archives of the CONLANG mailing list
------------------------------------

>From thorinn@tyr.diku.dk Mon Oct 11 11:02:29 1993
Date: Mon, 11 Oct 93 11:02:28 +0100
From: thorinn@diku.dk
Message-Id: <9310111002.AA18094@tyr.diku.dk>
To: conlang@diku.dk
Subject: You are still subscribed to CONLANG

Several subscribers have attempted to resubscribe to CONLANG. It is
not necessary---while the list has been thunderously quiet since
September 21st, it should still be fully operative.

But if you want to READ about conlangs, you have to WRITE something.

Lars Mathiesen (U of Copenhagen CS Dep)  (Humour NOT marked)



>From bhelm@majestix.cs.uoregon.edu Tue Oct 12 01:17:37 1993
Message-Id: <9310120017.AA26654@majestix.cs.uoregon.edu>
To: conlang@diku.dk
Subject: What is possible in pronoun systems that mark sex?
Date: Mon, 11 Oct 93 17:17:22 -0700
From: bhelm@majestix.cs.uoregon.edu


A question for linguists:

People on the Esperanto mailing lists are talking about purging their
language of sexism.  One aspect of this is a search for a sex-neutral
third-person singular pronoun.  I am curious: what pronoun systems
mark sex, but that also incorporate a neutral pronoun for persons?

Let me explain: the Esperanto system of third-person pronouns runs
something like the following:
						|
				       -------------------------
				      |				|
Specificity			  SPECIFIC		    NON-SPECIFIC
				 _____|_________		|
				|		|	      "oni" (one)
Personification		      PERSON	    NON-PERSON
				|		|
			 -------|-------       "gxi"(it)
			|		|
Number	  	    SINGULAR	      PLURAL
			|		|
			|	      "ili" (they)
			|
		 -------------------------------
		|		|		|
Sex	       MALE	      FEMALE	UNKNOWN/IRRELEVANT
		|		|		|
	       "li" (he)      "sxi" (she)      ?????

Zamenhof, the creator of the Esperanto fundamentals, implied one could
fill in the ???? with 'gxi', the singular pronoun for non-persons.
His proposal is resisted by many speakers.  Others have proposed
creating a new pronoun, or adopting the demonstrative for specific
objects and persons ('tiu') as a pronoun.  Colloquially, speakers of
English in the United States (and perhaps elsewhere) have adopted
"they" for this purpose.

In languages like German, there is a neuter grammatical gender, but
this is different: the neuter pronoun "es" can have a referent whose
sex is known to be (for instance) female.  Also, the masculine and
feminine genders apply to non-persons.

As I understand it, very few languages mark sex in their pronoun
systems the way English and Esperanto do.  Of those that do, have any
developed a neutral pronoun for persons?  If so, how?


Rob Helm (bhelm@cs.uoregon.edu)



>From C.J.Fine@bradford.ac.uk Tue Oct 12 14:03:43 1993
Date: Tue, 12 Oct 1993 14:03:35 +0100
Message-Id: <18877.199310121303@discovery.brad.ac.uk>
From: Colin Fine 
To: conlang@diku.dk
Subject: Re: What is possible in pronoun systems that mark sex?

All the languages I can think of that distinguish grammatical
gender have distinct masculine and feminine third-person
pronouns in the singular, and occasionally in the plural.
(The languages in question are almost all the Indo-European
languages, plus Hebrew and Arabic. I don't know whether 
other Afro-Asiatic languages meet this description or not).
Even Indo-European languages that have all but lost the 
masculine/feminine grammatical distinction (English, 
modern Scandinavian languages, and - I think? - Farsi)
retain this distinction in the pronouns.

I do not know of an example which has both these distinct
pronouns and a common one.

The problem is even greater in a language like Hebrew 
which in many cases marks the gender of the subject
in the verb!

Colin Fine
 
=================================================================
=======
There are no extraordinary people.  | Colin Fine
Whoever tells you otherwise is      |  Dept of Computing
      lying to you.                 |   University of Bradford
There are only ever ordinary people,|    Bradford, W. Yorks, England
Who do what they do -               |      BD7 1DP
The extraordinary thing is the      | Tel: 0274 733680 (h), 383915 (w)
  extraordinary things that they do!|   c.j.fine@bradford.ac.uk
                                    |   (cfine@cix.compulink.co.uk;
 .e'o ko sarji la lojban.           |      cjfine@gn.apc.org)
  EXLIB = EXpansion of LIBrary systems for the visually disadvantaged
=================================================================
=======


>From shoulson@ctr.columbia.edu Tue Oct 12 18:22:34 1993
Received: from sirius.ctr.columbia.edu by odin.diku.dk with SMTP id AA01304
  (5.65+/IDA-1.3.5 for /usr/spool/listserv/catmail -L CONLANG -f); Tue, 12 Oct 93 18:22:34 +0100
Received: from startide.ctr.columbia.edu by sirius.ctr.columbia.edu (5.67/5.10) with SMTP id AA00630; Tue, 12 Oct 93 13:22:20 -0400
From: (Mark E. Shoulson) shoulson@ctr.columbia.edu
Received: by startide.ctr.columbia.edu (5.67/SMI-4.1)
	id AA18473; Tue, 12 Oct 93 13:22:20 -0400
Date: Tue, 12 Oct 93 13:22:20 -0400
Message-Id: <9310121722.AA18473@startide.ctr.columbia.edu>
To: conlang@diku.dk
In-Reply-To: bhelm@majestix.cs.uoregon.edu's message of Tue, 12 Oct 93 12:20:08 +0100 <9310120017.AA26654@majestix.cs.uoregon.edu>
Subject: What is possible in pronoun systems that mark sex?

>Date: Tue, 12 Oct 93 12:20:08 +0100
>From: bhelm@majestix.cs.uoregon.edu


>A question for linguists:

>People on the Esperanto mailing lists are talking about purging their
>language of sexism.  One aspect of this is a search for a sex-neutral
>third-person singular pronoun.  I am curious: what pronoun systems
>mark sex, but that also incorporate a neutral pronoun for persons?

Oh, they're on that tack *again*?  Some things never change.

>Zamenhof, the creator of the Esperanto fundamentals, implied one could
>fill in the ???? with 'gxi', the singular pronoun for non-persons.
>His proposal is resisted by many speakers.  Others have proposed
>creating a new pronoun, or adopting the demonstrative for specific
>objects and persons ('tiu') as a pronoun.  Colloquially, speakers of
>English in the United States (and perhaps elsewhere) have adopted
>"they" for this purpose.

I don't know that using "tiu" is really an extension of Esperanto, or a new
usage.  It really seems to me to be a perfectly normal usage of the word,
and one that fills the bill admirably in just about alll cases not covered
by "si".  After all, "tiu", being the demonstrative, as you say, means
something like "that one" or "that person".  So what's wrong with "Se vi
renkontas iun, salutu tiun!" ("If you meet someone, greet that-person!")
The English sounds wordy, but that's just because we don't have a concise
pronoun that works like "tiu".  It seems to me that "tiu" (and "ties")
handles just about every case I've heard of needing a gender-neutral
third-person pronoun in Esperanto (except where it's already handled by
"si"/"sia").  Wellllll, maybe the Zamenhof aphorism mentioned above is an
exception, I think it goes something like "Avarulo deziras ke cxiu servu
gxin" ("A greedy person desires that all should serve it [sic]").  "Sin" is
plainly wrong, since that would mean everyone should serve himself.  "Tiun"
may or may not work, I'm not sure.

~mark



>From dan@atldbs.dbsoftware.com Tue Oct 12 23:13:05 1993
Date: Tue, 12 Oct 1993 17:23:08 -0400
From: dan@atldbs.dbsoftware.com (Dan McGinn-Combs)
Received: by atldbs.dbsoftware.com (AIX 3.2/UCB 5.64/3.1.090690-Dun & Bradstreet Software)
	id AA21552; Tue, 12 Oct 1993 17:23:08 -0400
Message-Id: <9310122123.AA21552@atldbs.dbsoftware.com>
To: conlang@diku.dk
Subject: Re: What is possible in pronoun systems that mark sex?

Colin Fine writes:
(English, modern Scandinavian languages, and - I think? - Farsi)
retain this distinction in the pronouns.

In Farsi, the pronoun catagorizes, but slightly differently.  While marking
either male or female (or neither male nor female) with the single pronoun 
"u," 
there is a distinction between animate (and sentient) beings and non-living
(or non-sentient) things ("an" or in the plural, "anha").
"An" and "anha" are the singular and plural forms of the word
for "that."

And this from an Indo-European language, influenced heavily by Arabic!




>From lojbab@access.digex.net Sun Oct 17 12:34:59 1993
Date: Sun, 17 Oct 1993 07:34:49 -0400
From: Logical Language Group 
Message-Id: <199310171134.AA27266@access.digex.net>
To: conlang@diku.dk, lojban@cuvmb.cc.columbia.edu
Subject: news from TLI (colored by the Lojban point-of-view)

Well, JCB just put out a new issue of TLI's publication "Lognet".  It
has a bit of interesting stuff, information, ideas, etc.  As usual, most
of these ideas are just ideas - JCB has always had a knack for making a
skeletal idea (like Loglan was) look like a complete, ready-to-roll
accomplishment.

But he has made one idea an accomplishment at last - he "beat us" in
getting out a new dictionary.  Well, sort of.  He has issued an
electronic dictionary (i.e. data files, and some support software on the
MacIntosh version - I'm informed that the PC version does not have much
support software).  There is no indication of a soon-to-come printed
dictionary.  Indeed, it looks like TLI is planning to limit itself to
electronic form for some time to come, issuing an updated on-line
dictionary as often as 2 or 3 times a year.

Lojban will have one advantage, when I get our dictionary done, though.
Our current policy is that our dictionary files, at least in raw form,
will be available free-of-charge, and indeed will be in the public
domain.  Updates will be posted to our ftp site(s) as they are created.
JCB/TLI is selling their dictionary for $50 ($30 for TLI members) and
charging $5 for updates.  I suspect that a free dictionary will cause
Lojban to spread a bit more rapidly than a $50 dictionary, though we
will have to hope that a lot of you are willing to buy the printed
dictionary and/or support us with donations to make up for the loss in
revenue.

In purchasing the dictionary, JCB informs his readership that "you have
the right -- yes, the _right!_ -- to edit your own dictionary, and to
send your edits and additions in to us for possible inclusion in the
next 'official' edition."  How generous .ianai

Although we are nominally at peace with TLI, I am informed that an order
for the dictionary from me or from LLG would not be welcomed by JCB.  I
don't know whether this means it would be refused, ignored, whether it
would prompt renewed hostilities, or whether it merely means that JCB
sees red when he views my name, and thus might be less amenable to other
efforts at enhancing cooperation between our groups.  In any event, I am
interested in having someone who is not well-known as a leader of
Lojban, purchase a copy of TLI's dictionary (PC form, not MacIntosh) and
donate it to LLG (for my/our use), not mentioning this intent in your
order, of course.  You will receive appropriate credit for such a
donation.  We could possibly use more than one copy (for Cowan and
perhaps Nick), but I don't want to enrich TLI by buying unnecessary
copies.  If you are interested in helping, please let me know.

