Re: Bootstrapping a cooperative conlang

--- MorphemeAddict@WMCONNECT.COM wrote:

<snip>
>
> A picture only shows what the animal looks like.  It doesn't give any of the
> rest of the meaning of the word, like how it moves, what sounds it makes,
> what
> kinds of food it eats, why it's a pest or pet.
> The "convoluted definition built from a minimal vocabulary" gives a
> comprehensive definition of the complex meaning of the word "mouse".  The
> minimal
> vocabulary ensures that it can be understood by anyone who knows those
> English
> primitives.
> This type of definition also allows one to compare the differences in
> definition between synonyms and other related words.  What part of the
> definition of
> "mouse" does not apply to "rat", e.g., and what parts are common to both?  I
> don't have a similar type of definition of "rat", so I won't try to answer
> that
> question, but these explications permit such fine discrimination.
>
> stevo
Let's back up one step and consider the unspoken assumptions:

If I am a visitor from a distant planet and ask the definition of "mouse" then
I have no idea what a mouse is and everything you say is perfectly valid.

If I am a theoretical linguist interested in the nature of "definition", or a
biologist interested in taxonomy, or a computer engineer interested in
knowledge representation structures, then everything you say is perfectly
valid.

If I am a visitor to Nepal and I ask what the local word for mouse is, I don't
need to be told what a mouse IS, I only need to be told what the WORD for mouse
is. The kind of dictionary project I'm discussing here begins with the
assumption that, in spite of not sharing a language in common, all users of the
dictionary will have a certain level of shared world experience. Thus the
conlang edition of the dictionary assumes I know what a mouse IS, and only
needs to point to some symbol that evokes that knowledge and binds it to the
given word.

While reading Wierzbicka's definition of "mouse" at some point I gathered
enough information to conclude which "thing-I-already-know" is being described.
From that point on, the rest of the definition is irrelevant, and I can safely
ignore it. The rest of the definition may, indeed, tell me things about what a
mouse IS, but I already know those things.

Anything beyond that should be looked for in the conlang edition of the
encyclopedia, rather than the dictionary.

--gary

Reply