighty in an assessment of the lived relations between Śaivism and Vaisika orthopraxy – is that the
Mantramārga’s scriptures themselves insist that the rules of the Vaidika socio-religious order are binding on Śaiva initiates. Śaivas were subject to that
order at the time
of their initiation, and to the extent that they chose to continue to live within it after their initiation they were enjoined to continue to adhere to its
That the Śaiva scriptures do indeed require this conformity is well illustrated in the following passage, much cited by the commentators:
So he should not transgress (na laṅghayet) the practices of
his caste-class and [Vaidika] discipline (varṇāśramācārān)
even in thought (manasāpi).
He should remain (tiṣṭhet) in the discipline (āśrame)
in which he was when he was initiated into the Śaiva religion
(dīkṣitaḥ śivaśāsane) and [at the same time] maintain
the ordinances of Śiva (śivadharmaṃ ca pālayet).
There is another respect in which the Vaidikas’ view of Śaivism and the Śaivas’ view of Vaidika religion were asymmetrical. For while the Vaidika tradition
made no attempt to justify its validity in Śaiva terms, the Śaivas, in their eagerness to establish themselves in what was by that time a fundamentally
brahmanical society, attempted to persuade the orthoprax that the Śaiva corpus was valid not only because it recognized the Vaidika ordinances as binding
on all including the Śaivas themselves but also by attempting to undermine the Vaidikas’ attacks on the legitimacy of their religious practices by pointing
to the abundant evidence of the promotion of the worship of Rudra or Śiva, by then considered one and the same, that is found both in Śruti texts and in
the secondary Vaidika scriptures.
Thus in his commentary on the Mṛgendratantra the tenth-century Kashmirian Saiddhāntika Bhaṭṭa Nārāyaṇakaṇṭha cites the presence of such practices
in the traditions of all four Vedas. The passage on which he is commenting is the narrative introduction to the Tantra. In the hermitage of Badarī,
and other sages install an image of Śiva and undertake asceticism before it. The god Indra comes to the hermitage and asks them why they are not following
the religion of the Veda (codanādharmaḥ). They reply that the method of propitiating Śiva with asceticism that they are following is indeed
Vaidika and point out (v. 6) that the Veda contains Mantras whose deity is Rudra and procedures for causing him to come into one’s presence. In his
commentary on this verse
Bhaṭṭa Nārāyaṇakaṇṭha elaborates, citing a six-month-long ascetic procedure for
the summoning of Rudra into the propitiator’s presence taught in the now lost Rudrakalpa that was a supplement (Pariśiṣṭa) of the
Śrautasūtra of the Kāṭhaka Yajurvedins, the use of the long Yajurvedic litany known as the Eleven Rudras (rudraikādaśinī saṃhitā), probably its
recitation while one inundates the Liṅga
(rudrābhiṣekaḥ), a practice still current among the Taittirīya Yajurvedins in the
Śiva temples of South India, sacrificial procedures using Mantras and chants of the Ṛgveda and Sāmaveda found in the Ṛgvidhāna
and Sāmavidhāna, and procedures
for the propitiation of Rudra found in the Atharvavedic corpus.
Similarly, in his commentary on Sadyojyotis’s Mokṣakarika Bhaṭṭa Nārāyaṇakaṇṭha son Bhaṭṭa Rāmakaṇṭha turns to the corpus of secondary
Vaidika scriptures, arguing that these contain abundant historical evidence that Śaivism was accepted by venerable figures of remote antiquity whose
standing as men learned in the Veda is beyond question. He cites the rule that Śaivas must remain in their castes and life-disciplines, not transgressing
the ordinances of those institutions even in thought, and then addresses the Vaidikas as follows: 
So this [teaching of Śiva] is not a forbidden form
of religion (na pāṣaṇḍatvam) even from your point of view [as Vaidikas].
This is because it does not conflict with the Vedas, and because there is [Vaidika]
scriptural evidence that it was accepted by men learned in the Vedas.
In the Purāṇas, the Mahābhārata, and the like we learn that Śveta, Upamanyu,
and other great sages undertook religious practice within this [teaching of Śiva].
