Message-Id: <m0rgXHY-000021C@xiron.pc.helsinki.fi>
Date:         Mon, 20 Feb 1995 07:30:15 -0500
Reply-To:     Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Sender:       Lojban list <LOJBAN%CUVMB.bitnet@FINHUTC.hut.fi>
From:         Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Subject:      status of lo/dapoi, here are some messages from the old thread
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 23569
Lines: 538

This is a compendium of messages from late last year regarding the
question of "lo" vs.  "da poi" and the question of whether "lo" implies
existence.  I THINK I was consistent in my position about the status
quo.  Cowan's of 10 November summarizes I think the concept of lo = da
poi, but changing the universe of discourse when talking about
non-existents (his proposal re negation however was found faulty and
retracted, but I think the rest stood).

Following from a private message from me to pc, refreshing his memory
on the historical context of "lo" when he first joined into the "any"
discussion.

|As for your earlier comment re "lo".  We KNEW what it meant, but as the
|language has evolved, it has changed, and some of the changes are
|probably incompatible with the original.
|
|
|1. Originally it was the veridical version of "le", derived from JCB's
|"lea" but with a default quantifier working with individuals rather than
|the entire set.  You and I had long confusing discussions about the
|default quantifier which at the moment is "su'o lo ro broda".
|
|Meanwhile we had what used to be "ba jia broda", now "da poi broda"
|which meant approximately the same thing.  This led into the "se sorme"
|(ze mensi) question, and we had to decide whether "ze mensi" meant "ze
|da poi mensi" (which it originally expanded to).  We decided that it
|3was better to expand it as a change of the default quantifier:  ze lo
|mensi.
|
|
|2. About a year later people started asking when to use "lo" and when to
|use "le".  I tended to answer on a 'faetures' basis - "lo" was
|+veridical and -definite, "le" was -veridical and +definite (without
|realizing that there is a difference between +definite and +specific,
|which is one point that has come out of this debate for Nora and me).
|AS a result, "lo" tends to be used a lot like English non-specific "A"
|or "Some".
|
|3. After Cowean got involved some discussion led to the "lo unicorn"
|discussion - whether "lo unicorn" statements are meaningless becausae
|the set is empty, or false because they don';t exist, or true because
|for the most part talking about unicorns implicitly moves you into a
|world where unicorns exist.  As I recall you indicated that there were
|positive aspects to all points of view, in a classic "pc wiggle".  You
|and Cowan chose one answer, we fiddled with the predicate "xanri" and
|"zasti" to fitt that answer, and I have been lost ever since, since I'm
|not sure I know what that answer was.  I have used the rule that "lo"
|means more or less "if broda exists then da poi broda" in my own usage.
|
|4. The first pragmatic use of "lo" was by Preston Maxwell in his
|trnaslations of some Chinese stories, which you commented on back in
|ancient days.  He used it in a story-like manner "A man came up to me"
|or something (lo nanmu pu klamu'o mi), where the man was +definite but
|-specific (I think that is correct) but this didn't enter into the
|discussion at the time.  I have since used "lo" for similar constructs
|where I wanted to assert that the thing in question really was a man,
|but that it really wasn't important to be able to indicate which one
|(whether or not there really was a particular one wasn't at issue).
|About a year ago, CVolin Fine introduced the "new information"/ "old
|information" dichotomy reflected in "bi'u" and generally used with "le"
|- thus "le" need not refer to something known to the listener if it is
|known to the speaker (and optionally marked with "bi'u").  This started
|to suck away from the Maxwell use of "lo" and in the recent discussion
|people have said that I should use "lebi'u" rather than "lo" for the
|story situation.
|
|5. Possibly related to 4., Nora has raised the question of "lo ninmu cu
|mamta mi" (A woman was my mother).  Here there is a specific woman, and
|in this case it is identifiable, but you can push things with "lo ninmu
|cu mamta le prenu", and derivative expressions like "You mother wears
|army boots - which might be purported to be a statement about your
|veridical mother without my having the slightest idea who she is or even
|if she is still alive" Here I think it is +specific and -definite (or am
|I mixed up in the middle of the night?)
|
|6. Nora has now read all of the traffic, and is convinced there is
|something missing and is trying to put her finger on it.  She has a
|spectrum of examples of usage with varying degrees of specificity and
|definiteness with me constantly getting sucked off on the tangent of "lo
|unicorn" i.e. veridicality/existence that keeps us from really getting
|too far.  There appears to be 2 situations not clearly handled, but I
|won't try to repeat them.  One is rare since I can only think of one
|example - quantum physics.  An electron is in "lo energy level" but you
|cannot even after the fact tell which one it really was in (that should
|be "pa lo energy level), so the discussion about looking back at the end
|of your life and asking about each book to see if it was the one you
|read until you get a "true", doesn't work.  (My explanation seems
|muddled to me, so feel free to be confused %^).
|
|The other is handled by "da poi", whatever it is, but this breaks down
|if the object doesn't exist.  I'll need her to present the example and
|how it differs from related cases.
|
|7. This whole question about opaqueness lsoes me, because I am not sure
|how it is distinguished from specificness and definiteness, whcih in
|turn also keep confusing me.
|
|So what does "lo" mean?  The definition is +veridical, but for me
|-specific and -definite and very +opaque %^)
|
|If you say that all of the answers spring from the deifnition, I'lll
|believe you, but I'll need you to help us figure out what that
|deifnition is again.


|Date:         Thu, 6 Oct 1994 17:24:44 BST
|From: i.alexander.bra0125@oasis.icl.co.uk
|Subject:      Re: A couple of questions
|
|> CS>1) Is there any difference between "lo broda cu brode" and "da poi broda
 cu> CS>   brode"?  If they are the same, the statement "lo [unicorn] cu brode"
|> CS>   should be false, since noda cu [unicorn].
|
|cu'u la lojbab.
|> lo broda is not the same as da poi broda, and this is specifically one of
|> the differences - there is no claim that the referent exists in the unoverse
|> of discourse.
|...
|
|I still can't help feeling that this is BOGUS.
|I never did understand this concept, and I think the recent discussions
|have helped clarify the situation.
|Let me try to change your mind.
|
|You seem to be saying that {lo broda cu brode}, e.g
| (1) {lo mlatu je nanmu cu blanu}
|
|could be true, even if there is no such thing as a cat-man ***in the
|universe of discourse*** (far less the real world).  This doesn't make
|any sense to me whatsoever - I can't think of any interpretation of (1)
|which doesn't imply existence.
|
|I believe that all the situations where you might think you needed such
|a concept are better handled in other ways.
|
|Intensional contexts:
|    e.g. {nitcu}, {djica}, where there is an abstraction which
|the typical NL elides, but we strongly encourage to be
|acknowledged in Lojban.
|Veijo's "You may choose two books" is similar.  The abstraction
|is compressed rather than totally elided in English -
|{curmi lenu do cuxna re cukta}.  (There are some other issues
|in this translation which I'll skip over for the time being.)
|
|Typical objects:
|    e.g. "I like an apple", which is {mi nelci lo'e plise}.
|Some of these might even use {ro}.
|
|Hypothetical objects:
|    e.g. the long-sought unicorn, which is {lo [da'i] pavyseljirna}.
|The {da'i} is recommended for clarity, but we often live without it.
|
|I may have forgotten some of the problem cases, but my instinct is that
|they will all be soluble.
|
|I strongly recommend that {lo broda} be defined as equivalent to {da poi
|broda}, at least as far as the formal definition goes.  If we feel the
|need to allow some sort of laxity in informal jbosku, that's a different
|matter - we all live with various kinds of indiscretion in the rough and
|tumble of live usage - but I don't like muddying the basics.
|
|> CS>3) Does the sentence
|> CS>
|> CS>     mi djuno ledu'u do djuno ledu'u makau blanu
|> CS>   mean "I know you know what is blue" or "I know what you know to be
|> CS>blue"?
|> CS>   Instinctively, the former should be correct, and the latter meaning can
|> CS>   be expressed by
|> CS>
|> CS>     mi djuno ledu'u do djuno ledu'u makauxire blanu
|> CS>
|> CS>   Am I right?
|
|> Since "kau" is a discursive, it cannot be subscripted, so your solution is
|> rather vague in meaning - you have really subscripted the "ma".
|
|Eh, what?  I thought he'd got it 100% right.  Last time I heard, that
|WAS the way we distinguish nested constructs, including {kau}.  Is the
|disposition of the subscript a serious problem?
|
|> I would do the second as
|
|> mi djuno tu'a makau poi do djuno ledu'u ke'a blanu
|
|I'm not sure either way about this as it stands, but it certainly
|doesn't extend well to deeper nesting.
|
|mu'o mi'e .i,n.


