Lojban's Biggest Problem or Why Still Nobody Speaks It

1067 views
Skip to first unread message

selpa'i

unread,
Oct 26, 2014, 1:51:26 PM10/26/14
to loj...@googlegroups.com
Simply put, there aren't enough words, and the few words that exist do
not enable us to make nuanced descriptions of the world.

I see the two main aspects of the problem as being:

1) The existing words (specifically the gismu) are too vague by
themselves (and yes, I know that this is by design)
2) Too many essential concepts are missing completely


1. "A logical language with the dictionary of Toki Pona"

gismu cover wide ranges of meaning, but specifying specific points on
those ranges is not as straighforward as the lujvo-fans advertise it to be.

Right now, if you want to speak about your emotions, or about your
clothes, or about almost anything, in depth and with nuance and
subtlety, then you'll find that Lojban is not much different from Toki
Pona. Both let you use tanru, but neither lets you be very exact. (This
is only a slight exaggeration; Lojban and Toki Pona would be in one
corner, and all the real languages in the other corner, with Lojban
standing only a few steps in front of Toki Pona)

The idea was/is to use lujvo to get specific meanings of tanru into the
lexicon, but there is more than one problem with that. For one thing,
there aren't enough veljvo to account for all the different nuances that
one might want to base on a root gismu.

Let's consider the following example of related words that the gimste
says should be made by combining {banli} with something else:
"extraordinary, illustrious, magnificent, impressive, awesome,
grandiose, august, inspiring, special, majestic, distinguished, eminent,
splendor, stately, imposing"

If each of those words is to be a different lujvo with {banli} as
tertau, what are the all the seltau going to be? This list isn't even
exhaustive, it could easily be extended further, and at some point the
lujvo would become nothing more than a random seltau + {banli}, and then
why even bother with lujvo?


2. "Not enough words"

What do the words traffic, traffic lights, traffic jam have in common?
They are all examples of everyday words that are completely missing from
the Lojban lexicon. I could name hundreds more. And it makes speaking
the language beyond a certain level impossible.

What's the fastest way to verify this? Try talking to yourself in Lojban
for one day, no English. Write down every object and situation you
encounter during the day that you cannot name let alone describe in
Lojban (and no, TokiPona-like tanru do not count as nuanced
descriptions). You will quickly fill your entire notebook. I did this
myself a while ago but I gave up pretty quickly.

There just aren't any words for the things we need words for if Lojban
is to be used autonomously.


It's doubtful whether lujvo can be the answer. Lujvo have many problems
(apart from the issue that nobody can agree on which lujvo to pick for a
given concept).

Lujvo that try to be exact by using a lot of components (rafsi)
naturally get very long, for two reasons: 1) obviously more components
means more length, but 2) many gismu have no short rafsi, which means
you get punished for using them as components.

Recently, more and more people have started to make zi'evla instead of
lujvo. There are different reasons for this. Length of the words is one,
and recognizability is another.

It's no coincidence there are more and more words like {sorpeka} (which
is short for {sorprekarce} (which is just "bus" (1 syllable!) in
English), {nonseka}, {pincivi}, {gapnifa}, etc. Normal lujvo are often
too long.

Try to make lujvo for all the things you wrote down on your list as you
went a day speaking nothing but Lojban to yourself. Or make lujvo for
all the parts of a computer, or of an engine. It's hard, sometimes it
feels impossible and the results are often not satisfying.

At that point I would stop and wonder if lujvo are really such a great
thing. Do they help us move closer to a speakable and spoken language?

- lujvo become very long once you try to inject nuance or specificity
- Nobody knows all the rafsi (and IMO nobody should have to know such a
list of irregular affixes in a language like Lojban), and the rafsi
cannot be derived algorithmically from their gismu or cmavo
- lujvo are made up of parts that describe a given concept, but
sometimes there aren't any good parts (i.e. gismu) to pick in the first
place (see {banli} above)

(and rafsi are bad design in a language that wants to be simple and have
no exceptions)

Do we really need rafsi? What about lujvo?

I say: Merge zi'evla with lujvo by getting rid of the concept of formal
rafsi assignments, thereby achieving so much more freedom on the word
creation level, which is Lojban's biggest construction site, and it
needs every bit of resources it can get.

(you can read more about this idea here:
http://selpahi.weebly.com/lojban/how-to-get-rid-of-formal-rafsi )


Anyway, what to do?

