me lu ju'i lobypli li'u 15 moi

From Lojban
Jump to: navigation, search
                                 
                                 
                                 
                                 
                                 
                                 
                                 
                                 
                                 
                 Number 15 - August-September 1991
         Copyright 1991,  The Logical Language Group, Inc.
        2904 Beau Lane, Fairfax VA 22031 USA (703)385-0273
 Permission granted to copy, without charge to recipient, when for
              purpose of promotion of Loglan/Lojban.
                                 

                First International Correspondence

                 JL to Become Subscription Journal

                         Lojban List Moves

                     Details Inside, and More.

     ju'i lobypli (JL) is the quarterly journal of The Logical
Language Group, Inc., known in these pages as la lojbangirz.  la
lojbangirz. is a non-profit organization formed for the purpose of
completing and spreading the logical human language "Lojban - A
Realization of Loglan" (commonly called "Lojban"), and informing
the community about logical languages in general.
     For purposes of terminology, "Lojban" refers to a specific
version of a logical human language, the generic language and
associated research project having been called "Loglan" since its
invention by Dr. James Cooke Brown in 1954.  Statements referring
to "Loglan/Lojban" refer to both the generic language and to Lojban
as a specific instance of that language.  The Lojban version of
Loglan was created as an alternative because Dr. Brown and his
organization claims copyright on everything in his version,
including each individual word of the vocabulary.  The Lojban
vocabulary and grammar and all language definition materials, by
contrast, are public domain.  Anyone may freely use Lojban for any
purpose without permission or royalty.  la lojbangirz. believes
that such free usage is a necessary condition for an engineered
language like Loglan/Lojban to become a true human language, and to
succeed in the various goals that have been proposed for its use.
     la lojbangirz. is a non-profit organization under Section
501(c)(3) of the U.S. Internal Revenue Code.  Your donations (not
contributions to your voluntary balance) are tax-deductible on U.S.
and most state income taxes.  Donors are notified at the end of
each year of their total deductible donations.
     Page count this issue: 88+2 enclosures ($9.00 North America,
$10.80 elsewhere).  Press run for this issue of ju'i lobypli: 275.
We now have about 620 people on our active mailing list, and 240
more awaiting textbook publication.

                        Your Mailing Label

Your mailing label reports your current mailing status, and your
current voluntary balance including this issue.  Please notify us
of changes in your activity/interest level.  Balances reflect
                                 2
contributions received thru 31 August 1991.  Mailing codes (and
approximate balance needs) are:

Activity/Interest Level:                        Highest Package
Received (Price Each)                           Other flags:
B - Observer     0 - Introductory Materials ($5)  JL JL
Subscription ($25/yr)
C - Active Supporter                            1 - Word Lists and
Language Description ($15)                      LK LK Subscription
($5/yr)
D - Lojban Student                              2 - Language Design
Information ($10)                               R  Review Copy (no
charge)
E - Lojban Practitioner                         3 - Draft Teaching
Materials ($30)  UP Automatic Updates (>$20)

Please keep us informed of changes in your mailing address, and US
  subscribers are asked to provide ZIP+4 codes whenever you know
                               them.

                                 3
                      Contents of This Issue

     Important:  Due to financial constraints, ju'i lobypli will be
fully converting to a subscription basis over the next few issues.
Be sure to read the financial news section if you wish to keep
receiving ju'i lobypli.
     We got a lot accomplished in the months leading up to LogFest,
and several major decisions were made at that meeting.  See the
news section.
     la lojbangirz. has made its first research proposal, to the U.
S. defense agency DARPA, and also has attended its first
linguistics conference.  See the news section.
     This issue contains a lot of material derived from the Lojban
List computer mailing list on the Internet.  Nearly all such
material has been edited, revised, and corrected from the original.
Included are discussions of grammar points, some more on Lojban and
linguistics, and a LOT of Lojban text.  I have Lojban text material
from over a dozen people to choose from for this issue, and it is
tough to choose.  Some will be saved for JL16.

                         Table of Contents

News
  Finances                                                    ---3
  Logfest 91                                                  ---5
  Lojban List Moves / Electronic Distribution Policy          ---7
  Language Development Activities                             ---8
  Using the Language                                          --10
  Research and Linguistics                                    --11
  Products Status, Prices, and Ordering                       --12
  International News                                          --15
  Publicity; News From the Institute                          --16
  New Loglans                                                 --17
le lojbo se ciska                         --18, 26, 42, 47, 57, 65
Is Lojban Scientifically Interesting?                         --20
Summary of gismu/rafsi Official Changes                       --23
Cleft Place Structures and sumti-Raising                      --32
Versions of the Theory of Linguistic Relativity               --42
On Loglan and Lojban Elidables                                --47
A History and Description of le'avla in Loglan and Lojban     --50
The Culture gismu Revisited:  Cultural Neutrality and the gismu
List                                                          --53
Grammar Notes:  On Observatives; Predications and Identities  --61
How to Say It - A New Regular? Feature                        --63
Translations of le lojbo se ciska                             --79

                     Computer Net Information

     Via Usenet/UUCP/Internet, you can send messages and text files
(including things for JL publication) to la lojbangirz./Bob at:
lojbab@grebyn.com    (This is a new address and supersedes the
prior "snark" address.)
     You can also join the Lojban List mailing list (currently
around 80 subscribers).  Send a single line message (automatically
processed) containing only:
"subscribe lojban yourfirstname yourlastname"
to:
listserv@cuvmb.cc.columbia.edu
If you have problems needing human intervention, send to:

                                 4
                                        lojban-list-
request@snark.thyrsus.com
Send traffic for the mailing list to:
lojban@cuvmb.cc.columbia.edu
     Please keep us informed if your network mailing address
changes.
     Compuserve subscribers can also participate.  Precede any of
the above addresses with INTERNET:  and use your normal Compuserve
mail facility.  If you want to participate on Lojban List, you
should be prepared to read your mail at least every couple of days;
otherwise your mailbox fills up and you are dropped from the
mailing-list.  FIDOnet subscribers can also participate, although
the connection is not especially robust.  Write to us for details
if you don't know how to access the Internet network.
     Whether you wish to participate in the news-group or not, it
is useful for us to know your Compuserve or Usenet/Internet
address.

                                 5
  We've been requested to more   chances - we have to pay the
explicitly identify people who   bills.
are referred to by initials in     We've found that a high per-
JL, and will regularly do so in  centage of people specifically
this spot, immediately before theordering material from us
news section.  Note that         contribute money to pay for it.
'Athelstan' is that person's realIt is ju'i lobypli and le lojbo
name, used in his public life,   karni, which we send people
and is not a pseudonym.          without a specific prepaid order,
                                 that people do not contribute
  'pc' - Dr. John Parks-Clifford,enough to cover.  Our financial
Professor of Logic and Philosophytracking system finally improved
at the University of Missouri -  to the point where we could
St. Louis and Vice-President of  identify this situation.
la lojbangirz.; he is usually      Yet ju'i lobypli is what people
addressed as 'pc' by the         provide most feedback on, a
community.                       product that people clearly like.
  'Bob', 'lojbab' - Bob            The answer, it seems, is to put
LeChevalier - President of la    JL on a prepaid, specific order
lojbangirz., and editor of ju'i  basis.  Then, presumably, those
lobypli and le lojbo karni.      of you who want JL will tell us
  'Nora' - Nora LeChevalier -    so with your checkbooks and
Secretary/Treasurer of la        credit cards.
lojbangirz., Bob's wife, author    This solution engenders its own
of LogFlash.                     new problems.  We can presume
  'JCB', 'Dr. Brown'- Dr. James  that not every JL subscriber will
Cooke Brown, inventor of the     subscribe if they have to pay for
language, and founder of the     it in full and in advance.  But
Loglan project.                  if we drop significantly below
  'The Institute', 'TLI' - The   200 U.S. addressees, we lose our
Loglan Institute, Inc., JCB's    reduced 3rd class bulk rates in
organization for spreading his   postage.  This reduction amounts
version of Loglan, which we call to about $2 per copy, or $400 per
'Institute Loglan'.              issue.  So, if by reducing our
  'Loglan' - This refers to the  subscriber list for JL does not
generic language or language     save us at least $400, we are
project, of which 'Lojban' is themerely serving less people for
most successful version, and     the same amount of money.  We
Institute Loglan another.        thus find that there is a gap
'Loglan/Lojban' is used in       between about 140 and 200 U.S.
discussions about Lojban where wesubscribers where we lose as much
wish to make it particularly     or more money than when we send
clear that the statement applies to additional people who are not
to the generic language as well. paying.  We suspect that going to
                                 a prepaid subscription basis will
              News               put us in the middle of that
                                 interval.
            Finances               Going to a fully paid basis
                                 also makes it more difficult for
  We may have gotten momentarily students, people out-of-work, and
overconfident in JL14, after     low-income Lojbanists to get JL.
raising a nice amount of money   Yet these subscribers are among
with our fund-raising letter lastour most productive volunteers,
fall.  Unfortunately, since that and have been more likely to
fund-raiser, income has been     spend the time and effort to
lower than the already depressed start learning Lojban.  Non-U.S
levels before the letter.  We aresubscribers are also hurt, having
hoping that this is due only to  a higher price to pay, but often
the recession, but cannot take   having a lower income because of
                                 their country's economy.

                                 6
  Finally, reducing our          the hoops and see whether we
subscriber list reduces our      could do so if it proves finan-
outreach - our ability to attractcially necessary.
new people and get them involved 
in learning and using Lojban.      Thus, at LogFest 91, we decided
People who buy our products oftenon the following steps:
learn about them through seeing  
how others use them productively   1. JL will be converted to a
in JL.                           prepaid subscription basis over a
  Fortunately, there is one      period of around a year.  If this
option that may eliminate the    means that we lose bulk rate, so
bubble.  In going to prepaid sub-be it.  Price will be $20-
scriptions, we may be able to    $25/year, payable in advance.
become a 'legitimate' periodical People with negative balances
qualifying for U.S. Second Class will be cut off (switched to le
(Periodical) postage rates.      lojbo karni), unless supported
Second Class doesn't require the either by volunteer credits (see
200 minimum mailing that our     below) or by direct donation by
current bulk rate permit does,   another person.
has even lower rates per piece,    2. The first step will be a
and offers faster and more       fund-raiser and direct-mail
certain delivery than bulk rate  announcement of the new policy in
mailings.                        the next month or two.  Every
  However, to establish          subscriber to JL will be sent a
legitimacy, we have to prove thatform to be signed and returned
our readers WANT to receive our  indicating that you want to
publication.  We can prove this  receive JL, and a signature line
either with formal audit         will be added to our order form.
procedures (which we cannot      If not signed and returned and
afford), or through having on    you have a negative balance, you
file explicit requests from each will be dropped as a JL
of our subscribers.  The latter  subscriber, but will receive LK
must be signed and dated, or we  instead.  If you have a positive
must have other proof that the   balance, we still need you to re-
request is bona fide (such as    turn the form to qualify for 2nd
electronic mail headers and      class mailing.
addresses).  The postal service    3. Thereafter, the negative
will audit us at least once a    balance cutoff for JL subscribers
year, and they check carefully.  will be raised each issue, and
  A side benefit/penalty         people not making the cut will be
(depending on whether you are thedropped to an LK subscription.
reader or the editor) is that 2ndWe will give people a one issue
class periodicals MUST be        advance notice of cutoff.  For
published regularly, and at leastthose with very negative
quarterly, so that JL would be   balances, you will be able to
coming out every 3 months with noavoid the cutoff by explicitly
slips like we've been making     subscribing and sending a signed,
lately.                          paid order for JL.
  A final factor is that it costs  There will be no exceptions.
$275 just to apply for a 2nd     Some of you with very negative
class permit, so we must have allbalances may wish to decide what
of our procedures in place BEFOREyou want your status to be, and
we apply.                        possibly to negotiate with
  We haven't decided for sure to someone or with us to continue to
go to 2nd class mailing - the    receive issues.  If you have done
rigor may be more than we can    things for us, including active
handle with one full-time worker,participation on Lojban List such
me, who has other things to do   that we use your material in JL,
besides publish JL.  But we are  you can possibly negotiate
going to start jumping through   delayed payment or a partial

                                 7
amount to be paid to zero out    should contact to request such
your balance.  We ask however    assistance in your company, to
that for those who can afford to,let us know.  We will also be
you pay most or all of your      directly seeking out ideas and
balances off so that we can help information from a couple of you
out others who cannot.           whom people have recommended that
  We do not intend to drop LK    we specifically ask.
subscribers until the books are    We are seeking donations,
done, except upon request.  It   probably in the $1000-$15000
isn't feasible to put LK on a    range, to support specific or
subscription basis, because the  general research projects in
response rate to our mailings is Lojban applications, and also to
so low.                          support publication of the
  4. Given the cutback, we hope  textbook and dictionary in
that our financial condition     amounts large enough to keep the
improves to the point that we    price down and allow wide
have a surplus.  If so, the      distribution.  Specifically from
following plan will aid the ones companies that manufacture and
who cannot pay for subscriptions sell computers, we also are
and other materials.             seeking unrestricted donations of
  A 'volunteer credit' donation  one or two small machines.  Unre-
fund will be set up.  People who stricted donation means that we
donate can specify donations for could use or sell the machine -
general expenses, or specificallyselling it to get money for
for this fund.  In addition, a   support or using it for research
specific portion of any excess   purposes.  Two machines would
revenues (profits) will be put inallow us to sell one and keep
this fund.                       one.  Donation of machines to la
  A committee will accept        lojbangirz. apparently benefits
recommendations of people who    such companies more than direct
have contributed in a wide       cash donations.  Again, ideas are
variety of ways from commentary  welcome in this area.
on JL, learning the language,      One such donation will greatly
participation on Lojban List,    ease our month to month financial
recruiting, overseas activities. pressure.  A larger donation or
They will also get a list from memore smaller ones would allow us
each issue of people whose       to make intelligent financial
balance is less than the         decisions on how to complete our
subscription cutoff, along with  projects and to get serious
notes on any special             research started, without the
circumstances that might allow   distorting effect of living hand-
them to be retained as JL sub-   to-mouth.  Please help if you
scribers.  The committee will    can.
allocate the funds among the pos-  6.  We plan to establish a
sible recipients, so as to allow 'Sustaining Membership' similar
the maximum number to be retainedto other non-profit organiza-
as JL subscribers.               tions.  Probably costing
  5. We will seek direct         $50/year, the benefits will be
donations of larger amounts of   minimal - perhaps acknowledgement
money from companies, especially in our books, periodicals, and
from computer companies who mightour annual reports, perhaps a 10%
profit by the positive image of  or 20% discount on purchases, and
supporting non-profit scientific higher priority on orders and
and educational research with    services.  The main 'benefit'
computer applications.  We are   will be knowing you are helping
asking ALL subscribers associatedmake Lojban a success.  Details
with a company who might be      will be announced.
willing to help support us, and    7. Finally, we have gotten a
who either have some influence inlocal computer network account
such decisions, or know who we

                                 8
which will significantly cut la    On Saturday, after a slow start
lojbangirz.'s phone bill.        due to late sleepers, we started
                                 doing 'serious' Lojban.  We had
  We believe these steps will be prepared for a couple of dozen
more than sufficient to right ourdifferent kinds of activities, so
tottering finances.  We've made aas to be ready for a range of
lot of progress so far, but as weLojban experience and interests.
continue to rapidly grow, it is  This year, attendees were almost
easy to lose control.  la        all active students who knew
lojbangirz. is now far larger    enough vocabulary and grammar for
than I can financially support byus to undertake intermediate
myself.                          activities.
  As a business, we need a safety  One activity that proved
margin so that financial crisis  moderately successful was
is not always knocking at the    translating aphorisms.  People
door.  And if we have to worry   seem much more comfortable trying
less about finances, that means  to translate single sentences
all the more effort that can be  both from English to Lojban and
put towards writing books and    vice versa, than with longer
software and otherwise making    texts.  Thus, every participant
sure Lojban continues to grow.   got a random aphorism out of a
                                 box (we pregraded the aphorisms
           LogFest 91            by grammatical difficulty, so
                                 people chose a line they had a
  Logfest 91, the annual         reasonable chance to translate),
gathering for celebration of     and worked on a translation to
Lojban, started Thursday night,  Lojban.  More experienced
June 20, with the arrival of the Lojbanists aided the less skilled
first three visitors, even thoughones.  Then each person presented
no organized activities (other   her/his translation to the group
than getting ready) were         as a whole, who then tried to
scheduled for Friday.  As happensfigure out what it meant.  In
when a good group of Lojbanists  general, everyone successfully
gets together, Friday was filled understood others' translations,
with a variety of lively and     using their word lists.
interesting discussions (not       A weakness of the activity was
limited to Lojban).  As people   the size of the group.  With over
arrived, the discussions got     a dozen participants, it took a
livelier, and a bit more serious.long time to go through all
  On Friday night, we turned to  translations.  We know next time
discussion of the financial      we have that many people to di-
situation, and a related matter -vide into groups, so that things
the distribution of Lojban       move quicker.  Still, everyone
materials electronically (via thelearned a lot, and many were sur-
computer networks).  Such distri-prised at how easily and well
bution helps our costs by        they could understand the trans-
reducing postage, and offers the lations.  You can try the
potential of more rapidly        activity yourself - aphorisms in
expanding the Lojban community,  both English and Lojban will be
but with a likely loss of income found in le lojbo se ciska this
since many people who receive    issue.
materials electronically will not  Less intense was a discussion
contribute to the costs of those on making tanru and lujvo.  We've
materials.                       tried this before, but working at
  The discussion ran all night,  the level of individual words
and was heated at times.  The    gets people bogged down in the
result, though, was a workable   semantics of English.  In this
policy that attendees were       case, working on lujvo for the
satisfied with.  This new policy English word "tyranny", we ended
is discussed below.              up with over a dozen tanru, each

                                 9
with its own subtle distinction  finances will improve to the
in meaning, and no real agreementpoint that we no longer have to
on a 'best' one.  My own opinion spend hours debating new
is that there is no 'best' lujvo strategies.
for any given English concept,     With such a long meeting, non-
because you will choose a        voting Lojbanists tended to drift
different emphasis depending on  in and out of the meeting into a
the context.  This exercise,     variety of discussions and
always educational but always    informal activities.  By the end
somewhat of a failure, reminds usof the meeting, a lively game of
that Lojban and English are very "la reno preti" (20 Questions)
different languages.             was being played, entirely in
  There were other activities on Lojban.  This proved to be the
Saturday, but the primary focus  most successful of the Lojban
outside of the above activities  activities, continuing well-into
was group discussion and         the evening.
socializing.  Art Protin and       By Monday, only 3 Lojbanists
David Twery, visiting from New   were left.  Two stayed until mid-
Jersey and the Philadelphia area week, with Bob Chassell joining
respectively met local Lojbanist in the regular Tuesday evening
Sylvia Rutiser, and agreed to    conversation group, reporting in
start writing Lojban letters to  Lojban on his touristy explo-
each other; there is now good    rations of Washington, and
hope that there will come to be  leading another round of "la reno
active Lojban social/study groupspreti".  One unfortunate problem
in those two areas.  Art and     with a weekend gathering is that
David also promised that every   so many (especially those from
once in a while they would pile  out-of-town) cannot arrive until
into the car and drive to the DC very late Frday (whereupon they
area for an informal Lojban      have to sleep half of Saturday in
social get together.             order to recover), and they then
  Sunday was dominated by the    have to leave by late afternoon
annual meeting of la lojbangirz.,Sunday.  Given that the annual
which started at 10:30 AM.  That meeting so dominates Sunday, this
meeting recessed for lunch, but  tends to give us less than a day
ran until 5 PM as we wrestled    for a variety of activities.
with financial issues and pri-   Thus the activities portion of
orities for the coming year.  A  LogFest has tended to be only
lot of decisions were made, and  mildly successful.
even more than previous years, I   We work more each year on pre-
think people were both satisfied planning activities, but planning
with the result and convinced    is inherently limited.  We never
that everyone had a meaningful   know till people arrive who is
voice in the process.  Since the coming, what their Lojban skill
latter was a major reason for    level is, and what activities
forming la lojbangirz., these    they find interesting.  Also, as
long meetings are worthwhile.    with the aphorism translations,
  We are taking some steps       activities that we test out
towards speeding up future       successfully in a weekly
meetings.  We will have more     conversation session may work
advance notice of agenda items soquite successfully with 5 or 6
people can be prepared for       people, but may bog down with a
discussion before LogFest starts.dozen or more participating.
We will also try to have a Board   Still, people noted and were
of Directors meeting perhaps a   pleased by the increasing
month before LogFest to weed out sophistication of the in-Lojban
issues and ensure group attentionactivities, and the general skill
to the most important, while     of everyone participating.  We
expediting routine business.  We still haven't reached the point
also hope, of course, that our   where Lojban conversations break

                                10
out spontaneously, but this may  lojbangirz.'s original charter in
happen next year given the rate  August 1987, when we started
of improvement in Lojban         "Lojban - The Realization of Log-
speakers.  More attendees will   lan", now also known as
make this more likely, and       "Loglan/Lojban" or just "Lojban",
improve the variety of activitiesla lojbangirz. has made repeated
going on at any one time.        efforts over the last several
  Total attendance was 17, most  years to mend the political split
of whom were there all weekend.  with The Loglan Institute, Inc.
7 were from out-of-town.  About    Earlier this year, we proposed
half were skilled enough to      a settlement that would have
converse at least minimally in   remerged the two current versions
Lojban, although such 'conver-   of Loglan into one.  The plan
sations' tended to be only       would have guaranteed an honored
snippets and remarks.  13        place for JCB, as well as organi-
attended the business meeting on zational and possible financial
Sunday.  John Cowan was elected  support for the Institute.  No
to the Board of Directors and    response was received.
Albion Zeglin dropped his Board    The Lojban design is
and voting membership due to lackessentially complete.  Time has
of time.                         run out on making changes to
  LogFest is supposed to be FUN, facilitate a merger - we can no
not all work.  A major differencelonger make significant changes
from previous LogFests is that   without corresponding impact on
the activities schedule didn't   those who have learned and will
include a mass of technical      learn Loglan/Lojban.  Our version
debates and decisions that had toof Loglan is now substantially
be made.  Of course, since the   better than the Institute's, and
major Lojban design decisions    we have people speaking and
have been made, only relatively  writing the language.
minor questions of style, seman-   As a result of this situation,
tics, and how we teach the       the LogFest attendees voted that
language remain to be resolved.  "Expending resources towards
These were decided in advance, orreconciliation with JCB or the
in a couple of cases, informally Institute is not a good use of
during the gathering (for exam-  resources at this time, but we
ple, the nest of issues we've    remain open to such
called "sumti-raising" - see     reconciliation should their
below - were satisfactorily      position change in the future."
resolved "in the halls" during   and "There is no longer special
LogFest).                        authority given to pronouncements
  Among minor decisions:         of JCB or the Institute about the
"?spero" as a culture word for   language."
"Esperanto" was voted down, and    It is unfortunate that we have
the baseline of the gismu was    had to go to such lengths in our
reaffirmed; few of the 'old-     dispute, but we have tried hard
timers' want even minimal change.and long for an alternative
"navni" is broadened to include  without success.  We cannot allow
"inert gas" in its meaning.      the ill will of one person, even
Finally, pending grammar         the language inventor, to prevent
proposals were adopted and the   us from freely using the language
grammar was rebaselined until af-he invented.  The language be-
ter the textbook is completed -  longs to the community now, as it
people are generally satisfied   must be to succeed.
with the grammar for now, and are  We hope that JCB and the
waiting to see how it is used andInstitute will change their
taught.                          position; we then can restore JCB
  la lojbangirz./Institute split to the position of honor and
- In accordance with a unanimous esteem that he once held among
vote taken at the time of la     the entire Loglan community.

                                11
                                 Languages Server' on the
  Electronic Distribution News   Internet.  Over the next few
        Lojban List Moves        months, as time allows, Bob will
                                 prepare materials for
  A major accomplishment of      distribution.  (We will also
LogFest was the adoption of a    supply data directly on diskette
policy for electronic            - current price is $10 per
distribution of materials that   uncompressed diskful, in any of
balances our desire to get these the 4 diskette formats we can
products to the public, thus     support:  5 1/4 and 3 1/2 high
aiding in the language growth,   and low density MS-DOS.)
with our need for income from our  For those with Internet access
publications, and a goal to      who wish to get materials, send a
fairly distribute our services tomessage containing, on separate
both computer people and non-com-lines, "help" and "index lojban"
puter people.                    to:
  The essential core of the      
policy benefits all Lojbanists,   listserv@hebrew.cc.columbia.edu
regardless of your access to     
materials:  All published        The Server will reply
"language definition materials"  automatically.  The index will
will be placed in the public     identify what files are available
domain, and will be distributable- a reading priority should be a
without restriction, in any      'read-me' file that will describe
medium.  These include word liststhe files officially put out by
and the language grammar.        la lojbangirz., and their status.
  Teaching materials, some draft The help file will tell you how
materials, and all JLs, will be  to request files to be sent to
distributable under our retained you - generally all you need to
copyright using a standard       do is say:
license - shown in the
distribution policy below.
  All materials, except those
that we rely on to show a profit
to support our other activities
(like software and the textbook),
will be posted for electronic
distribution.  Some materials,
like ju'i lobypli, will be posted
after considerable delay (6
months or more), so that we make
a current paid-for copy a
valuable service.  In addition,
the material as posted will
generally have minimal formatting
for electronic text.  Electronic
JLs and many other publications
will be difficult to read,
because standard electronic text
uses 80 characters per line, and
we use much higher print
densities in formatting our
publications.  As a result, an
electronic 'printout' of JL may
have sections that will be un-
readable without manual editing;
la lojbangirz. will not do that
editing.
  Our point of original
distribution will be the 'Planned

                                12
     "send lojban/filename".       Jerry Altzman is helping us out
                                 in another way.  Volume on the
  On an organized basis, we      Lojban List mailing group has
expect that much of this materialgrown so that it was straining
will be cross-posted to the Com- list-founder Eric Raymond's
puserve 'Foreign Language        network connection.  Jerry found
Education' forum by varying      room for us on one of the
Lojbanists with access to both   computers he manages, and Lojban
Internet and Compuserve.  Lojban-List was switched during the last
ists are welcome to distribute   week of August.  In addition, the
the material electronically in   list now uses a more advanced
keeping with the policy described"Listserv" process that allows
below - any restrictions will be people to sign up and remove
noted in the files themselves.   themselves from the list, tem-
  All materials will be released porarily suspend receiving
directly by me to Jerry Altzman  messages when overloaded or
of the PLS.  The read-me file    vacationing, and of course post
will contain my directory of     messages, all without human
dates and version numbers of all intervention.  See page 2 for
such releases.                   details.
  We eventually plan to include  
in the official directory an MD-4
(tamper-resistant 'message di-
gest' value) for each file so you
can verify that material you ob-
tain is authentic.  We will also
publish a printed MD-4 checksum
list separately, and will make
available for free a program to
determine the MD-4 checksum of
any file.  There are some hangups
in implementing the MD-4 support
because the checksum must be
calculated on the file as it
actually is sent by the Server,
which has UNIX-oriented line and
file conventions that differ from
the ones associated with the MS-
DOS version produced by la
lojbangirz.
  Others are encouraged submit
Lojban materials to the Server;
we will occasionally check these
materials and advise the Server
managers (Lojbanists Jerry
Altzman and Mark Shoulson) as to
which materials we think are
useful and current.  (We ask that
you send us a copy of all such
submissions, with a note that you
plan to so submit them.  Send
them either by paper-mail to the
la lojbangirz. address, or
electronically to:

        lojbab@grebyn.com

la lojbangirz. encourages
comments on draft materials that
are released to PLS.

                                13
 Logical Languages Group Policy  word lists and other language
   Electronic Distribution of    definition materials, will be
            Materials            copyrighted using a copyright
      Approved 23 June 1991      notice essentially similar to the
                                 one attached to this draft
         Copyright, 1991         policy.
The Logical Language Group, Inc. 
        (la lojbangirz.)         4) To assure the integrity of
2904 Beau Lane, Fairfax VA 22031-electronically distributed la
            1303 USA             lojbangirz. materials, every
      Phone (703) 385-0273       document distributed
                                 electronically will bear a
All rights reserved.  Permission message digest value computed
 to copy granted subject to your using the MD-4 algorithm, source
  verification that this is the  code for which is publicly
latest version of this document, available.
that your distribution be for the
 promotion of Lojban, that there 5) la lojbangirz. will make
is no charge for the product, andavailable, free of charge, a list
  that this copyright notice is  of the MD-4 message digest values
  included intact in the copy.   for all materials released in
                                 electronic distribution.  la
1) la lojbangirz. publications   lojbangirz. will also provide a
and materials are hereby divided program to compute message digest
into three groups:               values, free of charge with the
                                 purchase of Group C materials,
  Group A materials consist of   subject to technical limitations.
text, and are sold at or near    
cost.                            6) la lojbangirz. intends to use
  Group B materials consist of   the Planned Languages Server as
text, and are sold above cost.   the primary distribution medium
  Group C materials consist of   on the Internet.  Other
computer software, and are sold  distribution media on the same or
above cost.                      other networks may be established
                                 at la lojbangirz.'s discretion.
This division is independent of  
the division into Level/Package  7) Materials in Group B and Group
0-3 materials, which depends not C will not be distributed
on cost but on the presumed      electronically.  Group C
interest level of the reader.    materials in object form will be
                                 distributed on diskette and
2) The following are non-        whatever other media are
exhaustive lists of materials in technically available to la
each group:                      lojbangirz. (currently, none).
  Group A: JL and LK issues;     
draft textbook lessons; word     8) Source code to Group C
lists; language definition       software will be made available
materials; ancillary materials.  on diskette or other media to
  Group B: the (as yet unwritten)persons who sign a non-disclosure
textbook; the (as yet unwritten) agreement with la lojbangirz., at
dictionary.                      a cost equal to the cost of the
  Group C: Logflash for PC and   Group C software in object form.
Mac; the la lojbangirz. Lojban   
Parser (in beta release); lujvo- 9) This policy becomes effective
maker; random sentence generator.when ratified by la lojbangirz.'s
                                 official bodies.  (it has been.)
3) la lojbangirz. will provide   It may be altered at any time by
all materials in Group A for     la lojbangirz.
electronic distribution free of  
charge.  All materials, except   

                                14
 Language Development Activities   A last minute proposal assigns
                                 rafsi to fo'a, fo'e and fo'i
  Vocabulary - Many minor        (selma'o KOhA).  These assigned
vocabulary-polishing activities  to names with du or goi plus
occurred since last issue.  20   several other cmavo rafsi (mi,
gismu proposed and approved last do, vi, va, vu, ti, ta, tu) can
year were finally created using  be used along with names to allow
the 6-language algorithm.  rafsi more abbreviated expressions of
were assigned to as many of thesecultures not included in the
as possible, and the cmavo list  gismu list. e.g. fo'e du la
was examined to see how many     suomis (Finland).  .i mi cilre lo
cmavo that might be useful in    fo'enselsanga.  (I learn a
lujvo could be assigned rafsi.   Finnish song.)  Since the most
The revised rafsi have been      useful culture words are those
released in an updated list - seefor 'my' culture and 'your'
the products news below.  The newculture, "mi" and "do" will be
gismu and the changes to the     likely to be used in this way.
rafsi list are in the features     The last paragraph uses the
section of this issue.           word "selma'o", which may be
  The cmavo list has also been   unfamiliar.  We have adopted this
updated - reflecting the grammar lujvo for what we have previously
and usage developments of the    called a "lexeme".  The lujvo is
last year.  Extended definitions,based on the second place of
up to 96 characters long, are    "cmavo", which is the grammatical
incorporated into the new list.  role of the cmavo.  The things we
The cmavo list update will be    are calling "selma'o" are the
released at approximately the    basic grammatical types of cmavo
same time as JL16 in October,    and other words found in Lojban.
along with the Logflash 3 cmavo    (The definition of "selma'o"
instruction software and other   shows a little of the meaning
materials, giving time for last  variation permissible in lujvo,
minute reviews.                  since selma'o BRIVLA and CMENE
  The gismu place structure      are not grammatical units of
revision has been idling since   cmavo, although all other selma'o
last fall.  This project was     are.  The generalized meaning
intended to produce 96-character implicit in "selma'o" is
extended and clarified place     acceptable since people learn
structures/definitions for each  finely details of word meanings
gismu, thus providing clearer    by seeing how they are used, not
information for those learning   by some kind of rigorous
and using the words, and allowinganalysis.)
the new list to be used as input   Grammar - The proposed changes
for the updated LogFlash 1, now  to the grammar printed in JL14
scheduled for October release.   went without a single comment, or
  The place structure review willeven a question.  What little
almost certainly not be completedfeedback we got seemed to
before that October release      indicate that the discussion was
because of its relatively low    too technical for most readers,
priority, so we have decided thatand that without considerably
a version close to the present   more discussion and examples,
working list will be released in printing the proposals was not
October at the time LogFlash is  worthwhile.
updated, replacing the current     Additional proposals evolved
list.  The new list will become  after JL14 was published, finally
the official public domain       totalling 28.  All but one, the
language definition list upon re-'sumti-raising' proposal
lease, and we will recommend thatdiscussed below, passed without
people studying or using the     comment from Lojban List as well.
language start using that list as  Thus, at Logfest, the set of 28
soon as possible.                changes was adopted, and the

                                15
grammar was rebaselined until    has backed this discussion with
after the textbook is completed  the most prolific use of the
and reviewed.  We do not plan to language after Michael Helsem
consider any changes until then, (whose Lojban poetry is now truly
and very few are expected to     voluminous - he has published a
surface, anyway.                 volume of it).
  Even the 28 changes adopted are  Style and semantic issues that
quite minor:  almost nothing     have been raised and discussed on
written in the language in the   Lojban List are too numerous to
past two years became            mention here.  A lengthy
ungrammatical as a result of     discussion of relativistic tenses
changes, and a few things not    started the trend last winter.
grammatical became so, since manyMore recently, the primary topics
of the changes were designed to  have been the determination of
bring the formal description of  meaning of lujvo (stimulated by
the grammar more closely aligned Jim Carter's oft-rejected
with how people actually were    proposal for what he calls
using the language.              "dikyjvo" - regular mandatory
  Indeed, and this seems         rules for building lujvo based on
significant:  in the last few    the source gismu place
months it has become clear that  structures), the distinction
no longer is the language design between abstract and non-abstract
being driven by language         sumti values (tied in with the
engineers like myself who are    discussion of 'sumti-raising' -
trying to figure out how people  see below), the meaning and usage
WILL use the language.  Instead, of the various modals in selma'o
we have a group of people using  BAI, and the mass/set/individual
Lojban, and what they find out indistinction in Lojban
trying to express things in the  descriptors.
language has driven many, if not   Other 'old' issues are really
most, of the most recent changes.semantic ones.  Debate has
  The other significant factor incontinued on the necessity and
the grammar is that a complete-  value of the cultural gismu and
grammar Lojban parser has finallythe gismu that represent
been completed.  Not only does   elements.  Most often the debate
this provide a new standard for  derives from new people who are
what is grammatical in the       not familiar with the reasons why
language, but it serves as a     they were included, which include
stabilizing force motivating     historical reasons as well as the
against changes that might renderjustification of usage.  There is
this valuable tool outdated.     considerable fear that these
(The parser is expected to be    words will lead to cultural
released some time this fall.)   biases, fears not shared by Bob
  Semantics and style - A new    and others who have been working
entry in this discussion, becauseon the language longest.  We
the Lojban design plan excludes  expect that this issue will not
semantics and style being        be resolved until the dictionary
prescribed.  However, we have    is published, wherein the words
people actively using the lan-   for other cultures and elements
guage in conversation,           that did not get assigned gismu
translation, and new writings.   will be listed, along with the
The questions that come up in    rules for deriving new words of
actual usage of the language are those kinds as needed.  (An
generally not grammatical ones,  article later in this issue
but usage questions like "How do discusses cultural gismu.)
you say this?" and "Why doesn't    One recurring issue that
this work?".                     affects the community as a whole
  One Lojbanist, Nick Nicholas,  is the frequency and type of
has made discussion of style his translations presented with
primary theme on Lojban List.  HeLojban text.  We can give no

                                16
translation, or a block          the book, but if you like poetry,
translation for an entire text,  the English versions will have
or line by line translations     value and the enormous volume of
which are either colloquial      Lojban may inspire you, as well
English or word-for-word.  The   as provide ideas on what works
more literal the translation, theand what fails to communicate in
less need you have to look up    Lojban text.  We have several
words in words lists.  This can  copies left of this 'first Lojban
be both good and bad:  the trade-book', which we will send free
off is between learning the      upon request to anyone making a
vocabulary or understanding the  prepaid order over $20, or for
grammar.  Some people want text  postage costs only ($2-$3)
they can try to read and be      otherwise.  Michael seeks
challenged.  Others are just try-comments and suggestions from all
ing to get a feel for the        readers.
language.  What do you want?       John Hodges observes that
What do you think we should      Michael's publication, even with
change, if anything, in our      imperfect Lojban, is a "signifi-
Lojban text presentations in JL? cant event, symbolically and
                                 politically.  This is exactly the
                                 kind of thing [la lojbangirz.]
       Using the Language        wanted to make possible by
                                 insisting that the language be
  This is the most significant   public domain, and precisely what
area of news, in my opinion.  TheJCB wanted to prevent by keeping
number of people actively trying copyright control over the very
to speak and write in Lojban to  words of his language.  Helsem
communicate with others has ex-  did not ask permission to
ploded.  Since JL14, I have re-  publish.  You and he took it for
ceived or reviewed extensive textgranted that it was his right to
(more than a couple of paragraphspublish.  JCB would deny this.
of block text) in Lojban from BobTo defend the purity of the
Chassell, John Cowan, Ivan       language, JCB would insist that
Derzhanski, Coranth D'Gryphon,   Helsem correct his grammar before
Michael Helsem, Rory Hinnen, Nickpublishing.  (Not to mention,
Nicholas, Sylvia Rutiser, Mark   send royalties to JCB.)"
Shoulson, David Twery, and writ-   Sylvia Rutiser and Ernest
ten some myself.  By comparison, Heramia started an intermittent
only Jamie Bechtel, John Cowan,  'pen-pal' correspondence last
Sylvia Rutiser and myself sent inwinter.  Ivan Derzhanski
extensive text over the 8 month  (Bulgaria) and Nick Nicholas
period between JL13 and JL14.    (Australia) started the first
  This is not counting a couple  international correspondence ex-
of dozen people who have written change in May.  Recently Sylvia,
letters or sent messages elec-   David Twery, and Art Protin
tronically with a sentence or twostarted a round-robin letter
of understandable and often gram-exchange.  I have a list of
matical text.  Several other peo-several others interested in
ple have told me that they have  writing letters in Lojban - send
written some, or a lot of, Lojbanus a note with a few sentences
text (in some cases, I am waiting(or maybe a self-descriptive
to see before believing; the     paragraph) in Lojban with English
amounts claimed seem incredible).translation, and we will try to
  Michael Helsem has collected   match you with someone of
several of his Lojban poems, madecomparable skill.  Give us some
corrections, and published them  indication of how often you would
in an artistically decorated     expect to write - one problem we
cover - copies were given to     have experienced so far is people
every LogFest attendee.  There   prepared to write as often as
are still some Lojban errors in

                                17
once a week paired with people   other ongoing activity is the
who take months to respond.      construction of a Lojban
  The amount of Lojban text now  traveler's phrase book, after the
being posted on the Lojban List  style of Berlitz.
is rather overwhelming at times.   New Lojbanic activities seem to
Nick Nicholas first got materialssurface every week or two, and I
from us around the time JL14 was have no doubt that there will be
published.  He has recently been a new crop of them to report by
the most prolific and one of the JL16.  Why not let yours be among
most skillful among Lojban writ- them?
ers, posting paragraphs of text                  
to Lojban List virtually every                   
week.  Noting that Nick is also a    Research and Linguistics
full-time student AND one of the                 
leaders of the Australian Es-      The Loglan Project is starting
peranto organization, his        to become a real research
productivity makes me ashamed of endeavor again.  We have estab-
my own (but .ui what             lished a presence on several
inspiration).                    major forums for computer
  Also on the computer network,  linguistics information exchange,
Jack Bennetto has started a game and are making ourselves known to
of "telephone" (you may know thislinguists who are researching in
as "whisper down the line", or byareas where Lojban might be rele-
another name).  Starting with a  vant.  Among these areas are:
moderately complex sentence, each- linguistic expression of
successive person translates whatemotion;
he/she receives from English to  - word compounding;
Lojban or vice versa, and passes - predicate deep structure
the translation to the next      grammars;
person.  We've had no reports yet- the ISO standards for
on how well this activity is       international character set
proceeding.                        encodings;
  Weekly Lojban conversation     - semiotics;
sessions have continued here in  - representation of abstraction;
the Washington DC area, with any-- logical expression;
where from 3 to 6 attending each - computational linguistics;
session (about 10 people total   - machine translation;
have participated).  The amount  - abstract system specification
of conversation time has dropped   language;
a bit, because the group spent   - foreign language education.
time before LogFest planning ac- 
tivities for the gathering.      At least one well-known linguist
Since LogFest, we have started anhas expressed interest in Lojban,
intermittent group project -     and we hope to attract many more.
translating the entire board game  Bob wrote an essay on the
"Careers" into Lojban in honor oflinguistic research applications
Jim Brown, who invented both the of Lojban for posting to one of
language and the game.  (We may  these groups.  This essay appears
seek permission from Parker      later in this issue, slightly
Brothers Inc., which owns rights edited.  A new version of the
to the game, to distribute the   Lojban brochure will be issued in
game translation to those of you a couple of months, incorporating
who are interested.)             some of this material.
  Not all Lojban text is orderly.  Athelstan and Bob attended GURT
Next issue will contain a        (The Georgetown University Round
sampling of the Lojban graffiti  Table of Linguistics) this year.
that appeared on a wall of Bob   GURT is one of the more
and Nora's house (specially      prestigious linguistics
prepared to make this non-       conferences.  There were just
destructive) during LogFest.  Oneunder 800 attendees.  After

                                18
initially being hesitant for fearinitial bid, but preparing the
of adverse reaction from         proposal stimulated much new
linguists, on Wednesday we put   activity around here and opened
out about 30 brochures with a    options that look quite promising
short note on Lojban's           for the future.
applicability to linguistics       Bidding for research grants is
research.  They were gone within a learning experience. In today's
two hours.  On Thursday we put   competitive research environment,
out 110 more, and nearly all wereit may take several proposals to
gone when the conference ended atget one grant or contract.  The
4PM.  We got some great name     initial proposal not only serves
recognition out of this, even if as a basis for further proposals,
none of these brochure readers   which are now half written at the
decides to do something about    start, but every effort we make
Lojban just yet.                 teaches us more about how to do
  I suspect some will do so      things better the next time.
eventually.  Almost everyone we    For example, since submitting
talked to seemed at least mildly the proposal abstract, John Cowan
interested in the concept of an  has been researching and writing
artificial language designed for up a detailed analysis that shows
linguistics research, and a      that Lojban is a superset of the
couple of researchers thought we computer language PROLOG, often
had some interesting research    used in artificial intelligence
angles that they might like to   processing.  This means that
investigate.  I would say that   most, if not all, Lojban
Athelstan and I together threw upsentences could be processed into
more questions (usually good fromPROLOG statements and fed into a
the reaction of the audience and PROLOG processor.  This would
the speaker) than most people, sogreatly reduce the cost and risk
I'm sure we were noticed.        of developing a Lojban processor
  The primary topic at GURT was  from scratch.  (We seek PROLOG
foreign language education, but  experts among the community to
we also attended sessions on     review John's work.  Let us know
natural language processing.     you're interested!)
  la lojbangirz. is planning to    A major plus in our efforts to
attend at least one and possibly obtain research funding is John
two more linguistics conferences Cowan's completion of a full-
this year.                       language Lojban parser.  Still in
  la lojbangirz. is closer to    testing, this parser breaks all
initiating scientific research   Lojban text (including cmavo
using Lojban.  The new version ofcompounds) down to individual
LogFlash contains instrumentationwords and parses the results.
that will allow study of how     The ability to parse at the
people learn words, and whether  individual word level is a major
the recognition score algorithm  improvement over the best
used to build the words has any  accomplishments of the Loglan
relevance to their learnability. Institute before we started on
  More importantly, la           the Lojban redesign.  More
lojbangirz. in July prepared and importantly, it is better than
submitted its first research     anything that can be accomplished
proposal.  The proposal (actuallyin processing natural languages.
a proposal abstract since we did   Of course, our 'advantage' may
not request a specific dollar    be a problem with getting DARPA
amount) was submitted to DARPA   funding.  It turns out that
(US Defense Advanced Research    having bypassed the worst
Projects Agency), the primary    problems in natural language
government funding agency for    processing, the problems that we
artificial intelligence and      need and want to solve to process
natural language processing      Lojban text are quite different
research.  We didn't win this    than the ones considered on the

                                19
'leading edge' of research.  We  (Level B) Observer (old level
thus are required to write         0/B)
proposals extremely carefully to (Level C) Active
show how learning to process       Observer/Supporter (old level 1
Lojban text will lead to better    and 2)
processing of natural languages. (Level D) Lojban Student (old
  We hope to include portions of   level 3)
our proposal in JL16, in order to(Level E) Lojban Practitioner
give our supporters an idea of     (people demonstrating some
how we are presenting the          competency with the language,
language.  But also, we welcome    and actively using it in some
suggestions from the community on  regular activity)
how to better explain our        
research approach, and to prove
that it is sound.  (We also want
to hear of any alternate research
approaches that we may be
missing).
                
                
  Products Status, Prices, and
            Ordering

  With the decisions described in
the finances section, we are
making changes in our coding for
mailing status.  These changes
are summarized in the new mailing
label coding block on page 1.
  Most importantly, we have
separated JL and LK subscriptions
from the status codes (levels 0,
1, 2, 3, and B).  We have also
added an automatic update status
that is independent of the
others, indicating your desire to
receive updates and your
commitment to keep enough in your
balance to pay for them.
  Next, we are separating the
activity level implied in the
level numbers from the encoding
of the materials we actually have
sent you.  As people have moved
around in level, or been
downgraded, your 'mailing level'
no longer tells us what material
you have.
  The activity level portion of
your level will be converted to a
letter code indicating your
current interest level.  The
level numbers 0 through 3 will
refer to a series of packaged ma-
terials that will tell us what
we've sent you.
  The conversion to letter codes,
and their interpretation, is as
follows:


                                20
('A', in case you are wondering,   Products and Schedule - This
is used for people dropped from  past year has been one of change,
our mailing list, for whom we    of consolidation.  We haven't
maintain financial accounts      produced many 'new' things; we
because we've sent materials.)   have been enhancing and refining
                                 old ones.
  The difference between old       The fruits of that effort are
level 0 and old level B has      now starting to show up on our
merely been whether you were     order forms.  Even more will ap-
receiving le lojbo karni or not. pear over the next couple of
  The original difference betweenissues.  The following is a
old levels 1 and 2 was whether   summary of the current products
you automatically get updates of schedule (as well as the minor
materials when they are updated  releases since last issue):
(presumably a level 2 was more   
active and needed the latest in-             (Jun 91)
formation for active work).       Electronic postings to P.L.S.:
Since we went so long without      Baselined gismu list (old
issuing any updates, and have    version)
gotten into such a financial       Draft Proposed gismu Place
morass, the distinction became        Structure Revisions
insignificant.                     Review of Loglan 1 - Draft Long
  Then, during the last year, we      Version
started sending some additional  
materials to level 2 people that
we don't send to level 1 people,
in order to keep the level 1
price down.  Thus the original
distinction we intended between
the two levels was lost, and we
are restoring that information as
the automated update flag.  You
will not receive automatic
updates unless you keep suffi-
cient balance to pay for them.
  These codes will now appear
separately on your mailing label,
and with the start of paid JL
subscriptions, your subscription
expiration date/ issue will also
appear on your mailing label.
  There is increasing interest
among Lojbanists in contacting
and communicating with others of
equivalent skill levels.  Right
now, Bob makes these evaluations
subjectively, but as the numbers
of people actually using the
language increases, Bob's
evaluations become less reliable.
  Thus, we are also planning a
proficiency code system that will
tell us your demonstrated profi-
ciency level at reading, writing,
or speaking Lojban.  To minimize
confusion, we will delay im-
plementing this for about 6
months.  Suggestions are welcome,
though.


                                21
            (Aug 91)               Rebaselined gismu list
Printed:                              (updated)
  Updated rafsi list and lujvo-    Synopsis of Lojban Orthography,
     making guide                     Phonology, and Morphology
                                      (updated)
            (Sep 91)               Lojban and Machine Translation
Printed:                           Lojban and Esperanto - 16 Rules
  JL15                                Comparison and Commentary
  LK15                             Lojban, Sapir-Whorf and
  Synopsis of Lojban Orthography,     Semiotics
     Phonology, and Morphology                   
     (updated)                               (Nov 91)
  Attitudinal Paper (updated)    Printed:
  What is Lojban - la lojban. mo   JL16
     Brochure (revised)            LK16
  What is Lojban - la lojban. mo Software
     Brochure (Esperanto version)  Lojban Parser (PC and some UNIX
Software:                             versions)
  Revised Random Sentence        Electronic postings to P.L.S.:
  Generator                        New Textbook Lesson 1 Draft
  Revised lujvo-Making Program     JL14
Electronic postings to P.L.S.:     LK14
  What is Lojban - la lojban. mo   le'avla-making algorithm and
     Brochure (revised)               examples (Cowan)
  What is Lojban - la lojban. mo 
     Brochure (Esperanto version)
  Overview of Lojban (1991                   (Dec 91)
  update)                        Electronic postings to P.L.S.:
  lujvo-making guide               selma'o paper (Cowan)
  Updated rafsi list               Selected list of Lojbanized
  Re-baselined formal grammar         names
  E-BNF for re-baselined grammar   Revised Draft Lessons 1-6
  Reply to Arnold Zwicky's review  A comparison of Lojban and 1989
     of Loglan 1 (orig. review        Institute Loglan (Cowan)
     1969)                         Glossary of Lojban/linguistic
  Revised cmavo list                  terminology
  Back issues of JL #1-13        
  Back issues of LK #8-13                    (Jan 92)
  Summaries of sci.lang          Printed:
     discussions of Lojban         JL17
  The Lord's Prayer in Lojban      LK17
     (Revised 1991)                Lojban Learning Materials
  Negation paper                      (Book)
  Lojban Mini-Lesson (Athelstan)   Lojban Reference Materials
                                      (Book)
            (Oct 91)                             
Printed:                              Unscheduled But Planned
  Re-baselined formal grammar    
  E-BNF for re-baselined grammar Printed:
  Lojban Mini-Lesson (Athelstan)   Lojban Textbook
  Revised cmavo list               Lojban Dictionary
  Rebaselined gismu list           Lojban Pocket Reference
     (updated)                     Lojban Reader (Book)
Software:                          Lojban Phrase Book
  LogFlash 1 - gismu (Revision 7)Printed and Electronic:
  LogFlash 3 - cmavo (Revision 1)  Lojban gismu Etymologies
Electronic postings to P.L.S.:   Software:
  Lojban Tense Paper (Cowan)       Logflash 2 - rafsi (Revision 7)
  Lojban MEX Paper (Cowan)         Hypercard LogFlash/Mac -
  Attitudinal Paper (updated)         (Revised and New versions)

                                22
  Lojban Adventure Game          worst; page numbering will be
                                 incorrect, and graphics (like
  Now that we've shown the       Nora's cartoons) will not be pre-
overall plan, we can explain.    sented at all.  Some of these
  As with all of our schedules,  materials will thus be of
this one should be taken as a    marginal use; consensus is that
plan, not a promise.  We are a   different people judge usefulness
volunteer organization and the   by different standards.
schedule depends on the time       Software updates to all of our
availability of specific people. software nears completion.  In
We also are short of money, and  the case of the lujvo-making pro-
the scheduled publications dependgram and random sentence
heavily on significant numbers ofgenerator, this is merely an
you paying up your balances and  update to accept new data files
putting in additional money to   based on the latest language
cover these new products.        definition.
  The true goals are the items     As described in JL14, LogFlash,
listed for next January - two    our vocabulary teaching software,
books that will contain all of   is undergoing a major overhaul.
the teaching and reference       The new versions retain the
materials we have put out,       teaching algorithm that has
updated to the current language. proven so effective, but adds
  In the process of creating     colored screens, user
those books, all of our current  flexibility, and a new learning
products will be updated to re-  mode designed to help Lojbanists
flect changes in the language or quickly become familiar with the
the way we teach it.  As each is range of the gismu and cmavo
updated, there will be a heavy   without the time-consuming effort
emphasis on making it available  needed to master the lists.
on the Planned Languages Server.   Since the major guideline for
This helps fulfill our obligationthis schedule is the earliest
and commitment to place the      practical publication of two
language definition materials in books, let us look more closely
the public domain, enables more  at what they are and why we are
people to see detailed design    putting them out.
information about Lojban, and of   First, these two books may be
course gives us some last minute considered the prototype Lojban
feedback on these materials      textbook and dictionary.  The
before binding them.             word "prototype" is used because
  In the process of making these Bob has long had an idea of what
materials available, we will be  a Lojban textbook and dictionary
reviewing them for current       SHOULD be, and these short term
accuracy, and will make minor    products will not be anything
revisions and updates.  Some of  like the goal versions.
the printed products will thus     However, people in the
not be generally distributed - wecommunity are in need of books
won't waste your money (and Bob'scontaining materials for studying
time) sending you minor          Lojban, and reference materials
corrections to a publication you needed to use Lojban.  Even
may not be using.  We will,      serious Lojbanists who work with
however, send the latest version the language a lot are becoming
on new orders, and inform you in overwhelmed by the volume of
this column about other revisionsmaterials and updates that la
you may want to know about.      lojbangirz. has issued.  Time is
  As noted in the discussion of  being wasted hunting through
electronic policy above, we will accumulated ju'i lobypli issues
not be rewrite or specially      and enclosures, and other
format materials for electronic  materials you have obtained from
distribution.  Tables and some   us, looking for relevant material
example texts will suffer the    that has become a bit outdated.

                                23
  The language design is now firm     one, possibly with sample
enough that we can create up-to-      sentences for each;
date versions of all important   - a glossary of linguistic and
materials.  By collecting these       Lojban jargon terms;
materials in bound volumes, we   - a selected list of le'avla
give people actively working with     borrowings and an algorithm
the language the tools you need       to make more;
to do so - all in one place.     - a selected list of Lojbanized
  Bob's work on the textbook          names;
revision has dragged on far too  - a set of cultural/national
long, and the reasons are not         words for all countries in
going away.  Shifting away from       the United Nations, and se-
long-term targets to short-term       lected other places;
goals, he has already picked up  - a selma'o catalog describing
in productivity.                      the grammar of each word
  The materials being revised and     type, with many examples;
issued electronically (and       - as many sample lujvo as we have
occasionally in printed form)         time to verify and space to
over the next 6 months will           include.
become the contents of the new     This book will also run about
books.  The books will then be   300-400 pages and be bound.  It
assembled out of the revised     will also probably cost around
pieces and published, hopefully, $25, depending on orders.
around the beginning of next       Since we have had an excellent
year.                            record of recruiting Institute
  Under current plans, the       Loglanists who later find out
learning materials book will     about Lojban, la lojbangirz. is
contain the 1st lesson of the    planning a 'guide to Lojban for
revised textbook, Lojban mini-   Institute Loglanists', which may
lessons by Athelstan and John    be incorporated in one of the two
Hodges, the 6 draft textbook     planned books.  This will
lessons, the negation paper, the maximize people's use of the
attitudinal paper, the old       Institute's books that may have
grammar summary, and selected    been purchased, since much of the
short writings (mostly revised   material JCB has written applies
from JL articles) that teach the to Lojban equally as well as the
language.  This will be publishedInstitute version of the
as a bound book, probably Velo-  language.
bound.  Total page count will      Once the books are out, Bob
exceed 400 pages.  Price will be will then concentrate on
around $25-$30, depending on the producing refined versions while
number of advance orders.        you concentrate on learning and
  The reference manual will      using Lojban.  Your efforts will
     contain revised versions of:then provide the hundreds of
- la lojban. mo brochure;        examples needed for the properly
- Overview of Lojban;            completed books.
- Machine grammar and E-BNF;       The only immediately available
- Synopsis of phonology,         new product is an updated rafsi
     morphology, and orthography;list, incorporating the changes
- gismu list updated with new    listed later in this issue.
     place structures, and       Also, since people receiving the
     Roget's Thesaurus codes, andrafsi in the past have often had
     multiple English synonyms   no idea how to use it, the
     where applicable (instead ofsection on lujvo-making from the
     the English keyword index   Synopsis has been extracted and
     used now);                  heavily revised, and will now be
- rafsi lists and lujvo-making   distributed with the rafsi list.
     guide;                                      
- cmavo list with clearer                        
     definitions than the current       International News

                                24
                                   Australian Lojban Society - la
  Esperanto brochure -           lojbangirz. has effectively
Considerable effort by Paul      gained an affiliate in Australia.
Francis O'Sullivan, followed up  Major, in Perth, and Nick
by Mark Shoulson, Nick Nicholas, Nicholas, in Melbourne, are
and David Twery, have led to a   attempting to establish and keep
complete and up-to-date          contact with all Lojbanists in
translation of the la lojban. mo Australia and New Zealand.  In
brochure into Esperanto.  In     addition, because the cost of
addition, Mark is formatting it  mailing overseas is so high,
for a typeset-quality master, andMajor is serving as a focal point
we should have printed copies    for la lojbangirz. mailings, and
within a few weeks.  The brochurehe then redistributes copies to
will also be posted              all his correspondents.  Nick is
electronically on the Planned    becoming one of our most skilled
Languages Server, and possibly onLojbanists, and can answer most
Compuserve.                      questions about the language.
  Large numbers of our readers     This benefits la lojbangirz.,
are Esperantists interested in   because we lose money on most
Lojban.  We encourage you to     overseas mailings even with the
distribute copies of the         20% surcharge we require.  It
Esperanto version to other       benefits those who are part of
Esperantists.  This not only willthe new group, because it costs
spread knowledge of Lojban aroundless for all of you:  Major can
the world, but it will enhance   produce copies for you and get
our position as an artificial    them to you via local post much
language working with the        cheaper than we can.  Major does
Esperanto community, and not in  ask for reimbursing of his expen-
competition with it.  Indeed, we ses, or the group will not be
now see Esperanto as one of our  able to grow.
primary languages for spreading    Major and Nick both keep in
information about Lojban to othercommunication with the rest of la
countries.                       lojbangirz. via electronic mail.
  We don't yet have other          Major's address is:
materials about Lojban in        
Esperanto, but we expect that    Major
this will change.  As more and   Box T1680 GPO
more Esperantists who also speak Perth WA 6001
English join in with those who   AUSTRALIA
translated the brochure, our     
ability to produce Esperanto       Nick's address is:
translations of our other        
materials improves.              Nick Nicholas
  (We remind our readers that we 17 Renowden St.
also have a French translation ofCheltenham Victoria 3192
the brochure, although it has notAUSTRALIA
been updated yet to reflect new  
policies and new materials, and    It doesn't take a lot of people
is missing the newly added       to make this type of regional
section on Lojban and linguisticsgroup work (there are 7 on our
research.)                       lists in this region of the
  (We are constantly seeking     world, and only 5 are thus far
volunteers to translate any of   participating).  We require that
our materials into other lan-    one person is willing to take
guages.  Please contact us if    responsibility to get materials
interested.  Such volunteer work to the others, and also take the
is the type which we qualify for financial risk of supporting
credits in receiving materials   those who don't pay for materials
when you cannot pay for them.)   right away.

                                25
  We welcome others who would     Publicity - Bob and Nora's Trip
like to try to similarly organize                
the people of your country and     Bob and Nora travelled to the
possibly neighboring countries.  San Francisco area in late April
Already, we have a potential     for a vacation and Lojban pro-
volunteer in Sweden, Christopher motion trip.  We had an
Arnold, who hopes to organize andopportunity to meet with several
recruit other Lojbanists to join Lojbanists, though with a couple
the half dozen of you now in the we were dogged by an inability to
Scandinavian countries.          get schedules together.  We
                                 regret those of you we missed.
                                 (One way not to be missed is to
                                 make sure we have your telephone
                                 number - sometimes plans get made
                                 in a big hurry.  Specify if the
                                 number is unlisted or otherwise
                                 not for release to other
                                 Lojbanists.)
                                   Bob gave one lecture, to a
                                 group of students at St. Mary's
                                 College including Dr. Robert
                                 Gorsch's class in semiotics that
                                 has studied a small Lojban unit
                                 (see JL12 for more on this
                                 class), and gave two talks
                                 combined with mini-lessons to
                                 groups of Lojbanists.  Dave
                                 Cortesi organized and publicized
                                 the primary meeting, held in Palo
                                 Alto.  Donald Simpson organized
                                 the other at his house in Albany
                                 as a smaller event for those who
                                 couldn't get to the other
                                 meeting.
                                   A total of around 20 people
                                 showed up between the two
                                 meetings.  Special pleasure was
                                 Scott (Layson) Burson's
                                 attendance at the Palo Alto
                                 meeting.  Scott, now inactive in
                                 the Loglan community, did the
                                 final work to complete the first
                                 Loglan parser and the first
                                 version of the machine grammar
                                 accepted by the Institute.
                                   Jay Stowell arranged to
                                 videotape the Palo-Alto mini-
                                 lesson.  We have considered
                                 distributing copies of this, but
                                 the cost of videotape duplication
                                 is high enough that we want to
                                 use a better original (unedited
                                 videotapes have a 'home movie'
                                 quality about them, and we saw no
                                 easy way to turn Jay's tape into
                                 a salable product).  We are going
                                 to try to specially film a mini-
                                 lesson, hopefully later this
                                 year.  Brad Lowry, who does
                                 professional video filming, has

                                26
volunteered to film and edit thislegal position, that 'Loglan'
mini-lesson.                     cannot be a 'trademark', was
  Some new people attended the   sound, and was important to our
Palo Alto meeting, and at least  making the Loglan project a
one person signed up as a level 3success.  Contrary to what JCB
Lojban student.  All-in-all, the claims,  our legal fight started
meeting was a success, though we when Jim Brown sent a letter
always wish we could have done   threatening us with legal action
better at getting information to (a copy of this letter was
prospective attendees and helpingincluded in all issues of JL5);
more people to attend.           it is unfortunate for all of us
  Finally, Bob and Nora got      that his position on threats was
together for a brunch with Scott not then what it is today:
Burson and Doug Landauer, another  JCB reiterates his claims that
pioneer in Loglan machine grammarBob and Nora split off from the
work.                            Institute "presumably to ac-
                                 commodate their own
                                 entrepreneurial interests", using
     News From the Institute     "the threat of schism to try to
                                 make us [change Institute
  Legal - Last issue, we thought business policies]".  He then in-
the legal battles between la     sists that the two efforts "went
lojbangirz. and the Loglan Insti-their separate ways" because
tute had finally ended.  Alas,   "threats seldom work on human be-
the day after JL14 went to press,ings".
we heard from our lawyer that Jim  Our answer:  No!  Our purpose
Brown had informed him of his    in starting Lojban was to put
intent to appeal to the US Court Loglan in the hands of the people
of Appeals.                      who had been promised it, had
  At this writing, the appeal    paid for it, and had long assumed
process is well underway.  The   that they had the free right to
Institute has filed its appeals  use it as they choose, as "the
brief and we have responded; we  human use of any language is, of
see little chance of the appeal  course, in the public domain"
succeeding.  We won't go into the(Jim Brown, again, but this time
issues again at length - anyone  from a 1977 proposal).
interested can contact us for      JCB also claims that "there
details.                         were no substantial intellectual
  We are hoping for a ruling     differences between me and the
around the end of the year which proto-Lojbanists".
firmly closes the door on the le-  Response:  We consider our
gal battle.  Meanwhile, we are   commitment to intellectual
proceeding in accordance with thefreedom a substantial difference.
decision, using "Loglan" to refer  To stop these misstatements of
to the generic language of which our purpose and goals, and to
"Lojban" and what we have been   ensure that there is no further
calling "Institute Loglan" are   doubt or misconception of our
versions.                        true purpose, we have modified
  JCB claims in the new Lognet   our statements about Loglan and
that our initial challenge was   Lojban that appear on page 1 on
"an harassment designed to straineach issue of ju'i lobypli to
our resources" and that our suit more clearly indicate that the
"is a timewaster once based on   free use of Loglan as a human
the premise that The Institute   language is the sole reason for
couldn't or wouldn't be able to  the split and our existence.
respond to their attack."          Actually, of the statements in
  Our response:  No! We hoped    the new issue of Lognet, those of
that the dispute could be settlededitor Jim Smith are most of-
by negotiation, but fought at    fensive, and are indeed libelous.
this juncture because we knew ourMr. Smith accuses la lojbangirz.

                                27
with the false statements "LLG   publication in the Communications
has been around for just a few   of the ACM, a noted computer
years, but they are claiming all journal, albeit not a refereed
of JCB's work since 1955 as theirpublication.  This paper was
own ... I will not give free     finally submitted, and was
advertising to a competitor whoserejected.
primary technique is plagiarism    la lojbangirz. is considering
and whose product lacks any hint its own paper on Loglan/Lojban's
of originality."  Mr. Smith has  formal grammar, but not until
received a considerable set of   next year.
our publications and knows that    Declensions - Institute Loglan
we claim no work of JCB's as our added an 'animal' declension
own.  We have formally requested proposed over a year ago by Bob
that Mr. Smith issue a retractionMcIvor.  The change adds a large
and public apology for these     number of gismu to that version
uncalled for and unacceptable    of Loglan which differ from each
remarks.                         other only in the final vowel.  A
  Name of the language - We have broader proposal for a broader
been told that some supporters ofset of declensions, applying to
Jim Brown are offended by our useall Institute Loglan gismu, was
of "Institute Loglan" for their  never formalized, and is no
version of the language.  We havelonger being considered.
asked for an alternative other     Publications - There has been
than the generic name that would no further word of the
satisfy them, but have received  reincarnation of The Loglanist
no response.  We cannot agree to that was promised for late last
use the generic name "Loglan"    year.  Lognet has continued
only for their version - we need coming out quarterly to 'Members'
and use the term for our         of the Institute.  The May issue,
discussion of the evolutionary   indicated that the 'membership'
history of the language that     numbered 110.  The August issue
includes Lojban, and in reaching reported another 7 members were
out to people who have heard of  added.
Loglan through Robert Heinlein's   Bill Gober, in the new issue of
books or the 1960 Scientific     Lognet, criticizes the Institute
American article, and might not  use of the "lexemic pause".  This
realize that Lojban implements   is not a new criticism, and was
the language described.          one of the key language changes
  The Institute Moves - Shortly  we made in designing Lojban.
before publication, the InstituteWhat is new is that this lengthy
moved back to San Diego (actuallycriticism was printed without
Jim Brown moved - the Institute  rebuttal from JCB, who was less
proper will continue to be       open-minded on the subject four
incorporated in Florida and hold years ago.
annual meetings there).            Steve Rice is the chief word-
  "Careers" Lives - Jim Brown    maker for the Institute now.  He
reports that the board game      has made several dozen new words,
Careers, which he invented, is   both gismu and lujvo, and these
again on the market.  This       have appeared in the last couple
additional income is bolstering  of issues.  Some of these are
his capability to finance the    good, but most have the
Institute.  He indicates that    traditional failings of tanru
some money will be earmarked for (metaphors) made up by the In-
new Scientific American          stitute, incorporating English
advertising, which now costs     idiom.
$3500 for one advertisement.       Examples:
  CACM Paper - Since 1982, JCB     "criminal-quality-take", in
and others have been writing and Lojban:  "zekri ckaji lebna"
revising a paper on the Loglan   becomes the lujvo for "x1
machine grammar for intended     decriminalizes x2".

                                28
  "beyond-fast-record-use",
"bancu sutra vreji pilno" is
given to a lujvo meaning "x1 fast
forwards past x2 on record/tape
x3 on machine x4".
(The word "vreji", and its
Institute equivalent, refer to
any record of an event, not
merely sound recordings.)
                
                
          New Loglans?
                
  Nora noticed in passing that
Barbara Hambly's recent 'Star
Trek' novel, Ghost Walker,
contains several references to a
computer language called
'Loglan'.  She didn't note the
page numbers, so you will have to
read the book for yourself.
Hambly's book thus joins two
books by Robert Heinlein, one by
Robert Rimmer, and one by JCB
that mention a fictional Loglan
language.
  A new book, Loglan-88 is out,
describing Loglan, but the
language described has nothing to
do with ours or the Institute's
versions of the language.
Professor Salwicki and others at
the Polish Institute of
Informatics has been developing a
computer language for about 20
years that it calls 'Loglan'.
The newly published book reports
that the language is an ALGOL
derivative and has object-ori-
ented programming features.  The
book is published in hardbound by
Springer-Verlag; look under
'Loglan' in Books in Print for
more details.
                
_________________________________
           __________
                

                                      29
                                         Features
                                             
                                     le lojbo se ciska
                                             
            Our Lojban text will start this issue with commented sentences, and
            then a commented letter.  The sentences are offered by Rory Hinnen
            on behalf of a group of Lojbanists studying together in the Los
            Angeles area.  (If you live in this area, and wish to join this
            group, please contact Rory at 818-796-8096 (home) or 818-354-8128
            (work), or Gerald Koenig at 213-641-2905 (home) or 213-829-4156
            (work).  They are meeting irregularly, rotating the meeting place,
            because of the travel times, and offer to be very flexible in order
            to get new people to join in.
            
            In the following, Rory's submitted sentences are followed by an
            indented actual translation if it differs from his intended
            translation.  Corrected Lojban and any comments from Bob
            LeChevalier are also indented.
            
            le cmalu mlatu crane le bardu gerku
            the small-cat in-front-of thing, the large-dog, ... [an incomplete
            sentence with no selbri]
                 The small cat is in front of the large dog
                 le cmalu mlatu cu crane le barda gerku
            
            ti poi gerku cu prami le mlatu
            This which dogs, loves the cat. [This dog loves the cat.]
                 The dog loves the cat.
                 le gerku cu prami le mlatu
            
                 Bob:  "ti" should be used only when pointing.  The other and
                 more common way to say "this dog" is "le vi gerku"
                 
            .i ma prami le mlatu
            What loves the cat?
            .i ma se prami le mlatu
            What does the cat love?
                 Bob:  Excellent!
            
            mi cadzu vi le srasu
            I walk at the grass.
                 I walk on the grass.
                 mi cadzu le srasu
                 
                 Bob:  The correction uses the Lesson 4a revised place
                 structure of "cadzu", which specifies "on surface ..."  Your
                 version should be understandable, though.
            
            mi cadzu vi le srasu le ckule le ru'azda le klaji be va le zdani
            I walk on the grass to school from home via the street near the
            building.
            
                 Bob:  Using the baseline gismu list place structure of
                 "cadzu", this looks fine.  With the Lesson 4A change, just
                 replace "cadzu" with "dzukla" ("walkingly-go")
            
            do cadzu mo
            You are a walking _____.
            You are walkingly _____ing.
                 Where are you walking to?

                                      30
     do cadzu ma
     
     Bob:  "cadzu mo" is a tanru.  See above regarding "dzukla".

le ckule se cadzu mi
The school-walking_destination, I, ...
     To school I'm walking.
     le ckule cu se cadzu mi

                                      31
            *do cadzu fi mo  [Ungrammatical - no translation possible.]
                 Where are you walking from?
                 do cadzu fi ma
            le ru'azda te cadzu mi
            The assuming-nest type of walked-from thing, me, ...
                 I'm walking from home.  [Actually: "From home, walk I".
                 le re'azda cu te cadzu fi mi
            
                 I'm not sure what the lujvo is intended to mean, unless it is
                 a typo for "re'azda" ("human-nest"), which I assumed in the
                 correction.  "te" switches the x1 and x3 places, leaving x2
                 unchanged.
            
            do gasnu ma?
            What are you doing? [You do what?]
            
            mi gasnu lenu cilre la Lojban.           .i do gasnu ma
            I'm doing the event of learning Lojban.  What are you doing?
            
                 Bob: .i'e xamgu
            
            mi na cilre la Lojban.    .i mi zutse le stizu
            I'm not learning Lojban.  I'm sitting in the chair.
            
                 Rory:  (Jim [Carter] had a lot of problems with the last
                 little exchange.  He said I was taking for granted the
                 replication of the actor "mi". [in the "lenu" clause])
            
                 Bob:  He's right in that the x1 place of "cilre" was
                 elliptically unspecified.  In natural usage, "mi" would be the
                 obvious value, but formally the translation is:  "I'm doing
                 the event of ... learning Lojban.  What are you doing?".  The
                 most frequent assumption in the pattern [x1 broda <lenu ...
                 brode>] is to assume that the x1 is replicated in the
                 ellipsis.  But if the listener is unsure, it/she/he can always
                 ask:  ".i ma cilre la Lojban".
            
            ko catlu  .i le crino nanmu
            Look! The green man ...
                 Look!  Green man!
                 ko catlu  .i crino nanmu
            
                 Rory: (Jim made me put the "le" in there, and then convinced
                 me he was right.)
                 Bob:  He's wrong.  The "le" made it an incomplete sentence, a
                 bare sumti, leaving the listener hanging for the rest of the
                 sentence (.oi).  There is no implicit or explicit selbri.
            
            ko sisti .i mi na catlu le crino nanmu
            Cease!  I don't look at the green man.
                 Rory: We argued for a while "le" or "lo" in the above
                 sentence, but we eventually came to the conclusion that it
                 didn't matter because of the negation.  But without negation,
                 I should go with "lo").
                 
                 Bob:  Either could be correct with or without a negation.
                 "lo" claims that it really was a green man, rather than
                 possibly a picture of one.  Usually English speakers use "lo"
                 for indefinites.  "le" is definite - you DO have a specific
                 one in mind, but it might only be being described as a man for

                                      32
     convenience of conversation.  In this context, "le" would
     normally be taken as referring to the green man of the
     previous observative, since that is the logical 'thing
     described' that the speaker could expect the listener to
     assume.
          Note that I can't be sure of the intended meaning of this
     sentence for another reason, given the context.  What is
     supposed to be "ceased", the discussion, talking about the
     green man, or something else?  The second sentence also loses
     me, although it is quite grammatical.  Is it a mistake for "I
     don't see ...", or is it missing the attitudinal ".aunai" (I
     don't want!).

                                33
    Is Lojban Scientifically     about that system to a scientist.
           Interesting           I can see a few other
                                 possibilities:
  David Pautler                  
(pautler@ils.nwu.edu), challenged  a) in a highly complex system
the scientific relevance of      (which even an AL is), the
artificial languages.  The       interaction of the design
following is lojbab's (Bob       features displays properties that
LeChevalier's) response.         are 'more than the sum of the
                                 parts'.  Thus it is possible that
David wrote:                     all language is merely a system
  I did not say that ALs have no comprised of a bunch of neurons
good use.  I said there's nothingreleasing neurotransmitters.
particularly interesting about   Biochemistry may eventually
them (from a scientific viewpointdevise a complete explanation for
...) because they're artificial. the neuronic process (including
Some interesting sociological    genetic components), and we may
behaviors may appear if these    then say we "know the design
languages come into widespread   principles of the system".  But
use, perhaps even some in-       we won't know the system, because
teresting linguistic phenomena ifthe complexity of those neuronic
enough spontaneous innovation    interactions is so great that
occurs (although AL enthusiasts  knowing the pieces does not give
seem to want to prevent this).   a total understanding of the
But there certainly doesn't      system.  This indeed may be what
appear to be anything interestingdefines the concept of 'system'.
about them now, because AL         Knowing all the prescribed
enthusiasts in this group prefer rules of an AL does not tell you
to argue over which of several   how that AL is used communica-
(truly arbitrary) conventions aretively, and I don't mean in the
"better".                        sociological sense.  A sample
  I am willing to admit I am     question:  Given multiple ways of
wrong about all this if some of  communicating the same idea, do
you AL enthusiasts can give the  users of the language choose
rest of us some good reasons why particular forms over others, and
ALs are scientifically           why?  This is similar to a
interesting.                     question that presumably is
                                 commonly asked about natural
and later added in clarification:languages.
                                   I can come up with many other
I still believe that knowing the sample questions of science that
design principles of any system  can be applied to the system of
beforehand makes a scientific    an AL that are not compromised by
study of those principles        'knowing the design', but let's
silly...                         move on.  (Feel free to ask,
                                 though).
lojbab's response:                 b) A simpler system, which can
                                 be more fully understood, may
  The added comment definitely   serve as an excellent model for a
clarifies the problem, especiallyless understood, more complex
since it removes the loaded topicsystem.  Thus the simpler system
'AL' from the question.  I will  could be examined for parallels
answer primarily from the        to hypotheses about the more
standpoint of Lojban, though somecomplex system.  Examination of
of my points are applicable to   the simpler system may suggest
Esperanto and other ALs.         properties to look for in the
  David is taking a very limited more complex system, or it may
view of science, to presume that even suggest hypotheses that can
the design principles of a systembe tested in the more complex
are the only interesting thing   system.

                                34
  A 'hot' topic in parts of the    c) Another aspect of a simple
Lojban community is whether the  system is that it is easier to
language has or should have, an  perform experiments on than a
underlying semantic theory.  If  more complex system.  There are
one exists, it is certainly not  fewer variables, and if the
as developed or prescribed as thesystem is 'designed', some things
syntactic design and theory.     that are variables in complex
Filtering out syntactic ambiguitysystems are in effect tunable
allows a more direct examination constants in the simple,
of semantic ambiguities, in-     carefully-designed system.  You
cluding the properties of        can then rerun the experiment
modification and restriction,    with minor changes to explore the
resolution of anaphora, and      effects of those variables.
identification of ellipses.  Any   Experimental linguistics of
semantic theories proposed for   this sort is a virtually
natural language can be looked atunthinkable possibility with the
in terms of semantic usage in thenatural languages.  The Sapir-
simpler Lojban system.           Whorf Hypothesis is not really
  As a 'model of a natural       testable in the natural languages
language', it seems likely that  since we can't control any
any theory NOT true of Lojban is variables, and we don't know what
at least suspicious with regard  things about a language might be
to natural language, thus        determining to a culture.  Sapir-
allowing partial verification of Whorf may be more testable when
theories (not complete - I would you can reduce or even control
never say that ALs should be     the variables with a language
studied to the exclusion of      like Lojban.  Let me be specific:
natural languages, but rather in   Lojban is a predicate language,
relation to them).  If the theorywith no nouns, verbs, or
is true of natural language, thenadjectives.  What are the lin-
you have found evidence that     guistic (communicative)
Lojban is in some way unnatural. properties of such a system?  The
Then you try to explain which of answer has been partially
the (fully-known) design featuresexplored through symbolic logic.
of Lojban causes this un-        But do people thinking
naturalness.  By counterexample  linguistically in any way mimic
that design feature is not a     the processes of formal logic?
feature of natural languages.    What effects would a formal-
  You've learned something about logic-based language have on
natural language by studying an  those linguistic thinking
artificial one.                  processes.  Is the resulting
  As another example, pragmatic  language susceptible to the same
effects can be more easily       analysis as natural language in
recognized in the simpler Lojban terms of the various formal
system, and can be clearly       systems that have been developed
identified as pragmatic.  Thus,  by linguists over the past few
insights about pragmatic effects decades?
may be more visible in Lojban,     Given that natural language
insights that would then be      processing in computers usually
tested in the natural languages. involves converting natural
                                 language to some kind of
                                 predicate form in which
                                 deductions can be made, the
                                 validity of predicate logic as a
                                 tool for such analysis is already
                                 accepted.  But how to you
                                 identify the logical deductions
                                 that a human being makes from a
                                 natural language statement.  If
                                 thinking in Lojban, the human is

                                35
already thinking using predicate instead of the construct.  It
logic structures; thus the       seems that all manner of
deduction process is much more   linguistic universals could be
plain.                           investigated in this way.
  Let me pose an experiment.     
Take even a few children during    My remaining points are not
the critical period of language  necessarily specific to the
learning and teach them this     'system' nature of a language,
artificial language (at the same but deal with David's original
time as they learn their         question on whether artificial
traditional language).  Do they  languages are scientifically
become truly bilingual?  If they interesting.  In general they
are as fluently communicative in rely on the assumption argued
the AL as they are in their      above that a model of a system is
natural language, then the AL is valuable for learning about the
a suitable linguistic model.     system.
Then, ANY theory of language that
cannot extend to cover the fea-    d) I've mentioned only child
tures of the AL is inadequate.   learning as revealing the
You could perform a series of    essential nature of language,
experiments with ever more exoticbecause this is what many
artificial languages (obviously  linguists concentrate on.  But
you need new speakers for each   there is also the important
test).  Sooner or later, either  applied linguistics problems of
the model breaks and the AL is noteaching foreign languages.  It
longer acquirable by children    is much easier to test a method
and/or communicative as a        or theory of vocabulary
language, or the theory breaks,  teaching/learning with an
and you've learned where to look artificial language than with a
for improvements in the theory.  natural language; I don't think
  With only natural languages,   the statement that ALs are more
you have to devise theories basedquickly (I didn't say easily -
on the available data, and then  which is a subjective question)
go look in other natural         learned then NLs is particularly
languages for confirmation or    controversial; there have been
refutation.  But this isn't the  experiments verifying this in the
optimal kind of experimentation  literature for decades.
because you really cannot plan     The pragmatic problems of
the experiment or control the    language learning are alone
variables (the other language mayjustification into researching
have the same apparent feature   using ALs.  But ALs may provide
through a totally different pro- the solution as well as the means
cess that you won't recognize    of testing.
because you aren't looking for     It seems to be well accepted
it.)                             that in learning a second
  A language like Lojban is such language and then learning a
an ideal test bed for            third, you learn the third MUCH
experimentation, because it is   more quickly than the second.
flexible; you can evolve slightlyThe example I've heard is this:
different versions of the        
language very easily by simply        Assume that it takes 4 years
changing some features.  Forbid a  to learn French and then 2 to
given construct in the prescrip-   learn German thereafter; and
tion, and do not teach it to a     vice versa.  Let us assume that
child.  Does the child develop     you can learn an artificial
that construct anyway by analogy   language in 1 year to a
to other languages known, or does  comparable degree as you can
the child successfully adapt to    learn French.
whatever other processes you've       Then you can learn the AL
designed into the language         and German in 3 years instead

                                36
  of 4, and all three languages    I don't claim this example as a
  in 5 years instead of 6.  This fact - it should be easily
  gains a year EVEN IF YOU NEVER testable in a controlled experi-
  AGAIN USE THE AL.              ment, and this seems much more
                                 scientific than arguments about
                                 what ALs and NLs are 'easier to
                                 learn'.
                                 
                                   e) Lojban has one feature
                                 designed to explore a less-
                                 understood aspect of language -
                                 the expression of emotion.
                                 Lojban allows expressive
                                 communication of emotions in
                                 words without suprasegmentals
                                 (this presumably unlike all natu-
                                 ral languages, but not entirely,
                                 as many languages have a limited
                                 set of indicators of attitude in
                                 the form of interjections and
                                 some discursive function words
                                 e.g.  'but').  Can human beings
                                 manipulate the symbols of emotion
                                 in the same way they manipulate
                                 the comparable symbols of non-
                                 emotional expression?  There is a
                                 whole range of experimental
                                 questions raised by this design
                                 element, probably the most
                                 'unnatural' element of Lojban's
                                 design.
                                 
                                   f) The latter points to the one
                                 other aspect of a well-designed
                                 artificial language of scientific
                                 interest and value to linguistics
                                 - as a tool of analysis.
                                   I present an example, based on
                                 the 1991 Scientific American
                                 Library book The Science of
                                 Words, by George A. Miller of
                                 Princeton.
                                      In the book, a picture
                                 caption notes that Nootka (a
                                 Pacific Northwest language) has
                                 the single word:
                                 "inikwihl'minik'isit" meaning the
                                 equivalent of the entire English
                                 sentence "Several small fires
                                 were burning in the house."  I
                                 won't presume to know any more
                                 about Nootka than I've just told
                                 you, but in Lojban, I can express
                                 that sentence paralleling the
                                 English:
                                 
                                 so'i cmalu fagri puca      jelca
                                 vine'i    le
                                 Many small fires were-then
                                 burning at-within the

                                37
  prezda                           This order is also expressible
  person-nest.                   in Lojban:
                                 
and analytically as a single word  fagykemprezdanerso'icmapru
(though not with the same          fire-type_of-person-nest-
structure as Nootka)               inside-many_some-small-
                                   past_thing/event
  prezdane'ikemcmafagyso'ikemprun
  unje'a                           I don't know which of the two
  person-house-inside-type_of-   orders more accurately conveys
  small-fire-many_some-type_of-  how the Nootka speaker thinks of
  previous-burning               the concept expressed by the
                                 word, or whether others would be
(Yes, I can say it!)             better still.
                                   The Lojban in either case more
  Actually, according to Miller, accurately tracks the semantics
the Nootka breaks down as:       of the Nootka, demonstrating the
                                 inadequacy of the English - the
inikw     -ihl         -'minih - actual word as broken out did not
'is        -it                   require two separate particles
fire/burn in-the-house plural    for fire and burn as did the
diminutive past-                 English equivalent, and the
                                 English translation used the more
tense                            complicated tense "were-burning"
                                 instead of the simpler, and
                                 presumably more accurate "burnt".
                                 (I'll plainly admit that I'm
                                 relying on the given explanations
                                 by Miller, which are in English,
                                 but it seems clear that in
                                 translating the word-sentence
                                 into English there is a
                                 considerable ambiguity
                                 introduced.
                                   I won't claim that Lojban can
                                 express everything in the natural
                                 form of any language.  Lojban has
                                 a less-marked syntactic word-
                                 order, and expressing other
                                 orders requires marking particles
                                 that would not be found in the
                                 source language.  Thus there is a
                                 tradeoff between semantic
                                 representation and syntactic
                                 representation.
                                   Still, I think a convincing
                                 case can be made that, as a
                                 predicate language, Lojban is a
                                 much more effective tool at
                                 studying both the forms and
                                 semantics of other languages than
                                 is English, which has its own
                                 cultural, syntactic and semantic
                                 complexities to gum up the
                                 analysis.  This is especially
                                 true for analysis by non-native
                                 English speaking linguists - if
                                 there is any place where there is
                                 a justification for an interna-
                                 tional, minimal-culture language,

                                38
it is when linguists from        how the language creolizes in
different native language        contact with those other
backgrounds try to perform and   languages.  Because of the speed
communicate their linguistic     of learning, artificial languages
analyses.                        should tend to show effects more
                                 quickly (by being mastered to a
  g) There is also the 'other'   communicative level more
tool aspect of an artificial     quickly).  Anecdotal evidence
language, in computer and artifi-about Esperanto supports this
cial intelligence (AI)           idea.
applications.  I mentioned the     Does this mean that the
similarity in c) above between   conclusions are absolutely valid
Lojban and the internal          for natural language evolutionary
representations used in natural  processes?  I don't claim so.
language processing by computers.But again, we are performing
  A predicate language like      experiments with a model,
Lojban should be especially      somewhat idealized, of a natural
amenable to AI processes - the   language.  Unlike a paper-
programmers are familiar with    theoretic model (as all
predicate language expression andlinguistic theories must
manipulation, and often store theinherently be), this is a model
data in predicate form internallythat can be experimented with
for manipulation.  With Lojban,  using live speakers.  Provided
such storage becomes a fairly    that we understand the model as
trivial process.                 it evolves, that understanding
  If Lojban is proven by         much more approximates an under-
experiment (per above) to have   standing of natural language as
the systemic properties of a     time goes on.
natural language, and is easier  
to implement in computational      i) The large majority of
linguistics research problems, itlanguages have some degree, more
serves as a tool to bridge those or less, of prescription.  In
two disciplines, leading to more addition, some 'natural'
rapid and effective natural      languages, like modern Hebrew,
language processing.  But only ifformal Swahili, and some
it is tried.  Even if it proves  standardized dialects (e.g.
less than ideal, I have little   Mandarin, which has been noted as
doubt that study of natural lan- being related to but not
guage using computational        identical to the Beijing
linguistic techniques and a      dialect), are not all that far
Lojban-based tool will be        from being true artificial
productive in ways not possible  languages, but are much more
with any natural language.       interesting to linguists.  A
  (In effect, this argument is   predominantly prescribed language
the same as f), except that      would seem an especially
instead of two different-natural-effective tool for studying the
language speakers trying to      effects of prescription on lan-
communicate about language, you  guage development and use (again,
have a human and a computer, who I refer to linguistic and not
obviously speak different native sociological effects).
languages, trying to               Such studies may aid in first-
communicate.)                    language education as well as
                                 second-language acquisition.
  h) A highly prescribed languageThey may also aid in analyzing
is an ideal test bed for         the development of different
examining the processes of       registers (usages based on social
language evolution.  In the case class and situation) of a single
of an AL like Lojban, as the     language:  such registers can be
speaking community in each       interpreted as reactions to
culture grows, you can observe

                                39
prescriptive environments that   5. Add "hemp", to include natural
constrain language use.            rope, burlap, marijuana, and
                                   hashish - "marna";
  None of these scientific       6. Add "protein" - "lanbi";
applications of Lojban inherently7. Add "buckwheat" - "xruba";
requires a large fluent body of  8. Add "cassava", to include taro
speakers, or any solely-native     and yam, and other starchy
speaker of that tongue.  If any    roots (not tubers) - "samcu";
of the less scientific applica-  9. Add "sorghum" - "sorgu";
tions of Lojban serve to justify 10.-11. Add "magenta" and "cyan"
it developing such a speaker       as the missing two subtractive
base, the nature of Lojban's use-  primary colors - "nukni",
fulness as a model will change.    "cicna";
New applications, as yet not     12. Add "North America", the
really predictable, will turn up,  continent, as distinct from
aided by our no doubt increased    "merko", referring to the U.S.
understanding of language.  But    - "bemro";
the model, even if well under-   13. Add "South America", the
stood, no longer is as simple,     continent, as distinct from
and new Loglans and other          "xispo", referring to Latin
experimental linguistic tools,     America - "ketco";
all artificial languages, will be14. Add "Antarctica" - "dzipo";
developed to take the next step. 15. Add "glimmering" to cover the
  I have hopefully given a bit of  concepts of morning and evening
food for thought, yet with only a  twilight - "murse";
few hours preparation.  I also   16. Add "decrease" in parallel to
only thought about this as         a revised meaning of "increase"
somewhat an outsider to the        - "jdika";
profession of linguistics.  With 17. Add a different gismu to be
a different point-of-view others   the inverse of "panzi" -
should be able to find many more   "rorci";
questions of scientific interest 18. Add "elder/ancestor" for
using an AL like Lojban either as  family members of generations
a model, an experimental test      preceding the parents
bed, or a tool.  And if even a     (including non-direct line, the
small fraction of these ideas are  relationship is more so-
useful, then ALs have a valid      cial/ethnic than biological).
scientific role in linguistics.    Gender would be added via
                                   tanru, as would explicit
_________________________________  biological lineage.  The
         ______________            conversion would give "de-
                                   scendant" as well as
 Summary of gismu/rafsi Official   "grandkids" in the broadest
             Changes               sense - "dzena";
                                 19. Add "aunt/uncle/godparent"
New gismu as approved in June      for non-lineal (socio-ethnic)
1990 (see JL13):                   family members of the parental
                                   generation.  The conversion
1. Add "daytime", changing the     would give "niece/nephew" -
  keyword for "day" to "full day"  "famti";
  - "dinri";                     20. Add "cousin" for non-
2. Add "virtue", as distinct from  immediate (socio-ethnic) family
  "good", to parallel with "evil"  members of the same generation.
  - "vrude";                       The generalized family
3. Add "citrus" - "nimre";         relationship is still expressed
4. Add "cabbage", to include       by "lanzu", which can be
  broccoli, cauliflower, and       modified via tanru - "tamne".
  perhaps lettuce - "kobli";     

                                40
The following shows the new gismuketco   ket tco         South
actually made, which may be addedAmerican
to your gismu lists.                                     Quechua
                                   x1 reflects South American
gismu   rafsi           keyword    culture/nationality/geography
                        clue /     in aspect x2
synonyms                         
                                 kobli                   cabbage
bemro   bem             North                            cole- /
American                         cauli-, broccoli,
                        berti                             kale,
merko                            kraut
  x1 reflects North American       x1 is a quantity of
  culture/nationality/geography    cabbage/leafy vegetable of
  in aspect x2                     species/strain x2
                                 
cicna                   cyan     lanbi                   protein
                         /                               albumin /
turquoise                        amino
  x1 is cyan/turquoise/green-blue  x1 is a quantity of protein of
                                   type x2
dinri                   daytime  
                         /       marna                   hemp
daylight                                                 marijuana
  x1 is a daytime of day x2 at   / jute,
  location x3                                             cannabis
                                   x1 is a quantity of
dzena       dze         elder      hemp/marijuana/jute of
                         /         species/strain x2
grandparent, ancestor            
  x1 is an elder/ancestor of x2  murse
  by bond/tie/ degree x3         glimmering
                                                          /
dzipo   zip dzi zi'o             twilight, dawn,
Antarctican                                               penumbra
                        dzucipni   x1 is a twilight/dawn of day x2
(penguin?)                         at location x3
  x1 reflects Antarctican        
  culture/nationality/ geography nimre       mre         citrus
  in aspect x2                                           lime /
famti                   aunt or  lemon, citric
uncle                              x1 is a quantity of citrus
                         /         (fruit, tree, etc.) of
godparent                          species/strain x2
  x1 is an aunt/uncle of x2 by   
  bond/tie x3                    nukni   nuk             magenta
                                                         fuchsin /
jdika                   decrease fuchsia
                         / reduce  x1 is magenta/fuchsia/red-blue
  x1 is decreased/reduced in     
  property x2 by amount x3       rorci   ror             procreate
                                                          /
                                 engender, sire, dam,
                                                          beget
                                   x1 engenders/procreates/begets
                                   x2 with coparent x3
                                 
                                 samcu                   cassava
                                                          / taro,
                                 manioc, tapioca,

                                41
                         yam     xruba   xub             buckwheat
  x1 is a quantity of                                    rhubarb /
  cassava/taro/manioc/tapioca/yamsorrel grass
  (starchy root) of                x1 is a quantity of
  species/strain x2                buckwheat/rhubarb/sorrel grass
                                   of species/strain x2
sorgu   sog             sorghum  
  x1 is a quantity of sorghum of The following are rafsi changes
  species/strain x2              needed for the above changes.  In
                                 both cases, these are words that
tamne                   cousin   had been given an extra rafsi
  x1 is a cousin of x2 by        "because it was there", not
  bond/tie x3                    because it was needed.
                                 
vrude   vud     vu'e    virtue   dotco   dot     do'o German
  x1 is virtuous by standard x2  delete rafsi tco
                                 merli   mel     mei  measure
                                 delete rafsi mre
                                 
                                 The following is a previously
                                 approved change to the gismu
                                 baseline.
                                 
                                 ckamu-               less
                                 deleted, replaced by:
                                 mleca   mec     me'a less
                                 
                                 The following are previously
                                 approved changes to the gismu
                                 keyword baseline.
                                 
                                 mukti   muk     mu'i    motive
                                                         was
                                 purpose
                                 djedi   ded dje dei     full day
                                                         was day
                                 gismu   gim     gi'u    root word
                                                         was
                                 primitive
                                 tanru                   phrase
                                 compound
                                                         was
                                 metaphor
                                 lujvo   luv jvo         affix
                                 compound
                                                         was cpd
                                 predicate
                                 
                                   Several cmavo were newly
                                 assigned rafsi, but in some cases
                                 where a rafsi might be useful,
                                 there was no rafsi even remotely
                                 similar to the cmavo available.
                                 The following lujvo-making con-
                                 ventions are proposed as
                                 solutions for these cmavo used in
                                 lujvo.  Note that there may be
                                 lujvo with these patterns that
                                 are not convention-based
                                 

                                42
cu'o           preceded by number    dzi         dzipo
            rafsi is a               mre         nimre
            probability rather       mre-        merli
            than modification        tco         ketco
fi'u           with number rafsi     tco-        dotco
            is interpreted as    bem       bemro
            fraction rather than biz       bi'o or bi'i
            modification         caz       ca'a
ka'e           handled by kakne  caz-      cadzu (retains dzu)
li'i        lifrysucty-  in firstcel       ce
            position             cez       ce'i
mu'e        mulnynun-    in firstdaz       da'a
            position             dum       du'u
pu'u        prucynun-    in firstjoz       jo'e
            position             kep       ke'e
si'o        sidbysucty-  in firstket       ketco
            position             kuz       ku'a
su'u        sucty-  in first     mec       mleca
            position             mem       mei
za'i        zastynun-    in firstmom       moi
            position             nal       na'e
zu'o        zuktynun-    in firstnal-      na
            position             nar       na
                                 nuk       nukni
  The following is then a summarypez       pe'a  (figurative lujvo
of ALL changes to the rafsi list             - totally
since the gismu baseline list was            unpredictable place
published.  You may wish to manu-            structure)
ally update your gismu lists and piv       pi'u
rafsi lists.  New lists will be  piz       pi
published by the end of the year.puz       pu'i
                                 ror       rorci
( - after a rafsi means the rafsirov       roi
was deassigned for reuse.)       sog       sorgu
                                 soj       so'a
CVC CCV CVV     cmavo/gismu      sop       so'e
                                 sor       so'i  (also has "so'i"
        ce'o    ce'o                         added this one is for
        co'e    co'e                         making a series)
        fo'a-   forca (retains   sor-      skori (retains sko,
            for, fro)                        ko'i)
        fo'a    fo'a             sos       so'o
        fo'e    fo'e             sot       so'u
        fo'i    fo'i             vud       vrude
        ke'e    ke'e             xub       xruba
        le'e    le'e             zip       dzipo
        lo'e    lo'e             
        me'a    mleca              I received a question about the
        no'e    no'e             listing in the rafsi change
        nu'o    nu'o             summary about conventions for
        to'e    to'e             some abstractions.  So let me
        ve'e    ve'e             explain further.
        vu'e    vrude              For the example, we will use
        za'o    za'o             "salci", which has the place
        ze'e    ze'e             structure "x1 celebrates x2 by
        ze'o    ze'o             doing/being x3".
        zi'o    dzipo              "nu salci" is an abstraction
        zo'a    zo'a             selbri: "x1 is the event of (x1S
        zo'i    zo'i             celebrates x2S by doing x3S)".
    dze         dzena

                                43
  To make this into a lujvo, we  be representing the abstraction
have reserved a rafsi for 'nu',  since there may be more obvious
and in this case the lujvo-makingnon-abstraction meanings.  The
algorithm gives "nunsalci":  "x1 convention however tells you how
is the celebration-event of x2   to make a lujvo that will at
celebrating x3 by doing x4".  Allleast suggest the abstraction
we do is renumber the places.    rather than force you to try to
  We also have rafsi reserved forguess randomly how to do so.  In
'ka', 'ni', and 'jei' among the  general, the convention is close
abstractors.  We do not have     to the most obvious
rafsi for the four Aristotelian  interpretation anyway.
subcategories of 'nu' events:      As for nunsalci, the places of
'states', 'activities', 'process-the unabstracted selbri are
es', and 'achievements'.  Nor do renumbered shifted by 1. For a
we have rafsi for other members  couple of abstractors (ni and jei
of NU, including the idea ab-    come to mind) with more than one
stractor "si'o", the experience  place in the abstraction selbri,
abstractor "li'i", etc.          the trailing places are added to
  The convention described says  the end.
that for those members of NU that
do not have rafsi, we will use a   This convention is not part of
combination of rafsi for related any baseline, and is quite open
gismu in specific positions      for your comments.  Indeed, we'd
(generally at the beginning of   like to see some people try
the lujvo) to indicate the       making lujvo using them and using
abstraction.  We don't assume    them in example sentences.
that all lujvo in that form will
         _________________________________________________
                                 
                   le lojbo se ciska (continued)

Coranth D'Gryphon sent me a letter - an excellent beginning writing
effort, which I will share with all of you since it brings out some
neat points.  I'm sure Coranth's ego will stand public correction,
and all-in-all he did an outstanding effort even if I've commented
it to death.  I encourage beginning people to write to each other
in Lojban, preferably with English translation enclosed.  Send your
first writing to me (Bob) at the la lojbangirz. address, and I'll
match you with someone based on your skill level (and possibly your
interests).  A suggestion for a first letter is either a list of
sentences or a self-description.  There is NO BETTER WAY to learn
the language than to see how others misunderstand your attempts to
express in the language.  And if communication actually occurs,
you've achieved the purpose of language, which is more important
than being correct.  If you can figure out most of what Coranth was
TRYING to say on the following (there are errors and even typo-
nonexistent words that can make some parts virtually impossible to
a novice), you are ready.

Following is Coranth's uncorrected original text and his
translation.  My comments follow.

coi doi lojbab.

le lojbo ckule cu xamgu
.i di'u cu pluka mi
.i re le ci tardi cu djica troci le nu la Lojban cu se djuno
.i ji'a .i'ou'anai mi pu skami ciska piso'i le ti fasnu tebe'i do
.i mi'u pu fonxa tavla


                                44
.i la lojbo gerna cu frili mi
.i le lojbo valsi na go'i

.i mi cu troci le nu la'edi'u cu se djuno
.i .au.a'o mi baze'e djuno roru

                                45
.i paupei la logflac. cu kakne le sidju di'e .inaja jei di'u
.i pe'u.e'o ko cu tavla mi le pu'u cpacu ko'a

.i ju'e ki'e co'o lojbab.
                                        la korant.

Hi Bob

The Lojban-school is good.
This pleases me.
2 of the 3 students desire-ingly try-to-attain the-state-of Lojban
being-known.

I regret/sorrow-at I have-before not computer-written much-of these
events to-you.

Ditto have-before not telephone-talked.

Lojban grammar is easy for me.
The-set-of Lojban words is not.

I try-to-attain the-state-of them-previous being-known.
I hope-desire-that I will-after-some-time know them-all.

Question-opinion : Logflash is-able-to help?

The-following only-if the-previous-sentence is-true:
Request-please : you (imper) talk-to me about the-process-of
getting them.

I conclude. thanks. bye.
                                                Coranth
___________________________________________________________________
_________________________________

Commentary.  The unindented text and translation are Coranth's
versions:

coi doi lojbab.
Hi Bob

le lojbo ckule cu xamgu
The Lojban-school is good.

     Both excellent.  [Coranth, by the way was attempting to tutor
     some students in Lojban in the Boston area, which is another
     group that is ripe for group study as the LA people are doing.

.i di'u cu pluka mi
This pleases me.

     You translated the Lojban to English correctly, but the
     English is ambiguous.  A more exact English translation of
     your sentence is:
     
     The last sentence pleases me.

     Now you have a right to be pleased about the last sentence:
     it was well done.  But I suspect that you were trying to

                                46
     express that the state of affairs described by the last
     sentence is what pleased you, in which case you wanted to say:

     .i la'edi'u [cu] pluka mi.

     It is probably useful to memorize "la'edi'u" as a single word;
     you will use it in an English translation far more often than
     "di'u" alone, unless you are writing essays about language.


                                47
.i re le ci tardi cu djica troci le nu la Lojban cu se djuno
2 of the 3 students desire-ingly try-to-attain the-state-of Lojban
being-known

     Except "tadni" for "tardi", excellent.  I explicitly check or
     have Nora check all of my writings before sending them out to
     make sure I don't make word mistakes.  Until you master the
     vocabulary, you should do an extra check whenever possible.
     Of course, though, I would rather you write Lojban without
     checking the words, than not write Lojban at all.

.i ji'a .i'ou'anai mi pu skami ciska piso'i le ti fasnu tebe'i do
I regret/sorrow-at I have-before not computer-written much-of these
events to-you.

          A great teaching sentence.  It was grammatical and I
     understood what you intended, but you made lots of little
     errors.
          You left out translating the "ji'a" ("additionally"), and
     I'm not sure how it applies; if you are trying to non-
     specifically link this sentence to the previous one ".ije" is
     a logical joining, and ".i" is the simple run-on sentence
     "and".
          The attitudinal "appreciation+loss" does not convey to me
     your English - what did you "lose"; it sounds like the type of
     emotion one might feel when a trusted and valued employee left
     the company, or maybe what you might say in a letter of
     resignation.  Possibilities are the simple ".u'u", or
     ".i'anaise'i" (or attaching the ".i'anai" differently:  "mi
     .i'anai pu ..." which means roughly: "I, damn me, previously
     ..."
          Your English sentence is a negation - you have NOT
     PREVIOUSLY written much-of these events, but this negation is
     not in the sentence.  You could do so with "na", but I would
     prefer the more exact "punai".
          "ciska" is "inscribe"; you clearly noticed that its place
     structure wasn't that useful and used "tebe'i".  "cusku" is
     usually more applicable; writing is just a medium (which could
     go in the "ve cusku" place if it was important).  "tavla" is
     also useful here; it is not limited to verbal expression
     (which is "bacru"), and it has a 'talkee' too.  Finally, as
     implied by your "tebe'i", you could have used "benji" with "le
     datni/se skicu be le fasnu" or more simply "le fasnu datni".
          "piso'i" is a fraction of a whole.  If you are treating
     the events as separate reportable instances, you wanted "so'i
     le fasnu" = "many of the events".  If you wanted to talk about
     them as a mass, you wanted "piso'i lei fasnu" = "much of the
     mass of events".  Probably the latter is better.
          "ti" as a demonstrative does not work well in letters,
     and in any case I doubt if you could point to the "ti" that
     'possesses' the events.  "ti" is a sumti and in that position
     "le ti fasnu" it meant "this thing's event(s)".  For "this" in
     such a sumti you usually want the locator "vi":  ("le vi
     fasnu" = "the here event").  Now it turns out that a different
     possessive might actually have been appropriate here:  "le
     de'u fasnu" = "the earlier sentence's events", or possibly,
     "le ru fasnu" which in your writing can only refer to "the
     Lojban class"; therefore "the Lojban-class's events".  Of
     course, "ru" alone might have been clear since you would have

                                48
     been writing about the class as well as the events associated
     with the class.
          Adding all these comments together, I might have written
     the sentence as:
     
     .ije mi .i'anai punai .uu skami cusku le de'u fasnu do
     And I, (Damn me!) didn't (Regret!) computer-express the
     earlier sentence's events to you.

.i mi'u pu fonxa tavla
Ditto have-before not telephone-talked.

          Again missing the negation, only it is more blatant here.
     You said that we have talked.
          You don't say what we haven't talked about.  Since you
     and I have telephone talked before, you are putting a lot of
     semantic load on that "mi'u" to convey that you want me to
     transfer the subject of talking from the previous sentence (as
     well as the other places).
          I like "je" or "ji'a" here instead of "mi'u" though it
     seems as interesting concept to use a "ditto" discursive to
     copy the sumti from the previous sentence, while changing the
     selbri.  Hmmm.
          "je", with no discursive, linking a specified-sumti bridi
     with an observative, does strongly suggest the sumti copying.
     I do this a lot, according to Nora.  There are also more obvi-
     ous forms, based on a compound sentence using "gi'e", but I
     won't go into that here.
          My attempts:

     .ije punai fonxa tavla
     And didn't telephone talk.
     OR
     .ije punai fonxa tavla do le de'u fasnu
     And didn't telephone talk to you about the earlier sentence's
     events.

          You also could have combined the last two sentences into
     one, but the negated tense makes the translation apparently
     illogical for English speakers, who implicitly convert the
     selbri from "and" to "or" under DeMorgan's Law.  This is not
     recommended until people can truly think in Lojban:
     
     .ije mi .i'anai punai .u'u skami je fonxa cusku le de'u fasnu
     do
     And I, (Damn me!) didn't (Regret!) computer-(or)-phone express
     the earlier sentence's events to you.
     
.i la lojbo gerna cu frili mi
Lojban grammar is easy for me.
.i le lojbo valsi na go'i
The-set-of Lojban words is not.

     You are correct that "na" does not require "cu".  Excellent.

.i mi cu troci le nu la'edi'u cu se djuno
I try-to-attain the-state-of them-previous being-known

     "la'edi'u" here makes your sentence mean:
     

                                49
     I try-to-attain the-state-of (the-set-of Lojban words is not
     [easy for me]) being-known.
     
     You wanted "ri" to get your translation:
     .i le lojbo valsi na go'i .i mi troci lenu ri cu se djuno.

.i .au.a'o mi baze'e djuno roru
I hope-desire-that I will-after-some-time know them-all

     I corrected for publication what I presume to be a typo
".ava'o" in your original.

     You wanted "ze'eba" instead of "baze'e", which means here "I
     will for some interval in the future know all of them."
     Examine the parallel examples in "tense*" in the cmavo 'Lexeme
     order list', which are based on "pu".  "ra" is more correct
     than "ru" here; counting backwards in your version, "ri" is
     "la'edi'u" and "ra" is "le lojbo valsi".  In my rewritten
     versions, "ri" is still "le lojbo valsi".
     
     (See the following article for details regarding this
     comment.) Given the new place structure of "djuno", I think
     you want "tu'a roru".)

.i paupei la logflac. cu kakne le sidju di'e .inaja jei di'u
Question-opinion : Logflash is-able-to help?
The-following only-if the-previous-sentence is-true:

          This one was fun.  I had to write most of the following
     before I knew what you were trying.
          "paupei" isn't wrong but the "pau" is superfluous.  "pau"
     is used to let the listener reader know that a question is
     coming up later in the sentence that might not be expected,
     forewarning that attention is needed so that the answer can be
     provided.  When the question is the next word, the warning is
     redundant, but not wrong.  I think that your desired question
     was probably ".iapei" or "pe'ipei", asking about belief or
     opinion.
          I'm sure you wanted "le nu sidju di'e".  Otherwise "le
     sidju" ("the helper") is the x2 place of "kakne" and "di'e" is
     the x3 place.  But as for what it means:
          Your second sentence isn't grammatical; you may have
     wanted "di'u jetnu" based on your English.  But more pressing
     is that you have one awesome self-referential sentence loop
     here, and I can't honestly say I understand the English any
     better than the Lojban.  It's often a good idea to put a
     colloquial English translation along with a literal one if the
     structure is convoluted as in this case.
          I'll rephrase your English to reflect what you said
     (making the corrections already noted):
     
     .i la logflac. cu kakne lenu sidju di'e .inaja di'u jetnu
     Logflash is able at helping the following sentence.  [Only if]
     the previous sentence is true.
     
          Does this make the problem clearer?  You have a logical
     connective that makes some funky claims about the truth
     conditions of the combined pair of sentences.  (I think you
     get a tautology of a sort:  ".inaja" here effectively causes
     "not X or X", where "X" is the "di'u" of the second sentence.

                                50
     If the first is true the second must be; if the first is
     false, so is the second.)  Then the second sentence is talking
     about the truth of the first sentence, while the first
     sentence is talking about Logflash helping with the second
     sentence.  [Sounds of mental explosion as circuits are fried
     .oicairo'e].

     .ua.ue ki'anai [I think I finally understand!]
          I think I figured out now what you were trying to do (but
     only after all that analysis.  My clue is how you grouped your
     English with a line separating:
     
          Question-opinion : Logflash is-able-to help.
          
          The-following only-if the-previous-sentence is-true:
          Request-please : you (imper) talk-to me about the-
          process-of getting them.
     
          This makes it clear that "di'e" is supposed to be part of
     the same sentence as the "di'u", and actually refers to the
     sentence afterwards.  At which point all becomes clear:
          ".i" is an almost perfect sentence terminator.  It says
     that the following sentence is about to start, making all of
     the 'elidable' terminators of constructs shorter than sentence
     scope (i.e.  "vau", "kei", "ku", etc.) actually elidable at
     the end of a sentence.  BUT the ".i" can ONLY go between
     sentences.  It seems you were trying to use ".inaja" as the
     selbri.  But the "di'e" is irrevocably part of the first of
     two sentences in this case, and I merely spent my effort
     trying to figure out how it fit in.  But then, that analysis
     pointed out the need for "lenu sidju" vs. "le sidju" - which I
     think is what you wanted; I might have missed it if not for
     the hanging "di'e".
          At which point I can say GOOD TRY, especially since we
     nowhere cover logical connectives in text materials.  But:
     
     The members of "GOhA", "me"+"KOhA", and PA+MOI, are the only
     cmavo or cmavo compounds that come to mind as being valid as a
     selbri (there may be some complex equivalents of these, too.
     Lojban logical connectives connect constructs; they are not in
     themselves selbri.  If you want a predicated connective, you
     need a gismu or tanru.  We have provided "kanxe", "vlina",
     "dunli" and "nibli", and I think the latter is what you want
     here:
     .i pe'ipei la logflac. cu kakne lenu sidju .i di'e se nibli
     di'u
     (Opinion-question?) Logflash is able at helping.  The
     following is entailed by the preceding.
     
          (Nora would stop here; she likes short simple sentences,
     but I'll muck things up a little more.)  At which point we can
     actually eliminate the 2nd sentence entirely, using a causal
     connective:
     

                                51
     .i .pe'ipei la logflac. cu kakne lenu sidju
     .iseni'ibo pe'u.e'o ko cu tavla mi le pu'u cpacu ko'a
     
     which translates as:
     Question-opinion : Logflash is-able-to help?
     Necessitating therefore: request-please: you (imperative)
          talk-to me about the-process-of getting them.
     
          It is even possible to compress this to the single
     sentence (in which case the "pau" is useful, too):
     
     .i pau la logflac. cu kakne lenu sidju pe'ipei kei seni'i lenu
     pe'ue'o ko cu tavla mi le pu'u cpacu ko'a
     Question follows:  Logflash is-able to help (Your opinion
          please), which (the ability) would necessitate therefore
          (I request-please) that you (imperative) talk-to me about
          the-process-of getting them.
     
     Note the "kei" to terminate the "lenu" clause.  Without it,
     the now-more-complex sentence means something only roughly
     similar (note the angle brackets):
     
     Question follows:  Logflash is-able at <helping (your opinion
          please) thus necessitating therefore (I request-please)
          that you (imperative) talk-to me about the-process-of
          getting them.>
     
          There is no problem asking a Lojban question about a
     sentence while exploring the logical consequences of its
     truth.  Lojban presumes that, logically speaking, the truth
     value of a question is the same as that of the sentence with
     the question satisfied (the blank filled in, etc.)  Thus you
     can ask my opinion on the first sentence, while telling me
     what to do if it is true.  Note that you have to move the
     "pe'ipei" question itself around when you try to ask it all in
     one sentence, so that I clearly know that you are asking about
     Logflash's ability to help, and not whether (its ability to
     help necessitates talking about getting it).
          One other comment.  Just as "LogFest" Lojbanizes poorly,
     so does "LogFlash".  "*gf" is not a permissible medial pair in
     Lojban, since "g" is voiced and "f" is unvoiced.  Lojbanize it
     as "logyflac." or "logvlac."
     
.i pe'u.e'o ko cu tavla mi le pu'u cpacu ko'a
Request-please : you (imperative) talk-to me about the-process-of
     getting them.

          An excellent non-trivial concluding sentence, with only
     two minor flaws.  "pe'u" is a vocative member of COI, and
     expects a name or description afterwards - you can't quite use
     it like an attitudinal, unless you close it with "do'u".
     Without the "do'u", the vocative absorbs "ko" as the target of
     "pe'u", and you have approximately:  "Request-of-you
     (imperative), that (observative:  someone unspecified) talks-
     to-me about ...". This still gets your point across.  However,
     with the "ko" absorbed, there is no sumti to separate, and the
     "cu" is not needed or allowed.  Secondly, "ko'a" is undefined.
     Presuming that you mean Logflash, I would simply repeat "la
     logyflac."  Alternatives are the vague "ra" or assigning
     "ko'a" with "goi" - a waste for one reference.

                                52
     
     As an answer to your letter, call or write again regarding
     getting LogFlash.  I of course believe it is helpful - it is
     the ONLY reason I can lojbo cusku with any skill.
     
.i ju'e ki'e co'o lojbab.
I conclude. thanks. bye.
                                        la korant.
                                        Coranth

     You wanted "mi'e korant." for complete grammaticality.
Overall an excellent first effort.  It better than others argues
for some explanation of logical connectives at an early stage.  I
will modify my textbook outline as a result.  Keep it up!

                                      53
            Cleft Place Structures and sumti-  Thus, I'll explain some of the
                         Raising             problem, give simple examples of
                                             a couple of the points which you
              A minor excursion in "how to   may be able to use and general-
            say it in Lojban" turned into a  ize, but otherwise will just try
            major philosophical examination  to explain the changes.  If you
            of language and metalanguage (howdon't understand just yet, don't
            we talk about language) this     worry - we carried on
            spring.  The result of this      conversations in Lojban for a
            endeavor was a series of minor   couple of years without even
            word changes -  mostly additions noticing the problem.
            of cmavo, one minor grammar        (Difficulty warning:  this
            change, and a couple of major    material in places relies on all
            philosophical realizations about of the contents of the draft
            language that shook us to the un-textbook lessons.  You may not
            derpinnings and will have a      understand everything that
            profound effect on how we teach  follows without them, but I've
            the language.                    tried to make the explanation
              Unfortunately, we can't take   independent of your knowledge of
            the space here to discuss the    Lojban as much as possible.)
            question in depth, including the   You may know that Lojban has at
            various rationales for decisions least two major kinds of sumti
            made and not made.  The          (arguments) in its bridi (pre-
            discussion would be as long and  dications).  Only two are
            intricate as the negation paper  relevant here.  I will call these
            published last year, only more   'simple' and 'abstract' sumti.
            confusing.  Instead, we'll try to  A simple sumti is comparable to
            outline what was decided,        what in English are 'common
            emphasizing effects on Lojban    nouns' - objects that you can
            expression.                      point to.  Examples include "le
              English does not have very     stizu" ("the chair"), "le zarci"
            effective tools for talking about(the market).  But because Lojban
            language.  You have to teach a   doesn't distinguish nouns, verbs,
            whole specialized vocabulary for and adjectives, we can also have
            any aspect of language - a       "le blanu" ("the blue thing"), or
            vocabulary that reads like jargon"le kurji" ("the one taking care
            without a great deal of          of ...") as simple sumti.  All of
            explanation.  Classic problems   the examples so far are what we
            are how to describe the meaning  call 'descriptions' in the Log-
            of words like "of" and "the", or lan/Lojban project.  In a
            how to describe the meaning of a description, a selbri (the
            form of the verb "to be" combinedpredicate word or phrase that
            with the suffix "-ing" on a verb defines the relationship) is
            without merely giving an example.converted into a sumti, omitting
              However, for those questions,  the x1 place, using a descriptor
            you at least know the answer,    word like "le" or "lo".  The de-
            even if you can't easily phrase  scription then refers to some-
            it.  Topics like 'indirect       thing intended that would fill
            questions' (which are not really that x1 place.  Thus "le klama"
            questions at all), and 'object   is something that would fill the
            raising' (sumti raising when     x1 place of "x1 comes/goes to x2
            generalized in Lojban) are topicsfrom x3 via x4 using mode x5".
            for linguistic researchers.        An abstract sumti looks and
            Everyday people use these        acts differently.  In an abstract
            linguistic features all the time sumti, you take a whole bridi
            without realizing it AND, even   predication (including the x1),
            realizing it, find it difficult  i.e., a whole sentence, and turn
            to paraphrase and explain what   it into a sumti.  That sumti then
            they are doing, and why they     represents the abstract state or
            understand what it means.        event of the relation occurring

                                      54
("nu"), the characteristic prop- many ways.  An abstraction has
erty(ies) of that relation       the abstract marker from NU on
("ka"), or any of several other  the front and an elidable
abstractions, including "du'u" (aterminator "kei" on the back, to
predication about the relation), keep the language unambiguous,
and "jei" (the truth value of    allowing you to know whether a
such a relation).  These others  selbri is part of the abstraction
may be found in selma'o NU in    or is the main selbri of the
your cmavo list.                 sentence, or whether a sumti is a
  What is hidden in most usage ofsumti of the main sentence bridi,
these abstract sumti is that we  or of the abstraction bridi.
have created an entirely new       These descriptions and abstrac-
selbri relationship encompassing tions are long, possibly complex
the abstracted bridi and its     in grammar, and generally a pain
places as the selbri.  For most  to repeat when you are saying a
abstractions, this new selbri haslot about them.  So we have
only one place, though "jei",    symbols or 'anaphora' to stand
which talks about truth values,  for them.  You may be familiar
has an x2 place for epistemology,with "ko'a" which can be assigned
and "ni", and amount, has a      to represent any sumti, whether
scale.  When you use one of theseabstract or simple, as well as
abstractions in a sumti, you are "mi" ("me") and "do" ("you").
again filling in the x1 place of All of these are 'anaphora' -
one selbri, but at a higher, morewords that stand for something
abstract level than for a simple previously defined or obvious
sumti.  Thus there is a parallel from the context.  There are a
between these different levels oflot of others.  One other kind of
sumti such that both are tied    anaphora is names.  When you use
back to a bridi relation with onethe name "lojbab." for me, the
unspecified place.               name represents me for discussion
  An example of an abstract bridiin a sentence.  "la lojbab." is
is:                              thus grammatically equivalent to
                                 "do" and "le nu mi klama le zarci
x1 is the event ("nu") of        [ku] [vau] [kei] [ku]".
  <xk1 comes/goes (klama) to xk2   We can also use names as
  from xk3 via xk4 using mode    anaphora for events and other
  xk5>                           abstraction sumti.  "The
                                 Renaissance" is a name for an
In "le nu klama", the "le" means important historical period, and
that we are talking about the x1 "The Battle of Bunker Hill" is
just defined, just as "le" means the name of an event.
for simple sumti.  The other       By equating simple sumti and
sumti may or may not be          abstract sumti grammatically, we
explicitly expressed, but the    achieve some of the power of Loj-
grammar is that of a full bridi  ban's grammar.  Lojban allows the
terminated by the elidable       manipulation of both types of
terminator "kei":                arguments using its grammar as
                                 predicate logic does - you ignore
le nu mi klama le zarci kei ku   the representation when
[le nu <mi klama le zarci kei>   manipulating the symbols.
ku]                                The risk for human speakers is
The event of my going to the     the same as the advantage:  you
market ...                       may ignore the representation
(As a lujvo, "nunklama", the 6   when manipulating the symbols.
places would be renumbered x1    If you forget that you are
through x6, hiding the two levelsworking with abstractions, you
of grammatical structure.)       can end up mixing levels of
  The parallel effect of "le"    abstraction.  The result is
allows us to grammatically treat nonsense.  In natural language,
these two kinds of sumti alike inwhen we speak nonsense, the

                                      55
            listener tries to make sense of  that would be sumti in Lojban.
            it, and intuitively ignores      The relations that would be the
            errors of abstraction level,     selbri in Lojban are
            giving understanding that may    traditionally represented by a
            ignore logical errors.  We want  single word root ("gismu") in
            to avoid this in Lojban.         Lojban; respectively these are
              Some examples.  Here are some  "djuno" ("know"), "mulno"
            relations expressed as English   ("complete"), and "galfi"
            sentences:                       ("modify").
                                               In each pair of sentences, one
            (1)                              relates a concept using an
            It is good.  (It representing    abstract sumti, and the other a
            "the cat")                       concept using a simple sumti.
            It is good.  (It representing    But notice:  it is the "actor" of
                           "the long romanticthe event in the abstraction sen-
                           walk to the park")tence that serves as the simple
                                             sumti in the other sentence.  In
                                             (4) you may need to look at the
            (2)                              parenthesized versions to see
            Mary considered it. (It          this.
                           representing "the   In English, we typically
                           cat")             interpret both sentences as
            Mary considered it. (it          meaning the same thing.  But how
                           representing "the can this be - is an abstraction
                           long romantic walkreally the same as the actor of
                           to the park")     that abstraction?  Is it "I" that
                                             turns the water into steam, or is
              (1) serves to remind that      it "my boiling the water" that
            abstract sumti and simple sumti  turns it into steam?  It cannot
            are equivalent in Lojban bridi.  be both using a single definition
            You need to be able to manipulatefor "turns it into steam".
            them using their symbols, without  And indeed, the result is that
            worrying about what the symbols  the meaning of "turns water into
            represent, or whether you end up steam", "is done", and "knows
            with nonsense, as in (2).  At    about" in English have at least
            some point, however, you want to two meanings, and we figure out
            interpret "it", and if "it"      which one applies based on
            represents something illogical incontext.
            the context, you want to           Linguists say that in such
            recognize that you have nonsense.cases, we have in effect 'raised'
                                             the simple sumti out of the ab-
            I know about John.               straction and are using it to
            (3)                              represent the abstraction.  This
            I know about John sleeping with  feature is called 'subject-rais-
            Susan.                           ing' or 'object-raising' when
                                             used to describe English and
            The cooking is done.             other natural languages.  We call
               (My cooking something has     it sumti-raising when talking
            completed.)    (4)               about Lojban, which does not
            I'm done cooking.                distinguish between subjects and
               (I have completed the cook-   objects.
            ing.)                              Notice that Lojban can make it
                                             clear that there are really two
            I turn the water into steam by   distinct place structures
            boiling it.  (5)                 involved when you have a sumti-
            My boiling it turns the water    raising.  For the examples, we
            into steam.                      have:
            
              (3) through (5) are pairs of
            sentences with two 'arguments'

                                      56
x1 knows about x2 (an actor)     resulting in apparent non-se-
  being the actor in doing x3 (anquitors:
  activity abstract like         
  'sleeping with')]              I know about John that the dog is
x1 knows about x2 (an abstract   sleeping with Susan.
event or fact)                   I am done with you doing the
                                 cooking.
x1 (an event) is done/complete.  I turn the water into steam by
x1 (an actor) is done being the  John boiling Susan.
actor in event x2                
                                   These appear to be nonsense,
x1 (an actor) modifies x2 into x3but the human mind attempts to
  by being the actor in          make sense of them anyway, pos-
  event/process x4               sibly concluding that John is
x1 (an event/process) modifies x2literally or figuratively a dog,
  into x3                        that "my being done with you
                                 cooking" means that I will not
  The problem is explicit becausetolerate it any more and will eat
so much of Lojban semantics is   out instead, with "done" being
embedded in the place structures.interpreted figuratively.
For a variety of reasons -         Another example out of recent
logical integrity, ease of       events that shows the logical
learning, etc., we want to have  errors that can result from this
only a single place structure for(courtesy Art Protin):
each Lojban word, and we want to 
know what goes into each place.    Saddam Hussein modified the
  We have given a label to bridi borders of Iraq to include Kuwait
place structures where one of theby Iraqi soldiers invading
sumti places is defined to be an Kuwait.
actor (or some other place) in an
abstraction sumti found elsewhere  Here we have an abstract cleft-
in the same bridi.  We call thesestructured sentence (it uses the
'cleft place structures'.        same structure as "I turn the
  Even if there were no other    water ..."), where the actor,
reason, we dislike cleft place   Saddam, does not appear in the
structures because they are      abstraction.  To interpret this,
repetitive and redundant.  This  we can jump to all manner of
becomes evident in completely    conclusions that are in some way
expressed Lojban sentences, for  logically faulty.  We relying on
which I will give English        hidden assumptions to pull
equivalents:                     meaning out of the statement, as
                                 we try to decide whether Saddam
I know about John that John is   or his soldiers were the
  sleeping with Susan.           aggressors against Kuwait.
I am done with my doing the        For example, we may rely on the
  cooking.                       main predicate as pre-eminent,
I turn the water into steam by myinterpreting the statement as if
  boiling the water.             the Iraqi soldiers were Saddam
                                 Hussein, or else his direct
  Note that the last example has agents, puppets, or tools and not
two redundant sumti, "I", and    thinking and feeling human beings
"the water", thus showing that   capable of independent choice:
sumti-raising is not limited to  
'actor' places.                    The Iraqi soldiers are not
  Now, if you look at the Lojban   responsible for attacking
for these, you realize that it is  because they had no choice.
perfectly acceptable to put a      
different value in one of the two  Alternatively, we think of
supposedly equivalent places,    "invade" as the active verb
                                 making the soldiers the actual

                                      57
            'actors', while Saddam remains a completed it?  Is "le galfi" a
            nebulous motivational force (who modifier, or the modification
            actually 'did' nothing):         process?  You clearly want to be
                                             able to somehow access the actor,
              The Iraqi soldiers are         since he/she/it is likely to be
              responsible, because Saddam    used in a sumti.
              merely gave orders and they      In older versions of Loglan,
              were obligated to disobey an   there were many problem words of
              immoral or illegal order (the  this sort.  Jim Brown basically
              NЃrenburg judgement).          argued that place structures
                                             should be what is 'natural' for
              The converse of the first      speakers, including all
            version (seen from Saddam's pointinformation that is needed to
            of view perhaps) sees Saddam as  determine the truth conditions of
            actor, and the soldiers as       the relation.  Both of these
            impersonal tools that failed in  place structures include the
            their function:                  information, so he typically
                                             chose the more English-like
              Saddam is not responsible for  version of the place structure.
              Iraq's losing the war because  This led to all manner of subtle
              his soldiers failed him.       difficulties.  Since the actor is
                                             specified in one place, then also
              We thus end up depersonalizing in the event sumti, you typically
            either the soldiers or Saddam;   will elliptically omit the actor,
            they cease to be thought of as   as well as other places.
            real people because our          
            instinctive language use wants to?mi mulno
            recognize only one active agent    le  nu       [mi] jukpa   [le
            in an abstract statement.        cidja]       (6a)
            Carrying such statements to their I  complete
            illogical conclusion can           the event-of [me] cooking [the
            depersonalize any aspect of the  food].
            war:                              I finish cooking.
                                             
              The allied coalition forces    ?le  nu       mi jukpa   [le
              should/should-not punish the   cidja]
              soldiers (or Saddam, or the       cu mulno
              people of Iraq:  take your     (6b)
              choice) because they were/were- The event-of me cooking the food
              not responsible.                  is-complete.
                                             
              (I deliberately chose a current  The second sentence is often
            and controversial topic because  'shortened' in a couple of other
            opinion molders, whether         ways in colloquial English:  "I'm
            government or media, 'honest'    done cooking." and "The food is
            writers or propagandists, use    done cooking".  (6b) reveals that
            just such illogical arguments andin English we are merely con-
            hidden assumptions to convince   densing the abstract event in a
            readers of their point of view,  different way, by ellipsizing a
            often with deadly consequences.  different sumti of the event
            I intend none of the above       bridi "my cooking the food".  The
            expressions to be taken as being result was haphazard,
            mine or la lojbangirz.'s opinionsinconsistent place structures.
            on the matter.)                    It is important to realize the
              Another problem occurs when youhistorical roots of this problem,
            turn one of these words with     because they constrain the
            possibly cleft structures into a solutions.  The choice of gismu,
            simple description sumti.  Is "leand indeed the grammar of Lojban
            mulno" an action that is         itself, evolved from an earlier
            complete, or the actor that      version of Loglan.  That version

                                      58
did not recognize sumti-raising  a space as a pause (there is no
as a feature of the language, andsymbolic representation in
tended to obscure abstractions   Institute Loglan that a pause is
much the way we do in English,   required in "le po blanu".
based on Brown's concept of      Finally, Brown introduced a
'natural' place structures.      "poge" construction to make a
  In older versions of Loglan,   'long-scope' abstraction for use
most words had an actor in the x1with trailing arguments and
place.  When trying to express   logical connectives:  "mi viska
one of these bridi using the     le poge ta blanu" is a possible
given place structures, one      construction, though one never
naturally ellipsized the first   used because it is identical to
sumti of the abstract event,     "mi viska lepo ta blanu".  The
which was just a repetition of   web of spaces and "ge"s made a
the actor, just as in (6a) above.mess of the grammar description,
However, JCB apparently did not  especially since both are used in
recognize that the result was    other ways in the grammar as
logically identical to one with  well.
the abstract actor filled in, and  When we started Lojban, we
instead built the Loglan grammar rebuilt the grammar on our own.
considering an abstract with el- At first, we merely copied the
lipsized x1 actor as a totally   existing Loglan structures.  But
separate grammatical construct.  in 1989, we started teaching the
This construct has no basis in   language.  In what is now Lesson
logic and caused all manner of   3 of the draft textbook lessons,
ambiguities in Loglan,           we attempted to explain Loglan
ambiguities that were solved by  abstraction.  Lojbanists from
cheating in the Loglan machine   before March 1989 may remember
grammar.                         examples from that time using
  Institute Loglan STILL has this"*le nuke", which exactly matched
problem, which I'll describe for "le poge"; we also had "nu" and
those comparing the two versions "lenu" as distinct constructs.
(otherwise skip this and the nextThe widely distributed 'February
two paragraphs).  That version   1988' machine grammar contains
uses the word "po" where Lojban  these fossils.
uses "nu".  In Jim Brown's         While trying to explain
versions of Loglan, "po" ("nu")  abstractions, we demonstrated
does NOT change the nature of a  that there was only one real
bridi, as I discussed above.  "leconstruct involving "nu" and that
po blanu" was a simple sumti:    was a bridi turned into a selbri.
"le (po blanu)", where "(po      John Parks-Clifford (pc), noted
blanu)" is a description for of athat he and others had unsuc-
selbri.                          cessfully argued for this in the
  To get an abstract sumti, you  1970's with Jim Brown.  pc also
write "lepo" as a single word,   had discussed cleft place
which the computer parser would  structures with Brown, but with
then treat as a totally differentno resulting change.
selma'o (grammatical category)     In 1989, both the abstraction
than "le", turning a whole bridi grammar and cleft place
predication into a sumti:  "lepo structures issues came up
(ta blanu)".  But a human being  separately.  The abstraction
can't tell "lepo" from "le po" inproblem was resolved as described
normal speech, so Jim Brown in-  above, and pc's reasoning on the
troduced an arbitrary rule that  cleft place structures was
to separate the two words, you   convincing; we changed most place
had to pause between them - a    structures where x1 was the actor
"lexemic" pause.                 and x2 was an event to a single
  Computer languages often use   place.  tanru (metaphors) and
spaces to avoid ambiguities, and lujvo (complexes) were used to
Jim Brown was in effect treating access the actor.  "mulgau"

                                      59
            (mulno gasnu) is thus the actor  raising' from subject-raising and
            form of "complete" in current    object-raising.
            Lojban.                            Start with the English
              But the problem is not really  sentence:
            resolved.  We missed several     
            cleft place structures, and have I try the door.
            discovered them intermittently   (1)
            while doing the place structure  
            reviews.  "galfi" was a recent   Without recognizing sumti-
            discovery, with the x1 place     raising, we would translate this
            cleft from the x4 event.         into Lojban as:
              Meanwhile, in teaching Lojban, 
            we have found that when native   mi troci le vorme
            English speakers trying to speak (1a)
            Lojban guess at place structures,
            they presume actor forms of the  based on the place structure of
            words.  Thus some people,        "troci"
            (including those of us with more 
            experience) use "mi mulno" for   x1 tries to do/attain x2 by x3
            "I'm done", usually with humorous(1b)
            results when called on it.  (In  
            Lojban, "you" are presumably not which expands to:
            "done" until the "event" repre-  
            sented by the word "you" is com- x1 tries to bring about the
            plete; i.e. when you are dead, or  event/state/ process/activity
            even later, in some religions.)    x2 by method x3         (1c)
              Thus the problems:  how hard do
            we try to eliminate cleft place  (1c) clarifies that x2 is an
            structures and how do we solve   abstract clause.  I then view "le
            the natural language habit of    vorme" is a sumti-raising from
            sumti-raising while preserving   one of two possible x2 abstract
            Lojban's logical character?      sentences:
                                             
                      The Solution           mi troci lenu      mi karyri'a
                                             le  vorme    (2)
              The solution has evolved over  I  try   the-event I  open-cause
            the last year or so in several   the door.
            stages.  Only one grammar change 
            is involved, and that expands    mi troci lenu      le  vorme cu
            capabilities rather than changes kalri         (3)
            any existing constructs in the   I  try   the-event the door
            language.  In developing a       is-open.
            solution, we ended up running    
            into multiple distracting side-  (There is further possible
            issues, ranging from the place   complication in (2) in that rinka
            structure of "djuno", to         (cause), the basis of "karyri'a"
            'indirect questions', the changeswould normally take an event
            that resulted will also be       abstraction in its x1 place, thus
            listed.                          making the "mi" in that position
                                             a sumti-raising as well.  (2)
                      sumti-Raising          thus can be further expanded to:
                                             
              In Lojban, there is no         mi troci lenu
            difference between 'subject' and   lenu mi lacpu/catke le vorme
            'object', because free rearrange-  cu rinka
            ment of arguments (sumti) using       leka le vorme cu kalri
            conversion with selma'o SE can       /lenu le vorme cu kalri
            lead to any sumti being in the   
            first 'subject' position.  The   mi troci (lenu
            proposal thus generalizes 'sumti-I  try   the-event

                                      60
   <lenu      mi lacpu/catke le  abstracted bridi which is
vorme>                           ellipsized, much as we leave out
    the event I  pull /push  the unimportant trailing places.  We
door                             have assigned the cmavo "co'e" to
   cu rinka                      represent such an unspecified,
   causes                        normally ellipsized, bridi.
      <leka            le  vorme Thus, (1a) becomes:
cu kalri>)  (4a)                 
       the-property-of the door  mi troci tu'a le vorme
open(-ness)                      
       the door's openness.      which is equivalent to
      <lenu         le  vorme cu 
kalri>)     (4b)                 mi troci
       the-event-of the door     (5)
is-open                          I  attempt
     the door opening.              lenu      le vorme cu co'e
                                    the-event the door
  Clearly, much information is   being/doing something.
lost or hidden in sumti-raising -
we don't know in (2) whether     (To be formally correct, we
pushing or pulling the door is   should use "lesu'u" to get an
necessary (or hitting the        unspecified abstraction instead
elevator door button, for that   of "lenu")
matter).  Thus there clearly is a
lot of semantics hidden in "le   By looking back at (4a) and (4b),
vorme" in original sentence (1a).we see that "co'e" in (5) is
We want to mark this explicitly. actually the equivalent of the
  We want in Lojban to discourageEnglish:
unnecessary sumti-raising because
it is logically sloppy.  However,"being opened by my
sumti-raising allows brevity and pushing/pulling it".
a 'natural' feel to the language.
Clearly, (4a) and (4b) are too     We will urge that when people
much to ask of a speaker who is  speak Lojban, that they try to be
thinking (1) - that she/he merelyaware of the possibility that
wants the door open, and it      they may be sumti-raising, and
doesn't matter how.              mark it with "tu'a".
  We've thus accepted that sumti-  Sloppy speakers, and new
raising must be allowed in the   Lojbanists, will sometimes fail -
language.  But it is most im-    you have a lifetime of habit to
portant that people recognize    overcome.  Thus a listener may
when they are sumti-raising, and choose to assume that an unmarked
mark it so that the listener can concrete sumti in a place that
then allow for it, interpreting  normally takes an abstraction
'what the speaker really meant'  really is intended to be a sumti-
as being something involving an  raising - choosing to understand
unspecified abstraction.         (1) as meaning (5).  If this
  We therefore will explicitly   happens too extensively, however,
mark sumti-raising using the     the logical nature of Lojban is
cmavo "tu'a", which is assigned  compromised - thus we will not
to selma'o LAhE (thus not        teach unmarked sumti-raising as
requiring a grammar change).     valid, and will discourage it or
LAhE is the category of 'indirectcorrect it when we notice it.)
markers' that tell the listener    sumti-raising solves the
that the sumti as spoken is only problem of cleft place
an indirect symbol for what is   structures.  It allows uncleft
actually intended to fill the    place structures to mimic cleft
place.                           ones for user convenience.  A
  Any time there is a sumti-     sentence claim like:
raising, there is an implied     

                                      61
            lenu          mi cinfai cu galfi inherently defined to be cleft,
            le  bitmu                        with the actor/ agent extracted
            The event-of (I  paint)          from the action.  There is no
            modifies the wall                meaning to English "do" that
                                             avoids a cleft structure.
            corresponding to the place         "zukte" also has a mandatory
            structure:         (6)           cleft structure since an action
                                             with goal requires an 'actor' to
            x1 (an event/action/state)       adopt that goal.
            modifies x2 into x3                "lifri" is the corresponding
                                             (also cleft) word for a
            means the same as the cleft      passive/patient/experiencer.
            sentence:                        "Actor/ agent", "goal", and
                                             "passive/patient/experiencer" are
            *mi galfi  le  bitmu fo lenu     terms used in case theory
            mi cinfai                        semantics for the basic semantic
             I  modify the wall  by the      roles in a sentence.  It is
            event-of (I  paint)              likely that other places that
                                             correspond to such basic semantic
            based on place structure:        roles may have remain cleft, if
            (6a)                             only to support continued efforts
                                             to regularize Lojban semantics.
            *x1 (an actor/agent) modifies x2   We may find that a couple more
            into x3 by doing/being x4 (an    words must have cleft structures
            action/state)                    due to the inherent mental state
                                             of an actor that must be
              But if the act of painting is  identified to evaluate the truth
            irrelevant or obvious, and all   of the predication.
            you need to communicate was that   "ckaji" and "klani" are cleft
            it was YOU who changed the wall, because they express the basic
            then "tu'a" makes the ellipsis   semantic relation of a property
            possible for the uncleft         abstract to the thing having the
            structure (6), and makes the     property, and an amount abstract
            resulting sentence quite brief:  to the thing being measured.
                                               Having mentioned "gasnu", it us
            tu'a      mi   galfi    le  bitmuworth noting that there is
            (7)                              another way to extract an actor/
            Something I do modifies the wall.agent from an abstraction clause
                                             in an uncleft place structure.
              The consensus of Lojbanists is This other method is to use "gau"
            that uncleft place structures arefrom selma'o BAI, which is
            logically cleaner and are more   derived from "gasnu" and is tied
            concise, hence the preferred way to that word's actor-extracting
            to go in Lojban.  Thus, we are   cleft place structure.
            trying to identify and eliminate   Using the above example, we can
            as many cleft place structures asthus say:
            possible.                        
              We will not eliminate all of   gau        mi
            them.  In some cases, the cleft  (8)
            structure is inherent to the        galfi    le  bitmu
            meaning of the concept.  The mostwith-agent me (some-x1-event-
            noteworthy of these are "gasnu", unspecified)
            "zukte", "lifri", "ckaji", and      modifies the wall
            "klani".                         
              One place structure change that  This gives the same effect as
            has been formally adopted is the sumti-raising, but is more clear
            clarification that gasnu means   as to the role of 'me' in the re-
            "x1 is the actor/agent in doing  lationship.  sumti-raising need
            event/process/activity x2".  In  not always involve raising an
            other words, "gasnu" is          'actor/agent'.  Given that the

                                      62
unspecified abstract selbri is   The classic argument used by gun
cinfai, we could easily say:     control advocates, "Guns don't
                                 kill people; people kill people",
tu'a        lei blanu cinta      relies on just such confusion of
(9)                              abstraction levels.
   galfi    le  bitmu              It is thus important to
            The blue  paint      understand that there is no
doing/being something            explicitly marked link between a
            (i.e. being spread)  "gau"-added agentive place, and
   modifies the wall.            the ellipsized x1 sumti.  In
                                 causality statements like "Guns
 or even:                        don't kill people; people kill
                                 people", the claim about the
tu'a        le  bitmu            agent of a change may be indepen-
(10)                             dent of the event that physically
   galfi    le  bitmu            causes, motivates, justifies, or
            The wall  doing/beinglogically entails the result.
something                          "gau"-agents are thus logically
            (i.e. having paint   inaccurate but semantically clear
spread upon it)                  about the role of the marked
   modifies the wall.            sumti.  sumti-raising is more
                                 vague about the semantics of the
  "tu'a" is thus very vague and  "tu'a"-marked sumti, but more
relies on the speaker trusting   precise logically.  The only way
that the listener will be able toto be precise in both aspects is
determine from context what the  to explicitly identify the
hidden abstraction is and role   subordinate abstracted bridi.
the raised sumti plays (i.e. to    Put still another way, BAI
understand what the speaker meansclearly specifies the semantic
without her/him being explicit). relation between the sumti and
"gau mi" on the other hand says  the rest of the bridi, while
that I'm actually and actively   labelled sumti-raising clearly
doing something to bring about   specifies that hidden ellipsis is
the relationship, and we thus    present.  Both methods are a kind
would be surprised by "gau lei   of ellipsis, and both have a role
blanu cinta" or "gau le bitmu".  in the language.  But let it be
  On the other hand, "gau" takes recognized that only explicit
more thought than "tu'a".  The   elucidation of the hidden el-
reason the speaker might use     lipsis gives a complete
"tu'a" sumti-raising is to save  statement, just as explicit
time and mental energy for       elucidation of all places in a
communicating the important      bridi place structure makes a
instead of the 'obvious'.        bridi more complete.
  The "gau" form works           
differently from "tu'a", since it  All sumti are created equal.
is not actually sumti-raising.   
The x1 place of "galfi" remains    Let us suppose that you want to
unspecified.  By avoiding that   refer to the agent who modified
difficult place in the place     the wall in a sumti, rather than
structure, it weakens that place in a full bridi.  A mother says
structure.  More importantly, useto her child:  "whoever modified
of "gau" may indicate the        this wall (by putting paint on
speaker's failure to recognize   it) will be punished".  Until
the hidden logical structure -   this recent set of changes, there
that I am agent in a subsidiary  was no direct way to do this.
event (the painting) rather than However, in the one grammar
necessarily the agent in "galfi".change, introduced, we have added
  Ah, but aren't they the same   selma'o JAI, which only has the
thing?  Probably yes, in this    word "jai" in it.  Following is
case.  But in others, not so.

                                      63
            the approved change and an       abstraction sumti, one uses
            example.                         either explicit sumti-raising or
                                             implicit raising using a BAI
            PREVIOUS RULE:                   modal tag.  JAI-based 'modal
                                             conversions' allow description
            Official doctrine states that thesumti access to modal tag sumti-
            sumtcita of a bridi constitute   raising, as in:
            nonstandard places which are co- 
            equal with the regular numbered  le  jai gau  galfi     be le
            places.  However, there was no   bitmu
            way to make these places the     The actor-in modifying    the
            subject of a description by      wall.
            moving them into a numbered      
            (specifically, the x1) place.    
                                                           djuno
            APPROVED CHANGE:                 
                                               The place structure of "djuno"
            Add JAI+sumtcita as another      has been much-debated.  The
            variety of SE conversion.  (JAI  problem is made more difficult
            is a new selma'o.)  This is      because English combines at least
            usable on selbri in descriptions four different concepts in "to
            or main selbri, not in the other know" which are often broken out
            places where SE is legal (logicalinto separate words in other
            connectives, modals, etc.).  The languages.
            result is that the tcita sumti
            comes to occupy the x1 place, and
            the original x1 place is switched
            by the conversion to an un-
            numbered place which can be
            accessed with the cmavo "fai"
            (selma'o FA).  All other places
            numbers remain unchanged as in SE
            conversions.
            
            To make room for this usage, the
            current "fai" and "fi'e" (selma'o
            GOhA) are changed respectively to
            "nei" and "no'a".  (There is more
            explanation of these words
            below.)
            
            RATIONALE:
            
            It is currently messy to say "the
            time of my going to the store";
            this looks like an abstraction,
            but does not match any existing
            abstractor.  It can be handled
            quite neatly with:
            
            le                   jai ca
                 klama be le  zarci bei fai
                 mi
            the (thing which is)
                 simultaneous-with going to
                 the store     by  me.
            
            In addition, when attempting to
            access a place (such as an agent)
            that is actually found within a

                                      64
  These are:                     they contain a 'question word'
- "to understand something"      and imply a question that was or
  ("jimpe")                      could be asked, but are not
- "to be familiar with something"really questions.  They often
  ("it is na'e cnino to me")     involve knowledge, hence the
- "to know in general about      place structure of "djuno" fig-
  something" ("I know            ured heavily in the discussion.
  arithmetic")                   An English example:  "I know who
- "to know a specific fact is    went to the store."
  true" ("I know that '2+2=4'").   The term 'indirect question' is
  The latter two can be related  actually somewhat of a misnomer;
in a single gismu, and "djuno"   all Indo-European languages
represents those types of        overlap use of relative pronouns
knowledge.  The new place        with question words, and all use
structure for "djuno" recognizes these 'wh- words' in 'indirect
that the 'subject' of knowledge  questions'.  Grammarians could
(x3) may or may not be at a      also call them 'indirect
broader level than the knowledge restrictive clauses', but this
itself (x2), and that knowing a  would never sell.
fact (x2) entails knowing it in    The problem in Lojban is that
some larger context (x3), as wellwe cannot translate these using a
as according to an epistemology  question word, or there is an
(a means of knowledge, e.g.      ambiguity:
deduction, observation,
authority, etc. x4)).  Thus the
current place structure proposal
for "djuno" is:

x1 knows that abstract statement
x2 (a 'truth') about subject x3
is true under epistemology x4
                
Examples:
                
mi djuno zo'e
lei cmaci
I  know  some-fact(s) unspecified
about math
[by some epistemology].

mi djuno ledu'u         li vo
sumji li re   li re
I  know  the-assertion     4
sums     2 [and] 2
[about some subject by some
epistemology]
                
mi djuno
     fo le nu la .iuklid.
lojycipra
I  know something about something
     by          Euclid's
logical-proof.
                
                
       Indirect Questions

  There are a variety of kinds of
'indirect questions', most often
identifiable in English because

                                      65
            mi djuno le  du'u       ma       implicit in the context, knows
                 klama le      zarci         the correct value for that place:
            I  know  the statement (Who?     
              going to the  store) holds     mi djuno le  du'u         la
              (1)                              djan. klama    le
                                             I  know  the predication [John
            where "ma" is asking the listener  goes  to the
            to fill in the answer.  This is    zarci jikau              le
            akin to the English - perhaps      zdani
            said in surprise, with emphasis    store (Connection-known) the
            as underlined:                     house] holds.
                                             
            I know who went to the store?    I know whether John goes to the
                                             store or to the house.
            Loglan/Lojban must not use       (4)
            emphasis to distinguish such     
            usages.                            If the x1 place had been "la
              There are at least two ways of djan.", context would imply that
            expressing these now.  When the  it is John who knows the value,
            indirect question word is a form and not the speaker.
            of "ma" (who, what, when, where,   You can match an English
            why, and how questions), just usetranslation better sometimes
            sumti-raising:                   using a different word in the
                                             selma'o.  Using a non-question
            mi djuno tu'a                    word may imply additional
                 le  klama be le  zarci      information not expressible with
            I  know  [some statement about]  a question.  The initial "kau"
                 the goer  to the store,     example captures the 'indirect
                 [namely, identity]          question' aspect of the English
                 (2)                         "whether", but does not read very
                                             colloquially.  You can insert a
              Note that the place structure  hypothetical 'answer' where the
            for "djuno" allows us to avoid   question word goes for a better-
            sumti-raising entirely using a   reading English translation:
            'cleft' x3 subject.              
                                             mi djuno le  du'u         la
            mi djuno             fi    le      djan. klama    le
            klama be le  zarci               I  know  the predication [John
            I  know  (something) about the     goes  to the
              goer  to the store               zarci .akau
              [namely, identity].              le  zdani
              (3)                              store or-(correct value known)
                                                  the house] holds.
            This was one justification for   
            the x3 place.                    I know whether John goes to the
              "djuno"'s place structure does store or to the house.
            not make up for the need for     (5)
            "tu'a" sumti-raising with other  
            brivla that have no 'subject'      You might also choose to see
            place.                           'indirect questions' as
              When the question word is not arestrictive relative clauses:
            sumti, use the new "kau" marker  
            (memory hook: "pau"; "kau"       mi djuno tu'a
            belongs to selma'o UI).  "kau"     zo'ekau
            marks the previous word as being      (6)
            a placeholder that identifies the  poi  klama      be le  zarci
            selma'o of the correct value - itI  know  [something about]
            need not actually be that value.   the something unspecified
            "kau" then indicates that the         (correct value known)
            speaker, or some other person

                                      66
  that goes  to the store, namely  "kaunai", the negated form of
identity.                        "kau", will need some semantic
                                 exploration.  In the above
or even more preferably as the   sentence, I would interpret
simpler abstract bridi:          "jikaunai" to cause the meaning:
                                 
mi djuno le du'u                 I know that John goes to the
  zo'ekau                        store or to the house, but not
  klama    le  zarci             which.
I  know  the statement           
  the something unspecified        However, "kaunai" is more
     (correct value known)       useful in a sentence even more
  goes  to the store.            'indirect':
                                 
I know who went/is going to the  la .alis djica lenu
store.      (7)                     Alice wants the-event that:
  You can also express 'knowing' mi djuno le  du'u         la
more than one thing:               djan. klama    le
                                 I  know  the predication [John
mi djuno le du'u                   goes  to the
  zo'ekau                          zarci jikaunai             le
  klama    zo'ekau                 zdani
I  know  the statement             store (Connection-unknown) the
  the something unspecified        house] holds.
     (correct value known)       
  goes  to the store.            Alice wants me to know whether
                                 John goes to the store or to the
I know who went/is going where.  house (I don't).             (10)
(8)                              
                                 If "kau" had been used, the
  Another case of indirect       statement would imply that I do
question is the other            know.  Finally, by using the
interpretation of the English    discursive operator "se'inai"
(4).  This variety is more easily("other-centered") we could twist
handled:                         the meaning to imply whether
                                 Alice knows.
mi djuno tu'a              le                    
jei                                              
I  know  [something about] the            The role of BAI
truth-value of                   
   la djan. klama    le  zarci ja  Many of these changes are tied
le  zdani                        to the use of selma'o BAI, and we
   John     goes  to the store orwere forced to re-examine what
     the house, [namely the      these are.  Although it was not
     value].                     the original intent, BAI has
I know whether John goes to the  evolved towards being an exact
store or to the house.           equivalent of the gismu which we
(9)                              selected as a memory hook, or as
                                 an abbreviation for a specific
                                 FIhO construct.
                                   The original intent in creating
                                 BAI was to decide on certain
                                 useful or needed roles that could
                                 or should be useful in expanding
                                 bridi, and then to pick words for
                                 them.  As a basis for this we
                                 used Jim Brown's earlier work for
                                 Loglan, coupled with some
                                 research into case theory, and
                                 the everyday, if malglico

                                      67
            ("English-biased"), analysis of  mi nelci la  .apasionatas. pe
            English prepositional and             cu'u                    la
            subordinate phrases.                  artr. rubenstain.
              Institute Loglan has TWO sets  I  love  the Appassionata
            of these - case tags that are         performed (expressed) by
            usable only to label place struc-     Artur Rubenstein.
            ture places, and 'modal relative 
            phrases' which are used to attach  Note the "ne" vs. "pe"
            non-place structure terms.  Therecontrast, reflected in the
            is some overlap and some         English only in whether a comma
            commonalty between these.  The   appears after "Appassionata";
            two are not interchangeable - thethere is only one Appassionata as
            case tags are more like our FA   composed, but there are many
            tags.  We wanted to have the     performances of which I am
            capability that Jim Brown in-    specifying Rubenstein's.
            tended for 'case tags' without                   
            the restrictions and duality.               Loose Ends
            BAI was formed with the intention                
            that every place structure place   A couple of loose ends fell out
            could be labelled more or less   along with the above changes.
            accurately with one of these.      In addition to "co'e", we added
            We've since decided that there   the corresponding unspecified
            can be no all-inclusive set of   relation tag in selma'o BAI,
            'case tags' for Loglan/Lojban    "do'e".  As with "co'e", a memory
            since there is no theoretical    hook is "zo'e", the elliptically
            limit to the number of places in unspecified sumti.
            a bridi, and each place must have
            a different tag.                   "co'e" can be used as its own
              As a result of this evolution, rafsi in compounds.  Examples:
            some members of selma'o BAI have 
            been dropped, and one fairly         co'epre = "unspecified type
            useful one has been changed to   of person".
            clarify its meaning and to make  
            it still more useful.  The old   This could be used in parallel
            "ci'a" was eliminated, and "fi'e"and contrasting structures in
            was assigned (the old "fi'e" was lujvo, such as:
            moved elsewhere, as mentioned    
            above), tying the word to "finti"    ti   xaupre
            ("create/invent") instead of the     This is-a-good-person.
            malglico reference in the old    
            word to "ciska", whose keyword is    ti   xlapre
            "write", but refers to the           This is-a-bad-person.
            "inscribe" sense of that word.   
            Some uses of old "ci'a" are          ti   co'epre
            better expressed with "cu'u",        This is-an-unspecified-
            also in BAI and based on "cusku":person.
            
              "cu'u" refers to the
            "expresser", "fi'e" to the
            "creator".  Thus (example from
            John Cowan, who proposed this
            change):
            
            mi nelci la  .apasionatas. ne
            fi'e
              la betoven.
            I  love  the Appassionata,
            composed by
              Beethoven.
            

                                      68
  le lojbo se ciska (continued)
                                   jubme e lo loldi  .ije mi viska
  Next, a story written by long-
time Lojbanist Bob Chassell, with  lo nu lo canko cu kalri
a couple of corrections by John
Cowan, and Bob and Nora          .i mi catlu lo plita ke bartu
LeChevalier.  But the corrections
were minor.  The most significant  drudi noi lo'e prenu cadzu
change is the incorporation of
the sumti-raising changes in one .i mi cusku fi la tam. noi pulji
sentence, which was a superb
example of where such sumti-       ku'o fe <<lu pe'i le zekri
raising is needed to preserve the
logical character of the           prenu pu cpare le plita ke
language.  The translation
section later in this issue gives  bartu drudi le canko pe le
some stylistic comments that
would make things clearer or       kumfa li'u>>
perhaps more standard, but we
know the language is getting     .i la tam. cusku <<lu ia. ie.  .i
somewhere when comments are on
stylistics rather than communi-    ko catlu le kevna be le bitmu
cating basic ideas.  I'm
reasonably sure that most anyone   be'o poi ke'a trixe le pixra
can understand this story with
word list and only a most basic    .i le kevna cu vasru lo tanxe
understanding of the grammar.  It
is thus printed double-spaced for  .ije ri kunti zo'e li'u>>
those who wish to write
translation notes as they read.  .i la tam. cusku <<lu ju'e le
(Note that "tu'a" is defined in
the preceding article.)            tanxe pu vasru loi rupnu li'u>>

            lo zekri             .i mi catlu le vorme pe le tanxe
                
      fi'e la bab. tcySEL.         pe le bitmu

                                 .i mi cusku <<lu ba'a le stela cu
.i mi cadzu pagre le vorme le
                                   porpi  .i .ua .ue mi facki lo
  kumfa  .i lo xadni pe le nanmu
                                   za'i ge lo vorme gi lo stela na
  cu vreta lo loldi
                                   porpi li'u>>
.i mi viska le flecu be loi ciblu
                                 .i mi cusku fi la tam. fe <<lu le
  bei fo le xadni  .i mi sisku
                                   minra pu farlu lo bitmu lo
  loi sinxa be le zekri
                                   loldi gi'e pu porpi li'u>>
.i mi viska loi kevna pe loi
                                 .i la tam. cusku <<lu pe'i le
  danti ge'u be lo sefta be lo
                                   morsi nanmu pu lacpu le minra
  jubme  .ije mi viska lo nu loi
                                   lo loldi lo bitmu  .i se'o mi'o
  cukta pu farlu lo kajna lo
                                   catlu lo sinxa da poi ke'a

                                   vajni lo nu sisku li'u>>

                                      69
                                             of European civilization over the
            .i la tam. cusku <<lu ra'u ju'e  past couple of centuries to take
                                             other cultures and their "world-
              lo prenu poi ke'a pu sazri le  views" seriously, not only as
                                             curiosities of interest to
              stela lo za'i kalri ku'o djuno scholars (especially
                                             anthropologists), but as evidence
              fi lo tadji be lo pu'u kalri   of the range of possible human
                                             experience.  The formulation of
              sazri le tanxe vorme li'u>>    the hypothesis, associated with
                                             the names of Edward Sapir and
            .i mi cusku <<lu ganai tu'a le   Benjamin Lee Whorf, had to await
                                             what Noam Chomsky has called the
              zekri prenu goi ko'a ge kalri  "Boas tradition" of
                                             anthropological linguists, early-
              rinka le stela tanxe ginai     twentieth century scholars
                                             engaged in empirical studies of
              spofu rinka tu'a ri gi ko'a cu American Indian languages. [See
                                             Chomsky, "Linguistic
              djuno fi lo pu'u kalri sazri leContributions to the Study of
                                             Mind: Future," rpt. in Language
              tanxe vorme  .i .ua ru'a ko'a  in Thinking: Selected Readings,
                                             ed. Parveen Adams (Harmondsworth:
              catlu le se minra be le nu     Penguin, 1973), pp. 336ff.]  The
                                             hypothesis is emphatically not
              kalri sazri le stela tanxe     the a priori doctrine of
                                             linguists seduced by a
              vorme sepi'o lo darno ke catlu philosophical tradition, but a
                                             proposal advanced by investiga-
              cabra li'u>>                   tors who actually took the
                                             trouble to confront "alien"
                                             languages and cultures.
            _________________________________  What does the Sapir-Whorf
                     ______________          Hypothesis claim?  If it were
                                             true, what phenomena would we en-
                Versions of the Theory of    counter and be equipped to
                  Linguistic Relativity      explain?  In a fairly recent
                    by Robert Gorsch         article in the American
                                             Anthropologist (1984), Paul Kay
                      INTRODUCTION           and Willett Kempton reduce the
                                             Hypothesis to three propositions:
            The Sapir-Whorf Hypothesis       
              The "Sapir-Whorf Hypothesis,"    I.  Structural differences
            which asserts that one's native  between language systems will, in
            language determines in some      general, be paralleled by non-
            fashion the nature of one's      linguistic cognitive differences,
            experience and that members of   of an unspecified sort, in the
            different linguistic communities native speakers of the two lan-
            will necessarily inhabit         guages.
            different experiential worlds,     II.  The structure of anyone's
            has its roots in the ideas of    native language strongly
            eighteenth- and nineteenth-      influences or fully determines
            century thinkers like            the world-view he will acquire as
            Giambattista Vico and Wilhelm vonhe learns the language.
            Humboldt.  [See George Steiner,    III.  The semantic systems of
            After Babel: Aspects of Language different languages vary without
            and Translation (London: Oxford  constraint.
            Univ. Press, c. 1975), pp. 73ff.]
            The emergence of this hypothesis ["What is the Sapir-Whorf
            reflects the growing willingness Hypothesis?"  American

                                      70
Anthropologist 1984 (86), 66.    omissions, discourages other,
Kay and Kempton's formulation is logically possible distinctions.
based upon the thinking of Roger   In short, according to the
Brown and, through him, Eric     Sapir-Whorf Hypothesis, one will
Lenneberg.]                      "see" what the structure of one's
  As this series of propositions language asks one to see and one
suggests, one can distinguish twowill "see" -- as separate things
possible sources of "Whorfian    -- what the semantic system of
effects": (1) differences in     one's language defines as
"linguistic structure" and (2)   discrete semantic units.
differences in "semantics."      
(Strictly speaking, of course,   Saussurean Sign-Theory
the "semantic system" of a         It is sometimes thought that
language, the division of        the Sapir-Whorf Hypothesis has
experience embodied in its       been discredited and relegated to
lexicon, is a part of its "struc-the trash-heap of intellectual
ture."  For, in linguistics,     history.  Certainly, it is true
"structure" is really a synonym  that mainstream linguists, influ-
for "system.")  Whorfians typi-  enced by Noam Chomsky, tend to
cally emphasize linguistic       dismiss the Sapir-Whorf
"structure" in a fairly limited  Hypothesis no matter how
sense.  Thus, they tend to argue judiciously it is reformulated.
that the structure of one's      One could hardly expect any other
native language will, by         response, since Chomskian
encouraging a particular manner  linguists are committed almost as
of structuring one's report of   a matter of faith to the notion
experience, have the effect of   that the differences between
shaping one's perception of the  human languages must be
world.  One will tend to note in superficial and even trivial.  If
perception, that which one's     one accepts the Chomskian theory
grammar asks one to report in    of a "universal grammar," one
utterance.  "Structure" embodies,will be compelled to dismiss any
and imposes upon the speaker, a  attempt, no matter how empirical
metaphysics.                     its grounds, to justify the
  The semantic organization of   Whorfian argument that "grammars"
one's language will similarly    vary enough to affect the
shape one's experience of the    structure of human experience.
world.  This is the implication    Whatever mainstream linguists
of Whorfian arguments that make  say, the Sapir-Whorf Hypothesis
appeal to such facts as the num- is alive and well in the popular
ber of words that the Eskimos    mind and in the academic mind --
have for the English concept     at least outside of the
"snow."  If one approaches the   discipline of linguistics.  Many
semantic system of language in a feminists, for example, believe
Whorfian spirit, this system willthat the structure of English
be viewed as an arbitrary seg-   imposes upon its speakers a
mentation of the experienced     patriarchal metaphysics.
world.  We divide up the         (English customarily subsumes the
continuum of experience in       feminine under the masculine in
"culturally pertinent" ways, to  its pronoun system, as in
use a phrase borrowed from the   expressions like "To each his
semiologist Umberto Eco, in      own.")  In the disciplines
accordance with our needs as     customarily termed the humani-
members of cultural groups       ties, particularly those that
confronting particular physical  investigate literature and
and social environments.  The    culture, versions of the Sapir-
lexicon of our language, by the  Whorf Hypothesis are widely taken
categories it defines, affords usfor granted; the Sapir-Whorf
ways to make distinctions in the Hypothesis, in some version, is
experienced world and, by its

                                      71
            the premise of many currently    continuum of experience into
            dominant methodologies.          units -- "things," "states,"
              Take for instance modern "sign-"processes," and so forth -- and
            theory."  Semiology or "sign-    links these units, one to
            theory," popularized by struc-   another, in a web of relations of
            turalism and post-structuralism, opposition and affinity.  Green
            embraces an equivalent of the    is, for instance, differentiated
            Sapir-Whorf Hypothesis.  Some    from yellow on the one hand and
            "sign-theorists" even look back  blue on the other:  green exists
            to Whorf as a precursor.  Modern as a unit in opposition to
            "sign-theory," rooted in the workadjacent units in the same
            of the Swiss linguist Ferdinand  semantic field.  At the same
            De Saussure, posits an initial   time, green is linked
            moment when the human subject    metaphorically, in relations of
            enters "language" and at the sameaffinity, to units belonging to
            time a certain culture-bound     different semantic fields, for
            experiential world.  In first    instance, such units as nature,
            language acquisition an arbitrarylife, youth, and jealousy.
            system for organizing raw          In suggesting that "raw
            experience begins to be imposed  experience" -- what Whorf calls
            upon the mind.  Subjects learn   "the kaleidoscopic flux of
            how to segment experience into   impressions" -- is organized by
            the units specified by the       the human mind after its embrace
            language they acquire as infants;of a particular sign-system,
            they divide the continuum of     Saussurean sign-theory simply
            experience into the "semantic    reformulates the Whorfian
            units" that semiologists call    Hypothesis.  According to this
            "signifieds" -- i.e., the con-   reformulation, the lexicon of
            ceptual elements of "signs."     one's native language imposes a
            [According to semiological       system of categories on one's
            theory, every "sign" consists of experience; the lexicon imposes
            a "signifier" or "expression" andon the speaker an arbitrary
            a "signified" or "content":      differentiation of the continuum
            every linguistic sign, for       of experience into semantic units
            instance, unites a combination of-- or, in the terminology of
            sounds or a series of written    semiology, "signifieds" or
            symbols (the signifier) with a   "culturally pertinent units."  At
            concept (the signified).]        the same time each language
              Semiologists typically pay     imposes on the speaker a network
            special attention to the array ofof relations of affinity between
            "signifieds" posited by a lin-   these semantic units.  This
            guistic community, i.e., the     system of categories and the
            units into which the community   accompanying network of asso-
            divides the continuum of the     ciations constitute the "map" of
            experienced world, and to the    experience offered by each
            network of relations by which    language to its native speakers.
            these "signifieds" are                           
            interrelated, i.e., the system of     A WORKING BIBLIOGRAPHY
            connotative links by which units 
            belonging to different semantic  Note on the bibliography:
            fields are linked with one         In this bibliography I attempt
            another.                         to trace the development of the
              Thus, semiology takes for      "Sapir-Whorf Hypothesis" from the
            granted one of the crucial       early decades of the twentieth-
            corollaries of the Sapir-Whorf   century to the present.  The
            Hypothesis, namely, that in      items included in the
            acquiring the semantic system of bibliography range in date from
            a language one embraces a        1911 to 1990.  While the bib-
            particular "map" of experience.  liography makes no claims to
            A semantic system divides the    completeness, it does represent

                                      72
an attempt (1) to clarify the    1a.  The Sapir-Whorf Hypothesis:
role of earlier ethnologists,      Formulation
including Boas and Sapir, in the 
formulation of what is often     Ben G. Blount, ed., Language,
called simply "the Whorfian        Culture, and Society: A Book of
Hypothesis," (2) to chart the      Readings (Cambridge, Mass.:
career of the Hypothesis from the  Winthrop, 1974).
1940's to the 1980's, and (3) to   This sourcebook includes
draw attention to the kindred      important selections from Boas,
thinking of semiologists working   Sapir, Whorf, and Hoijer.
in the tradition of Saussurean   Franz Boas, "Theoretical
linguistics.                       Importance of Linguistic
  The bibliography is not          Studies," in "Introduction" to
alphabetical; entries are          the Handbook of American Indian
arranged by category and date.     Languages, F. Boas, ed.,
  In compiling this working        Bulletin 40, Part II, Bureau of
bibliography I have cannibalized,  American Ethnology (Washington,
without shame, the following       D. C.: Government Printing
lists of references:  Wallace L.   Office, 1911). Reprinted in
Anderson and Norman Stageberg,     Blount, pp. 23-31.
eds., Introductory Readings on   Lucien Levy-Bruhl, How Natives
Language (New York: Holt,          Think (N.Y.: Knopf, 1925), pp.
Rinehart, and Winston, 1975), pp.  139-180.
38ff.; Ben G. Blount, ed.,       Willis D. Wallis, An Introduction
Language, Culture, and Society: A  to Anthropology (N.Y.: Harper
Book of Readings (Cambridge,       and Row, 1926), pp. 416-431.
Mass.: Winthrop, 1974);  Ralph   Edward Sapir, "The Unconscious
Dumain, "Bibliography on Language  Patterning of Behavior in
and Thought," ju'i lobypli (The    Society," in The Unconscious: A
Logical Language Group), March,    Symposium, ed. E. S. Drummer
1990, 36-38; John J. Gumperz,      (New York: Knopf, 1927).
"Reader" for "Interactional        Reprinted in Blount, pp. 32-45.
Sociolinguistics (Anthropology   - - - - - - , "Conceptual
270B)," University of California,  Categories in Primitive
Berkeley, Fall, 1986; John Parks-  Languages," Science 74 (1931).
Clifford, [Note], ju'i lobypli   - - - - - - , "Language,"
(The Logical Language Group),      Encyclopedia of the Social
Dec., 1989, p. 44; and Bob         Sciences, ed. Seligman and
LeChevalier [and Alan Munn], ju'i  Johnson (New York: Macmillan,
lobypli, March, 1991, pp. 57ff.    1933).  Reprinted in Blount,
I want to thank Bob LeChevalier    pp. 46-66.
and the Logical Language Group   Benjamin Lee Whorf, Language,
for arguing incessantly about the  Thought, and Reality: Selected
Sapir-Whorf Hypothesis and my      Writings of Benjamin Lee Whorf,
colleague Barbara Grant for loan-  ed. John B. Carroll (Cambridge,
ing me a copy of Gumperz's         Mass.: MIT Press, 1956).
"Reader."                          The most revealing essays are,
                                   in my opinion, "Science and
                                   Linguistics" (1940) and
                                   "Languages and Logic" (1941).
                                   Another interesting essay,
                                   reprinted in Blount as well as
                                   in Carroll's selection, is "The
                                   Relation of Habitual Thought
                                   and Behavior to Language"
                                   (1939).
                                   See also the essays "An
                                   American Indian Model of the
                                   Universe" (c. 1936), "A
                                   Linguistic Consideration of

                                      73
              Thinking in Primitive Com-     Paul Hanle, Language, Thought,
              munities" (c. 1936),             and Culture (Ann Arbor: Univ.
              "Linguistics as an Exact         of Michigan Press, 1958).
              Science" (1940), and "Language,  Summarizing the results of a
              Mind, and Reality" (1941).       conference held at U. Mich. in
                                               1951-2.
            1b. The Sapir-Whorf Hypothesis:  Roger Brown, Words and Things
            Career                             (N.Y.: Free Press, 1958), pp.
                                               229-63.
            M. J. Herskovits, Man and His    J. Fishman, "A Systematization of
              Works (N.Y.: Knopf, 1947), pp.   the Whorfian Hypothesis,"
              440-457.                         Behavioral Science 5 (1960),
            Clyde Kluckhohn, "The Gift of      232-39.
              Tongues, in Mirror for Man: A  James Cooke Brown, "Loglan,"
              Survey of Human Behavior and     Scientific American 202 (1960),
              Social Attitudes (New York:      53-63.
              McGraw-Hill, 1949), Chapter VI.  Describes an effort in
            John B. Carroll, The Study of      linguistic engineering designed
              Language (Cambridge, Mass.,      to create an artificial
              1953), pp. 43-48.                language that would permit the
            Harry Hoijer, "The Relation of     Whorfian Hypothesis to be
              Language to Culture," in         tested.
              Anthropology Today, ed. A. L.  John B. Carroll, "Language and
              Kroeber (Chicago: Univ. of       Cognition," in Language and
              Chicago Press, 1953), pp. 554-   Thought (Englewood Cliffs, N.
              573.                             J., 1964).
            Harry Hoijer, ed., Language in     See especially 106-110 ("The
              Culture, Comparative Studies of  linguistic-relativity thesis"),
              Cultures and Civilizations, No.  which offers a critique of the
              3; Memoirs of the American       strong version of the Whorfian
              Anthropological Association,     Hypothesis.
              No. 79 (Chicago: Univ. of      James Cooke Brown, Loglan I
              Chicago Pr., 1954).              (Gainesville, Fla.: The Loglan
              The proceedings of a 1953        Institute, 1966).
              conference on the Sapir-Whorf    Brown's book was revised in
              Hypothesis.                      1975 and 1989.
            Harry Hoijer, "The Sapir-Whorf   Dell Hymes, "Two Types of
              Hypothesis" (1953), reprinted    Linguistic Relativity," in
              in Hoijer (1954) and in Blount   Sociolinguistics: Proceedings
              (1974).                          of the UCLA Sociolinguistics
            R. Brown, "Linguistic Determinism  Conference (1964), ed. W.
              and Parts of Speech," Journal    Bright, Janua Linguarum Series,
              of Abnormal Social Psychology    20 (The Hague: Mouton, 1968),
              55 (1957), 1-5.                  114-167.
            R. Brown and E. Lenneberg,       Arnold M. Zwicky, Review of
              "Studies in Linguistic           Brown's Loglan I, Language 45:2
              Relativity," in E. Macroby, T.   (1969), 444-457.  See also John
              H. Newcomb, and E. L. Hartley,   Cowan (1991), below.
              eds., Readings in Social       Roger Brown, Psycholinguistics:
              Psychology, 3rd edition (New     Selected Papers (N.Y.: Free
              York: Holt, Rinehart, and        Press, 1970), pp. 235-256.
              Winston, 1958), 9-18.
            John B. Carroll and Joseph B.
              Casagrande, "The Function of
              Language Classification in
              Behavior," in Readings in
              Social Psychology (1958), 18-
              31.

                                      74
John MacNamara, "Bilingualism and  selected and introduced by A.
  Thought," Georgetown University  S. Dil (Stanford: Stanford
  Round Table on Languages and     Univ. Press, 1979).
  Linguistics 1970: Bilingualism   "Friedrich disagrees with
  and Language Contact, ed. by     Whorf's views on language and
  James E. Alatis (Washington:     metaphysics, but accepts the
  Georgetown University Press,     strong thesis in the realm of
  1970), pp. 25-45.                poetic language and its
  Critical of the Whorfian         relation to the imagination"
  Hypothesis.                      (R. Dumain).
Ferruccio Rossi-Landi, IdeologiesPaul Kay and Willett Kempton,
  of Linguistic Relativity (The    "What Is the Sapir-Whorf
  Hague: Mouton, 1973).            Hypothesis?" American
  Includes consideration of the    Anthropologist 86 (1984), 65-
  sociological roots of the        79.
  doctrine of linguistic relativ-  Discusses the content of the
  ity, including white guilt over  Hypothesis and reviews
  the extermination of the         empirical research that
  Indians.                         attempts to test it; reports
Noam Chomsky, Introduction to      experimental confirmation of a
  Adam Schiff, Language and        modified version of the
  Cognition (1964), tr. Olgierd    Hypothesis in the area of color
  Wojtasiewicz and ed. Robert S.   perception.
  Cohen (N. Y.: McGraw-Hill,     Frederick J. Newmeyer, The
  1973).                           Politics of Linguistics
  Critique of the Whorfian         (Chicago: Univ. of Chicago
  Hypothesis.                      Press, 1986).
Adam Schiff, Language and          A history of linguistic theory
  Cognition (1964), tr. Olgierd    that attacks the Whorfian
  Wojtasiewicz and ed. Robert S.   Hypothesis as racist.
  Cohen (N. Y.: McGraw-Hill,     David McNeill, "Linguistic
  1973).                           Determinism: The Whorfian
  Historical account of            Hypothesis,"      in
  linguistic theory (from the      Psycholinguistics: A New
  18th century on): background to  Approach (New York: Harper and
  the Sapir-Whorf Hypothesis.      Row, 1987), Ch. 6, pp. 173-209.
Ronald W. Langacker, "Semantic   The Logical Language Group, ju'i
  Representations and the          lobypli (1988-1991).
  Linguistic Relativity            A variety of discussions of the
  Hypothesis," in Foundations of   Sapir-Whorf Hypothesis from the
  Language 14 (1976), 307-357.     perspective of Lojbanists: see
  The author "tries to formulate   Aug.-Sep., 1988; Dec., 1988;
  the hypothesis in a non-vacuous  June-July, 1989; Nov.-Dec.,
  manner, and ultimately rejects   1989; March, 1990; May, 1990;
  the strong version, basing       August, 1990; and March, 1991.
  himself on a distinction       John Cowan, "Loglan and Lojban: A
  between primary conceptual       Linguist's Questions and an
  structures and the semantic      Amateur's Answers," ju'i
  representations into which       lobypli (March 1991), pp. 21ff.
  thought is coded" (R. Dumain).   Responding to Zwicky's review
Danny K. Alford, "The Demise of    of Brown's Loglan I.
  the Whorf Hypothesis (A Major  
  Revision in the History of Lin-2. Semiology and the Thesis of
  guistics)," Proceedings of the   Linguistic Relativity.
  4th Annual Meeting of the        The following list by no means
  Berkeley Linguistic Society, 4   represents the field of
  (1978), 485-99.                  semiology as a whole; I have
Paul Friedrich, Language,          limited myself to a handful of
  Context, and the Imagination:    texts that I have found useful
  Essays by Paul Friedrich,        in the classroom.

                                      75
                                             3.  Related Studies
            Ferdinand de Saussure, Course in 
              General Linguistics (1915), tr.B. Comrie, ed., The World's Major
              Wade Baskin (New York:           Languages.
              Philosophical Library, 1959),    Descriptive text used in the
              pp. 7-17, 65-78, and 111-122.    design of Lojban.
              Seminal sections from          Brent Berlin and Paul Kay, Basic
              Saussure's lectures, laying the  Color Terms (Berkeley: Univ. of
              foundations for modern sign-     Calif. Press, 1968), esp. pp.
              theory (semiology or             1-14.
              semiotics).                    George Steiner, After Babel:
            Pierre Guiraud, Semiology (1975).  Aspects of Language and
              A reasonably good primer,        Translation London: Oxford
              introducing sign-theory and its  Univ. Press, c. 1975), esp. pp.
              application to various areas of  73-109: Linguistic relativism
              human experience.                (including Whorf) vs.
            Umberto Eco, "Social Life as a     linguistic universalism
              Sign System," Structuralism: An  (Chomsky).
              Introduction, ed. David Robey,   Useful for its discussion of
              (1973), pp. 57-72.               the philosophical tradition
            - - - - -  , "How Culture          that lies behind the Sapir-
              Conditions the Colours We See,"  Whorf Hypothesis.
              On Signs, ed. Marshall Blonsky Eleanor Rosch, "Classification of
              (1985), pp. 157-175.             Real-World Objects: Origins and
              This essay and "Social Life as   Representations in Cognition,"
              a Sign System" provide a useful  MS, University of California,
              introduction to the semio-       Berkeley, c. 1975.
              logical equivalent of the        Criticizes, on empirical
              Sapir-Whorf Hypothesis.          grounds, the idea that
            Takao Suzuki, Words in Context: A  experience is a continuum
              Japanese Perspective on          arbitrarily segmented by the
              Language and Culture (1973),     mind.  Available from E. Rosch,
              tr. Akira Miura (Tokyo:          c/o Dept. of Psychology, Univ.
              Kodansha International, 1978;    of Calif., Berkeley, Berkeley,
              rev., 1984).                     CA 94720.
              A richly suggestive comparison George Lakoff and Mark Johnson,
              of the languages and cultures    Metaphors We Live By (Chicago:
              of Japanese speakers and         Univ. of Chicago Press, 1980).
              English speakers.  The book      "The authors make an important
              presents, and offers empirical   study of the metaphorical basis
              evidence for, a theory of lin-   of language.  In the final
              guistic relativity similar in    chapters they argue for an
              spirit to those of Whorfians     extreme relativism" (R.
              and Saussurean semiologists.     Dumain).
            John Lucy, "Whorf's View of the  Alfred H. Bloom, The Linguistic
              Linguistic Mediation of          Shaping of Thought: A Study in
              Thought," in Semiotic            the Impact of Language on
              Mediation: Sociocultural and     Thinking in China and the West
              Psychosocial Perspectives, ed.   (Hillsdale, N.J.: L. Erlbaum,
              E. Mertz and R. J. Parmentier    1981), pp. 13-36.
              (Orlando: Academic Press,        "The Distinctive Cognitive
              1985).                           Legacies of English and
                                               Chinese," especially the
                                               sections "Counterfactuals in
                                               English and Chinese" and
                                               "Theoretical Extensions."
                                             George Lakoff, Women, Fire, and
                                               Dangerous Things: What
                                               Categories Reveal About the

                                      76
  Mind (Chicago: Univ. of Chicago  le lojbo se ciska (continued)
  Press, c. 1987).               
  How human beings segment and     Now for a lighter piece of
  order their experience.        Lojban text.  The following bit
                                 is from John Cowan, and he uses
CONCLUDING NOTE:  This is only a "rinka" in the intended uncleft
working bibliography; I welcome  place structures manner.  He
the assistance of other inter-   noted in submitting it that he
ested scholars.  Please send     made no grammatical errors - the
comments, criticisms,            parser accepted it the first
corrections, and suggested       time.  For those trying to read
additions and deletions, to the  it, "clupe'as. xarengus." is a
following address:               Lojbanization of the Linnean bi-
                                 nomial for "herring".  A
Robert Gorsch                    translation will be found later
Department of English            in this issue.
St. Mary's College
Moraga, Calif.  94575
                                    pamoi xamrei ra'a lo verba
_________________________________
        ________________
                                 ni'o la paf. cusku lu pau mazo'o
                                   crino gi'e dandu le bitmu gi'e
                                   siclu li'u


                                 .i la ver. cusku lu .uanai mi na
                                   djuno li'u


                                 .i la paf. cusku lu .ui lo me la
                                   clupe'as. xarengus. finpe li'u


                                 .i la ver. cusku lu ia ri goi
                                   ko'a na crino li'u


                                 .i la paf. cusku lu fu'i le nu ko
                                   gasnu cu rinka le nu ko'a ba
                                   crino li'u


                                 .i la ver. cusku lu .iasai ko'a
                                   ba'e na dandu le bitmu li'u


                                 .i la paf. cusku lu fu'isai le nu
                                   ko gasnu cu rinka le nu ko'a ba
                                   dandu li'u


                                 .i la ver. cusku lu iacai ko'a
                                   ba'e ba'e na siclu li'u


                                 .i la paf. cusku lu fu'icai mi pu
                                   cusku lo jitfa li'u

                                      77
            _________________________________  Each grammatical construct
            ________________                      consists of a leading
                                                  LeftBracket token, a
             On Loglan and Lojban Elidables       trailing RightBracket token,
                                                  and some number of sub-
            The following paper was written       structures trapped between
            by Jeff Prothero as an answer to      these two tokens.
            criticism of the use of elidable   Every token is either a
            tokens in Loglan formal grammars.     LeftBracket or a
            The argument applies to Lojban as     RightBracket.
            well as to any other version of    No token is both a LeftBracket
            Loglan grammar, provided that the     and a RightBracket.
            grammar abides by the defining     Every LeftBracket has a unique
            rules of Bracket Languages (I am      matching RightBracket.
            not sure that current Institute    
            Loglan still abides by these     Note that we do not require that
            defining rules - comment is      each RightBracket have a unique
            sought from anyone who has such  matching LeftBracket.
            knowledge.)  The reference to GU 
            in the title is to the older     Sample Bracket Language:
            Loglan RightBracket selma'o that 
            in Lojban was changed to KU.  Thestart   ->   bracket | broket |
            title is thus a bit of a pun for mixed
            Lojbanists, since the 'GU' is    
            gone from our selma'o list as    bracket ->   '['   ']'
            well.                                    |    '[' start ']'
                                                     |    '[' start start ']'
                     The GU is Gone!         
             Elidable Terminators in Logical broket  ->   '<'   '>'
                        Languages                    |    '<' start '>'
            Copyright (c) 1989 Jeff Prothero         |    '<' start start '>'
             Reprinted with permission from  
                       the author.           mixed   ->   '{'   '>'
                                                     |    '{' start '>'
              The elision of trailing                |    '{' start start '>'
            terminators has been a prime
            problem for everyone seriously
            working to understand the Lo**an
            grammar.  This paper is a first
            attempt to deal with this
            problem.  The major questions to
            resolve:
            
              When can terminators be elided?
              When would such elision
                 introduce ambiguity?
              How does one recover the full
                 syntax of a sentence
                 containing such elisions?
            
              The first step is to establish
            a simple analytical model, which
            exhibits the relevant problems
            without extraneous detail and
            complexity.  We consider the
            Bracket Languages BL, defined by
            the following grammatical
            properties:
            

                                      78
  This grammar specifies the     added to BL as a result of the
infinite set of strings:         Augmentation Rule.
                                 
   []                              Question 1:  Does the
   <>                                 Augmentation Rule introduce
   {>                                 ambiguities?
   [[]]                          
   [[][]]                          Let us make the question more
   [{><>]                        precise.  Each e in E was derived
   . . .                         from some parent p in BL (not
                                 BL'!) by one or more applications
  For terseness, we would like toof the Augmentation Rule.  We
omit some of the trailing        want to know if this derivation
terminators "when no ambiguity   was unique, or if some such e has
would result".  The problem is totwo possible parents in BL.
formally specify the latter      Formally:  can there exist a pair
constraint.                      <e,p>, e in E and p in BL with p
  From a mathematical-linguistic -> (via repeated Augmentation
point of view, dropping some of  Rule) e, such that r is in (BL-
the trailing terminators corre-  p)'?  If so our Augmentation Rule
sponds to adding various strings has introduced an ambiguity into
to the above language, such as:  the language by erasing an
   [                             essential token, rather than
   {                             merely a redundant token.
   [[]                           
   [[][]                           Answer   1:  No such ambiguity
   [{><>                              is introduced by the
   [{><                               Augmentation Rule.
   [{><]                         
   . . .                         PROOF:  Let us assume that such
                                 an ambiguity exists.  Then either
  How do we specify the full set e, or some other string along the
of strings to be added?  Given   path from p to e, has two
such a string, how do we recover possible legitimate parents under
the full syntax?                 the Augmentation Rule.  Let us
  I propose the following        call this child c, and the two
Augmentation Rule for adding the possible parents p0 and p1.
strings:                           Then we have:
                                    p0 == "a]Bc" for some a,],B,c
   IF "a]Bc" is in the language,    p1 == "a>Bc" for some > != ],
where:                                same a,B,c.
     "a" is any sequence of      
tokens                             Now > and ] both match the same
     "]" is any RightBracket     LeftBracket in "a", since p0 and
     "B" is any token            p1 are both strings from our
     "c" is any sequence of      Bracket Language.  But each
tokens                           LeftBracket has a UNIQUE matching
   AND IF "aB" is not a prefix ofRightBracket, by our definition
     any string in the language, of Bracket Languages, hence we
   THEN we add "aBc" to the      must have > == ], hence p0==p1,
language.                        hence no such distinct parent-
                                 pair is possible. QED.
  Given a Bracket Language BL,   
application of the Augmentation    Question 2:  Is a LALR(1)
Rule until closure is achieved        parser capable of detecting
results in a new language BL'         such elided tokens?
which contains BL as a subset.   
Let us call BL'-BL "E" (for "Eli-  Answer   2:  Yes.
sions").  These are the strings  

                                      79
              If "a]Bc" has been reduced to  
            "aBc", then we must have the          *le nanmu joi le ninmu
            condition that "aB" is not a     
            legal prefix of any string in BL.is ungrammatical (though
            But an LALR(1) parser has a tableperfectly understandable to
            which tells it, at any given in- humans) because of the '1' in
            stant, the legal set of lookaheadLALR(1).
            tokens.  If the current lookahead  After "nanmu", "joi" is legal
            token is not in that set, there  and moves the grammar to a state
            will be at most one RightBracket (using the formal grammar rule
            in the lookahead set, by a simplefor "joikjeks") where it expects
            variant of the above argument.   a token valid in a selbri (i.e.;
            The parser can then insert that  it expects something like the
            unique RightBracket in its input valid:
            stream and continue.             
                                                    le nanmu joi ninmu
            _________________________________      = the hermaphrodite).
                    ________________         
                                               Not finding a selbri word (of
                  How Elidables Work in      which some 20 or 30 selma'o are
                      Loglan/Lojban          legal in the first position by
                   by Bob LeChevalier        SOME rule or another), it then
                                             tries to stick in the elidables
              I would love it if someone     from shortest scope termination
            could solve the problem of       to longest.  But none of these
            specifying elidability rules, butare legal in the position AFTER
            I can only do so in generalities joi where we are looking:
            without making specific reference
            to YACC's LALR(1) algorithm.        *le nanmu joi ku/cu/vau/...
            Specifically, the elidables are  
            optional if, using the YACC      so the parser rejects the phrase.
            algorithm, a parser looking at   To non-logically join two sumti
            the next token after an omitted  with joi under the formal grammar
            elidable, does not find it valid.rule for "joikek", the "ku"
            Instead it performs error pro-   cannot be elided before the "joi"
            cessing which sticks in the      and the following is grammatical:
            elidable and this in turn moves  
            it to a new state.                   le nanmu ku joi le ninmu
              Thus in "le nanmu klama le      The team of the man/men joined
            zarci", after "nanmu" a parser         with the woman/women
            will read "klama" and determine  
            that a tanru is continuing.  It    Presumably a LALR(2) (looking
            reads "le", which is not legal   ahead 2 tokens) parser would be
            after "klama" by any rule, and   able to handle this particular
            inserts "ku".  It inserts "ku"   elision, but Lojban is defined so
            and other elidable terminators   as to satisfy LALR(1).  There are
            based on the order of constructs;other places that even 2 is not
            it sticks in the 'tightest bound'sufficient, but they less often
            elidable - in this case the      involve elidables.  Far more
            terminator for the first "le"    often, if you omit an elidable
            construct.  The result is a      incorrectly, you will end up with
            grammatical parse - as two sumti another sentence/fragment that is
            without a selbri.                valid but grammatically
              Probably the speaker intended  different.  We thus recommend
            "le nanmu cu klama le zarci", butthat where in doubt, or in noisy
            the parser cannot determine this,environments, use the added
            because YACC will not stick in anredundancy of including the
            elidable unless it finds an      elidable.  We thus WANT the
            invalid token.                   language to be defined so that
              For another example:

                                      80
elision is not mandatory if it isdo fine.  There is no stigma
possible.                        against including in an elidable
  Thus, the question:  "Is there terminator unnecessarily.
always enough information for    Because of this, you do not need
someone to decide exactly when a to know 'the whole grammar' to
"cu" or some other elidable is orspeak Lojban.  Only in complex
is not required?" can be         nested constructs which you
answered:  yes, it is always pos-shouldn't be using if you don't
sible.  But you must know the    know most of the grammar, are you
entire grammar to always be able likely to find situations where
to decide.  With an incomplete   you might erroneously elide a
knowledge of the grammar, you mayterminator.  Even there, by
end up incorrectly eliding, and  concentrating on just a few
should err when uncertain on the 'most-frequent errors', you will
side of not eliding.             seldom make an error.
  This is not as bad as it seems,
because most problems that might _________________________________
arise (the "joi" problem is un-          ________________
usual in this regard) occur      
because of multiple elision.       A History and Description of
Thus in:                           le'avla in Loglan and Lojban
                                        by Bob LeChevalier
le nu mi klama le zarci cu xamgu 
                                   Those with 1975 dictionaries
the complete specification with  will find every chemical element
all elidables added is:          was included twice, as a name,
                                 and as an 'S-prim' ("le'avla that
lenu mi klama le zarci ku vau keilooks like a gismu" if you are
           ku cu xamgu           new to the project).
                                      Though JCB disagrees, I
and the "cu" separator, acting asbelieve it was a conversation
the longest scope elidable, makesbetween him and me in 1980 that
a wall that forces a parser to   led to the "3rd lineage" of
keep sticking in all the         borrowings (translation - the
optionals until there are none   creation of a third form of
left.                            brivla besides gismu and lujvo -
  If we had omitted the "cu", thethe le'avla).
parser would add in elidables      His response proposal - the
only at end of text giving the   birth of le'avla, was reported a
non-elided equivalent:           month or so later in TL3/4.
                                   At the time of GMR, JCB moved
le nu mi klama le zarci xamgu ku MOST of the 'S-prims' into
vau kei ku vau                   borrowing space.  The 'algorithm'
                                 for le'avla was discussed in
  It turns out that there is a   TL6/1 (1983).  JCB then launched
valid Lojban interpretation that what he called the "Sciwords"
is not the one intended if you   project, to massively borrow
use any single one of those      words from many fields into the
terminators besides "cu" in the  language.  If there were any
place where the "cu" would go.   volunteers at the time, their
With no "cu" and two elidables,  work was never reported because
you can get the correct          TL folded followed by Lognet a
interpretation with "ku ku", "vauyear later after the 1983-4
ku", or "kei ku", but not with   political squabbles.
"ku vau", "ku kei", or "vau kei".  JCB continued to work on the
                                 borrowings, and translated a few
  But the average Lojbanists needparagraphs of Scientific American
not worry about these other forms(reprinted in 4th edition Loglan
- just use the "cu" when in even 1) that were heavy in scientific
the slightest doubt, and you'll  jargon to be borrowed.  He once

                                      81
            reported making borrowings for 50the elements to be borrowed from.
            kinds of cheese one night after  I still subscribe to this idea,
            reading an article on the        though John Cowan does not.  Ger-
            subject.  There have been some   man has made non-international
            reports in recent Lognets that   forms of some elements, and
            others have made some le'avla andChinese, with its word-forming
            that the Sciwords project finallyrestrictions, has non-in-
            accomplished something, but no   ternational forms for all (but
            list has been published because  they often try to make a word
            of the Institute's trade secret  that suggests the chemical
            policy.                          symbol).  The "Latinate forms"
              When Rebecca Bach and I visitedare really the English/ French
            JCB is May 86, we discussed      forms, since those two languages
            borrowings, and specifically     have dominated the scientific
            JCB's then current effort on     publication field during the time
            remaking the element words into  of internationalization of sci-
            le'avla as a test for his        ence.  We can't get around this
            attempts to devise "fast-tracks  'Latinate bias', but feel that if
            to borrowing" that would evade   a truly international standard
            the mind-stretching "*slinku'i"  exists, we should use it.
            test.  They didn't.  It turned     After the 1986 visit to JCB, I
            out that we went through all of  went home, and reworked the
            the elements and remade them, butelement words, which were left
            found that there were few simple hanging.  The UL2 publication was
            guidelines.  (We did notice that 4 months later.  Other than a
            -CVCV word-endings frequently    discussion in JCB's Notebook 3,
            give good le'avla, as well as    and Loglan 1 4th edition, and
            that it is easier to avoid       ensuing responses to my criticism
            "*slinku'i" problems by making   of the latter, there has been
            the initial consonant cluster notminimal discussion of le'avla
            a permissible initial.  Rebecca, until recently - although the
            a Loglan novice, proved better   culture words have been
            than either JCB or me at         questioned by many new Lojbanists
            detecting flaws in le'avla-      (who have generally been
            making, but none of us were      satisfied with my answers -
            really good at it.)              again, until recently).
              JCB at this time made clear    
            that a standard for scientific        Four Flavors of le'avla
            le'avla making, unlike gismu, was
            visual recognition rather than     Here was my 1979 argument and
            aural recognition, since         proposal for 4 'flavors' of
            technical words are used in      borrowing.  In attempting to
            written language more than spokentranslate the song "Man of La
            language.  JCB introduced        Mancha", which as I've reported
            borrowing-and-name-only lerfu forwas my own first attempt to use
            "W", "Q", "X", and "Y" to make   Loglan, there was no word for
            visually recognizable borrowings "trumpet", "gauntlet", etc.  Even
            easier.  (His version of the     if there had been, in the context
            language uses "H" where we use   of the song, these words convey
            "x".)  JCB's published examples  specific cultural values that are
            show the priority on visual      not inherent to the musical
            recognition of borrowings rather instrument or the piece of
            than aural recognition, but some medieval armor.  I tried to make
            rules seem to contradict this    a lujvo for each, but we're obvi-
            trend.                           ously talking 6-7 terms - really
              JCB also felt that the         ugly!
            beginning of the chemical element  I had already noted that
            words should reflect the interna-many/most of the gismu proposals
            tional symbol - the closest thingbeing made were for plants/ani-
            to an international 'word' for   mals, etc.  The limits on this

                                      82
set were effectively infinite,   not conflict with another
but gismu space wasn't.  There   'official' meaning of the
did not seem to be any way to    wordform.  This is also a
determine which plants or animalssolution when you want to use an
should get gismu.                existing word but are afraid that
  I thus proposed to JCB a seriesyour place structure usage may be
of 4 steps to borrowing words.  Itotally unlike the dictionary
still stand by these steps,      definition.
though even in Lojban we haven't    Though not an approved
gotten past the third for any    practice, you could even make
words yet, (and shouldn't).  The le'avla in the form of gismu or
element words and the culture    lujvo if you mark them with
words are the most likely        "za'e".  The permitted word-forms
candidates to be the first       for le'avla are defined primarily
le'avla of the fourth step:      by exclusion (it can't break down
                                 into two words, it can't be a
  1. Most borrowings are little  lujvo, or a gismu, it can't fail
more than names, and indeed are  something called the "*slinku'i"
used as sumti.  Thus to use a    test), and coining nonce words is
current example, la kromium. willdifficult, so this freedom is
do for most instances of the     worth something for spontaneity
concept "Chromium".  In a rare   in the use of Lojban by non-
instance where you need to use itfluent speakers.  I do not recom-
in a selbri, you have "me la     mend intentionally invading lujvo
kromium."                        space with le'avla because, even
                                 in a nonce lujvo, the listener
  2. When a borrowing will be    will presumably try to take the
frequently used as a selbri, you word apart into component rafsi.
want to coin a word, but don't   But let's face it; the people
want to go through the 6 (or 8 inmaking nonce le'avla will often
old Loglan) languages research   be less than expert, and "za'e"
effort.  So you just make up a   allows a good bit of margin for
nonce word, probably borrowing   error.
from your native language word      It turns out that our design
for the concept, and then OVERTLYmade "za'e"-form le'avla a bit
mark it as a borrowing.  The     useless anyway.  "za'e" will now
marker was to be an unassigned   be used more with lujvo than with
cmavo, probably from the then    borrowings because it turns out
partially unused 'hV' set.       that it is virtually as easy to
   This proposal survived into   make step 3 le'avla as step 2
the initial Lojban design.  A    nonce forms as described next.
cmavo "le'a" (no longer used for Because step 3 forms are limited
this purpose) would mark the fol-to specific fields, and the
lowing word as a nonce borrowing.method for making them is so
This particular version lasted   simple. it is not necessary to
until a couple of months ago whenmark these with "za'e".  (It is
John Cowan proposed the general- of course permitted and may be
ization to mark ANY nonce word   recommended that you do so if you
usage using (instead) "za'e", noware a using a word in a field in
in selma'o BAhE.                 which you are non-expert.  This
   This current design says that is like putting quote marks
you coin a word, which must be a around the word to show that you
legal brivla (a lujvo or le'avla,may be being non-standard in the
though one could coin an         technical terminology.)
unofficial gismu as well, in     
theory) and not break down into    3. Most of the le'avla you see
multiple words.  Marking this    nowadays are step 3 le'avla.
word by preceding it with "za'e" These are names for concepts in
means that you have just now     specific semantic fields,
coined the word, it may or may   Lojbanized into brivla-form by a

                                      83
            most trivial process, and then   sound stream is ambiguous and the
            marked with a classifier rafsi.  listener hears "paslinku'i", a
            Originally I proposed that       valid lujvo.  Since le'avla are
            classifier rafsi go on the end,  defined so as never to cause
            making things look like lujvo:   conflict with gismu or lujvo,
            "kromium-xuki" (chromium-        this form of le'avla is invalid.
            chemical).  This is the proposal   The virtue of step 3, then, is
            that was printed in UL2, the     that almost anyone can make nonce
            early version of JL from before  le'avla with minimal learning.
            the Institute/la lojbangirz.     The resulting words are known not
            split.  The reaction to UL2 was  to fail the "*slinku'i" test,
            that people did not like the     they are flagged so that a lis-
            ending rafsi partly because thesetener knows he/she is hearing a
            rafsi were CVCV form and were    le'avla, that the word is some
            thus a second type of rafsi that kind of nonce word, and that it
            had to be memorized.  I then cameis restricted to a specific
            up with the current design, whichjargon field which is identified.
            is described lightly in the      This is often all that is needed
            Synopsis.  John Cowan has        - since someone familiar with the
            codified the current process for jargon field will recognize the
            step 3 le'avla, and proposed a   borrowed portion, and someone who
            new list of chemical element     doesn't can ask.
            le'avla that are presumably valid  Step 3 and step 4 le'avla can
            by that process.  (The list is   be used in lujvo.  However they
            too long to reprint here.)       are always joined to adjacent
              Simply, the step 3 le'avla     rafsi by the hyphen syllable
            process is to use any standard   "iy", and NO letters are deleted
            rafsi, or even more than one likefrom the le'avla:  the "rafsi"
            a lujvo (with some restrictions),form of a le'avla is the le'avla
            as a classifier.  You then take aitself.  The reason is shown by
            Lojbanized form of the word to bean example Nora invented while
            borrowed, which must have:       reviewing the 4th edition of
            - a final vowel                  Loglan 1 (1989), which had the
            - no letter 'y'                  type 4 le'avla "protoni" and
            - no impermissible medial        "*nukli" (not valid in Lojban,
            consonant pairs.                 hence the asterisk).  Without
              The classifier is 'glued on'   "iy" hyphenation in le'avla-based
            with a vocalic consonant 'l',    lujvo, the le'avla compound
            'n', or 'r', which also conve-   "*protonynukli" breaks into rafsi
            niently makes the first consonantas "pro-ton-nukli" and is thus
            cluster in the word NOT a        invalid.
            permissible initial ('m' would be  For easing recognition of
            acceptable except when it might  le'avla compounds, it has become
            form the consonant clusters "ml" standard to surround the "iy"
            or "mr").  Most of the problems  joints with close-commas is
            in le'avla coining result from   print, ensuring that the hyphen
            the possibility of parts of the  is treated as a separate syllable
            word absorbing sounds that are   and aiding visual recognition.
            supposed to be part of adjacent  An example is:
            words, with the combination then "djarspageti,iy,sanso"
            breaking up into different words ("spaghetti sauce").
            than you intend.                 
              The most well-known (because     4. There are as yet no official
            difficult to check for) such     step 4 le'avla in Lojban,
            le'avla problem is that associ-  although some will probably be
            ated with the aforementioned     proposed as samples when we
            "*slinku'i" test - "*slinku'i"   publish the reference book in a
            can be seen to be an invalid     few months.  An example might in-
            le'avla because if you use it    clude "protoni" ("proton"), as
            with "pa":  "pa slinku'i", the

                                      84
mentioned above, which breaks no endeavor, that it violates Zipf's
rules.                           law to have such a frequent word
  These words should be made by  be as long as step 3 le'avla must
people skilled in Lojban word-   inherently be.  Step 4 le'avla
making, and familiar with the    are a formal alternative to
previous body of such words to   having Lojban suffer the
prevent conflicts.               irregular Zipfean shortenings
  The words need no classifier   that occur in natural language -
rafsi, and may utilize any of thelike "teevee" for "television".
valid le'avla wordform space.  As  With no usage history yet,
stated above, there is no simple we've never bothered to make
algorithm for this space, and    standards for step 4 le'avla.
making these words correctly is aRecent discussion of culture
trial-and-error-aided-by-growing-words suggests that any culture
experience process.              whose name acquires any
  Lojban currently disallows     significant use in Lojban will
le'avla from invading gismu spaceget a step 4 le'avla, giving more
to allow type 4 words like       equality with the historical
"*nukli".  This is primarily an  culture gismu.  This will then
aesthetic principle, since gismu answer most, if not all of the
like the culture words are in    criticisms of the cultural gismu.
effect just such le'avla.  But we
call them gismu, and they gain   _________________________________
the key advantage of gismu-form  ________________
in having shorter rafsi.         
  (A secondary principle which                   
necessitates that any such         The Culture gismu Revisited:
le'avla be made only under       Cultural Neutrality and the gismu
tightly controlled conditions, is              List
that gismu must be prevented fromby John Cowan and Bob LeChevalier
what is called 'packing', being  (with comments by Arthur Hyun and
so alike in sound that noise or            Bob Chassell)
slight errors in pronunciation   
makes them easily misheard.  Even  The following breaks down the
with our experience in gismu     cultural gismu, by category.  All
making and with relatively simplethese words end in "-o" and we
rules on packing, we've needed a attempted to draw them from the
computer check on the gismu      relevant language directly rather
making process that has found    than being manufactured through
many conflicts missed by this    the usual process of combining
tricky test.)                    the six source languages.  Some
  We have few useful standards   words fall into more than one
yet for deciding that a word de- category.
serves this privilege enough to  
invade gismu space, other than   1) Lojban itself:
the class decisions that were      lojbo   Lojbanic
made for culture words and       
"cmavo" and "lujvo" which are    2) The words for six source
themselves self-borrowings from    languages used in Lojban gismu-
malglico Lojban tanru - "cmalu-    making:
valsi" and "pluja-valsi" for       jungo   Chinese   glico
JCB's English "little words" or  English
"LWs" and "complexes" or "Cpxs".   xindo   Hindi spano   Spanish
So for a while at least, the bar   rusko   Russian   xrabo
against gismu-shaped le'avla willArabic
continue.                        
  The qualification for a step 4 3) The words for six other widely
le'avla is that it be a word used  spoken languages that were on
sufficiently often, probably       the list of candidates for
outside of a single field of       gismu-making:

                                      85
              bengo   Bengali   fraso        4)              xurdo   Urdu
            French                             is the name for Hindi written
              dotco   German    ponjo          in Arabic script.  It is
            Japanese                           culturally unacceptable to the
              porto   Portuguese     baxso     Moslem speakers of the language
            Malay-Indonesian                   to refer to it as Hindi,
                                               although linguists classify
              (The word for Japan is from      them as the same tongue.
              "Nippon"; legal Lojbanizations 
              of that word starting with 'n' 5) Large countries (gugde) which
              would not have been              speak any of these 14
              pronounceable by the Japanese    languages, where their names
              with buffering the consonant     differ from the language names:
              cluster.  The word for Malay-       5a) glico gugde:
              Indonesian is from their word    merko   American  sralo
              "bahasa" for "language",       Australian
              because they have no word for    brito   British   kadno
              their common heritage other    Canadian
              than that one.)                  skoto   Scottish
                                                  5b) spano gugde:
                                               gento   Argentinian    mexco
                                             Mexican
                                               xispo   Hispanic (generic term)
                                                  5c) rusko gugde:
                                               softo   Soviet (The current
                                               upheaval in the Soviet Union
                                               may affect the usefulness of
                                               this word, and may require a
                                               few new rusko gugde words for
                                               seceding states.  We do not
                                               expect to make official changes
                                               before the reference book is
                                               published.)
                                                  5d) xrabo gugde:
                                               jerxo   Algerian  misro
                                             Egyptian
                                               rakso   Iraqi jordo   Jordanian
                                               lubno   Lebanese  libjo
                                             Libyan
                                               morko   Moroccan  sadjo   Saudi
                                               sirxo   Syrian
                                               filso   Palestinian    (include
                                               d for historical reasons and to
                                               be neutral in the ongoing cul-
                                               tural dispute in the Middle
                                               East
                                                  5e) baxso gugde:
                                               bindo   Indonesian     meljo
                                             Malaysian
                                                  5f) porto gugde:
                                               brazo   Brazilian porto
                                             Portuguese
                                                  5g) xindo gugde:
                                               kisto   Pakistani xindo   India
                                               (The Hindi name for India -
                                               "Bharat" could not be used due
                                               to rafsi packing.
                                             
                                             6) The continents of the Earth:

                                      86
  friko   African   dzepo          merko bangu     merbau American
Antarctic                          English
  xazdo   Asiatic   sralo          merko kulnu     merklu American
Australian                         culture
  ropno   European  bemro   North  merko gugde     mergu'e     ter
American                                           ritory of
  polno   Polynesian/Oceanian)                     American people
  ketco   South American           merko jecta     merjecta    ter
                                                   ritory ruled by
7) A few smaller cultures with                     the American
widespread historical or cultural                  government
influence:                         merko natmi     mernai American
  xelso   Greek xebro   Hebrew                     ethnos/ nation
  latmo   Latin srito   Sanskrit   merko turni     mertru American
                                   government
8)              semto   Semitic    merko ke ralju lidne
(Judeo-Arabic)                              merli'e  American
is a major language family         president
encompassing two of the major    
cultures included in the list.     These will typically used as
"Indo-European" is               lujvo by assigning merko as a
internationally a compound word, gismu, and ensuring it has a
and was not given a gismu.       rafsi, we make it easier to do
                                 so.  Such words will be shorter,
9) The major religions:          and hence will satisfy the need
  budjo   Buddhist  xriso        of speakers who want/need them.
Christian                        We have taken the cultures that
  xebro   Hebrew    muslo        are part of Loglan/Lojban's
Islamic                          definition as being most populous
  jegvo   Jehovah/Yahweh(-ist)   for such gismu.  This is NOT for
          = Judeo-Christian      the purpose of denigrating Bul-
  dadjo   Tao(-ist)              garia, Hungary, Persia/Iran, Swe-
  (Note that the deities of theseden, nor Kurdistan or Moldavia.
  religions are NOT represented  While none of these have gismu,
  by the gismu.  "la jegvon." canand hence do not have rafsi, they
  be the Judeo-Christian deity.  can be made as le'avla, and those
  Note that while the Islamic    le'avla can be used in lujvo too,
  deity "Allah" is considered thebut they won't be as short.
  same as the Judeo-Christian      Typically, as a Type 3 le'avla,
  one, cultural reasons require  these will incorporate the second
  "la .alax."                    term of the above tanru as a
                                 classifier.  For example, in the
All cultural words have the placecase of Iran (we'll presume Farsi
structure:                       as the logical choice for
                                 borrowing):
x1 is <adjective> in property x2 
                                 rupnrfarsi          fepnrfarsi     
but these words are expected to  banrfarsi
be seldom-used as bare selbri.   gugrfarsi kulrfarsi turnrfarsi
Instead, they will be used in    etc.
tanru and lujvo.
  The primary cultural
tanru/lujvo are the obvious ones.
For example:

  merko rupnu     meryru'u    Ame
  rican dollar
  merko fepni     merfe'i     Ame
  rican cent/penny

                                      87
              Things only get moderately     gismu were the shortest content
            'inequitable' when you have to   words they should be used for the
            make 'real' lujvo out of these   most frequent concepts.  He made
            le'avla.  If a Persian rug is a  gismu for most of the first 1000
            "kulrfarsi lolgai" as a lujvo it concepts, unless there was an
            becomes "kulrfarsi,iy,lolgai"    obvious 2-term lujvo based on
            where as an "American rug" (if ithigher frequency words.  He then
            rated a lujvo) might be the      continued to the 2000 and 3000
            shorter "merlolgai".             concept levels, and ended up with
              The set of gismu is certainly  about 750 gismu.
            in one sense arbitrary - I can't   From 1962-82 this list grew to
            state any external standard      about 950.  Because there were no
            justifying the entire selection, le'avla in the language design at
            and indeed we do not claim       that point, all of the elements
            perfectly objective judgement.   were added as gismu, and many
            But I still claim that for all   other rather idiosyncratic words
            practical purposes the set is    like 'billiards'; if someone
            culturally neutral.  Such a claimwrote something in Loglan and
            is always relative - there could needed a word, a gismu was often
            me 'more perfect' neutrality in  the result.  After the 1982
            theory; I think we did a good    revision of the language, there
            job, and I do not think the list was the capability for le'avla,
            is 'slanted towards a particular and some of these gismu were
            culture', unless that culture is removed, but JCB's Loglan still
            the non-existent Lojban culture. has a lot of historically
              The set of gismu were derived  idiosyncratic gismu which are
            over 35 years.  Jim Brown        gismu only because they had no
            selected the first set based on 3obvious 2-4 term tanru/lujvo 20
            or 4 sources, including BASIC    years ago.
            English, some studies of words     This is the list we inherited
            that are 'biologically primitive'when we remade the list for
            in that they appear to be        Lojban.  Among the words were
            primitive in most every language,culture words for the 8 source
            etc.  He then used the Helen     languages for Institute Loglan,
            Eaton study of the most          (as well as separate gismu for
            frequently used concepts in 4    the people and the culture) plus
            languages                        some idiosyncratic cultures that
            (English/German/French/Spanish). had been added haphazardly,
            This list is of course European  including Italian, Scottish,
            biased, but it is the only such  Roman, and Amerind.  We decided
            comparative study across several to regularize the set based on
            languages for word/concept       some external standard - the
            frequencies, and Helen Eaton was culture words we used were those
            doing so for AL research and was for JCB's 8 languages, and the
            presumably aware of the          other 4 we considered for Lojban
            neutrality issue.  In any case,  (we once planned to use 12
            there is reason to believe that  languages instead of 6, then cut
            the list is more biased in its   back to 6 for several reasons).
            obsolescence (being 60 years old)We added the religions that were
            than it is toward a specific     primary in the source cultures,
            culture - key concepts in scienceand separate words for the sev-
            and medicine are unknown in the  eral countries that used the
            list, while certain concepts no  source languages.  Because we had
            longer important rate highly.  Itle'avla, if we could not assign a
            is still a standard, and the onlygood rafsi to any recognizable
            one.                             form of a country's culture word,
              Brown assumes that Zipf's law  we left it out - the assignment
            holds.  Zipf noted that word     of a short rafsi was the main
            length was inversely proportion- justification for these words.
            ate to word frequency.  Since    

                                      88
  The point of all this is that  cover any concept that is
the culture words were added     important across cultures
according to a standard that is  (reasonable I set at about 4
inherent in the history of the   terms, the longest lujvo ever
language and its design - thus nomade and used as a 'real word'
one really had to be an 'objec-  for Loglan).  Words that are
tive judge'.  If it is accepted  specific to one culture, or are
that our "Chicken MacNuggets"    part of the international
word formation algorithm is      vocabulary of science are
culturally neutral, being based  relegated to le'avla.
on 6 languages, then the culture   BUT, in going to this
words meet the same criteria of  definition of our gismu coverage,
neutrality.  In addition, the    we did not claim the need to
words are not slanted towards oneeliminate every gismu that had no
culture - if so, we would have   obvious intercultural use.
not used the Egyptian word for   Indeed, if it was already made as
Egypt, the German word for       a gismu, we kept it unless
Germany, etc.  Yes, we had to    someone explicitly proposed its
leave some cultures out, and somedeletion accompanied by (usually)
countries that have speakers of  a 2-term lujvo for the concept.
the languages we do have.  But   About 20 odd words were so
the decision was not wholly      deleted before the baseline.
arbitrary.                       There is no intent to delete any
  The rest of the gismu were     gismu prior to the 5-year usage
selected to complete various     baseline, because the only
incomplete sets recognized by a  meaningful criteria now are that
Roget-like study of the gismu by would justify a deletion in the
Paul Doudna.  Later, when        baseline period would be
Athelstan joined the project, we something like the word being
conducted two further reviews    impossibly vague (not likely
against Roget's Thesaurus lookingsince we have place structures
to achieve 'completeness' in thatfor each).  Arguments of usage -
the gismu could be used to form  either potential or actual are
lujvo covering every concept in  irrelevant; that is the point of
Roget.  Roget is of course En-   a usage baseline, to see whether
glish-biased, but it also        they are used.
purports to be a comprehensive     As a result of this long
survey of the semantic word spaceevolutionary process, it is clear
and it is in that mode that we   that the list is not an arbitrary
used the list.                   representation of one or two
  In the course of doing so we   persons' biases.  Being based on
recognized that the rationale forthe concept of 'semantic space',
gismu has changed since JCB firstwith some verification of
started Loglan (and in his       usefulness in a few cultures, the
versions this is also true,      list is close to comprehensive
though he has never so-stated).  (occasional new words will be
At one point Brown thought his   proposed when we find a gap).
words were in some absolute senseThe list is not angled towards a
'primitive', partly based on his specific one or even identifiable
biological primitive research.   set of cultures, except that if
This is not the current practice some culture has a truly
in assigning gismu.  gismu are inimportant concept that is not
no way assumed to be the 'most   shared by any of the Eaton lan-
basic', 'most important', or     guages, it may currently be
'most' anything for one or       omitted.  In which case, it will
several cultures.                likely become a gismu later when
  We now claim ONLY that the     recognized.  Beyond this, I do
gismu we have are sufficient,    not see the claims that the Loj-
using the lujvo-making rules to  ban list is biased in some
make reasonable length lujvo to  recognizable way towards any

                                      89
            language.  It can only be claimed  developers, sufficient to meet
            that it is possibly biased away    what they think are the needs
            from some less common              of enough Lojban speakers and
            languages/cultures in the most     experimenters for Lojban to
            trivial sense, since we are        meet the purposes of the
            talking about exactly one word     language;
            per such culture.  No doubt if   
            any of these less common cultures* to extent:
            develops a significant Lojban      as best the developers can,
            speaker base during the formative  particularly with reference to
            years of the language, the         Zipfian and other practical
            culture will be assigned a gismu.  concerns.
              The remaining element words may
            plausibly be biased towards      Arthur:
            English or at least toward Euro-   Whoever has to judge what is
            pean cultures.  These were       "important" or less so will have
            justified by their use in        a bias.
            metaphors before we had the now  
            clear policy against heavily     Bob Chassell:
            figurative metaphors.  Even so,    Yes, indeed.  So therefore,
            there are ways to define these   whoever judges what to include is
            words based on the properties    obligated to do the best she can.
            adjectivally attributed to the   Since there is little desire to
            substance in non-technical       give everyone his own gismu, the
            compounds, leaving the 'chemical question is "to whom to give
            word' either for a lujvo (using  gismu?"  If the selected gismu
            "curve" or "xukmi") or a le'avla.turn out to be sufficient for
            Thus "nikle" is fine as is,      enough of those who are experi-
            "romge" is any highly reflective menting with and speaking Lojban,
            non-tarnishing metal, "navni" is then the selection will have been
            an inert gas (this one change wasOK.
            adopted officially at LogFest),    Of course, the decision is not
            "kliru" can be used for all the  easy; to me, the best beginning
            halogens, (people put kliru      is to use three criteria:
            tergu'i on their cars), etc.     population, wealth/power,
            This eliminates the most obvious frequency of occurrence in the
            part of the bias, but more       kinds of speech the language
            importantly allowing the words todevelopers anticipate for Lojban.
            be useful.                       These criteria are obviously
            _________________________________unfair, and they have a bias; but
                     _______________         enough Lojban speakers may be
                                             sufficiently satisfied by them
            Arthur Hyun comments:            for Lojban to succeed.
              The point is that if anyone    
            wishes to maintain the pretense  Arthur Hyun:
            of Lojban being "culturally neu-   Therefore, either give up
            tral", then every culture is     trying to claim "neutrality" or
            going to have to be considered   treat them all the same.
            the same.                        
                                             Bob Chassell:
            Bob Chassell responds:             This statement is a
              The consequence does not followmisunderstanding of what neu-
            from the `if' clause.  The       trality is about.  Neutrality is
            predicate "culturally neutral"   not and cannot be absolute.
            has several places so far        Consider as an analogy, Swedish
            unexpressed.                     neutrality during WWII.  At the
                                             beginning of the war, when it
            * by standards:                  appeared that Germany would
              in this case, standards of     become the hegemonic power in
              mostly middle-class, American  Europe, Sweden cooperated more

                                      90
with Germany than with the       
Allies.  (Remember:  the invasion  Another way to tackle the issue
of Norway was designed to protectis to turn my questions around:
shipments of Swedish iron ore in 
coastal freighters; very likely, * Is the current design of Lojban
the Swedish mining regions would   too culture bound for
have been occupied by the Germans  experiments relating to tense
if the Swedes had not              in grammar to succeed?
cooperated.)  When the Allied    
coalition was seen to be winning,* Is the current design of Lojban
Sweden cooperated more with it,    too culture bound for speakers
permitting, for example, people    in Asia, Africa, and Europe to
such as Niels Bohr to be flown     accept?  (One possibility is
out of Sweden by the British.      that Lojban's logical bias may
  Similarly, Lojban cannot be      be more significant than its
absolutely neutral, neither as a   other biases.)
practical matter nor as a matter
of being able to define what
`absolute neutrality' means.
(Obviously, to me at least,
`treating them all the same' is
NOT neutral, but is very biased;
is that not apparent to you?)
  Lojban has several goals
relating to cultural neutrality:

* to serve as a vehicle for
  intercultural experiments;

  For Lojban to be useful, this
means that the undesired
influences of Lojban should both
be small enough for the
experiments and be less than the
undesired influences of
alternative vehicles, such as
Urdu or Esperanto.
  If you are testing some group
whose culture lacks a gismu,
perhaps you would invent a nonce
gismu; not all five-letter forms
are currently used, just as not
all shorter forms are assigned
cmavo.

* to serve as an international
language;

  For Lojban to be useful, this
means that enough people must not
be overly offended by Lojban's
biases.

* to serve as a test for `logical
language';

  This means that Lojban must be
biased towards logic in some way,
which biases it against languages
such as English.

                                91
                   le lojbo se ciska (continued)

And now, how about some poetry?

     Michael Helsem sent me a translation of the first stanza of
the Esperanto Hymn into Lojban.  The original (1), and Michael's
English translation (2) are given first.  Then comes his Lojban
translation into a limerick!  He made only minor errors, none of
them grammatical (e.g. the place structure of fasnu has changed,
making it non-useful for his purposes.  So my correction of his
translation (3) is immediately after with Michael's intended
English translation (4) along with what he actually said.  Finally,
I have a pronunciation guide for those who want to read it aloud.

(1)        La Espero      (2)

En la mondon venis nova sento,     Into the world came (a) new
feeling,
Tra la mondo iras forta voko  through the world goes (a) strong
call;
per flugiloj de facila vento  by (the) wings of (a) gentle wind,
nun de loko flugu gxi al loko now fly it from place to place.

(3)
le cnino menjvinu goi ko'a
pu selfri le terdi ku .o'a
.i ta'i lo brife
ku leko nunjimpe
ca selbei fi role selvo'a

(4)        (Intended)                 (Actual)
A new mind-view           A (specific) new mind-view (ko'a)
happened on the Earth (pride!)     is-experienced-by the Earth
(Pride!)
Like a breeze             Having-form a breeze
let your understanding    Your (imperative! - Make it true!)
thing_understood
be transferred to all the voiced-ones.  now is-sent to all-the-
voiced_ones.

(5)
 le  cnino       menjvinu         goi ko'a
/leh,SHNEE,noh   mehn,ZHVEE,noo   goi,KOH,hah/

 pu  selfri      le  terdi        ku .o'a
/poo,SEHL,free   leh,TEHR,dee     ku. OH,hah/

 .i  ta'i       lo brife
/.ee,TAH,hee    loh,BREE,feh/

 ku  le  ko     nunjimpe
/koo,LEH,koh    noon,ZHEEM,peh/

 ca   selbei     fi  ro  le       selvo'a
/shah,SEHL,bei   fee,ROH,leh      sehl,VOH,hah/

     But I wasn't happy with this.  Not only might some
Esperantists take offense at having their song of great traditional
importance turned into a limerick (which in English culture is
almost universally used for non-serious verse), it was not a very

                                92
exact translation of the Esperanto.  At Nick Nicholas's
instigation, I've lately been translating a bit of poetry, so it
was only natural that I tried to do this one.  I was surprised to
find that my result was rhythmic (though not the exact same rhythm
as the original), and it was fairly easy to make it rhyme a little
bit too.  My new translation is given as (6), and I've added
pronunciation and translation (7).


                                93
(6)
                      la'ede'e cu se pacna  .i

le cnino selcinmo cu se lifri loi terdi
.isemu'ibo fe'eroroi vliki'a
.i .e'o ko sepi'o le mlibrife voknalci
fairbevri ru rode da'ada

(7)
                      The following is-hoped-for

The new thing-emotionally-felt is-experienced-by of-the-mass-of-
the-Earth
Therefore motivationally, everywhere powerful-cry_out
(Petition!) (imperative!) You, usingly, mild-breeze voice-wings
distribute-carry (the earlier) it from every-something to all-
other-somethings.

 le  cnino        selcinmo           cu   se  lifri       loi terdi
/leh,SHNEE,noh    sehl,SHEEN,moh     shoo-seh,LEE,free
loi,TER,dee/

 .i  se  mu'i    bo      fe'e    ro  roi     vliki'a
/.ee-seh,MOO,hee-boh     feh-heh,ROH,roi     vlee,KEE,hah/

 .i   .e'o        ko  se  pi'o       mlibrife             voknalci
/.ee  .EH,hoh     koh-seh,PEE.hoh    mlee,BREE,feh
vohk,NAHL,shee/

 fairbevri               ru  ro  de          da'a    da
/fai-rr,BEHV,ree         roo,ROH,deh         dah,HAH,da/

     But I didn't stop there.  I'd been carrying the seeds of an
idea for an ode to Lojban, and this exercise got my writing urge
going.  Within an hour, I had written my first original Lojban
poem, which can be seen to be both rhythmic and, with poetic
license, rhyming.  This is given as (8) with pronunciation and
translation in (9).
     This is the first 'original' poem I've written since 6th
grade, some 25 years ago.  I'm actually quite pleased with it, and
with the fact that it was so easy.  Maybe I'll even write more (and
inflict it on the world).  But one thing is for sure:  if >I< can
write Lojban poetry, anyone can.

(8)
doi bangu co satci joi ji'a na'e satci
do'u dunda le ko selkai co sidbo velcusku
loi terdi selvo'a noi nitcu lo nintadji
be lo nu menbenji fo loi prenu co simxu

.i doi selmenli co cfipu be le so'imei poi djica
leka jimpe do'u bevri ko le cfari lei zgatirna
.i .e'o ko cu klina selcusku gi'e jicla
le no'e farvi pe ri'a lenu seljimte loi malgerna

(9)
 doi bangu              co   satci         joi  ji'a           na'e
satci
/doi,BAHN,goo           shoh,SAH,chee      zhoi,ZHEE,hah       nah-
heh,SAH,chee/

                                94

 do'u    dunda          le  ko  selkai     co   sidbo
velcusku
/doh-hoo,DOON,dah       leh-koh,SEHL,kai   shoh,SEED,boh
vehl,SHOO,skoo/

 loi terdi              selvo'a            noi nitcu           lo
nintadji
/loi,TEHR,dee           sehl,VOH,hah       noi,NEE,choo        loh-
neen,TAH,jee/

 be  lo  nu             menbenji           fo  loi prenu       co
simxu
/beh,LOH,noo            mehn,BEHN,zhee     foh-loi,PREH,noo
shoh,SEEM,khoo/

O, language of-type exact mixed-with-in-addition non-exact
(something) gives your (imperative - make it true!) characteristic
of-type idea expression-mode
to of-the-mass-of-the-Earthly voiced-ones, who incidentally need
the new-method
of events-of mind-sending to of-the-mass-of-persons of-type mutual.

O, language, both exact and non-exact,
Let your characteristic mode of idea-expression
be given to the voiced ones of the earth, who need this
new method of mutually communicating between minds.

 .i  doi selmenli       co   cfipu         be  le  so'imei     poi
djica
/.ee-doi-sehl,MEHN,lee  shoh,SHFEE,poo     beh-leh,SOH-HI,mei
poi,JEE,shah/

 le  ka  jimpe          do'u    bevri      ko  le  cfari       lei
zgatirna
/leh-kah,ZHEEM,peh      doh-hoo,BEH,vree   koh-leh,SHFAH,ree   lei-
zgah,TEER,nah/

 .i   .e'o              ko  cu   klina     selcusku            gi'e
jicla
/.ee  .EH,hoh           koh-shoo,KLEE,nah  sehl,SHOO,skoo      gee-
heh,ZHEESH,lah/

 le  no'e    farvi      pe  ri'a    le     nu  seljimte        loi
malgerna
/leh-noh-heh,FAHR,vee   peh,REE,hah-leh    noo-sehl,ZHEEM,teh  loi-
mahl,GEHR,nah/

O thought-about-things of-type confusing to-the-many-some who
desire
understanding-ness,  (something) carries you (imperative! - make it
true!)
      from-the initiators to-the observing-hearers (listeners)
(Petition!) You (imperative!) be clearly-expressed-things, and stir
the neutral-non-developing (stagnant) which-are-caused-by the-
event-of being
      limited-by Derogative-Grammar.

O ideas which confuse the many who desire understanding,

                                95
Let yourself be carried from those who initiate you to those who
listen.
Please be clearly expressed, stirring
the stagnation caused by the limits of (Ugh!) Grammar.

___________________________________________________________________
                 _________________________________
Another poem.  Well, actually part of a song.  The following is
Mark Shoulson's first Lojban translation attempt:  the beginning of
the Doors' song People Are Strange.

loi prenu cu cizra       .inaja do ca fange
.i loi flira cu simlu to'e melbi   .inaja do ca na se kansa
.i loi ninmu cu mabla simlu   .inaja do ca na se djica
.i loi klaji cu to'e xutla    .inaja do ca badri fa'o

The English:
People are strange       / When you're a stranger;
Faces look ugly          / When you're alone.
Women seem wicked        / When you're unwanted;
Streets are uneven       / When you're down.

Bob's comments (mostly answers to questions from Mark):
     A most excellent first effort!  Bravo!  I find little to
     complain about.  I might have used the less literal "pluta"
     instead of "klaji", or even "tadji", thus conveying what I
     think the intent is better, but this is 'stylistics', which
     Lojban has none of yet.  Of course, Nora found one big mistake
     I missed (see below).

Mark:     I use 'loi' all the time.  Should it be "lo'i"?  Or
something else?

Bob: You done good!  This is about as good an English text for
     "loi" as there is.  "lo'i" would give you a set.  Sets
     generally are not ugly or wicked, and seldom are described as
     strange or uneven - but you don't want the set anyway, but the
     members.

                                96
Mark:     I don't much care for ".inaja".  I want a way to say "if
     but not necessarily only if".  I assume there's a better way.
     I copied this usage from lojbab's translation of Language. [by
     Suzanne Vega - see JL14].

Nora:     With ".inaja", what you have is "IF people are strange,
     THEN you are a stranger".  I think you wanted ".ijanai":
     "People are strange, IF you are a stranger."  This seems a
     virtually exact translation.  The Lojban statement is false
     only when people aren't strange but you ARE a stranger.  It
     makes no claims about what happens if you are not a stranger.
     (Bob: I completely missed this - which may be why I did so
     poorly in logic when I was in school.)

Mark:     Does the use of "ca" make sense?  I mean to get across
     the sense that faces look ugly if you're down at the time
     (hence the English "when").

Bob: Seems fine to me.  You could also do something with one of the
     "under conditions" modals (selma'o BAI), used exactly in the
     same place you used "ca".

Mark:     Should "mabla" in line 5 be "palci"?  Am I using "to'e"
     right?  Is there a better way to express these things?

Bob: Yes, "mabla" is the right word, given the English semantics.
     You certainly are not judging women as morally evil ("palci").
     You might use "xlali", but the English use of "bitch" in such
     situations is most certainly "mabla", not "xlali", and I
     suspect that the usage here is more suggestive of such
     cursing.  "to'e" is fine.

Mark:     Should I be using the tanru I use?  What would be better?

Bob: I would have omitted "simlu" in both the second and third
     lines, or I would have included them in all four; they are
     implicitly there.  You could have done a couple of things to
     fiddle around.  For example, the 2nd half of each line, rather
     than being ".inaja ... ca ...", could have been a subordinate
     clause attached to the selma'o BAI word for "under
     conditions":

     loi prenu cu cizra va'o le nu do fange

     or you can even take out the "do" - the English really just
     means "someone", and use the "observer" modal of BAI:
     
     loi prenu cu cizra ga'a lo fange

     or you can make all four based on "simlu", which has an under
     conditions place and an observer.  Note that "simlu" has a
     cleft place structure and may need changing.  x1 and x2 are
     redundant, and a revision would be "x1 seems to be so to x2
     under conditions x3".  The first version following uses the
     current cleft structure; the second version uses the possible
     revised structure:

     loi prenu cu simlu le ka ri cizra ku roda le nu da fange
     x1      }    simlu {x2             } x3   {x4          }
     

                                97
     le nu loi prenu cu cizra cu simlu roda le nu da fange
     {x1                    }    simlu x2   {x3          }

     To me these last seem very analytical, not poetic, and I
     prefer the "ga'a" version.

Mark:     I realize that some selma'o UI words would probably
     belong here, but I'm not positive which to use or even if I'd
     want them there.  There's something unsettling about the
     unemotionality you get without them which fits the mood of the
     song.  Or not.

Bob: I personally think they don't belong, because the author has
     used the impersonal "you"/"someone".  If the pronoun had been
     "mi", the attitudinals are vital.  There are SOME attitudinals
     that might apply anyway, and perhaps some discursives, but
     there is no vital need for any.

                                98
One last poem:

            Nick Nicholas's translation of a Greek poem

ko doi loi tarci na fegycrugunta mi mu'i lenu mi ru'inai sanga ca
le nicte  .imu'ibo le nu mi cordri kei cei broda (to .u'anairo'i
doi lemi se xe'ikre toi) gi'e klama fi le zdani gi'e bacru lu'e
broda

Translation: literal Lojban

Imperative you, O Stars [make-it] not that [you] angry-utter-attack
[=scold] me motivated-by the-event I occasionally sing during the
night.  Because the-event I am pain-sad in-the heart ((Emotional
loss!) O my black-haired one) and [I] come from the nest [=house]
and uttered the symbol for it (the event of me being pain-sad).


.i .ai mi bacru lu'e broda ga'a loi tarci noi mipri gi'eji'a na
pante (to .uinairo'i do mo'u bapli mi ti toi) ca so'ilo cacra fi le
nuntirna fe'o

(Intent!) I utter the symbol for it (the event of me being pain-
sad) observed by Stars who keep-secret and-additionally not protest
((Emotional unhappiness!) you are-at-the-completion-of forcing me
to do this-here [presumably this poem]), during many hours, about
this event-of-hearing.

Literal English from the Greek

Stars, not-imperative me you-scold that I-sing the night because I-
had pain in-the heart (ach dark-diminutive-neuter my) and I-came-
out and it I-said.

To-the stars future-tense I-say the pain my that not it they-
witness [metaphorically, bear witness] that they-have and [also]
patience (ach how me you-rendered) with the hours and they-listen.

Colloquial English from the Greek, with notes

Stars, do not scold me, that I sing in the night.  (clearly an
imperative.  The "pou" in the original is no clearer that "that",
and sorta corresponds to "va'o" or "tesau", but of course means
"for singing") Because I had a pain in my heart (apostrophe:  Oh,
my dark-haired little one), and I came out and uttered it (the
pain).

I will utter my pain to the stars, who don't betray it (present
tense; the verb literally means "witness", and implies that the
stars keep one's confidence) and which also have the patience
(apostrophe:  Oh how you've rendered me/ Oh whata bad state I am in
because of you!) to listen for hours.

___________________________________________________________________
                 _________________________________

         On Observatives         Jim Carter writes:
by Bob LeChevalier in response to  At the L.A. group meeting we
           Jim Carter            discussed "observatives",
                                 Initially we had trouble anal-

                                99
yzing the meaning of the bare    'subject') adds strong emphasis
selbri "nanmu"; we concluded thatto the selbri as the critical new
it meant "manliness is happening information being pointed out in
here", but the distinction       expressing the sentence.
between that and "a man", while    Other than this strong
obviously real, is hard to       emphasis, "nanmu" is treated as
explain.                         any ellipsized sentence is, all
                                 unspecified sumti are still
  You are correct that this is anactually there, but are
'observative'.                   unexpressed.  They thus have the
  For the benefit of those who   implicit value of "zo'e"
may have old copies of The       (something I'm not bothering to
Loglanist:  in TL3, Scott Layson specify because it isn't
(supported by Chuck Barton)      important in this pragmatic con-
proposed that the bare selbri be text.)  In the normal observative
interpreted as an observative,   case, with a physical "subject"
after the natural language usage ellipsized as for "fagri", a more
reflected in shouting "Fire"     accurate specification of the
("fagri") upon sight of smoke (assumti would be "(pointing) ta
compared to the then current barefagri").  But "ta" would cause
selbri as an imperative, a       the speaker to look at the
command).  (At that time there   pointer (to see what is pointing
was no clear way to declare an   where, not look for the fire and
observative, though "le nanmu"   run - in hearing the observative
and "da nanmu" (using current    "karce" while standing in the
Lojban words) were considered.   street, the distraction of having
(For newer Lojbanists, please    to look at the pointer could be
forgive my extensive references  fatal.)
to old Institute Loglan in this    There thus is nothing about
response.)                       looking at some "manliness is
  Scott and Chuck argued that in happening here".  That would
most languages, an imperative is either be the observative "nu
in some way inflected, whereas   nanmu" or "ka nanmu", which are
observatives are not.  We        in turn equivalent to
researched further in designing  
the Lojban version and found thatzo'e      [cu] {nu       <zo'e
children first learn to speak      [cu]            nanmu [vau]>
essentially in observatives:       [kei]}
"Mama!", "Doggy!", and occasion- Something is-an-event-of
ally in attitudinally inflected    something (else)'s    being-a-
observatives:  "Milk?"  (".au      man.
ladru").  My invention of the im-
perative pro-sumti "ko"          zo'e      [cu] {ka         <zo'e
solidified the change, and this    [cu]           nanmu [vau]>
is now one of the two major        [kei]}
identifiable differences between Something is-a-property-of
Institute Loglan and Lojban, that  something (else)'s   being-a-
is not simply an expansion of the  man.
language or a correction of      
hidden syntactic ambiguity.      where different "zo'e"s can have
  "nanmu" is an observative      different values.  (I use
because the selbri has been      "manhood" for "nu nanmu" and
atypically brought to the front  "manliness" for "ka nanmu"; it is
of the sentence.  (Indeed in thisnot clear from Jim Carter's
case it is the only thing in the example which he intends.)
sentence, but this is beside the   "A man" would be "pa nanmu",
point.  "klama le zarci" is also which is a shortened form of "pa
an observative.)  This movement, lo nanmu" "One something that
and the explicit elliptical omis-really is a man".  This is a
sion of the x1 sumti (the        sumti, not a complete sentence.

