Received-SPF: pass (google.com: domain of seladwa@gmx.de designates 212.227.15.19 as permitted sender) client-ip=212.227.15.19;
Message-ID: <519C13CF.1090406@gmx.de>
Date: Wed, 22 May 2013 02:39:43 +0200
From: selpa'i <seladwa@gmx.de>
User-Agent: Mozilla/5.0 (Windows NT 5.1; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: lojban@googlegroups.com
Subject: [lojban] Tags and bridi operators
Reply-To: lojban@googlegroups.com
Precedence: list
Mailing-list: list lojban@googlegroups.com; contact lojban+owners@googlegroups.com
Sender: lojban@googlegroups.com
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
X-Spam_score: 0.0
X-Spam_score_int: 0
X-Spam_bar: /

We know that na(ku), quantifiers, connectives and tenses are bridi 
operators, but what about non-tense tags? Some clearly are bridi 
operators, for instance the causals:

     (1) mu'i lo nu mi tatpi kei mi na klama
         "Because I'm tired, I don't go."

     (2) mi na klama mu'i lo nu mi tatpi
         "It is not the case that: I go because I'm tired."

Obviously, (1) and (2) have very different meanings. This happens 
everytime the order of bridi operators is reversed.

When using certain BAI, it's not so clear if they should be treated as 
bridi operators:

      (3a) mi na sanga bau lo lojbo

      (3b) bau lo lojbo mi na sanga

Do they mean the same? There are some cases where it is very useful for 
BAI to be operators, and this is also true for {fi'o} constructs. 
However, I see the following problem:

Either we make every tag, no matter what selma'o it belongs to, a bridi 
operator, or we pick some that are operators and some that aren't 
depending on what is the most useful / easiest to use. The problem with 
the second option is that it is a bit annoying to have to memorize which 
tags are and which aren't bridi operators. On the other hand, the 
problem with the first option is that it would actually invalidate a lot 
of usage! For example, from The Little Prince:

      (4) do pu djuno noda fi'o fuzme mi

The intended meaning is (more or less) "It's my fault that you didn't 
know." However, if {fi'o fuzme} is treated like any other tag, then the 
scope is wrong and the sentence suddenly means "In the past, there was 
no thing such that: you know it because of me", which is backwards from 
the intended meaning.

Even worse, one of the most useful constructs I've been using, and which 
I think xorxes came up with, is {tai ... ja'e ...}, but upon closer 
examination, it, too, seems to be wrongly scoped:

      (5a) mi tai tatpi ja'e lo nu mi na ka'e sanli
           intended: "I'm so tired that I can't stand."

But the scope should look like this:

      (5a') mi [tai [tatpi ja'e lo nu mi na ka'e sanli]]

The bridi operators have scope over what's to their right. Thus, [tatpi 
ja'e lo nu mi na ka'e sanli] ends up filling tamsmi2:

      (5b) lo nu mi tatpi ja'e lo nu mi na ka'e sanli cu se tamsmi zo'e
           "The event of me being tired with the outcome that I can't 
stand is like something."

While the intended expansion is more like:

      (5c) lo nu lo nu mi tatpi cu se tamsmi zo'e
	      cu se jalge lo nu mi na ka'e sanli

That is, {jalge} is supposed to be the "main" claim, not {tamsmi}...

This is sad news. Should we say {tai} (or even {ja'e}) is not a bridi 
operator and thus save the construction (and usage) or do we need to 
update our usage? Both solutions have their pros and cons, but in the 
long run, consistency seems more important.

I'd be particularly interested in xorxes' opinion.

mu'o mi'e la selpa'i

-- 
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.