Date: Thu, 20 Nov 1997 15:05:39 -0500 (EST) Message-Id: <199711202005.PAA15595@locke.ccil.org> Reply-To: And Rosta Sender: Lojban list From: And Rosta Organization: University of Central Lancashire Subject: Re: Indirect questions X-To: LOJBAN@cuvmb.cc.columbia.edu To: John Cowan X-Mozilla-Status: 0011 Content-Length: 4397 X-From-Space-Date: Thu Nov 20 15:05:40 1997 X-From-Space-Address: LOJBAN@CUVMB.CC.COLUMBIA.EDU Jorge: > >> Then I could say: {lu le mi mlatu cu morsi li'u cu danfu > >> lu xu do badri li'u}, but when I say {ko'a djuno le du'u xukau > >> mi badri} I don't mean {ko'a djuno le du'u le mi mlatu cu > >> morsi}. > >OK. But why are you so sure that {danfu} is the appropriate > >selbri? {la`e lu le mi mlatu cu morsi} is not what one knows > >if one knows le du`u xu kau do badri. I seem to be missing a bit > >of your reasoning. > > I suppose we need to know what {danfu} is supposed to mean. > I think that you either have { danfu } or { > danfu }, or maybe we can have both, but I don't think > you can have a mix. If I'm right then you must have either: > {ko'a djuno le danfu be la'e lu ... li'u} or {ko'a djuno la'e le > danfu be lu ... li'u}, because the x2 of djuno is a du'u, not a > text. OK: I agree that either both these versions are OK or neither are. I'm still not convinced that {le danfu be la'e lu ... li'u} or {la'e le danfu be lu ... li'u} are what ko`a knows when ko`a knows whether such-and-such. > >I take {xu do badri} to mean > > > > Bring it about that for every x, a truthvalue of {do badri}, > > I know that x is truthvalue of {do badri}. > > That's asking for too much. For example, you are asking > the person not only that they respond with the truth but that > they convince you that they're saying the truth (otherwise > you wouldn't _know_ that what they say it true). Maybe that > really is implicit in questions? I don't know. I'm surely asking them to respond with the truth in such a way that it changes the state of my knowledge. > I would have said {xu do badri} means: repeat this > statement replacing the question word so as to make it > a true statement. The replacement for {xu} is in a first > instance either {na} or {ja'a}, and ususally you will repeat > by using {go'i}. I know you think this, and I don't think it incorrect, but I do think it an unnecessarily metalinguistic characterization of the meaning of questions. At any rate, I at least understand why you're not sure what {la`e lu xu do badri li`u} means but you are sure what {le danfu be la'e lu ... li'u}/{la'e le danfu be lu ... li'u} mean. > >I suppose the "answer" to that command would be some information > >such as a list of everything that is a truthvalue of {do badri}. > > To me, as I said, it's either: {lu mi badri li'u danfu lu xu do badri li'u} > or {la'e lu mi badri li'u danfu la'e lu xu do badri li'u}. > > The second one could be {le du'u mi badri cu danfu la'e lu > xu do badri li'u}, assuming the question was made to {mi}. I understand what you're saying. > In the case of { danfu } there is no claim that any > statements were made by anybody. Sure. I think the answer to a question (a piece of text = an act by a speaker) is either an act of supplying information, or a piece of information. We seem to agree on this, but not on whether either kind of answer can be x2 of djuno. > >So if {xu do badri} means > > > > Bring it about that for every x, a truthvalue of {do badri}, > > I know that x is truthvalue of {do badri}. > > > >then {la`e lu xu do badri} means one of the following: > > > > - For every x, a truthvalue of {do badri}, I know that x is > > truthvalue of {do badri}. > > - x is truthvalue of {do badri}. > > - do badri > > > >To me, the first of these makes the most sense. > > To me, my knowledge has little to do with the meaning of > {la'e lu xu do badri}. Perhaps it's not such a good idea to use > examples with mi/do. What would you say {la'e lu xu la djan badri} > means? "I know whether John is sad"? I reckon so. But the other two of my three suggestions are a bit more intuitive. So I wouldn't want to take a firm stand on this. > >{la`e lu ko > >klama li`u} would be "do klama". And {xu do badri} would > >be equivalent to something like {gau ko mi djuno le du`u > >xu kau do badri}. > >["gau" is a guess at the appropriate BAI] > > How do you make reference to the meaning of > {lu xu la djan badri}, which is the same as the meaning > of {lu xu se badri la djan}, even though they're two different > text-questions? If you mean "propositional content", then just stick {la`e} in front. If you mean "discourse function" or something like that, you don't need to stick anything in front: the two different texts have the same discourse function. --And