Printer Friendly

The Citrasutra of the Visnudharmottara Purana.

The Citrasutra of the Visnudharmottara Purana. By PARUL DAVE MUKHERJI. Delhi: INDIRA GANDHI NATIONAL CENTRE FOR THE ARTS, 2001. Pp. xlv + 293. Rs. 750.

For more than seventy-five years, ever since Stella Kramrisch brought out the first printed edition in translation, the Citrasutra of the Visnudhamottara Purana has been recognized as an important treatise for art historians and, to some degree, for historians of religion and literary developments in India. The Citrasutra refers to Khanda III, Adhyayas 35-43, of the Visnudharmottara Purana. In its entirety, Khanda III propounds an early and remarkable view of the interrelationship of the arts (see, for example, III.2.1-9 on the interrelationship between painting, dancing, music, and singing), and it articulates the framework of an aesthetic theory relevant to these arts. The Citrasutra, named after the first line of Adhyaya 35.1a (atah param pravaksyami citrasutram tavanagha), concentrates on the theory of painting, although occasional reference to sculpture is also made throughout the text. Adhyaya 35 considers the mythic origin of painting and the five types of males together with their differing proportions. Adhyaya 36 discusses measurements and proportions of the different parts of the body and the colors and other distinguishing features of the five male types. Adhyaya 37 deals with the measurements of the five types of females, hair and eye types, and the general characteristics of a Cakravartin. Adhyaya 38 gives details on auspicious marks that divine images, both plastic and painting, will possess. Adhyaya 39 treats the different postures (sthanas) for figures. Adhyaya 40 describes how to mix paints, prepare the surface, and apply the paints. Adhyaya 41, of cardinal importance, defines the four types of paintings. Adhyaya 42, equally significant, prescribes the manner in which a large number of beings--royalty, priests, nature and heavenly sprites, demons, wives, courtesans, attendants of vaisnava deities, warriors, merchants, sudras and more--should be depicted. We may infer that all these subjects were, or could be, depicted in art, otherwise prescriptions such as these would not be provided. Adhyaya 43 talks about the nine rasas in painting, strengths and defects in painting, as well as sculpture in different materials. In closing, III.43.37, as if to underscore the unity and interdependence of the arts, states that whatever has been left unsaid about painting can be understood from the section on dance, and what is not given there can be supplied from painting.

An impressive group of Indologists and pioneers in the history of Indian art have analyzed this text. They seem to agree that Kramrisch's 1924 translation of the Citrasutra and her revised version of 1928 (1) have many problems, due to the faulty edition she worked with and her reliance on others; however, her introduction is still worth reading, since such issues as art contemporary to the text (e.g., Ajanta), its date (between the fifth and seventh century A.D.), and the text's general characteristics are perceptively outlined. A. K. Coomaraswamy agreed with Kramrisch on dating and the text's contemporaneity with Ajanta; his annotated translation solely of III.41 in this journal (2) was undertaken, I would think, by a desire to fathom the nature of Indian painting treated in an early text. His analysis is more philologically oriented than Kramrisch's. C. Sivaramamurti's 1978 translation of the Citrasutra is accompanied by an extensive discussion of the text based on his immense knowledge of Indian literature and art. (3) Although Sivaramamurti's textual and visual comparisons are often ahistorical, he too is concerned with the date of the text (somewhere between the Kushan and Gupta periods). The Visnudharmottara (Khanda III.1-118) brought out in two volumes by Priyabala Shah, a Sanskritist, has strong merits. (4) Her edition incorporated the evidence from six new manuscripts, with the variant readings of a seventh given in an Appendix. Shah's arguments for an early date (450-650) influenced Sivaramamurti's position. These four are but the main in-depth analyses of the text; many more scholarly works are cited in the edition under review.

The critical edition by Parul Dave Mukherji was produced under the guidance of Professor Alexis Sanderson, for an Oxford University Ph.D. Dr. Mukherji is now a Reader of Art History and Aesthetics at M S University, Baroda. In other words, Mukherji had chosen to be an art historian with a solid background in textual criticism and Sanskrit. This background prompts her goals in the new edition of the Citrasutra. The editor aims to reconstruct the text on the basis of two new manuscripts she located, and she wishes to rework the text in the face of post-colonial concerns. Whereas the first succeeds very well, the second falters.

