The reuse of the iconography of Hayagrīva in texts and images* Elisa Freschi May 8, 2015 Contents 1 On Reuse 1.1 State of the art . . . . . . . . . . . . . . . . . . . . . . . 1.2 Some basic dichotomies . . . . . . . . . . . . . . . . . . 1.2.1 Simple re-use vs. adaptive reuse . . . . . . . . . 1.2.2 Quotation, reference and interlanguage . . . . . . 1.2.3 Recycling, conventional re-use and new life re-use . . . . . 2 3 4 4 5 5 2 Introduction on Hayagrīva 2.1 Major changes in the iconography of Hayagrīva . . . . . . . . . . 6 6 3 Textual history of Hayagrīva and his iconography 3.1 Origins . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Mahābhārata and Purāṇas . . . . . . . . . . . . . . . 3.2.1 Hayagrīva vs. Madhu and Kaiṭabha . . . . . 3.2.2 Hayagrīva vs. Hayagrīva . . . . . . . . . . . . 3.2.3 Iconography of Hayagrīva in the Purāṇas . . 3.3 Hayagrīva and Vaḍavāmukha . . . . . . . . . . . . . 3.4 Hayagrīva in Tantric Buddhism . . . . . . . . . . . . 3.5 Hayagrīva in the Pāñcarātra Saṃhitās . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 8 9 9 10 12 13 14 16 4 Hayagrīva in art history (apart from the area of the Vijayanagara empire after Veṅkaṭanātha) 17 4.1 Geographic diffusion of images . . . . . . . . . . . . . . . . . . . 19 5 Hayagrīva in the Hayaśīrṣa Saṃhitā 20 5.1 Descriptions of Hayagrīva in the Hayaśīrṣa Saṃhitā . . . . . . . . 21 * Research for this article has been financed by the FWF M-1437 project and the FWF V-400 project. I am grateful to Katharine Apostle for having reviewed my English. 1 6 Hayagrīva as supreme deity 6.1 Veṅkaṭanātha’s mentions of Hayagrīva and his Hayagrīvastotra 6.1.1 The Hayagrīvastotra . . . . . . . . . . . . . . . . . . . . 6.1.1.1 Hayagrīva’s dissociation from Lakṣmī . . . . . 6.2 Attestations possibly independent of Veṅkaṭanātha . . . . . . . 6.2.1 Sudarśanasūri . . . . . . . . . . . . . . . . . . . . . . . . 6.2.1.1 Other Vaiṣṇava schools after Veṅkaṭanātha . . 6.3 Post-Veṅkaṭanātha standardisation . . . . . . . . . . . . . . . . 6.3.1 In Icons . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.1.1 Yoga-Hayagrīva . . . . . . . . . . . . . . . . . 6.3.1.2 Lakṣmī-Hayagrīva . . . . . . . . . . . . . . . . 6.3.2 In copyists’ and editors’ maṅgalas . . . . . . . . . . . . 6.3.2.1 Authors’ maṅgalas . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 22 24 26 26 26 26 27 27 27 27 28 30 7 Conclusions on the post-Veṅkaṭanātha diffusion of the standardised Hayagrīva 30 8 Various types of reuse 30 9 Abbreviations 32 Abstract The present articles discusses the iconography of Hayagrīva as found in sculptures, paintings and in the texts describing them. It argues that determinate elements of this iconography have been the object of a process of adaptive reuse, in which they were consciously reused with a new purpose and that this process took place around the time of the well-known Śrī Vaiṣṇava theologian Vedānta Deśika (1269– 1370), who adopted Hayagrīva as the supreme deity within his pro-Vedic agenda. The study on Hayagrīva is preceded by a methodological introduction on reuse in general and on the terminology adopted in the article. 1 On Reuse Reuse as conceived in this article1 is a phenomenon involving a reused item (be it concrete material or a concept), a reusing agent, a community of experiencers and an interruption in use (one cannot speak of the reuse of something which has been continually used, without discontinuance). In this sense, reuse is just not an accidental phenomenon independent of human agency, which could be better described by terms such as “hybridisation” or “continuation and changes”.2 1 This short introduction is the result of the research leading to the panel on “Re-use at the Borders of South Asia: Himālayas and South India” at the first conference of the EAAA (Olomouc, September 2014: http://www.ea-aaa.eu/news/2014/04/first-eaaa-conference-september-25-27-olomouc-czech-republic). I am grateful to the participants of the panel for their feedback before, during and after it. 2 Cf. the definition of reuse in Hegewald and Mitra 2012: “Re-use is a conscious and selective process in which existing elements are borrowed or salvaged and taken out of their former environment in order to be applied to a new context, or they are left within their old milieu but filled with new meanings, or they get manipulated and react to new external influences. […] For this to happen, a disruption or change has to take place, favouring a confrontation with 2 1.1 State of the art The topic of reuse in the history of art and architecture has been the theme of important research in western and South Asian art. As for the former field, due to the apparent omnipresence of reuse especially in medieval art and architecture, important studies have focused on both specific instances and on the general topic of the reuse of materials and of its underlying typology. Todisco 1994 (in Italian) focuses on the Roman reuse of Greek motifs but offers also important insights into the reuse of Roman elements in the Middle Ages (see, e.g., fn. 47, p. 96 on the reuse of ancient statues to represent Christian saints) and an interesting bibliographical list of references for the topic of reuse (Todisco 1994, pp. 268–269) in classical and medieval settings. Andreae and Settis 1983 (in Italian, with contributions in English, German and Spanish) outlines a typology of medieval reuse and offers interesting methodological suggestions. AA.VV. 1999 (in Italian) is a comprehensive study in two large volumes of the topic of reuse in the early Middle Ages and includes a vast bibliography on the topic. The topic of reuse in South Asian art has also been the object of several studies focusing on concrete instances of reuse, especially in the context of Islamic reuses of Hindū materials and motifs. Of particular relevance are the seminal volume Asher and Metcalf 1994 and the volume 59 of Archives of Asian Art, which is a special issue dedicated to the topic of reuse with a special focus on Hindū-Islamic reuse. Its Introduction, Patel 2009, summarises previous studies on the topic of reuse in South Asian art, highlighting the alternate fortunes of the case of Hindū-Islamic reuses, also in connection with the sensitive issue of their political background. This historicopolitical concern is clearly present throughout the special issue, since “Underpinning this project was the belief that the phenomenon of reuse furnished a unique entrée to specific historical processes, such as the establishment and consideration of new rulerships, changes in religious and nonreligious iconography and their reception through time, and the emergence of new ways of building” (Patel 2009, p. 2). Less common are the studies focusing on more theoretical issues connected with reuse, among which one needs to mention Julia Hegewald’s studies, especially Hegewald and Mitra 2012, which further offers the specific advantage of employing reuse as a concept bridging history of art and social sciences due to its applicability also to the field of political studies. The book does indeed focus also on cases such as the reuse of the Indian myth in Nehru’s rhetoric (Schöttli 2012) or of the term “Pakistan” in the Indian Muslim press (DiCostanzo 2012). More in general, the overarching idea is that reuse (found in the orthography re-use in the volume) is an answer to anxiety and offers a solution to it, both in politics and in art, since it may allow the permanence of motifs in a new garb, acknowledged by the dominant elite. The book also includes a chapter dedicated to a theory of reuse (Hegewald 2012) the contents of which have been partially sketched above and which also discusses the differences and affinities between reuse and other terms. On the one hand, terms such as hybridity, diffusion and osmosis are deemed to focus too much on the automatism of the phenomenon: “The term excludes the element of the human actor or agent, who is so important when it comes to re-use” (Hegewald and Mitra 2012, p. 39). On the other hand, terms such something different. […] Re-use only materialises when agents and actors make it happen. […] How there re-used items are perceived by various audiences depends on their relationship with the individual items that have been fused. Re-use can create anxiety and it can contribute to integration. Items and concepts which are based on re-use often persist longer because they establish continuity and involve different groups of society”. (2012, pp. 48–49) 3 as “ eclecticism”, Hegewald continues, presuppose that the authors investigating such phenomena are “conscious and sensitive to the element of selectivity when it comes to the combining of elements”, but “the basic term eclecticism treats all combined elements equally and does not establish structural hierarchies” between the artist who is reusing and that which is being reused (Hegewald and Mitra 2012, pp. 41–42). Further investigated terms (such as syncretism and pluralism) regard more directly the political sphere. 1.2 Some basic dichotomies I am myself not an art historian, but have long been working on the concept of reuse, both in texts and in images. This introduction is meant as an attempt to bridge the terminologies developed in both contexts (e.g., in Freschi 2015b, Bignami 2014, Freschi and Maas forthcoming and Hegewald and Mitra 2012). 1.2.1 Simple re-use vs. adaptive reuse Adaptive reuse is a concept developed in the history of architecture in order to discuss cases such as the adaptation of a pre-existent architecture in view of a new purpose (see Plevoets and Van Cleempoel 2013 and Plevoets and Van Cleempoel 2011). The dichotomy between simple3 re-use and adaptive reuse is presented in Freschi and Maas forthcoming. Simple re-use is dictated by sheer economic reasons. It is the kind of reuse which is only determined by economic and pragmatic reasons, e.g., when one buys a used car because it is the cheapest available option. In simple re-use, the artist does not want the audience to recognise the reused elements as such and the fact that they are reused is not an explicit asset of the new composition. A typical example is the re-use of Roman building materials in order to build the first Christian churches —assuming that it was exclusively or at least primarily due to the simple fact that re-used building materials were cheaper and more easily available than new ones. By contrast, adaptive reuse4 involves the intentionality of the artist performing it and the awareness of her or his audience. Thus, adaptive reuse implies an explicit underlining of the reused element. The artist wants the audience to recognise what is happening and the fact that the element has been reused is part of the value of the new composition. Contemporary history of art is full of such examples, in which a new frame gives a complete new meaning to reused material. See, for instance, Petr Motycka’s Christ made of old shoes. It is worth noting that simple re-use and adaptive reuse are not aut-aut alternatives, but rather two extremes on a grey scale. It is hardly the case that a reuse is just an instance of simple re-use, with no further value added: Reusing columns from a Roman building for a Christian church is probably also linked with the subtle idea of the new hegemony. Vice versa, adaptive reuse would often be unthinkable if it became too expensive. 3 Cristina Bignami has also suggested to avoid “simple” and speak instead of “linear” or “pragmatic” re-use (http://elisafreschi.com/2014/10/03/downcycling-and-pragmatic-reuse/). 4 The lack of hyphen hints at the fact that this is not just a repeated use, but something altogether new, with new purposes. 