response to coranth - gismu and 'h'

[lojbab (Bob LeChevalier)]

Coranth makes several comments on the gismu list and place structures.
General comment - having gotten comments on place structures from 3
people in one day, I have decided to set up a file to save them so they
don;t get lost, but I can;t promise to look at them all now - they will
be considered eventually - evry one of them.  So by all means contribute.

It is useful, but not mandatory, that you check your comments against
the in-progress longer-definition gismu list, which is available on the
P. L. S.  Some of Coranth's comments I can see without checking myself,
did not use this list.

Note that the gismu list is baselined, and suggestions to add or delete
are unlikely to be accepted without very strong argument.  Changes to
keywords are nearly as strong a taboo.  But you can propose any of these
as well.  Place structures are fair game, and also words that you think
need short rafsi that don't have them (a minimum of 10, and preferably
20 lujvo are likely to be needed to cause a change, though).

>"smadi"/"sruma"/...     1 guesses/assumes/.... 2 is true
>
>(there's a whole group of related words), add [by clues/evidence... 3]

If you are guessing or assuming then you may be doing so apriori without
evidence.  Use one of the causals in BAI to attach this place.

>speni" 1 married 2     add [under law/custon/system 3]

worth considering - Comments welcome

>"panzi" 1 is offspring of mom 2 and dad 3
>
>                add [at time 4, place 5]        eliminate "jbena"

You can claim that any non-eternal event needs a time and place of
initiation.  But this is a truism.  I suspect jbena may end up
metaphorically being stretched farther than panzi, just as it has in
English.  The time/place are included specifically because birthdate and
birthplace are important in many cultures (bureaucracy and astrological
and celebratory reasons, at least).

>"citno" 1 is young              add [, age 2]

age is "ni slabu"

>verba   1 is a child    add [ of species 2]

This may have been done and will be considered, as well as the corresponding
for cifnu, if necessary.

>"fatci" 1 is an absolute fact - I don't understand this concept.

jetnu and the places of djuno relate to things knowable and therefore subject
to epistemology.  Some philosophers argue that all truth is relative, others
argue or assume some unarguable 'truths' that are absolute.  The latter is
fatci, and is needed at the very least to argue just such philosophical
points.

>"benji"/"muvdu"/"bevri"  1 xfers/moves/carries 2 to 3 from 4 via path 5
>
>        why have three different gismu for the same relation...?

Having the same place structure doesn not mean they are the same
relation or the same concept.  I am sending this to you (benji), but not
muvdu or bevri it.  There are many ways to move something without
carrying it (push/pull/teleport)

>"notci"  1 is a message about 2 to 3 from 4
>"xatra"  1 is a letter  about 4 to 2 from 3
>
>        same relations, different ordering

Again different concepts.  There is a big distinction between a business
message (a memo) and a business letter.  Except in some diary styles you
don;t write letters to yourself, but you might write notes.

The place structure ordering is significant - in a note, the subject is
most important, in a letter, it is the receiver.

Note that if there is semantic difference between two concepts, there
might be good reason for them to be separate gismu justified by their
differing use in lujvo-making.  There are several gismu, including the
soon to be discussed gender-specifics, that are separate because some
are separately useful in lujvo-making in many cultures.

>"bajra"  1 runs(/walks ?) to 2 from 3 via path 4
>"klama"  1 goes           to 2 from 3 via path 4 by means 5
>
>        these I can see as different iff "bajra" also means walk (by mean foot)

cadzu means walk.  The place structure of these was changed as of Draft
Textbook Lesson 4A, and are also in the working list.

>"cliva" 1 goes                 from 2 via path 3 by means 4
>
>        they have to be coming to somewhere, which makes the relation
>        the same at "klama"

Why? have you never left somewhere with no specific destination.  Have
you never told someone in anger to "Leave!"? litru is also the same
relation but with no origin OR destination - just a route.  The planets
litru, but one would be hard pressed to say they klama or cliva.

>"prenu" 1 is a person          add [of species/race/culture/planet 2]

worth dicussion, but I want more justification.  I personally don't like
the change because it suggests that the classification is important to
person-hood.  I consider my cat a person, but the species is not
relevant, indeed that is the whole point.

