Re: whether (was Re: ni, jei, perfectionism)

[And Rosta <a.rosta@UCLAN.AC.UK>]

Lojbab:
> >> In English we are prone to saying "I know whether x is true",  But if
 indeed
> >> we do know, then why do we not say "I know THAT x is true" or "I
> >> know THAT x is false."
> >
> >Because they don't mean the same thing, and the "whether" meaning
> >is some times more appropriate. For example, "She knows whether
> >x is true" is a perfectly sensible thing to say (especially if
> >I don't know whether x is true).
>
> But SHE knows, so why is it not "She knows the truth value of x" which is no
> longer an indirect question, and still does not require the speaker to also
> know what the truth value is.

It has the semantics of an indirect question, but not the syntax of
a subordinate interrogative. "She knows the truth value of x"
does not mean "Y is the truth value of X and she knows Y".
Rather, it means (in perhaps simplified terms) "she knows that
Y is the truth value of X".
This ambiguity is not possible in Lojban, where if you literally
said "She knows the truth value of x", it would unambiguously
mean "Y is the truth value of X and she knows Y" (which
would be nonsense).

> I suspect that it is possible to turn any factual statement using djuno
> into an indirect question in the way that English does, or even more than
> one.

I'm not sure exactly what you mean. But I'm pretty sure that
whatever you mean, it is wrong. (zo`o i ku`i je`a...)

> mi djuno ledu'u la djan. klama le zarci
> Means  I know THAT John went to the store literallym

Yes.

> but this also
> can be reflected in English with 3 indirect questions,

No so, if "reflected" means "translated with truth-conditional
meaning preserved".

> one of which
> is unquestionably to me the same in meaning:
> I know WHETHER John went to the store.
> and the other two, seem to be the same but with different emphases:
> I know where John went.
> I know who went to the store.

Not truth-conditionally equivalent, though these are entailments.

> If the person who knows is other0than the speaker, and the speaker does not
> know the value, then the speaker is precluded from filling in the
> predication completely and in English has to use an indirect question
> or a relative clause construction.  I don't see much difference between:
> I know where John went (indirect question).
> I know the place where John went (relative clause).

the NP complement has the semantics of an indirect question
(as explained above).

> Even a whether indirect question can become a relative clause
> I know the truth value that pertains to predication X.

See above.

You are correctly reporting certain facts about the grammar
of English. But I do not think that they advance our discussion
at all. Indeed, I am not even sure what overall point you
are trying to argue.

> These are restrictive relative clauses, I think.

Yes. Not that it matters, though.

> Let us take "John sees the man with a hat on".  If the speaker does not know
> which man has a hat on, he could say

Yes. Though what the speaker does or doesn't know is immaterial to
the semantics of these sentences - though if it helps you to imagine
certain hypothetical utterance contexts then do so.

> equivalently "John sees which man
> has a hat on".  with regard to the predication, I see them being identical
> in meaning - the only difference is the indication of knowledge on the part
> of th4e speaker.

They are not remotely identical in meaning. If you want then I
will point out the difference, but before I do take the time
and trouble to do so, let us be sure that your desire to know
is sufficient to justify my efforts.

> I write all this without a great deal of analysis,

All too evident.... [Is there an UI that means "I am being
sincerely but affectionately rude"? If there is, please insert
it here, and sprinkle it wherever else it might be called for.]

> but was rather stimulated
> by discussing this debate with Nora last night,.  Nora will never have time
> to read the endless thread, so I did a quick summary for her, and she
> came back with " there is no question in an indirect question" - it is to
> her misnamed because we use question words in the construction in English.

Even if it is misnamed, the misnomer should not distract from
substantive issues. It is the name for a certain grammatical
phenomenon.

However, as it happens, if one sits down and works out the semantics
of direct questions and indirect questions, it is clear that they
are different varieties of the same phenomenon.

> It seems to her not necessary to follow that practice and use question word
> constructions in Lojban, even though it might be permissible.