4 1.2.2 Quotation, reference and interlanguage The categories of “quotation” (literal or semi-literal and acknowledged reuse), “reference” (paraphrase, often unacknowledged) and “interlanguage” (floating ideas common to a whole cultural milieu) have been distinguished and discussed in Freschi 2015b. Bignami 2014 has suggested to apply them to the history of art. This might lead to the following applications: • the term quotation could cover cases such as Andy Warhol’s reuse of well-known works of art (notably the Mona Lisa) within his creations. In fact, in this case the reuse is acknowledged and the viewers need to be aware of the original painting for the mechanism to work. • the term reference could cover cases such as the reuse of a content without a specific form, as in the abovementioned case of Motycka’s Christ which reuses the motif of the crucified Christ although it does not reuse a specific representation of him. • the term interlanguage could cover cases such as the diffusion of Corinthian columns throughout the Roman Empire. Their use outside of Greece was in fact no longer linked to a specific geographic area and readers were not reminded of a single building the style of which would have been reused. They were just the shared common language for prestige buildings.5 1.2.3 Recycling, conventional re-use and new life re-use Julia A.B. Hegewald adopted a different terminology in her studies on re(-)use, starting from Hegewald and Mitra 2012, where she distinguished between: • recycling (when only the raw materials are re-used, e.g., while melting a statue) • conventional re-use (when the purpose is not changed, e.g., a temple is re-used again as a temple) • new life re-use (when the purpose is changed, i.e., a water tank is re-used as a prison) Concerning “recycling”, in order to avoid the ambiguity between cases in which the reused object is no longer recognisable (such as the gold reused from melted jewels) and cases in which it is still recognisable (see the example of Roman columns in PalaeoChristian churches mentioned above), one might want to add the subcategories of downcycling, for melted jewels and other cases in which the object is lost and only its material is reused, and upcycling for cases in which simple materials acquire a deeper meaning (as in the case of ready-made art).6 “New life re-use” seems quite close to what was called “adaptive reuse” above (section 1.2.1), stressing the extreme of the grey scale on the side of adaptiveness. “Conventional re-use”, by contrast, is not exactly tantamount to simple re-use, since the former points to the continuity of the purpose, whereas the latter points to the lack of awareness of the reuse by artist and audience. 5 By contrast, the reuse of the same Corinthian columns in Washington D.C. is a case of reference, since it did not represent the obvious way of building and it rather clearly referred to the classical model of ancient Greece, trying to evoke democracy and other classical ideas. 6 For these categories see McDonough and Braungart 2002. 5 I will now focus on the core of my article, namely the adaptive reuse of the iconography of Hayagrīva. My application to it of the terminology discussed so far can be read in section 8. 2 Introduction on Hayagrīva Hayagrīva is a minor pan-Indian deity but at a certain point (after 13th c.) and in a certain place (a region of the Vijayanagara empire excluding today’s Kerala and extending roughly across the area of what is presently the northern part of Tamil Nadu until Mysore in today’s Karṇaṭaka7 ) he becomes the supreme deity. In the following I will discuss the reasons for this major change and argue that it only happened in the 17th c. and due to precise historical reasons. The Buddhist history of Hayagrīva is not the main focus of my paper but I will keep it in the background for the sake of reference. 2.1 Major changes in the iconography of Hayagrīva The pan-Indian Hayagrīva is a minor avatāra of Viṣṇu and as such he is rarely represented. His main distinctive feature is the one which corresponds with his various names (Hayagrīva, Hayavadana, Hayaśīrṣan, Hayamukha, Hayāsya, Hayavaktra, Aśvaśīrṣan, Aśvamukha, Turagavadana, Vaḍavāmukha8 …), i.e., the fact that he has a horse’s head. In Tantric Buddhism, like other non-Buddhist deities, it evolves into a terrific deity (see below, section 3.4). From an iconographic point of view, the pan-Indian and minor deity Hayagrīva as we find it, for instance, in Khajuraho (see Fig. 1),9 is characterised as follows: head he has a horse’s head arms he has four or eight arms position he stands10 (it is worth noting that also the Buddhist Hayagrīva stands) attributes he has pan-Vaiṣṇava attributes. In his Khajuraho relief, he has a mace (gada) and the dānamudrā, ‘gesture of bestowing boons’, while the other two arms are mutilated. In other cases, he carries the conch, the wheel, the mace and only sometimes and in a later stage a book representing the Vedas mudrā abhayamudrā ‘gesture of non-fear’ or dānamudrā (also called varadamudrā)11 association he is connected with generic attendants. In Khajuraho, with two servants holding a fan (identified in Desai 1973, pp. 143–144, as āyudha puruṣas ‘personified weapons’)12 7 Just for practical reasons, I will refer to this area as the “Vijayanagara empire”, although it is not identical with the extension of the historical Vijayanagara empire, which was moreover only established in 1336. 8 This last name presents a specific problem, see below, section 3.3. 9 In the northern niche of the pradakṣiṇā-patha of the Laksmana temple. See Desai 1973, pp. 143–44, and Gail 2013, Figs. 4–5. For further similar depictions see below, section 4. 10 According to Desai 1973, the samabhaṅga posture is the most common one. 11 According to Desai, the first images of Viṣṇu in general usually displayed the abhayamudrā (Desai 1973, pp. 8–9), but the varadamudrā seems to be the second most common gesture (Desai 1973, p. 13). 12 According to Desai 1973, āyudhas become increasingly common. 6 Figure 1: The pan-Indian Hayagrīva at Khajuraho, Lakshman Temple, Sculpture 12 — The sculpture is located at the inner walls of the sanctum area. Source: Wikipedia The first characteristic remains constant in all representations of Hayagrīva. By contrast, the Vijayanagara and supreme deity Hayagrīva is characterised as follows: arms he has four arms position he sits (in padmāsana or rājalalitāsana, see immediately below) attributes he always holds the same Vaiṣṇava attributes, i.e., conch and discus in his upper arms, whereas the lower left hand holds a book representing the Vedas, at times together with a rosary mudrā the lower right hand performs the vitarkamudrā ‘gesture of reflection’ or another mudrā connected to teaching (e.g., the jñānamudrā or vyākhyānamudrā) association two depictions of the supreme deity Hayagrīva are known: the Yoga-Hayagrīva, who sits alone, and the Lakṣmī-Hayagrīva in which he is depicted in rājalal- 7 itāsana, with his divine spouse Lakṣmī sitting on his left knee Figure 2: Two recent popular images of Yoga- and Lakṣmī-Hayagrīva 3 Textual history of Hayagrīva and his iconography R.H. van Gulik (1935) and Kamala Elisabeth Nayar (1994) have traced an accurate history of nearly all occurrences of Hayagrīva in Sanskrit literature, so that in the following I will only highlight the elements which are relevant for the purpose of this article. 3.1 Origins Hayagrīva might be connected with the importance of the horse in the Vedic culture (and perhaps also with its importance in the Indo-European one, as argued in Van Gulik 1935, pp. 9–10).13 However, according to Suvira Jaiswal (Jaiswal 1981), Hayagrīva rather had a local origin, since it was a local demon of Assam connected with fever and was only later incorporated in Mantrayāṇa Buddhism and then in pan-Indian Hinduism in general and Vaiṣṇavism in particular. However, even the theory of his local origin would not be of any aid in explaining the major change discussed in section 2.1, given that Hayagrīva would not have had any initial connection with the relevant area of the Vijayanagara empire.14 13 On the antecedents of Hayagrīva in Vedic literature see Sridhara Babu 1990, chapter I. the Vedic vs. tribal origins of Hayagrīva, see Nayar 1994, chapter 2. For his connection with fever see also Sridhara Babu 1990, chapter II.3 (where the connection with fever and with Assam is discussed within the precinct of the Kālikā Purāṇa). 14 On 8 3.2 Mahābhārata and Purāṇas 3.2.1 Hayagrīva vs. Madhu and Kaiṭabha The occurrences of Hayagrīva in the Mahābhārata (henceforth MBh) have been neatly summarised in Van Gulik 1935, pp. 10–15, Sridhara Babu 1990, chapter II.2, and in Nayar 1994, chapter 3. Van Gulik notes that in different portions of the Mahābhārata one finds Hayagrīva connected with the recitation of the Vedas and that in MBh 12.335.43–69 a horse-headed Viṣṇu brings back the Vedas and kills the two asuras Madhu and Kaiṭabha, who had stolen them from Brahmā. The following is an excerpt of Hayagrīva’s rescue of the Vedas:15 Having entered the mythical stream, [Viṣṇu-Hayagrīva] performed the supreme Yoga | Performing the sound according to the rules of phonetics, he pronounced the Oṃ || 12.353.50 || The sound was resonant and went in each direction and was charming | It was in the whole earth and had all good qualities || 12.353.51 || Then, the two asuras made an agreement regarding the Vedas (presumably: regarding when to come back and pick them up) | and having thrown them onto the bank of the mythical stream, they ran whence the sound came from || 12.353.52 || At that point, the king god carrying a horse’s head, | Hari, grasped all the Vedas which had arrived at the bank of the mythical stream || 12.353.53 || He gave them back to Brahmā and went then back to his own nature | […] Then, the two [demons] sons of Danu, Madhu and Kaiṭabha, who did not see anything [as the source of the charming sound they had heard before] | went back quickly to the place [where they had left the Vedas] and they looked || 12.353.55 || Where the Vedas had been thrown, the place was empty! | […] Then there was a fight between them and Nārāyaṇa | The two Madhu and Kaiṭabha, whose bodies where filled with rajas and tamas, | were killed by the [now become] ‘Killer of Madhu’ (Madhusūdana, a name of Viṣṇu), who thereby pleased Brahmā || 12.335.64 ||16 15 My translation does not mean to improve on the preceding ones, but only to guide the readers through the myth of Hayagrīva. Thus, I have not dealt in detail with the meaning of yoga, etc., in this passage. 16 rasāṃ punaḥ praviṣṭaś ca yogaṃ paramam āsthitaḥ | śaikṣaṃ svaraṃ samāsthāya om iti prāsṛjat svaram || 12.353.50 || sa svaraḥ sānunādī ca sarvagaḥ snigdha eva ca | babhūvāntarmahībhūtaḥ sarvabhūtaguṇoditaḥ || 12.353.51 || tatas tāv asurau kṛtvā vedān samayabandhanān | rasātale vinikṣipya yataḥ śabdas tato drutau || 12.353.52 || etasminn antare rājan devo hayaśirodharaḥ | jagrāha vedān akhilān rasātalagatān hariḥ | prādāc ca brahmaṇe bhūyas tataḥ svāṃ prakṛtiṃ gataḥ || 12.353.53 || […] atha kiṃ cid apaśyantau dānavau madhukaiṭabhau | punar ājagmatus tatra vegitau paśyatāṃ ca tau | yatra vedā vinikṣiptās tat sthānaṃ śūnyam eva ca || 12.353.55 || […] atha yuddhaṃ samabhavat tayor nārāyaṇasya ca || || 12.353.63 || rajastamoviṣṭatanū tāv ubhau madhukaiṭabhau | brahmaṇopacitiṃ kurvañ jaghāna madhusūdanaḥ || 12.353.64 || 9 The connection with the Vedas, perhaps both with their oral (insofar as Hayagrīva is described as neighing the Vedas) and written form (although it is quite possible that what is rescued is still an oral version of the Vedas), is here very evident. It is also interesting that this version of the rescue of the Vedas is the only one which will be referred to in Pāñcarātra and in Viśiṣṭādvaita Vedānta texts. I will also come back to the motif of the ocean (see section 3.3), which is sometimes connected with Hayagrīva (although the word rasā might also mean ‘lower regions, hell’, its connection with tala ‘bank’ as well as the evidence derived from parallel texts seem to suggest the meaning ‘stream’). However, the reason why Viṣṇu took on a horse’s head of all things is altogether absent (unlike in the case of his Matsya or Varāha-avatāras, where the transformation had to do with the task to be accomplished). 3.2.2 Hayagrīva vs. Hayagrīva Another mention of Hayagrīva in the MBh has it figure as the name of a demon slaughtered by Viṣṇu: The two Madhu and Kaiṭabha have been slain by [Viṣṇu], who lies on the ocean | Having reached a different birth, Hayagrīva has also been slain in the same way || 5.128.49 ||17 Here, no details about the demon Hayagrīva and his misdeeds are present. His slaughter is simply compared to that of Madhu and Kaiṭabha, a detail which probably paved the way for the successive identification of the two events. In fact, the following one is a summary of the Hayagrīva story in one of its Purāṇic forms:18 A horse-headed Asura called Hayagriva once invoked Brahma and sought from him […] a boon by which he could be defeated by none other than another being who also had a horse’s head, also called Hayagriva. Such a creature did not exist […] The Devas did not know what to do. […] When they went to Vishnu, they found him taking a nap, resting his chin on his bow. Taking the form of termites, the Devas ate into the bowstring so that the bow shaft snapped with such force that it severed Vishnu’s neck. To save the headless Vishnu, the Devas sacrificed a horse and placed its head on his neck. Vishnu thus transformed into a horse’s headed being. […] Vishnu challenged Hayagrīva to a duel, smote him with his mace and restored the Veda. […] Brahma then restored Vishnu’s head. (Skanda Purāṇa). (Pattanaik 2006, pp. 86–87, s.v. Hayagrīva) There are various versions of this story (other versions, e.g., have Viṣṇu lose his head because of a curse and include no purpose for it, see Nayar 1994, chapter 3), and in any case the story looks somewhat strange, since: • it looks like an ad hoc explanation for Viṣṇu’s horse’s head 17 ekārṇave śayānena hatau tau madhukaiṭabhau | janmāntaram upāgamya hayagrīvas tathā hataḥ || 18 Further information on Hayagrīva in the Purāṇas can be read in Sridhara Babu 1990, chapter II.2. 