In the [Mahā]bhārata we learn that Nara,
Nārāyaṇa [=Arjuna and Vāsudeva], and Aśvatthāman did the same, in the words
‘The god that you [Aśvatthāman] have worshipped in an anthropomorphic image
in every age those two have worshipped in the Liṅga,’ and also that it was by propitiating Śiva
that the Lord Vāsudeva achieved his goal in Suvarṇākṣa, as is related in the verse
‘O Kṛṣṇa, you will be the man most dear to me in the world;
and the whole world will turn to face you [in adoration]. Of this there is no doubt,’
[and Paraśurāma received the axe with which he slew Kārtavīrya.]
Moreover in the Smṛtis we have references to such pious acts for the benefit of the public (pūrto dharmaḥ)
as that of establishing a temple [of Śiva, as in] ‘He who makes a temple of Śiva, built with baked bricks’ 
and ‘in pious acts for the benefit of the public one should know [that the reward
Then there is the evidence of our own eyes in the form of the Pṛthukeśvara [of Pṛthu], the Rāmeśvara
[of Rāma], and [many] other [Śivas that have been installed in temples by exemplary Vaidikas in ancient times].
Furthermore, the Veda confirms the validity of the teaching of Śiva in such Upaniṣads as the Śvetāśvatara
and in Mantra-texts such as the Atharvaśiras. So none of the [three] faults that would entail
the invalidity [of the Śaiva scripture] from your point of view applies:
there is no disagreement [concerning the omniscience of Śiva, the creator of our scriptures],
there is no lack of proof [of their validity], and they have not been adopted by a small minority.
So Mantramārgic Śaivism, while claiming to transcend Brahmanism by offering true liberation, was nonetheless closely tied to it. It was the Vaidika world
rather than the Buddhist or the Jaina through which Śaivas rose to salvation; and on their path thereto they continued to be bound by its rules, adding
their Śaiva obligations rather than substituting them for the Vaidika. They looked, moreover, to the Vaidika scriptural corpus to provide proof of the
validity of their own scriptures. The evidence adduced as proof is not cogent, since none of it refers to the specific practices of initiatory Śaivism; it
refers only to forms of propitiation that had long been part of Vaidika observance. Nonetheless, the attempt reveals the concern of the Saiddhāntikas to be
considered valid by the adherents of the Vaidika tradition that they claimed to rise above.
Ganesha Holds Court cca.1800 Walters Art Museum
The Properly Śaiva attitude of Śaivas towards their Vaidika rites
Now this extension from the purely Śaiva domain of the ascetic in the Atimārga into the Vaidika domain of the Śaiva householder added in the Mantramārga
opened the door to a process of Śaiva-Vaidika hybridization, in which rites of both kinds were maintained and co-ordinated without a proper sense of their
distinctness. The Śaivas’ theoreticians, aware that this development had the potential to produce a blurring of the boundary between the two domains that
would undermine the faith of Śaivas in the independence and supremacy of Śaivism, ruled that while initiated Śaiva householders were thus subject to two
bodies of injunction – both the Śaiva and the Vaidika – their attitude towards adherence to the latter was to differ fundamentally from that of the
Two passages of early Saiddhāntika Śaiva scripture much cited by the commentators clarify this attitude. The first of these is in the Sarvajñānottara. In the context of its prescription that only persons in two of the four Vaidika disciplines, those of the unmarried scholar and
the married householder, may be consecrated as Śaiva officiants, it tells us that the Śaiva should maintain his Vaidika observances after initiation but
without believing that they are fully real. He is to do them but without subjective commitment. He should not think that by accommodating Vaidika rites
beside the Śaiva he brings about a doubling of the benefit
that he will derive or that if he were to omit them he would damage that benefit. He is to see himself as acting in this regard not for his own advantage
but so as not to undermine through a pointless non-conformity the Vaidika order within which Śaivism is embedded. 
