Re: [lojban] non-ka properties

[Felipe Gonçalves Assis <felipeg.assis@gmail.com>]

doi xorxes

2011/6/18 Jorge Llambías <jjllambias@gmail.com>:
> 2011/6/18 Felipe Gonçalves Assis <felipeg.assis@gmail.com>:
>>
>> These non-ka properties are then just lambda expressions
>> which return different kinds of values. It appears only logical
>> that {ce'u} should be used.
>
> "ce'u" is not used for the return value, it is used to mark the slot
> of the property holder. The slot for the return value is marked with
> "ma kau".
>

Clarifying: I was not marking the return value with {ce'u}, I understand
it marks the argument. I was expressing the return value as the x1 of
{broda} when saying {lo broda be ce'u}. Anyway, that was just a guess.

{kau} would be my second one. My basic consideration was that,
even in English, "property" in the sense of
  "X has property ka-P when P(X) holds"
is understood as different from
  "Y is the/a P-property of X when P(X,Y) holds".
In the first definition, a property is an abstraction, while in the second
one, a property is really just a simple object, so I felt like avoiding {ka}
or any abstractor.

These were semantic considerations. Now to the syntactic ones, which
only now have I been making.

1. The syntax of abstractions requires less BE, BEI. It can also avoid
SE and {ce'u} at the cost of deferring the {kau}, and possibly confusing
the receiver about what is being referred to.

  {lo xe klama be ko'e bei ko'i bei ko'o bei ce'u} (non-ka) vs
  {lo ka klama ko'e ko'i ko'o makau} (ka-kau)

(note the difference in meaning when the {makau} is removed)

  {lo se stuzi be ce'u} (non-ka) vs
  {lo ka stuzi makau} (ka-kau)


2. The non-ka approach fits more naturally the use of the full
power of descriptors, including quantifiers.

  {loi panzi be ce'u} (non-ka): "One's children, considered collectively"
  {lo pa panzi be ce'u} (non-ka): "One's single child"
  {le panzi be ce'u} (non-ka): "The thing I describe as one's child/children"

Of course, {loi ka makau panzi ce'u} and friends can be given
the same interpretation. This is actually the semantic issue
written above. Were it not for the {kau}, {loi ka} would be a very
unusual construction. This highlights the semantic difference
between the x1 of a regular ka-abstraction and a non-ka/ka-kau one.


3. When nesting properties, ka-kau requires indexing for
disambiguation, while non-ka works out-of-the-box.

  {lo ka ce'u traji lo ka ce'u zmadu makau fo makau} (ka-kau, ambiguous)
is ambiguous between
  {lo ka ce'u traji lo ka ce'u zmadu makauxipa fo makauxiro}
(ka-kau-xi, plausible)
  {lo ka ce'u traji lo ka ce'u zmadu makauxiro fo makauxipa} (ka-kau,
implausible)

Note the clear difference between
  {lo se zmadu poi ce'u traji lo ve zmadu be ke'a be fo ce'u} (non-ka,
plausible)
  {lo ve zmadu poi ce'u traji lo se zmadu be ce'u be fo ke'a} (non-ka,
implausible)

Of course, the same problem already happens with {du'u}.
The bridi
  {ko'a djuno lo du'u ko'e djuno lo du'u makau prami makau}
has four plausible interpretations.

Things get even messier when you mix {du'u} and ka-kau.

ni'o
doi rodo
I would very much appreciate if you commented on the above,
so I can feel more confident in using the language. Also, I would
be very content if any of you could point uses of this in texts.

Thanks.

mu'o
mi'e .asiz.

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.