|To: i.alexander.bra0125@oasis.icl.co.uk
|Subject: Re: A couple of questions
|
|IA>Hypothetical objects:
|IA>    e.g. the long-sought unicorn, which is {lo [da'i] pavyseljirna}.
|IA>The {da'i} is recommended for clarity, but we often live without it.
|IA>
|IA>I may have forgotten some of the problem cases, but my instinct is
|IA>that they will all be soluble.
|
|
|But if this is "da poi pavyseljirna" you are not being hypothetical at all
|you are CLAIMING existence:  da zo'u da
| pavyseljirna.
|
|I can accept discursive marking with "da'i" in non-logical discussions,
|but da'i seems incompatible with 'the present universe of discourse', it
|specifically implies to me that we are moving OUT of said universe.  At
|which point "da poi pavyseljirna" is highly questionable to me.
|
|lojbab


|Date:         Thu, 6 Oct 1994 20:03:15 EDT
|From: Jorge Llambias <jorge@PHYAST.PITT.EDU>
|Subject:      Re: A couple of questions
|To: Bob LeChevalier <lojbab@access.digex.net>
|
|> 1) Is there any difference between "lo broda cu brode" and "da poi broda cu
|>    brode"?
|
|I agree with Iain and And, and disagree with lojbab. They should mean the same
|thing.
|
|> If they are the same, the statement "lo [unicorn] cu brode"
|>    should be false, since noda cu [unicorn].
|
|If no unicorns exist in the world where the statement is used, then the
|statement is false in that world, yes.
|
|> 3) Does the sentence
|>
|>      mi djuno ledu'u do djuno ledu'u makau blanu
|>
|>    mean "I know you know what is blue" or "I know what you know to be blue"?
|>    Instinctively, the former should be correct, and the latter meaning can
|>    be expressed by
|>
|>      mi djuno ledu'u do djuno ledu'u makauxire blanu
|>
|>    Am I right?
|
|pe'i go'i, you're right.
|
|Jorge


|To: jorge@PHYAST.PITT.EDU
|Subject: Re: A couple of questions
|
|JL>> If they are the same, the statement "lo [unicorn] cu brode"
|JL>>    should be false, since noda cu [unicorn].
|JL>
|JL>If no unicorns exist in the world where the statement is used, then the
|JL>statement is false in that world, yes.
|
|1. Therefore the statement "Elves have pointed ears" is false since
|there is no such thing as an elf.  Likewise definitional statements
|"Elves are humanoid" is also false even if definitional.  How can you
|describe the properties of a hypothetical but non-existent object if any
|statement about such an object is false.
|
|2. If statements about non-existent objects are false, then their
|negation is true.  We can possibly weasel around this with "na" negation
|(and I think I did in the negation paper), but I am not sure.
|
|3> And then there is the argument that all statements about non-existent
|objects being equivalent to each other, since all are statements about
|the members of the empty set.
|
|I don't pretend to know the answers, but this is one of those questions
|that comes up again and again and I never am satisfied enough with
|whatever explanation is proposed to internalize it.
|
|But the status quo remains, as far as I know, that "lo [unicorn] cu
|brode" is not the same as da poi [unicorn] cu broda.  Cowan or pc are
|welcome to correct me, since they supposedly reolved this once before.
|
|lojbab


|Date:         Fri, 7 Oct 1994 20:34:08 +0100
|From: ucleaar <ucleaar@ucl.ac.uk>
|Subject:      Re: A couple of questions
|
|Iain:
|> if you really want to say that some-but-not-necessarily-all elves
|> have pointy ears, then you have to allow that such things as elves
|> exist, if only for the purposes of the discussion.
|
|This bit I don't go along with.  It should be possible to talk about >0%
|of all elves without allowing there are any elves, just as talking about
|100% of elves doesn't entail there are any.  [This happens to be what
|I'd prefer for the default interpretation of "lo":  i.e.  I agree with
|you & Jorge about what "lo" means now, but I agree with Lojbab about
|what it ought to mean.]
|
|---
|And


|From: Logical Language Group <lojbab>
|Subject: Re: A couple of questions
|Date: Tue, 11 Oct 1994 09:36:26 -0400 (EDT)
|
|la .and. cusku di'e
|
|> The lojbanic solution in such cases is usually to invent ways to
|> express both meanings (& to make both expressions "Zipfean" - i.e.
|> verbose in proportion to their infrequency). So I conclude that
|> we need:
|>   (1) all, not implying existence
|>   (2) all, implying existence
|>   (3) some-but-not-necessarily-all, not implying existence
|>       [This is the ">0%" I've advocated.]
|>   (4) some-but-not-necessarily-all, implying existence
|> (1) is "ro" & (4) is "lo" & "da". It would be nice to have a convenient
|> expression for (2) & (3).
|
|I believe that by the current interpretations "lo" is #3.  #2 can be handled
|by something like "rosu'o", "all of the at-least-one".
|
|--
|John Cowan              sharing account <lojbab@access.digex.net> for now


|Subject: Re: context in Lojban
|Date: Tue, 8 Nov 1994 15:51:10 -0500 (EST)
|
|la .and. cusku di'e
|
|> replying to Bob Chassell:
|> > So as not to confuse anyone with jargon like `+specific' and
|> > `-specific', remember, if the context is that there are a real and a
|> > non-real box in front of us, and our contextual range is constrained
|> > to those boxes, then
|> >
|> >     .i mi nitcu lo tanxe
|> >
|> > is *specific* as to which box, and
|>
|> So even if there exists a real box that you do need, but you need neither
|> of the boxes in the "contextual range", then the utterance is false
|> - according to you. I am incredulous that this really is the official
|> line on LO.
|
|It isn't.  "lo" binds the universe of discourse, but not the local context.
|"lo -unicorn" may refer in a universe where there are unicorns, but it
|would be absurd to suppose that "lo gerku" has no referent for me, merely
|because there are no dogs in the room (or, I suppose, the building) which
|I currently am in.


|From: Logical Language Group <lojbab>
|Subject: Cowan's summary #2: "lo" vs. "da poi"
|Date: Thu, 10 Nov 1994 17:45:35 -0500 (EST)
|
|The "official" line on "lo" and "da poi" has always been that they don't
|mean the same thing, because "lo -nonexistent" could be valid, whereas
|"da poi -nonexistent" was self-contradictory, as "da" can be glossed
|"there exists an X".  I now believe this to have been a mistake: "lo"
|under current definitions is the equivalent of "da poi", simply syntactic
|sugar.  However, I am going to propose a small change in interpretation
|that will give it added value.
|
|Historically, there were two kinds of cases for using "lo -nonexistent";
|those involving opaque contexts like "John wants a giant box" where the
|giant box might not exist, and those like "Elves have pointed ears" where
|elves have properties even though there are no elves.
|
|I believe that these can both be resolved, but in different ways.  The first
|case involves the opaque contexts I discussed in the previous part of this
|discussion:  we can use an embedded prenex to get the variable bound within
|the abstraction only.  Thus:
|
|1)      la djan. djica lo brabra tanxe
|        John desires a colossal box
|
|means the same as:
|
|2)      da poi brabra tanxe zo'u la djan. djica da
|        There-is-an-X which is-a-colossal box : John desires X
|
|and can only be true if there really is something which is a colossal box.
|On the other hand,
|
|3)      la djan. djica tu'a lo brabra tanxe
|        John desires something-about a colossal box
|
|converts to a "da poi" within the abstract bridi, and so is limited in scope,
|and needn't really exist.
|
|The case of elves is quite different.  I believe that merely by talking of
|elves, we (normally) put ourselves into a universe in which elves exist.
|In the >Midsummer Night's Dream< universe, the sentence  "Some elf is a king"
|is true; in the >Lord of the Rings< universe, it is false; and in the
|real universe, it is vacuously false.  (Yes, I know about Wood-Elves.)
|
|In any case, a statement about "lo -elf" works the same as "da poi -elf".
|There is absolutely no difference in meaning, though there is a noticeable
|difference in grammar; any sumti following "da poi broda" will be eaten
|by the "poi", whereas "lo broda" is self-contained.
|
|This is also a good result in that it allows the outer quantifier of "lo"
|to be "su'o" = "at least one" without restriction; "lo -nonexistent" either
|indicates a shift in the universe of discourse so that the set referred to
|becomes non-empty, or involves the speaker in a vacuously false statement.
|
|However, I would like to propose instituting one difference between "lo"
|and "da poi":  that "lo" be given an implicit outside quantifier which
|mutates across a negation boundary.  This means that:
|
|4)      lo nanmu klama le zarci
|        Some men go to the store.
|
|and
|
|5)      da poi nanmu cu klama le zarci
|        Some X's which are-men go to the store
|
|mean the same thing, but
|
|6)      lo nanmu na klama le zarci
|        Some men don't go to the store.
|
|and
|
|7)      da poi nanmu cu na klama le zarci
|        It is false that some X's which are men go to the store.
|
|mean different things:  Example 6 is true as long as at least some men don't
|go to the store (on the given occasion), whereas Example 7 require that
|no men go.  In effect, "lo broda" transforms to a "da poi broda" with
|widest scope, even wider than sentential negation.
|
|Providing this feature is not strictly necessary, but may make the use of
|negation somewhat simpler, because it means that both "lo" and "le" commute
|with negation, i.e. are in effect singular terms.  It remains true, as
|Jorge and And have said since the beginning, that "le" is +specific and
|"lo" is -specific (pc's claim that "lo" was +specific turns out to have been
|founded on a misunderstanding of the terms).
|
|Comment on this proposal?
|--
|John Cowan              sharing account <lojbab@access.digex.net> for now


