Print

Print


Dear TEI list,

I have a question about the handling of terms, particularly East Asian
terms in English scholarly writing.  I am encoding a journal article,
containing conventional glosses  like this:

    At this meeting, they agreed on the inauguration of the Dōjinkai (同仁会,
    Association of Universal Benevolence), a non-profit medical philanthropic
    organization.

Here, "Dōjinkai" is the romanization of the Japanese term "同仁会", and
"Association of Universal Benevolence" is the translation of the term.
 So we have 3 elements to describe: the romanized form, the Japanese
glyph form, and the English translation.  (Note that subsequent to
this definition, the author refers to this association simply as "the
Dōjinkai".)

However, it is also common to see glosses in a different order:

    The organization recruited Japanese doctors and other medical workers
    to organize epidemic prevention teams (bōeki han 防疫班) to work in China.

Here, the English term is provided first, and the romanization second,
followed by the Japanese glyphs.  Clearly "medical teams" is an
English translation of a Japanese term, but the term itself is in
parentheses.  Subsequent to this definition, the author refers simply
to "epidemic prevention teams", since the Japanese term itself has no
particular nuance or significance.

I would appreciate any guidance on how TEI might approach the encoding
of these terms.  I think the most appropriate elements are <term> and
<gloss>.  I can see two approaches to these examples:

a) Follow the author's implicit order, applying the gloss to what
comes in parentheses:

    ... the <term xml:lang="jp-Latn" xml:id="dojinkai">Dōjinkai</term>
    (<gloss xml:lang="jp" target="#dojinkai">同仁会</gloss>,
    <gloss target="#dojinkai">Association of Universal
Benevolence</gloss>), a ...

    ... organize <term xml:id="medical-teams">medical teams</term>
    (<gloss xml:lang="jp-Latn" target="#medical-teams">shinryō han</gloss>
    <gloss xml:lang="jp" target="#medical-teams">診療班</gloss>) and ...

b) Impose a linguistic normalization, always treating the Japanese
glyphs as *the term*, and treating the romanization and translation as
mere glosses, regardless of the order:

    ... the <gloss xml:lang="jp-Latn" target="#dojinkai">Dōjinkai</gloss>
    (<term xml:lang="jp" xml:id="dojinkai">同仁会</term>,
    <gloss target="#dojinkai">Association of Universal
Benevolence</gloss>), a ...

    ... organize <gloss target="#shinryo-han">medical teams</gloss>
    (<gloss xml:lang="jp-Latn" target="#shinryo-han">shinryō han</gloss>
    <term xml:lang="jp" xml:id="shinryo-han">診療班</term>) and ...

(I don't provide xml:lang="en" for the English since that is implicit,
being defined globally on the root TEI element in the document.  Also,
"jp-Latn" is the BCP47 style of indicating romanized Japanese.)

The advantage to approach #1 is that I can blindly accept the order
used by the original author, while the advantage to approach #2 is
that I can impose some uniformity and rely on every <term> being the
Japanese glyph form, thereby easing things like index creation, etc.

I'd be interested to hear what people think about these two options,
or whether there are other factors I haven't considered yet.

Thanks in advance,
Joe