Script doctors' blunders mar valuable texts

November 11, 2005

Misconceived typography spoils John Smith's delight in a collection of Sanskrit classics accessible to modern readers

The Clay Sanskrit Library Edited by Richard Gombrich

The Birth of Kumára By Kalidasa ranslated by David Smith 360pp ISBN 0 8147 4008 1

The Emperor of the Sorcerers, Volume One By Budhasvamin Edited and translated by James Mallinson 452pp ISBN 0 8147 5701 4

The Epitome of Queen Lilávati, Volume One By JinaratnaEdited and translated by R. C. C. Fynes 543pp ISBN 0 8147 41 7

The Heavenly Exploits: Buddhist Biographies from the Dívyavadána, Volume One Edited and translated by Joel Tatelman 444pp ISBN 0 8147 8288 4

Love Lyrics By Amaru, Bhartrhari and Bilhana Edited and translated by Greg Bailey and Richard Gombrich 326pp ISBN 0 8147 9938 8

Maha-bharata Book Three: The Forest, Volume Four Translated by William J. Johnson 374pp ISBN 0 8147 48 5

Much Ado about Religion By Jayanta Bhatta Edited and translated by Csaba Dezsõ 320pp ISBN 0 8147 1979 1

Ramáyana Book One: Boyhood By Vaxlmiki Translated by Robert P. Goldman 424pp ISBN 0 8147 3163 5

Ramáyana Book Two: Ayodhya By ValmikiTranslated by Sheldon I. Pollock 652pp ISBN 0 8147 6716 8

The Recognition of Shakúntala By Kalidasa Edited and translated by Somadeva Vasudeva 350pp ISBN 0 8147 8815 7

Three Satires By Nilakantha, Ksemendra and Bhallata Edited and translated by Somadeva Vasudeva 403pp ISBN 0 8147 8814 9

What Ten Young Men Did By Dandin Translated by Isabelle Onians 651pp ISBN 0 8147 6206 9

All published by New York University Press/JJC Foundation Distributed by Eurospan £14.95 each

If Greek is green and Latin is red, what colour is Sanskrit? The answer, it transpires, lies somewhere between cyan and Cambridge blue, for this is the colour of the dustjackets adorning a handsome new series of dual-language Sanskrit texts that is clearly inspired by the venerable Loeb Classical Library. The Clay Sanskrit Library was launched in February with the publication of the first twelve titles, to be followed by six others in 2005; the aim is to produce about 100 volumes within the next five years.

No effort has been spared to make these little volumes as attractive as possible to readers: the paper is of high quality, the typesetting immaculate. (The Loeb logo depicting Athena has been replaced by one showing Sarasvatl-, the Hindu goddess of learning and literature.) The founders of the series are John and Jennifer Clay, and Sanskritists can only thank them for an initiative intended to make the classics of an ancient Indian language accessible to a modern international audience.

The first 12 offerings form a mixed bag. There are sections of the two great epics, Mahabharata and Ramayana ; two dramas; a cycle of picaresque tales; Kalidasa's "court epic" on the birth of the god of war; as well as love lyrics, satires and story collections of Buddhist, Jain and Hindu provenance.

These very varied volumes are the work of 11 different scholars overseen by a general editor, Richard Gombrich, so it is hardly surprising that they embody a number of slightly different approaches to the basic task of presenting a text with a facing English translation. The Ramayana volumes, for example, are abridged reprints of the heavily annotated translation that has been appearing from Princeton University Press over the past two decades. That work was based on the standard Baroda critical edition of the epic, and the text of that edition is accordingly printed on the left-hand pages of the Clay volumes. The Clay version of the Mahabharata , by contrast, is based not on the standard Poona critical edition but on the old "Bombay edition" - that is, the version known to the commentator Nl-lakan.t.ha. Perhaps copyright problems prevented the use of the scholarly text - no explanation is offered for this slightly puzzling choice.

Various of the texts are said to have been edited, as well as translated, by the scholars who have produced them. It turns out, though, that the verb "edit" can mean more than one thing. Somadeva Vasudeva's text of Ksemendra's Kalavilasa is based on three manuscripts and three earlier editions, and Csaba Dezsõ's text of Jayanta Bhatta's A gamadambara on two manuscripts and one earlier edition. In other cases, however, the editorial aspect of the translator's task appears to be limited to proposing occasional emendations to a previous scholarly edition. Probably the most striking interpretation of the editorial role is made by Gombrich himself, who has edited and translated Bilhan.a's Caurapañcasika , and who writes, for example, that he has preferred the version of the 12th stanza found in an edition other than the unnamed "best" one on which he generally relies, because "stanzas 12 and 13 of the best text are so empty and repetitious, partly of notions difficult to render tasteful to a modern reader, that I have combined them in my stanza 13".

It is inevitable that translation style also varies from volume to volume. Gombrich is the only translator to attempt the transformation from metrical Sanskrit into metrical rhymed English. Most of the translators aim for a fairly literal correspondence with the Sanskrit original, as seems appropriate in a dual-language edition. Prose and simple narrative verse are generally translated as English prose, more poetic verse as English free verse or "shaped" prose. In those cases where the Sanskrit author has used the punning s lesa figure to allow his words to yield more than one sense, a simple typographic device is used to indicate the multiple meaning - a useful convention, and one that has to be quite heavily exercised in some of the texts because playing on words was not something that Sanskrit authors looked down on or regarded as appropriate only for jokes.

The translations are not, by and large, intended to read as literature, but they are nonetheless the focal point of the Clay volumes. In the case of the non-Sanskritist reader, the translation is the book; in the case of readers who do know Sanskrit, the fact that they have opted for a Clay edition presumably indicates that they would still welcome some assistance with the language. It is therefore particularly important that the translations provided are clear and correct. Once again, the picture is mixed.

