I can no longer resist chiming in; apologies for rambling and for stating what may be obvious.
The IAST scheme is indeed much more widely used, but I am not at all sure if future standards should be based on precedent and if counting the number of existing publications using one standard or another is any help. I think there are two main reasons why IAST is widely used. One, as Harry Spier has pointed out, is that the ISO standards are not openly accessible. Without having any insight into the workings of ISO standards in general, as a layman I find this frankly ridiculous and if Jan Kučera can apply any pressure to change this, this restriction should be lifted. The other principal reason is that characters with an undercircle are supported only by a very small number of fonts, and even some that do support such combinations render them poorly. It is thus much more convenient for everyone, authors to publishers, to stick to underdots (which my spell checker immediately changed to underdogs!).
That said, there are good reasons against IAST too, primarily ambiguity. That is to say, R and L with an underdot are suitable for representing vocalic R and L only so long as you are dealing with classical standard Sanskrit. As soon as Vedic enters the picture, you might use l+underdot for the retroflex flap as well as for the vocalic L. The retroflex L is also present in Dravidian languages and Marathi, and in many regional Sanskrit manuscripts and inscriptions. There is also the (different?) retroflex flap used e.g. in Hindi and normally transliterated with r+underdot. According to Wikipedia, IAST uses ḻ (l with line below) for the former kind of retroflex flap and I have no idea if it uses anything for the latter kind. That could work, but then how do we transliterate the retroflex central approximant (ೞ) of some Dravidian languages? One could argue that this is not required, because we are talking about Sanskrit transliteration. But ideally, a transliteration system for Sanskrit ought to be able to transliterate (unambiguously) any language written in a Brāhmī-derived script. With this, I am entering murky waters and I would rather not go much further; it would be too much to hope for a system that can really do this, but perhaps supporting the major modern and premodern languages of South and Southeast Asia is not a dream. It is theoretically also possible to agree on language-dependent standards, as is already the case for instance with e and o, which (in IAST and in a permitted optional variant of ISO15919) represent long vowels in Sanskrit, but short vowels in Dravidian languages. The DHARMA project has chosen to follow (and expand upon) the ISO standard primarily for this reason, because it is better suited for a textual corpus in a variety of languages.
As for anusvāra as m with overdot or underdot, it seems largely a question of taste. That said, there is a good rationale for not choosing the underdot, which it implies some commonality either with vocalic trills or with retroflex consonants.
On the whole, one question we must ask ourselves is which situation we would rather live with: the chaos involved in introducing a new standard that will not immediately, if ever, become the number one global standard - or the chaos involved in maintaining multiple standards in parallel. Those who work primarily or only with classical standard Sanskrit may prefer the latter, as they are not really affected by its drawbacks.
Another relevant question is what can be done to improve and propagate any standard. IAST is not being actively maintained at all beyond a vague implicit consensus of scholars using it. It is not even a standard in fact, just a set of conventions that are clearly defined at the centre, but loose at the periphery:
whenever phonemes outside the ambit of classical standard Sanskrit come into the picture, authors (and/or publishers) seem to choose transliterations inconsistently on the basis of conventions found in earlier publications and of whatever diacritical marks they can produce on their systems. Given that ISO15919 is actively maintained, and is now to be reconsidered with the involvement of the scholarly user community, this may be a good time to settle on something that most of us are willing and able to follow in the future.
I personally would certainly be happy to jump on the ISO bandwagon
provided that the revised ISO standard will actually be made
known to the public, and not accessible only to those who pay an
exorbitant price for it, get a pirated copy, or rely on incomplete
extracts such as that on Wikipedia. But even then, I think one of the considerations in revising the standard should be a preference for character combinations that can actually be displayed properly in more than just a few fonts.
Dan Balogh
Thank you for the clarification Madhav. Since your book predates the 15919 standard, I'm wondering what sanskrit books after creation of the 15919 standard have chosen it over the IAST standard. The two Clay Sanskrit library books I have use the IAST transliteration scheme and as far as I can see the Sanskrit etexts in GRETIL also use IAST. Muktabodha uses IAST.
Thanks, Harry, but while writing my संस्कृतसुबोधिनी, which goes back to mid-1980s, I did not consult "
ISO 15919 standard" or any such documents. I was following, what seemed to me at the time, to be the prevalent practice. If my memory serves me correctly, to use r̥, r̥̄, l̥, with small circles under r and l, I was influenced by Wackernagel's Altindische Grammatik. I had used the same in designing my diacritics font Manjushree-CSX. While the ancient fonts used for the संस्कृतसुबोधिनी going back to mid-1980s and the pre-Unicode Manjushree-CSX are no longer usable, I am generally continuing to use these diacritics today. Probably just by acquired habit.
Madhav
Madhav M. Deshpande
Professor Emeritus, Sanskrit and Linguistics
University of Michigan, Ann Arbor, Michigan, USA
Senior Fellow, Oxford Center for Hindu Studies
Adjunct Professor, National Institute of Advanced Studies, Bangalore, India
[Residence: Campbell, California, USA]
To download a pdf of the current ISO 15919 standard (a 30 page document) costs 145 Swiss francs = 160 US dollars. I'm wondering if this is one of the reasons that most people use IAST for transliterated Sanskrit. The only place I've seen the ISO 15919 standard used in a book is Madhav Deshpande's sanskrit primer संस्कृतसुभोधिनी .
_______________________________________________
INDOLOGY mailing list
INDOLOGY@list.indology.info
https://list.indology.info/mailman/listinfo/indology
_______________________________________________
INDOLOGY mailing list
INDOLOGY@list.indology.info
https://list.indology.info/mailman/listinfo/indology