> On Dec 5, 2017, at 10:01 AM, Hugh Cayless <[log in to unmask]> wrote:
> I'm currently working on encoding an ancient commentary—that is, a work that annotates another work, in this case Vergil's Aeneid. The source is arranged thus: there is a lemma, or headword(s), quoting the source, followed by commentary on that word, phrase, or line. This commentary can be lexical, literary, historical, etc. So these works have something in common with dictionaries, but maybe not to the extent that I could use the dictionary elements, even though they might sometimes be apt.
> My question is: has anyone worked on this sort of source, and how did you mark up lemmas, etc.? My current thinking is to use <quote type="lemma"> for the headwords, but I'm hopeful there's something better. Or maybe we need a new element?

Interesting question; thank you!  

It's not completely clear what you are denoting with the word “lemma”

If by “lemma” you mean a word or phrase appearing in the source text,
I believe I have used ‘q’ for that in the past.  That's the sense in
which I would normally interpret 'lemma' in the context of a
commentary (text-critical or other).

If by “lemma” you mean the dictionary or lookup form of a word, as it
might appear in a discussion of the word’s usage, I’ve used
‘mentioned’ for that.  That's the sense suggested by your meditation
on whether to repurpose the element from the dictionary tag set.

To take a concrete example, here is an entry from Max Black, A
Companion to Wittgenstein’s ‘Tractatus’ (Ithaca: Cornell UP, 1964),
p. 115.  For simplicity in the example I've used fairly shallow
markup; in a more serious effort I would mark the bibliographic
citations as units, identify the works from which quotations come (if
not the Tractatus), and schedule a first review of the project's
criteria for distinguishing term, q, mentioned, and gloss (and any
other elements that come elbowing their way into entries), once 1% of
the work has been transcribed, with another review at the 10% mark.

<div class="entry">
    <ref target="TLP#s3.262" rend="b">3.262 application</ref>: or,
    <q>use</q>. Cf. <ref target="TLP#s3.326">3.326</ref> on
    <q>significant use</q> and <ref target="TLP#s3.32">3.32</ref> on
    the difference between symbol. and sign. In the light of <ref
    target="TLP#s3.5">3.5</ref>, <mentioned>applied</mentioned> =
    <q>By application I understand whatever makes combinations of
    sounds or strokes into a language. In the sense in which it is the
    application which makes a rod with strokes on it into a
    <term>ruler</term>. The <term>laying</term> (<gloss
    xml:lang="de">Anlegen</gloss>) of language against reality</q>
    (<title>Phil. Bem.</title> 20, 12).
    In the <title>Tractatus</title>,
    <mentioned>application</mentioned> does not mean
    <mentioned>use</mentioned> in the sense of the
    <title>Investigations</title> (cf. Anscombe,
    <title>Introduction</title>, p. 91&mdash;I agree with her that
    <mentioned>application</mentioned> means <gloss>that kind of
    difference between the syntactical roles of words which concerns a
    logician</gloss>). Cf. <ref target="TLP#s3.327">3.327</ref> on
    <q>logico-syntactic application</q>.

C. M. Sperberg-McQueen
Black Mesa Technologies LLC
[log in to unmask]