I share Lou's unease about this distinction between "name" and "word"
as mutually exclusive categories. There was a 17th-century German
intelligencer called Johannes Brun who, however, was almost always
referred to by his pseudonym 'Unmuessig', which means something
like 'diligent' or 'never idle'. So it was at once the man's name and
a description of him - a proper noun, but also an adjective. I'm sure
listers could come up with countless similar examples.
Quoting Gabriel Bodard <[log in to unmask]>:
> At least two points here:
> Lou Burnard a écrit :
> > (1) makes me feel much more uneasy than (2), though I can
> understand th> e
> > preference for (1) if epigraphers really do divide their strings
> > mutually exclusive classes of "name" and "word". But is this
> At least in one sense we need to treat words and names differently:
> lemmatization and indexing (and searching). A lexical index or
> dictionary would not contain "Gabriel" or "Bodard" or "Lou".
> also has a different sense for names and words: words take @lemma
> give the normalized and lemmatized nominative-singular-masculine
> 1st-person singular-present-active-indicative, i.e. the
> headword, whereas names take a <reg> giving the usual form of the
> (which may still be unique, or unusual, and not in any controlled
> > Isn't the "name"-ness, a property additional to the "word"ness?
> > <name><w>Gabriel</w> <w part="y">Boda</w></name> <w>fecit</w>
> > makes perfectly good sense to me -- and also simplifies the
> > since a processor interested only in words can reliably just look
> > <w>s, without having to be told also to consider <name>s.
> Actually for our purposs this makes the processing harder, not
> because we'd need to tell both the lemmatizer and the indexer to
> words when they appear inside names, or something. But anyway.
> In fact we use a similar markup to what you propose, but with names
> words still distinguished within personal names:
> <persName><name>Apollonius</name> <w>son</w> <w>of</w>
> (We don't use the specialized <foreName>, <surname>, etc. because
> feel that Greek and Roman names don't fit easily into that
> categorization> .)
> > I also feel that incompleteness and metricality are properties
> > words, not of names. A name might spread across two verse lines,
> > example. And "completeness" of a name might mean something
> > from "completeness" of its constituent tokens. Is "L. Burnard" an
> > incomplete name?
> Which brings me to a clarification of my original point: we
> two different senses of "name" in our texts: (1) a reference to a
> person, which might contain multiple names, rolenames, words,
> placenames, etc., (2) an onomastic name or token, "Gabriel",
> "Apollonius", etc. For the first, which is the sense you are using,
> use <persName>, and we don't indeed have any reason to mark them
> segmented or incomplete. For the second we use <name>, as I pointed
> above, and these are the names that I am identifying as basically on
> same level as words.
> (Another way of expressing this: we look up <w> for linking to
> Perseus Greek dictionary; we look up <name> for linking to the
> of Greek Personal Names; we look up <persName> for linking to our
> prosopographical <listPerson> table.)
> Still uncomfortable about calling names words. And having to tag
> as <seg> as well as <name> instinctively feels like a fudge.
> Dr Gabriel BODARD
> (Epigrapher & Digital Classicist)
> Centre for Computing in the Humanities
> King's College London
> 26-29 Drury Lane
> London WC2B 5RL
> Email: [log in to unmask]
> Tel: +44 (0)20 7848 1388
> Fax: +44 (0)20 7848 2980