On Sun, Aug 12, 2012 at 3:10 PM, Bob LeChevalier, President and Founder - LLG
<lojbab@lojban.org> wrote:
".ai mi dunda la lunra do" is simply (literally) false
when uttered by any non-delusional interlocutor.
iff it is false when the ".ai" is omitted, then it is false with the "ai" included.
But the emotional _expression_ of ".ai" could still be quite honest, even if it would be delusional to think the underlying proposition to be true.
It's not quite that simple as Xorxes has pointed out. Some attitudinals definitely do affect the truth conditions of the bridi they're applied to, namely the ones that shift the bridi into what in natlangs would be called an irrealis mood. Perhaps "a'o" is the archetypical example.
Emotions are NOT "logical", nor truth-functional. Most people probably prefer it that way, even if it makes them sometimes seem a bit delusional. So long as we can clearly distinguish between the claim and the emotional _expression_, this causes no problem in communication.
When you start trying to make attitudinals truth-functional, you kill the whole point in having them in the language, which is to allow _expression_ of emotions without having to worry about "truthiness". Assigning truth to attitudinals INVITES people to lie using them, whereas the expressions of attitude in natural language generally are not subject to such analysis.
The example I like to use for this are most uses of obscenities in English. When my dad talked about the "f***ing door being left open" he was not attributing reproductive activity on the part of an inanimate object, and indeed there was no truth functionality to that adjective - it was expressing an attitude towards the state being described. We might argue about what attitude he was expressing, (and the point of Lojban attitudinals is to enable one to be clear in expressing one's attitudes if one wishes), but one would not legitimately be able to say that my dad was lying either about the door or about his emotions in making that _expression_.
I hear what you're saying throughout this missive and I think I understand your reasoning (You don't want attitudinals to unnecessarily complicate the semantics, I dare presume), but what concerns me even in English is if someone talks about the about the "f***ing door being left open" when he's secretly delighted that the door was left open. An implied inner state of the speaker's mind is being messaged through language, interjections, and even nonverbal cues, and the accuracy of those messages could _hypothetically at least_ be mapped to truth values based on whether those messages are accurate ones, or whether they're what you'd call disingenuous or false or whatever. I agree with you that none of that affects the truth conditions of "someone left the door open" spoken in any language and in any register. But times may come when it is desirable to be able to talk metalinguistically about disingenuous play-acting, especially given the enormous toolbox that Lojban attitudinals comprise.
Best,
-Mike