The needed words (thousands of words) must be defined, but it hasn't
been done in the last 25 years. A lot of the time, people are unable to
agree on a lujvo that they deem good enough as a translation of a given
concept. Long discussions ensue about single words, and often result in
no addition of a new word into the dictionary.
Even if they did result in new words, we cannot take that long for each
word or it's going to take another 25 years for all the essential words
to exist, by which time the language might already have died and
replaced by a language with a better dictionary.

If the goal is to have true fluent speakers of Lojban, then this
lexicon-shaped hole in the language must be filled as soon as possible.

For some people this is the only thing keeping them from reaching
fluency, and for the language as a whole it is the one thing that makes
it impossible for anyone to be fluent. Having fluent speakers would be
the best way to show that Lojban is both human-usable and human-used, it
would be better advertising than anything else, yet it's the most
lacking part of the language (for people who want to use the language
and not only talk about it in English).

Unfortunately, only being aware of the problem is not enough, because
the solutions aren't obvious. People will not suddenly stop arguing
about which lujvo is the best (will they?). People will not suddenly
accept without double and triple checking (and then re-inventing) the
lujvo someone else came up with and put in the dictionary (will they?).

If you know a good solution, good, because I don't.

Though what I find interesting (or bizarre?) is that we have the ability
to define all the words we need using random word shapes and giving them
the desired meanings very easily, but we don't. Not that I propose using
randomly generated words, but it's interesting to note that we
apparently prefer a state in which we can't say things to one where we
can but with non-perfect words.

We'd rather have no words than have bad words.

And thus we remain silent.

mi'e la selpa'i mu'o

Gleki Arxokuna

unread,
Oct 26, 2014, 2:13:19 PM10/26/14
to loj...@googlegroups.com
ta'oru'e HSK lists have a word for "traffic jam" although sei mi stace no word for "traffic". If anybody is learning Chinese and has HSK lists with usage examples WITH translations to English it might be a help to Lojbanistan.

By usage examples I mean something like this:




--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to loj...@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.

Robin Lee Powell

unread,
Oct 26, 2014, 2:33:24 PM10/26/14
to loj...@googlegroups.com
On Sun, Oct 26, 2014 at 06:51:45PM +0100, selpa'i wrote:
> Simply put, there aren't enough words, and the few words that
> exist do not enable us to make nuanced descriptions of the world.

I would say it's a mixture of lack of words and lack of phrasal
idiom.

Frankly, though, I consider the self-similarity of existing words,
and the vast array of accidental tongue twisters, to be more serious
problems than the lack of vocab. The language is *physically* hard
to speak, to an extent that is distinctly unpleasent.

--
http://intelligence.org/ : Our last, best hope for a fantastic future.
.i ko na cpedu lo nu stidi vau loi jbopre .i dafsku lu na go'i li'u .e
lu go'i li'u .i ji'a go'i lu na'e go'i li'u .e lu go'i na'i li'u .e
lu no'e go'i li'u .e lu to'e go'i li'u .e lu lo mamta be do cu sofybakni li'u

Gleki Arxokuna

unread,
Oct 26, 2014, 2:37:22 PM10/26/14
to loj...@googlegroups.com
2014-10-26 21:33 GMT+03:00 Robin Lee Powell <rlpo...@digitalkingdom.org>:
On Sun, Oct 26, 2014 at 06:51:45PM +0100, selpa'i wrote:
> Simply put, there aren't enough words, and the few words that
> exist do not enable us to make nuanced descriptions of the world.

I would say it's a mixture of lack of words and lack of phrasal
idiom.

Frankly, though, I consider the self-similarity of existing words,
and the vast array of accidental tongue twisters, to be more serious
problems than the lack of vocab.  The language is *physically* hard
to speak, to an extent that is distinctly unpleasent.

And what is your personal solution ?


--
http://intelligence.org/ :  Our last, best hope for a fantastic future.
.i ko na cpedu lo nu stidi vau loi jbopre .i dafsku lu na go'i li'u .e
lu go'i li'u .i ji'a go'i lu na'e go'i li'u .e lu go'i na'i li'u .e
lu no'e go'i li'u .e lu to'e go'i li'u .e lu lo mamta be do cu sofybakni li'u
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.

John E Clifford

unread,
Oct 26, 2014, 5:01:07 PM10/26/14
to loj...@googlegroups.com
An interesting (though slightly absurd) proposition.  Bringing toki pona in just stresses the absurd side.  Lojban has ten times the vocabulary of tp, just counting gismu straight, forty times taking in the permutation of places, and Lord knows how many times once cmavo and constructed forms come in.  Yet tp is spoken (at least written) much more frequently per capita than Lojban and usually within the first week of exposure.  Factors in its favor are clearly the small vocabulary, mostly immediate usable for common things, and a small, easy grammar.  The whole vocabulary can be absorbed more or less at one go but is, in fact, fed in small doses (a dozen words a lesson, more or less).  The corresponding claim holds for the grammar.  By contrast, Lojban vocabulary is typically dumped at once, all 1500 gismu, with no significant lesson separation (largely because there are no really significant lesson books at all) and the same applies to the grammar.  It is pretty easy to check that a tp utterance is grammatical and that what you meant to say is at least one of the things you did say.  With Lojban, neither of these is simple and many are open to debate even after a formal parser has done its damnedest. Indeed, this uncertainty was, in my experience, the main reason for not using Lojban; not only was it difficult to speak with confidence checking on the fly, but even after due diligence one could not be confident one had said what one intended.

All that being said, there are gaps in the vocabularies of both languages and, while Lojban has means to fill those gaps eventually, tp does not officially (but some word creep is already in evidence).  And both are unZipfy and getting worse.  Thirty some years ago, the creation of rafsi (not then so called) was considered a great advance for Loglan, in that it reqularized compound words (which were previous thrown together any which way and usually ending up looking like gismu, e.g. -- one of JCB's own -- {bedgo} from iirc {betsu gotso}) and made interpretation much more nearly possible (up to the usual vagueness of tanru).  But, of course, many very common notions in areas Loglanists (and now Lobanists) were likely to talk about involved long compounds.  Nothing has been done about this to speak of and what little has been suggested (mainly various sort of acronymization, whether of the strict DARE sort or the freer Gestapo /Ogpu /kenken) are organized to preserve the transparency that the earlier Great Morphological Revolution introduced, but rather are returns to the status quo ante,  Short of the LoCCan3 solution (scrap the structure of Lojban, take the statistics, and build from scratch -- dropping the common languages bases in favor of neatly distributed forms and a list of common concepts discovered in actual usage), some transparent system of abbreviations seems the best hope.

Btw, the absence of idioms is an effect, not a cause of lack of chat (or, at best, arises with it): it is hard to have a fac,on de pareler without a lot of parler.

selpa'i

unread,
Oct 27, 2014, 5:44:01 AM10/27/14
to loj...@googlegroups.com
la .camgusmis. cu cusku di'e
> On Sun, Oct 26, 2014 at 06:51:45PM +0100, selpa'i wrote:
>> Simply put, there aren't enough words, and the few words that
>> exist do not enable us to make nuanced descriptions of the world.
>
> I would say it's a mixture of lack of words and lack of phrasal
> idiom.
>
> Frankly, though, I consider the self-similarity of existing words,
> and the vast array of accidental tongue twisters, to be more serious
> problems than the lack of vocab. The language is *physically* hard
> to speak, to an extent that is distinctly unpleasent.

I have certainly come across phrases that are difficult to pronounce
quickly, though it has gotten much easier for me (as a rapper I'm forced
to learn to say things quickly).

Logically speaking, however, I would say that not having any words is
infinitely worse than having words that are hard to pronounce or that
cause tongue twisters now and then, since in one case you can say what
you mean (even if you may have to slow down in places to keep your
tongue from twisting) and in the other you can't no matter how hard your
tongue tries.

Pierre Abbat

unread,
Oct 27, 2014, 10:40:26 AM10/27/14
to loj...@googlegroups.com
On Sunday, October 26, 2014 18:51:45 selpa'i wrote:
> Simply put, there aren't enough words, and the few words that exist do
> not enable us to make nuanced descriptions of the world.
>
> I see the two main aspects of the problem as being:
>
> 1) The existing words (specifically the gismu) are too vague by
> themselves (and yes, I know that this is by design)
> 2) Too many essential concepts are missing completely

li'o

> 2. "Not enough words"
>
> What do the words traffic, traffic lights, traffic jam have in common?
> They are all examples of everyday words that are completely missing from
> the Lojban lexicon. I could name hundreds more. And it makes speaking
> the language beyond a certain level impossible.

"Traffic" is pretty clearly "ma'efle" in that sense. (Drug traffic is something
else.)

> If the goal is to have true fluent speakers of Lojban, then this
> lexicon-shaped hole in the language must be filled as soon as possible.

I agree.

Now let's consider some other languages that have undergone an expansion or
modernization of vocabulary in recent centuries: Hungarian, Turkish, Hebrew,
and Navajo. Hungarian and Turkish are both agglutinative; long words usually
consist of a single root with several affixes (like English
"antidisestablishmentarianism", which consists of the root "(e)sta" and a
bunch of affixes) or are compounds of such words. It is not difficult to map SAE
vocabulary onto a word formation like that.

Hebrew words consist of a root (usually three consonants, some have two or
four), a pattern of vowels and possibly inserted consonants called a binyan
(similar to grammatical voice but there are at least seven), and a few affixes.
(The Arabic-derived vocabulary, much of which was discarded in the Turkish
language reform, is similar.) Much of the vocabulary for which new Hebrew
words were coined doesn't fit this form and was borrowed.

Navajo is the most different from English, of these languages. I don't know
much about Navajo morphology, but I do know it belongs to an ancient language
family, Dene-Yeniseian, which originated among hunter-gatherers in Siberia and
spread across the Bering land bridge. Until the Spanish arrived, the Navajos
had no domestic animals. If you look through the Navajo entries in Wiktionary,
you'll see that they have very different ways of naming things.

Lojban vocabulary formation is different from SAE in (at least) two ways: it
has a different idea of what concepts are verbs and what are nouns (there are
no adjectives), and long words typically contain almost as many roots as
syllables. This results in different ways of dividing up semantic space (one
word means both "crowbar" and "nutpick", and if you're looking for an exact
equivalent of "august", forget it).

As to "traffic light", I suggest "jezystisni". That doesn't have to be a light,
and what it stops doesn't have to be traffic.

What are some more essential concepts that are missing?

Pierre
--
Jews use a lunisolar calendar; Muslims use a solely lunar calendar.

Gleki Arxokuna

unread,
Oct 27, 2014, 10:43:22 AM10/27/14
to loj...@googlegroups.com
I think the problem is not in the our ability to solve such questions but in that people have to think every time they face such missing concept.
Instead a fuller dictionary must preexist to stop diverting people from Lojban.
Definitely thousands of words missing until we can stop.
Solving them on demand is not an option.
How fast will such single solutions enter jvs is another problem.


Pierre
--
Jews use a lunisolar calendar; Muslims use a solely lunar calendar.

Oleksii Melnyk

unread,
Oct 28, 2014, 12:47:25 PM10/28/14
to loj...@googlegroups.com
> if you want to speak ... in depth and with nuance and subtlety

We can just use the words for that nuances/subtleties, leaving the rest for the context. For example, in the context of the road, traffic is le flecu or loi klama, it's lights are le fendi(jitro/sisti/...), jam is le jmaji (± some SE). That way we can keep it short, easy and manageable.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.

To post to this group, send email to loj...@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.



--
mi'e lex mu'o

MorphemeAddict

unread,
Oct 28, 2014, 1:25:30 PM10/28/14
to loj...@googlegroups.com
I have always assumed that increased fluency will help a speaker both create and decipher new words on the fly, especially lujvo. Is this not the case? 

stevo

selpa'i

unread,
Oct 28, 2014, 3:37:52 PM10/28/14
to loj...@googlegroups.com
la'o me. MorphemeAddict .me cusku di'e
> I have always assumed that increased fluency will help a speaker both
> create and decipher new words on the fly, especially lujvo. Is this not
> the case?

I would say it is very much the case (I know hundreds of rafsi and can
make up lujvo on the fly if I want to), but there are way too many words
missing for that to be practical. I really encourage anyone who thinks
there is no vocabulary problem to try the experiment I outlined in the
original post.

And it's not just a technical problem of picking the right rafsi, but
also a problem between different speakers of agreeing that a particular
choice is okay. It usually doesn't go that smoothly, and certainly too
slowly for that to be the optimal way to fix this hole.

There must be a faster way; a bootstrapping phase that adds a few more
thousand words. Only then is the on-the-fly method really practical,
because then the new needed coinings per day, month or year are going to
be much much fewer. Think about how many new words are added to the
English language dictionary each year; it's not a large number. And how
often do you see an everyday object and can't name it in English? How
many of those times do you *invent* a word for it?

v4hn

unread,
Oct 28, 2014, 8:06:16 PM10/28/14
to loj...@googlegroups.com
On Tue, Oct 28, 2014 at 08:37:51PM +0100, selpa'i wrote:
> And how often do you see an everyday object and can't name it in English?
> How many of those times do you *invent* a word for it?