Other news (mostly) from TLI:

While not directly saying so, JCB indicates that his community is around
160 US people in size, with maybe another 120 inactives.  This is
deduced from the fact that he is now using a bulk mailing permit, and is
mailing about 30-40 'extra copies' to non-paying people, who are in turn
getting a copy every 3 or 4 issues.  Our paying community is somewhat
larger, but not much, but we know our people are real supporters of the
language who have in general renewed their subscriptions after some time
receiving them.  Most TLI people are still on their initial
subscriptions, and he has a history of very low renewal rates (as low as
10-20%).

It turns out that JCB and TLI did not participate in the SanFrancisco
WorldCon in August, though the editor of Lognet was apparently in charge
of registration at that convention.  Keith Lynch took several dozen
brochures to WorldCon, all of which disappeared quickly (as I expected;
we went through some 600 brochures at Noreascon, the 1989 Worldcon in
Boston), but I have had no responses indicating Worldcon as a source for
their contact.

(I did discover yesterday that Lojban List has made it into a "new"
(1993 edition) book on Internet mailing lists that is popular enough to
be in a Crown Books bookstore, though it gives the list outdated snark
address for the list.  Since it gives our snail address and Cowan's
snark address is still (or rather hopefully will soon again be) good,
this should provide some new visibility for Lojban and this list.)

JCB answers a letter, providing the interesting information that Loglan
was originally an NA order (Noun-Adjective) language, but that he
changed it to conform with Greenberg's language universals after they
were published in 1963.  Unfortunately, other than the snippets of
Loglan in the 1960 Scientific American article, the only documentation
of Loglan as it was before 1966 lies in JCB's head and files.  There was
a locally-circulated (in Gainesville FL) document on the language in
1955 but the language truly was sparse then, having less than 300 words
and only the beginning of a grammar.  I know of no one besides JCB who
has a copy of this (though I'd love to find one for the archives.)

There is a short chunk of original Loglan text in this Lognet, from Bob
McIvor.  I'm asking Bob for an on-line copy which we would post on
Lojban List with a Lojban equivalent/translation, whereupon all of you
can tear it apart as you so aptly do all posted Lojban text.  It might
be interesting.  This text, and indeed this issue, shows the development
of the first significant stylistic difference between TLI Loglan idiom
and Lojban idiom.  While Lojban supports it, few Lojbanists make use of
lerfu (letter name) words as sumti anaphora (back-referencing pronouns);
it is coming to be heavily used in TLI Loglan, since they have realized
the weakness of their other form of anaphora, which is the equivalent of
our ri/ra (but with 5 members to the set, all strictly counted).

(Nora actually has always like using lerfu, but she seldom writes in the
language, and the technique never seems to occur in our spoken Lojban
here.  The closest to significant use has been the occasional use of
lerfu as 'meaningful' delimiters on zoi quotes of non-Lojban text. lerfu
anaphora have some advantage over the ko'a series in that generally the
letter indicates the first letter of the name or brivla being
referenced, a useful memory hook, and one that does not require explicit
assignment with 'goi' unless there is ambiguity due to multiple possible
referents.  Anyone who wishes may experiment with this usage in your next
text(s)).

Finally, and perhaps most interesting, JCB wrote and presented a paper
on the lessons he has learned in inventing and teaching Loglan for 35
years.  He intends to publish this paper in a TLI publication, which I
look forward to seeing.  But he reports that in writing this paper, he
came to believe that Loglan is now ready for the development of a
speaking community.  He then proposes a way to bring it about, one which
could actually work, though I suspect not right away in the case of the
TLI version of the language.  Still, JCB and TLI might have the means to
bring it about and if they get the people to participate, they COULD (I
won't say "will") quickly catch up and surpass our accomplsihments with
Lojban as a spoken language.

Specifically, JCB proposes setting aside his Florida home as a
Loglan-using center.  People would visit this place for anywhere from 1
to 4 weeks for the purpose of learning and using the language.  In order
to have it work, he says that he needs to have enough people sign up so
as to maintain a continuous community of people there year-round.  He
suggests 8 people at one time.

He argues that the TLI Loglan community is ready for this because there
are a dozen or so people who know his language well enough to solve most
any usage question "within a very few days -- well, weeks at the most",
or weeks at most".

I doubt if JCB realizes how far TLI is from where they are (about where
Lojban usage was in early to mid 1989) to where we are now, much less
from where he would like to be.

Now I suggest that such an attempt would probably not work for Lojban
yet, much less for TLI, though we could come close to pulling it off.
Unlike TLI, we HAVE sustained real conversation for hours at a time.  We
have a dozen people who could probably in a face to face communication
setting solve any usage question communicatively within a few minutes,
at most.  So if we actually could get a group of our people in one place
for a couple of weeks, they probably could try, and succeed, by the end
of that period in not merely sustain conversation in Lojban, but living
at a "vacation level" in Lojban, performing normal days' activities in
the language.  It would probably be one step more difficult to live a
full life in a language, including work, etc., as Rob Brady's jargony
texts showed a couple of months ago, even if one could get enough people
together who could work fully in the language for any signififcant
period of time.

I think we would be very lucky if we could get enough people to keep
this going for a month, because to make it work, you need enough people
from your first group to overlap the next group long enough to teach
them enough everyday usages that the wheel doesn't need to be reinvented
every week.  Having 8 people in your group, with say, 4 person turnover
per week, requires that you have 20 people for one month alone.  For 2
weeks per person twice a year (rather more than most people can really
manage), you would need 120 people, and more rationally about 4 times
that number, to sustain such a group continuously, as JCB thinks he can
do.  While merely trying to do so, would get a lot more people
interested in learning the language, that is a lot more people than >I<
expect to get involved at this level of intensity for another year or
two.

This does not allow for the fact that, as far as I know, no one in the
TLI community other than JCB is committed to his language >as a family<
(and I'm not sure about the level of interest in speaking the language
JCB's wife has).  Most of us with husbands/wives and kids haven't gotten
the rest of our family interested at the same level we are.  My family
is, as far as I know, is the furthest along in this way, since Nora and
I are among the best speakers of the language, but our kids are just
beginners, and we cannot use the same techniques in teaching them that
we have used on adults.  For us to go cold turkey to pure Lojban would
be virtually impossible.  I dare say, most others with a
non-Lojban-speaking family member would have even greater problems.  And
families would have a greater difficulty committing a great deal of time
to such an experiment - I can't see my kids interacting only with each
other and adults for 2 days, much less 2 weeks, without someone going
crazy %^).  And enforcing Lojban-only on kids is really impossible.
Thus you need a sizable portion of your 'community' as single people.
That means mostly college students, which in turn constrains the ability
to sustain a continuous group to summer months.

(And this does not allow for the conviction that I and others in the
Lojban community share, that JCB overestimates the skill of himself and
his people in the language.)

I think the Lojban community might be able to pull something off like
this for several weeks or a couple of months, provided that there is
some way to get a stable core, and with the added proviso that there
would be (at least at first) some non-Lojban periods.  We've talked, for
example, of having an area at LogFest deemed Lojban-only (with another
area deemed English-permitted).

If enough people said they wanted to try it, we could perhaps start next
year at LogFest here (which would be mixed Lojban and English in
whatever manner people choose - I don't want people to stay away from
LogFest because they aren't ready for such an ambitious undertaking in
the language), and extend through the following week or so trying to
increase the percentage of Lojban used (possibly by shrinking the
English-permitted environment).

If such an experiment were even moderately successful, I suspect
>Lojban< might be ready for a more ambitious experiment of several weeks
or more by the following year.

I can't see TLI's language being ready for such a challenge within the
next year or so, much less their people.  But they have the advantage of
a place that can be dedicated to Loglan use which people might be
willing to go to for a vacation (albeit in winter - JCB's Gainesville's
place is a muggy swamp by the end of May), and JCB's fairly deep purse
(stoked by new Careers royalties).

But the idea is good, and WE can think about it.  Would anyone be up to
trying to do this next summer?

lojbab

lojbab
----
lojbab                           Note new address:    lojbab@access.digex.net
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA                        703-385-0273
 
For information about the artificial language Loglan/Lojban, please
provide a paper-mail address to me.  We also have information available
electronically via ftp ( casper.cs.yale.edu, in the directory
pub/lojban) and/or email.  The LLG is funded solely by contributions,
and are needed in order to support electronic and paper distribution.



>From EZ-as-pi@cup.portal.com Sun Oct 17 13:01:21 1993
To: lojban@cuvmb.cc.columbia.edu, conlang@diku.dk
From: EZ-as-pi@cup.portal.com
Subject: Re: news from TLI (colored by the Lojban point-of-view)
Lines: 33
Date: Sun, 17 Oct 93 05:01:16 PDT
Message-Id: <9310170501.2.28694@cup.portal.com>
X-Origin: The Portal System (TM)

The proposal by Lojbab is quite inter-
esting. While I myself could never
conceive of being a part of it (my
facility in Lojban is such that I could
scarcely engage in continuous conversa-
tion for any significant time in it,
even if I had the time to spare) I am
not uninterested in finding out how
successful a project like this could be
-- it is an attempt to create a
Volapuek-only environment that proved
to be the undoing of Volapuek, and by
contrast one of the successes of 
Esperanto is that they do manage to
function in Esp'o-only environments,
despite my dislike of that language.
So the ability to pull this off would
mean that Lojban is demonstrating that
it can in a practical sense function
as alanguage.
If enough people do sign on to this
idea to carry it off, I hope that a
report on how successful it proved to
be will be posted onto the net. While
I have some doubts that it can work,
I wish Bob well in his attempt to
advance the language to this state,
because from my point of view it is a
great experiment: Can a language whose
grammatical structure is as unconven-
tional as Lojban really develop a
speaking community?
                        Bruce





>From ram@netcom.com Wed Oct 20 13:03:05 1993
Date: Wed, 20 Oct 93 05:01:35 -0700
From: ram@netcom.com (Rick Morneau)
Message-Id: <9310201201.AA27019@netcom6.netcom.com>
To: conlang@diku.dk, ram@netcom.com
Subject: Ideas for Discussion


Howdy conlangers!

Since this list has been awfully quiet for several weeks, perhaps we
need a spark to get the discussions rolling again. Below is a list
of some topics that I would enjoy discussing. Note that all of the
following topics are relevant to the design of NEW conlangs, rather
than to the use of EXISTING conlangs. Here's the list...

     1. How should semantic case roles be marked?
     2. Should the argument structure of verbs be marked?  If so, how?
     3. Should nouns be classified?  If so, how?
     4. Should parts-of-speech be marked?  If so, how?
     5. How should tense, aspect and modality be implemented in a conlang?
     6. How should voice (active, passive, middle, etc.) be implemented
        in a conlang?
     7. How should metaphor be dealt with in a conlang?

If anyone has an interest in any of these topics, speak up! Also, if
you're interested, but you're not sure what I mean by some of the
questions, I'm willing to post a brief introduction to any of the above
topics to get things started.

And if you're not interested in the above topics, how about providing
some topics that YOU would like to discuss? This group used to be
lots of fun. Let's see if we can make it so again.

Regards,

Rick

*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
=*   Rick Morneau                 ram@netcom.com            =*
*=   Idaho Falls, Idaho, USA      "Make love - not laws!"   *=
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*



>From C.J.Fine@bradford.ac.uk Wed Oct 20 13:42:58 1993
Date: Wed, 20 Oct 1993 13:42:51 +0100
Message-Id: <14899.199310201242@discovery.brad.ac.uk>
From: Colin Fine 
To: conlang@diku.dk
Subject: Re: Ideas for Discussion

Rick asks:
++++++>
      1. How should semantic case roles be marked?
     2. Should the argument structure of verbs be marked?  If so, how?
     3. Should nouns be classified?  If so, how?
     4. Should parts-of-speech be marked?  If so, how?
     5. How should tense, aspect and modality be implemented in a conlang?
     6. How should voice (active, passive, middle, etc.) be implemented
        in a conlang?
     7. How should metaphor be dealt with in a conlang?
>++++++

I respond:
	1. Should semantic case roles be marked?
	2. What are verbs?
	3. What are nouns?
	4. What are parts-of-speech?
	5. What are tense, aspect and modality?
	6. What is voice?
	7. Should metaphor be dealt with in a conlang?

(No, I'm not saying I don't understand the questions. I'm responding
the equivalent of the lojban "na'i", ie "there is an error not in
but underlying your utterance" - specifically that in asking your
questions you are making many assumptions that are not necessarily
warranted.)


	Colin Fine



>From thorinn@tyr.diku.dk Wed Oct 20 16:41:41 1993
Date: Wed, 20 Oct 93 16:41:40 +0100
From: thorinn@diku.dk
Message-Id: <9310201541.AA00710@tyr.diku.dk>
To: conlang@diku.dk
In-Reply-To: Colin Fine's message of Wed, 20 Oct 93 14:20:09 +0100 <14899.199310201242@discovery.brad.ac.uk>
Subject: Ideas for Discussion

   From: Colin Fine 

	   1. Should semantic case roles be marked?

Well, the semantic roles of the basic lexical items in an utterance
must be conveyed somehow. I think.

	   2. What are verbs?
And what are argument structures?
	   3. What are nouns?
	   4. What are parts-of-speech?
	   6. What is voice?

These are all structural information associated with or imposed on an
lexical item, allowing the semantic role of the item or associated
items to be deduced in the absence of explicit marking.

	   5. What are tense, aspect and modality?

Information about the relation of the utterance to its context. The
question is whether this information should be distinguished and
specially implemented.

	   7. Should metaphor be dealt with in a conlang?

A language design should explicitly consider whether to copy the
widespread use of metaphor in expanding the semantic range of lexical
items. Cf. the discussion of ``looking _through_ windows.'' I see no
reason to regulate the use of metaphor for effect, in cases where the
language does have a ``normal,'' but boring, way to express something.

I have an embryonic idea for a language where 2-6 above are eliminated
by associating each lexical item in a sentence with an explicit ``role
descriptor.'' Not just case markers, but also things like "nexus verb"
and "time adverb". There's a couple of other radical ideas in there as
well.

Lars Mathiesen (U of Copenhagen CS Dep)  (Humour NOT marked)



>From jrk@sys.uea.ac.uk Wed Oct 20 17:26:04 1993
Date: Wed, 20 Oct 93 17:29:21 BST
Message-Id: <18742.9310201629@s5.sys.uea.ac.uk>
X-Sender: jrk@139.222.1.5
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: conlang@diku.dk
From: jrk@sys.uea.ac.uk (Richard Kennaway)
Subject: Re: Ideas for Discussion

Another issue in conlang design:

        8. How should one choose the concepts for which the language
           will contain words?

(Colin, do you want to na'i this as well?  "What are words?"  "What are
concepts?"  "What is a language?" :-))

My question is sparked off by a message Lojbab posted to the Lojban list on
the subject.  To have a word for a concept embodies the assumption that the
concept is meaningful, an assumption which one might (depending on one's
purpose in designing the conlang) be undesirable.  Lojban has no word for
phlogiston, not (I assume) because there is no such thing, but because
historically, the concept broke down into incoherence under the weight of
experimental evidence.  The concept developed, changed, and disintegrated,
and there is now no useful definition which one could give to the word in a
conlang, other than "any of a range of notions which were at certain times
erroneously thought to underlie the phenomena of fire and heat"; a concept
of little utility except in the history of science.

In contrast, there's no such thing as a unicorn, but while I doubt Lojban
has a word for it at present, it is easy to think of a compound (e.g.
"one-horned horse") which will be understood at once by anyone familiar
with the concept of a unicorn.  "Unicorn" is meaningful; "phlogiston" was
once meaningful, but is so no longer, except in a context of imagining for
fictional or historical purposes that we do not know certain things which
in fact we do know.

Which words in common use today denote meaningless concepts?  Is it
possible to avoid incorporating such concepts into the vocabulary of a
conlang?  Of those concepts which are to be included, how shall they be
precisely defined?  How precisely shall they be defined?

The conlangs I know of, with the exception of Loglan/Lojban, appear
semantically naive.  A word in the conlang is typically defined by a single
word or phrase in some natlang, with the assumptions that (a) this is
sufficient to make the meaning clear, and (b) all the concepts one would
need can be specified in this way.  A few conlangs, such as Laadan,
deliberately include concepts that can only be explained in a more
discursive manner (e.g. "a supposed holiday, the preparations and execution
of which are more burdensome than one's everyday work").

Lojban is an exception, and it is there that the question of choosing the
definition of the vocabulary has recently come to the fore.

Consider the Lojban word "klama".  This is typically glossed in English as
"come".  The true meaning is much more complicated.  It denotes a relation
among five things: an object which moves, a destination, a starting place,
a route, and a means or vehicle. If any of these things is not present in
the scene one is describing, the relation "klama" does not apply.  They
need not all be specified when the word is used, but any left unspecified
are assumed to be understood from context and must exist for the use of the
word to be correct.

Why has klama been defined in this way?  Why this particular concept of
motion-with-source-and-destination?  The choice of concepts is based to a
large extent on Eaton's list of the most frequent concepts in several
European languages.  But when Eaton gives words for "amusing" in four
languages, what is the precise concept she is referring to?  Here is a
Lojban definition:

zdile x1 (abstract) is amusing/entertaining to x2 in property/aspect x3; x3
is what amuses x2 about x1

Why does Lojban separate the thing which is amusing and the specific
features of that thing which are amusing?  Is this separation implicit in
the notions denoted by the European words?

These are rhetorical questions to illutrate some of the issues; discussion
of Lojban vocabulary belongs on the Lojban list.

--                                  ____
Richard Kennaway                  __\_ /    School of Information Systems
Internet:  jrk@sys.uea.ac.uk      \  X/     University of East Anglia
uucp:  ...mcsun!ukc!uea-sys!jrk    \/       Norwich NR4 7TJ, U.K.




>From KNAPPEN@VKPMZD.kph.Uni-Mainz.DE Wed Oct 20 18:37:25 1993
Date: Wed, 20 Oct 1993 18:34 +0200
From: KNAPPEN@VKPMZD.kph.Uni-Mainz.DE
Subject: Verbs, Nouns and other parts of speech
To: conlang@diku.dk
Message-Id: <01H4CG1LHLMO8WVZMS@VzdmzA.ZDV.Uni-Mainz.DE>
X-Envelope-To: conlang@diku.dk
X-Vms-To: GATEWAY"conlang@diku.dk"

Rick triggers interesting questions. It seems that all natlangs have a 
distinction between verbs and nouns. In inflecting or agglutinating 
languages this distinction is often clear by form, in isolating languages 
you need more subtle methods of discrimination (can be counted/ cannot be 
counted, can be inserted in pattern X/ cannot etc.). However, natlangs are 
very different in the emphasis of verbs and nouns. The european languages 
we are most exposed to put most of the information into the nouns, in 
headline style the verb is often entirely missing. There are lots of nouns, 
compund nouns and means of modifying nouns.

If I understand east asian languages right, they are very different in this 
respect: Most information lies in the verb, and they use compound verbs 
quite often, forming even series of verbs which are quite impressive. Nouns 
can be entirely missing, if clear from context.

No other parts of speech are really required -- adjectives, adverbs, 
adadjectives, pre- and postpositions and other particles may, but need not 
occur. I don't know if there is a prefered order (say: If a language 
contains adverbs, it also contains adjectives). Most striking for europeans 
is the potential absence of adjectives. They can be united either with 
verbs or nouns (The sky is blue may be expressed as: There is the blue of 
the sky or as: The sky blues.)

I think a conlang becomes easier, the less parts of speech it contains, 
allthough the way lojban does handle this problem is quite mind-boggling to 
me. And yes, lojban has verbs and nouns, allthough they are not called this 
way. 

Enough for now,

J"org Knappen.



>From ucleaar@ucl.ac.uk Wed Oct 20 18:42:55 1993
From: ucleaar 
Message-Id: <42939.9310201742@link-1.ts.bcc.ac.uk>
To: conlang@diku.dk
Subject: Re: Ideas for Discussion
Date: Wed, 20 Oct 93 18:42:21 +0100


In reply to Rick:
>      1. How should semantic case roles be marked?

Do they exist?

>      2. Should the argument structure of verbs be marked?  If so, how?

Maybe. Inflectionally, perhaps.

>      3. Should nouns be classified?  If so, how?

No.

>      4. Should parts-of-speech be marked?  If so, how?

Perhaps by a certain correspondence of phonological representation
& syntactic category. You might also ask: what is the minimum
number of parts of speech necessary, and what is the optimal number.
(E.g. Lojban has a two or 3 dozen.)

>      5. How should tense, aspect and modality be implemented in a conlang?

By words that take words referring to events as their arguments.

>      6. How should voice (active, passive, middle, etc.) be implemented
>         in a conlang?

It shouldn't. 

>      7. How should metaphor be dealt with in a conlang?

It shouldn't be dealt with.

These answers are made under the assumption that grammatical simplicity
is highly desirable, and that the invented bit of the conlang is its
grammar.

---
And



>From ucleaar@ucl.ac.uk Wed Oct 20 18:56:36 1993
From: ucleaar 
Message-Id: <37514.9310201756@link-1.ts.bcc.ac.uk>
To: conlang@diku.dk
Subject: Re: Ideas for Discussion
In-Reply-To: (Your message of Wed, 20 Oct 93 17:19:14 N.) <9310201541.AA00710@tyr.diku.dk>
Date: Wed, 20 Oct 93 18:56:05 +0100


Lars says:
> I have an embryonic idea for a language where 2-6 above are eliminated
> by associating each lexical item in a sentence with an explicit ``role
> descriptor.'' Not just case markers, but also things like "nexus verb"
> and "time adverb". There's a couple of other radical ideas in there as
> well.

Are we to hear more of these ideas?

