What is Shaivism?

What is Śaivism? (excerpted from my doctoral dissertation)

            Śaivism is the name scholars use for a religion that flourished in India from about the beginning of the common era through to the Muslim period, when it was subsumed, along with Vaiṣṇavism, under the umbrella of Brāhmanism (vaidika-dharma),[1] giving rise to Hinduism in the 15th century.

            This first sentence already demands an explanation of terms and of methodology before we can focus in on our specific subject. Firstly, as an Indologist, my primary methodological orientation is that of philology. As a corollary of that orientation, I hold that the application of terms of identity to historical and religious agents that were not used self-referentially by those agents are invalid in our enterprise, that of historically accurate description and analysis of Indian culture. Another way of saying this is that as a philologist I hold that “emic” identifiers are valid, not “etic” ones. Back-projecting complex and heavily weighted terms from one historical era to a previous one in which they were not used creates obscured perception, selective reading of the evidence, and significant errors.

            Therefore I hold that Hinduism came into existence in the 15th century. Though the exact Anglicism “Hinduism” was not used until 1816 (by Rammohan Roy; see Lorenzen 1999: 631), “Hindu” was used as a self-designation by Indians beginning in the 14th century, and as a specifically religious self-designation throughout the 15th century and beyond.[2] Since “Hindu” in the 15th century denotes much the same collection of religious phenomena as “Hinduism” does in the 19th, as shown by Lorenzen (1999), I am not concerned about the addition of an –ism to form an English abstract noun, either here or in the case of Śaivism, to which we will come.

The term “Hindu” emerges specifically in contrast to “Muslim,” and in the context of Muslim rule. Lorenzen anticipates my argument when he says, “In practice, there can be no Hindu identity unless this is defined by contrast against such an Other. Without the Muslim (or some other non-Hindu), Hindus can only be Vaishnavas, Saivas [sic], Smartas or the like” (Lorenzen 1999: 648). Therefore, I argue that Śaivas, and hence Śaivism, existed until the 14th or 15th century; and this is not merely terminologically true, but also institutionally, for a substantial number of Śaiva institutions, both literal (temples, maṭhas, etc.) and figurative (lineages, schools, etc.) had been destroyed by the Muslims by this period, both violently and through lack of patronage.

The next question that emerges in the explication of the first sentence is, What is it about Śaivism and Vaiṣṇavism that allowed them to be subsumed under the umbrella of Brāhmanism, unlike Buddhism and Jainism, and thereby give rise to Hinduism? The answer to that question also serves as a definition of what I mean by Brāhmanism. We may define the latter simply in two dimensions: valid sources of knowledge and social structure. In the first instance, Brāhmanism (derived of course from brāhmaṇa) is a religion (dharma) based on the Veda and the traditions and conduct of those who know the Veda, as clearly stated in both the Gautama-dharma-sūtra (1.1-2: vedo dharmamūlaṃ tadvidāṃ ca smṛtiśīle) and the Manu-smṛti (2.6ab).[3] Therefore, we can also call Brāhmanism by the name Vedism (vaidika-dharma). In the second instance, that of social structure, we can define Brāhmanism in terms of the varṇāśrama-dharma. Medhātithi, a 9th-10th cen. authoritative commentator on the Manusmṛti, tells us that a territory fit for Vedic rites (yajñiyo deśaḥ) is one in which a conquering king who observes brāhmanical practice (sādhvācaraṇa) enforces the varṇāśrama-dharma, imposing the status of caṇḍāla (untouchable) on all those indigenous to the region who do not fit into that system.[4] Therefore, the reader will be unsurprised to learn, the reason that Śaivism and Vaiṣṇavism were able to be subsumed by Brāhmanism in the late medieval period is precisely because a) both acknowledged the authority of the Veda to some degree, and b) they did not, in that period, transgress the norms of the varṇāśrama-dharma.[5] But this situation was not always the case for either, and most especially for Śaivism, as well shall see. In fact, the latter had been undergoing a process of increasing Veda-congruence, with the progressive etiolation or separation out of its transgressive Śākta elements, since at least the 12th century.