10 • it looks like the conflation of three different stories, i.e., the slaughter of the demon Hayagrīva, the slaughter of Madhu and Kaiṭabha, who had stolen the Vedas, and the slaughter of the demon Hiraṇyakaśipu19 One could object that if one does not accept the Purāṇic versions of the story, it is difficult to make sense of Viṣṇu’s horse’s head. In fact, this might be due to either an ancient (Vedic or perhaps Indo-European) attribute of a deity, linking it to the horse because of the latter’s importance in the Vedic mythology, or the inclusion of a pre-existing deity20 in the Smārta pantheon through the device of turning it into an avatāra of Viṣṇu. Thus, in my opinion Hayagrīva is a (perhaps Vedic) deity, perhaps assimilated to Viṣṇu or always identical with him, and the horse’s head is linked to the importance of the horse in the Vedic culture. The same importance has led to the invention of several demons with horse attributes, until someone conflated the two stories into one, with added details from other demons’ slaughters (Madhu and Kaiṭabha and Hiraṇyakaśipu).A further possible conflation regards the transposition of Brahmā’s Vedic attributes onto Hayagrīva, discussed in section 4.21 In the Purāṇic version of the story, two elements are particularly relevant for the present research, namely Hayagrīva’s connection with the Vedas and the ambiguity of Hayagrīva, which is the name of both a demon and the god who defeated it. One might imagine that this demonic nature might have been part22 of the reason of the development of Hayagrīva as a terrific deity in Tantric Buddhism (see below, section 3.4). Furthermore, as already hinted at, this characteristic might be due to the conflation of three different myths and to the need to justify Viṣṇu’s horse’s head and indeed, as I will show, this demonic aspect is completely absent in the Vaiṣṇava reuse of Hayagrīva. As for the Vedas, it is noteworthy that the connection with them occurs mainly through their phonic form, insofar as Hayagrīva is said to recite the Vedas by neighing, with a pleasant voice (see the MBh passage quoted above in this section),23 and also to rescue them. It is in the iconography that, due to the difficulty of representing an audible Veda, the connection with the Vedas evolved into the motif of Hayagrīva holding a thin book which should represent the Vedas. The interesting point here is that in this way the connection with the Veda evolves into a non-Vedic ideal, that of the Vedas as books instead of being only orally transmitted (on this transformation of the Vedas into written books, see Larios 2011, who also quotes the example of Hayagrīva). 19 According to the Bhāgavata Purāṇa, after years of ascesis, the demon Hiraṇyakaśipu had obtained from Brahmā a boon of his choice and asked for immortality but Brahmā refused. Therefore, Hiraṇyakaśipu asked to be killed neither by a human being nor by an animal, nor by a demon, nor by a God. He is at last killed by Viṣṇu in the form of Narasiṃha, who is neither a human being, nor an animal, nor a God. Hayagrīva’s request seems very similar. 20 Please note that, as already discussed in section 3.1, horses are very significant already in the Vedic mythology, so that “pre-existing deity” does not mean “non-Indo-Aryan”. 21 For further possible conflations regarding Hayagrīva, see Van Gulik 1935, p. 19. 22 Van Gulik 1935 discusses the transformation of Hindu deities into terrific deities of the Buddhist Tantric pantheon, so that the demonic nature of Hayagrīva surely does not exhaust the reasons for this transformation. 23 In this sense, it is important to note that Hayagrīva did not “create” the Vedas, as maintained in Gail 2013, p. 142, but rather recites the permanent Vedas. The Bhāgavata Purāṇa passage (BhP 2.7.11) quoted by Gail as his only evidence for the above claim only says that Hayagrīva is “made of the Vedas” (chandomaya) and that these sounds come forth out of his nostrils (vāco babhūvur uṣatīḥ asya nastaḥ). 11 3.2.3 Iconography of Hayagrīva in the Purāṇas More interestingly for my perspective, which mixes art history and textual history, are texts prescribing Hayagrīva’s iconography (i.e., not just texts like the MBh passage examined above, in which Hayagrīva is described, without aiming at prescribing how icons of him should be made). In the Viṣṇudharmottara Purāṇa (henceforth VDhP) Hayagrīva is described as a form of Saṅkarṣaṇa.24 Moreover, arms he has eight hands. In four of his hands he holds a conch, a wheel, a mace and a lotus. The rest of the hands should be placed on the four personified Vedas (kartavyo ’ṣṭabhujo devas tatkareṣu caturṣv atha || śaṅkhacakragadāpadmān sākārān kārayed budhaḥ | catvāraś ca karāḥ karyā vedānāṃ dehadhāriṇāṃ || devena mūrdhni vinyastā sarvā bharaṇadhāriṇā | VDhP 3.80.3cd–5ab) colour he should be shown wearing a blue garment, fair complexion (?) position he should have his feet on the hands of the earth goddess (?), he should smile (?) (nīlāmbaradharaḥ, VDhP 3.80.2c) The last three iconographic details (marked with a question mark) are difficult to ascertain, insofar as on the basis of the text found in Viṣṇudharmottaramahāpurāṇa 1912 and Priyabala 1958 one might say that Hayagrīva should have a smiling face,25 whereas the same passage appears with substantial variants in Gopinatha Rao 1968, where the passage states that Hayagrīva has fair skin and has his feet on the hands of the earth [goddess].26 Mallmann (1963, p. 40) derives from Rao the same information, whereas Desai (1973, p. 143) repeats it without any indication of its sources. In the Agni Purāṇa, Hayagrīva is described as arms he has four arms, carrying conch, wheel, mace, Veda position he stands with his left foot on the snake Śeṣa and his right foot on the back of a tortoise.27 24 saṅkarṣaṇāṅgam, VDhP 3.80.3a. 25 mūrtimatpṛthivīpālahastapādasmitacchaviḥ, VDhP 3.80.2ab. Emphasis on the word ‘smiling’ added. 26 mūrtimān pṛthivīhastanyastapādas sitacchaviḥ (Gopinatha Rao 1968, appendix C, p. 58). Rao does not list his sources, however, in the first footnote of the Appendix (Gopinatha Rao 1968, appendix C, p. 1, in Sanskrit) he explains that his sources contained many mistakes (bhūyiṣṭham apaśabdaliṅgavyatyayādikam upalabhyate) and that at times he had to correct the ones which would have made the text meaningless (arthāvabodhoparodhakānāṃ param apaśabdānāṃ sādhusvarūpam adhastād upadarśyate). Since the appendix presents several other interpretive footnotes, it is not clear whether Rao intended to emend the texts directly or whether he rather suggested his emendations only in the footnote. Therefore, it is impossible to evaluate the authority of the passage of the VDhP he reports. As for the latter content, however, Hayagrīva’s smile is never mentioned in other non-Buddhist sources and the word chavi ‘complexion’ is strangely left without any attribute in Viṣṇudharmottaramahāpurāṇa 1912 and Priyabala 1958. Moreover, the misreading between s and m can easily occur in several North Indian scripts, so that the reading °s sitachaviḥ ‘he, who has a fair complexion’, as in Rao, seems justified, although it is absent from the two editions of the VDhP. The issue regarding the other point, namely, the reference to the earth, is much more complicated, although Rao’s reading seems interesting and partly harmonises with the wording of the Agni Purāṇa and in general with the presence of the earth goddess at the feet of other Viṣṇu images (see Maxwell 1992–1993, images 01 and 02. 27 śaṅkhacakragadāvedapāṇiś cāśvaśirā hariḥ | vāmapādo dhṛtaḥ śeṣe dakṣiṇaḥ kūrmapṛṣṭhagaḥ || (Agni Purāṇa 49.26). The passage is discussed in Mallmann 1963, p. 39. 12 This passage of the Agni Purāṇa belongs to the portion allegedly influenced by the Hayaśīrṣa Saṃhitā (see section 5), and it contains elements typical of the pan-Indian Hayagrīva, including his standing. The connection with the Veda is, however, already crystallised in the form of its manuscript. In the much later Bhāgavata Purāṇa28 Hayagrīva is described as attributes connected with the Vedas and with the sacrifice, neighing colour golden (tapanīyavarṇa) 3.3 Hayagrīva and Vaḍavāmukha Among the many seeming synonyms of Hayagrīva, one is particularly interesting, namely Vaḍavāmukha (or Vaḍavāvaktra) ‘having the face of a mare’.29 In fact, the peculiar element is the accent on the female horse, represented in the iconography by the presence of breasts. Why should a male deity be described specifically as having the face of a female animal? As with the case of the fusion of elements described above (section 3.2.2), it is possible that also in this case two different stories were conflated into one. On the one hand, there is the myth of Hayagrīva as described in the previous section (3.2.2), on the other, the myth of the submarine fire erupted out of a volcano identified with the mouth of a mare, a myth repeatedly found in the Harivaṃśa supplement to the MBh and in various Purāṇas30 and connected with the cosmos-destroying deflagration which takes place at the end of a kalpa. Iyanaga 2002 further connects the femininity of this mare with other Vedic and Indo-European myths about female horses as found in Doniger 1980. Evidence of the initial separate nature of Vaḍavāmukha and Hayagrīva is also the fact that they are mentioned separately in a passage of the VDhP discussing the places in which each aspect of Viṣṇu is worshipped: [People worship] Varāha in the Sindhu region and Trivikrama at Śāligrāma | In the same way, [people worship] Matsya in the Kashmirian regions and Vaḍavāmala in the ocean || 3 || In the same way, [people worship] Rāma in Ayodyā and Dharma in the Naimiṣa pilgrimage place | And [they worship] Aśvaśiras in Karṇāṭaka and Nṛkeśari in the Madra region || 4 || 31 28 BhP 2.7.11, see Gail 2013, p. 142 and Van Gulik 1935, p. 18. am grateful to Kiyotaka Yoshimizu, who pointed out to me the problem of the femininity of Vaḍavāmukha and referred me to Iyanaga 2002, which, unfortunately, I am unable to read directly. 30 For the Matsya and Padma Purāṇas, see Gail 1977, p. 73, for the Viṣṇudharmottara, Viṣṇu and Bhāgavata Purāṇas see Gail 2013, pp. 140–141. 31 sindhukūle varāhaṃ ca śāligrāme trivikramam | kāśmīreṣu tathā matsyaṃ sāgare vaḍavānalam || 3 || ayodyāyāṃ tathā rāmaṃ naimiṣe dharmam eva ca | karṇāṭe cāśvaśirasaṃ madradeśe nṛkeśarim || 4 || (VDhP 3.121.3–4). Note that the context allows one to interpret karṇāṭa as indicating a wider region than today’s state of Karṇāṭaka. This passage is in any case interesting evidence of the diffusion of the cult of Hayagrīva towards South India, even before the time of the Tamil Āḷvārs. Gail (2013, p. 145, fn. 17) suggests a different interpretation, namely: “the ocean (Sāgara) seems to represent the original or main site of the submarine site (Vaḍavānala), while horse-headed images (aśvaśiras) can be found (preferably) in Karṇāṭaka”. But this differentiation between a “main site” and a “(normal) site” is 29 I 13 The fusion32 of the two myths might have been easier through the fact that ViṣṇuHayagrīva is said to have slain Madhu and Kaiṭabha in the ocean (see above, section 3.2.2). A further element in favour of the distinction of Vaḍavāmukha and Hayagrīva is the fact that they are separately listed (as Vaḍavāvaktra and Vāgīśvara) in the list of 38 vibhavas ‘manifestations’ of Viṣṇu in the Sattvatā Saṃhitā and in various other Pāñcarātra Sacred Texts —something which does not happen with any of the other names of Hayagrīva (Rastelli 2006, pp. 361–362).33 The Sattvatā further describes Vaḍavāmukha as bearing no attribute at all (alāñchana, see Rastelli 2006, pp. 391– 392). For further details on the Pāñcarātra depiction of Hayagrīva see below, section 3.5. 