Note: I wrote this some time ago as part of another discussion. It’s a question that comes up from time to time, so I thought I’d make it more visible by making it it’s own thread. I have made a few changes to the original.
In the original discussion, Ven Brahmali recommended adding footnotes to my translation, arguing that:
Good. A translator’s work should be invisible. One of the biggest problems with footnotes is that you end up listening to the translator’s voice instead of the Buddha’s.
I always come back to my own experience, long before I was a monk. My girlfriend and I loved Dostoevsky. We’d sit around for ages reading The Idiot, The Brothers Karamazov, and the rest. Of course it was in translation and we knew it, but we never paid any attention to that. I couldn’t say who the translator was, or how they worked, or whether there were other translations, or what the issues in the translations were. All I know is that I loved the books and they moved me and informed me.
I’m doing these translations for the 99%. For the people who’ve never heard of the Suttas, who have no idea what the difference is between “sutta” and “commentary”, no idea of doctrinal or terminological controversies—and no need to know of these things. I want people to read the suttas so that they will understand the Dhamma, to be inspired, moved, and informed.
Now, sure, if I was doing a University thesis on Dostoevsky, then I’d have to address the translation issues. What is lost or distorted in translation, what misconceptions persist, and so on. And it’s important for serious students to address these issues. However, notice that in this case, even though the work of Dostoevsky is highly complex and must raise all manner of issues, the translators don’t provide footnotes at all, unless it is a specialized academic edition. It’s up to the scholar to glean what they can from the text they have.
No footnotes is, of course, the normal situation in Buddhist texts, too. For 2,500 years such notes were unknown. There were introduced by western scholars a little over a century ago, as part of the critical apparatus of the academic book. They are a feature made possible by the technology of printing; but their very name (at the “foot” of the page) reveals that they are also bound to and limited by that same technology.
In doing this translation, so far I’ve checked most of BB’s footnotes for his AN text, and I think there are few that are really useful for understanding the text. Most of them are technical, or cross-references, or paraphrases from the commentary, and so on. Which is all fine, but hardly essential. There are, on the other hand, many that are actively confusing and contribute to misunderstanding the text, such as explaining suttas in terms of dry insight.
I don’t say this because I think BB’s work is bad, but because I think it is the best. And even so, I don’t really think the footnotes add all that much to the work. I’ve edited over ten thousand footnotes in various texts, and my overwhelming impression is that they’re mostly a waste of time, especially for the general reader.
Now, having said this, as I think I mentioned earlier in the thread, I have in fact been making notes as I proceed. These aren’t meant for publication, as they’re usually pretty scrappy. But they will be there to help translators, and of course if anyone else wants to read them they can always find them. Personally I don’t think they’re all that important, but it’s not hard to jot down a few notes as I go.
In terms of what we’re doing on SC, there’s a few things to bear in mind from the beginning. The first is that SC is essentially an HTML app, and there is no support for footnotes in HTML. There are good reasons for this. Footnotes have been pushed for inclusion in HTML forever, and the basic reason they’re not is because the web is an intrinsically different environment to books. A footnote is essentially one piece of text that connects with another piece, and this is basically just a hyperlink, the very basis of HTML. Footnotes are a limited subset of the web’s basic functionality, one that binds it unnecessarily to the past of paper.
But while not having a dedicated footnote or ref tag is sensible, people still want to use them, which means that every method of doing footnotes on the web is essentially a hack. Footnotes are done in all kinds of ways, most of which are excrescences voided from the bowels of Microsoft Word, or other such minions of terror and darkness.
We are currently working on exporting our HTML texts to EPUB (which as of the latest EPUB 3 does support footnotes) and LaTeX (which of course has good footnote support). On a technical level it is far from trivial to suck in tens of thousands of texts from different sources, and convert them to sane markup, then out again to something else. Add footnotes to this and it becomes nigh on impossible.
Allow me to indulge a little here. When I complain about the markup, most people don’t really know what it means. Let me give one example of actual code that I’ve worked with in converting suttas. I’ve used dummy text to avoid embarrassing those who produced the source code!
