anti-ka (was: machine translation)

[John Cowan <cowan@DRV.CBC.COM>]

la lojbab. cusku di'e:

> >>? I've never even HEARD of "anti-ka". Sounds toxic. :)
> >
> >Indeed, I hope lojbab explains what he meant by that.
>
> I defer to Cowan.  You guys talked about this around 2 years ago, and
> at some point we decided that we didn't need an explicit construct,
> but my only concept of it is as an inverse operation for the abstraction
> that is "ka".

It goes something like this.  "ka"-bridi with a single "ce'u",
implicit or explicit, represent properties of an object, and
can be applied to the object with "ckaji":  "lo broda cu
ckaji le ka ce'u broda" expresses a truth for any value of "broda".

When we extended "ka"-bridi to represent reified relationships,
with multiple "ce'u" markers, there was no way to say
"A and B stand in relationship C" where "C" is "le ka ce'u
broda ce'u".  The "anti-ka" was a proposed device to do this.

Several people, including Jorge and I, noted that "bridi" does
the right thing:  "le ka ce'u broda ce'u kei cu bridi zo'e
A ce'o B".

--
John Cowan      http://www.ccil.org/~cowan              cowan@ccil.org
                        e'osai ko sarji la lojban