Mukherji's reconstruction marks a considerable step forward in understanding difficulties in this text. In her introduction she provides profiles of the new manuscripts: one is written in Newari script and is located in the National Archives, Kathmandu, Nepal; the other is written in Bengali script and is in the Dhaka University Library, Bangladesh. Mukherji fits these two manuscripts into a genealogy of known manuscripts, showing why the two new ones figure so importantly in her reconstruction of the text. Mukherji postulates a lost archetype from which descended two branches. The oldest known MS of the first branch is a birch bark MS, whose Sarada paleographic features support its antiquity. The birch bark MS and the other MSS Shah used in her edition belong to this lineage. The Newari and Bengali MSS represent a line of descent on a separate branch of transmission from the lost archetype. Mukherji argues well for the descent of the Newari and Bengali MSS from a lost MS on a par with the birch bark manuscript. Thus the Newari and Bengali MSS relate to the first branch, though they have not directly descended from it. The editor, therefore, not only has two new MSS to work with, but a new exegetical tool, for wherever the Newari and Bengali MSS differ from each other, Mukherji chooses the reading that agrees best with the birch bark MS. Separating the two lineages descending from the archetype, of course, helps in establishing preferential readings. The way readings are chosen is made clear in the way the translation is organized and presented. Each adhyaya has the critical text on the left side, with variant Sanskrit readings in the footnotes below, and the translation on the right side. Notes following each adhyaya explain how and why Mukherji's translation differs from that of her predecessors, and provide citations from other silpa texts. So, for example, out of the forty-seven verses in Adhyaya 39 "On Postures" the notes inform that the new material assists in approximately twenty readings. In my opinion, the most significant reconstruction occurs in III.41.1 dealing with the four types of paintings.

Adhyaya 41.1 reads: satyam ca daisikam caiva nagaram misram eva ca / citram caturvidham proktam tasya vaksyami laksanam//. Mukherji translates: "Painting is said to be of four types: Satya ('Naturalistic'), Daisika (Provincial/Local), Nagara (Urban/Professional) and Misra (Mixed). I shall speak about its (their) characteristic(s)." None of the other translations reads "daisika"; Kramrisch here reads "vainika" ('lyrical'), which is followed by Coomaraswamy' and Sivaramamurti, in spite of recognized difficulties; Shah accepts vainika and says it means "born or coming from Vena," a place name she associates with regions in South India, and thus concludes that the type of painting is the Dravida type (pp. 119-24 in Shah, vol. II, and Mukherji, pp. 164-66, where both scholars also refer to the many other scholars who commented on this important passage). Mukherji conjectures daisika on the basis of "sense and paleography." Sense requires a term signifying a type of painting contrasting with the Nagara type. If the Nagara type is understood as relating to a city or an urban context, the counterpart would be Daisika as in Desi or provincial or local context. With regard to paleography, she argues that if the archetype MS was in Newari script, the reconstruction would be as follows; daisika (Mukherji's conjecture) ... > vaisika (as in her new Newari MS) ... > vainika (as in MSS known to earlier editions) ... > venika (as in the birch bark MS plus another in Sarada script). "Moreover," says Mukherji, "in Newari script, -da- and -va- do not resemble one another as much as they follow the same direction graphically. By repeated copying from one MS to another, if the gap in -da- is not maintained, it would turn into -va-" (164). Thus, Mukherji's emendation improves our understanding of the four types (? styles) of paintings; in today's art-historical parlance the pair daisika/nagara reminds of types relating to "urban center and provincial periphery."

Mukherji's exegesis of the term satya in III. 41.1 provides a good example of the colonial context she aims to eliminate by reworking the text from the vantage point of "post-colonial concerns." Her translation of satya as "naturalistic" differs from Coomaraswamy's rending of satya as "pure and sacred" and Kramrisch's "true and realistic." Mukherji contends that the translations of these predecessors were colored by a politico-social desire to defend Indian art against the prevailing colonial opinion that "naturalism" was lacking in Indian art. Thus she discusses at considerable length (pp. xxxiv-xxxviii) why Coomaraswamy chose not to render satya as "naturalistic" painting, stating that according to a "nationalistic agenda" Indian art was not to be evaluated according to Western artistic norms. A balanced reply to this viewpoint is presented in the foreword written by Kapila Vatsyayan, at the time Director of the Indira Gandhi National Centre for the Arts and general editor of the Centre's Kalamulasastra Series in which this volume was published. Vatsyayan illustrates Coomaraswamy's careful philological approach in defining terms by quoting entries on sadrsya ('similitude') from his works. Further insight, I believe, into Coomaraswamy's choice of "pure and sacred" for satya painting may be hidden in footnote 1 of his JAOS article. Coomaraswamy notes that he sought the help of Professor William Norman Brown in solving textual difficulties. Having myself been a student of Brown and part of many Vedic translation sessions, I can quite imagine Brown's input for satya as "pure and sacred" here, understanding satya in terms of sat, a metaphysical concept that continually engaged his attention. (5) Coomaraswamy's textual commentary reflects this idea, for he believes that satya painting contains subject matter derived from "the world of the gods or any other sphere" (JAOS 52 [1932]: 15). I see little evidence for Mukherji's contention of an interpretation according to a "nationalistic agenda." At this point in our critical understanding of this silpa text, perhaps we should use the neutral definition cited by Sivaramamurti (on p. 66): satya is painting "portraying some object of the world that it intends to represent."