Quite often. Usually I choose "thing" for it.


mu'o

Pierre Abbat

unread,
Nov 1, 2014, 7:21:05 AM11/1/14
to loj...@googlegroups.com
On Monday, October 27, 2014 17:43:20 Gleki Arxokuna wrote:
> I think the problem is not in the our ability to solve such questions but
> in that people have to think every time they face such missing concept.
> Instead a fuller dictionary must preexist to stop diverting people from
> Lojban.
> Definitely thousands of words missing until we can stop.
> Solving them on demand is not an option.
> How fast will such single solutions enter jvs is another problem.

A couple of words that need translations:
driveway
sinus (in the body, such as maxillary sinus)
Anyone have some more words you don't know how to say in Lojban?

Pierre

--
ve ka'a ro klaji la .romas. se jmaji

Jorge Llambías

unread,
Nov 1, 2014, 11:09:41 AM11/1/14
to loj...@googlegroups.com
On Sat, Nov 1, 2014 at 8:20 AM, Pierre Abbat <ph...@bezitopo.org> wrote:

A couple of words that need translations:
driveway

vorklaji?
 
sinus (in the body, such as maxillary sinus)

zbixle?

mu'o mi'e xorxes
 

Pierre Abbat

unread,
Nov 2, 2014, 12:11:54 AM11/2/14
to loj...@googlegroups.com
On Saturday, November 01, 2014 12:09:39 Jorge Llambías wrote:
> On Sat, Nov 1, 2014 at 8:20 AM, Pierre Abbat <ph...@bezitopo.org> wrote:
> > A couple of words that need translations:
> > driveway
>
> vorklaji?

Sounds good. Would x2 be the street and x3 the building?

> > sinus (in the body, such as maxillary sinus)
>
> zbixle?

"zbidaski" sounds better to me.

What are some other words that need translations?

Jorge Llambías

unread,
Nov 2, 2014, 8:26:05 AM11/2/14
to loj...@googlegroups.com
On Sun, Nov 2, 2014 at 1:11 AM, Pierre Abbat <ph...@bezitopo.org> wrote:
On Saturday, November 01, 2014 12:09:39 Jorge Llambías wrote:
> vorklaji?

Sounds good. Would x2 be the street and x3 the building?

I think it would be:

x1=klaji1=vorme1 (the driveway)
x2=klaji2=vorme4 (the property/real estate)
x3=klaji3=vorme2 (the building)
x4=vorme3 (the street)

or "lajvro" for reordered places:

x1=vorme1=klaji1 (the driveway)
x2=vorme2=klaji3 (the building)
x3=vorme3 (the street)
x4=vorme4=klaji2 (the property/real estate)

TR NS

unread,
Nov 2, 2014, 1:23:30 PM11/2/14
to loj...@googlegroups.com


On Sunday, October 26, 2014 1:51:26 PM UTC-4, selpa'i wrote:
Simply put, there aren't enough words, and the few words that exist do
not enable us to make nuanced descriptions of the world.


One the big advantages of natural language is connotation --they can use the same (short) word for different meanings.  e.g. a `bus` for transporting people vs a `bus` for transporting data in computer hardware. There is a related concept between the two, but they are very different things. Lojban doesn't seem to work well for creating a single word for such disparate things. (maybe it could but people tend not to?)

It also kind of shame that there isn't a way to speak in abbreviated forms, e.g. if their were a two part word "vehicle-bus" and another "computer-bus", we could just use "bus" once we set forth the context as "vehicle" or "computer".

Gleki Arxokuna

unread,
Nov 2, 2014, 1:28:43 PM11/2/14
to loj...@googlegroups.com
2014-11-02 21:23 GMT+03:00 TR NS <tran...@gmail.com>:


On Sunday, October 26, 2014 1:51:26 PM UTC-4, selpa'i wrote:
Simply put, there aren't enough words, and the few words that exist do
not enable us to make nuanced descriptions of the world.


One the big advantages of natural language is connotation --they can use the same (short) word for different meanings.  e.g. a `bus` for transporting people vs a `bus` for transporting data in computer hardware. There is a related concept between the two, but they are very different things. Lojban doesn't seem to work well for creating a single word for such disparate things. (maybe it could but people tend not to?)
yes, it's people that dont tend to or just dont realize that the two concepts forma continuum, not a polysemy. so you are free to create as many such words as you need/want.
 

It also kind of shame that there isn't a way to speak in abbreviated forms, e.g. if their were a two part word "vehicle-bus" and another "computer-bus", we could just use "bus" once we set forth the context as "vehicle" or "computer".

well, here we should think of the place structure of "bus" to fill it once and then refer to what we defined by just using unfilled brivla. 