<table width="70%" border="0" cellspacing="5" cellpadding="0">
<tr>
<td width="61%" height="2"><br>
</td>
<td width="38%" height="2">
<div align="right"><i> <font color="#996633"> <span style="font-size:8pt;font-family:Arial;"> <span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;">Some Text: 08.10.2014<br>
</span></span></span>Some Text: SV<br>
<br>
Some Text: <br>
<span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;"><span style="font-size:8pt;font-family:Arial;">"Anguttara Nikaya by Bodhi, p. 219"</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span> </span> </font> </i></div>
</td>
<td width="1%" height="2" bgcolor="#F4EFBD"></td>
</tr>
</table>
If anyone reading this knows any HTML, they probably need a good hug right about now.
So one of the things that SC has done is to rescue Buddhist texts from this mess and preserve them all in clean, sane, consistent markup. And we’ve done this for over 60,000 texts. Yay us! But of course we have our limits and we have to work with what is achievable. You simply can’t do things in the same way if you need to work with such a large corpus as compared to working on individual texts.
Okay, so within these constraints, what can we do? Well, my new translations work with PO files, where the text is segmented into nice, short bits: a sentence or less (broken on major punctuation). That means we have matched text, source and translation, keyed off the same ID. And we can apply this same system to translations in any number of languages. In each case, we will be able to register equivalent ID, i.e. equivalent segments of text.
So, when I am doing the translations, I make comments or remarks from time to time, using our browser-based Pootle software. These are recorded as plain text in the PO file. The crucial thing is that they are matched to the same ID as the text segment. This means we can always associate a given comment with a given text segment, in any language. The limitation is that we can’t assign the comment to any particular word: it has to be to the segment. However, this doesn’t matter too much for notes, as you can always just mention the term in question. (Stylistically, it’s best to have notes at the end of a sentence or after major punctuation anyway, so this works out fine.)
This opens up many possibilities, especially since comments that are made in, say, an English text can be made available for, say, Finnish readers. And vice versa. So this holds the promise of becoming a very powerful tool.
What I am thinking of, as a long term goal, is providing for web annotations on this basis. Annotations are a cutting edge web technology, you can use them on, say medium.com, but they are still evolving and do not have a rigorous standard. One of the big problems with them is keeping the correct note associated with the correct text. However, with our hard-coded text segments this won’t be an issue.
This technology would make it possible to develop, across languages, a network of genuinely useful notes. I’d love to see a wiki-style system, where people can give notes to explain a term or use, and others can edit or expand them. The notes could include rich material: videos, images, maps … They could have different categories: “linguistic”, “beginners”, “historical” and so on. You could distinguish between “expert” notes, like say those made by the translator, which are fixed, and community notes, which can be edited wiki-style. You could follow the notes made by someone you like, or a teacher could set up notes for their students. You could make available an English note for an Italian reader, let it be machine-translated for them, or let them do their own translation. Compare what you can do today with song lyrics, for example. This is all discussed on SC’s issues list.
This system of notes would complement what we have here on Discourse, which mainly deals with whole texts, rather than individual segments.
This is all in the future, but we are laying the groundwork for it right now. Rather than trying to make a poor copy of footnotes on the web, we’ll be able to use the possibilities of the digital environment to do entirely new kinds of things.
To get back to your original point, one of the kinds of things that such notes could do is to offer discussions of different translation choices. Rather than such discussion being essentially me (or another translator) justifying their choices, this would become more objective and more interesting. Different translations can be presented, pros and cons discussed, and, crucially, if desired the translation can be amended and corrected. In this way, a translation is no longer a didactic presentation—“the Buddha according to Sujato!”—but is a conversation in a community. This is, of course, normal now in journalism and other creative fields, where the creators interact more directly and intimately with their audience than ever before.
Part of this process is that the author seeks not to win authority, but to grant it. I could play the academic game of doing papers and peer review and all that, but I would be much happier if I could encourage others to have a say about the suttas, become learned in them, and argue with me about them. I want to recapture that which is so powerfully evident in the suttas themselves: they are a conversation, not a lecture.