Another characteristic of this edition is that the editor limits her evidence for solving textual issues to evidence within the Citrasutra alone, thereby rarely using the Visnudharmottara Purana as a whole. This approach influences her date for the text. Mukherji postulates that the original composition of the text was between 500 and 900 A.D. Thus, she is not influenced by the date of the first half of the fifth century A.D., to which Pingree assigns the Paitamahasiddhanta in Khanda II of the Visnudharmottara, (6) believing that it is difficult to ascertain the gap in time between the two khandas. She further justifies her date by citing the close correspondence between a description of Visnu in III.44 and a Kashmiri vaisnava sculpture, published by P. Pal in 1973-74; (7) the sculpture, called Vaikuntha, or Para Vasudeva-Narayana, has the human head of Vasudeva in the center, the lion's head to the right, and the boar's head on the left. However, in 1997 I published an inscribed bronze example of this sculptural type which Lore Sander dated, by the inscription, to 427. (8) Para Vasudeva-Narayana expresses beliefs pertaining to the Pancaratra sect within Vaisnavism. A characteristic tenet of this sect is the vyuha doctrine, according to which the transcendental god differentiates himself into four vyuhas (emenations), namely Vasudeva, Samkarsana, Pradyumna, and Aniruddha, in that order. That order can be recognized in the sequence of names in Khanda III.42.21-22, wherein the Citrasutra prescribes the appearance of attendants to these four vaisnava gods. The same sequence of names also occurs before the Citrasutra (III.32 and 33), and after it (III.44 and 85). Manifestly, the Visnudharmottara seems to be a Pancaratra text originating in northern India. Recurrence of the sequence of names, indicative of the vyuha doctrine, suggests not only that the Citrasutra is probably familiar with the doctrine and its sculptural representation, but also that interpretation of the Citrasutra benefits from knitting the text into the fabric of the Visnudharmottara Purana as a whole. This example has dating implications. For, if the Citrasutra has knowledge of the vyuha doctrine, and if vyuha imagery described in a post-Citrasutra passage corresponds with an image dated 427 A.D., then a date for the Citrasutra closer to Pingree's date seems to gain ground. Further support for an early date is based on the text's seeming contemporaneity with the fifth-century paintings at Ajanta. But scholars have found reason to date the Visnudharmottara anywhere between the fifth and the eleventh centuries, and more work is needed to settle this question. (9)

This critical edition and translation of the Citrasutra is a very useful tool for art historians and a welcome study for Indologists, but it would have benefited from additional material. More extensive references to art corresponding to textual prescriptions would have enhanced the work considerably. Another lacuna is charts illustrating the orthography of Sarada, Newari, and Bengali aksaras. Mukherji supports numerous emendations on orthographic properties in the scripts of the various Citrasutra MSS; she frequently cites orthographic similarities of certain Sarada aksaras as the cause for confusion or corrupt forms in MSS (e.g., pp. 73, 98, 101-2, 106-7, 110, 222, 234, etc.). Mention of orthographic similarities in the Newari and Bengali scripts occurs too, though less frequently. Without charts, many readers may not be able to visualize the orthographic properties of which she speaks and therefore not follow her arguments. Also, the book could have used one more check of the proofs, as some bibliographic typos have crept in. Nonetheless, as it stands, Mukherji's work, a careful undertaking of a difficult text, advances our understanding of citra in ancient India and represents a welcome position from which to continue refining our understanding of this text.



1. "The Vishnudharmottaram: A Treatise on Indian Painting," The Calcutta Review, 3rd ser., 10. no. 2 (Feb. 1924): 331-86; The Vishnudharmottaram (Pt. III): A Treatise on Indian Painting and Image-Making (revised and enlarged edition). (Calcutta: Univ. of Calcutta, 1928).

2. "Visnudharmottara, Chapter XLI," JAOS 52 (1932): 13-21.

3. Chitrasutra of the Vishnudharmottara (New Delhi, 1978).

4. Visnudharmottara-Purana, Third Khanda; vol. I: Text, Critical Notes, etc. (Baroda, 1958); Visnudharmottara-Purana, Third Khanda, vol. II: Introduction, Appendixes, Indexes, etc. (Baroda, 1961).

5. See, e.g., Brown, Man in the Universe: Some Continuities in Indian Thought (Berkeley and Los Angeles, 1966), 16-42.

6. David Pingree, "The Paitamahasiddhanta of the Visnudharmottarapurana," Adyar Library Bulletin 31-32 (1967-68): 472-510.

7. "A Brahmanical Triad from Kashmir and some Related Icons," Archives of Asian Art 27 (1973-74): 33-45.

8. Doris Meth Srinivasan with Lore Sander, "Visvarupa Vyuha Avatara: Reappraisals Based on an Inscribed Bronze from the Northwest Dated to the Early 5th Century A.D.," East and West 47 (1997): 105-70.

9. See Ludo Rocher, The Puranas (Wiesbaden, 1986), 252. I now feel less convinced of the eighth-century date I gave in "Visvarupa," 106.
COPYRIGHT 2004 American Oriental Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2004 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Srinivasan, Doris Meth
Publication:The Journal of the American Oriental Society
Article Type:Book Review
Date:Jul 1, 2004
Previous Article:Early Tamil Epigraphy: From the Earliest Times to the Sixth Century A.D., vol. 62.
Next Article:The Dravidian Languages.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters