Re: Dictionary formats

In a message dated 03/30/02 6:29:47 PM, grey@FAS.HARVARD.EDU writes:

<< Those of you with detailed dictionaries (are there any of you?), >>

    I guess by now you've seen there are quite a few.  :)

<< how do you list everything in your notes? Everyone here can't be making
one-to-one
wordlists. So how do you make comments indicating various usages,
compounds, and so on?>>

    The way I've done it for Kamakawi is to actually list one-to-one
correspondances, much the way a non-technical, run-of-the-mill two-way
language dictionary works.  I won't list things like productive affixation
(for instance, you won't find the entry "dance" in a dictionary with "danced"
right below it as a separate entry and "dances" right below that), but pretty
much everything else gets its own entry, even if it would make sense to list
it all under one root.  Examples:

lama (v.) to make (usually used to refer to something that one makes that can
fit in
    one’s hands and that is made up of various parts, such that it couldn’t
be made
    by accident or in the wild, as a bird’s nest or a beaver’s dam, but
which a human
    had to put together, but which isn’t considered big enough to have been
built);
    (n.) making, construction; (adj.) made, constructed, built, produced

This would be in the Kamakawi-English section.  In the English-Kamakawi
section, I'll just use one word, such as "to make (v.) lama", and that's it,
with the understanding that if you need a better explanation, you can go to
the other section, and that's generally the way my dictionaries work.  The
X-English section details usage, variants, etc., whereas the English-X
section is merely for quick, one-to-one reference.  A different example:

ikeli (n.) wake (as with a boat)

    And the separate entry...

keli (v.) to flutter, to trail behind, to leave a trail, to leave a wake (as
with a boat), to fly
    (as in a flag), to wave (as in a flag); (n.) tail (as in an animal)

In this example, the prefix /i-/ has a kind of object derivation pattern
thing going on, yet it's not always that x root is a transitive verb, and i-x
is the thing that is x'd.  Thus, the meanings of roots with /i-/ prefixed to
it are not completely predictable.  For instance, this word might also have
come to mean "trail", "fluttering", "kite tail", "kite" or "streak" or
something like that.  It doesn't mean any of those, though, and it can't; it
can only mean the wake that a boat (or a fish) produces.  Thus, it has enough
weight to be listed as its own entry and not under the main root "keli".  Why
I didn't also list it under "keli"...?  I don't know.  To avoid redundancy?
Is "unhappy" listed under "happy" in English dictionaries?  <checking>  No,
it's not.  (Funny: They have a list of synonyms, and then also a list of
antonyms, which has only one member, and it's not "unhappy", but "sad".)  It
does list "-pier" and "-piest" as the comparative and superlative endings,
and that's probably because there's a spelling change.  A piece of productive
morphology in Kamakawi would be the passive suffix /-?u/, but it's
invariable, so it'd be unncessary to list it for every verb.  If, however, a
passive verb came to be used more often than it's counterpart, or it took on
a different meaning, then I'd list it.
    I don't know what WordNet is, but I think I'll look it up.  And I've
never used another dictionary before now, where I'm using my Hawaiian
dictionary to see what kinds of words were developed naturally, and which
borrowed (and I'm also using one of my other languages to borrow from--it's
quite fun!).  And then, of course, when the dictionary fails, it's always
useful to use you all as resource, as many of you (including he to whom I
reply now) well know.  ~:D

-David

"fawiT, Gug&g, tSagZil-a-Gariz, waj min DidZejsat wazid..."
"Soft, driven, slow and mad, like some new language..."
                    -Jim Morrison