Some ideas/questions (long)

["oskar2379" <oskar2379@hotmail.com>]

Hello, I am new to this group but I have been studying lojban for a 
while (actually, I've been paying most of my attention to the grammar 
rather than vocabulary). I don't have any specific points of the 
language that I need explained, but rather I have questions about 
broad aspects of the language. Four in particular...

1. I noticed that one must not only learn the (~3,000 word?) 
vocabulary of gismu, but also the several forms of rafsi for each 
gismu. This seems a little unnecessary; why not have just one rafsi 
for each word? Or better yet, why not use the three-letter rafsi all 
the time? That would shorten text considerably and, when the words 
are only one syllable long, it could reduce the need for all the word 
shortcuts that may also be complicated to learn. Not only that, but 
using rafsi full-time means that, in order to make lujvo, all you 
must do is jam words together.

2. Speaking of lujvo, I noticed some things I don't understand. First 
of all, they are put together in English-order when it makes much 
more sense for them to go in the opposite order. For example,

the tanru 'cmalu prenu' means small person, but lojbanically it looks 
like 'cmalu' is going in the x1 place of 'prenu'. It would make more 
sense to go the other way (I know there's a way how to, but it should 
be the default order).

3. Also, I thought of a way in which you could ease the ambiguity of 
their meanings. Your book uses the tanru example 'klama jubme'. Does 
it mean 'table which goes', 'table of a goer' ('table used by a 
goer'?), or 'table when it goes'?

Couldn't the distinction be made by placing a cmavo inbetween 
them? 'table when it goes' could be 'le jubme jo klama', 'table that 
goes' could be 'le jubme klama' (no cmavo), and 'table of a goer' 
could be ... 'le jubme pe le klama' (though that wouldn't be a 
tanru...).

The problem arises with something like 'fenki litki ' (crazy liquid, 
or liquid that causes you to be crazy). 'fenki' is {x1 is crazy by 
standard x2}, there is no causative agent in the predicate, so this 
tanru is false. If we made it a lujvo with gasnu, fekygau (or gaufek, 
using my word order), it would read 'x1 causes x2 to be crazy by 
standard x3', we could make the tanru 'le litki gaufek' where 'litki' 
falls in the x1 place of gaufek, meaning that the liquid is what 
causes the craziness (and then 'le litki fenki' would mean liquid 
that is crazy). Of course, I am using the backwards word order that I 
just made up when I put together these tanru.

4. Since we are on the subject of changing predicate places, there 
were some discrepencies that I saw in the way they are. Your book 
said there were three types of relationships; static (John is the 
father of Sam), active (John hits Sam), and attributive (John is 
taller than Sam).

  4.1 In static relationships, there is only one argument, but in 
lojban this number can vary.

	It can have one...
	djacu - x1 is made of/contains/is a quantity/expanse of water

	It can have two...
	cidja - x1 is food/feed/nutriment for x2
	
	It can have three...
	cavni - x1 is a/the god/deity of people(s)/religion x2 with 
dominion over x3 [sphere]
	cipra - x1 (process/event) is a test for/proof of 
property/state x2 in subject x3
	
	It can have four...
	botpi - x1 is a bottle/jar/urn/flask/closable container for 
x2, made of material x3 with lid x4

The x2 place is almost always the possesor (...bird of 
species...cheese of source...cloud of material...deity of 
religion...test of...). But what about

x1 is a cup/glass/tumbler/mug/vessel/[bowl] containing contents x2, 
and of material x3 

This predicate seems to jam two different words into one...'x1 is a 
cup' and 'x1 contains x2'. This word seems to be an oddball and sort 
of confused me when I was skimming your word lists. Another oddball 
is the bottle/jar/urn example above...'x1 is a bottle' plus 'x1 is 
made of x2'? And why must lid be specified in this predicate?
  
