John E Clifford, On 27/10/2012 18:07:
> *From:* And Rosta <and.rosta@hidden.email>
> For me (& I know others such as Mike and Martin and you don't share
> my view), the loglang should unambiguously encode *logical form* but
> needn't engage with logics, where by logics I mean ways of formally
> modelling the world in order to make it possible to state
> propositions about the world. By logical form, I mean the relations
> between operator and variable and between predicate and argument: I
> mean that not as a definition but as a rational claim that all
> formulae are fashioned from these elements, and if I'm wrong in this
> claim I'd be delighted to be shown where I err. I have very much not
> read any of the 6000 pages of the handbook of philosophical logic,
> but I suppose them to> be about logics rather than logical form.
John:
> I guess I don't see the distinction you are making very clearly. So
> you mean by "logics" particular theories about what entities (or
> sorts of entities) there are and how they interact (i.e, what is
> ordinarily called metaphysics and in Lojban epistemology)?
I mean tense logic, mereological logic, and so forth.
> If so, then I pretty much agree that that ids not very much of our
> business, except as it is a way of seeing that a language "needs"
> certain sorts of predicates and arguments (or some other means to
> deal with the situations involved). I take it the what I mean by
> logical form, the representation in a formal logical language, is
> precisely a reasonably efficient way to present this in a systematic
> way (for theory, not for practice, of course) with all the issues
> taken care of in advance by the choice of logical language.
I agree the project should consider what predicates are necessary for talking about tense, about colour, about meals, about force dynamics, and everything else, prioritizing the most important and most fundamental. I don't agree that they have to be highly formalized. But my crucial point is that I don't see this as part of loglanghood (i.e. what makes a loglang a loglang), tho there's more to any actual usable loglang than its loglanghood. I was saying this in response to your question about what your interlocutors see as being the goals of Xorban.
--And.