In this welcome addition to the illustrious Collection Indologie of the Institut français de Pondichéry and Ecole française d'Extrême-Orient, R. Sathyanarayanan presents a first critical edition and annotated translation of the Prāyaścittasamuccaya, a compendium (samuccaya) of passages on methods of expiation (prāyaścitta) compiled by Trilocanaśiva, a disciple of the famous Saiddhāntika exegete Aghoraśiva. Trilocanaśiva lived in south India in the twelfth century ce and is known to have composed a number of other important Saiddhāntika texts. His Prāyaścittasamuccaya contains a total of 828 verses, which are primarily quotations from texts ranging from the c. sixth-century Puṣkara-Pārameśvara to the c. twelfth-century Uttara-Kāmikāgama describing atonements for faults of commission or omission by Śaiva initiates. The quotations are not identified by Trilocanaśiva, but an earlier text also called the Prāyaścittasamuccaya and compiled by a certain Hṛdayaśiva, which survives in a Nepalese palm-leaf manuscript dated 1157–58 and which, unlike Trilocanaśiva's compendium, presents and identifies entire chapters from a wide range of Śaiva texts, includes many of the passages found in Trilocanaśiva's text, thus allowing their source texts to be identified. A transcription of this manuscript is included as a lengthy appendix to the edition and translation of Trilocanaśiva's text.

The edition and translation are preceded by a rich introduction by Dominic Goodall which not only introduces the text, its author and its edition, but also succinctly contextualizes the text's teachings within the broader history of Śaivism by showing how its concerns pertain to a Śaiva social reality quite different from that which can be inferred from earlier Śaiva texts. The Śaivasiddhānta of Trilocanaśiva was no longer a private cult for male brahmin initiates but had become a more public and hierarchized religion centred on monasteries and rich householder gurus. Goodall's introduction also includes a survey of the role of women in the Śaivasiddhānta inspired by their relative prominence in Trilocanaśiva's work, which he sees as evidence of greater participation by women in initiatory Śaiva religion as a reflex of Śaivism acquiring a broader social base. Reflecting on the history of prāyaścitta itself, Goodall shows how the earliest Śaiva tantras taught rules (samaya) but not expiations for their trangression (prāyaścitta) and suggests that the subsequent inclusion of prāyaścitta in Śaiva tantras was the result of influence from Brahmanical dharmaśāstras, adducing the absence of a similar development in Buddhist tantric works, whose authors would have been less in the thrall of the prescriptions of dharmaśāstra, in support of this theory. Towards the end of the introduction, Goodall contextualizes Trilocanaśiva's Prāyaścittasamuccaya within the history of temple worship in south India. In somewhat complex discussions about the authorship of the Prāyaścittavidhi (which he shows to have been falsely attributed to Aghoraśiva) and parallel passages in the Prāyaścittasamuccaya and Uttara-Kāmikāgama, he shows that Trilocanaśiva probably knew the Uttara-Kāmikāgama, one of the first south Indian Temple Āgamas, and so was aware of the new Saiddhāntika concern with public religion, despite concerning himself only with rites related to individuals.

Sathyanarayanan's edition is based on eleven south Indian witnesses, none of which appears to be especially old. These are eight palm-leaf manuscripts in Grantha script, one paper transcript in Grantha script and two paper transcripts in devanāgarī. A tentative stemma of the witnesses is provided, but because of contamination between them readings have not been adopted through simple mechanical means. The text is presented together with a positive critical apparatus and a register identifying the sources of the passages quoted by Trilocanaśiva. In general the text is well supported by the witnesses; conjectural emendations are only occasionally required. A pāda-index is included as an appendix, as in all editions in the Collection Indologie. In this age of e-texts such indexes seem unnecessary, but in this case, by also indexing the pādas of Hṛdayaśiva's Prāyaścittasamuccaya, it functions as a useful concordance.

The faults for which atonements are prescribed in Trilocanaśiva's Prāyaścittasamuccaya include many that are not specific to initiates of the Śaiva Siddhānta, such as those committed while eating, bathing, urinating, excreting, having sex or menstruating (the detailed list of the latter being the inspiration for Goodall's analysis of the role of women in the Śaiva Siddhānta), those pertaining to the use of books or those incurred by being born or dying, and the five great sins or mahāpātakas. There are also offences relating to the practices of Saiddhāntika initiates such as mistakes in the performance of worship, defilement of a śivaliṅga and misbehaviour in the presence of the guru. The methods of atonement themselves may consist of ways of purifying objects that have been defiled or of purifying the person who is at fault. The latter most commonly involves the repetition of a mantra, usually one of the five brahmamantras; of them, it is the Aghora mantra which is most commonly prescribed. Indeed, the Aghora mantra is a cure-all; verse 496 says “As for when a combination of many sins arises, one should recite the Aghora mantra until one's mind is set at ease”. Bathing, fasting and consuming the pañcagavya, the five products of the cow, are also regularly prescribed as methods of atonement, and details of different types of fast and the preparation of the pañcagavya are given in the text. The specific types of fault described attest to a rigidly stratified and sectarian society: penances are graded not only according to which level of initiate is at fault, but also according to their varṇa and, where the fault involves interaction with someone else, that person's religion, varṇa and jāti.

This is an excellent work firmly in the tradition of the publications of the Institut français de Pondichéry and Ecole française d'Extrême-Orient in that it presents a truly critical edition of an important but recently neglected text together with an annotated translation and introduction that use the meticulous text-critical scholarship involved in producing the edition to draw conclusions about the historical context of the text's production.