The second passage is in the Mataṅgapānameśvara. This clarifies the matter in the language of the Mīmāṃsā by saying that though the initiate
his Vaidika duties, here called ‘the mundane observance’ (laukikācāraḥ), he must not conceive of them as ancillary elements (aṅgam) of
his Śaiva observances, which is to say, as elements without which those observances would be incomplete and therefore inefficacious. 
The Śākta-Śaiva attitude towards Vaidika observance among initiates
The view seen in the Sarvajñānottara that the Śaiva should conform to Vaidika injunction only for the sake of others tended not to be emphasised
in the later Saiddhāntika exegetical literature, which seems to be more eager to stress conformity with Vaidika injunctions than to justify this from a
properly Śaiva standpoint. But the Śākta-Śaiva scholars of such traditions as the Krama and Trika, whose ritual practice was further distanced from Vaidika
norms than the Saiddhāntikas’, preserved a strong emphasis on Śaivism’s transcendence,  even arguing that the true reason for conformity with Vaidika observance was
spiritual immaturity. Thus Abhinavagupta’s pupil Kṣemarāja (fl. c. 1000–1050) asserts that one should continue to perform the Vaidika ritual of
Juncture of the day (brāhmī saṃdhyā) before one venerates it in the Śaiva manner
(śaivī saṃdhyā) only so long as one’s mind is in thrall to one’s constructed social
identity as a member of a caste. In support of this position he cites a passage from the Saiddhāntika Svāyambhuvasūtrasaṃgraha to the effect that
ritual of the Juncture is optional but the Śaiva compulsory. But he leaves out that part of the passage in which we learn that the option applies only in
the case of ascetics. His reason for doing so is evidently that he wishes the text to support the view that one should aspire to drop the Vaidika elements
of one’s ritual programme even if one is not an ascetic, seeing persistence in these as symptoms of a contracted state of mind that all Śaivas,
householders included, should strive to transcend.
Nonetheless, it is unlikely that this Śākta-Śaiva view of the accommodation of Vaidika ritual had a negative impact on the integration of such Śaivas
Vaidika society. Kṣemarāja speaks here of an ideal adjustment within the largely
private domain of the Śaiva householder’s daily ritual rather than a wholesale rejection of conformity in the socio-religious domain. And this is in line
with other elements of transcendence that set the Śaivas engaged in Bhairava and goddess worship apart from the Saiddhāntikas. Thus, for example, when a
Saiddhāntika participated in a collective meal with other initiates he was not to sit in a line that contained persons of a caste other than his own. If he
did so he was to do penance, its severity determined by the degree of the caste difference, being doubled if the contaminator was a Vaiśya and trebled if
he was a Śūdra. But according to the
, the principal scripture of the non-Saiddhāntika Dakṣiṇa system of
Bhairava worship, initiates may be arranged in separate rows on such occasions only according to their rank as initiates. They must never separate
themselves in accordance with the distinctions between the castes in which they were at the time of their initiation (prāgjātiḥ). They must see
themselves as having become equal
members of a single ‘caste’ of Bhairava (bhairavīyā jātiḥ), and they must know that if they make any reference to the former caste of an initiate
they will be guilty of a sin that will lead them to hell. In short, says the text, if one wishes to attain one’s goal, be it salvation or Siddhi, one must
be free of all caste discrimination (avivekī). But there is no
suggestion that this transcendence of caste distinctions should be applied on the socio-religious level in such matters as marriage. The fact that
Abhinavagupta cites the Sarvajñānottara with approval for its view of the necessity of general conformity, and does so even in the context of an
argument that the considerations of relative purity and impurity that dominate Vaidika behaviour are subjective (pramātṛdharmaḥ), makes clear that
no such subversive transcendence was envisaged.
The same is the point of a Kaula verse cited by the Śākta-Śaiva Jayaratha in this context and also by the Smārta Aparāditya, though in that case with the
contrary purpose of demonstrating the insincerity of the Śaivas’ conformity:
He should be a Kaula in private (antaḥ kaulo), a Śaiva in outward appearance
(bahiḥ śaivo), but a Vaidika in his mundane observances (lokācāre tu vaidikaḥ),
keeping the essence [of his religion hidden behind these two outer layers],
just as the coconut fruit [keeps its milk within its flesh,
which in turn is
enclosed by its hard outer shell].