>"nanmu"/"fetsi"/"nanla"/"nixli"
>
>        these are (resp.) male/female adult, male/female child
>
>        add [of species 2], pushing the childs' place [age 3]
>
>Personally, I think adding seperate gismu only to allow gender distinctions
>is a bad idea unless necessary ("mamta"/"patfu")

Define "necessary" in a culturally-neutral manner.  Why are the two you
mentioned necessary and not the others?  The move towards
gender-neutrality may be politically correct but is not culturally
neutral.  For some cultures the distinctions given are very important,
for some not.  Some make greater distinctions (Chinese distinguishes
between older and younger brother and sister as separate roots.  Latin
used to distinguish between father's siblings and mother's siblings as
separate roots, and the uncle/aunt of English are the descendants of
those roots (whereas most Romance languages chose just one e.g.  Spanish
tio/tia)).

Kinship terms are among the most varied and most studied concepts in
linguistics, and I'd rather Lojban have a surplus than a shortage.  Same
with colors.

Your pairing was wrong by the way  - it is nanmu/ninmu nakni/fetsi
nanla/nixli, and you left out bruna/mensi.

Finally think about how your tanru would appear in lujvo.  There is a
strong aesthetic thus far against >4 term lujvo.  If man is "male adult
human", you've used 3.  Now try making lujvo specific to men.

>In those cases where a distinction is necessary, a tanru or lujvo should
>be used
>
>        fetre'a (female human - woman)
female-adult-human/adult-female-human
>        nakre'a (male human - man)
male-adult-human/adult-male-human
>        feipre  (female person)
>        nakpre  (male person)
not now a gismu - these seem good and useful lujvo

etc.

>All these are examples of cases where a gismu has been given just to make
>the gender distinctions simpler to translate.

Yep.  And since people make the distinctions in every language, then we
are engaging in Newspeak engineering to make them unnecessarily
difficult, especially since they are among the most common of words.

>BEGIN SOAPBOX
>
>A bridi is a RELATIONSHIP.  There should not be seperate bridi for the
>same relationship.  Gender distictions which do not affect the
>relationship should be left out by default, and expressed only if
>necessary.
>
>END SOAPBOX

You may be confusing bridi and gismu. You've been arguing about gismu
which are not bridi.

In any event, guessing what you mean, I disagree.  There may end up
in Lojban hundreds of roughly equivalent lujvo for a given concept,
each differing only in subtle shades of meaning.  Example:  is there
a difference betwenn adult-female-human and female-adult-human.  The
two lujvo will probably both be coined and used - and sooner or later
will develop a distinguished meaning.

The rule is that every brivla should have only one meaning, NOT that
every meaning have one brivla.

You have every right to use gender neutral words.  For each of the
gender specific words we have a gender neutral equivalent if gender is
unimportant (which is why they do not have gender as a place).

>Another way to handle the problem is to add an additional place to
>"remna","prenu","verba","cifnu","tunba", etc.  [gender .....]

see last comment

>Freeing up many of the redundant gismu would also allow slightly more
>even distribution of rafsi for those which gismu which SHOULD get used
>more (the ones listed above).

As noted, gismu changes are not being accepted, and changing rafsi is at
this stage not a good reason.  The tuning that went into the rafsi list
was done on the set as a whole and took a LONG time.  If I were to do
anything more than incremental changes, I would do the whole thing over
again using the much better usage data we could probably generate now.

Note that many gismu changes offer no benefit in rafsi assignment.  The main
problems in rafsi assignment are tight packing in the b's c's, j's, k's,
p's and s's.

>lojban does not use the symbol 'h' for any purpose, and all the
>functionality of it which la lojbangirz feels is necessary is taken by
>the apostraphe.  I have no real problem with this.
>
>However, lojban also does not have a 'th' sound.  Why not use the
>symbol 'h' for that...?

You answered your own question - because Lojban does not have the 'th'
sound.  Why give a letter to a sound we don't recognize?  Especially one
that is rare outside of English - and there are two sounds anyway the
voiced th of the and the unvoiced of thistle.

Note that we also do not use q or w.

If we ever did make a change, it would be to assign two letters to the i
and y in diphthongs.  This was considered once but rejected by the high
standard for baseline changes, and the fact that we would have to use 'q'
or 'h' for a vowel sound, when neither are used for vowels in any
language I know of.

lojbab