---
And



>From lojbab@access.digex.net Wed Oct 20 19:38:39 1993
From: Logical Language Group 
Message-Id: <199310201837.AA23934@access.digex.net>
Subject: Re: Ideas for Discussion
To: conlang@diku.dk
Date: Wed, 20 Oct 1993 14:37:38 -0400 (EDT)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1354      

And Rosta writes:

> In reply to Rick:
> >      3. Should nouns be classified?  If so, how?
> 
> No.

I agree, if you refer to semantic classification; however, I believe that the
issue still remains open for a more "syntactic" kind of classification, where
nouns are put into anaphora reference classes based on their phonological
or morphological representation.

For example, Lojban/Loglan has 17 grammatical genders, with a gendered
reference pronoun for each, plus 10 non-gendered reference pronouns.
They are the "b-" gender, the "c-" gender, and so on, and the reference
pronouns are the names of the initial letters themselves.  It seems to me that
this system helps to clear up anaphora reference ambiguities without requiring
any extra burden of learning.

> >      4. Should parts-of-speech be marked?  If so, how?
> 
> Perhaps by a certain correspondence of phonological representation
> & syntactic category. You might also ask: what is the minimum
> number of parts of speech necessary, and what is the optimal number.
> (E.g. Lojban has a two or 3 dozen.)

Lojban has 123 parts of speech at present: one kind of proper-name,
one kind of content-word, and 121 kinds of particles.  Many of the
"parts of speech" have only a single word in them.

-- 
John Cowan		sharing account  for now
		e'osai ko sarji la lojban.




>From ucleaar@ucl.ac.uk Wed Oct 20 19:54:49 1993
From: ucleaar 
Message-Id: <28499.9310201854@link-1.ts.bcc.ac.uk>
To: conlang@diku.dk
Subject: Re: Verbs, Nouns and other parts of speech
Date: Wed, 20 Oct 93 19:54:34 +0100


Jo"rg says:
> No other parts of speech are really required -- adjectives, adverbs, 
> adadjectives, pre- and postpositions and other particles may, but need not 
> occur. 

The traditional parts of speech are not necessary, sure. But the 
question is, is it necessary to group lexemes (= dictionary words)
into larger categories on the basis of their shared grammatical
behaviour. 3 possible answers: 
  * the grammatical behaviour of each lexeme is wholly idiosyncratic
  * all lexemes behave the same
  * we need other word classes to capture generalizations about
    shared patterns of grammatical behaviour. (Lojban calls these
    "selmaho".)

So by all means abolish adjectives, adverbs, nouns, prepositions, and
what have you. But don't then conclude that all you're left with is
verbs. (Example: what word class will your word for "every" have,
& what will its grammatical behaviour be like?)
 
> I think a conlang becomes easier, the less parts of speech it contains, 

True, because the grammar gets simpler.

> allthough the way lojban does handle this problem is quite mind-boggling to 
> me. And yes, lojban has verbs and nouns, allthough they are not called this 
> way. 

Well, Lojban has verb-cum-nouns, but not two distinct classes. It has
one class that corresponds to the "lexical" or "content" words of
a lg like English, & lots of other classes that correspond to the
"functional" or "grammar" classes. For example, the Lojban equivalents
of "happy", "dog", "quickly" and "resemble" all belong to the same
class (called brivla, I think), but the equivalents of "a" and "the" 
belong to a different class (called selmaho NU).

---
And



>From KNAPPEN@VKPMZD.kph.Uni-Mainz.DE Thu Oct 21 01:44:20 1993
Date: Thu, 21 Oct 1993 01:43:47 +0200
From: KNAPPEN@VKPMZD.kph.Uni-Mainz.DE
Subject: Re: Verbs, Nouns and other parts of speech
To: conlang@diku.dk
Message-Id: <01H4CUGTC88I91VVBB@MZDMZA.ZDV.UNI-MAINZ.DE>
X-Envelope-To: conlang@diku.dk
X-Vms-To: MZDMZA::IN%"conlang@diku.dk"
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
Content-Transfer-Encoding: 7BIT

Nouns in lojban are in fact quite rare, the best are the aequivalents to 
the classical pronouns like lojban `mi' etc. We can extend the class of 
nouns be words (by the way, what is a word?) which can replace `mi' in a 
given syntagma, this adds proper names like `la pier.' and the `le ...' 
constructions. From a classical point of view one can regard `le' as a 
substantivizing prefix.

Yes, the brivla are verbs classically. And they can serialise, like in thai 
or other south-east asian languages.

On the number of classes: Usually you stop the process of classifying parts 
of speech at some pragmatic point and call the rest `particles' or stuff it 
into some other class. No-one speaks of adadjectives in english or german, 
allthough there are some in these languages, usually booked under adverbs.
I remember from my latin lessons the splitting of conjunctions in 
co-ordinating and subordinating ones, and those are further split for 
semantic reasons -- temporal, causal, local, ... at last only few 
conjuctions were rest in each subclass and some times, a conjunction was 
split into more than 6 homophones belonging to different classes and 
demanding different translations.

Yours, J"org Knappen.



>From EZ-as-pi@cup.portal.com Thu Oct 21 02:45:30 1993
From: EZ-as-pi@cup.portal.com
Subject: Re: Ideas for Discussion
Lines: 88
Date: Wed, 20 Oct 93 18:45:08 PDT
Message-Id: <9310201845.2.26998@cup.portal.com>
X-Origin: The Portal System (TM)

ram@netcom.com (Rick Morneau) writes:

>     1. How should semantic case roles be marked?

I've had varying thoughts on this.
In the Voksigid project, _all_ case
relationships were explicitly marked
with prepositions. This has a number
of advantages, but it does make for a
longer sentence than one sees in other
conlangs. Ihe system of Ido/Novial/
Interlingua/Intal, using word order to
mark the _subject_ and _object_, with
prepositions to mark other relation-
shops, is more compact than Voksigid,
without requiring memorization of vast
numbers of place structures as in Lojban
(for example).
Whether the Voksigid method or the Ido
etc., method is better needs experimen-
tal testing, I think.

>     2. Should the argument structure of verbs be marked?  If so, how?

This really relates to the same thing
as #1 above. Cases and the argument
structure of verbs are intimately
connected; after all, a transitive
verb is one with 2 arguments (subject
and object), a 3-place verb (such as
"give" in English) takes 3 case rela-
tionships (what in Latin are called
nominative, dative, and accusative), 
etc.

>     3. Should nouns be classified?  If so, how?

I think they should at least be 
classified into animate and inanimate.
This is a fairly weak preference,
however.

>     4. Should parts-of-speech be marked?  If so, how?

I think so. I am not sure which of
a number of approaches I prefer, how-
ever. In Novial, we have a refinement
of the vowel-ending system of Esperanto
which I think is one of the few things
Zamenhof did right. However, the Lojban
approach, using C/V patterns and number
of syllables, might work; something
similar was attempted in Voksigid. I
cannot really say one is better than
the other, but I _do_ believe POS mark-
ing is helpful.

>     5. How should tense, aspect and modality be implemented in a conlang?

I used to favor the Esperanto/Ido
suffix-vowel method of expressing tense
but, after looking at the particles 
used in Novial and only slightly 
changed in Intal, I have come to prefer
that system. The original Loglan 
system which allows compounding tense
particles (I do not know whether Lojban
retains it) has some nice regularities.
I have not given as much thought to
aspect and modality, but using parti-
cles for tense suggests using them for
these two categories as well. 

>     6. How should voice (active, passive, middle, etc.) be implemented
>        in a conlang?

This depends on how case relationships
are expressed. If subject and object
are expressed via particles (as in 
Voksigid) then no special passive form
is needed. If not, I would favor the
approach of Novial (as in the case of
tense, having a passive particle).

>    7. How should metaphor be dealt with in a conlang?

I won't try to touch this one.




>From j.guy@trl.oz.au Thu Oct 21 05:01:26 1993
From: j.guy@trl.oz.au (Jacques Guy)
Message-Id: <9310210401.AA23780@medici.trl.OZ.AU>
Subject: Re: Verbs, Nouns and other parts of speech
To: conlang@diku.dk
Date: Thu, 21 Oct 1993 14:01:03 +1000 (EST)
In-Reply-To: <9310201933.3.26998@cup.portal.com> from "EZ-as-pi@cup.portal.com" at Oct 21, 93 04:27:56 am
X-Mailer: ELM [version 2.4 PL20]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1014      

> >No other parts of speech are really required -- adjectives, adverbs, 
> >adadjectives, pre- and postpositions and other particles may, but need not 
> >occur. 
> 
> I can't see how you can dispose of
> pre- and postpositions. 

Many languages dispense almost entirely with pre- and
postpositions. Prepositions and postpositions, in fact, often
derive historically from verbs (eg. Chinese gen1 with (comitative)
< "to follow", Hungarian -ban in < "belly", Lolovuevue lo- in 
< lolo "viscera, innards".

I counted 10 prepositions in the language of my thesis (Sakao).
Of those ten, seven, yes SEVEN, translate our "of" (possession)
or "to" (benefactive), and one is made up of the two remaining
prepositions, of which one is derived from a noun and the other
one probably from a verb. I don't know if there are in fact,
languages without any prepositions (or postpositions). I'm
pretty sure you'd find one in Papua-New Guinea if you looked
hard enough (you find every kind of weird thing there, language-wise).



>From C.J.Fine@bradford.ac.uk Thu Oct 21 10:55:06 1993
Date: Thu, 21 Oct 1993 10:54:57 +0100
Message-Id: <5016.199310210954@discovery.brad.ac.uk>
From: Colin Fine 
To: conlang@diku.dk
Subject: Re: Verbs, Nouns and other parts of speech

EZ-as-pi says:
++++>
 I can't see how you can dispose of
pre- and postpositions.
>++++

It depends what you mean.  In Turkish most postpositions are
actually nouns (exactly comparable to the compound English
'preposition' "in front of"). On the other hand, there is a 
small class of bound suffixes (case endings) which are used
on their own (eg dative -de) or in these postpositional phrases.

	Colin




>From wolfe@gboro.rowan.edu Fri Oct 22 05:21:41 1993
Date: Fri, 22 Oct 93 00:22:13 -0400
From: wolfe@gboro.rowan.edu (Jon Wolfe)
Message-Id: <9310220422.AA04154@gboro.rowan.edu>
To: conlang@diku.dk
Subject: Sci-Fi conlangs
Cc: wolfe@gboro.rowan.edu

Keep an eye out for the British sci-fi sitcom "Red Dwarf".
It is hillarious.  One of the characters is always trying
(but never suceeding) to master certain subjects to improve
himself.  One of his goals is to learn Esperanto (the other
"official" language of the space ship).  Even though it is
set so far in the future that all other possible speakers
are long dead.



>From dasher@netcom.com Fri Oct 22 08:10:18 1993
Date: Fri, 22 Oct 93 00:10:45 -0700
From: dasher@netcom.com (D. Anton Sherwood)
Message-Id: <9310220710.AA26127@netcom4.netcom.com>
To: conlang@diku.dk
Subject: Cornish
Cc: bas+@andrew.cmu.edu (Bruce Arne Sherwood)

My ex is home from two weeks of shopping and theatre-going in London,
bringing me (inter alia) 'A Grammar of Modern Cornish', put out by something
called The Cornish Language Board!  I thought Cornish died with Dorothy
Pentreath in 1777, but evidently it is being revived.