3.4 Hayagrīva in Tantric Buddhism The Buddhist Hayagrīva34 (about whom see Linrothe 1999, especially pp. 85–141) strongly differs from the Vaiṣṇava Hayagrīva. His first depictions in the Buddhist milieu date to the sixth century (cave 7 at Aurangabad, see Fig. 12 in Linrothe 1999) and throughout his Buddhist pre-Tantric life, notes Linrothe, Hayagrīva is found close to Avalokiteśvara and his appearance “is marked by significant formal dependence on the ancient images of Yakṣa, Gaṇa and āyudhapuruṣa; indeed, at this stage Hayagrīva is hardly discernible from these non-Buddhist demigods” (Linrothe 1999, p. 85). More importantly for my purposes is what follows: head he has a normal human head arms he has two arms position he sits or stands attributes club (daṇḍa), snakes, necklace association he is an attendant of some forms of Avalokiteśvara or of Tārā (Linrothe 1999, p. 85) Note that this pre-Tantric Buddhist Hayagrīva is the only form of Hayagrīva which has no trace of a horse’s head. Within Tantric Buddhism in India, Tibet, China and Japan,35 Hayagrīva may increase in size and importance but “never rival[s] [his] master bodhisattva” (Linrothe 1999, p. 85). He is depicted with a small horse’s head in the crown over is human, terrific head (or, less frequently, with three small horse’s heads in the crown over his human terrific one36 ) and is a terrific deity. Texts describe him as an independent not presupposed by the initial question to which these verses are an answer and it would be present, within the passage, only in the case of Hayagrīva. 32 Also Gail, who calls the Vaḍavāmukha aspect the “Ādi (i.e., Ur-)-Hayagrīva”, maintains that “The Mahabhārata joins the horse’s head with explicit Vaiṣṇava mythology” (Gail 2013, p. 140, emphasis added). 33 I am grateful to Marion Rastelli who pointed out this passage. 34 I am grateful to Gerald Kozicz for many interesting discussions and suggestions concerning the Buddhist Hayagrīva. 35 An elaborate discussion of this history can be found in Van Gulik 1935. 36 I am grateful to Gerald Kozicz, who showed me some examples of the latter type; see also his contribution within this volume, fig. 14. 14 deity (i.e., not as a form of Viṣṇu) and identify him as a vidyārāja.37 This Buddhist evolution is not the focus of the present paper but one might suggest that the terrific aspect may have evolved from Hayagrīva’s connection with the demon of the same name (although see some words of caution above, fn. 22) and that his presence among the vidyārājas could have to do with his being connoted as the God of the Vedas and of knowledge in general. Also interesting is the (re-?)emergence of the horse’s head, which might be due to a crossing influence from non-Buddhist milieus, so that the Tantric Buddhist Hayagrīva would reuse some of the elements of the non-Buddhist Hayagrīva. This influence could also explain the presence of some new attributes, such as the bow and arrows found in the Tsatsapuri temple (see Kozicz’ contribution, Fig. ADD), which were not found in the pre-Tantric Buddhist Hayagrīva but were common in Vaiṣṇava depictions of Hayagrīva as a minor deity (see below, section 5.1). However, a cursory description,38 already shows that the terrific Hayagrīva shares only some of the elements with his Vaiṣṇava counterpart (as described in section 2.1): head he has one or three horse’s heads over his own human and terrific one (often as part of his crown), he has one or three human faces arms he has four arms or eight arms position he stands attributes tiger skin, club, snakes colour green or blue (according to Van Gulik), red (plates 5 and 9 in Linrothe 1999), the principal face is red, the other two are blue and white, the horse’s head is green (Mallmann 1986, pp. 180–181) Figure 3: The Buddhist, terrific Hayagrīva 37 A vidyārāja ‘king of knowledge’ is in Tantric Buddhism a wrathful manifestation of the Buddha. 38 Full details can be found in the very informative and well thought-out Van Gulik 1935. Many Tibetan images of Hayagrīva are reproduced in Chandra 1991. 15 3.5 Hayagrīva in the Pāñcarātra Saṃhitās The Pāñcarātra Saṃhitās are a constellation of Sacred Texts devoted to Viṣṇu, the composition of whom probably started in the last centuries of the first millennium in Kashmir and migrated to South India after the 11th c., where it was prosecuted for some centuries. In this sense, they are more a genre, like the Upaniṣads, than a group of texts bearing specific relations with each other. However, most Pāñcarātras are also characterised by a high degree of intertextuality, so that entire portions, especially of the most ancient and revered texts, are reused in later ones (see Rastelli 2006, Leach 2012, D฀bicka-Borek 2015). Further common elements are the chiefly ritual focus of most texts and the so-called Tantric elements of such rituals. By contrast, the relationship between the devotee and the God is differently configured in the various texts, oscillating between monism (advaita, there is no other reality but God) and dualism (dvaita, the human soul is different from God), with all possible intermediate positions being implicitly upheld (explicit statements are comparatively rare, given the ritual character of these texts). Also the type of ritual described varies over time in the Pāñcarātra constellation. The first texts describe private rituals performed by householders at home, whereas later texts describe temple rituals. It is perhaps due to this focus on householders’ private rituals that the Vedas are at first simply ignored in the Pāñcarātra Sacred Texts. However, some of the Pāñcarātras refer to and quote from a so-called Ekāyana Veda, which would have been a Pāñcarātra text and the common root of the extant four Vedas. In the Pāñcarātras, already in the early, Kaśmīri ones, Hayagrīva appears relatively frequently as an avatāra of Viṣṇu (Ahirbudhnya, Pauṣkara, Pādma, Parāśara, Īśvara, Śeṣa, Lakṣmī Tantra, see Nayar 1994, chapter four). His past connection with the demon bearing the same name is not debated, whereas the connection with the fact of having rescued the Vedas from Madhu and Kaiṭabha is highlighted. However, different details are found in the various Pāñcarātra Saṃhitās in terms of Hayagrīva’s iconography.39 For instance, arms a different number of arms is mentioned, from two to twelve attributes different objects are said to be held by him: disc, conch, club, lotus, noose, goad, staff (aṅkuśa), firebolt (agnivajra), book —not further specified as the Veda—, rosary, only some texts mention a specific mudrā, which is either the varadamudrā or the jñānamudrā. Note that apart from the last ones, all other attributes are generic Vaiṣṇava ones colour different colours —but usually white and shiny— are connected with his complexion association The Parāśara Saṃhitā (27.6–23) mentions both Śrīdevī and Bhūdevī CHECK! More in detail, the Sattvatā describes Vāgīśvara ‘Lord of Speech’ as varavājimukha ‘having the face of a noble horse’ and then lists: arms six attributes lotus, rosary, Vedic altar, two ritual spoons, a ritual seat made of darbha grass and antelope skin and with Soma and a string to delimit the altar (in the right hands); book —not further specified—, conch, stick, water jar (kamaṇḍalu), ritual ladle, and various ritual oblations (in the left hands)40 39 For a complete overview, see Nayar 1994, chapter four, section 1.1.3 and Sridhara Babu 1990, chapter III and chapter IV.2. 40 See Rastelli 2006, p. 393, where the entire passage is reproduced and translated. 16 This description clearly shows that the identification of Vāgīśvara and Hayagrīva predates Veṅkaṭanātha and so does the attempt to connect this deity with Vedic elements. The Sattvatā account is, however, clumsy insofar as too many attributes are mentioned for each hand. An early Pāñcarātra, which has unfortunately only been partially edited, goes under the name of Hayaśīrṣa Saṃhitā and describes Hayagrīva with distinct characteristics, as will be explained below in section 5. 4 Hayagrīva in art history (apart from the area of the Vijayanagara empire after Veṅkaṭanātha) Although Hayagrīva is not frequently represented (as can be expected, given that he is only a minor avatāra of Viṣṇu), there are some depictions of him both alone (see above, section 2.1) and in complex images (e.g., together with Brahmā and Śiva in the Viśvarūpa image of Śamalājī, Gujarat or more frequently together with other avatāras of Viṣṇu, see Gail 2013, p. 146). According to Adalbert Gail (2013), the oldest image of Hayagrīva comes from Mathurā and hails from the Kuśāna period (2–3 c. AD). This Hayagrīva conforms to the characteristics described in section 2.1 insofar as he stands, and holds a club and a disk in his upper two arms. Out of the lower two arms, one is described as being “empty”41 and the other as carrying an object which might be identified as a kamaṇḍalu ‘water jar’ or conch (see Joshi 1972–73, p. 37) or as a book (see Gail 2013, p. 143). An image from the ardhamaṇḍapa (a porch leading to the main hall) of the Vaikuṇṭha Perumāḷ temple (8th c.) in Kāñcīpuram depicts, again, a standing Hayagrīva, which seems to be holding in his four arms a water jar and perhaps a conch and a disc (the stucco is damaged, which makes an exact identification is difficult) (Gail 2013, Fig. 8). Gail presents also a specimen of Hayagrīva from Kaśmīr, which he dates to ca. 9th c AD. The depiction of Hayagrīva is part of a metal frame and is significantly preceded by Viṣṇu resting on the waters and followed by Viṣṇu killing Madhu and Kaiṭabha (Gail 2013, p. 143 and Fig. 3). This Hayagrīva has: arms eight arms attributes varadamudrā, lotus, sruc ‘large Vedic ladle’, sun, moon, sruva ‘small Vedic ladle’, club, and conch As for these attributes, the sun and the moon might have no specific meaning, since sun and moon are generic attributes indicating a deity at least in Central Asian art.42 The concomitant presence of sruc and book, instead, might suggest that this relatively early iconography of Hayagrīva has been directly transposed from that of Brahmā, who is often represented with sruc and book, with the same pro-Vedic symbolism. Maxwell suggests in fact that this transposition from Brahmā to Hayagrīva may have indeed taken place in Kaśmīr (see below, section 4.1): 41 Perhaps because it is partly mutilated, given that hands are never left without attributes in Indian sculptures —at least as far as my knowledge reaches). 42 See the description about the Western Wall of the Buddhist shrine D13 in Lo Muzio forthcoming. I am grateful to Ciro Lo Muzio for having shared with me a pre-print copy of his paper. 17 The ladle and manuscript in North Indian tradition refer to Brahmā, of course, but in this Kashmiri image43 they were transferred to Hayagrīva, as can be seen in the image of this deity in the 3rd position on the left side of the frame. Three mediaeval images have been preserved in the Kotah Museum. One is a four-handed standing image with the face of a horse. The lower right hand is in the varadamudrā, the upper right holds the mace, the upper left holds the book and the lower left holds the kamaṇḍalu ‘water jar’. Another image is also similar to this.44 Figure 4: The depiction of Hayagrīva preserved in the Kotaḥ museum (from Agrawala 1961) As for composite images depicting Hayagrīva together with other avatāras, according to Gail the second most ancient depiction of Hayagrīva is a stele built around the 7–8 c. and displaying Viṣṇu, Varāha, Hayagrīva and Narasiṃha (see Gail 2013, Fig. 2). Here, again, Hayagrīva is standing, he displays the abhayamudrā and touches with his lower arms the personified disc and the personified club. He seems to be carrying a book, although he is holding it tight in his fist instead of having it lying on his open palm, like the Hayagrīvas found in the area of the Vijayanagara empire. A four-headed Viṣṇu is conserved in the Khajuraho Museum and displays on its rear a horse’s head (on the sides are the lion’s and the roar’s heads) (Gail 2013, Fig. 6–7). Of particular interest are late images of Veṅkaṭanātha in areas far away from the Vijayanagara empire. Among them, Gail mentions one from the Svathanārāyaṇa temple, Pāṭan (Kathmandu Valley, 1666–1674), where a standing Hayagrīva is found in a pillar opposed to Varāha. He has eight arms and carries four Vaiṣṇava attributes (disc, conch and lotus are recognisable, a club is probably lost, see Gail 2013, p. 145, see also his Fig. 10, where the four left arms are, however, outside of the photograph). Gail notes that “This is the only image of Hayagrīva that appears in full accordance with the description of the V[iṣṇu]dh[armottara Purāṇa] pratimālakṣaṇa, section III, 43 Maxwell 44 All refers to the image 02 in his book, available on line. information from Agrawala 1961. 18 80 that enumerates the classical attributes of Viṣṇu and the embodied Vedas” (Gail 2013, p. 145, fn. 18). A partial exception to the pan-Indian standing Hayagrīvas is a late painting from the Pahari area (Mankot, Himachal Pradesh) and dated to 1720 AD (according to Gail 2013, p. 146, see also his Fig. 11) where Hayagrīva sits on a lotus and displays conch, club, disc and lotus in his four hands. To sum up, the depictions of Hayagrīva throughout India apart from the postVeṅkaṭanātha Vijayanagara empire confirm the fact (already described above, see section 2.1) that Hayagrīva is depicted as standing, with four or eight arms, with pan-Vaiṣṇava attributes (or even more generic ones, as in the case of the Hayagrīva found in Kaśmīr and described in Gail 2013, p. 143) and with mudrās not directly linked to knowledge. The attendants are also not linked to knowledge, nor is Lakṣmī present. Particularly striking is the case of the Hayagrīva image found in Kāñcīpuram, that is, in a site closely connected to Veṅkaṭanātha (see Freschi forthcoming(c)), where the iconography of Hayagrīva is nonetheless very distant from that of Hayagrīva as praised by Veṅkaṭanātha. 