A reader contributed this comment to the feedback area:
Not sure whether individual words are the most important things in e.g
poetics. Lit-crits have argued for years, even suggesting that relations
are important---opposition and comparison of meanings---whereas any given
meaning is utterly irrelevant. Would this mark-up dare to extend to e.g.
<metaphor> or <synecdoche> tags, demonstrating relationships between words?
:) At what semantic level does TEI back off, or should it back off? Just
telling us when a line is metrically incomplete? Or telling us when a line
echoes symbolism from a previous verse?"
SystemAdmin 110000D4XK29 Posts
Re: Poetics annotations2003-10-10T20:31:50ZThis is the accepted answer. This is the accepted answer.I wanted to post the comment separately, since I think it is very interesting--certainly I am interested in other readers' experience... particularly from literary criticism.
Oh... my attempt to post the prior note failed to escape some brackets: the "e.g." referred to <metaphor> and <synecdoche> tags.
Although I wrote the article, there is much more I do not know about TEI than I know. But it looks to me like TEI provides some decent framework for semantic or analytic annotations. I think the <note> tag is the relevant one, see http://www.tei-c.org/P4X/CO.html#CONO. Using the 'target' or 'targetEnd' attributes can specify a span thought to have a particular semantic value. Moreover, the 'resp' (responsibility) attribute can annotate WHO thinks a given thing is a metaphor--precisely the sort of thing literary critics argue about :-).
There doesn't seem to be any prior categorization of tropes that might be so annotated. But within a given project or institution, the 'type' attribute of a note seems easily utilizable for such a schematization. Presumably, you could have multiple commentators add their own annotations (perhaps conflicting) about the tropes utilized in given passages. XPATH or other techniques would work for searching for every, e.g. 'type=metaphor' annotation.