If Cornish was indeed dead, several issues come to mind.  Should the revived
language be based on that of the last speakers, or of a 'classical' period? 
Should dubious forms be resolved in favor of resemblance to Welsh (the
nearest living cousin to Cornish), or distinction from it?

Has English phonology anything to offer to New Cornish?  If New Cornish
is to gain wide use, it should not give English-speaking adults gratuitous
hurdles.  Perhaps, though, old Cornish phonology has left traces in the
English of Cornwall.

Anton Sherwood                                             DASher@netcom.com
+1 415 267 0685         1800 Market St #207, San Francisco, California 94102




>From C.J.Fine@bradford.ac.uk Fri Oct 22 12:45:00 1993
Date: Fri, 22 Oct 1993 12:44:43 +0100
Message-Id: <14005.199310221144@discovery.brad.ac.uk>
From: Colin Fine 
To: conlang@diku.dk
Subject: Re: Cornish

Dolly Pentreath indeed died in 1777 (and there is a memorial
to her in Paul Churchyard, near Penzance, erected some
years later by the local vicar and Prince Louis Bonaparte), 
but it is almost cetain there were still a few cornish speakers
who survived her: being common people (unlike Pentreath 
who had a claim to be considered gentle), they were ignored
by antiquarians.

Nonetheless, die the language did, certainly by the early
years of the last century; but there has been a revival
movement now for many years. One of the leaders of 
this was R Moreton Nance, whose 'Kernewek Sempelhes'
(Simplified Cornish) I have. I believe the revival was 
consciously based on the Middle Cornish of the surviving literature
(mostly miracle plays).

For more information, see the relevant chapter of 
Peter Trudgill's book "Language in the British Isles"


Colin Fine
(who composed this entire message from memory and
hasn't checked any of his facts).



>From EVANS@BINAH.CC.BRANDEIS.EDU Fri Oct 22 14:18:38 1993
Date: Fri, 22 Oct 1993 09:14 EDT
From: Ronald Hale-Evans 
Subject: Re: Sci-Fi conlangs
To: conlang@diku.dk
Message-Id: <01H4EP1RXD8W9FSFNK@BINAH.CC.BRANDEIS.EDU>
X-Vms-To: IN%"conlang@diku.dk"
X-Vms-Cc: EVANS

I have a lead on what looks like a cool SF book. It's supposed to be
about a cabal of mutant geniuses who speak in a conlang very much like
Heinlein's Speedtalk (in the short story "Gulf"; read it if you
haven't -- very good). The person who recommended it on the net said
that it went into a great deal of detail on the language. Here's the
citation as it appeared in the Library of Congress records:

Smith, George Oliver
_Highways in hiding_
New York, Gnome Press, 1955?, 1st ed.
223 p., 21 cm.
LoC #: PZ3.S64753 Hi                       

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * 
"The Sea refuses no river; remember that when a beggar buys a round."
--Pete Townshend * * * * * Ron Hale-Evans, evans@logos.cc.brandeis.edu
PGP 2 public key: finger evans@logos.cc.brandeis.edu



>From donh@netcom.com Fri Oct 22 15:38:05 1993
Message-Id: <9310221438.AA02575@netcom4.netcom.com>
To: conlang@diku.dk
Cc: donh@netcom.com, donh@netcom.com
Subject: Re: Sci-Fi conlangs 
In-Reply-To: Your message of Fri, 22 Oct 93 14:25:09 +0100.
             <01H4EP1RXD8W9FSFNK@BINAH.CC.BRANDEIS.EDU> 
Date: Fri, 22 Oct 93 07:38:27 -0700
From: Don Harlow 

> _Highways in hiding_
> New York, Gnome Press, 1955?, 1st ed.

I've read _Highways in Hiding_ several times (the main characters are 
not mutants, but victims of a disease -- Mekstrom's syndrome -- that, 
if not properly treated, basically turns you to stone, but, if properly 
treated, turns you into superman), both in the Gnome Press edition and 
in the later Lancer incarnation. Unfortunately, I don't remember any 
particular linguistic side-forays. In fact, I don't remember George O. 
_ever_ paying much attention to language in his books.

Don Harlow			donh@netcom.com
Esperanto League (Info only)    (800)828-5944 or elna@netcom.com
Turnig^as la Rado de la Tempo, 
kaj postlasas multajn vojkadavretojn. (Lau^ Robert Jordan)



>From ucleaar@ucl.ac.uk Fri Oct 22 16:24:49 1993
From: ucleaar 
Message-Id: <122808.9310221524@link-1.ts.bcc.ac.uk>
To: conlang@diku.dk
Subject: Re: Verbs, Nouns and other parts of speech
Date: Fri, 22 Oct 93 16:24:36 +0100

Jo"rg says:
> Nouns in lojban are in fact quite rare, the best are the aequivalents to 
> the classical pronouns like lojban `mi' etc. 

Good point.
 
> On the number of classes: Usually you stop the process of classifying parts 
> of speech at some pragmatic point and call the rest `particles' or stuff it 
> into some other class. 

But the reason for stopping is not some grammatical principle, but 
rather laziness, lack of rigour, and general uncertainty on the
part of the grammarian. (So it is admirable that John Cowan 
can say (truly) that there are 123 parts of speech in Lojban.)

----
And





>From KNAPPEN@VKPMZD.kph.Uni-Mainz.DE Fri Oct 22 18:47:32 1993
Date: Fri, 22 Oct 1993 18:45:56 +0200
From: KNAPPEN@VKPMZD.kph.Uni-Mainz.DE
Subject: Re: Verbs, Nouns and other parts of speech
To: conlang@diku.dk
Message-Id: <01H4F8Q2J2HE91VVGI@MZDMZA.ZDV.UNI-MAINZ.DE>
X-Envelope-To: conlang@diku.dk
X-Vms-To: MZDMZA::IN%"conlang@diku.dk"
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
Content-Transfer-Encoding: 7BIT

And,

yes, it is really impressive that Lojban has exactly 123 parts of speech. 
This statement is only possible because the designers of Lojban have taken 
much effort on this part of the language.

However, it is not only sloppinees preventing linguists from doing a 
lojban-like classifacation of parts of speech in a natlang. In a natlang 
usage typically varies with dialect, sociolect, etc. Therefore it is much 
harder to decide to which class a given word belongs. And what about 
examples like "Down with !" ?
Is down a verb (imperative mode) in this utterance ? Or something special 
to create a new class for ? Or still a preposition ?

Unfortunately most conlang designers didn't care about this aspect but 
rather imported a modell for parts of speech from their favorite 
natlang(s).

--J"org Knappen.



>From ucleaar@ucl.ac.uk Fri Oct 22 20:09:42 1993
From: ucleaar 
Message-Id: <91041.9310221909@link-1.ts.bcc.ac.uk>
To: conlang@diku.dk
Subject: Re: Verbs, Nouns and other parts of speech
In-Reply-To: (Your message of Fri, 22 Oct 93 19:28:24 N.) <01H4F8Q2J2HE91VVGI@MZDMZA.ZDV.UNI-MAINZ.DE>
Date: Fri, 22 Oct 93 20:09:24 +0100

> Is down a verb (imperative mode) in this utterance ? Or something special 
> to create a new class for ? Or still a preposition ?

No I didn't mean it's only sloppiness: it's also the inherent difficulty
of grammatical analysis. Working out grammatical categories is a major
component. I'm not aware of any current theories that say how many
word classes any particular language has. I think Government-Binding
puts a limit on non-functional categories, because the categories
are defined by combinations of two binary features [+/-N], [+/-V],
but since in that theory there has been an explosion of functional
categories, that restrictiveness doesn't count for much. 

I would be inclined to treat your "down" as an instance of a special
class. It may also be a preposition, but since most prepositions
don't behave this way (& nor does the imperative of the normal verb
DOWN), we still have to recognize it as a special class.

The peculiarities of the construction have been discussed in an
old article by McCawley in _Studies out in left field_, a collection
of articles from c.1971 (I forget the editors' names). [The
collection became notorious, &, I understand, has recently
been republished, 1991.] McCawley also looks at constructions like
"Fuck you", where the "fuck" superficially looks like an
imperative, but isn't a normal imperative, since if it were
we should get "Fuck yourself" rather than "Fuck you". 

----
And



>From bhelm@majestix.cs.uoregon.edu Fri Oct 22 22:37:37 1993
Message-Id: <9310222137.AA02122@majestix.cs.uoregon.edu>
To: conlang@diku.dk
Subject: Interlinguites, Glosies, et al.
Date: Fri, 22 Oct 93 14:37:04 -0700
From: bhelm@majestix.cs.uoregon.edu

A friend and I are looking into non-Esperanto conlangs
that have speaking populations, and that have a claim to
being suitable for international use.

I know where to find Lojbanists :-)  But recently, I heard
that speakers of Glosa and Interlingua attended a symposium
related (somehow) to the European Parliament on constructed
languages.  I'd be interested in contacting the speakers and
their organizations.  I have a postal address for Glosa, but
nothing on Interlingua.  I don't have electronic addresses
for anyone.

Any help would be appreciated.  Other conlangs that have more
than one living speaker would be interesting as well, provided
they are intended for international use.

Rob



>From urban@cobra.jpl.nasa.gov Fri Oct 22 23:28:55 1993
Message-Id: <9310222228.AA07365@odin.diku.dk>
To: conlang@diku.dk
Subject: Re: Interlinguites, Glosies, et al. 
Date: Fri, 22 Oct 1993 15:28:47 PDT
From: Michael P Urban 

Your message dated: Fri, 22 Oct 1993 23:21:09 PDT
> 
> A friend and I are looking into non-Esperanto conlangs
> that have speaking populations, and that have a claim to
> being suitable for international use.
> 

I have an old address for Ido (mid '80s) at home.  If nobody else
comes up with anything, I will try to dig it out for you.  I imagine
that Bruce Gilson will know something.  

	Mike




>From EZ-as-pi@cup.portal.com Sat Oct 23 02:43:16 1993
To: conlang@diku.dk, ez-as-pi@cup.portal.com
From: EZ-as-pi@cup.portal.com
Subject: Re: Ideas for Discussion
Lines: 39
Date: Fri, 22 Oct 93 18:43:10 PDT
Message-Id: <9310221843.1.19072@cup.portal.com>
X-Origin: The Portal System (TM)

And Rosta (ucleaar ) writes:


>Bruce writes:
>> ucleaar  (And Rosta) writes:
>> >>      4. Should parts-of-speech be marked?  If so, how?
>> >Perhaps by a certain correspondence of phonological representation
>> >& syntactic category. You might also ask: what is the minimum
>> >number of parts of speech necessary, and what is the optimal number.
 
>> Sometimes, different descriptions of
>> the same grammar use different numbers
>> of POS. In English, if you compare
>> Fries' description with a traditional
>> one, you find a vast difference.

>You are quite right. But one description can be better than another,
>and one figure for the number of POS a given lg has can be better
>motivated than another. Yes, the figure is not available from
>cursory inspection, but it is nevertheless (I believe) ultimately
>decidable.