I have not read through the whole of the first set of translations, so it would probably be invidious to name names, but their quality seems to be rather uneven. Some appear extremely accurate (and often rather well phrased); but others contain regrettable levels of error. In the worst case, the first two verses I picked at random each contained a simple misunderstanding of the meaning of the Sanskrit, and further reading did little to suggest that this had been an atypical sample - clearly the translator had not been well chosen, and the unfortunate student who decides to use this particular volume to improve their Sanskrit is likely to run into problems.

Most of the translations are carefully done and will serve students well.

Supplementary apparatus is (no doubt deliberately) small in quantity, but it, too, will be found useful: the volumes are well indexed and are provided with textual notes, glossaries and the like as appropriate.

There is one respect in which a major effort has been made to ease the student's burden. I consider the effort misconceived, and others I have spoken to mostly seem to share my opinion.

Sanskrit may be a "difficult" language, but its difficulty does not reside where it is sometimes perceived to. What makes Sanskrit difficult is partly what makes all natural languages difficult - the complexities of usage, the differing registers used in different contexts and so on - and partly the fact that it is a highly inflected language that exhibits a complex morphology and an unnerving number of irregularities. These are real problems for students: they can be overcome only by mastering the orthography and formal grammar and by reading widely (and the Clay series offers much material).

But the form in which the Clay texts are presented suggests that quite different kinds of "difficulty" have preoccupied the editorial mind; the aim appears to be abolition, rather than mastery, of certain seemingly problematical aspects of the language. The script used is not the Devanagari which Sanskrit is traditionally printed, but rather an accented Roman transliteration. A complex set of typographic devices is employed to indicate the way in which adjacent vowels often fuse with each other, so that, for example, tatha api always appears as tathapi . (This process of sound change at word junctions is known as sandhi.) And compound nouns and adjectives, a linguistic feature of which Sanskrit authors made great use, are printed with their members separated by thin vertical bars.

These three preoccupations - script, sandhi and compound analysis - are very familiar to anyone who teaches Sanskrit. They are, of course, serious issues for the elementary student. The Devanagari syllabary is not an inherently difficult script to learn, but it operates on principles very different from alphabetic scripts such as Roman. The processes whereby sandhi modifies adjacent sounds at word junctions can make it hard for the novice to recognise the words that he sees. But neither of these features of the language should take a reasonably competent student more than a few weeks to master. In my experience, it is precisely those who have never attained a reasonable competence who tend to make a fuss about them.

The policy of the Clay Sanskrit Library appears to be to attempt to abolish both problems. The problem of the script is abolished by simply not using it (despite the fact that it is now easy to print high-quality Devanaxgarl- - easier than it is to print Roman with complex accents). The sandhi problem is abolished by telling the student what the "hidden" forms are so that he does not have to work them out for himself. (A "sandhi grid" is also printed on the inside of the back cover of every volume.) Compounds are treated in the same way. This is a somewhat larger topic in Sanskrit grammar, and I would not expect a first-year student to be comfortable with compound analysis until well into the course. But it is no deep mystery, and it should be taught - and, more important, learnt - in the first year. To publish a series of texts in which the analysis is done for the student is to suggest that this is not something that really needs to be mastered - or possibly that it is something beyond mastery by ordinary folk.

A homely English comparison may be helpful. When I was a child of about four years old, my parents bought a weekly comic for me. It was called Chicks' Own , and its objective was to help very young children to learn to read. It sought to do this by hyphenating words of more than one syllable: in the online copy that Google found for me were examples such as "Per-cy Pig does some ex-er-cises" and "Din-ner is read-y". This seems like a perfectly sensible approach - for children of about four years old. Once I knew how to read, I graduated to more grown-up reading printed without superfluous hyphenation.

The texts contained in the Clay Sanskrit Library are not elementary reading primers: they are texts , some of them quite advanced texts, and no one can hope to make anything of them without having learnt at least basic Sanskrit. The Clay policy of printing in Roman, with sandhi and compounds pre-analysed, is the equivalent of hy-phen-at-ing Shake-speare for old-er stu-dents. It should not be necessary, and if any given reader does find it necessary, that can be only because similar policies have been applied in teaching they have, with the result that he has been discouraged from learning these particular basics.

Yet further complication is introduced by a bizarre decision to omit the normal accents from Sanskrit names occurring in an English context, but to add centred dots indicating compound junctions and acute accents indicating stress. Thus, for example, the epic generally known as Mahabharata , which as a Sanskrit word would appear in the Clay series as Maha|bharata , is spelt Maha·bharata whenever it appears on a title page, in the text of an introduction, footnote and the like.

Why it should be imagined that someone capable of reading an epic in Sanskrit (with or without the aid of a translation) is likely to require such aids to help him pronounce its name is more than I can say. And one can look only with some astonishment at a series of volumes that needs not one but two highly non-standard sets of conventions to represent the language that is its raison d'être .

A sad side-effect of these strange editorial decisions is that these very attractive little books will find no favour with Indian Sanskritists, and very little with Westerners whose grasp of the language is even moderately advanced. This is a great shame, as the concept of the series is such a good one and the overall initiative so welcome.

Rumour has it that parallel Devanaxgarl- versions are likely to appear from an Asian publisher, and it is to be hoped that this is true. I like the idea of catching a plane with a Clay volume in my pocket to help pass the time; but I would want it to look like Sanskrit and not like the work of a typographer suffering from an attack of brain-fever.

John D. Smith is reader in Sanskrit, Cambridge University.

Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Please Login or Register to read this article.


Featured jobs