4.1 Geographic diffusion of images There seems to be some disagreement in the secondary literature concerning the diffusion of the images of Hayagrīva.45 This is probably due to the paucity of images, so that a small difference in the number of images available to the one or the other author leads them to very different conclusions. More in detail, D. Desai notes that Khajuraho must have been a centre of the Hayagrīva cult (Desai 1993), whereas K.S. Desai writes: “It seems that the Hayagrīva worship was acknowledged only by a narrow group of people. His worship must have been prevalent in Rajasthan, for Rajasthan has produced his images, as far as the present discoveries go.” (Desai 1973, pp. 143–44). Next, Gail observes: Hayagrīva is a minor avatāra of Viṣṇu, not represented in all regions of India. From Himachal Pradesh, Bihar/Bengal, Gujarat/Rajasthan [!], Maharashtra, Andhra Pradesh, Kerala, not a single image of Hayagrīva is known to me. (Gail 2013, p. 146) A pivotal role in the diffusion of the cult of Hayagrīva may have been played also by Kaśmīr, since more than one image of Hayagrīva was found in Kaśmīr (see section 4), which might be also the original place of composition of the relevant section of the VDhP (as maintained in Maxwell 1992–1993 on figure 02, and as suggested by the parallels to the early Pāñcarātra vyūha doctrines which can be traced in the VDhP, see Srinivasan 2004, p. 572) and which was the place of composition of various Pāñcarātra texts mentioning Hayagrīva (see section 3.5) and especially of the Hayaśirṣa Pāñcarātra (see section 5). Gail also indirectly refers to what I call the Vijayanagara empire-Hayagrīva, insofar as he notices that Karṇāṭaka must have been the “homeland of the Hayagrīva cult”. As for his understanding of “Karṇāṭaka”, Gail’s references to the VDhP passage quoted above, section 3.3, and to “Hoysala temples” allows one to conclude that he might be referring to the same area I defined in footnote 7. Last, a damaged Śrī Vaiṣṇava pre-Veṅkaṭanātha inscription in the Śrī Raṅganātha temple ad Śrīraṅgam offers a further interesting perspective on Hayagrīva: 45 See also Sridhara Babu (1990, chapter II.4), who discusses the geographic references to Hayagrīva in the Purāṇas. 19 There are a number of inscriptions of the Hoysaḷas in the Śrī Raṅganātha Svāmi temple in Śrīraṅgam, many of them belonging to the reign of VīraRāmanātha.46 One of the most important of these is a damaged epigraph in the third prākāra [wall, EF] of this very last temple-complex dated CE 1269 of the time of Vīra Rāmanātha. This inscription is particularly important as it refers to the existence of a library (Sarasvatī-bhaṇḍāram), attached to the temple of Śrīraṅgam. […] This record of Vīra-Rāmanātha mentions that the images of Sarasvatī Devī, Vedavyāsa Bhagavān and Hayagrīva […] had been newly installed in the maṇṭapa constructed as an adjunct to the library. (Madhavan 2013, pp. 138–139)47 No further details are present but one is tempted to imagine that Sarasvatī represents learning and language in general, whereas Hayagrīva and Vyāsa are meant as deities presiding over Vedic and post-Vedic Vaiṣṇava literature respectively (Vyāsa is identified with the author of the Vedānta Sūtra at least since the time of Veṅkaṭanātha, see Freschi forthcoming(b)). 5 Hayagrīva in the Hayaśīrṣa Saṃhitā Of particular interest is the fact that the iconographic portions of the Agni Purāṇa are frequently attributed to Hayagrīva. This is in turn due to the fact that among the sources for such passages is the ancient and elusive Hayaśīrṣa Pāñcarātra Saṃhitā (about which see Rastelli 2007).48 More in detail, Agni Purāṇa 39–70 stems from the Hayaśīrṣa Pañcarātra and this part of the Agni Purāṇa is in fact ascribed to Hayagrīva. The Hayaśīrṣa Saṃhitā is probably an early Pāñcarātra Saṃhitā of which only the first book has been edited (Dutta Sastri 1976). It has been little studied by scholars and seems to have originated in North India,49 possibly in an early age.50 Further, It consists of four sections (kāṇḍa), of which only the first, the ādikāṇḍa, has been edited and published. It derives its name from the fact that according to its framestory (ādikāṇḍa 1) it was revealed by God in the form of Hayaśiras, the Horse-Headed One. At the beginning of each chapter of this text, the narrator is identified as Bhagavat. (Rastelli 2007, p. 190) 46 The author adds here a reference to ARE 139 of 1938–9 and SII, vol. XXIV, no. 279, pp. 297–98. 47 I am obliged to Marion Rastelli, who pointed out this reference. 48 I am grateful to Robert Leach and Marion Rastelli for discussing this Saṃhitā with me. I also recently attended a paper on the Hayaśīrṣa Saṃhitā which Valdas Jaskūnas presented at the first conference of the European Association of Asian Art and Archaeology (Olomouc, September 2014): http://www.ea-aaa.eu/conference2014/program. Unfortunately, I have not been able to receive any news concerning the publication plans of Jaskūnas. 49 “From the evidences of the names of the provinces forbidden in religious consecration ceremonies viz. Kacca, Kāveri, Koṅkana, Kāmarupa, Kaliṅga, Kāñcī, Kāsmira, Kosala and Maharāṣṭra we can conjecture that this work was composed somewhere in the Northern part of India. This is corroborated by the fact that Hayaśīrṣa Pāñcarātra is available in the Kāsmira and Nāgari (Northern Indian) scripts only. These scripts only are recommended for transcriptions of this sacred text” (Sridhara Babu 1990, p. 50). Although I am not convinced by the first argument (see Rastelli forthcoming, section “Names, Places and Motives” for a discussion of how sacred geography might deviate completely from one’s concrete local environment), the second one appears persuasive. 50 Cf. Smith 1978: 166: “(…) there are many clues in it which suggest an early date; yet other details indicate that, albeit written early, it passed through the hands of late redactors.” Cf. also Smith 1975: 549ff. 20 The HS is by no means a text predominantly dedicated to Hayagrīva but nonetheless one is automatically led to the question regarding the reasons which made a secondary deity gain a more important role at a certain point of his history. In general, different Pāñcarātra Saṃhitās stress different aspects of Viṣṇu. More in particular, however, one might wonder whether certain belligerent aspects of Viṣṇu (such as Nṛsiṃha and Hayagrīva) gained more popularity because of the influence of the coeval Śaiva and Śākta emphasis on furious deities.51 5.1 Descriptions of Hayagrīva in the Hayaśīrṣa Saṃhitā The HS describes Hayagrīva at least twice.52 First, in the first paṭala ’chapter’ of the first book, it describes Hayaśiras as follows: With four arms, carrying the club, the discus, the lotus and the bow caturbhujaṃ gadācakrapadmaśārṅgadharaṃ […] (HS 1.1 first part of v. 22ab) However, in the 25th paṭala of the first book, the Hayaśīrṣa Saṃhitā describes Hayagrīva (in fact, Aśvavaktṛ) in a form very similar to the one found in Veṅkaṭanātha’s Hayagrīvastotra: Rather, one should let [an artist] make me with conch, discus, club and Vedas in the hands || 24 || Distinguished as having the face of a horse and four arms | seated in padmāsana and connected in the upper part of the body with two (?)53 Goddesses || 25 || śaṅkhacakragadāvedapāṇiṃ vā kārayīta mām || 24 || aśvavaktraṃ caturbāhum evam eva vyavasthitam | puṣkarāsanam adhyasthaṃ devīdvitayasaṃyuktaṃ || 25 || Readers will note that the two forms do not harmonise and that the first one seems more old-fashioned, insofar as it is closer to the pre-Veṅkaṭanātha iconography. The latter description, by contrast, is somewhat intermediate between the Khajuraho and the Veṅkaṭanātha models of Hayagrīva, since in it Hayagrīva: arms has four arms holding conch, discus and Vedas but also the club position seated as in the Hayagrīvastotra but connected to Lakṣmī and Bhūdevī (?) This shows that Veṅkaṭanātha probably had a precise model and that he chose to focus on what was according to him the real essence of Hayagrīva, with some specific attributes (I could imagine that the club was eliminated also because Veṅkaṭanātha wanted to be sure that the jñānamūdra and the Vedic book were always present) and without Lakṣmī. 51 This suggestion is developed in Sumant 2010, p. 65, which refers to Tripathi 1978, p. 44, and to Sanderson 2007, pp. 226–227, for the extra-Tantric outreach of the Tantric motif of fearful deities, especially Nṛsiṃha. I have already dealt above (section 3.2.2) with the structural similarities between the myths in which Hayagrīva slaughters a demon and the Hiraṇyakaśipu myth, in which Nṛsiṃha is the main figure. 52 A further passage, which could be the source of AP 49.26, see section 3.2.3, might be in the unedited portion of the HS. 53 If devīdvitaya actually means ‘the two goddesses’, then this passage would harmonise with the passage of the Parāśara Saṃhitā where Śrīdevī and Bhūdevī are mentioned, see above, section 3.5. 21 6 Hayagrīva as supreme deity Hayagrīva becomes the supreme deity for Veṅkaṭanātha (1269–1370). This philosopher, also known as Vedānta Deśika, is among the most important thinkers of the Viśiṣṭādvaita Vedānta school of philosophy and of the corresponding religious movement known as Śrī Vaiṣṇavism. Today, two currents of Śrī Vaiṣṇavism oppose each other and one of them regards Veṅkaṭanātha as its founder, although the split between the two occurred only much later.54 Veṅkaṭanātha was the first to write a stotra ‘eulogy’ to Hayagrīva, in which He is praised as the supreme deity. Why did Veṅkaṭanātha raise Hayagrīva to the level of a supreme deity? I will come back to this topic, but for the time being my tentative answer is that Hayagrīva was a way for the philosopher and theologian Veṅkaṭanātha to frame his relationship with God under the shield of an intellectual form of God. Moreover, Hayagrīva’s link to the Vedas was a further positive element for Veṅkaṭanātha, who had always tried to frame Śrī Vaiṣṇavism as a non-ant-Vedic movement and to reinterpret in a non-anti-Vedic way the Ekāyana-Veda statements present in some Pāñcarātras (see above, section 3.5). Further evidence of this attitude can be found in Veṅkaṭanātha’s approach to the validity of Pāñcarātras, which are said to derive their authority from the Veda (SM ad 1.1.2). 6.1 Veṅkaṭanātha’s mentions of Hayagrīva and his Hayagrīvastotra Veṅkaṭanātha mentions Hayagrīva both in his philosophical texts and in his religious hymns. Whereas authors after him start using standardised maṅgalas ‘auspicious verses opening a work’ (see below, section 6.3.2) praising Hayagrīva, Veṅkaṭanātha’s mentions of Hayagrīva are still freely shaped and Hayagrīva is not a fix presence in each maṅgala. As an example, let me quote the maṅgalas that Veṅkanātha (presumably55 ) put at the beginning of some of the chapters of his Śatadūṣaṇī. Interestingly, they focus on different aspects of Hayagrīva. The first one focuses on His being connected with the Veda and speech, the second on the latter connection only, the last two on Him as the supreme deity, while the middle one (opening the 29th vāda) is a sort of threshold between Hayagrīva’s connection to knowledge and Hayagrīva as supreme deity: Let the shout of the neighing of Hayagrīva —which collects the Sāmans (melodies of the Sāmaveda) and all the words of the Ṛcs (the strophes of the Ṛgveda), which is the abode of the Yajus (the ritual formulas of the Yajurveda) and destroys all obstacles, which is the collection of the waves of cognition in the water of knowledge— destroy the ignorance which is present in the tumult (kolāhala) of the quarrelling disputants who are trembling because of pride in the discussion!56 54 On the two currents, see Raman 2007. For more on Veṅkaṭanātha in general, see Freschi forthcoming(c). For more on Veṅkaṭanātha’s theology, see Freschi 2015a. 55 Caution is needed since the opening verses of a book or a chapter are particularly open to additions by copyists, who often add altogether new maṅgalas or may add new verses or new deities to existing ones. 56 samāhāras sāmnāṃ pratipadam ṛcāṃ dhāma yajuṣāṃ layaḥ pratyūhānāṃ laharivitatir bodhajaladheḥ | kathādarpakṣubhyatkalikathakakolāhalabhavaṃ haratv antardhvāntaṃ hayavadanaheṣāhala- 22 Let the