In the same spirit the Tārātantra, a text from the later East Indian Śākta tradition, tells us concerning the orgiastic gathering of Śākta
initiates known as a ‘circle of Bhairavī’:
Once the circle of Bhairavī has commenced (ārabdhe bhairavīcakre)
all [the participants, whatever their] caste-classes are the best of brahmins (sarve varṇā dvijottamāḥ).
But as soon as it has ended each returns to his or her separate station.
If a person, being deluded, makes distinctions of caste (jātibhedaṃ...
karoti) within the sacred circle,
then without doubt he will fall into a terrible hell from which it will be hard to escape.
Kama Shiva circa 1820
A Śaiva-Vaidika socio-religious hierarchy under royal authority
The Śaivas, then, advocated a two-tiered Śaiva-Vaidika socio-religious system, with the Vaidika subordinate to the Śaiva. To achieve this new order they
exerted themselves to secure the support of royal patrons, without which no such aspiration could be realised, and though the details of how they proceeded
to this end in particular cases are mostly inaccessible to us now, epigraphical and other evidence reveals that they were successful in their endeavour in
many parts of the subcontinent and Southeast Asia.
It is at least clear from their prescriptive literature that they expanded their ritual repertoire to support this endeavour, by introducing a form of
initiation for rulers that not only promised the benefit of liberation at death and, in this life, the heightening of the initiand’s temporal power, but
also departed radically from tradition by exonerating the beneficiary from all the time-consuming and arduous ritual duties that were the usual consequence
of initiation. A king who was devoted to Śiva (paramamāheśvaraḥ) could receive the prestige of initiation
without any of its inconveniences, being required thereafter only to maintain the
support of the faith that is the duty of any lay (uninitiated) devotee (śivabhaktaḥ, upāsakaḥ).
The Saiddhāntikas also developed a Śaiva version of the royal consecration ritual (rājyābhiṣekaḥ) to be given to a king after he had received this
Śaiva initiation. The Naimittikakriyānusaṃdhāna of Brahmaśambhu, completed in ad 938/9,  the earliest surviving guide to the Saiddhāntika Śaiva rituals, states that the
purpose of this ceremony is to qualify the king for his office as the guide and guardian of the system of the castes and disciplines. This is none other than the role assigned to him by purely Vaidika authorities;  and accordingly the Mantra recited at the climax of this empowerment, as the
water of consecration is poured, is not Mantramārgic but rather the long-established verse text of the (royal) consecration Mantra prescribed for this
purpose by Varāhamihira in the first half of the sixth century on the authority of the Older Garga.  But since it is as an initiated Śaiva that the king is to assume this role, it
is evident that the socio-religious order entrusted to his care is not just that envisaged by the Vaidika authorities but rather the expanded religion that
comprised both the Vaidika and the Śaiva traditions. For the Śaiva literature elsewhere requires him to ensure that the strata of this complex of
injunction are maintained in the proper order of relative authority, with the Vaidika subordinate to the Śaiva, promising him that to do so
will guarantee him a long reign and the prosperity of his kingdom, and implying thereby that failing to do so will have the opposite consequences.
We see all this clearly enunciated in a passage of the Mohacūrottara, one of a number of as yet unpublished scriptures of the Śaiva Mantramārga
known as Pratiṣṭhātantras.
These texts, as their name indicates, are concerned to regulate the practice specific to the class of Mantramārgic Śaiva officiants known as Sthāpakas, who
specialized in the installation (pratiṣṭhā) of temples, their images, and monasteries, and in the planning of settlements and royal palaces, and
the layout of the towns around them. After prescribing the proper disposition of the habitations
of the various castes around the palace of an emperor (mahārājādhirājaḥ) it says:
Tradition declares that the king is the protector of his subjects.
Therefore it is right that he should protect the caste communities
and ensure that they are instructed in their duties, each according to its station.
The sources that convey these duties are Śruti, Smṛti, Purāṇa, and the [Śaiva] scriptures (āgamāḥ).
If the king abides by these he enjoys a long reign.
