Date: Sun, 2 Nov 1997 07:00:38 -0500 (EST) Message-Id: <199711021200.HAA06011@locke.ccil.org> Reply-To: Geoffrey Hacker Sender: Lojban list From: Geoffrey Hacker Subject: Re: le/lo X-To: bob@rattlesnake.com X-cc: lojban@CUVMB.CC.COLUMBIA.EDU To: John Cowan In-Reply-To: X-Mozilla-Status: 0011 Content-Length: 6779 X-From-Space-Date: Sun Nov 2 07:00:49 1997 X-From-Space-Address: LOJBAN@CUVMB.CC.COLUMBIA.EDU Initial disclaimer: I will probably learn to anticipate your own conventions regarding the use of {le} and {lo} in time, to avoid confusion, and that will make some of the following discourse probably a little academic. On Thu, 30 Oct 1997 bob@megalith.rattlesnake.com wrote: > My thesis is that {lo} provides more information to a listener than > {le}; and that sometimes one is misled by the common English glosses > of `the' for {le} and `a' for {lo}. I agree that the mechanics of Lojban allow you to use {lo} to indicate specific things; however, this usage is totally inconsistent with all the examples in the refgramm, and is highly contrastive to the common gloss of {lo} as a nonspecific article. I think it's unfortunate that you can't more easily express specificity plus veridicality in Lojban, because I take it that if you could, there would be no reason to use {lo} to refer to something specific, even if you could constrain your universe of discourse to something small enough only to include specific items. I think that would lead to much clearer thinking and expression in this area of semantics. > And if the context reduces to one real cat and one real but small dog, > with no further distinguishing context, {lo mlatu} conveys more > information to you, the listener, than does {le mlatu}. In this > case, to the speaker, {lo mlatu} means `the cat', whereas {le mlatu} > means `what the speaker describes as one or more specific entities to > be called cats', a more puzzling designation to a listener. I don't think it needs to be more puzzling, though. Again, pragmatics seems to dictate that you will tend to describe a cat as a cat. Describing a cat as a small dog, when there is a cat that you could be talking about, and in the absence of any context to the contrary, is rude. It is the deliberate obfuscation of your meaning to your listener. > Until there is more context, she may not be referring to what we in > English call `the cat', for two reasons: > > * {le mlatu} may be plural; > > * {le mlatu} may refer to other than veridical cats. Default context is important. But it would seem extraordinary that the default context would allow you to use {le} with rampant non-veridicality without some clear guidelines for figurative or approximating use in place. If she really has two cats, there will be strong pragmatic pressure on her to describe them as cats. > > Specificity does not mean `a specific cat' in the conventional English > sense of the words. 'A specific cat' in conventional English is a claim both of specificity AND veridicality, I can see that. You lose the guarantee of veridicality with {le}. Yet I see no compelling reason why someone would want to trick you by using {le} to refer to something different enough from the {le} description that it would cause you to infer the wrong referent. You seem to be making a big deal of what the literal meanings are of {le} and {lo}, but you seem to be creating problems with their conventional usage where none actually exist, because their default contexts carry the presupposition that you will not try to mislead your listener about that to which you are referring. I certainly would not have THOUGHT that that was an unreasonable assumption myself. You claim that the default context confuses you about what {le} refers to, and then you solve this problem by creating a new default context that confuses other people regarding what {lo} refers to. Six of one and half a dozen of the other. > > Hence, in the context in which we are in a room with a cat and a small > dog, {le mlatu} may refer to either or both. I think this statement makes it very clear that your default context for interpreting {le} and {lo} referents is radically different from that of most other Lojbanists. You seem to be trying to justify this difference on the grounds that your interpretation is more consistent with the literal meaning of {le} and {lo}. I just don't agree with that. I don't give a damn about the English language or English translations for this purpose, either. It is simply that on pragmatic grounds alone, I would have great difficulty trying to justify using a non-veridical descriptor with the kind of rampant licenciousness that seems to scare you off using {le} more often. The non-veridicality of {le} seems really only to be there to enable you to use figurative language more easily in Lojban, as well as to indicate that you might be mistaken, on occasion, about the identity of something specific that you wish to discuss with your listener. Commonsense and sympathy with your listener are necessary to use {le} clearly, as they are necessary for clear communication in any other aspect of any language. But to be left in the dark about what {le} refers to, when the literal meaning of the description is veridically consistent with likely candidates, seems far more skepical than I could possibly warrant. > > As I say, Lojban is weird. > > I agree, but I'm still not so sure you're not making it much more weird than it perhaps needs to be. > Hence, when my friend says {lo mlatu}, the question comes down to what > mechanism we are using to determine veridicality in the conversation? But if your friend says {le mlatu} there will be the corresponding question of what mechanism you are using to determine designations in the conversation. Basic veridicality, with certain commonsensical side constraints such as what I have already discussed, seems only logical. > Finally, I should note what John says in Chapter 6.2: > > In all descriptions with ``le'', the listener is presumed to > either know what I have in mind or else not to be concerned at > present (perhaps I will give more identifying details later). I think this is excellent confirmation of what I'm saying. You don't need to be in the dark about {le} because it should only be used when you will know what the speaker does have in mind for a referent in the first place, or at least when you will know that you won't have to worry about what it is if you haven't got it quite yet. Although this latter requirement is not necessitated by the refgrammar, it is a compulsory part of all human communication. Any concerns that you have about what a {le} description could refer to, shouldn't be a problem under the correct use of {le} by someone who actually cares about communication rather than simply trying to bewilder you. Finally, I do not seriously expect anything I say here today to convince you, since you have obviously survived the argument of a great many other people before me. At least, I hope, we have clarified our own views to each other, as well as to the other people who will read this message. In the meantime, a'o ko gleki lobypli Geoff