                                100
We rejected such a bare sumti as snime cu carvi" = "The snow
a version of observative, as wellrains" and "loi mlatu je gerku cu
as "lo nanmu" and "le nanmu" for carvi" = "Cats and dogs rain." -
three reasons.                   the literal statement - NOT a
  First, as incomplete sentences,figure of speech.)
the listener has to wait to be   
sure that the speaker isn't just _________________________________
hesitating before continuing with        ________________
a selbri:  "pa nanmu ...", "lo   
nanmu ...", "le nanmu ...".         Predications and Identities
These are equivalent to the             by Bob LeChevalier
trailing-off incomplete English  
sentences "One man ...", "A man    A discussion on Lojban List
..."/"Some men ...", "The man/meneventually hinged on clarifying
...".  In Lojban such incomplete the differences between 'predica-
sentences are defined to be      tions' and 'identities'.  In his
grammatical, and are typically   writings on Loglan, Jim Brown has
used to answer "ma" questions.   long stressed this distinction,
The latter two English transla-  which is basic to predicate
tions point out that Lojban      logic.
descriptors make no                It has turned out, however,
singular/plural distinction.     that in both the Institute's and
  The second reason is that the  our versions of Loglan, there is
descriptors end up being the     no grammatical difference between
first word heard, not the selbri.predications and identities.  In
Shouting "A fire!" has less      Lojban, we kept the two gram-
impact than shouting "Fire!".    matically distinct until this
  Finally, the versions with the last spring, when John Cowan
extra cmavo have just that littleshowed using his E-BNF that the
bit extra grammar and semantic   distinction was illusory.  Now,
interpretation implicit in the   identity sentences look like
extra word.  Observatives are    predications and can be
generally used in situations     understood like them, and one
where people don't want to take  must recognize them by the use of
the trouble (or in the case of   specific cmavo that indicate the
children, don't know how) to con-difference.
struct a sentence with more        Jim Carter observed that Lojban
elaborate grammar, and/or don't  "binxo" (keyword "become") could
want the listener to take the    be seen as a kind of identity
time to interpret the grammar.   claim, since in English, "become"
                                 is a future tense of "to be".
Jim Carter:                        We must clarify that "binxo" is
But we came up with a better     rather to be contrasted with
example:                         "cenba" ("vary") and "galfi"
  carvi         It's raining     ("modify").  This trio of gismu
  lo carvi      Look, raindrops  were assigned because old Loglan
                                 "cenja" ("change") which means
  The first English is a         what "binxo" does, was often used
reasonable colloquial translationin tanru and lujvo as if it meant
of its Lojban.  More exact is    one of the other two words.
"[Something] rains", or "Rain!"  Institute Loglan solved the
The second Lojban, a sumti, is   problem by misusing its "madzo"
the incomplete sentence "A rain- for the 'transitive' "change" of
ing thing/Some rain ... [is doing"galfi"; "madzo" has the same
something]", whereas Jim's       meaning as Lojban's "zbasu" (to
English would be expressed in    make/construct ... out of ...).
Lojban as "ko catlu .i carvi     We separated the English word
[dirgo]" or "ko catlu lenu carvi "change" into its three distinct
[dirgo].  (The thing(s) raining  meanings.
need not be 'drops'.  We say "lo 

                                101
binxo     changes into/becomes   intending, but not saying "Bantha
...                              is a cat".
cenba     changes/varies in
property ...
galfi     changes ... into ...

  The keywords were chosen to
maximize the distinction.
  A similar problem was
recognized with "gasnu"  "do"
last year.  As noted in the
discussion of cleft structures
and sumti-raising above, "gasnu"
is clarified to mean that x1 is
the actor/agent in an event or
action x2.  We kept the keyword
as "do" because "actor" and
"agent" are much more familiar in
English with meanings that have
nothing to do with the Lojban
"gasnu".
  "du" IS an identity
'predicate', and its morphology
alone flags it as different from
other predicate words.  It claims
that the two sumti on either side
are alternate and equivalent
designations for the same thing.
Translate it best as the
mathematical "=" sign.
  "du", other than in a
mathematical context, has a
somewhat metalinguistic effect.
It equates two labels for the
same thing.  No other words in
Lojban, other than the
relativizers "po'u" and "no'u",
and the assigners "goi" and
"cei", have this metalinguistic
effect.)
  As Lojban has grown, the role
of "du" in Lojban has shrunk.
Most noteworthy, the practice of
using "du" for self-
identification as taught in draft
textbook lesson 1, is now frowned
on:  use "mi'e. .atlstan.", or mi
se cmene zo .atlstan. (if your
name happens to Lojbanize as
".atlstan.").  It isn't wrong,
but we do not want new Lojbanists
concentrating on the use of "du"
early in learning the language.
Otherwise we get such
unacceptable statements as:

 la banthas. du  lo        mlatu
(1)
 Bantha      =   a/some    cat(s)


                                102
  Here we have a                 and more experienced Lojbanists
legal/grammatical but probably   try to answer.  We prefer to see
false statement.  "lo mlatu" is ayour attempts at expressing it,
description that can apply to a  because 1) it means you tried and
cat, or the members of any       this isn't a trivial question;
collection of cats, in the       and 2) it helps us slant our
universe of discourse (possibly  explanation to fit your needs.
including the non-domesticated     Word translations need some
species).  I doubt that there    kind of context for them to be
exists anyone that would apply   included in this feature.  Thus,
the name "la banthas." to all of some questions from Ivan
these cats.  If we were          Derzhanski in his Lojban letter
expressing the Linnean           are not answered here.
(Latin/Greek) name for the cat   
family, well, maybe ...          From Coranth D'Gryphon:
  There is a 'predication' (as     I have a few English statements
opposed to 'identity') "predicatethat I'd like the appropriate
word" that is near-equivalent to meanings for:
"du", and that is "mintu" - "x1  
is identical to x2" ("du" while  1) some people (plural of person,
etymologically tied to "dunli" is  referring to existing
not really related due to place    individuals);
structure differences).  There   2) some people (plural of person,
have been some probably            referring to hypothetical
legitimate but inconclusive        individuals);
debates about whether "du" and   3) people (the set of all
"mintu" are the same predicate.    persons, treated as a lump
Nora and I currently feel that     unit);
"mintu" can be used more broadly,4) people (the set of all
as in "this plate is the same as   persons, treated as in-
that one", when the two are        dividuals);
interchangeable for the intended 5) person (the set of all the
function.  "du" would not be       things that make up a person,
correct in translating such a      considered a unit).
statement, since presumably "this
plate" and "that one" refer to   Bob LeChevalier:
different objects.)              These 5 mass statements are
  Predicate logic does not write simple, but look carefully at the
identity sentences as predicates.quantifiers in my answer.  I have
Lojban's predicate grammar       put normally ellipsized
requires even an identity        quantifiers in brackets - they
sentence to be phrased as a      are needed to properly understand
predication.  As such, Lojban is what is going on.  Afterwards I
a mirror image of natural        summarize the default quantifiers
languages.  But the use of a     for the 4 descriptors involved.
cmavo in an identity sentence    Note that none of your examples
should alert you to the very realinvolve "lo'i" or "le'i" the set
difference between them.         descriptors. "lo'i vinji" is the
                                 set of all things that really are
_________________________________airplanes, and does not relate to
        ________________         their components.
                                 
          How to say it          1.   su'ore lo [ro] prenu
     A New Regular? Feature        or
                                   su'ore le [su'ore] prenu
  We are happy to take good ideas  or
from the Loglan Institute, and     su'ore da poi prenu
are thus instituting (!?) a new  
feature, where you ask how do say2. Either of the first two above
something which is not obvious,    for 1; "da poi ..." claims

                                103
  actual existence of something  Compare this with:
  that meets the restrictive     
  bridi that appears after "poi".su'o lo ci mela studjez.
  The distinction between "lo"   su'o            lo             ci
  and "le" is that "lo" refers to                 mela studjez.
  things that have the relevant  at-least-one of the-set-of-all-
  property, whereas "le" refers    Stooges (which set has
  only to the speaker's intended   cardinality 3)
  referent which is presumed to  
  be understood by the listener  ro le su'o prenu
  or the speaker would have givenro      le
  more information to restrict   Each of the-set-of-things-that-I-
  the referent.  We have also    describe-as
  given the two descriptors        su'o prenu
  different default quantifiers,   persons (which set-in-mind has
  as shown in the bracketed           cardinality at-least-1)
  values above.                  
3.   piro loi [ro] prenu         pisu'o loi ro prenu
4.   ro lo [ro] prenu            pisu'o           loi
  [ro] le ro prenu                                ro prenu
5.   [piro] lei pa prenu         at-least-some of the-mass-of-all-
Using "loi' with "pa" after the    who-are-persons (cardinality
descriptor like this would         'all')
incidentally claim that there is 
only one person in the universe).piro lei su'o prenu
                                 piro   lei
  The big 'secret' in all of thisall of the-massified-set-of-the-
is the default quantifiers - the   things-that-I-
numbers inside and outside of the  describe-as
descriptor.  The inside number                  su'o prenu
enumerates the set meeting the                  persons
description, while the outside     (cardinality 'at least 1')
quantifiers selects from that    
set.                             le/lei/le'i must have at least
  Examples:                        one in the set.
                                 lo/loi/lo'i need not have any in
su'o lo ro prenu                   the set (in which case the
su'o            lo                 "su'o" means "at least 0" since
                 ro prenu          "ro" is also = "0").
at-least-one of the-set-of-all-  
  who-are persons (which set has   In normal usage, all of the
  cardinality 'all')             above implicit quantifiers are
                                 left unstated.  You only put in a
                                 quantifier if it differs from the
                                 default value.  The resemblance
                                 of "lo" to English indefinites is
                                 purely a result of our choice for
                                 the implicit quantifier.  In
                                 JCB's Loglan the equivalent word
                                 was "lea" which had the default
                                 quantifier "ro *lea ro prenu"
                                 ("all of the set of all who
                                 really are persons") which is
                                 only useful for logically risky
                                 universal claims, whereas "lo" is
                                 useful for indefinites, where the
                                 speaker has no particular
                                 referents in mind.  But "lo" is
                                 still not quite the same as
                                 English indefinites ("a" or

                                104
"some" as articles).  If you haveHow to say: "bias", as in "x is
even the slightest restriction onbiased/directed/ influenced in
the set of persons being         direction y by applied force z";
described and do not make the    
restriction explicit with        John:
poi/pe/po'u etc., the you should I'm not sure if the physical or
use "le" instead of "lo", and usethe metaphorical sense of "bias"
explicit "su'o" to replace the   is wanted here.  For the latter,
implicit outside quantifier "ro":"se xlura" = "x1 is influenced by
"su'o le ro prenu" ("some of all x2 to do or be x3 under
persons that I have in mind") -  conditions x4" seems to do the
usually shortened to "su'o le    trick.
prenu".                          
                                 Dave:
Dave Cortesi:                    How to say: tend, as in x tends
I ran off down the following     toward y (naturally, of itself)
dead-end alleys and would        
appreciate anybody's comments on Bob:
how to escape them...            "jinzi" = "x1 is the innate
                                 property of x2"
-  How to say: "habit" and/or    The other portion of the tanru
"habitual".                      could be either "lakne" = "x1 is
-  How to say: "customary".      likely under conditions x2"
tcaci = "custom"; is it enough to"tarti" = "x1 behaves as x2 under
use the quality abstractor "ka"? condition x3"
Is ka tcaci = "customary"?       depending on context, or possibly
                                 other choices.
John Cowan:                      
"Habit" and "custom" are the sameDave:
gismu:  "tcaci".  You would say  How to say: "thoughtless".
"customary" by using this gismu  Negation of "sanji" = "aware"?
in a tanru:  "a habitual walker" And then abstracted?
= "lo tcaci cadzu". [Bob adds:   How to say: "unwise". Negation of
You can also use "ta'e" as a     "prije" and abstracted?
tense-like inflection for        
"habitually".]                   John:
                                 No need to abstract here.
Dave:                            Abstraction corresponds to things
How to say: "cultural", "x       like "-ness" and "-ity".  "lo
springs from culture y". Here "kana'e prije" = "an unwise person";
kulnu" is clearly not adequate.  "lo na'e prije cusku" = "unwise
                                 statements".  "Thoughtless" has
John:                            several English meanings, I
"lo se kulnu" are the people who think.
exhibit a culture.  [Bob adds:   
loi kulnu can be used for some   Bob adds: "ka sanji" is thus
portion or element of culture.]  "consciousness" or "awareness".
                                 "nu prije" and "ka prije" are
Dave:                            different interpretations of
How to say: "tilt", as in "x     "wisdom".  I would do most
tilts/leans at angle y in frame  varieties of "thoughtless" as
z";                              combinations of "claxu" = "lack"
                                 and "pensi" = "think" or "kurji"
John:                            = "taking care of", or as "na'e",
"salpo" = "x1 is sloped/inclined "no'e" and "to'e" negations of
with angle x2 to horizon/frame   the latter two.
x3"                              
                                 Dave:
Dave:                            And here's a biggie: how do you
                                 say "idiom"? An idiom is not

                                105
simply a metaphor, it's a        best if space is too limited?
metaphor that through constant   There are a lot of aphorisms that
usage has lost its metaphoric    can be translated, and a lot of
indirection and simply means whatconcepts to explore Lojbanically.
it originally suggested.  (Like    The first exercise is a study
"red herring".)                  in happiness.  It all started
                                 with Nick Nicholas lecturing
John:                            about "happiness":
We don't have those in Lojban.   
(zo'o)                             In fact, to bring in a
                                 parenthesis, there is a massive
Bob: A lot of possible tanru     history to the verb to be used
  here.  How about:              corresponding to Happy.  It
"se farvi smuni valsi" =         alludes, of course, to the Sermon
  "evolved_into-meaning-words"   On The Mount.  Now here are some
"kulnu smuni valsi" = "cultural- distinct types of happiness:
  meaning-words"                 
"tcaci smuni valsi" = "customary-English   Esperanto   Greek
  meaning-words"                 (Modern)
                                 ---------------------------------
_________________________________----------------
        ________________         Fortunate Felicxa     Eutuxhs,
                                 Eutuxismenos
  le lojbo se ciska - Your turn  Pleased   (Kontenta)
                                 Euxaristhmenos
  Nick Nicholas has translated   Happy     Gaja        Xaroumenos
the next paragraph of the ongoingJoyful    Gxoja       (can't
Jim Carter science fiction story,think of one)
but partly due to space, and     Blissful  (Sengxena)  Makarios
partly because of the other large(cf. Latin Beatus)
translation here from Nick, we   
are holding that for next issue.   The Esperanto has 'Fortunate'
Instead, I'm going to put some   because this makes the link
challenges to the readership (in between cause of happiness and
case all this text, and Ivan's   the happiness explicit:  Gxojas
requests for tanru or lujvo,     tiuj, kiuj... would sound like
aren't enough.                   their joy was incidental to their
  I seriously want to see as manyseeing God.  It would be even
people as possible try one of theworse with Gajas tiuj, kiuj...
following two Lojbanic exercises.which reads somewhat like "Those
The first is merely an exercise  who have seen God are running
in creative word manipulation,   around smiling".
which any language lover can do.   But the original Greek had
(You can complicate the exercise Makarioi, and the Latin
by using it as practice in lujvo-translation has Beati.
making, but this is not obliga-  Admittedly some semantics would
tory.)  The second is a set of   have been influenced by the
aphorism translation exercises   Church's use of the term; but Oi
that anyone can work on.         Makarioi Nhsoi, the Isles of
  We would like feedback on theseBliss, the late-pagan-Greek
exercises, whatever you choose toequivalent of heaven, predates
do with them.  Are they in-      Christian theology (I think).
teresting?  Which did you try to What this implies to me is that
do, how well did you do, and how Christ meant something along the
do you feel about your level of  lines of 'They will have no
success compared to your         worries, no disquiet', not 'they
expectations?  Do you want       will run around smiling' (Happy)
exercises like this to be a      or 'they will run around
regular feature of ju'i lobypli, hurrahing' (Joyful) or 'they will
and which exercises did you like

                                106
say "ain't we lucky"' (Fortunate,  First you wander through the
Pleased).                        gismu list pulling out words
  Does la Lojban distinguish     related to the concept.  I didn't
between these happinesses (it    pull the following out in order,
doesn't have to, and I've heard  but about 15 minutes gave me all
SapirWhorfish mumbles against    of them.  You can probably find
such distinctions), and which    more using your own lists.  As
would it have picked here?       you will see, don't be picky -
_________________________________let your mind play word associa-
________________                 tion games.
  Bob:  Yes and no.  We can        Some key source words (I'll use
distinguish between any concepts,the published gismu list; some
but no one has done so yet.  The places of some of these are pro-
"SapirWhorfish mumbles" you heardbably going to change, but not to
are wrong - one of the areas     significantly affect this
where Lojban may exhibit S-W     exercise - feel free to suggest
effects is in the ability to makechanges, in fact):
such distinctions, and the       
creativity that results from the Idea   Place structure  rafsi or
free combination of ideas.  In   lujvo
fact, I proposed a massive effortgleki  x is happy about ygle gei
like the following way back in   pluka  x pleases y      puk pu'a
the first issue of JL, under the  se pluka      x is pleased by y                       
name 'complexing'.  In a sense,  selpu'a
doing this is what made me       salci  x celebrates y        sal
interested in Loglan enough to   xalbo  x is levity/non-serious
tackle the dictionary project    about y                        -
(which is how I got started).  (Izdile  x is amusing to y zil zdi
love playing with words and their se zdile       x is amused by y                       
corresponding ideas, to see how  selzdi
they interact.)                  
  So let's do it.  Let's see how These are 5 different basic
many ways there are to be happy  'kinds' of happiness. They can
(do happy?).  I'll give some     stand alone or modify each other:
hints and guidelines, and see    
what our readers can come up
with:

salci gleki                                            x is
celebratingly-happy about y                            salgei
           gay/joyful
se zdile gleki                                         x is
amusedly-happy about y                                 selzdigei
           one kind of enjoy, but see below.

Idea       Place structure                             rafsi or
lujvo
gleki      x is happy about y                          gle gei
pluka      x pleases y                                 puk pu'a
 se pluka  x is pleased by y                           selpu'a
salci      x celebrates y                              sal
xalbo      x is levity/non-serious about y             -
zdile      x is amusing to y                           zil zdi
 se zdile  x is amused by y                            selzdi


                                107
These are 5 different basic 'kinds' of happiness. They can stand
alone or modify each other:

salci gleki                                            x is
celebratingly-happy about y                            salgei
           gay/joyful
se zdile gleki                                         x is
amusedly-happy about y                                 selzdigei
           one kind of enjoy, but see below.

But we needn't stop here.  There are related words that are useful
for specific kinds of happiness:

bebna      x is foolish in y                           beb
cando      x is idle/at rest/inactive                  cad
cizra      x is strange/bizarre to y in z              ciz
cunso      x is random/chance                          cun cu'o
dimna      x is the fate/destiny/doom of y             -
 se dimna  x is doomed to y                            seldimna
fenki      x is crazy/insane in doing/being y          fek
jgira      x has pride about y                         jgi
kanro      x is healthy                                ka'o
kufra      x is comfortable in environment y           kuf
lifri      x experiences y                             lif fri
lijda      x is the religion of people y with tenets z lij jda
 se lijda  x follows religion y with tenets z          seljda
mansa      x is satisfied with y                       -
panpi      x is at peace                               pap
pensi      x thinks about y                            pen pei
prije      x is wise/sage about y by standard/observer z    pij
racli      x is sane/rational                          -
ranxi      x is ironic in that y                       rax
sanga      x sings y to z                              sag
siclu      x makes whistling sound y with z            sil
stodi      x is constant/invariant in y                tod sto
tarti      x behaves/conducts self as y under conditions z  tat tai
tcaci      x is the custom/habit of y under condition z     cac tca
tinbe      x obeys/follows command y by z              tib
xamgu      x is good for y by standard z               xag xau
zabna      x is the ameliorative of y                  zan za'a

     All 24 of these can be applied as modifiers to "gleki",
"selpu'a", "selzdi", "xalbo" and "salci", and in some cases each
other. That gives over 120.  Many will bring to mind a situation
where they would be useful.  Some, not all will suggest an English
word equivalent (or possibly to Esperantists an Esperanto word, or
to Nick, a Greek word.)  All are valid tanru in Lojban.  All more
or less mean "happy".  All can be made into lujvo.  Anyone want to
tackle the complete set systematically, giving us hundreds of words
for the dictionary in one fell swoop?  If you aren't that ambi-
tious, try a few dozen, put together as you see fit (which may take
a while) or a systematic subset (after doing this for a few
minutes, you'll find you can't write them as fast as you can
analyze them and put them together).

I'll give several examples:
bebna gleki                                            x is
foolishly-happy about y                                bebgei
           one kind of giddy

                                108
fenki gleki                                            x is
crazily/insanely happy about y                         fekygei
           another kind
zabna lifri                                            x
Experiences! y                                         zanfri
           enjoys (my preference for this English word in most
contexts)
siclu gleki                                            x is
whistlingly-happy about y (whistling z with w)         silgei
           (think Snow White and the 7 Dwarves)
panpi kufra                                            x is
peacefully-comfortable in environment y                papkufra
           content
panpi gleki                                            x is
peacefully-happy about y                               papygei
           serene
se lijda gleki                                         x is
religiously-believing happy about y                    seljdagei
           beatific
           With the "happy" word first:
gleki panpi                                            x is
happily-at peace (happy about y)                       glepanpi
           blissful                                    geirpanpi

Not enough?  We have intensities:
carmi      x is intense in y                           cam cai
milxe      x is mild/gentle/unextreme in property y    mli
mleca      x is less than y in property z by amount w  mec me'a
mutce      x is toward the y extreme in property z     mut mu'e
traji      x is superlative in property y              taj rai
zmadu      x is more than y in property z by amount w  zad zma mau

carmi gleki                                            x is
intensely-happy about y                                camgei
           implies a particularly emphatic happiness
mutce gleki                                            x is very-
happy about y                                          mu'egle
           a little broader happiness than camgei, but still
extreme
gleki zmadu                                            x is happy-
more than y is (about z by amount w)                   glemau or
           happier                                     geizma
zmadu gleki                                            x is more-
happy about y (than z is by amount w)                  maugle or
                                                       zmagei
se gleki zmadu                                         x is
happy_making-more than y (than z is by amt w)          selgeimau
           which can then be converted without lujvo making to:
se selgeimau                                           x is made-
happy-more by y than by z by amt w
           a proper and fairly exact lujvo is          selkemselgei
mau
           but the following will probably be adopted  selselgeimau
           because any other interpretation of "selsel-" is a
nullity

     Finally, we can also define happiness in terms of what it is
not, or what it lacks:

to'e       polar opposite (c.f. Esperanto mal- ?)      to'e

                                109
no'e       neutral scalar negation                     no'e
badri      x is sad/depressed about y                  dri
claxu      x lacks/is without y                        cax cau
dunku      x is anguished/distressed by y              duk du'u
fanza      x annoys/irritates/bothers y by doing/being z    faz
 se fanza  x is annoyed by y doing/being z             selfanza
fengu      x is angry at y for z                       feg fe'u
junri      x is serious about y                        jur
pante      x protests/objects/complains to y about z by doing w  -
raktu      x troubles/disturbs y by z                  ra'u
 se raktu  x is troubled by y doing z                  selra'u
steba      x feels frustration about y under conditions z   seb
xanka      x is nervous/anxious about y                -

           Examples:
to'e badri x is polar-opposite of sad/depressed about y     to'erdr
i
to'e pante x opposite-of-protests to y about z by doing w   to'erpa
nte
           compliments y ...
no'e pante x doesn't-protest to y about z by doing     no'erpante
           has nothing to complain about to y
           is a reasonable interpretation of this as a lujvo,
           but there may be others
se raktu claxu                                         x is
troubled-lacking of y doing z                          selra'ucau
           a specific kind of bliss or serenity suggesting:
           x is blissfully unaware of y doing z

     I'll close by listing some others Jim Brown had in his old
dictionary, (updated to Lojban standards).  I don't always agree
with Brown's tanru but they may give ideas (a version of his
'English equivalent' is in parentheses):

gleki culno                                            x is happy-
full of y (x is joyful about y)                        geiclu
gleki xendo                                            x is happy-
kind towards y (x hearty/warm-hearted toward y)        glexe'o
gleki rinka                                            x is happy-
causer of y (x cheers y up giving)                     gleri'a
se rinka gleki                                         x is
causedly happy about y by cause z                      selgleri'a
           (x is cheered up/cheerful about y cheered by z)
           (which we can use other causals in:)
se mukti gleki                                         x is
motivatedly-happy about y by motive z                  selglemu'i
se krinu gleki                                         x is
justifiedly happy about y with reason z                selgleki'u
           or even
se nibli gleki                                         x is
logically-necessitated to be happy about y by logic z (?!)  selglen
i'i

     Most English words will map to more than one Lojban word,
because there are so many Lojban words.  Thus we can make finer
distinctions in our words than English can!
     For those not familiar with lujvo-making rules, here's the
easy version.  Replace the final vowel of all but the last word
with 'y', and write it as one word. Thus the 'easy' form of

                                110
'bebgei' is 'bebnygleki', and the two word versions are absolutely
the same in meaning, place structure, etc.
     If you want to try to shorten them, the following is a short
set of rules omitting a couple of special cases that you'll rarely
run into.  See the Synopsis for complete rules. (For those few
people who have gotten the new issue of the rafsi list, these rules
are included).
     Remember that the final rafsi must end in a vowel,
incompatible consonant junctures between rafsi (voiced/unvoiced
like "bp", same consonant like "bb", and both fricative "c, j, s,
or z", and a few others - see the Synopsis or the back page of the
Lojban-order gismu list) require a "y" 'hyphen' between the
consonants.  Also a CVV form requires 'r' as a hyphen in the first
rafsi, unless immediately followed by a CCV affix (this to form the
mandatory consonant cluster - use an 'n' if the following affix
starts with 'r').  There are a few other rules, but if you get this
far your first try without making any mistakes, I'll be more than
******* (Well - you find the word I want!!!)

___________________________________________________________________
_________________________________

     Our other exercise is in translation of aphorisms.  This can
be a game of almost any level of difficulty or skill required, so
we will provide both Lojban and English data so that it can be max-
imally used.
     What follows are three sets of sentences:
- The first set of sentences consists of Lojban translations of
     some aphorisms and other 'pithy sayings', the product of
     various members of the Lojban conversation group that meets
     here in the Washington DC area.
- The second set of sentences are the original English versions of
     each of the Lojban translations, not in the same order as the
     Lojban sentences.  This set of aphorisms is graded in
     difficulty from 1 (easiest) to 4 (hardest), in terms of amount
     of Lojban skill needed to translate them into Lojban.
- The third set of sentences are also English aphorisms, but these
     have no translations.  They are 'exercises'.  This set of
     aphorisms is graded in difficulty from 1 (easiest) to 4
     (hardest) as is the second set.  These are estimates, since
     the sentences haven't been translated.

     The translation section at the end of this issue has the
complete set of matches between the first two sets of sentences, so
that you can study them in comparison to each other.  Feel free to
comment on or criticize any of these translations, or try to do
better yourself.

Here's what you do with these sentences:
- The simplest exercise is to attempt to understand the first set
of aphorisms, matching them up with their corresponding members of
the second set.
- Slightly more difficult is to go the other way.  Attempt to
translate sentences in the second set into Lojban sufficiently that
you can match them to the corresponding members of the first set.
You can also mix the two directions, trying some in each direction,
making sure you mark sentences off so that you don't repeat
yourself.  Note that the Set #1 translations may not be the best
possible.

                                111
- You can attempt to fully translate the first set into English,
seeing if you can figure out colloquial translations of the Lojban.
You can compare your answer with the real quotation, which will
help you judge your skill in reading Lojban, and our skill in
communicating to you in Lojban.
- Still braver, you can attempt to completely translate the second
set of sentences into Lojban, and then compare your answer with the
one that our group came up with.  If you don't agree, you may still
be correct - there is certainly more than one way to translate
these types of sentences.  If you think your version is as good or
better, send it to us, and we may use yours when these are used as
exercises in the Lojban textbook or samples in the first Lojban
reader.
- Finally, you can translate the third set of sentences into Lojban
and send them to us.  Include on a separate sheet the identifying
number or complete English for each translation.  The conversation
group here will then attempt to back-translate your versions, which
will help us learn the language.  We will collectively respond to
all submissions with comments and corrections of your attempts.
The best translations will appear in future la lojbangirz.
publications.  We of course recommend working from easier aphorisms
to harder ones, but work at whatever level you feel comfortable.
- If you are studying together with at least one other Lojbanist,
you can also do exactly what our group did in developing the first
set.  Cut individual English aphorisms into strips and put them
into a box.  Each person takes a slip and translates their sentence
into Lojban.  When all are done, say or write your results in turn,
and have the other(s) try to back-translate into English - then
compare with the original, and discuss the problems you
experienced.  We successfully did this with people of a variety of
Lojban expertise at LogFest.  You can do it too.


Set #1 - Pre-translated Lojban Aphorisms

Note: Two Lojban sentences below correspond to one single English
in Set #2.  Thus there are 20 Lojban sentences, and only 19 English
ones in these two sets.

A. mi pu ricfu  .ije mi pu pindi  .i la'ede'u cu xagmau

B. le tordu temci morji cu traji se sinma sera'a le turni

C. le nu clira ckaklama gi'e clira ckacliva cu rinka le ka kanro
     joi ricfu joi prije

D. le ka terpa cu me lo pa drani se terpa

E. le banli to'e jetnu cu zmadu le jetnu le ni se krici

F. lo ka krici le cevni cu noroi mintu lo ka prami le cevni

G. mi pagbu ro le se tcidu be mi

H. lo cevni ka'enai galfi loi purci  .iku'i lo circtuca ka'e go'i

I. lo no'e cevni krici cu krici le nu ri cu snuti

J. te'inai mi zmadu djica le nu zvati la Filydelfias

                                112
K. lo sidbo cu na fuzme le prenu poi krici ri

L. le clira cipni cu cpacu le curnu

M. na curmi le nu djuno roda

N. pa nunfenso pe ca lo zantemci cu fanta so nunfenso

O. fe le nu catra xaksu lo temci kei fa lo kamni nu penmi cu prane
     xarci

P. ro le nu kusru cu se rinka le ka ruble

Q. da poi renvi sepi'o loi balre baca'a mrobi'o sepi'o loi sodva
     vanju se pinxe

R. lo kumte cu simlu lo xirma poi se plafinti lo kamni

S. le verba poi terpa le manku ku'o ji le prenu poi terpa le se
     gusni cu bebna traji

T. no da pe le turni cu dunli se sinma lo tordu ni morji


Set #2 - Difficulty-Sorted English Originals for Set #1

1. (1) An idea isn't responsible for the people who believe in it.
     - Don Marquis

2. (1) The early bird gets the worm.

3. (1) I am a part of all that I have read. - John Kieran

4. (1) I've been rich and I've been poor; rich is better. - Sophie
     Tucker

5. (2) A camel looks like a horse that was planned by a committee.
     - Vogue magazine, July, 1958

6. (2) God cannot alter the past, but historians can. - Samuel
     Butler

7. (2) An atheist is a man who believes himself an accident. -
     Francis Thompson

8. (2) A stitch in time saves nine. - Benjamin Franklin

9. (2) To know all things in not permitted. - Horace

10. (2) All cruelty springs from weakness. - Seneca

11. (3) Nothing is so admirable in politics as a short memory. -
     John Kenneth Galbraith

12. (3) The only thing we have to fear is fear itself. - Franklin
     Delano Roosevelt

13. (3) On the whole I'd rather be in Philadelphia. - W. C. Fields

14. (3) Faith is never identical with piety. - Karl Barth

                                113
15. (3) Early to bed and early to rise / Makes a man healthy,
     wealthy and wise.

16. (3) A big lie is more plausible than truth. - Ernest Hemingway

17. (3) To kill time, a committee meeting is the perfect weapon. -
     Laurence J. Peter

18. (4) Who is more foolish, the child afraid of the dark or the
     man afraid of the light? - Maurice Freehill

19. (4) He who lives by the sword shall perish by the champagne
     cocktail. - Saul Alinsky


Set #3 - No-Translations Given:  Aphorisms in Difficulty Order

1. (1) A page of history is worth a volume of logic. - Oliver
Wendell Holmes, Jr.

2. (1) All our knowledge has its origins in our perceptions. -
     Leonardo da Vinci

3. (1) You can fool most of the people most of the time - P. T.
     Barnum

4. (1) Examine the contents, not the bottle. - The Talmud

5. (1) History is only a confused heap of facts. - Earl of
     Chesterfield

6. (1) I hear and I forget.  I see and I remember.  I do and I
     understand. - Chinese proverb

7. (1) If it is not erotic, it is not interesting. - Fernando
     Arrabal

8. (1) Lo!  Men have become the tools of their tools. - Henry David
     Thoreau

9. (1) Native ability without education is like a tree without
     fruit. - Aristippus

10. (1) Not to decide is to decide. - Harvey Cox

11. (2) A single death is a tragedy, a million deaths is a
     statistic. - Joseph Stalin

12. (2) Art is not a thing; it is a way. - Elbert Hubbard

13. (2) Doubt is not a pleasant mental state but certainty is a
     ridiculous one. - Voltaire

14. (2) History is a cyclic poem written by Time upon the memories
     of man. - Percy Bysshe Shelley

15. (2) If thought corrupts language, language can also corrupt
     thought. - George Orwell

                                114
16. (2) If you scoff at language study ... how, save in terms of
     language, will you scoff? - Mario Pei

17. (2) If we are not ashamed to think it, we should not be ashamed
     to say it. - Marcus Tullius Cicero

18. (2) In war there is no substitute for victory. - Douglas
     MacArthur

19. (2) My father gave me these hints on speech-making: "be
     sincere...be brief...be seated." - James Roosevelt

20. (2) Nothing in life is to be feared.  It is only to be
     understood. - Marie Curie

21. (2) No man is a failure who is enjoying life. - William Feather

22. (2) Poetry is the rhythmical creation of beauty in words. -
     Edgar Allan Poe

23. (2) Seeing is deceiving.  It's eating that's believing. - James
     Thurber

24. (2) Shake and shake / The catsup bottle, / None will come, /
     And then a lot'll. - Richard Armour

25. (2) Think like a man of action, act like a man of thought. -
     Henri Bergson

26. (2) Those who cannot remember the past are condemned to repeat
     it. - George Santayana

27. (2) Those who write clearly have readers; those who write
     obscurely have commentators. - Albert Camus

28. (2) The thoughtless are rarely wordless.

29. (2) These Macedonians are a rude and clownish people; they call
     a spade a spade. - Plutarch

30. (2) The heart has its reasons which reason does not understand.
     - Blaise Pascal

31. (2) There is nothing permanent except change. - Heraclitus

32. (2) To be a success in business, be daring, be first, be
     different. - Marchant

33. (2) We are tomorrow's past. - Mary Webb

34. (2) What the country needs are a few labor-making inventions. -
     Arnold Glasow

35. (2) Who shall guard the guardians themselves. - Juvenal

36. (2) You'll find in no park or city / A monument to a committee.
     - Victoria Pasternak

37. (2) A belief is not true because it is useful. - Henri Frederic
     Amiel

                                115
38. (2) A person gets from a symbol the meaning he puts into it. -
     The United States Supreme Court

39. (2) All the fun's in how you say a thing. - Robert Frost

40. (3) Be obscure clearly. - E. B. White

41. (3) I am free of all prejudices.  I hate every one equally. -
     W. C. Fields

42. (3) Leadership is action, not position. - Donald H. McGannon

43. (3) Liberty means responsibility.  That is why most men dread
     it. - George Bernard Shaw

44. (3) Passions are vices or virtues to their highest powers. -
     Johann W. von Goethe

45. (3) Power tends to corrupt and absolute power corrupts
     absolutely. - Lord Acton

46. (3) The race is not always to the swift, nor the battle to the
     strong, but that's the way to bet. - Damon Runyon

47. (3) Sometimes I've believed as many as six impossible things
     before breakfast. - Lewis Carroll

48. (3) The man who strikes first admits that his ideas have given
     out. - Chinese Proverb

49. (3) Those who deny freedom to others deserve it not for
     themselves. - Abraham Lincoln

50. (3) The day will come when everyone will be famous for fifteen
     minutes. - Andy Warhol

51. (3) We can lick gravity, but sometimes the paperwork is
     overwhelming. - Wernher von Braun

52. (3) What is honored in a country will be cultivated there. -
     Plato

53. (3) When I use a word, it means just what I choose it to mean -
     neither more nor less. - Lewis Carroll

54. (3) When an idea is wanting a word can always be found to take
     its place. - Johann W. von Goethe

55. (4) 'Tain't what a man don't know that hurts him; it's what he
     knows that just ain't so! - Frank McKinney Hubbard

56. (4) A word is not a crystal, transparent and unchanged; it is
     the skin of a living thought and may vary greatly in color and
     content according to the circumstances and time in which it is
     used. - Oliver Wendell Holmes, Jr.

57. (4) Do you realize if it weren't for Edison we'd be watching TV
     by candlelight? - Al Boliska

                                116
58. (4) For every person wishing to teach there are thirty not
     wanting to be taught. - W. C. Sellar and R. Y. Yeatman

59. (4) It's a poor sort of memory that only works backwards. -
     Lewis Carroll

60. (4) Production is not the application of tools to material, but
     logic to work. - Peter Drucker

61. (4) The foolish and the dead alone never change their opinion.
     - James Russell Lowell

62. (4) The past always looks better than it was.  It's only
     pleasant because it isn't here. - Finley Peter Dunne

63. (4) The most valuable of all talents is that of never using two
     words when one will do. - Thomas Jefferson

64. (4) When a man has pity on all living creatures then only is he
     noble. - Buddha

___________________________________________________________________
                 _________________________________
     For our untranslated Lojban text this issue, I'll share the
floor with Ivan Derzhanski, who was studying in Boston when he
wrote the following letter.  My answer follows.  Ivan wrote his
text and understood mine with only level 1 materials.  There ARE
grammatical errors in Ivan's letter, but I was able to figure out
most of what was intended, and my response corrects those errors.

coi lojbab. do'u
ni'o di'e pamoi xatra ci'a mi bau la Lojban.
   .i .oinai.o'u mi cpacu le'i pelji

ni'o mi sidju ledo nu zbasu le reno valsi
   .i ku'i baze'inai la'edi'u ne ki'u ca nu mutce cutyzu'e

ni'o .a'ocai do pu'i jimpe se mi ciska
   .i u'u na mi mutce djuno Lojban.  .ije lerci  .ije mi tatpi
   .i .au.a'u mi cpacu zoi <draft textbook lessons> zoi

co'omi'e  .iVAN.

___________________________________________________________________
                 _________________________________
.ue do xamgu troci lenu cusku bau la Lojban.
.i la nik. ji'a puzi cpacu le selprina du'i xaunro'i (xamgytroci
     ta'unai)
.i mi jimpe ledo selsku  .iku'i do milxe srera  .ijeseku'ibo mi na
     birti le smuni be ledo vomoi jufra noi pilno le pluja ke temci
     cmavo  .iku'i mi smadi
.i lo lojbo cmene cu nitcu lo cmene valsi tcita  .iseni'ibo le do
     xamoi jufra cu se srera  .ije do pu djica <<lu .i .u'u na[ku]
     mi mutce djuno la Lojban. li'u>>
.i xu do djica lenu mi ca benji le zasni tadycku kei ji lenu mi
     denpa ledo benji le krasi bangu selfanva terfanva
.i mi djica lenu do cmima binxo le lojbo ke skami xe mrilu po'u la
     lojban. list.