Thus, even if we were using an etic definition of “Hinduism” and applying it to the early medieval period, I argue that it would be inappropriate to include Śaivism under that rubric. The reasons for this will be given below; here I will confine myself to the generalization that the broad sweep of the history of Śaivism may be seen as a millennium-long process (roughly, 400 to 1400 ce) in which an explicitly anti-brāhmanical religion, that initially characterized itself in terms of a wholesale rejection of the brāhmanical worldview, was slowly brought within the confines of the brāhmanical system of values, to the point where it could rightly be lumped together with the latter under the name Hinduism.

Finally, I intentionally called Śaivism a religion in the first sentence, by which I mean “an institution consisting of culturally pattern interactions with culturally postulated superhuman beings” (Spiro 1966). In the context of South Asia, where religious boundaries are often blurred in various ways, we must also ask what distinguishes a religion from other religions. I hold that in the period in question (500-1300 ce), Śaivism possessed the traits I associate with a distinct and self-contained religion, which are: 1) a body of texts that belong to that system and no other; 2) authoritative teachers consecrated in that system and no other, propagating pan-Indian doctrine from within a transregional ecclesiastical hierarchy; 3) the fact that the system itself makes an effort to distinguish itself from others;[6]  4) competition with other religious systems, including especially the claim to offer definitive salvation above and beyond them.[7] Space does not permit me to explore these four points in detail here, but the reader is referred to the oeuvre of Alexis Sanderson, which provides ample evidence. One such piece of evidence may be mentioned briefly: the doctrine of the unity of the Śaiva scripture (over and against the Vaidika), typified in the assertion that the entire Śaiva canon is a single complex statement made by God (Sanderson 2005: 23).[8]

ABSTRACT of the dissertation:

The present work comprises a detailed study of specific terms of discourse in the pre-twelfth century sources of esoteric “Tantric” Shaivism, both scriptural and exegetical, some of which are still unpublished and others of which are published only in the original Sanskrit. As a dissertation in South Asian Studies using the philological method, the primary purpose of the study is to ascertain the range of meanings of certain technical terms of great importance to the theology and practice of the Śaiva religion, namely āveśa, samāveśa, and śaktipāta. The work focuses on both the independent meaning and the intersection of these key terms, incorporating also the terms dīkṣā and vedha in the latter endeavor. The intersection of these terms constitutes a complex set of relationships, a nexus of ideas that lie at the very heart of the Śaiva tradition and which, due to the latter’s widespread influence, came to be important in Tantric Buddhism and later forms of Hinduism as well. This thesis contends that samāveśa—meaning the fusion or commingling of one’s self with the energy of one’s deity and/or the consciousness of one’s guru—is the key term that distinguishes Tantric Shaivism from mainstream (esp. Vaidika) Indian religion. This constitutes a reinterpretation and overcoding of the earlier meaning of āveśa, i.e. self-induced controlled possession by a deity.

Samāveśa is important to all forms of Shaivism, whether dualistic and ritualized (the Siddhānta) or nondual subitist charismatic forms (the Kaula). This thesis further contends that a philological study of samāveśa and related terms like śaktipāta demonstrates that religious experience (or evidence thereof) was considered central and indispensable to initiatory Shaivism throughout the medieval period. Śaktipāta was requisite to receive the basic level of initiation, and in the Kaula branch of the tradition, samāveśa denoted forms of religious experience that were necessary for aspirants to demonstrate in order to receive higher-level initiations. The former term is still commonly used in many Hindu communities today to designate a “spiritual awakening” or initiatory experience that is transmitted by a qualified guru.

Part One of this work is a comprehensive overview of the nature and structure of the Śaiva religion, providing important context to what follows. Part Two studies the key terms of (sam)āveśa, śaktipāta, etc. in a) early Sanskrit literature generally, b) Śaiva scriptures, and c) the abundant exegetical literature based on those scriptures.


