Lojban gismu semantic roles and valency slots

[Larry Sulky]

Hello!

I have a question about semantic roles and where they fit in gismu valence slots; specifically, what are the criteria that determine which semantic roles typically go in which slots? For example:

• The gismu "xebni" ("hates") has the experiencer in slot 1.
• The gismu "rigni" ("disgusts") has the experiencer in slot 2.

Why the difference?

Thank you for any insight you can provide.

Cheers!

-- Larry Sulky

[Bob LeChevalier]

On 8/14/2023 11:57 AM, Larry Sulky wrote:
> Hello!
>
> I have a question about semantic roles and where they fit in gismu
> valence slots; specifically, what are the criteria that determine which
> semantic roles typically go in which slots? For example:
>

> * The gismu "xebni" ("hates") has the experiencer in slot 1.
> * The gismu "rigni" ("disgusts") has the experiencer in slot 2.

>
> Why the difference?
>
> Thank you for any insight you can provide.

The short answer is that when the gismu list was created, we were not
trying to systematize semantic roles. Rather, we were building off of
wordlists used in earlier versions of Loglan, which were in turn based
on lists of the most frequent concepts found in different languages.

xebni was based on the verb concept "x hates y"
rigni was based on the adjectival concept disgusting/repugnant (x
disgusts y).

To have them parallel, you would probably want either "y is hateful to
x" which is se xebni, or in a lujvo selxei which would have the
experiencer in slot 2 or se rigni, in lujvo form selrigni, which would
be "x feels disgusted in response to y" with the experiencer in slot 1.

se (or sel- as a prefixing affix) causes the 1st and 2nd slots to be
reversed, whatever they mean

Because it is so trivially easy to create words that have the semantic
roles in whichever order you want, we never made an attempt to
systematize the gismu list. The assumption was always that the gismu
list would just be a starting point, and people who would actively use
the language would acquire a vocabulary that was mostly lujvo (and
fu'ivla borrowings for technical terms), and these would be haphazard
and unpredictable in semantic roles because the specific words one would
coin/memorize would be based on what bridi relationship one was trying
to communicate, and not on a system of semantics. (we explicitly avoided
choosing a semantic system or theory as the basis for the language
because 1) I don't really know enough about semantic theory to do so and
2) presuming one particular semantic theory to be valid for language use
would in itself skew the usage experience and probably render the
language in violation of the design principle of metaphysical parsimony.

lojbab
Bob LeChevalier
(Founder)