One problem is that there is a question
as to "how different in usage" two 
words need to be to constitute differ-
ent parts of speech.
For example, most treatments of English
recognize a POS called "verb" (or, in
Fries' terminology, "class 2 word").
However, one could easily make the case
that intransitive verbs, transitive
verbs, copulative verbs that can take
noun and adjective complements, copula-
tive verbs that can only take adjective
complements, dative verbs, etc. can all
be considered separate POS. After all,
the difference in how they can occur is
as salient as between the different
sorts of Lojban particles.




>From EZ-as-pi@cup.portal.com Sat Oct 23 03:19:00 1993
To: conlang@diku.dk, ez-as-pi@cup.portal.com
From: EZ-as-pi@cup.portal.com
Subject: Re: Interlinguites, Glosies, et al.
Lines: 23
Date: Fri, 22 Oct 93 19:18:54 PDT
Message-Id: <9310221918.1.20114@cup.portal.com>
X-Origin: The Portal System (TM)

Michael P Urban  writes:

>Your message dated: Fri, 22 Oct 1993 23:21:09 PDT
> 
>> A friend and I are looking into non-Esperanto conlangs
>> that have speaking populations, and that have a claim to
>> being suitable for international use.
> 

>I have an old address for Ido (mid '80s) at home.  If nobody else
>comes up with anything, I will try to dig it out for you.  I imagine
>that Bruce Gilson will know something.  

Unfortunately, I have recently moved
and I'm not sure where to find that; I
do have recent issues of Ido-Vivo 
around, but just can't find them.
However, for Interlingua let me
suggest the British Interlingua Society,
14, Ventnor Court, Sheffield, England
S7 1LB. They supposedly have an office
in this country too, in NYC, but nobody
seems to know the address.



>From 71174.2735@compuserve.com Sat Oct 23 05:48:32 1993
Date: 23 Oct 93 00:42:45 EDT
From: Rick Harrison <71174.2735@compuserve.com>
To: 
Subject: spoken auxlangs
Message-Id: <931023044244_71174.2735_DHQ54-1@CompuServe.COM>

 > From: bhelm@majestix.cs.uoregon.edu
 
> A friend and I are looking into non-Esperanto conlangs
> that have speaking populations, and that have a claim to
> being suitable for international use.
 
Many proposed international languages have a number of people
who can read and write them.  However, I don't know of any hard
evidence indicating that anyone can actually converse in any 
auxiliary languages (auxlangs) other than Esperanto.  It is very 
difficult for an auxlang to make the leap from a paper project to 
a spoken language.  Some auxlangs that were spoken in their "glory 
days" have now fizzled to the point that they are not spoken any 
longer, e.g. Volapu"k, Ido, Occidental.
 
The contact address for Glosa is P O Box 18, Richmond, Surrey,
TW9 2AU, England.
 
=============
Rick Harrison
editor, Journal of Planned Languages (soon to return from hiatus)
Box 54-7014, Orlando FL 32854 USA       71174.2735@compuserve.com
 


>From lojbab@access.digex.net Sat Oct 23 13:23:55 1993
Date: Sat, 23 Oct 1993 08:23:47 -0400
From: Logical Language Group 
Message-Id: <199310231223.AA08459@access.digex.net>
To: conlang@diku.dk
Subject: Re: RE: Ideas for discussion/Purpose of a conlang
Cc: lojbab@access.digex.net

Bruce is correct that nobody, not even I, am fluent in Lojban.  However, it
has been pointed out to me that fluency itself is a relative thing.  I am
fluent in talking to my kids in LeChevalier Russian, a subset of the langauge
that appears to have most of the essential vocabulary that 6 year-olds need
for everyday communication - it is the language that was spoken around here
primarily for the last year - but has many errors in it according to the
standard Russian textbooks, and is not very robust in that I can't say much
to adult Russians due to lack of vocabulry (not to mention the errors that
I keep learning that I need to correct).

I am told that the number of people who are fluent in Esperanto to the extent
that I am fluent in English is probably quite small.  But many people have
an intermediate fluency:  I can carry on substantial conversations with
people who have come to the US from other countries 3 or 4 years ago - I can
tell that they are struggling at times to understand my words or to find the
words that they want, but they do get their ideas across and understand mine.
So a conlang may be usable "as a means of communication" even if no one is
fluent in it.  

And indeed Lojban is so usable, since we have multitudinous times had conver-
sations lasting from an hour to several hours solely in Lojban.

But fluency is the mark necessary to use a language for linguistics work,
and fluent speakers (if not native speakers) are the only ones that the
Linguists will care about.

lojbab



>From ram@netcom.com Sat Oct 23 13:40:39 1993
Date: Sat, 23 Oct 93 05:41:15 -0700
From: ram@netcom.com (Rick Morneau)
Message-Id: <9310231241.AA09870@netcom6.netcom.com>
To: conlang@diku.dk, ram@netcom.com
Subject: Re: Ideas for Discussion

Richard Kennaway writes:
>
> Another issue in conlang design:
>
>	8. How should one choose the concepts for which the language
>	   will contain words?
>

How to divide up concept space is one of the most difficult problems
semanticians have to deal with. Currently, most semantic research limits
itself to relatively small domains of meaning and usage. The big picture
is still unclear, and I doubt if it will clear up anytime soon. :-(

There is a way, though, to help guarantee that the vocabulary of your
conlang will be as neutral, rich, semantically transparent and
semantically precise as possible. What's especially nice about this
approach is that words practically design themselves. Here, very briefly,
is how to do it...

     1. Design a derivational morphology for your conlang
	that is as productive as you can possibly make it.
	This will almost certainly require that you mark
	words for part-of-speech, mark nouns for class, and
	mark verbs for argument structure (i.e., valency and
	case requirements) and voice.

     2. Start with a common verb, noun, adjective, etc. and
	decompose it into its component concepts using the
	above system. For example, the verb "to know" has a
	valency of two, the subject is a semantic patient and
	the object is a semantic theme. (The theme provides a
	focus for the state "knowledgeable". Unfocused, the
	state "knowledgeable" would be closer in meaning to
	the English word "intelligent".)

     3. The root morpheme meaning "knowledgeable/intelligent"
	can now undergo all the morphological derivations
	that are available in the language. Some of these
	derivations will have little or no use. Many, though,
	will be useful. For example, this SINGLE root morpheme
	could undergo derivation to produce the following
	English words: "know", "intelligent", "teach", "study",
	"learn", "review", "instruct", plus words derived from
	these words, such as "student", "intelligence",
	"education", etc. You will also be able	to derive
	words to represent concepts for which English
	requires metaphor or periphrasis, such as "to broaden
	one's mind", "to keep up-to-date", etc. It is important
	to emphasize that ALL of these words can be derived
	from a SINGLE root morpheme.

In other words, use a back door approach - start with a powerful
derivational system, and iteratively decompose words from a natlang
and apply all derivations to the resulting morphemes. In doing so,
many additional useful words will be automatically created, making it
unnecessary to decompose a large fraction of the remaining natlang
vocabulary.

This approach won't guarantee that concept space will be perfectly
subdivided, but it will be as close as you can get. If you know of
a better system, please tell us about it.

Another fairly obvious advantage is that your conlang will be easier
to learn, since you'll be able to create many words from a small number
of basic morphemes. Ad hoc borrowings from natlangs will be minimized.

Also, such a rigorous approach to word design has some interesting
consequences that may not be immediately obvious. If you use this kind
of approach, you'll find that many of the words you create have close
(but not quite exact) counterparts in your native language. However,
this lack of precise overlap is exactly what you ALWAYS experience
whenever you study a different language.

In fact, it is this aspect of vocabulary design that seems to frustrate
so many conlangers, who feel that they must capture all of the subtleties
of their native language. In doing so, they merely end up creating a
clone of the vocabulary of their natlang. The result is inherently
biased, semantically imprecise and difficult to learn for speakers of
other natlangs. It is extremely important to keep in mind that words from
different languages that are essentially equivalent in meaning RARELY
overlap precisely.

Fortunately, all of this does NOT mean that your conlang will lack
subtlety. In fact, with a powerful and semantically precise derivational
morphology, your conlang can capture a great deal of subtlety, and can
go considerably beyond any natural language. The only difference is that,
unlike a natural language, the subtleties will be predictable rather
than idiosyncratic, and the results will be eminently neutral.

So, do you want to create a clone of an existing vocabulary? Or do you
want to maximize the neutrality and ease-of-learning of the vocabulary
of your conlang? You can't have it both ways.

Regards,

Rick


*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
=*   Rick Morneau                ram@netcom.com             =*
*=   Idaho Falls, Idaho, USA    "Make love - not laws!"     *=
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*



>From ucleaar@ucl.ac.uk Sat Oct 23 15:37:37 1993
From: ucleaar 
Message-Id: <118871.9310231437@link-1.ts.bcc.ac.uk>
To: conlang@diku.dk
Subject: Re: Ideas for Discussion
In-Reply-To: (Your message of Sat, 23 Oct 93 03:25:30 N.) <9310221843.1.19072@cup.portal.com>
Date: Sat, 23 Oct 93 15:37:28 +0100

Bruce writes:
> One problem is that there is a question
> as to "how different in usage" two 
> words need to be to constitute differ-
> ent parts of speech.
> For example, most treatments of English
> recognize a POS called "verb" (or, in
> Fries' terminology, "class 2 word").
> However, one could easily make the case
> that intransitive verbs, transitive
> verbs, copulative verbs that can take
> noun and adjective complements, copula-
> tive verbs that can only take adjective
> complements, dative verbs, etc. can all
> be considered separate POS. After all,
> the difference in how they can occur is
> as salient as between the different
> sorts of Lojban particles.

Again, you're quite right that this is a tricky problem.
But again, it's decidable. For example, the category
'transitive verb' is only justified if there are more
generalizations to be made about the category than
"every transitive verb has an object". This is because
"HIT is a transitive verb" is no simpler than 
"HIT has an object". I personally think that such categories
are in fact justified, since further generalizations can
be made about the categories (leading to simplifications
in the rest of the grammar).

----
And



>From KNAPPEN@VKPMZD.kph.Uni-Mainz.DE Sat Oct 23 17:27:04 1993
Date: Sat, 23 Oct 1993 17:26:54 +0200
From: KNAPPEN@VKPMZD.kph.Uni-Mainz.DE
Subject: Re: spoken auxlangs
To: conlang@diku.dk
Message-Id: <01H4GKGME2KY90MYAO@MZDMZA.ZDV.UNI-MAINZ.DE>
X-Envelope-To: conlang@diku.dk
X-Vms-To: MZDMZA::IN%"conlang@diku.dk"
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
Content-Transfer-Encoding: 7BIT

Hello Rick, fine to see you back again on the net!

Allthough my preferance goes to Esperanto, I must admit that there are some 
speakers of Interlingua around who are able to arrange smaller conferences 
in that language. However, I never heard of native speakers of Interlingua 
or any other conlang than Esperanto.


--J"org Knappen.



>From EZ-as-pi@cup.portal.com Sun Oct 24 13:18:45 1993
To: conlang@diku.dk, ez-as-pi@cup.portal.com
From: EZ-as-pi@cup.portal.com
Subject: Re: RE: Ideas for discussion/Purpose of a conlang
Lines: 91
Date: Sun, 24 Oct 93 05:18:43 PDT
Message-Id: <9310240518.2.9142@cup.portal.com>
X-Origin: The Portal System (TM)

Bob LeChevalier  writes:

>I am told that the number of people who are fluent in Esperanto to the extent
>that I am fluent in English is probably quite small.  But many people have
>an intermediate fluency:  I can carry on substantial conversations with
>people who have come to the US from other countries 3 or 4 years ago - I can
>tell that they are struggling at times to understand my words or to find the
>words that they want, but they do get their ideas across and understand mine.

The level of fluency that I would like
to put forth as my criterion would
probably be somewhere in between the
two levels we're talking about. I do
not consider myself fluent in German.
However, if I could improve my German
a bit, to the point that I could engage
in conversation, without resort to a
dictionary and without more than occa-
sionally struggling to come up with a
word (or struggling to recall the
meaning of a word spoken to me) I would
consider it "adequate fluency." I think
I could develop this level of fluency
in Novial if I had people to talk to
in it. I admit that there are probably
this fluent in Esperanto; there were
apparently no people who reached that
level in Volapuek, as that is what 
brought down the Volapuek movement when
they tried to hold a convention in the
language. I don't know whether Bob and
Nora LeC have reached this point -- the
only time I've heard him speak Lojban
was in the presence of enough new 
learners that he was, perhaps very 
deliberately, speaking quite slowly.
I also do not know if anyone but the
LeChevaliers is anywhere near that
point.

>So a conlang may be usable "as a means of communication" even if no one is
>fluent in it.  

Perhaps "usable as a means of communi-
cation" may not be the words I want.
I might amend it to "usable as a non-
emergency means of communication in 
other than a learning situation." If
I need to make myself understandable in
German, or French, or even Spanish, I
will do so. But if I find the person
I am speaking to is better in English
than I am in the other language, I will
certainly switch to English.
In the case of people at a Lojban 
meeting, I suspect that desire to learn
carries things a long way. However, if
the facts come out, I imagine that a
number of times, resort must be made
to English when someone fails to get a
message across. If so, Lojban has 
failed (in that situation, mind you; I
am not claiming that Lojban has failed
in general, but I think the jury is
still out.)


>And indeed Lojban is so usable, since we have multitudinous times had conver-
>sations lasting from an hour to several hours solely in Lojban.

Have they ever ground to a halt because
one or the other had to go look up a 
word? If a conversation of several
hours solely in Lojban has taken place
without such a break to go look some-
thing up, I would accept this as evi-
dence. 

>But fluency is the mark necessary to use a language for linguistics work,
>and fluent speakers (if not native speakers) are the only ones that the
>Linguists will care about.

Actually, linguists are interested in
studying people's acquisition of
second languages and the errors they
make. (I have read such articles.) 
It depends _which_ linguists and
what sub-branch of linguistics.




>From ram@netcom.com Sun Oct 24 14:09:35 1993
Date: Sun, 24 Oct 93 06:10:06 -0700
From: ram@netcom.com (Rick Morneau)
Message-Id: <9310241310.AA28087@netcom6.netcom.com>
To: conlang@diku.dk, ram@netcom.com
Subject: Anaphora


Howdy conlangers!

[I apologize in advance for what follows. It actually started out as
a brief response to And Rosta. Unfortunately, I had a little time on my
hands, and I got carried away. What follows sounds much more like a
tutorial than a "brief response". Feel free to flame me if it makes
you feel good  :-]

Andrew Rosta writes:
>
> Anaphora, or Reference Tracking wd make a good topic for discussion here.
> It is not something nat lg does very well, yet really good alternatives
> (easy to use but unambiguous) are hard to think of. The basic problem
> is:
>   you want to refer to something you've already referred to, & want
>   to communicate to the hearer this identity of referents.
> 
> As Lojbab says, use of gender is a partial solution. I'm sure Rick
> Morneau has ideas on this too.
> 
> ----
> And
>

I don't think that the design of a comprehensive yet simple anaphoric
system is especially difficult. All natural languages have one. However,
as is typical of natural languages, the anaphoric systems are clouded
in idiosyncrasy and irregularity.

One of the problems that many people have is that they tend to think
of anaphora as belonging to a special, closed class of words. In English,
we think of third person pronouns ("he", "she", "it", etc.), demonstratives
("this", "those", etc.), auxiliaries ("be", "have", and "do") and a
handful of oddballs ("herself", "each other", "so", "such", etc.) as
most of the available anaphora. Here are some examples:

	I love anchovy ice cream. Do you?
	(Anaphor: "do")

	William Shakespeare lived in a small town with his pet rock
	and his wife Fifi Yokohama. He would not eat veggies, she
	would not eat vegemite, and IT didn't eat at all.
	(Anaphora: "his", "he", "she" and "IT")

	John said he'll definitely attend the class on Creative
	Suffering. Louise will too.
	(Anaphor: "will")

However, these "closed class anaphora" are not the only ones. Consider
the following:

	1. Ten theoretical physicists and eight sanitary engineers
	attended the seminar. They were constantly heckling them.

Obviously, we can't use the anaphora "they" and "them" in the second
sentence of (1). Instead, we need something like:

	2. The engineers were constantly heckling the physicists.

The point, though, is that the words "engineers" and "physicists" in (2)
are anaphora, and they can continue to be used as such throughout the
remainder of the dialog. Thus, the head word of a phrase is used as a
referent for the entire phrase. I'll call these "open class anaphora".

[For the GB nitpickers: Obviously, I am using the word "anaphor" in a
loose functional sense, rather than in a strict syntactic sense. Whether
an anaphor is legal because node A c-commands node B holds little
interest for me. For your sake, I hope it holds little interest for
you as well. :-]

Sometimes, especially when writing, we define new open class anaphora
explicitly, as in:

	3. This contract is between Timothy TackyTie (henceforth the first
	party) and Wendall WeeWilly (henceforth the second party)...

In (3) the anaphora are explicitly defined as "the first party" and "the
second party". But we can also do it in informal writing and speech:

	4. Ten computational linguists and twelve theoretical linguists
	attended the seminar. The comps were constantly heckling
	the theos. Finally, the theos got so angry that they
	mooned the comps and left.

Another common way to create open class anaphora is to use single
letters or abbreviations:

	5. In discussing the "Best Artificial Language Linguists
	Ever Designed" (BALLED), the designers forgot that there
	were many other lingwackos out there, who were out to get
	BALLED and who would ridicule it at every opportunity.

Of course, once an abbreviation becomes recognizable without introduction,
it will no longer be an anaphor - it will be a proper noun (like USA,
IBM, etc.)

The major difference between the open (O) and closed (C) classes of
anaphora is that the Os tend to keep their referents throughout the
discourse, while the referents of the Cs are constantly changing. Thus,
the anaphor "BALLED" in (5) will refer to the same thing throughout the
dialog, while anaphora such as "he", "do" or "each other" will
continually take on new meanings.

One other thing should be mentioned. Most anaphora are "backward-
referring"; that is, the anaphor refers to something that was mentioned
earlier. It is also possible to have "forward-referring" anaphora, as in:

	6. After ordering a pint of his favorite ale, Robert was
	perplexed when the barmaid replied that the fishmonger
	was next door. The Great English Vowel Shift had begun.

In (6) "his" precedes its referent "Robert".

So, how do you handle anaphora in a conlang? One solution would be to
create a lot of noun and verb classes. But this will not always solve
the problem. You'll often have situations when you want to differentiate
between two or more members of the same class (such as "physicists"
and "engineers").

A better way, in my opinion, is to design the phonotactics and
morphotactics of your conlang to allow the head word of any phrase
to be contracted. The result would always be immediately recognizable
as an anaphor by its form. The contraction could then be used as an
anaphor for the entire phrase from that point on. You could modify
this rule to allow the contraction to take on a new meaning if its
pattern matches the head word of a new phrase. Here's how something
like this might sound in English:

	The Sheboygan Bandits and the Milwaukee Dragoons
	faced off at Lovemud Stadium on Sunday. The Mil'goons
	beat the She'its out of their expected title.

Unfortunately, English is not really suited for this. A conlang,
however, can be designed to allow such an anaphoric system, and there
are many ways to do it. Here's one possible approach:

	Let open class words have the following form:

		stem + classifier + part-of-speech

	where

		stem = [CV]		[] = 1 or more of the enclosed item
		classifier = CC		C = consonant
		part-of-speech = V	V = vowel ('o'=noun, 'e'=verb, etc.)

	Thus, examples of open class words would be "mande", "kitusta",
	"jonabefti", etc. (You would probably want to exercise some
	restraint in your choice of legal consonant clusters to make
	pronunciation as easy as possible.)

	An anaphor can be created from the head word of any phrase by
	replacing the classifier and as many preceding syllables of the
	stem as needed by a glottal stop, represented here by an
	apostrophe. Thus, the following noun phrase:

		Timanodendo janasuski tupya
		engineer    sanitary  ten
		"Ten sanitary engineers"

	could be abbreviated to "ti'o", "tima'o" or "timano'o".

	And the following verb phrase:

		Jujushimpe makitundo  bubuski
		heckle     fishmonger illiterate
		"To heckle illiterate fishmongers"

	could be abbreviated to "ju'e", "juju'e" or "jujushi'e".

Anyway, that should give you the idea. And there are many, many
other ways of accomplishing the same thing. Have fun!

Regards,

Rick


*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
=*   Rick Morneau                 ram@netcom.com            =*
*=   Idaho Falls, Idaho, USA      "Make love - not laws!"   *=
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*




>From ucleaar@ucl.ac.uk Mon Oct 25 16:42:04 1993
From: ucleaar 
Message-Id: <48868.9310251541@link-1.ts.bcc.ac.uk>
To: conlang@diku.dk
Subject: Re: Anaphora
Date: Mon, 25 Oct 93 15:41:24 +0000

Rick Morneau does indeed have ideas about how to handle anaphora
(i.e. reference to something already referred to in the previous
discourse). Essentially his suggestion is that to re-refer X,
you abbreviate the linguistic expression first used to refer to X.
Example:

> 		Timanodendo janasuski tupya
> 		engineer    sanitary  ten
> 		"Ten sanitary engineers"
> 
> 	could be abbreviated to "ti'o", "tima'o" or "timano'o".

Like Jacques, I think this is a nice idea (Lojban uses it too, 
in a less elaborate way). But like all solutions to anaphora, 
it only works partially. The main problem in this case is contexts 
like:

   A dog was attracted to a dog. But its owner kept it away from it.

Two dogs, here, and the same word ("dog") is used to initially
refer to each. I guess Rick would have:

   A dog was attracted to a dog. But dogs owner kept dog away from dog.

Which is not very helpful.