waves of the delightful neighing of Hayagrīva in the blissful ocean, | which have thrown away the erroneous views of the outsiders at the end of a discussion, win! ||57 As it could have been expected, the middle maṅgala performs a difficult role and its translation is thus trickier: The Upaniṣads, by repeating what has been understood, properly distribute all of this (this whole knowledge), which consists purely of Him: | Let He, treasure of good things,58 with the face of a horse, whose opulence is not understood, take perpetually place close to us ||59 There is an evident echo between vidita/avidita ‘understood/not understood’ and vidadhati/antarvidhattām ‘to place or distribute/to place internally’. Given that the second part of the verse refers directly to Hayagrīva and the first part to the Upaniṣads, the gist of the passage appears to lie in the idea that the Upaniṣads are an excellent device for gathering knowledge but Hayagarīva surpasses all possible human knowledge. As already mentioned, the last two maṅgalas dedicated to Hayagrīva focus on Him as the Supreme Deity: Honour to Hayagrīva, the Brahman, who is the inner-Self of the three types of self (the normal souls, the liberated ones and the ones liberated ab initio) [and] is the single cause for the liberation from bondage ||60 He who is accompanied by Lakṣmī, separates through various souls who are swallowed (nigal-) by the innate māyā and the three guṇas (sattva, rajas and tamas), which are hard to overcome | This merciful [and] great god with a horse face, may he rescue us, who are slaves of the destruction of the basis which is the creation of the world ||61 Hayagrīva is also mentioned, interestingly enough within Veṅkaṭanātha’s “agenda” at the beginning of his Seśvaramīmāṃsā, after two maṅgala verses to God in general and to the Mīmāṃsā teachers: halaḥ || 1|| (beginning of the first vāda ‘discussion’). The same verse is found in the HGS as v. 3. 57 hayagrīvasudhāsindhuharṣaheṣāravormayaḥ | jayanti vādavelāntakṣiptabāhyakudṛṣṭayaḥ || (beginning of the ninth vāda). 58 The pun is based on the fact that the second sannidhi has its usual meaning of ‘closeness’, whereas the first one should be interpreted as a compound of sat and nidhi. I am grateful to Harunaga Isaacson for suggesting to me this interpretation. 59 viditam anuvadanto viśvam etad yathāvad vidadhati nigamāntāḥ kevalaṃ yanmayatvam | aviditabahubhūmā nityam antarvidhattāṃ hayavaravadano ’sau sannidhis sannidhiṃ naḥ || (beginning of the 29th vāda). 60 namas tridhā vibhaktānām ātmanām antarātmane | brahmaṇe hayavaktrāya bandhamokṣaikahetave || (beginning of the 36th vāda). 61 ya eko durlaṅghyatriguṇanijamāyānigalitair vicitraiḥ kṣetrajñair viharati sarojāsahacaraḥ | jagatsargakṣemakṣapaṇaparikarmīṇamahimā dayālur devo ’sau turagavadanas tārayatu naḥ || 42 || 23 These unselfish ones (nirmatsara) shall consider (niśam-, caus.) this [opinion of mine] with respect. [This] correct path (gati) through the sūtras about the rules (naya) regarding (adhikāra) the ritual action (i.e., the PMS), | this one (ayam) God, horse-faced (i.e., Hayagrīva) and inner self of the world, he unites it with the Śārīraka (i.e., the Vedāntasūtra) by means of my voice || 3 ||62 In the Adhikaraṇasarāvalī, the Nyāyasiddhāñjana, the Pāñcarātrarākṣā (notwithstanding the Pāñcarātra context), the Śrīśaraṇāgatidīpikā, and the Saccaritrakṣā there is no mention of Hayagrīva in Veṅkaṭanātha’s maṅgala. Nor is Hayagrīva mentioned in the maṅgala of Veṅkaṭanātha’s commentary on the Adhikaraṇasarāvalī and on the Mīmāṃsāpādukā, nor in the maṅgala of the commentary on the Saccaritrakṣā.63 These data point to the conclusion that the standardisation of the use of Hayagrīva in the maṅgalas occurred after Veṅkaṭanātha and at a significantly later stage. 6.1.1 The Hayagrīvastotra The Hayagrīvastotra is the first hymn dedicated to Hayagrīva. It started a new genre and the later stotras to Hayagrīva embedded verses from Veṅkaṭanātha’s original (on these various praises and invocations, see Sridhara Babu 1990, chapter VII). The central verses for the purpose of the current volume are the following ones (the elements which could be iconographically relevant are in bold face): We revere the God Hayagrīva, who is made of cognition and bliss, whose appearance is an immaculate crystal, [and] who is the receptacle of all Knowledges || 1 || We praise the glory64 which has the face of a horse, is self-established, rivals with a mountain of pure crystal, who cleanses the three worlds with his rays endowed with nectar, whose call is a neighing imitated by the Upaniṣads which will never end, who has destroyed all vices || 2 ||65 […] May the form of Viṣṇu called “Lord of Speech” (Vāgīśa), having the face of a horse, who utters the Vedas shine to me! […] || 4cd || I seek refuge in the God Hayagrīva, whose own nature is a heap of pure knowledge, who is an initiation (yielding salvation) for the bound [souls] through his gift of knowledge, who is a receptacle of mercy, [and] who is the proper refuge for all the embodied (living beings) || 5 ||66 62 nirmatsarā niśamayantv idam ādareṇa karmādhikāranayasūtragatiṃ samīcīm | śārīrakeṇa ghaṭayaty ayam asmaduktyā devas turaṅgavadano jagadantarātmā || (Viraraghavacharya and Nainaracarya 1971). 63 BY WHOM? CHECK! 64 mahas could also refer to a Vedic oblation, probably an intended double meaning, given the context of a praise of Hayagrīva. 65 jñānānandamayaṃ devaṃ nirmalasphaṭikākṛtim | ādhāraṃ sarvavidyānāṃ hayagrīvam upāsmahe || 1 || svataḥsiddhaṃ śuddhasphaṭika maṇubhūbhṛtpratibhaṭaṃ sudhā sadhrīcībhir dyutibhir avadātatribhuvanam | anantais trayantair anuvihita heṣāhalahalaṃ hatāśeṣāvadyaṃ hayavadanam īḍīmahi mahaḥ || 2 || 66 vaktrī vedān bhātu me vājivaktrā vāgīśākhyā vāsudevasya mūrtiḥ || 4 || viśuddhavijñānaghanasvarūpaṃ vijñānaviśrāṇanabaddhadīkṣam | dayānidhiṃ dehabhṛtāṃ śaraṇyaṃ devaṃ hayagrīvam ahaṃ prapadye || 5 || 24 […] In the fire of sacrifice, whose flame is blazing, you, who have assumed a body made of mantras, | have pleased the deities by giving them oblations […] || 10 ||67 […] The wise people visualise in their mind your form, which produces the nectar of bliss, and is alluring (vilobhanīya) [like] the stream (niṣyanda)68 of the young moon, like the vast horizon of the ocean of milk || 13 || The words (gir) of the one who contemplates you —who are the best of the geese in the Kailāsa lake69 which is the mind of the wise— continuously in his mind, spontaneously (svayam)70 participate in debates and excel and generally act as it is fit [for him] || 14 ||71 […] He has four lotus hands, with one in the mode of bestowing knowledge; another holds books of wisdom, and the other two hold the conch and discus. […] May this Lord of Speech who showers such cooling rays of grace on me be forever manifest in my heart! || 32 ||72 Verse 10 hints at an iconographical motive which I could not identify at all in any extant description of Hayagrīva, but if we leave it aside, the following characteristics of Hayagrīva appear preeminently: arms four attributes Vedas, conch and discus colour white and brilliant (i.e., shining white or transparent) mudrā jñānamudrā association The absence of Lakṣmī, which represents something unusual for Śrī Vaiṣṇavism, in which —as the name goes— Viṣṇu is always together with Śrī (Lakṣmī)73 67 agnau samiddhārciṣi saptatantoḥ ātasthivān mantramayaṃ śarīram | akhaṇṣasārair haviṣāṃ pradānaiḥ āpyāyana., vyomasadāṃ vidhatse || 10 || 68 I am not sure about the translation of niṣyanda. 69 mānasa is a pilgrimage place on mount Kailāsa, which is believed to be the native place of the wild geese. But mānasa can also be an adj. deriving from manas ‘mind’. 70 That is, without his conscious efforts. 71 mugdhenduniṣyandavilobhanīyāṃ mūrtiṃ tavānanda sudhā prasūtim | vipaścitaś cetasi bhāvayante velām udārām iva dugdhasindhoḥ || 13 || manogataṃ paśyati yaḥ sadā tvāṃ manīṣiṇāṃ mānasa rājahaṃsam | svayaṃ purobhāva vivādabhājaḥ kiṃkurvate tasya giro yathārham || 14 || 72 vyākhyāmudrāṃ karasarasijaiḥ pustakaṃ śaṅkhacakre bibhradbhinn. A devotional translation of the entire hymn is available in Raghavan, Lakshmi Kumari, and Narasimhachary 1995. A more scholarly translation is currently under preparation by me. 73 This absence might have caused troubles also to later Śrī Vaiṣṇavas looking at the HGS as the standard text on Hayagrīva. For instance, in the introduction of a devotional edition of the hymns by Veṅkaṭanātha, one can read: “Although no specific reference to the Lord’s consort Lakshmi is made in this stotra [namely in the HGS, EF], a veiled remark to her eternal association with the Lord may be found in verse 32 (amlanasrih= of unfading splendour). The remarks made in this hymn to the compassion and love of the Lord may also be taken to substantiate the view that his benign aspect alone is described here. Tradition also recommends meditation of Hayagriva in the company of Lakshmi (Lakshmi-Yahagriva)” (Raghavan, Lakshmi Kumari, and Narasimhachary 1995, p. 12). 25 characteristics Hayagrīva’s connection with speech in general and not just with the phonic form of the Veda, Hayagrīva’s connection with victory in debates All of these elements were already present before Veṅkaṭanātha, although some of them were less pre-eminent (most notably, Hayagrīva’s connection with speech is only found in the Sattvatā Saṃhitā and in the other Pāñcarātra texts depending on it (see section 3.5) and I could not locate any reference to victory in debates). What is new is their balanced synthesis in an ideal image (cf., by contrast, the over-attribution found in the Sattvatā, section 3.5) which will perhaps also because of that gain incredible popularity. 6.1.1.1 Hayagrīva’s dissociation from Lakṣmī The dissociation of Haya- grīva from Lakṣmī in the Hayagrīvastotra might appear surprising, given the emphasis of Śrī Vaiṣṇavism on the connection between Viṣṇu and Lakṣmī. This theological point is also most probably the reason why Veṅkaṭanātha decided to reintroduce Lakṣmī in the praise to Hayagrīva found in the Śatadūṣaṇī (see section 6.1). The reason for the dissociation of the two, by contrast, might lie in the yoga form of Hayagrīva praised in the HGS, which implies that the deity is worshipped in isolation. A distinction between a yoga and a Lakṣmī form of an avatāra of Viṣṇu has been discussed at least also in the case of Nṛsiṃha in Sumant 2010. 6.2 Attestations possibly independent of Veṅkaṭanātha 6.2.1 Sudarśanasūri Sudarśanasūri was a much more senior contemporary of Veṅkaṭanātha and he wrote commentaries on several works by Rāmānuja.74 The edition of Sudarśanasūri’s commentary on the Vedārthasaṅgraha by Rāmānuja starts with the following maṅgala. Although words of caution concerning the authorship of maṅgalas (see fn. 55) apply all the more in this case, it is noteworthy that the maṅgala is not the standardised one which will be examined in section 6.3.2 as it does not mention Veṅkaṭanātha. Honour Honour Honour Honour to to to to the the the the venerable Rāmānuja! venerable Hayagrīva! honourable teacher Śrīnivāsa! previous teachers!75 Śrīnivāsa was a very common name among Viśiṣṭādvaitins, but I could not identify this one (the maṅgala of the VS honours only Viṣṇu). 6.2.1.1 Other Vaiṣṇava schools after Veṅkaṭanātha Hayagrīva is found also in the Madhva-sampradāya, that is, in the Vaiṣṇava dualist school founded by Madhva (1238–1317), but together with other avatāras of Viṣṇu. For instance, Vādirāja (16th c.) opens his Gurvarthadīpikā commentary on Jayatīrtha’s Nyāyasudhā with the following maṅgala: After having honoured Nārāyaṇa, Hayagrīva, Vyāsa, Vāyu [and] Sarasvatī, […]76 74 For some more words on Sudarśanasūri, see Freschi forthcoming(a). rāmānujāya namaḥ | śrīmate hayagrīvāya namaḥ | śrīśrīnivāsamahāgurave namaḥ | pūrvācāryebhyo namaḥ | 76 nārāyaṇaṃ hayagrīvaṃ vyāsaṃ vāyuṃ sarasvatīm | natvā tatkaruṇāsattvād rāme nyāyasudhām budhau || 75 śrīmate 26 What is represented here is a constellation similar to the one evoked in the epigraph about the library attached to the temple at Śrīraṅgam: Nārāyaṇa is the main deity and Sarasvatī, Hayagrīva and Vyāsa are connected to learning, Vedic and post-Vedic Vaiṣṇava literature (see above, section 4.1). Hayagrīva is also praised (in the standard form “Honour to the venerable Hayagrīva”) in a rubric found in the first folio of an 18th c. manuscript recording a Mīmāṃsā text and compiled by a Vaiṣṇava copyist or scribe (as it is proved by the running invocation “rāma” on each folio). This might be due to the local origins or religious affiliation of the scribe, since the author of the text does not include any invocation to Hayagrīva.77 6.3 Post-Veṅkaṭanātha standardisation 6.3.1 In Icons Recent icons of Hayagrīva are extremely frequent, both in temples, and as depictions on cloth and paper to be distributed to devotees (see two such depictions in section 2.1). They fall into two categories: the Yoga- and the Lakṣmī-Hayagrīva. 