What does indentation mean in suttas?

Hi,

I was wondering whether there are general principles that explain why indentation is used in the suttas. For instnace, I was looking at:

https://suttacentral.net/sn2.18/en/sujato

The sutta is split in two, with the second half being indented:

“Kacci tvaṁ anagho bhikkhu,
“I hope you’re untroubled, mendicant

It seems that the second half, like the first, is a dialog, so it’s not the case that it’s setting off an extended monologue. Is this an editorial decision in every case, or is there something about the Pali itself that suggests that parts of the sutta should be distinguished?

1 Like

The second part of the sutta is a verse section.

2 Likes

Ah, I see. Is this the general reason for indentation?

1 Like

As far as I know, yes.

1 Like

Yes, it’s a fairly universal practice. I have often wondered how many people realize that.

When the lines are broken into short phrases it might be more obvious, but I bet even then many have no idea.

3 Likes

Thank you both!

Now I’m wondering how the distinction is stored in the data files. :thinking:

1 Like

There’s an “html” folder next to all the other ones in SC bilara-data.

Here’s the html for your sutta:

{
  "sn2.18:0.1": "<article id='sn2.18'><header><ul><li class='division'>{}</li>",
  "sn2.18:0.2": "<li>{}</li></ul>",
  "sn2.18:0.3": "<h1 class='sutta-title'>{}</h1></header>",
  "sn2.18:1.1": "<p><span class='evam'>{}</span>",
  "sn2.18:1.2": "{}</p>",
  "sn2.18:1.3": "<p>{}",
  "sn2.18:1.4": "{}</p>",
  "sn2.18:1.5": "<p>{}</p>",
  "sn2.18:1.6": "<p>{}</p>",
  "sn2.18:1.7": "<p>{}</p>",
  "sn2.18:1.8": "<p>{}</p>",
  "sn2.18:1.9": "<p>{}</p>",
  "sn2.18:2.1": "<blockquote class='gatha'><p>{}<br>",
  "sn2.18:2.2": "{}<br>",
  "sn2.18:2.3": "{}<br>",
  "sn2.18:2.4": "{}</p>",
  "sn2.18:3.1": "<p>{}<br>",
  "sn2.18:3.2": "{}<br>",
  "sn2.18:3.3": "{}<br>",
  "sn2.18:3.4": "{}</p>",
  "sn2.18:4.1": "<p>{}<br>",
  "sn2.18:4.2": "{}<br>",
  "sn2.18:4.3": "{}<br>",
  "sn2.18:4.4": "{}</p>",
  "sn2.18:5.1": "<p>{}<br>",
  "sn2.18:5.2": "{}<br>",
  "sn2.18:5.3": "{}<br>",
  "sn2.18:5.4": "{}</p>",
  "sn2.18:6.1": "<p>{}<br>",
  "sn2.18:6.2": "{}<br>",
  "sn2.18:6.3": "{}<br>",
  "sn2.18:6.4": "{}</p></blockquote></article>"
}

It’s a complex system!

1 Like

“gatha” means “verse”.

Huh, interesting. So an HTML template is stored for every sutta in addition to the data file. This means that they both have to be maintained, right?

Somewhat off-topic nerd time enclosed… I would have imagined that the classification of lines as verse or prose would be part of the data itself.

So just taking the window where the transition to verse takes from the root text JSON file, we have:

{
  "sn2.18:1.9": "“Evamāvuso”ti. ",
  "sn2.18:2.1": "“Kacci tvaṁ anagho bhikkhu, ",
  "sn2.18:2.2": "kacci nandī na vijjati; ",
}

(Ah, beautemous JSON.)

And then there’s also a file like:

{
  "sn2.18:1.9": "<p>{}</p>",
  "sn2.18:2.1": "<blockquote class='gatha'><p>{}<br>",
  "sn2.18:2.2": "{}<br>",
}

And I guess the final HTML is generated by replacing the brackets with the root source text…

I can’t help but wonder why something like this might not have worked:

[
  {
    "id": "sn2.18:1.9",
    "content": "“Evamāvuso”ti. ",
  },
  {
    "id": "sn2.18:2.1",
    "content": "“Kacci tvaṁ anagho bhikkhu, ",
    "startVerse": true
  },
  {
    "id": "sn2.18:2.2",
    "content": "kacci nandī na vijjati; "
  }
]

I mean, there are a zillion ways that idea could be expressed, but it seems like prose-versus-verse status is part of the core data. (Like, I could imagine wanting to quantify and compare how much verse each book contains versus prose.)

:nerd_face:

Nerd moments.

1 Like

:laughing: I am not a nerd really. What I have understood is that the explicit aim is to have the text files as plain text files, with nothing else in them. If I understood correctly, this way it allows the most flexible usage. Applications can combine languages, or use just one language, use the HTML or ignore it, or do whatever combination suits their purpose best.

Voice for example only uses the “root” and “translation” files and does not make use of the HTML. For listening, it doesn’t make a difference if something is indented or not. The less clutter on the website the better for screen reader users. Therefor it’s great that we can just ignore certain things.

1 Like

Oh I hope it didn’t seem I was calling you a nerd, I was calling myself one! :rofl:

These sound like good goals. My one nitpick is that the data about whether a particular line is verse or prose seems to be stored only in the HTML. That means an application that wants to refer to that attribute needs to parse the HTML and merge it back into the content lines. Not a huge deal.

Thanks for your explanations!

1 Like

I don’t understand “nerd” to be an insult. It’s rather that my knowledge in this field is rather superficial from what I pick up here and there while doing my work. But there’s quite a substantial lack of background knowledge, and I don’t feel too much inclined to learn all these things. Though I find it fascinating to get some glimpses of how, for example, website development works.

2 Likes