                                117
.ije ko benji le notci poi vasru ledo skami judri fo zoi <<.uniks.
     lojban-list-request@snark.uu.net .uniks.>>  .i ko cpedu bau la
     gliban .uu ki'u la erik. na tadni la lojban
.i mi djica lenu do mrilu lo fukpi be le do xatra joi le mi se
     spuda xatra le lojbo ke skami girzu po'u zoi <<.uniks. lojban-
     list@snark.thyrsus.com .uniks.>>
.i .u'u mi puza srera  .ije ledo pamoi se mrilu cu te mrilu zoi
     <<.uniks. lojban-list-request@snark.thyrsus.com .uniks.>>
.i ko ranji lenu troci le lojbo cusku
.i co'omi'e lojbab.

[Note that the network address is no longer correct.  See page 2 of
this issue for the new address.]

     Those letters were written in April.  In July, Ivan, now back
in Bulgaria, sent me a postcard and the following letter.  The
letter is virtually perfect, having only three minor errors that
hardly affect understanding.  1) Ivan had a "zo" quote on the name,
which I've replaced with lo'u/le'u quotes since there is more than
one word.  2) Lojbanized names should end in a consonant, but
inside 'ungrammatical Lojban text' quotes, this does not matter.
3) Ivan's original: "pa le tercfi lemi natmi beme'e zo ..." meant
that his nation, and not the author, had the indicated name.


                                118
     A portion of the text is embedded English.  Ivan asks for help
in expressing some concepts, which will be obvious in the text,
even if you can't read the Lojban.  Please suggest tanru or lujvo
for any or all of them.  If you are not confident at lujvo-making
from a tanru, but feel that the concept should be expressed in a
single word, put the Lojban words together separated by a hyphen.
I encourage that proposed lujvo come with a plausible place
structure.  I'll collect the suggestions and send them to Ivan.
     I note in passing that "LogFest" has the impermissible medial
'gf' which makes it a bad Lojbanized name.  Maybe it is time we
switch the name to Lojban since people are starting to write about
it in Lojban.  What do people think of "la jbosalci"?

de'e xatra tu'i la pijyta'u de'i li pa pi'e ze pi'e sopa

coi lojbab.

ni'o mi rinsa do tebe'i lemi tcadu no'u le la bulgariax. ralta'u
.i .oiru'e  mutce lenku  .i le djacu carvi cu na sisti

ni'o mi troci ke lojbo fanva lo lisri be ci'a pa le tercfi pe lemi
natmi zi'e peme'e lo'u XRISto. SMIRnenski. le'u be'o
.iku'i pi su'o loi selsku cu dukse nandu mi  .i mi benji lo pagbu
be ri do  .i .e'o  ko stidi le lojbo velsku

.i  zoi. problem.

(1)  dedicate (a story) to (sbd.)
(2)  clench one's fist;  fist (n.)
(3)  bribe; ransom
(4)  betray
(5)  tear, rend (cloth); rag (n,
(6)  bend, lean over (sbd.)
(7)  threat, menace (n.)
(8)  rush, dash
(9)  shrill, piercing (sound)

  .problem.

ni'o.a'o  la LogFest. pe le cabnanca pu snada

co'omi'e .iVAN.

___________________________________________________________________
_________________________________

     Nick Nicholas, our Australian/Greek/Esperantist Lojban star is
interested in exploring Lojban stylistics.  Although he has worked
on learning Lojban only a few months, he has a command of the
language about as good as anyone has.  Perhaps too good - his
writings may be beyond the capability of most readers' following.
This is sad because the stylistic variation in the Lojban that he
was trying to achieve is quite obvious.  Equally important, it is
quite interesting.
     I urge everyone willing to spend a bit of time to try various
portions of the text (some are more difficult than others).  Feel
free to liberally cheat and look back at the translation section,
or perhaps just read the text with translation there.  Skip around
and sample the five sections, and see if you too can perceive the
stylistic variation in the Lojban.

                                119
     A warning with a suggestion - this is not an easy text;
indeed, it goes far beyond the draft textbook lessons in use of the
grammar.  Do not get too hung up if you cannot figure out a word
(Nick even uses a couple that are not in the published cmavo list,
but are noted in the JL14 change pages, or in the discussion above
on sumti-raising.  I've tried to identify these and work around
this problem, but may have missed something.)  Try to get the gist
of what is being said, and the words may become clear from context.
And don't get hung up on one sentence too long - move on to the
next one.

     The following is translated from modern Greek.  Freddy
Germanos, (1934- ) is a newspaper columnist, and this is taken from
a book collection of his columns for the Mesimvrin'i (Midday) pa-
per, printed in 1967 by Galaxias publishers.  The title of the book
is 'To Dhis Examartein' ('Sinning Twice' - alluding to the ancient
Greek saying, "Sinning twice is not [a characteristic] of a wise
man".  The parody is on pp. 12-13 of the 7th edition, dated March
1974, by Grigoris Publishers, 73 Solonos St., Athens.  There is no
copyright notice on the book.

lonu reroi pacnunzu'e cu na se ckaji lo prije

     ni'oni'oni'oni'o la fredis. germaNOS. pu finti  .i la kir. pu
te pinxe pamo'o  .i memi'a poi te prosa


     ni'oni'oni'o nuzba

     ni'oni'o le briju cu so'iroi se klama so'olemi pendo poi co'a
se jibri zu'i po mi gi'e preti cpedu le tarmi po'e lo nuzba nu
cusku


     ni'o .i'a loi nuzba ne semau roda vajrai ci'elo karni nu cupra
.iku'i cinri fa lo karni poi mu'i lenu ke'a pilno ku cusku le go'i
.i rolo karni cu ckaji leri tadji be le nuzba nu cusku  .i la'edi'u
nibli lenu ra se tcidu co frili gi'a se tcidu naku





     ni'o mupli lo nuzba poi mutce sampu  .i lo nixli be li mu ca
lepu'u kelci levo'a bolci cu farlu lo skuro poi karbi'o se kakpa
lei gunka pole ta'utru


.i tarmi le ve cusku be le nuzba bei pi'o la karni be de'i roboi lo
djedi be'oku (to la roldei toi)  .i
     <<lu di'o lo paboi te kruca be le klaji poi se cmene la
     .adriaNOS. vauku'o bei le klaji poi se cmene la paleologos.
     ku'obe'oku ko'a goi paboi lo nixli be muboi lo nanca be'oku
     ge'u pu se xrani ca lepu'u ko'a cu kelci da noi ke'a se tarmi
     lo bolci vauku'ovaukeivau  .i le nu xrani vaukei cu diklo
     fe'eba'o va paboi le zdani be lai paleologon. benizelon.
     be'oku noi la kamBUroglus. pu skicu ke'a ta'i lo se ciska ku
     tai loka lanli vaukeivauku'ovau  .i la'edi'e cu cizra zo'e
     lenu fasnu vaukei ne sekai leka na xlura vaukeige'uvau  .i
     paboi lo reboi rirni be la kamBUroglus. be'oku pu binxo le

                                120
     speni be paboi lo reboi rirni be la kamBUroglus. be'oku poi na
     du vo'a ku'o soivo'avo'e se'u be'oku vi le zdani be lai
     varvatis. be'oku noi jibni le zdani be lai paleologon.
     benizelon. be'okuvauku'ovau  .i le nixli ki na se ckape vau
     li'u>>


















.i ke'unai vecu'u la akROpolis. le nuzba cu ka'e ckaji loka zmadu
cinri bo cfipu .i
     <<lu pu zgana ne'i lo skuro lo se xrani. goi ko'a poi mebrai
     nixli gi'e jarco lo frili se viska jalge be lonu vlile  .i
     ri'anai lenu le zekri lifri po'u ko'a na pujaca skicu fi lei
     pulji kei ca cipra fa ri lejei na'e snuti (to pupu jinvi
     to'ebori toi) gi'e cinse zekri  .i lo kamni poi se cmima le mi
     karni zbasu caca'a cipra fi le tcini mu'i lenu djica co danfu
     fo re preti ra'u po'u di'e  .i pamai xu le nixli pu cinse
     vlile lifri  .i remai fau lenu ko'a na'eke cinse vlile lifri
     kei ko'a na'eke cinse vlile lifri ki'u ma li'u>>









.i li'a vecu'u la cermurse [see the new gismu added this issue]
sesau se basna fa loi drata tcila  .i
     <<lu nixli  .i ko'a pe leni slabu  .i ri du mu nanca  .i xrani
     ca le purlamdei  .i farlu lo skuro po la ta'utru ca lenu kelci
     le bolci  .i ko'a ze'iba tavla palemi karnypra  .i ko'a fatci
     xusra  .i
          <<lu mi farlu le skuro  .i ni'ibo mi na citka ca re djedi
          .i besna se cfipu  .i lemi mamta cu selfu seljibri  .i
          lemi patfu cu na'e dinycpa  .i ni'ibo ri roroi pantydzu
          li'u>>
     li'u>>
     
     
     
     
     
     

                                121
     
.ice ka'e zgana cusku fo la natmi nu'arki'a
     <<lu lenu xrani cu se rinka lonu pimo'ale skuro cu te sabji lo
     gacri  .i ni'inai lenu na catni se xusra cu cusku lenu fuzme
     fa relo gunka pe le ta'utru ge'u zi'e noi ze'u .o'onai
     kaurposysi'orpre  .i lei pulji cu cipra lejei zasti fa loi
     flana'etinbe gripre noi pu zukte lo drata nu jecyselxrazei
     li'u>>
     
     
     
     
     
     
     
     
.i romai la deimid. cfari le ni'onrai tarmi be loi nuzba
     .i <<lu se tirna fa lo voksa  .i se viska fa lo xance poi se
     desku tai loka ti'e pacna zi'epoi cpedu lenu sidju  .ibabo
     smaji  .i ?ma pu fasnu  .i mi ciksi  .i lo nixli pu sakli
     mo'ine'i lo skuro  .ibazabo le nixli cu bacru
          <<lu mi farlu ri'a lenu la paf. na di'i tcidu la deimid.
          li'u>>
     li'u>>
                                 
                                 
                                 
                                 
                                 
                                 
                                 
      Nick asked for a return of "lei lojbo", Nora's comic strip.
 In honor of Nick's efforts, and his Esperantist background, enjoy
   the following page.  Esperanto text was aided by David Twery.

                                122
                   lei lojbo by Nora LeChevalier

                                123
                 Translations of le lojbo se ciska
                                 
                 lo zekri  fi'e    la bab. tcySEL.
                 A  crime, created by Bob Chassell

     Probably just "zekri", an observative, would be better.
     Stand-alone sumti in Lojban suggest an answer to a "ma"
     question.  With the "fi'e" author label, "me la'ezo zekri ..."
     might be still better.

.i mi cadzu     pagre        le  vorme      le  kumfa
   I  walkingly pass through the door  into the room.

     We need a word for doorway, probably as a place of "vorme".  I
     doubt that the door itself was passed through.
     
     Bob several times uses "lo" and "le" in ways I can't quite
     figure out.  As a reader, whenever I see "le" attached to a
     new description, I expect to see some restrictive relative
     clause that tells which one "the" door is, and which one "the"
     room is.  When I don't, as here, I get narrative suspense,
     which may be the intent.  But one normally expects the
     suspense will resolve.  I have made the English translate in a
     literal fashion that suggests what Bob's descriptor choices
     would be interpreted as.

.i lo xadni pe le  nanmu cu vreta         lo loldi
   A  body  of the man      reclines upon a  floor.

     "pe" is an unnaturally loose 'possessive' for this relation.
     Bob could have used "be" since the man could be the x2 of
     "xadni".  Normally one thinks of a body as inalienably
     associated with a particular person, so "po'e" would seem more
     natural than "pe".
     
.i mi viska le  flecu   be loi ciblu bei fo       le  xadni
   I  see   the current of     Blood flowing from the body.

     Bob C.'s Note: I suggest that "flecu" be given the same form
     as "fall":
     "x1 flows to x2 ..." instead of the current:
     flecu  fec fle      flow       current of/in..flowing
     to..from..'flush'
     
     Bob L. responds: That would be a different, and narrower
     meaning of "flow".  This is the 'noun' version of "flow",
     which can include oceanic currents, etc.

.i mi sisku loi sinxa         be le                           zekri
   I  seek      Signs/Symbols of the described-as-a (alleged)
crime.

.i mi viska loi kevna pe              loi danti ge'u be lo sefta
be lo jubme
   I see        Holes associated with     Bullets    in a  surface
of a  table.

     Presumably the bullets made only one hole each, so "lo kevna"
     and "lo danti" might be better than "loi".  "lo" can apply to
     plural things, as long as the statement is true of each sepa-

                                124
     rate item. can apply to plural things, as long as the
     statement is true of each separate item.

.ije mi viska lo nu           loi cukta pu farlu      lo kajna
lo jubme e   lo loldi
 And I  see   an event that       Books fell     from a  shelf  to
a  table and a  floor.

     There is no inference permitted that this "lo loldi" is the
     same floor as the one that the body was on.  This is one
     problem with "lo", which is never restricted unless explicitly
     so.

.ije  mi viska lo nu         lo canko cu kalri
  And I  see   an event that a  window   is open.

                                125
.i mi catlu   lo plita ke bartu    drudi noi   lo'e      prenu
cadzu
   I  look at a  flat  ,  exterior roof  which a typical person
walks (on it).

.i mi cusku fi la tam. noi              pulji  ku'o    fe lu   pe'i
le  zekri prenu
   I  say      to Tom, who-incidentally polices,          "    I
think [I opine] the crime person
     pu cpare      le  plita ke bartu    drudi   le  canko  pe le
     kumfa li'u
     climbed  over the flat ,   exterior-roof to the window of the
     room. "

.i la tam. cusku lu ia.           ie.
      Tom  says  "  Yes (belief), I agree.

.i ko catlu   le  kevna  be le bitmu be'o poi   ke'a trixe     le
pixra
      Look at the cavity in he wall       which it   is behind the
picture.

.i le  kevna cu vasru    lo tanxe .ije  ri                   kunti
zo'e                  li'u
   The cavity   contains a  box     and The last referent is empty
of something unspecified."

.i la tam. cusku  lu ju'e            le  tanxe pu vasru
loi rupnu li'u
      Tom  says,  "  I conclude that the box   contained  [full of]
Money.    "

     The "full of" is not implied in the Lojban.  Perhaps "vasru
culno" would do so.

.i mi catlu   le  vorme pe le  tanxe  pe                le  bitmu
   I  look at the door  of the box in [associated with] the wall.

     Again, "vorme be le tanxe" seems better.  ('Use the place
     structures, Luke')

.i mi cusku lu ba'a     le  stela cu porpi
    I say   "  I expect the lock     is broken.

.i .ua        .ue        mi facki         lo  za'i     ge   lo
vorme gi  lo stela na     porpi  li'u
  (Discovery! Surprise!) I  discover that the state of both a  door
and a  lock  is not broken."

.i mi cusku fi la tam. fe lu le  minra  pu farlu      lo bitmu   lo
loldi  gi'e pu porpi li'u
    I say      to Tom,     " The mirror fell     from a  wall to a
floor  and  broke.   "

.i la tam. cusku lu pe'i              le  morsi nanmu pu lacpu le
minra     lo loldi      lo bitmu
      Tom  says  "  I think [I opine] the dead  man   pulled   the
mirror to a  floor from a  wall.


                                126
.i se'o           mi'o          catlu      lo sinxa   da        poi
     ke'a vajni            lo nu
        sisku   li'u
   I have a hunch you and I are looking at a  sign of something
     that [it] is important for an event of seeking."  ("se'o" is
     listed in the JL14 changes.)

.i la tam. cusku lu ra'u            ju'e          lo prenu   poi
ke'a     pu sazri le  stela
      Tom  said, "  Most important, I conclude    the person who
(he/she) operated the lock
              lo za'i           kalri ku'o    djuno fi    lo tadji
be lo pu'u       kalri sazri
 with goal of a  state-of-being open          knows about a  method
of a  process of open  operating
      le  tanxe vorme li'u
      the box   door. "

     I made some minor changes to Bob's original of this and the
     line so it would fit the place structure of "djuno" described
     in the sumti-raising article above.

.i mi cusku lu ganai tu'a                 le  zekri prenu goi ko'a
ge   kalri rinka  le  stela tanxe
   I  said   " If    (in doing something) the thief           (x1)
both open  caused the lock  box

    ginai       spofu  rinka  tu'a ri
    and did not broken causes      it (the lock box) (to be
something; i.e., broken).

    gi   ko'a cu djuno fi    lo pu'u       kalri sazri     le
tanxe vorme
    then he      knows about a  process of open  operating the box
door.

     This is an excellent example of both forethought logical
     connectives (which Bob figured out how to properly do even
     though that textbook lesson has not yet been written), and of
     the necessity for sumti-raising (I added the "tu'a"s).
     Without marking the sumti-raising, it is especially easy to
     see that "spofu  rinka  ri" does not translate properly:
     "ri", the lock box is the x2 place of "rinka", an event
     caused.

.i .ua          ru'a        ko'a catlu     le  se minra  be le  nu
kalri sazri     le  stela
   (discovery!) I postulate x1   looked at the reflected of the
event of open  operating the lock
     tanxe vorme  sepi'o     lo darno ke      catlu      cabra
li'u
     box   door   using tool a  far   type-of looking-at apparatus
[telescope]."

___________________________________________________________________
_________________________________

Following is the translation of John Cowan's story:

pamoi xamrei ra'a lo verba

                                127
first funny-question associated-with a child
ni'o la paf. cusku <<lu pau mazo'o crino gi'e dandu le bitmu gi'e
siclu li'u>>
(New subject) Dad says "What (funny!) is-green and hangs-on the
     wall and whistles?"
.i la ver. cusku <<lu .uanai mi na djuno li'u>>
Kid says "(Confusion!) I do-not know".
.i la paf. cusku <<lu .ui lo me la clupe'as. xarengus. finpe li'u>>
Dad says "(Happiness!) A Clupeas-harengus type-of fish [a
     herring]".
.i la ver. cusku <<lu .ia ri goi ko'a na crino li'u>>
Kid says "(Belief!) That-last, or it1, is-not green".
.i la paf. cusku <<lu fu'i le nu ko gasnu cu rinka le nu ko'a ba
crino li'u>>
Dad says "(Easy!) The event-of you (Imperative!) acting is-the-
     cause-of the event-of it1 will-be green".
.i la ver. cusku <<lu .iasai ko'a ba'e na dandu le bitmu li'u>>
Kid says "(Belief-moderate!) It1 does-not hang-on the wall".
.i la paf. cusku <<lu fu'isai le nu ko gasnu cu rinka le nu ko'a ba
dandu li'u>>
Dad says "(Easy-moderate!) The event-of you (Imperative!) acting
     is-the-cause-of the event-of it1 will hang".
.i la ver. cusku <<lu .iacai ko'a ba'e ba'e na siclu li'u>>
Kid says "(Belief-intense!) It1 does-not whistle".
.i la paf. cusku <<lu fu'icai mi pu cusku lo jitfa li'u>>
Dad says "(Easy-intense!) I (past) express a false-thing."

     ________________________________________________________
                                 
            Matches between Set #2 and Set #1 Aphorisms

1. (1) An idea isn't responsible for the people who believe in it.
     - Don Marquis
K. lo sidbo cu na fuzme le prenu poi krici ri

2. (1) The early bird gets the worm.
L. le clira cipni cu cpacu le curnu

3. (1) I am a part of all that I have read. - John Kieran
G. mi pagbu ro le se tcidu be mi

4. (1) I've been rich and I've been poor; rich is better. - Sophie
     Tucker
A. mi pu ricfu  .ije mi pu pindi  .i la'ede'u cu xagmau

5. (2) A camel looks like a horse that was planned by a committee.
     - Vogue magazine, July, 1958
R. lo kumte cu simlu lo xirma poi se plafinti lo kamni

6. (2) God cannot alter the past, but historians can. - Samuel
     Butler
H. lo cevni ka'enai galfi loi purci  .iku'i lo circtuca ka'e go'i

7. (2) An atheist is a man who believes himself an accident. -
     Francis Thompson
I. lo no'e cevni krici cu krici le nu ri cu snuti

8. (2) A stitch in time saves nine.
N. pa nunfenso pe ca lo zantemci cu fanta so nunfenso


                                128
9. (2) To know all things in not permitted. - Horace
M. na curmi le nu djuno roda

10. (2) All cruelty springs from weakness. - Seneca
P.  ro le nu kusru cu se rinka le ka ruble

11. (3) Nothing is so admirable in politics as a short memory. -
     John Kenneth Galbraith
B.  le tordu temci morji cu traji se sinma sera'a le turni
    or
T.  no da pe le turni cu dunli se sinma lo tordu ni morji

12. (3) The only thing we have to fear is fear itself. - Franklin
     Delano Roosevelt
D.  le ka terpa cu me lo pa drani se terpa

13. (3) On the whole I'd rather be in Philadelphia. - W. C. Fields
J.  te'inai mi zmadu djica le nu zvati la Filydelfias

14. (3) Faith is never identical with piety. - Karl Barth
F.  lo ka krici le cevni cu noroi mintu lo ka prami le cevni

15. (3) Early to bed and early to rise / Makes a man healthy,
     wealthy and wise.
C.  le nu clira ckaklama gi'e clira ckacliva cu rinka le ka kanro
     joi ricfu joi prije

16. (3) A big lie is more plausible than truth. - Ernest Hemingway
E.  le banli to'e jetnu cu zmadu le jetnu le ni se krici

17. (3) To kill time, a committee meeting is the perfect weapon. -
     Laurence J. Peter
O.  fe le nu catra xaksu lo temci kei fa lo kamni nu penmi cu prane
     xarci

18. (4) Who is more foolish, the child afraid of the dark or the
     man afraid of the light? - Maurice Freehill
S.  le verba poi terpa le manku ku'o ji le prenu poi terpa le se
     gusni cu bebna traji

19. (4) He who lives by the sword shall perish by the champagne
     cocktail. - Saul Alinsky
Q.  da poi renvi sepi'o loi balre baca'a mrobi'o sepi'o loi sodva
     vanju se pinxe

___________________________________________________________________
                 _________________________________
                                 
              Nick's translation of a Greek newspaper

lonu reroi pacnunzu'e cu na se ckaji lo prije
     ni'oni'oni'oni'o la fredis. germaNOS. pu finti  .i la kir. pu
te pinxe pamo'o  .i memi'a poi te prosa
     ni'oni'oni'o nuzba
     ni'oni'o

[Title] Events of twice sinning [evil-events-of-acting] are not
     characteristic of the wise ones.
[Author] Freddy Germanos created.
[Section] Kir is Drunk From, Section I. We Who Write Prose.

                                129
[Article] News


                                130
le briju cu so'iroi se klama so'olemi pendo poi co'a se jibri zu'i
po mi gi'e preti cpedu le tarmi po'e lo nuzba nu cusku

The office is oftentimes come-to-by several of my friends who
initially are be-jobbed by that typical of us and who
questioningly-request the form which is inalienable to news
expressings.

To the office often come friends who are just getting started in
our profession, and who ask me how to write news.


     ni'o .i'a loi nuzba ne semau roda vajrai ci'elo karni nu cupra
.iku'i cinri fa lo karni poi mu'i lenu ke'a pilno ku cusku le go'i
.i rolo karni cu ckaji leri tadji be le nuzba nu cusku  .i la'edi'u
nibli lenu ra se tcidu co frili gi'a se tcidu naku

(Acceptance) News, more than all-other-somethings is superlatively-
important in-system journal-producing.  However, interesting is the
journal which, motivated by it using, (something) expresses News
[the x1 of the previous bridi].  Each journal is characterized by
its method of news-expressing.  This [that they are so
characterized] necessitates that it [uncertain - could be the news-
expressing, the method, each journal] is a read-thing of-type easy,
or is a read-thing not-so.

News is of course the Alpha and Omega of journalism [Greek: public-
writing].  But what's important is which paper you write it for.
Each paper has its own style of writing news.  This results in them
being read more easily - or not at all.

     Bob: "News, more than all-things, is superlatively-important."
     - This is highly redundant.  There are alternate phrasings
     using the places of "traji", but the minimal change is "loi
     nuzba ne semau da'ada vajni" = "News, more than all-other-
     things, is important."
     
     I modified one phrase, where Nick used "lo karni poi jaipi'o
     cusku le go'i", since the usage is not on anyone's cmavo list,
     and was so vague, I had trouble figuring it out WITH knowledge
     and the English.  "jaipi'o" is the new tagged-sumti-place
     converter/extractor (the equivalent of a SE conversion)
     intended to help clarify sumti-raising.  Since Nick's version
     elliptically omitted the "ke'a", you have to guess where it
     was supposed to go.  I made the following transformations,
     which are the obvious deductions and equivalences:
     
     "lo karni poi <jaipi'o cusku le go'i>"
     "lo karni poi <ke'a jaipi'o cusku le go'i>" (inserting "ke'a"
          at the beginning)
     "lo karni poi <pi'o ke'a cusku le go'i>" (The "jaipi'o" means
          that "ke'a" is the "pi'o"-tagged sumti of the selbri
          (which is "cusku")
     "The journal which used by it [something] expresses the x1 of
          the previous sentence (News)".
     This can possibly be interpreted to mean the colloquial
     English, but has more the sense of "which paper uses your
     writing".  Clearly the emphasis in this article is on the what
     the writer writes and why, and the motivational implication
     ("mu'i") of "who you write it for" is lost.  Even if Nick's

                                131
     original means what he wanted, the "jaipi'o" is far more
     opaque than the equivalent "pi'o ke'a".


     ni'o mupli lo nuzba poi mutce sampu  .i lo nixli be li mu ca
lepu'u kelci levo'a bolci cu farlu lo skuro poi karbi'o se kakpa
lei gunka pole ta'utru

Example of a news item which is much simple.  A girl of age #5,
simultaneous with the process of playing with x1's [her] ball,
falls to a groove which open-becomingly is-dug-by workers owned by
the city-government.

Let's examine a very simple news item.  A girl, aged 5, while
playing with her ball, falls into a trench opened by a Municipal
crew.

     The age place on "nixli" seems wrongly-expressed.  I think
     more correct is "lo nixli be lo nanca mumoi", optionally
     omitting the "nanca".


                                132
[There follow five different reports, one upper-class in archaic
prose concentrating on the famous socialites living next door to
the trench, one sensationalist speculating on rape, one communist
calling for a class struggle, one anti-communist hinting at Red
sabotage, and Germanos' own paper claiming she fell in because her
father did not buy said paper.]

.i tarmi le ve cusku be le nuzba bei pi'o la karni be de'i roboi
lo djedi be'oku (to la roldei toi)

Form of the medium of expression of the news, used by the-one-
called Journal Associated with Date Each-Day (The Each-day).

[Here's] how you'd write the news in the Daily.

     [Bob: I changed Nick's journal title to match the style of the
     newspaper.  Nick had originally used the parenthetical name]


{Very archaic Greek follows.  I've emulated it by expanding all
tanru, and lots of terminators.}


.i <<lu di'o lo paboi te kruca be le klaji poi se cmene la
.adriaNOS. vauku'o bei le klaji poi se cmene la paleologos.
ku'obe'oku ko'a goi paboi lo nixli be muboi lo nanca be'oku ge'u pu
se xrani ca lepu'u ko'a cu kelci da noi ke'a se tarmi lo bolci
vauku'ovaukeivau

"At the locus of the single crossing-point of the street which is
named 'Adrianos' with the street which is named 'Paleologos', it1,
defined as one girl of age 5 years was-injured, simultaneous with
the process of it1 playing with something, which it has form of a
ball.

"At the intersection of Adrian and Palaeologus streets a five year
old maiden was injured while she did ludificate with her sphere.


.i le nu xrani vaukei cu diklo fe'eba'o va paboi le zdani be lai
paleologon. benizelon. be'oku noi la kamBUroglus. pu skicu ke'a
ta'i lo se ciska ku tai loka lanli vaukeivauku'ovau

The event of injury is local to beyond [something] near one of the
nests of the mass called Paleologon-Benizelon, which [house]
Kamburoglus described it in form inscribed, in manner analytical.

The accident occurred directly opposite the domicile of the
Palaeologus-Benizelos family, about which [the house] Kambouroglou
has written analytically.


.i la'edi'e cu cizra zo'e lenu fasnu vaukei ne sekai leka na xlura
vaukeige'uvau

That referred by the following utterance is bizarre to unspecified
one(s) in occurrence characterized-by not-influencer-ness.

By a most strange coincidence,


                                133

.i paboi lo reboi rirni be la kamBUroglus. be'oku pu binxo le speni
be paboi lo reboi rirni be la kamBUroglus. be'oku poi na du vo'a
ku'o soivo'avo'e se'u be'oku vi le zdani be lai varvatis. be'oku
noi jibni le zdani be lai paleologon. benizelon. be'okuvauku'ovau

One of the two parents of Kamburoglus became the spouse of one of
the two parents of Kamburoglus which was not = x1 [the first 'one
of the two parents of Kamburoglus'], and x2 to x1 [or "vice versa";
the latter became the spouse of the former, too] at the nest of the
mass called Varvatis which is near the nest of the mass called
Paleologon-Benizelon

The parents of Kambouroglou were bewedded in the Varvatis domicile,
which is next to the P-B domicile.

                                134
.i le nixli ki na se ckape vau li'u>>

The girl is presently not imperiled."

The maiden is out of danger."


.i ke'unai vecu'u la akROpolis. le nuzba cu ka'e ckaji loka zmadu
cinri bo cfipu

Continuing, in medium the Acropolis, the news can be characterized
by more interestingly-confusing.

On the other hand, in the Acropolis the news item could take on a
more enigmatic character.  {Note very dreadful tanru in the
following.}


.i <<lu pu zgana ne'i lo skuro lo se xrani. goi ko'a poi mebrai
nixli gi'e jarco lo frili se viska jalge be lonu vlile

"Observed, inside a groove, an injured, hereinafter it1, who is
beautiful-mostly girlish, and displays easily-seen results of
events of violence.

"There has been found wounded, in a ditch, a most beautiful maiden
bearing obvious signs of abuse.


.i ri'anai lenu le zekri lifri po'u ko'a na pujaca skicu fi lei
pulji kei ca cipra fa ri lejei na'e snuti (to pupu jinvi to'ebori
toi) gi'e cinse zekri

Despite the-event the crime-experiencer who is it1 not before-or-
now describes to the police, simultaneously testing by them [the
police] the truth value of other-than-accidental (had earlier
opined opposite-of-this [accidental]) and sexual-crime.

Although the victim has not yet made her statement, the police is
investigating whether this was no accident, as had been thought
initially, but a crime of a sexual nature.


.i lo kamni poi se cmima le mi karni zbasu caca'a cipra fi le tcini
mu'i lenu djica co danfu fo re preti ra'u po'u di'e

A committee which is bemembered by our journal-makers is presently
testing among the situation motivated by the event of desiring of
type answers to two questions, chiefly, which follow.

A team of our editors is already investigating the matter to answer
two main questions.


.i pamai xu le nixli pu cinse vlile lifri

First, Is it true that the girl was a sexually-violent-experiencer?

a) was the maiden raped?


                                135

.i remai fau lenu ko'a na'eke cinse vlile lifri kei ko'a na'eke
cinse vlile lifri ki'u ma li'u>>

Second, in the event of it1 [the girl} other-than sexually-violent-
experiences, it1 other-than sexually-violent-experiences justified
by what?"

b) if the maiden was not raped, why was she not raped?"


                                136
.i li'a vecu'u la cermurse sesau se basna fa loi drata tcila

Clearly, in media-of-expression the-one called morning-twilight,
necessarily, are-emphasized Other Details.

Of course in the Dawn other points must be emphasized.  {Very
folksy/dialectical Greek follows}


.i <<lu nixli  .i ko'a pe leni slabu  .i ri du mu nanca  .i xrani
ca le purlamdei  .i farlu lo skuro po la ta'utru ca lenu kelci le
bolci

"Girl.  It1 [not defined, but obvious.  This fits the style.]
associated with age.  Age = 5 year-intervals.  Injury simultaneous
with the past-adjacent-day [yesterday].  Faller to a groove owned
by the-one(s)-named City-Government, simultaneous with the playing
with the ball.

"A li'l girl, 5 yr old, got hurt yesterday, falling, while she was
playin' with her ball, into a City ditch.


.i ko'a ze'iba tavla palemi karnypra  .i ko'a fatci xusra

It1 after-the-momently talks to one of our journal-producers.  It1
factually asserts.

Talking later to one of our editors, she put things straight.

     There is current debate going on that may change the
     interpretation of "ze'iba".  I left it unchanged on the basis
     of the published cmavo list.  Whatever is eventually decided
     will be reflected in the in-progress paper on interpreting
     tenses, and in updates to the cmavo list.


.i <<lu mi farlu le skuro  .i ni'ibo mi na citka ca re djedi  .i
besna se cfipu

"I fall to the groove.  Logically-necessary because I not eat
during 2 days.  Brainily-confused-by.

'I fell in that ditch cause I hadn't eaten for two days.  I bin
dizzy.


.i lemi mamta cu selfu seljibri  .i lemi patfu cu na'e dinycpa  .i
ni'ibo ri roroi pantydzu li'u>> li'u>>

My mother is servant bejobbed.  My father other-than money-gets.
Logically-necessary because he always [Nick had a complex tense
here that wasn't quite right, added nothing, and seemed like more
than a 5-year-old's mouthful] protest-walks.'"

My momma works as a servant.  My poppa can't get a day's wages
cause he's always on protest marches.'"


.ice ka'e zgana cusku fo la natmi nu'arki'a

                                137

And next (not in any order) can observing-express in-medium the
National News-Crier.

While in the National Herald it could be noted that:


<<lu lenu xrani cu se rinka lonu pimo'ale skuro cu te sabji lo
gacri

The event of injuring is caused by an event of too-little-of the
groove being supplied with a cover.

"The accident is due to the incomplete covering of the ditch.

                                138
.i ni'inai lenu na catni se xusra cu cusku lenu fuzme fa relo gunka
pe le ta'utru ge'u zi'e noi ze'u .o'onai kaurposysi'orpre

Logically despite [Note how Nick has both political extremes prone
to invoking "Logic" in attacking each other.  Very cute, Nick!] the
event of it-being-false that [something] is-authority-asserted,
[something] expresses the-event-of Responsible are two workers
associated with the city-government and incidentally, for-a-long-
time (Anger!) common-owner-idea-persons.

Although there has been no formal announcement, it is said that
those responsible are two City workers, who are longtime
communists.

     Nick had omitted the "zi'e", which made the second relative
     clause apply to "le nu fuzme".  Multiple relative clauses and
     other sumti modifiers attached with selma'o GOI and NOI is one
     place that the language can get clumsy, and you can easily
     make errors.  So I recommend using "zi'e" to join such
     multiple modifiers.
     
     Nick's lujvo for "communists" is not the best I've seen.  JCB
     used "ownerly-common-believer".  I think this is close, but
     backwards - this is best shown by using the inverted form of
     the tanru:  it should be "ownership of type common", and not
     "common things of type owner".  I prefer the similar "common-
     owner-believe" or better-by-massifying:  "community-owner-be-
     lieve".  However from the standpoint of the conservatives who
     wrote this article, perhaps "worker-own-believer" or "public-
     owner-believer" might convey some of the apparent distaste for
     the 'lower-class' orientation.  (Anyone want to work on lujvo
     for a variety of political and social credos?  What is the
     linguistic difference between "Democrats" and "Republicans"?
     
     I also think "mabla" (or "mal-", I'm sure the newspaper would
     use it enough to lujvo-ize it) on the front of whatever the
     word for "communists" would be better than the attitudinal,
     which seems rather to fit the style of the sensationalist
     tabloid.  (Incidentally, I think all quoted text from
     supposedly 'real people' would be richer in attitudinals in
     all versions of the story.  We use a lot of attitudinals in
     conversation here.  This would heighten the distinction
     between the repertorial voice and the voice of the people
     being interviewed.)


.i lei pulji cu cipra lejei zasti fa loi flana'etinbe gripre noi pu
zukte lo drata nu jecyselxrazei li'u>>

Police are testing the truth of existence of Law-non-obeying Group-
People [massed] who acted at other events of state-injured-crime."

The police is investigating whether this is an illegal network,
which has acted out other acts of sabotage in the past."

     Nick should consider variations and compounds of "sisku"
     ("seek") or fakro'i ("discover-try") or "lanli" ("analyze")
     "cliro'i" ("learn-try") for "investigate", which he has
     translated as being the same as "test" in all versions of the
     story.  I liked his 'conservative' translation of "sabotage"

                                139
     as a crime that injures the state, though I added the "sel"
     rafsi to make sure that no one thinks that the polity did the
     injuring.


.i romai la deimid. cfari le ni'onrai tarmi be loi nuzba

Finally, the-one-called Day-Middle initiates the new-most form of
News.

The Midday has established a totally new style in news:



                                140
.i <<lu se tirna fa lo voksa  .i se viska fa lo xance poi se desku
tai loka ti'e pacna zi'epoi cpedu lenu sidju

Is-heard, a voice.  Is-seen, a hand which [it?] is-shaken-by by-
method a quality (I hear!) of hope and-which requests the-event-of-
help.

"A voice was heard.  A hand was seen floundering desperately,
asking for help.

     Nora suggests "slilu" for "se desku" but I'm not sure I agree.
     The "tai loka ti'e pacna" meant nothing to me.  (What did
     people who tried reading this guess it meant?  This kind of
     feedback will be helpful to Nick, and indeed to all of us, in
     learning to think from the listener's point of view.  I
     welcome such comments on any other places in the text where
     you read something different in the Lojban than Nick or I
     represented in our translations.  Help us learn Lojban!)  I
     would suggest using "mutnitcu" ("much-needer") instead of
     "pacna" for "desperate", "seci'o" for "tai", and moving the
     attitudinal to show that it is the emotion expressed that is
     hearsay.  "se desku seci'o ti'e loka mutnitcu" is "shaken-by
     [something] expressing-emotion (I hear!) much-need".


.ibabo smaji  .i ?ma pu fasnu  .i mi ciksi

And-then silence.  What occurred?  I explain.

And then, silence.  What was happening?  [I answer my own
question:]


.i lo nixli pu sakli mo'ine'i lo skuro  .ibazabo le nixli cu bacru
<<lu mi farlu ri'a lenu la paf. na di'i tcidu la deimid. li'u>>
li'u>>

A girl slid moving-to-the-inside a groove.  And a bit later the
girl uttered 'I fall because the-event Pop not regularly reads the
Day-middle'".

A girl had slipped into a ditch.  Later the girl said:  'I fell,
because Dad doesn't read the Midday!'"

     An awesome effort, comparable to and perhaps as germinal as
     Athelstan's translation of Saki's "The Open Window" (see
     JL10).  Let's hear a round of applause for Nick, such that he
     might hear it Down Under.

___________________________________________________________________
                 ________________________________
                                 
                    Translation of "lei lojbo"

Nora: (Completion!) I am able to go, justified by Hope taking care
          of Katrina.
Friend:   (I hear!) Hope speaks Esperanto and not Lojban.
Nora: True. But Katrina, also, speaks Esperanto.
Friend:   Approval!
Hope: (in Esperanto) Takingly-pull (imperative).

                                141
Sam:  (to himself) "prEnutrEne" is-the same-as "prenu trene" (=
          "person-train").  (Confusion!) I don't sense that
          represented by this [text].
Sam:  (aloud) The station is excessively far.
Katrina:  (in Esperanto) Mystery
Sam:  (To himself): "mistEro" is-the same-as "mi stero" (= I am
          measured in steradians as (ellipsized amount).
Sam:  (thinking) Katrina expresses the symbol for a falsehood.
          This state-of-affairs is bad for Computers.
Sam:  (aloud) Users! (pejorative) (which sounds in Esperanto like
          "less than")
Hope: (in Esperanto) Less than what?

                      Till next issue.  co'o.