-----
And




>From j.guy@trl.oz.au Tue Oct 26 01:38:48 1993
From: j.guy@trl.oz.au (Jacques Guy)
Message-Id: <9310260038.AA07243@medici.trl.OZ.AU>
Subject: Re: Anaphora
To: conlang@diku.dk
Date: Tue, 26 Oct 1993 10:38:37 +1000 (EST)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 461       

> 
>    A dog was attracted to a dog. But its owner kept it away from it.
> 
> Two dogs, here, and the same word ("dog") is used to initially
> refer to each. I guess Rick would have:
> 
>    A dog was attracted to a dog. But dogs owner kept dog away from dog.
> 
> Which is not very helpful.
> 
(I should have made the subject line "puppy love", no?)

What about: But a[ttracted]'s owner kept a[ttracted] away from do[g]?

When there's a will there's a way...



>From ram@netcom.com Wed Oct 27 12:42:12 1993
Date: Wed, 27 Oct 93 04:42:45 -0700
From: ram@netcom.com (Rick Morneau)
Message-Id: <9310271142.AA15538@netcom6.netcom.com>
To: conlang@diku.dk, ram@netcom.com
Subject: Re: Anaphora


Howdy conlangers!

And Rosta takes me to task because my proposed anaphoric system cannot
deal with the following:
>
>   A dog was attracted to a dog. But its owner kept it away from it.
>

I agree, but I don't understand why you would WANT your anaphoric
system to be able to deal with this kind of situation. It will only be
used when the speaker is being humorous or intentionally ambiguous.
As far as I'm concerned, if the speaker wants to have fun, then let him!
Besides, you could always distinguish between "the first dog" and "the
second dog", or "the former" and "the latter". In my opinion, this is
a non-problem, and I see no reason to waste time on it.

There may be cases, however, in which distinct phrases share a common
head word. My proposal did not deal with these, and so I will now post
a revised system.

To refresh your memory, here again is one possible basic structure for
open class words:

	stem + classifier + part-of-speech
	[CV]       CC            V

where C = consonant, V = vowel, and [] indicates 1 or more of the
enclosed item.

There should be two types of anaphor: simple anaphora and compound
anaphora. A simple anaphor will be formed from one or more initial
syllables of the head word followed by a glottal stop followed by
the final part-of-speech vowel. Thus, simple anaphora will have the
structure:

	[CV]'V

where both parts of the anaphor are taken from the head word. This
is exactly the same as what I posted earlier.

Compound anaphora will be formed from one or more initial syllables
of a significant MODIFIER of the head word, followed by a glottal
stop followed by the final VCCV of the head word. Thus, compound
anaphora will have the structure:

	[CV]'VCCV

where the first part of the anaphor comes from a significant
modifier of the head word, and the second part of the anaphor comes
from the head word itself. (Note that both forms are consistent with
a self-segregating morphology.)

Thus, my sample noun phrase:

	Timanodendo janasuski tupya
	engineer    sanitary  ten
	"Ten sanitary engineers"

could have the simple anaphora "ti'o", "tima'o" or "timano'o", or the
compound anaphora "ja'endo", "jana'endo" or "janasu'endo".

We can now deal with a more reasonable version of And's sentence, such
as:

	A big dog was attracted to a little dog. But its owner kept
	it away from it.

Using compound anaphora, one possible permutation would be:

	A big dog was attracted to a little dog. But li'og's owner
	kept bi'og away from li'og.

Finally, as I mentioned in my previous post, this approach will work
best if the phonotactics and morphotactics of your conlang are
designed for it. Unfortunately, if you are designing a yachecle
(Yet Another CHauvinistic Euro-CLonE :-), the above system may not
work very well for you. :-(

Regards,

Rick


*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
=*   Rick Morneau                 ram@netcom.com            =*
*=   Idaho Falls, Idaho, USA      "Make love - not laws!"   *=
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*



>From ucleaar@ucl.ac.uk Wed Oct 27 19:52:38 1993
From: ucleaar 
Message-Id: <83635.9310271852@link-1.ts.bcc.ac.uk>
To: conlang@diku.dk
Subject: Re: Anaphora
In-Reply-To: (Your message of Wed, 27 Oct 93 13:19:20 N.) <9310271142.AA15538@netcom6.netcom.com>
Date: Wed, 27 Oct 93 18:52:10 +0000


Rick writes:
> And Rosta takes me to task 

(I only pointed out that it was a nice but not comprehensive proposal.)

> because my proposed anaphoric system cannot
> deal with the following:
> >
> >   A dog was attracted to a dog. But its owner kept it away from it.
> 
> I agree, but I don't understand why you would WANT your anaphoric
> system to be able to deal with this kind of situation. It will only be
> used when the speaker is being humorous or intentionally ambiguous.
> As far as I'm concerned, if the speaker wants to have fun, then let him!
> Besides, you could always distinguish between "the first dog" and "the
> second dog", or "the former" and "the latter". In my opinion, this is
> a non-problem, and I see no reason to waste time on it.

I am mystified. What is strange about wanting to talk about two
dogs, or two apples or two people, or whatever? It is a common
problem. One solution is indeed to use "the former" and "the
latter", and according to my memory of an article by Comrie,
whose details I forget, it is a very common method of anaphora
in narrative. 

Perhaps you misunderstood me and thought that I was suggesting
that you should be able to ambiguously say "its owner kept
it away from it". I meant this as an example of something English
does badly & a conlang shd do better.
 
> Using compound anaphora, one possible permutation would be:
> 
> 	A big dog was attracted to a little dog. But li'og's owner
> 	kept bi'og away from li'og.

This would work so long as you remember to describe each dog
(or whatever) using a different phrase, e.g.

        Two identical twins were chatting. One twin had just
        returned from Tashkent. Tashke'in smiled at not-tashke'in.

---
And



>From ram@netcom.com Fri Oct 29 12:09:31 1993
Date: Fri, 29 Oct 93 04:09:38 -0700
From: ram@netcom.com (Rick Morneau)
Message-Id: <9310291109.AA23560@netcom6.netcom.com>
To: conlang@diku.dk, ram@netcom.com
Subject: Re: Anaphora

Howdy conlangers!

And Rosta writes:
>
> Perhaps you misunderstood me and thought that I was suggesting
> that you should be able to ambiguously say "its owner kept
> it away from it". I meant this as an example of something English
> does badly & a conlang shd do better.
>

I strongly feel that a properly designed anaphoric system should be able
to provide an unambiguous index to any referent.  The system I proposed
does this very well.  Furthermore, if the referent is ambiguous, then
the anaphor should also be ambiguous.  In other words, the anaphoric
system should not be given the additional duty of disambiguating an
ambiguous phrase.  Disambiguation should be handled explicitly by the
speaker.  Thus, "a dog and a dog" is intentionally ambiguous (in
addition to being unnatural).  I do not feel that the anaphoric system
should be required to resolve an intentional ambiguity.

In the case of "two identical twins", you've only provided one referent,
and the system I proposed can deal with it very well.  The referent is
"two identical twins", and one possible anaphor would be "id'ins".  Now,
if you are saying that the anaphoric system must also provide an
unambiguous index to EACH of the twins, then you are saying that the
anaphoric system must provide an index to a referent that has not even
been mentioned.  If neither of the twins has been mentioned separately,
then you do not have a referent, and I see no reason to provide an index
to a non-existent referent.

What you seem to want is an anaphoric system that can also provide
semantic decomposition.  I do not feel that this should be the purpose
of an anaphoric system, even though it is occasionally possible in
natural languages.  Considering the many, many possible kinds of
groupings (twins, clubs, choirs, companies, orchards, boxes of spare
parts, etc), such a system would be very complex, and I'm not even sure
it would be possible.

In summary, I feel that an anaphoric system should be rich enough to
provide an unambiguous index to any unambiguous referent.  Such a system
should NOT have the additional duties of disambiguation or semantic
decomposition.  I suspect that our differences are essentially
philosophical.

Regards,

Rick


*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
=*   Rick Morneau                 ram@netcom.com            =*
*=   Idaho Falls, Idaho, USA      "Make love - not laws!"   *=
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*



>From C.J.Fine@bradford.ac.uk Fri Oct 29 12:41:43 1993
Date: Fri, 29 Oct 1993 11:41:30 GMT
Message-Id: <26097.199310291141@discovery.brad.ac.uk>
From: Colin Fine 
To: conlang@diku.dk
Subject: Re: Anaphora

Rick says to And:
++++++++>
 I strongly feel that a properly designed anaphoric system should be able
to provide an unambiguous index to any referent.  The system I proposed
does this very well.  Furthermore, if the referent is ambiguous, then
the anaphor should also be ambiguous.  In other words, the anaphoric
system should not be given the additional duty of disambiguating an
ambiguous phrase.  Disambiguation should be handled explicitly by the
speaker.  Thus, "a dog and a dog" is intentionally ambiguous (in
addition to being unnatural).  I do not feel that the anaphoric system
should be required to resolve an intentional ambiguity.
>++++++

I think you are confusing translation with expression. And's point was that 
the text is rather ambiguous in English, but I understood him to be
saying that a conlang should resolve the ambiguity, not perpetuate
it. "a dog and a dog" is neither ambiguous nor necessarily unnatural
(it is unusual in English, certainly, but we do not have a monopoly on
ways to express things).

+++++++
 In the case of "two identical twins", you've only provided one referent,
and the system I proposed can deal with it very well.  The referent is
"two identical twins", and one possible anaphor would be "id'ins".  Now,
if you are saying that the anaphoric system must also provide an
unambiguous index to EACH of the twins, then you are saying that the
anaphoric system must provide an index to a referent that has not even
been mentioned.  If neither of the twins has been mentioned separately,
then you do not have a referent, and I see no reason to provide an index
to a non-existent referent.
>++++++++
This is a contentious claim. I accept that it is possible to construe 'two
identical twins' as one referent - but I suggest that it can equally be 
construed as two.

	Colin



>From ucleaar@ucl.ac.uk Fri Oct 29 19:16:52 1993
From: ucleaar 
Message-Id: <50291.9310291816@link-1.ts.bcc.ac.uk>
To: conlang@diku.dk
Subject: Re: Anaphora
In-Reply-To: (Your message of Fri, 29 Oct 93 12:19:21 N.) <9310291109.AA23560@netcom6.netcom.com>
Date: Fri, 29 Oct 93 18:16:18 +0000


Rick Morneau writes:

> I strongly feel that a properly designed anaphoric system should be able
> to provide an unambiguous index to any referent.  The system I proposed
> does this very well.  Furthermore, if the referent is ambiguous, then
> the anaphor should also be ambiguous.  In other words, the anaphoric
> system should not be given the additional duty of disambiguating an
> ambiguous phrase.  Disambiguation should be handled explicitly by the
> speaker.  Thus, "a dog and a dog" is intentionally ambiguous (in
> addition to being unnatural).  I do not feel that the anaphoric system
> should be required to resolve an intentional ambiguity.

We are at cross purposes. My example was "A dog was attracted to a
dog". This refers to two dogs, but uses the same words ("a dog") 
to refer to each. All I want is a way to refer unambiguously in
the subsequent discourse to the first dog, the fancier, or the
second dog, the fancyee. This is the only point I've been trying
to get across. I entirely agree with you about what a properly
designed anaphoric system should be able to do.

----
And

    Source: geocities.com/raiu_harrison/conlang

               ( geocities.com/raiu_harrison)