  4.2 In active relationships, you see more variation in the number 
of places.
	It can have one...
	cmila - x1 laughs

	It can have two...
	sisti - x1 ceases/stops/halts activity/process/state x2

	It can have three...
	sanga - x1 sings/chants x2 [song/hymn/melody/melodic sounds] 
to audience x3
	klaku - x1 weeps/cries tears x2 about/for reason x3 
(event/state)

	It can have four...
	frati - x1 reacts/responds/answers with action x2 to stimulus 
x3 under conditions x4
	minra - x1 reflects/mirrors/echoes x2 [object/radiation] to 
observer/point x3 as x4

	It can have five...
	klama- x1 comes/goes to destination x2 from origin x3 via 
route x4 using means/vehicle x5
	fanva - x1 translates x2 to language x3 from language x4 with 
translation-result x5

Here it becomes more confusing because sometimes the receiver is in 
the x2 place, and sometimes in the x3 place. x2 is the receiver in 
sisti, x3 is the receiver in sanga. In these cases, the extra place 
is the direct object (to sing the song x2...to weep the tears x2...to 
respond with action x2...to reflect the object x2...).

klaku offers a reason for the action in x3, yet I believe lojban 
already has words for 'for the purpose/reason of...' so I don't see 
why this is necessary.

x4 and x5 are more complicated. The 'under conditions' in frati 
sounds like a connective to me (the logical connective IFF) and thus 
shouldn't be in the predicate. The 'as x4' in minra refers to the 
state of the direct object after the action. Similarly, the x2 in 
klama refers to the state of x1 after the action (or more 
specifically, the location). But in the latter case, it is not the 
direct object. To explain why, lets compare it to another predicate...

x1 comes/goes to destination x2 from origin x3 via route x4 using 
means/vehicle x5
x1 transfers/sends/transmits x2 to receiver x3 from 
transmitter/origin x4 via means/medium x5

Both these predicates are the very same; the difference is that the 
x5 in the first one is moved to x1 in the second one (the cause). So 
if we changed klama to 'x1 transports x2 to x3 from x4 by means x5', 
it would all work out.

minra seems to be another oddball that combines two predicates into 
one, that is, 'x1 reflects x2 as x3' and 'x1 sends/transmits x2 to 
x3...' and thus should be seperated. fanva is another oddball because 
it has two later states; the language and the actual text 
(translation-result). This could probably be altered to fit the 
system as simply 'x1 translates x2 to x3'.

To sum this up, here are all the different active relationships I can 
think of (there are probably more but they will just be different 
combinations of direct object, audience, previous state, later state, 
and means). I put brackets around the parts that don't fit my system 
(><):

x1 smiles (no direct object, no audience)
x1 stops x2 (no direct object)
x1 weeps x2 [for reason x3] (no audience)
x1 reacts with action x2 to stimulus x3 [under conditions x4] (both a 
direct object and audience)
x1 reflects x2 [to x3] as x4 (a direct object and a later state)
x1 expresses/says x2 for audience x3 via expressive medium x4 (direct 
object, audience, means)
x1 sends x2 to x3 from x4 by means/medium x5 (direct object, previous 
state, later state, and means)

  4.3 Attributive relationships can have a causative agent too, 
although lojban leaves it out; 'x1 is white' vs. 'x1 whitens x2'. 
Again there is a differing number of places...
	It can have one...
	blabi - x1 is white / very light-coloured
	
	It can have two...
	badri - x1 is sad/depressed/dejected/[unhappy/feels 
sorrow/grief] about x2 
	bebna - x1 is foolish/silly in event/action/property [folly] 
x2
	culno - x1 is full/completely filled with x2
	
	It can have three...
	cmalu - x1 is small in property/dimension(s) x2 as compared 
with standard/norm x3
	berti - x1 is to the north/northern side [right-hand-rule 
pole] of x2 according to frame of reference x3 
	
	It can have four...
	drani - x1 is correct/proper/right/perfect in property/aspect 
x2 in situation x3 by standard x4

In order for lojban to include a causative agent, you must combine 
the word with 'gasnu' (to do). This is fine, its just that some words 
seem to come with it built-in, which is confusing. Take jabre:

x1 brakes/causes to slow motion/activity x2 with 
device/mechanism/principle x3 

Compare it to masno:

x1 is slow/sluggish at doing/being/bringing about x2

masno includes the agent and the property that is slow. jabre 
includes the cause, the activity (basically the x1 and x2 of masno 
combined?) and the means. jabre should be a lujvo.

To analyze similarities in the places: x2 seems to always be the 
property of x1 that is attributed, and x3 seems to always be the 
standard/frame of referance.

An oddball that I don't think fits is fengu (x1 is angry/mad at x2 
for x3), in which the receiver of the anger (x2) and the action that 
caused the anger (x3) are seperated when it should be 'x1 is 
angry/mad about x2 (abstraction)'. Compare it to badri: x1 is sad 
about x2 (abstraction).