|Date:         Sat, 26 Nov 1994 14:23:35 -0800
|From: "John E. Clifford" <pcliffje@CRL.COM>
|Subject:      Ol Uncle Tom Cobleigh
|
[...]
|        One of the realms of (usually) friendly disagreement is about the
|range of the quantifiers.  The extreme positions here are that bound
|variable range only over actually existing things in this world (the
|extreme extreme would not even allow numbers and suchlike mathematical
|entities) and that they range over the existents of all possible worlds
|(and maybe even impossible ones). Lojban has officially (I think) taken
|the view that quantifiers in a world range over the things that exist
|(zaste) in that world.  McCawley* reminds us that ordinary languages are
|more generous than that, though less than a rabid substitutionalist would
|allow, and Lojban practice has tended to follow natural language practice
|when no one hauls it back into line with theory.  As McCawley notes, we
|tend to allow the characters of established fiction and even those loosely
|related to them -- Ulysses and Sherlock Holmes and Data and even
|Sherlock's mistress and my son are all in (having a name is a sufficient
|condition in almost all of the expansionist logical views) and even,
|occasionally, unnamed members of classes from such tales (unicorns, for
|example, none of which has a name -- that I know of, anyhow).  We say such
|paradoxical things (for a strict constructionist) as "There are
| mythical beasts" which we then instantiate, when challenged, with
|"Unicorns".  And we surely allow that, if both Rembrandt and Picasso drew
|pictures of Zeus, that there is something (even someone) that they both
|drew pictures of -- even if they illustrated different events in the
|tales, so the quantification is not over events in which Zeus appears.
|And, of course, we quantify -- in English often but constantly in Lojban
|-- over events that never obtain.  (I personally have no objection to
|saying that events exist even if they never obtain, but even some
|expansionists are not so liberal.)
|        Even if we allow a richer range for variables than just the real
|existents, we still need to bring our focus back to them from time to
|time: for science if nowhere else but for many practical matters as well.
|The laws of physics nor aerodynamics need not account for flying carpets
|nor is even the reformed Scrooge a source for a loan. And many predicates
|require that their various arguments have the same ontic status: we can
|only hit what is in the world with us, for example.  We could, within an
|expanded quantifier range, always haul back to the real with an
|appropriate predicate, _zaste_ in Lojban, "real" or "exist" in English.
|But in a logical language, it is often more convenient to use a different
|quantifier for the two cases. And right now we seem to have two quantifier
|sets floating around, one of which is already being used (pretty much
|unofficially) for what bes and the other for what exists -- except that
|the two concepts are not separated consistently.  Using _da_ alone and
|with _poi_ (for predicates non-empty in this world) for the strong
|quantifier ("there exists") and _lo_ for the weak ("there is" or "bes")
|would solve a number of current controversies and a few old ones and head
|off a few that have been long abrewing but have not yet -- or only just
|now -- come on the scene: the problem with event descriptions as needed
|for intensional predicates, for example.  I recommend that we change the
|official line accordingly.
|        All of which has almost NOTHING to do with opacity.
[...]


|From: Logical Language Group <lojbab>
|Subject: Re: existential quantification
|Date: Mon, 5 Dec 1994 14:49:02 -0500 (EST)
|
|la .and. cusku di'e
|> I have been told, in the last few months, that "nu" doesn't entail
|> its complement bridi is true, but I should have thought that the
|> existentially quantifying preceding "lo" does require there to
|> be an event.
|> Have I gone wrong?
|> What is the solution?
|
|In one sense you are right, in another sense you are wrong (as you might
|expect).  Saying "lo nu" = "da poi nu" does entail that the event exist.
|However, an event can exist independently of whether the encapsulated
|bridi actually happens.  Thus, you can speak of "lo nu mi ninmu" even
|though you are not a woman.  To assert that some event actually takes
|place, use "fasnu":
|
|        da poi nu mi nanmu cu fasnu
|
|So I think that your proposed use of "si'o" is not necessary.
|This is not official, merely my opinion.
|
|--
|John Cowan              sharing account <lojbab@access.digex.net> for now