[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >


[Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: [jboske] sumti grammar oddity



At 01:56 PM 9/11/03 -0700, Jorge "Llambías" wrote:

I just noticed an oddity in the grammar of sumti. The EBNF rule for
sumti-tail (what comes after a gadri) is:

   sumti-tail-1 = [quantifier] selbri [relative-clauses]
                | quantifier sumti

I expected it to be:

   sumti-tail-1 = [quantifier] selbri [relative-clauses]
                | quantifier sumti-6

where sumti-6 is an unquantified, not connected sumti. A sumti is more
general than sumti-6, it can be quantified and connected with other
sumti-6.

So the grammar allows things like:

lo ci (ko'a .e ko'e)
lo ciboi re broda

I could make some sense of {lo ciboi re broda} interpreting it
as {lo ci lo re broda},

I won't guess what you interpret the latter to be. The former under the version of the language I know would mean "at least some of the 3 pairs of broda in the universe", which I would loosely equate to "lo ciboi re lo [xa] broda, since there would have to be exactly 6 for there to be 3 pairs of broda in the universe.

The former is difficult because the prosumti are undefined. It might expand out into logically connected sumti, in which case ko'a might mean re broda and ko'e might mean re brode, and you have two claims parallel to the one you can grasp. But a corresponding *intended* usage might be "lo so'eboi pa tarci joi so'u plini" (the many planetary systems).

 but what is {lo ci (ko'a e ko'e)}?

(The real answer is that we tried to make the grammar as unrestricted as possible, while still being able to resolve the structure unambigously per YACC.

There is all kinds of nonsense possible in number strings too, so we should expect some amount of nonsense to be grammatical in Lojban sumti structures.

At one time I was one of those fighting to find interpretations for as many possible strings, but Cowan convinced me that this was a waste of time.)

--
lojbab                                             lojbab@hidden.email
Bob LeChevalier, Founder, The Logical Language Group
(Opinions are my own; I do not speak for the organization.)
Artificial language Loglan/Lojban:                 http://www.lojban.org