It just occurred to me that this is what I’ve been doing as a translator. Every word in Pali, every phrase, everything present and absent, is a fuzzy and indeterminate ball of probability. It’s a cat resting or playing or stretching unseen in a box. A translator opens the lid, quietly and delicately, so as to not disturb the sensitive beast. But alas, with excruciating inevitability, the poor cat drops dead, stiff and lifeless as soon it’s seen.
A thousand possible renderings die, and only one may be chosen.
And as soon as you’ve recovered from the remorse of slaying so many possibilities, there is another word, another box to be opened, another unsuspecting cat.
We use the Mahasangiti edition, which does indeed use the Burmese readings by default. Stylistically it’s not my preference, but it is, I think, the most accurate and carefully proofread digital Pali text.
Unfortunately it’s not possible to simply switch from one set of readings to the other. This is a limitation in using XML/HTML for markup. The differences between different editions may be subtle and not readily captured in such a way. For example, each spelling of a particular word might vary, but the markup only identifies the first occurrence.
The new version of SC does in principle support multiple editions, so it is possible that in future we will offer a choice of Pali editions. I’m not hugely inspired to work on this, however, as I don’t believe any of the other digital texts offer a compelling alternative. Nevertheless, I have not looked into it in detail and we may work on this at some stage.
Bhante, isn’t the Mahasangiti edition based on VRI’s Chaṭṭha Saṅgāyana CD? If that is so, that edition already contains the variant readings, set by default on Burmese (although it seems some variant readings have been missed in some places). Sometimes, one reading clearly makes much better sense than the others, and it is not necessarily the Burmese one. I can guess also that there are many cases where variant readings seem equally probable (that’s where ven. Bodhi usually choses the Sinhala one, and ven. Thanissaro the Thai one, it seems).
We have variant readings, you can see them by enabling Textual Information. These inherit the variants from the VRI text, as well as those added by the Dhamma Society revision of that text.
It’s unfortunately the case that no one text always contains the best readings, so in difficult cases we simply do our best. On the whole, like Ven Bodhi I tend to prefer the Sinhalese readings; the Burmese text often looks a little more “normalized”, whereas the Sinhalese preserves the quirks of the manuscript tradition more faithfully.
But I believe—and don’t quote me on this, as I haven’t studied it in detail—that the available Buddha Jayanthi text, which represents the Sinhalese manuscript tradition, is poorly proofread. It’s good as an occasional reference, but not as a mainline text.
Venerable Indacandra—a Vietnamese monk who lives in Colombo and is quietly translating or retranslating the whole canon into Vietnamese—has made many corrections to the Buddha Jayanthi text. However these are simply mentioned as notes in his translations, and have not been incorporated in the text itself.
The PTS text, of course, is quite erratic, some good, some poor, so again we prefer to avoid it for the mainline text.
Leaving aside the considerations of manuscript fidelity, for a digital text consistency is a big advantage. For example, it makes searching much easier and more powerful. In this respect, the Mahasangiti is useful for us, precisely because it normalizes the more erratic manuscripts.
Right, indeed, it is far from perfectly consistent.
Ideally, the best thing would be to have multiple digital texts, each precisely representing a single manuscript tradition. Rather than resolving quirks and standardizing, leave them all exactly as they are, and cope with the fuzziness in the application. Then we could use a diff program to analyze variations.
I’ve found that i can [as a reader rather than translator] often have/find 3 to 5 renderings in english and then if i’m really intrigued with something i’ll pull out my dictionaries and look into the Pali as well.
I’ve found this engaging with the material, really helpful in my getting a sense of what’s being said.
If i’m right and these -below- are all accusatives?
Ajjhattasantiṃ pavinaṃ adassaṃ.
then the line would read ‘The inner peace, the seeing, the searching’.
4 english versions plus some Pali dictionary work suggests to me
Searching, i saw inner peace.
The -aṁ ending is commonly used across a wide range of forms, including neuter nominative, -ant masculine nominative, many indeclinables, and so on. So while it is probably true that the accusative masculine singular is the most common form that uses this ending, it is far from the only one.
The noun form dassana appears in the accusative as dassanaṁ.
In our verse, the prefix -a indicates that it is an aorist, and is listed as such in the dictionary.
Neither, it is “searching, inquiring”. I translated adassaṁ as “found”, but “saw” would also be fine.