Re: Systematic Word Relationships (Was: Arabic and BACK and a whole lot of other things.)

--- Jim Henry <jimhenry1973@...> wrote:

> On 12/21/05, Gary Shannon <fiziwig@...> wrote:
> > --- Jim Henry <jimhenry1973@...> wrote:
>
> > > Method, system, way of doing action:
> > >
> > > to fight -> martial art, fighting style
> > > to program  -> programming methodology
> > >
> > I added this as an enumeration since there are
> more
> > than one system or method.
>
> OK.  I intended this as a way of deriving a
> _general_ term for all such fighting styles,
> programming methodologies, schools of conlanging,
> etc.  Maybe terms for specific styles,
> methodologies,
> schools, etc. could be derived from that
> general term with appropriate modifiers.

Ah, yes. I see. I'm thinking this goes in one
directiong as a single relationship (karate->fight)
but in the opposite direction it seems like an
enumaeration to me. One could also say that "Freanch"
is a "style" of "language", so I'm not sure how to
distinguish this from a regular subclass type of
enumeration.

> > > to get into NOUN, to put oneself into NOUN
> > > (enlitigxi "to get in bed", etc.)
<snip>

>
> That's similar, but what is different about the
> Esperanto
> examples I gave (and others like "surtabligi",
> "devojigxi", etc.) is that they incorporate a
> preposition
> and an object of the preposition into a
> becoming-verb
> or causative-verb.
So actually this is a combination of TWO wordsm,
rather than a direct derivation from a single word. I
hadn't added any of those types of relationships yet,
but they clearly need to be included.

<snip>

> > My intention was the very next level of
> abstraction,
> > colloquially, so that apple->fruit as opposed to
> > apple->mass-of-protons-and-electrons.
>
> Even with the "very next level of abstraction"
> the derivations would be idiomatic -- for instance
> for one person lemon-GNR would suggest
> "citrus fruit", to another just "fruit" in general.
On the other hand setting up this system for the use
of a person creating a lexicon for a conlang implies
that the conlang creator would set the precident. At
any rate, for each instance a standard would be
established either by usage or buy edict.

<snip>

>
> Each of these derivation patterns is of the form
> ( specifier prefix ) + ( root ) + ( person suffix )
> + (noun ending)
>
> So:
> land-o : country
> land-an-o : citizen, inhabitant
> sam-land-an-o : inhabitant of the same country
> ali-land-an-o : inhabitant of another country
Again, these look like compounds of multiple words
rather than derrivations from a single word. I haven't
started adressing those yet.

--gary

Reply