[1] By Brāhmanism we mean the body of practice and belief that claims to derive its authority from Śruti and Smṛti (where smṛti effectively means vedavit-parigraha).

[2] Cynthia Talbot has shown that Andhra inscriptions use the phrase “Sultan among Hindu kings” beginning in 1352 ce—our first documented use of the word “Hindu” in an Indian language—in response to Muslim incursions into the region beginning in 1323 (1995: 700, cited in Lorenzen 1999: 652). This corresponds to one of the earliest Muslim uses of the word Hindu in a specifically religious sense (‘Abd al-Malik ‘Isami, 1350; Lorenzen 1999: 653). In the early 15th century the Apabhraṃśa author Vidyāpati contrasts Hindu and Turk dharma, i.e., religion (hindū dhamma and turaka dhamma), as does Kabīr a couple of generations later (Lorenzen 1999: 650-52). Note that a) from the beginning, the term Hindu is defined in contradistinction to a Muslim “other” and b) both “Hindu” and “Turk”, often paired in late medieval poetry, are originally ethno-geographical terms that became religious designators.

[3] References provided by Sanderson. See also Medhātithi’s Manubhāṣya on 2.6, where he specifies that whatever other practices are followed by those who correctly enact the rites of śruti are also Vedic by extension, for the primary reason to think any practice valid is that it is adopted by knowers of the Veda (prāmāṇyakāranaṃ mukhyaṃ vedavidbhiḥ parigrahaḥ).

[4] Manusmṛti-bhāṣya on 2.23, cited at Sanderson 2005: 400; see also Sanderson 2009: 41n1, where many citations of inscriptions are given to prove that a righteous (dhārmika) king is precisely one who rigorously imposes the varṇāśrama-dharma on his subjects.

[5] A passage from a lost text much cited by conservative exegetes in Śaivism’s classical period (the 9th-11th centuries) reads: varṇāśramācārān manasāpi na laṅghayet | yo yasminn āśrame tiṣṭhan dīkṣitaḥ śivaśāsane | sa tasminn eva saṃtiṣṭhec chivadharmaṃ ca pālayet (See Sanderson 2007b: 231n1). This injunction came to be followed by virtually all Śaivas beginning in the 13th century.

[6] For example, Śaiva exegete Bhaṭṭa Rāmakaṇṭha distinguishes “those of our own religion” (samāna-tāntrikas) from the views of other religions (tantrāntaras). See Watson et. al. 2013: 40.

[7] These four criteria emerged in discussions with Prof. Sanderson while I was his M.Phil. student. For primary sources in which this sense of a discrete religious identity is evident, see (e.g.) Kṣemarāja’s Pratyabhijñā-hṛdayam and Rāmakaṇṭha’s Paramokṣanirāsakārikā-vṛtti (Watson et. al. 2013).

[8] This doctrine is exemplified in an interesting quote from Vairocana, Rājaguru of King Nirbhara and author of the Pratiṣṭhā-lakṣaṇa-samuccaya, where he describes himself as a lion whose eyes are the Siddhānta-tantras, whose huge sharp fangs are the Gāruḍa-tantras, whose tongue and hair are the Vāma- and Dakṣiṇa-tantras, and whose massive claws are all the Bhūta-tantras, for these are the principal divisions of the Śaiva canon (siddhānta-dvaya-dṛk ca gāruḍa-bṛhat-tīkṣṇogra-daṃṣṭraś ca yaḥ | jihvā + + ca keśarāṇi + tathā savyāpasavyāgamau | vistīrnākhila-bhūta-tantra-nakharo vairocano keśarī |) Reference provided by Sanderson. Note that Vairocana also uses the phrase śaivāntaḥpātinaḥ, “those who fall within the Śaiva religion” (e.g. 2.169).