[The correct order of authority in which they should be applied is as follows.]
The Vedas [comprising both Śruti and Smṛti] take precedence over the Purāṇas,
and the [Śaiva] scriptures take precedence over the teachings of the Vedas.
There is the common [Vaidika authority of Śruti, Smṛti, and Purāṇa] (sāmānyam),
and then there is the special (viśeṣam). The Śaiva [scriptures] (śaivam) are the latter.
[So] the learned should not doubt their authority when they find that they conflict with [a Vaidika injunction].
The all-knowing [master] should adjudicate each case objectively [by this criterion].
Given the plurality of scriptural authorities, whenever there is a question as to which of two [conflicting]
statements takes precedence, he should adopt that which has been taught by Śiva.
He should reconcile the two, whether self-sufficient or depending for the understanding of their meaning on
[examination in the light of] other sources of the same kind, related sources, and [, where they fail,]
learned exegesis, by applying such modes of reasoning as presumption (arthāpattiḥ).
Understand this, O Indra, and thereby attain the ultimate bliss.
When the king understands the duties of religion in this way his realm will always prosper.
This model, in which the Vaidika ordinances are maintained under the aegis of Śaivism, might be suspected to have been more ideal than real were it to rest
on this prescriptive evidence alone. However, it is in harmony with what is conveyed by the historical records of the period. They certainly do not support
a position that the rise of Śaivism during these centuries led to a corresponding decline in the hold of the Vaidika order. On the contrary, they point to
a renaissance in that sphere; and they show that Śaiva kings were active in promoting it.
A good part of the inscriptions that have come to light from this time consists of thousands of copperplate charters in which kings, including those who
were Śaiva, have recorded their establishing Vaidika brahmins in their territories through grants of tax-exempt land, thus fulfilling one of the principal
duties imposed on them by Vaidika scripture, extending the penetration of Vaidika observance, while facilitating the administration of their territories
and promoting agricultural development.
Further, numerous kings, Śaivas prominent among them, have been commended during this period, particularly at its beginning, for having imposed the system
of castes and disciplines (varṇāśramadharmaḥ) in their newly established
kingdoms, this frequently being presented as a restoration after a period of decline.
Nor was this promotion of Brahmanism by Śaiva kings restricted to the socio-religious level. It extended on occasion to the commissioning of the horse
sacrifice (Aśvamedha) and other solemn (Śrauta) Vaidika rituals. These were associated with the acquisition and celebration of sovereignty; but their
also a signal of a king’s desire to be seen as an exemplary supporter of the Veda, dedicated to the revival of Vaidika religion in its entirety.
So the epigraphic record indicates that there were numerous Śaiva kings throughout our period who fully accepted their role as the guardians of the Vaidika
social order and thus conducted themselves in matters of religion in the
manner envisaged by Brahmaśambhu and the Mohacūdottara. Indeed, in many
cases their panegyrists have portrayed them as zealous propagators of that order, praising them for their efforts to reverse the decline in the hold of the
Vaidika order on society.
Moksakārikāvrtti on v. 146ab.
3.82.18; text and translation in brackets suspect at this point.
Source not located.
170.33[c]d. The category of pious action termed pūrtam or pūrto dharmaḥ comprises such actions as establishing fountains, wells,
step-wells, reservoirs, dams, and gardens, and planting fruit trees and the like; the installing of deities; and the building and renovation of
monasteries. See, e.g., Varāhapurāṇa 168.21; 170.33–58.
These three faults are specified by Kumārila in ŚlokavārttikaCodanāsūtra 133 as reasons for rejecting the Buddhists’ argument that their
claim that the Buddha, the author of their scriptures, was omniscient is proved by the existence of an unbroken tradition to that effect from his
time to the present. For a detailed discussion of this verse and the three that follow and their interpretation by Kumārila’s commentators see KATAOKA 2011, pt. 2, pp. 358–366.
, Liṅgoddhārādiprakaraṇa N f. r5–v5; P pp. 97–98.
, Caryāpāda 2.2–7b.
