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

Re: [lojban] Time-(non)local sumti (was: Mixing tenses, on the beginners list)

So, the stock loglang definition: a language (this is loaded, note) in which each discourse has a single possible parse and that parse gives the logical structure (as defined somehow, FOPL on steroids for the nonce) of the discourse.  

From: Ian Johnson <blindbravado@gmail.com>
To: lojban@googlegroups.com
Sent: Thursday, May 2, 2013 7:52 PM
Subject: Re: [lojban] Time-(non)local sumti (was: Mixing tenses, on the beginners list)

On Thu, May 2, 2013 at 7:42 PM, Jorge Llambías <jjllambias@gmail.com> wrote:

On Thu, May 2, 2013 at 8:02 PM, Ian Johnson <blindbravado@gmail.com> wrote:
This amounts to saying it can be local or nonlocal, but doesn't specify how to *explicitly* state the distinction. The ability to make such specifications (if the language supports them at all) is pretty much my third axiom for Lojban design :)


I don't think the language supports that disticntion via grammar (it can make it semantically just as you are making it in English, of course). If you want a grammatical support for the distinction, I guess the first step would be to show why it would be desirable. Just because a distinction can be made is not reason enough to make it.
A more precise statement of the axiom is that if there's a distinction that can be made, you should be able to make it explicitly, and (for the most part) should have the option to just not make it, as well. In other words, nothing we care about should be required to be implicit, to the extent possible. English more or less fails to meet this requirement all the way down at the grammatical level, since you can't really maintain control of how clauses attach to one another.

(What are your other axioms?)
First axiom: the grammar is formally parsable, with a significant amount of "structural semantics" being reflected in the parse tree. A formally parsable language with only one grammatical class meets the first requirement but not the second. su'o so'a natlangs meet the second requirement (i.e. if you have a parse tree you can make inferences about low-level semantics) but not the first. 

Second axiom: referential and semantic transparency to the extent possible. This is not an expectation of perfection, it's a goal rather than a requirement.

mi'e la latro'a mu'o
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
 
 


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