Loading ...
Sorry, an error occurred while loading the content.
 

Lojban's prepositional phrases

Expand Messages
  • Patrik Austin
    Hey guys, I ve been looking at Lojban grammar and I ve noticed that the syntax is actually not tied to predicate logic. What I mean is that in predicate logic
    Message 1 of 7 , Aug 24, 2014
      Hey guys,

      I've been looking at Lojban grammar and I've noticed that the syntax is actually not tied to predicate logic. What I mean is that in predicate logic you would always use predefined argument places and never prepositional phrases, right? However, there are 65 prepositions in Lojban's cmavo list.

      An example of the usage of the preposition 'vi' (near) is in chapter 10 of the reference grammar:

      le ratcu cu citka le cirla vi le panka
      The rat eats the cheese [short distance] the park.
      The rat eats the cheese near the park.

      where "vi le panka" is an object noun phrase. Could Lojban be spoken without using any prepositional phrases?

      Patrik
    • selpa'i
      la o me. Patrik Austin .me cusku lo se lidne ... Lojban s prepositions are really shorthands for predicates. Especially the tense-prepositions have a very
      Message 2 of 7 , Aug 24, 2014
        la'o me. Patrik Austin .me cusku lo se lidne
        > Hey guys,
        >
        > I've been looking at Lojban grammar and I've noticed that the syntax
        > is actually not tied to predicate logic. What I mean is that in
        > predicate logic you would always use predefined argument places and
        > never prepositional phrases, right? However, there are 65
        > prepositions in Lojban's cmavo list.

        Lojban's prepositions are really shorthands for predicates. Especially
        the tense-prepositions have a very clear semantic relationship to the
        sentence they are in (expanding them out into an explicit form is easy).

        > An example of the usage of the preposition 'vi' (near) is in chapter
        > 10 of the reference grammar:
        >
        > le ratcu cu citka le cirla vi le panka The rat eats the cheese [short
        > distance] the park. The rat eats the cheese near the park.
        >
        > where "vi le panka" is an object noun phrase. Could Lojban be spoken
        > without using any prepositional phrases?

        Yes, but it would make sentences much longer.

        Let's use a simpler case: "Rats are eating in the park." You can either
        use the location tense "bu'u" (at), or use its predicate equivalent
        "zvati" (to be located at):

        (1) lo ratcu cu citka bu'u lo panka
        "Rats are eating in the park." (I'll pass on trying to produce
        an acceptable direct gloss)

        And here in super-explicit logical form:

        (2) da zo'u da nu lo ratcu cu citka kei gi'e fasnu gi'e zvati lo panka
        "There exists some X such that X is an event of rats eating and
        X occurs and X is located in the park."

        I claim that (1) and (2) are equivalent. Clearly, (1) is more practical
        from a human-language standpoint, but fortunately (2) exists for when we
        want to examine the underlying logic of tenses.

        --selpa'i
      • Logan Kearsley
        ... Not necessarily. Look up neo-Davidsonian event semantics ; the whole point of the neo-Davidsonian notation is to allow you to continue to unambiguously
        Message 3 of 7 , Aug 24, 2014
          On 24 August 2014 16:07, Patrik Austin <patrik.austin@...> wrote:
          > Hey guys,
          >
          > I've been looking at Lojban grammar and I've noticed that the syntax is actually not tied to predicate logic. What I mean is that in predicate logic you would always use predefined argument places and never prepositional phrases, right?

          Not necessarily. Look up "neo-Davidsonian event semantics"; the whole
          point of the neo-Davidsonian notation is to allow you to continue to
          unambiguously represent predicate-argument structure while allowing
          for flexibility in ordering and simple representation of optional
          arguments.

          -l.
        • Patrik Austin
          Thanks, that s all very helpful. Also, I ve been wondering why the separator cu is needed between le ratcu and citka, but not between citka and le cirla?
          Message 4 of 7 , Aug 26, 2014
            Thanks, that's all very helpful.

            Also, I've been wondering why the separator 'cu' is needed between le ratcu and citka, but not between citka and le cirla?

            > le ratcu cu citka le cirla vi le panka
          • selpa'i
            la o me. Patrik Austin .me cusku ... citka and cirla are selbri (predicates). When two predicates are adjacent they form a tanru (compound). le ratcu
            Message 5 of 7 , Aug 26, 2014
              la'o me. Patrik Austin .me cusku
              > Also, I've been wondering why the separator 'cu' is needed between le
              > ratcu and citka, but not between citka and le cirla?
              >
              >> le ratcu cu citka le cirla vi le panka

              "citka" and "cirla" are selbri (predicates). When two predicates are
              adjacent they form a tanru (compound).

              "le ratcu citka" could be translated as "the rat eater(s)", while "le
              ratcu cu citka" means "the rat(s) eat".

              You don't need (and in fact cannot put) "cu" after a predicate (its
              purpose is to mark the main predicate of a sentence). "citka le cirla"
              is a selbri followed by a sumti (a predicate followed by an argument),
              there is no chance of those two forming a compound.

              --selpa'i
            • And Rosta
              ... A regrettable design, IMO. Better to do without cu and instead have an overt tanru-forming operator; tanru, being a departure from logical form, ought by
              Message 6 of 7 , Aug 26, 2014
                selpa'i, On 26/08/2014 11:22:
                > la'o me. Patrik Austin .me cusku
                >> Also, I've been wondering why the separator 'cu' is needed between le
                >> ratcu and citka, but not between citka and le cirla?
                >>
                >>> le ratcu cu citka le cirla vi le panka
                >
                > "citka" and "cirla" are selbri (predicates). When two predicates are adjacent they form a tanru (compound).
                >
                > "le ratcu citka" could be translated as "the rat eater(s)", while "le ratcu cu citka" means "the rat(s) eat".

                A regrettable design, IMO. Better to do without "cu" and instead have an overt tanru-forming operator; tanru, being a departure from logical form, ought by rights to be marked. The unmarkedness of tanru encourages their use, with the result that a logical language is used as if it were Riau Indonesian.

                There are no tanru in my conlang.

                --And.
              • Gleki Arxokuna
                ... Not a regrettable design imo since it saves one syllable. Conciseness of the opposite design has to be proved.
                Message 7 of 7 , Aug 26, 2014
                  2014-08-26 17:47 GMT+04:00 And Rosta <and.rosta@...>:

                  > selpa'i, On 26/08/2014 11:22:
                  >
                  > la'o me. Patrik Austin .me cusku
                  >>
                  >>> Also, I've been wondering why the separator 'cu' is needed between le
                  >>> ratcu and citka, but not between citka and le cirla?
                  >>>
                  >>> le ratcu cu citka le cirla vi le panka
                  >>>>
                  >>>
                  >> "citka" and "cirla" are selbri (predicates). When two predicates are
                  >> adjacent they form a tanru (compound).
                  >>
                  >> "le ratcu citka" could be translated as "the rat eater(s)", while "le
                  >> ratcu cu citka" means "the rat(s) eat".
                  >>
                  >
                  > A regrettable design, IMO. Better to do without "cu" and instead have an
                  > overt tanru-forming operator; tanru, being a departure from logical form,
                  > ought by rights to be marked. The unmarkedness of tanru encourages their
                  > use, with the result that a logical language is used as if it were Riau
                  > Indonesian.
                  >

                  Not a regrettable design imo since it saves one syllable.
                  Conciseness of the opposite design has to be proved.


                  > There are no tanru in my conlang.
                  >
                  > --And.
                  >
                Your message has been successfully submitted and would be delivered to recipients shortly.