See, for example, Abhinavagupta, Tantrāloka 4.221c–253; 15.162c–179b; Parātriṃśikāvivaraṇa p. 266, l. 4–p. 267, l. 7, edited and
translated in SANDERSON 2005, pp. 111–112.
Kṣemarāja on Svacchanda 2.14c: ‘This veneration of the Juncture (sandhyāvandanam) is done with the Mantras of Śiva, but first it
is done with the Mantras of [one’s] Veda. That is the duty of those in whom there lingers the deep-seated mentality of identification with the
caste that was theirs [before they entered the casteless “caste” of Bhairava (bhairavajātiḥ) through initiation] ( anivṛttaprāgjātivāsanaiḥ kāryam). The rest should do it with the Śaiva Mantras [alone], immediately after they have completed the ritual
bath. As has been taught [by Śiva in Svāyambhuvasūtrasaṃgraha 7.9cd]: “He may or may not do the Vaidika [Sandhyā ritual]; but it is
compulsory that he should do the Śaiva”.’ On the early Śākta-Śaiva attitude to caste, and other Vaidika dualities, such as
that of the pure and the impure, see SANDERSON 1985, pp. 198–205 and endnote 69; SANDERSON 2009a, pp. 292–297;
and SANDERSON 2009b.
Trilocanaśiva, Prāyaścittasamuccaya p. 25: ‘He should always avoid when eating sitting in the same line (ekapanktih) as persons
of a different caste (bhinnajātibhih). A brahmin who eats unknowingly with persons of a Ks˙atriya, Vaiśya, or Śūdra
caste and abandons his meal in the middle as soon as he realizes ˙ this, should declare this, and then [as his penance] repeat
the Aghoramantra ten, twenty, or thirty times ˙ respectively. If he realizes [what he has done only] after the meal has been finished
[he should repeat it] one, two, or three hundred times respectively.’
Tantrāloka 4.248–251. On Abhinavagupta’s doctrine that purity and therefore impurity are subjective and not real properties of things
see SANDERSON 2013a.
Rājānaka Jayaratha, Tantrālokaviveka on 4.251ab. For Aparāditya’s version see Yājñavalkyasmṛtiṭīkā, p. 10, ll. 12–13.
Tārātantra quoted in the Sarvollāsatantra, p. 80. I conjecture ārabdhe for ārambhe in the first verse.
f. 103r–v2: ‘On the tenth day of the bright fortnight of the first month of autumn in the 860th year of the king of the Śakas I, disciple of the
abbot of Mattamayūra, have declared this procedure for initiation that adheres to the teaching of the Dviśata[kālottara] and should be
given by a Guru to Gurus initiated in his own lineage to terminate his holding of his tenure of office.’
f. 74v1 [4.118]: ‘I shall now teach in addition the consecration ceremony to empower an initiated king as the guide and guardian of the castes and
āśramāṇāṃ ca gurubhāvāya bhūpateḥ
| yo ’bhiṣekavidhiḥ so ’pi procyate dīkṣitātmanaḥ).’ See SANDERSON 2009a, p. 255, fn. 593.
See SANDERSON 2009a, p. 255, fn. 594.
See Bṛhatsaṃhitā 47, especially 47.55c–70 Varāhamihira himself says only that the Mantra was ‘taught by the Muni’ (47.51d:mantro ’tra munigītaḥ). It is his commentator Bhaṭṭa Utpala who in a comment on this statement identifies the Muni as the Older Garga: munigīto muninā vṛddhagargeṇoktaḥ.
. On the canon of these texts see SANDERSON 2014, pp. 26–27 and fn. 100.
The Vedas here must be understood to include Smṛti, that is to say, the Dharmaśāstras, if this
statement is not to contradict the preceding assertion that the (non-Śaiva) sources of the knowledge
of duty are not only Śruti (the Vedas in the narrow sense) and Purāna but also Smṛti.
Mohacūrottara ff. 21v6–22r2 (4.275–281).
. Space prevents me from setting forth here the epigraphical evidence of the engagement of Śaiva kings in these efforts to promote Brahmanism in
their kingdoms. It has been presented in detail in my forthcoming Śaivism and Brahmanism.