6.3.1.1 Yoga-Hayagrīva Some time after Veṅkaṭanātha the icons of Hayagrīva also started to become much more frequent. They also became standardised according to what was found in the HGS, i.e., Hayagrīva held in them conch, discus and book and the gesture of teaching, with the latter two being typical of Hayagrīva alone. No such icon prior to Veṅkaṭanātha has been preserved, but nonetheless it is impossible to rule out the possibility that Veṅkaṭanātha already had an icon of Hayagrīva as an example for his stotra. The Hayagrīva Pāñcarātra Saṃhitā itself constitutes a hint at the possibility of the existence of such icons (see above, section ??). What remains certain is that the standardization of the images is a result of the success of Veṅkaṭanātha’s stotra. Gail presents for instance an image of what I refer here to as Yoga-Hayagrīva from the Vidyāśaṅkara temple, north wall, Śṛṅgeri. Hayagrīva sits in padmāsana and has disc and conch in the upper arms and jñānamudrā and rosary in the lower right hand and Veda books in the lower left hand (Gail 2013, Fig. 9). A further instance of Yoga-Hayagrīva appears to be present in the temple Devanathan Divya Desam at Tiruveṇḍipuram (West of Cuddalore, Madras), which is also closely connected with Veṅkaṭanātha, who is traditionally believed to have lived there, and which is traditionally believed to be the home of the Vaṭakalai sect (see below, section 7). Hayagrīva is found in a smaller shrine in the hill opposite the main temple. He sits in padmāsana and has disc and conch in the upper arms and jñānamudrā in the lower right hand and Veda books in the lower left hand 6.3.1.2 Lakṣmī-Hayagrīva The other standardised image of Hayagrīva initi- ated by Veṅkaṭanātha is that of him in sitting in rajalalitāsana, with Lakṣmī sitting on his left knee and the same attributes as seen above. A Lakṣmī-Hayagrīva image is found also in the Tiruveṇḍipuram temple (see Sridhara Babu 1990, fig. 12a). 77 The manuscript is preserved in the Cambridge Manuscript Library and has been catalogued by Hugo David, see http://cudl.lib.cam.ac.uk/view/MS-ADD-00894/1. 27 Figure 5: Hayagrīva preserved in the Tiruveṇḍipuram temple 6.3.2 In copyists’ and editors’ maṅgalas Well after Veṅkaṭanātha, Hayagrīva started to find a fixed place in the maṅgalas in most Viśiṣṭādvaita Vedāntin works. In this section I will in particular examine the maṅgalas added by copyists and later also by editors to the works of Veṅkaṭanātha. Given that Veṅkaṭanātha is among the figures praised in such maṅgalas, the maṅgala cannot have been composed by him. These maṅgalas can therefore be attributed to the copyists and to the editors of his works, and it is in this connection worth noting that Veṅkaṭanātha’s works have been for the most part copied and edited by Śrī Vaiṣṇavas (see the “State of the art” section in Freschi forthcoming(c)). What is Hayagrīva’s role in this case? Veṅkaṭanātha assumes in the maṅgalas the function which is performed by Gaṇeśa in Śaiva and Smārta texts, i.e., that of a God of learning and of the overcomer of obstacles, the right god to be invoked at the beginning of an intellectual enterprise. This functional equation might have been based on the connection —operated by Veṅkaṭanātha— of Hayagrīva with speech and with knowledge, and might have been also suggested by the similar appearance of the two deities, who both have an animal head and a human body. A further characteristic of Hayagrīva which is frequently mentioned in the maṅgalas is his neighing, which is said to destroy demons (perhaps again for his Vedic connection?). Like in the case of the invocations to Gaṇeśa within a maṅgala, even the invocations to Hayagrīva are quite short. Moreover, copyists and perhaps editors often connect Hayagrīva in the initial praises to their lineage of teachers. See, for instance, the copyist’s or editor’s maṅgala, preceding the editio princeps of Veṅkaṭanātha’s Śatadūṣaṇī : Glory! Honour to Lakṣmī! Honour to the venerable Hayagrīva! Honour to the venerable Rāmānuja! Honour to the venerable great Vedānta Deśika! [This is] the Śatadūṣaṇī, composed by the venerable teacher of Vedānta, who was at home in each system [of thought]. May the venerable Veṅkaṭanātha, the lion among poets and thinkers, the 28 best among the teachers of Vedānta, be always near to me, in my heart!78 Note that the invocations have a gradually narrower focus, suggesting a decreasing structure, from the supreme God to the founder of Viśiṣṭādvaita Vedānta and to its most important teacher. The maṅgala (presumably by the editor) at the beginning of the Śatadūṣaṇī, Vedāntadeśika Granthamālā edition (1940), has the supreme Brahman instead of Lakṣmī and is apart from that identical with the previous one79 The maṅgala (by the editor or copyist?) at the beginning of the printed version of Veṅkaṭanātha’s Nyāyasiddhāñjana is almost identical but here the supreme Brahman is identified with Hayagrīva.80 The maṅgala (by the editor or copyist?) at the beginning of the printed version of Veṅkaṭanātha’s Nyāyapariśuddhi is identical except for the enunciation of the title.81 An almost identical maṅgala is found also at the beginning of the printed version of Veṅkaṭanātha’s Saccaritrarakṣā.82 The maṅgala (by the editor or copyist?) at the beginning of the printed versions of Veṅkaṭanātha’s Adhikaraṇasārāvalī is almost identical, but adds before the invocation to the supreme Brahman in the form of Hayagrīva an invocation to the supreme Brahman in the form of Lakṣmī and Viṣṇu (in his Nṛsiṃha-avatāra).83 A condensed version is found in the editor’s (?) maṅgala to the SM (Viraraghavacharya and Nainaracarya 1971), where only Hayagrīva as supreme Brahman is praised before Veṅkaṭanātha.84 Last, a few other printed editions of works by Veṅkaṭanātha (such as that of the Śrīśaraṇāgatidīpikā, of the Mīmāṃsāpādukā and of the Pāñcarātrarākṣā) open with just the latter part of the same maṅgala, i.e., the praise of Veṅkaṭanātha.85 78 śrīḥ / śriyai namaḥ / śrīhayagrīvāya namaḥ / śrīmate rāmānujāya namaḥ / śrīmate nigamāntamahādeśikāya namaḥ / sarvatantrasvatantraśrīmadvedāntācāryaviracitā śatadūṣaṇī. śrīmān veṅkaṭanāthāryaḥ kavitārkikakesarī | vedāṃtācāryavaryo me sannidhattāṃ sadā hṛdi || (Anantācārya 1901). Here and in the next footnotes the slash indicates a paragraph break in the original not indicated by punctuation. 79 śrīr astu || śrīmatpraṇatārtiharavaradaparabrahmaṇe namaḥ | śrīmate hayagrīvāya namaḥ | śrīmate rāmānujāya namaḥ | śrīmate nigamāntamahādeśikāya namaḥ | śrīmadvedāntadeśikagranthamālā || kavitārkikasiṃhasarvatantrasvatantraśrīmadvedāntācāryaviracitā | śatadūṣaṇī || śrīmān veṅkaṭanāthāryaṃ kavitārkikakesarī | vedāntācāryavaryo me saṃnidhattāṃ sadā hṛdi | 80 śrīḥ / śrīmate hayavadanaparabrahmaṇe namaḥ / śrīmate rāmānujāya namaḥ / śrīmate nigamāntamahādeśikāya namaḥ / śrīmān veṅkaṭanāṭhāryaḥ kavitārkikakesarī / vedāntācāryavaryo me sannidhattāṃ sadā hṛdi / śrīmannigamāntamahādeśikair anugṛhītam / śrīnyāyasiddhāñjanam (Vīrarāghavācārya 1976). 81 śrīḥ / śrīmate hayavadanaparabrahmaṇe namaḥ / śrīmate rāmānujāya namaḥ / śrīmate nigamāntamahādeśikāya namaḥ / śrīmān veṅkaṭanāṭhāryaḥ kavitārkikakesarī / vedāntācāryavaryo me sannidhattāṃ sadā hṛdi 82 śrīḥ / śrīmate hayavadanaparabrahmaṇe namaḥ / śrīmate nigamāntamahādeśikāya namaḥ / śrīmān veṅkaṭanāthāryaḥ kavitārkikakesarī | vedāntācāryavaryo me saṃnidhattāṃ sadā hṛdi || śrīmadvedāntaguruviracitā / saccaritrarakṣā 83 śrīmate lakṣmīnṛsiṃhaparabrahmaṇe namaḥ | śrīmate hayavadanaparabrahmaṇe namaḥ / śrīmate rāmānujāya namaḥ / śrīmate nigamāntagurave namaḥ / śrīmān veṅkaṭanāthāryaḥ kavitārkikakesarī | vedāntācāryavaryo me sannidhattāṃ sadā hṛdi || śrīman nigamāntamahādeśikānugṛhītā / adhikaraṇasārāvalī 84 Śrīḥ śrīhayavadanaparabrahmaṇe namaḥ / śrīmān veṅkaṭanāthāryaḥ kavitārkikakesarī | vedāntācāryavaryo me saṃnidhattāṃ sadā hṛdi || 85 śrīmān veṅkaṭanāṭhāryaḥ kavitārkikakesarī / vedāntācāryavaryo me sannidhattāṃ sadā hṛdi (Pāñcarātrarākṣā); śrīmate nigamāntamahādeśikāya namaḥ | śrīmān veṅkaṭanāthāryaḥ 29 6.3.2.1 Authors’ maṅgalas The maṅgala of the commentary by Abhinava Deśika Vīrarāghavācārya, a 20th c. Śrī Vaiṣṇava scholar, on Veṅkaṭanātha’s SM, is, by contrast, completely different and original. Its originality could be further evidence of the fact that the standardised maṅgalas seen above are due to copyists and not authors: May the venerable Hayagrīva, by whom the whole Veda has been taught, the Veda which had been transferred to earth, which is [Hayagrīva’s] own self, after [Hayagrīva] had taken it back and killed the two asuras, spread out prosperity!86 Similarly, Abhinava Deśika Vīrarāghavācārya’s Satpathasañcāra commentary on Veṅkaṭanātha’s MP opens with a free praise of Hayagrīva: Let me salute the venerable Hayagrīva, praised as the celestial sage, the tutelary deity of all knowledge, who at the beginning manifested [the Vedas] for the sake of prescribing the fruitful which is in the [Vedic] prescriptions [and] for the sake of teaching the Vedas which had been destroyed.87 7 Conclusions on the post-Veṅkaṭanātha diffusion of the standardised Hayagrīva When exactly did this standardisation occur? And when did Hayagrīva become a fixed presence in the maṅgalas? An exact terminus post quem is difficult to settle, given that we have a vast amount of icons of Hayagrīva after the 17th c. which display these standardised traits but this cannot rule out the possibility of earlier, lost, icons with the same characteristics. However, the idea that the major change was realised by Veṅkaṭanātha and was implemented much later fits with other background data. In fact, when Śrī Vaiṣṇavism split into the two schools of Vaṭakalai and Teṅkalai (about whose name and doctrines see Raman 2007), each school recurred to an illustrious predecessor and named him its founder. In the case of the Vaṭakalai, the illustrious predecessor was Veṅkaṭanātha and the Vaṭakalai reinforced this connection by endorsing all of Veṅkaṭanātha’s distinctive points. It is probably also not a coincidence that the famous Yoga-Hayagrīva temple at Tiruveṇḍipuram was in fact only built much later than Veṅkaṭanātha’s time, namely in 1667, at the time of the Teṅkalai-Vaṭakalai split, when one needed to show off one’s connection with Veṅkaṭanātha. The main icon of this temple has been discussed before, Fig. 5. 8 Various types of reuse From the point of view of reuse, the reuse of Hayagrīva acquires different meaning in each context: kavitārkikakesarī | vedāntācāryavaryo me saṃnidhattāṃ sadā hṛdi || (Śrīśaraṇāgatidīpikā); śrīḥ śrīmān veṅkaṭanāthāryaḥ kavitārkikakesarī | vedāntācāryavaryo me saṃnidhattāṃ sadā hṛdi || kavitārkikasiṃhasarvatantrasvatantraśrīmadvedāntācāryaviracitā mīmāṃsāpādukā (Viraraghavacharya and Nainaracarya 1971). 86 śrīmān vājimukhaḥ śriyaṃ vitanutāṃ yenopadiṣṭo ’khilo vedaḥ svātmabhuve ’rpitaś ca punar apy āhṛtya hatvāsurau | […] || 1 || (Viraraghavacharya and Nainaracarya 1971). 87 vandeya śrīhayāsyaṃ vidhihitavidhaye naṣṭavedopadiṣṭaye labdhāvirbhāvam ādau suramunivinutaṃ sarvavidyādhidevam | (Viraraghavacharya and Nainaracarya 1971). 30 • In the late Vedic literature, in the Purāṇas and in the Mahābhārata: Hayagrīva is a minor form of Visṇu among many. • In the Pāñcarātra texts: Hayagrīva is a fixed avatāra of Viṣṇu. The reference to the demon with the same name (see section 3.2.2) seems to have disappeared. • For Veṅkaṭanātha: Hayagrīva is himself the supreme deity. • After Veṅkaṭanātha: Hayagrīva as described by Veṅkaṭanātha becomes a fixed reference point. The reasons behind Veṅkaṭanātha’s turn charged the latter two kinds of reuse with new meanings: 1. belonging to an intellectual tradition 2. belonging to a religious tradition The reuse of Hayagrīva as a mark of one’s belonging to a precise intellectual tradition might lie at the basis of Veṅkaṭanātha’s choice, i.e., Veṅkaṭanātha’s pro-Vedism. Later, the presence of Hayagrīva was probably a sort of trade mark to be shown immediately (e.g., already in the maṅgala) by the adherents of the Vaṭakalai school as a mark signalling one’s belonging to Veṅkaṭanātha’s tradition As for the terminological introduction regarding the various forms of reuse, imagining the departure point as Hayagrīva as found in the Mahābhārata:88 • The reuse of Hayagrīva in the Purāṇas and in Vaiṣṇava texts is (at the present stage of research) a case of simple re-use, insofar as the purpose of Hayagrīva is very similar and the modifications are probably due to gradual changes rather than to conscious shifts. • The reuse of Hayagrīva in Veṅkaṭanātha, by contrast, is a case of adaptive reuse, since Hayagrīva shifts from minor avatāra to supreme deity and acquires new functions (connection with speech and with knowledge, standardisation of his Vaiṣṇava attributes). • The reuse of Hayagrīva in the Vaṭakalai school is again a case of adaptive reuse, since he becomes a sort of sectarian mark. Further: • The hymns to Hayagrīva written after Veṅkaṭanātha offer examples of quotations of Veṅkaṭanātha’s Hayagrīvastotra. • The maṅgalas to Hayagrīva after Veṅkaṭanātha resemble each other so much that it is difficult to say that they are quoting something. They are rather a case of interlanguage, since the reference to Hayagrīva was extremely common among Vaṭakalai scholars. • The situation with the image of Hayagrīva is more complicated. It is possible that there was a Yoga-Hayagrīva icon before Veṅkaṭanātha and that he openly referred to it while describing Hayagrīva in his stotra, although nothing remains which could validate this hypothesis. • Last, the icons of Hayagrīva after Veṅkaṭanātha might have at first been an explicit reference to his stotra, but they nowadays represent an interlanguage way of iconographically representing Hayagrīva. 