I was the first and only person to demonstrate this, in 1996.
However, if you exclude the methods I demonstrated, the only
successful method ever used by any Lojbanist has been the Q-kau
method.

> >> In the forner case, what appears to me an indirect question is not really -
> >> it is an English idiom, and there is a non-indirect-question that can
> >> substitute.
> >
> >It's not an English idiom. The semantics is compositional, so it's
> >not an idiom,
>
> It is idiomatic, if people use the construction without thinking about its
> meaning,

That, in the words of my students, is pants.

> and the meaning (even if compositional by some theory) does not
> match the average speaker's understanding of what the words mean.

You are free to misuse the term "idiomatic", and to propound
some incoherent and empirically inadequate account of semantics,
but I shall ignore it.

> Indirect
> questions use question words, but are not really/necessarily questions.

You are terminologically confused. Your use of "question" is
equivalent to what would be called a "direct question", if the
term "indirect question" is being used.

If one goes beyond terminology and looks at the nature of these
grammatical phenomena, is is apparent that "direct questions"
are a subtype of "indirect questions". (e.g. "Is it raining?"
means, roughly, "Tell me whether it is raining".)

> >and it is not peculiarly English, since other
> >languages do it too (e.g. Italian _so se_, "I know if", vs. _so che_
> >"I know that")
>
> Languages may have constructions that make a disticntion, but I have not seen
> any sign in American usage that the two are truly distinct in meaning (not
> that I have studied the question - this is my "language instinct" talking %^)
> Maybe I have poverty of the stimuklus and need to read more of your
 discourses.
> %^)

You have not consciously *noticed* any sign of the distinction. Nor
have most Americans. Only those immersed in the study of grammar
have.

> >It can't mean that a fact is being discussed.
> >The meaning of "whether" as complement of "discuss" is indeed
> >different from the meaning of "whether" as complement of "know"
> >(and most other verbs that can have an interrogative clause
> >complement).
>
> OK, Try "say" instead of "discuss".  Presume that the sayer is
> other than the speaker and knows the truth value.  Then John says
> that x is true. and John says whether x is true. are identical in
> predicative meaning.  The only difference is a discursive
> implicature that the speaker does not know the truth value in the
> whether form, and hence cannot make the distinction between "John
> says that  x is true" and "John dsays that x is not true".

"John says that x is true" is true iff John says that x is true.
It makes no difference whether x is true. I am not sure
whether "Say whether" presupposes the veracity of John's
assertion.

Let's change the example from "say" to "know".

"John knows that x is true"
  is true iff x is true and John knows that x is true.

"John knows whether x is true"
  is true iff either x is true and John knows that x is true
    or x is false and John knows that x is false

- ergo they are plainly distinct in truth-conditional meaning.

> But there are other possible implicatures, since the speaker could
> also know but not feel it necessary to tell the listener. But the
> event being discussed remains the same- John is making a statement
> which communicates the truth value of x.
>
> >At any rate, discussing whether x is
> >the case involves enquiring whether x is the case, while
> >discussing the fact that x is the case does not.
>
> In my usage, discussing "the fact that" indeed has no question compnent.
> But "discussing whether" is ambiguous bettwen the question and non-question
> forms.

I know what you mean: you are alluding to your "We discussed who
took out the trash" example. I think you have an accurate intuition
here. I've been thinking about "discuss" and certain other
indirect question contexts in my spare time, but am finding
"discuss" contexts the hardest nut to crack.

> I can imagine that the type of distinction you make is the sort of
> thing that grammar prescriptivists came up with in days when they were
 listened
> to, and  are the sort of things up with which I shall not put.

You may indeed be able to imagine this. But it has no bearing on
anything I say. If what I say happens to coincide with what some
prescriptivist has said, it simply shows that sometimes the
prescriptivist's prescriptions happen to be descriptively
accurate. If the head of the KKK says Idi Amin was an evil
madman, the beliefs of the asserter do not make the assertion
any the less true.

--And