88 One might speculate on the reuse of the horse motif after the Vedic period in the Mahābhārata but due to the scarcity of documents it would be difficult to substantiate such speculations. 31 9 Abbreviations AP Agni Purāṇa HS Hayaśīrṣa Pāñcarātra, see Dutta Sastri 1976 HGS Hayagrīvastotra by Veṅkaṭanātha MP Mīmāṃsāpādukā by Veṅkaṭanātha SM Seśvaramīmāṃsā by Veṅkaṭanātha VDhP Viṣṇudharmottarapurāṇa VS Vedārthasaṅgraha by Rāmānuja References AA.VV. (1999). Ideologie e pratiche del reimpiego nell’alto Medioevo : settimane di studio del Centro italiano di studi sull’alto medioevo, 46. : 16-21 aprile 1998. Spoleto: Centro italiano di studi sull’Alto Medioevo. Adhikaranasāravalī śatadūsanī [by Veṅkaṭanātha] (1940). Vedāntadeśika granthamālā. Madaras: Libarti mudranalaye. Agrawala, R.C. (1961). “Three Unique and Unpublished Sculptures of Hayagrīva from Rajasthan”. In: Journal of the Oriental Institute XI.3, pp. 281–282. Anantācārya, Prativādibhayaṅkara, ed. (1901). Śatadūṣaṇī [by Vedānta Deśika]. Śrīkāñcī: Sudarśanamudrākṣaraśālā. Andreae, Bernard and Salvatore Settis, eds. (1983). Colloquio sul reimpiego dei sarcofagi romani nel Medioevo, Pisa 5–12 Settembre 1982/Marburger Winckelmann-Programm 1983. Marburg/Lahn: Verlag des kunstgeschichtlichen Seminars. Asher, Catherine B. and Thomas R. Metcalf, eds. (1994). Perceptions of South Asia’s Visual Past. New Delhi: American Institute of Indian Studies. Bignami, Cristina (2014). “Re-use in the Art Field: The Iconography of Yakṣī”. In: Journal of Indian Philosophy (Special Issue). Ed. by Elisa Freschi, pp. 1– 24. doi: 10.1007/s10781- 014- 9250- 7. url: http://dx.doi.org/10. 1007/s10781-014-9250-7. Chandra, Lokesh (1991). Buddhist Iconography. Delhi: International Academy of Indian Culture and Aditya Prakashan. D฀bicka-Borek, Ewa (2015). “To Borrow or not to Borrow? Some Remarks on vaibhavīyanarasiṃhakalpa of Sātvatasaṃhitā”. In: Journal of Indian Philosophy (Special Issue) 43. Ed. by Elisa Freschi. Desai, D. (1993). “Hayagrīva in the Vaikuṇṭha pantheon of Khajurao”. In: Viṣṇu in Art, Thought and Literature. Ed. by G. Kamalakar and M. Veerender. Hyderabad: BIrla Archaeological & Cultural Research Institute, pp. 47–50. Desai, Kalpana S. (1973). Iconography of Viṣṇu (In Northern India, Upto the Mediaeval Period). New Delhi: Abhinav Publications. DiCostanzo, Thierry (2012). “Use and Re-use of ‘Pakistan’ in the Indian Muslim Press (1932–47)”. In: Re-use. The art and politics of appropriation and anxiety. New Delhi; Thousand Oaks, Calif.: SAGE. Chap. 11, pp. 227–255. Doniger, Wendy (1980). Women, androgynes, and other mythical beasts. Chicago: Univ. of Chicago Press. 32 Dutta Sastri, Kali Kumar, ed. (1976). Hayaśīrṣa Pañcarātra. Ādi Kāṇḍa. Bibliotheca Indica — A Collection of Oriental Works. Calcutta: The Asiatic Society. Freschi, Elisa (2015a). “Between Theism and Atheism: a journey through Viśiṣṭādvaita Vedānta and Mīmāṃsā”. In: Puṣpikā: Tracing Ancient India Through Texts and Traditions. Contributions to Current Research in Indology. Number 3. Ed. by Robert Leach and Jessie Pons. Oxford: Oxbow Books Press, pp. 24–47. — (2015b). “The reuse of texts in Indian Philosophy”. In: Journal of Indian Philosophy (Special Issue) 43.2–3. Ed. by Elisa Freschi, pp. 85–108. — (forthcoming[a]). “Are the limbs of God’s body free? Yes, if He wants so — Free will in and before Veṅkaṭanātha”. In: Fate, Freedom and Prognostication in Indian Traditions. Ed. by Marcus Schmücker and Sven Sellmer. Wien: Verlag der Österreichischen Akademie der Wissenschaften. — (forthcoming[b]). “Reusing, Adapting, Distorting. Veṅkaṭanātha’s reuse of Rāmānuja, Yāmuna and the Vṛttikāra in his commentary ad PMS 1.1.1”. In: Adaptive Reuse of Texts, Ideas and Images in Classical India. Ed. by Elisa Freschi and Philipp A. Maas. Abhandlungen für die Kunde des Morgenlandes. Wiesbaden: Harrassowitz. — (forthcoming[c]). “Veṅkaṭanātha”. In: Internet Encyclopedia of Philosophy— Indian Philosophy. http://www.iep.utm.edu/. Freschi, Elisa and Philipp André Maas (forthcoming). “Introduction”. In: Adaptive Reuse of Texts, Ideas and Images in Classical India. Ed. by Elisa Freschi and Philipp André Maas. Wiesbaden: Deutsche Morgenländische Gesellschaft. Harrassowitz. Gail, Adalbert J. (1977). Paraśurāma, Brahmane und Krieger. Untersuchung über Ursprung und Entwicklung eines Avatāra Viṣṇus und Bhakta Śivas in der indischen Literatur. Wiesbaden: Harrassowitz. — (2013). “Hayagrīva or the making of an avatāra”. In: Pandanus 7.1, 139–149 (plus 11 pp. of images). Gopinatha Rao, T.A. (1968). Elements of Hindu Iconography. 2nd (1st Madras 1914). Vol. I—Part II. New York: Paragon Book Reprint Corp. Hegewald, Julia A.B. (2012). “Towards a Theory of Re-use: Ruin, Retro and Fake versus Improvement, Innovation and Integration”. In: Re-use. The art and politics of appropriation and anxiety. Ed. by Julia A.B. Hegewald and Subrata K. Mitra. New Delhi; Thousand Oaks, Calif.: SAGE. Chap. 2, pp. 30–54. Hegewald, Julia A.B. and Subrata Kumar Mitra, eds. (2012). Re-use: the art and politics of integration and anxiety. New Delhi; Thousand Oaks, Calif.: SAGE. Iyanaga, Nobumi (2002). Kannon henyō tan: Bukkyō shinwa gaku 2 ฀฀฀฀฀:฀฀฀฀฀ II (Metamorphoses of the Bodhisattva Avalokiteśvara. Essays of Buddhist Mythology 2). Kyōto: Hōzōkan. Jaiswal, Suvira (1981). The Origin and Development of Vaiṣṇavism (Vaiṣṇavism from 200 B.C. to A.D. 500). Delhi: Munshiram Manoharlal. 33 Joshi, N.P. (1972–73). “Hayagrīva in Brahmanical iconography”. In: Dr. Vasudev Saran Agrawala Commemoration Volume. Part II. Ed. by U.P. Shah. Vol. V. Calcutta: The Indian Society of Oriental Art, pp. 36–42. Larios, Borayin (2011). “Sacred sound becomes sacred scripture: the Veda Mandir in Naśik, Mahārāṣṭra”. In: Travaux de Symposium International Le Livre. La Roumanie. L’Europe, Troisième édition - 20 à 24 Septembre 2010- Tome III: La troisième section — Études Euro- et Afro-Asiatiques. Ed. by Jan E.M. Houben and Julieta Rotaru. Bucarest: Éditeur Bibliothèque de Bucarest, pp. 233–244. Leach, Robert (2012). “Textual Traditions and Religious Identities in the Pāñcarātra”. PhD thesis. Edinburgh: University of Edinburgh. Linrothe, Rob (1999). Ruthless Compassion. Wrathful Deities in Early IndoTibetan Esoteric Buddhist Art. London: Serindia Publications. Lo Muzio, Ciro (forthcoming). “Skandha and the Mothers in Khotanese Buddhist Painting”. In: Interaction in the Himalayas and Central Asia: processes of transfer, translation and transformation in art, archaeology, religion and polity from antiquity to the present day. Proceedings of the Third International SEECHAC Colloquium, Austrian Academy of Sciences, Vienna, 25-27 November 2013. Ed. by F. Grenet et al. Vienna. Madhavan, Chithra (2013). Sanskrit education and literature in ancient and medieval Tamil Nadu. An epigraphical study. New Delhi: D.K. Agencies. Mallmann, Marie-Thérèse de (1963). Les einsegnements iconographiques de l’AgniPurana. Paris: Presses Universitaires de France. — (1986). Introduction à l’Iconographie du Tântrisme Bouddhique. Paris: Librairie d’Amérique et d’Orient. Maisonneuve. Maxwell, T.S. (1992–1993). Vaikuṇṭha-Viśvarūpa Vol. III. Katalog des archäologischen Materials. August 1992–September 1993. url: http://ignca.nic. in/visv0305.htm. McDonough, William and Michael Braungart (2002). Cradle to Cradle: Remaking the Way We Make Things. North Point Press. Nayar, Nancy A. (1994). Praise-poems to Viṣṇu and Śrī: The Stotras of Rāmānuja’s Immediate Disciples : a Translation from the Sanskrit with Introduction and Notes. Ananthacharya Indological Research Institute series. Ananthacharya Indological Research Institute. Patel, Alka (2009). “The Historiography of Reuse in South Asia”. In: Archives of Asian Art 59, pp. 1–5. Pattanaik, Devdutt (2006). Myth = Mithya. A handbook of Hindy mythology. New Delhi: Penguin Books India. Plevoets, Bie and Koenraad Van Cleempoel (2011). “Adaptive reuse as a strategy towards conservation of cultural heritage: a literature review”. In: Structural Studies, Repairs and Maintenance of Heritage Architecture XII. Ed. by C. Brebbia and L. Binda. Chianciano Terme: WIT press. — (2013). “Adaptive reuse as an emerging discipline: an historic survey”. In: Reinventing architecture and interiors: a socio-political view on building adaptation. Ed. by G. Cairns. London: Libri Publishers, pp. 13–32. 34 Priyabala, Shah, ed. (1958). Viṣṇudharmottarapurāṇe tṛtīyakhaṇḍaḥ. Vol. 1: [Critically edited] Text, Critical Notes etc. Gaekward’s Oriental Series CXXX. Baroda: Oriental Institute. Raghavan, S.S., M.S. Lakshmi Kumari, and M. Narasimhachary, eds. (1995). Sri Vedanta Desika’s Stotras (With English Translation). Chennai: Sripad Trust. Raman, Srilata (2007). Self-Surrender (prapatti) to God in Śrīvaiṣṇavism: Tamil cats and Sanskrit monkeys. RoutledgeCurzon Hindū Studies Series. London: Routledge. Rastelli, Marion (2006). Die Tradition des Pāñcarātra im Spiegel der Pārameśvarasaṃhitā. Wien: Verlag der Österreichischen Akademie der Wissenschaften. — (2007). “The “Pāñcarātra Passages” in Agnipurāṇa 21–70”. In: Mélanges tantriques à la mémoire d’Hélène Brunner. Ed. by Dominic Goodall and André Padoux. Pondichéry: Institut français de Pondichéry, pp. 187–229. — (forthcoming). “Narratives as a medium for appealing to the royal court: A look into the Ahirbudhnyasaṃhitā”. In: Tantric Communities in Context. Sacred Secrets and Public Rituals. Ed. by Nina Mirnig, VIncent Eltschinger, and Marion Rastelli. Wien: Österreichische Akademie der Wissenschaften. Sanderson, Alexis (2007). “Atharvavedins in Tantric Territory. The Āṅgirasakalpa Texts of the Oriya Paippalādins and their Connection with the Trika and the Kālīkula. With critical editions of the Parājapavidhi, the Parāmantravidhi, and the *Bhadrakālīmantravidhiprakaraṇa”. In: The Atharvaveda and its Paippalādaśākhā. Historical and Philological Papers on a Vedic Tradition. Ed. by Arlo Griffiths and Annette Schmiedchen. Vol. 11. Geisteskultur Indiens. Texte und Studien. Indologica Halensis. Aachen: Shaker Verlag, pp. 195– 311. Schöttli, Jivanta (2012). “Myth, Idea, Dream and Vision: Nehru’s Discovery of India”. In: Re-use. The art and politics of appropriation and anxiety. New Delhi; Thousand Oaks, Calif.: SAGE. Chap. 10, pp. 209–226. Sridhara Babu, D. (1990). Hayagrīva. The Horse-Headed Deity in Indian Culture. Tirupati: Sri Venkatesvara University. Oriental Research Institute. Srinivasan, Doris Meth (2004). “Review of The Citraūtra of the Viṣṇudharmottara Purāna, by Parul Dave Mukherji”. In: Journal of the American Oriental Society 124.3, pp. 569–572. Sumant, Shilpa (2010). “Iconography of Nṛsiṃha represented in the Karmapañjikā of the Paippalādins”. In: Journal of the Oriental Institute 60.1–2, pp. 57–66. Todisco, Luigi (1994). Scultura antica e reimpiego in Italia meridionale. Bari: Edipuglia. Tripathi, Gaya Charan (1978). “On the Concept of ‘Puruṣottama’ in Āgamas”. In: The Cult of Jagannatha and the Regional Tradition of Orissa. Ed. by Anncharlott Eschmann, Hermann Kulke, and Gaya Charan Tripathi. New Delhi: Manohar. Chap. 2, pp. 31–59. Van Gulik, R.H. (1935). “Hayagrīva. The Mantrayānic Aspect of Horse-Cult in China and Japan”. In: Leiden: E.J. Brill. 35 Vīrarāghavācārya, T., ed. (1976). Nyāya Siddhāñjana by Vedānta Deśika with two old commentaries [Saralaviśadavyākhyā by Śrīraṅgarāmānujasvāmi and Ratnapeṭikā by Śrīkāñcī Kṛṣṇatātayārya, including a Ṭippaṇa by the editor]. Ubhayavedāntagranthamālā. [Madras]. Viraraghavacharya, Uttamur T. and Nainaracarya, eds. (1971). SeśvaramīmāṃsāMīmāṃsāpaduke, Seswara Mimamsa and Mimamsa paduka [by Veṅkaṭanātha]. Madras: Ubhaya Vedanta Granthamala. Viṣṇudharmottaramahāpurāṇa (1912). Bombay: Śrī Veṅkaṭeśvara. 36