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.

Gleki Arxokuna

unread,
Nov 2, 2014, 1:32:44 PM11/2/14
to loj...@googlegroups.com
2014-11-02 7:11 GMT+03:00 Pierre Abbat <ph...@bezitopo.org>:
On Saturday, November 01, 2014 12:09:39 Jorge Llambías wrote:
> On Sat, Nov 1, 2014 at 8:20 AM, Pierre Abbat <ph...@bezitopo.org> wrote:
> > A couple of words that need translations:
> > driveway
>
> vorklaji?

Sounds good. Would x2 be the street and x3 the building?

> > sinus (in the body, such as maxillary sinus)
>
> zbixle?

"zbidaski" sounds better to me.

What are some other words that need translations?

today i wanted verbs for "to roast", "bake", "stew", "grill", "steam", "fry".
WordNet mentions these hyponyms of "to cook":

bake, brown, coddle, souse, microwave, micro-cook, zap, nuke, blanch, parboil, overcook, fricassee, stew, roast, braise, fry, grill, steam, pressure-cook, poach

Not sure I understand all of them.



Pierre

--
ve ka'a ro klaji la .romas. se jmaji

MorphemeAddict

unread,
Nov 2, 2014, 2:33:01 PM11/2/14
to loj...@googlegroups.com
I don't understand all of them either, although I'm a native speaker. My wife would have understood all of them. I don't cook. I do recognize all of them as cooking words, although micro-cook is new to me, and coddle is new in this context. 

stevo

Pierre Abbat

unread,
Nov 3, 2014, 5:31:13 AM11/3/14
to loj...@googlegroups.com
On Sunday, November 02, 2014 21:32:41 Gleki Arxokuna wrote:
> today i wanted verbs for "to roast", "bake", "stew", "grill", "steam",
> "fry".
> WordNet mentions these hyponyms of "to cook":
>
> bake, brown, coddle, souse, microwave, micro-cook, zap, nuke, blanch,
> parboil, overcook, fricassee, stew, roast, braise, fry, grill, steam,
> pressure-cook, poach
>
> Not sure I understand all of them.

fricassee: frikasa (made up for the song)
fry: rasyjukpa (ditto)
overcook: dusyjukpa

To roast, broil, or bake is "tokyjukpa", as long as it's done in an oven.
Roasting on a spit is a different word, "carjukpa".

Grill is "gredile" but that's the thing you grill on, not the act of grilling.

"Micro-cook", "microwave" (as a verb), "zap", and "nuke" are all synonyms in
this context.

Pierre
--
I believe in Yellow when I'm in Sweden and in Black when I'm in Wales.

Gleki Arxokuna

unread,
Nov 3, 2014, 6:38:09 AM11/3/14
to loj...@googlegroups.com
gredilyjukpa then?

Jorge Llambías

unread,
Nov 3, 2014, 6:42:12 AM11/3/14
to loj...@googlegroups.com


On Mon, Nov 3, 2014 at 8:38 AM, Gleki Arxokuna <gleki.is...@gmail.com> wrote:
gredilyjukpa then?

That's pagre zei dilnu zei jukpa.

 gredile zei jukpa is gredile'yjukpa

Jacob Errington

unread,
Nov 3, 2014, 10:11:49 AM11/3/14
to loj...@googlegroups.com

I had thought of a system to achieve this with type-3 and contextual type-4 fu'ivla. The idea is that each type-3 fu'ivla would suggest a (shorter) contextual type-4 fu'ivla to use in its place.

e.g. zuktriorlo = x1 does x2 (ka) because YOLO |>>> suggests {iorlo} as a type-4 to use in its place when the context is clear.

Typically, the suggested type-4 can be constructed by pulling off the type-3 head, and injecting a consonant if necessary to make a cluster.

.i mi'e la tsani mu'o

TR NS

unread,
Nov 8, 2014, 8:16:03 AM11/8/14
to loj...@googlegroups.com


On Sunday, October 26, 2014 2:33:24 PM UTC-4, Robin Powell wrote:
On Sun, Oct 26, 2014 at 06:51:45PM +0100, selpa'i wrote:
> Simply put, there aren't enough words, and the few words that
> exist do not enable us to make nuanced descriptions of the world.

I would say it's a mixture of lack of words and lack of phrasal
idiom.

Frankly, though, I consider the self-similarity of existing words,
and the vast array of accidental tongue twisters, to be more serious
problems than the lack of vocab.  The language is *physically* hard
to speak, to an extent that is distinctly unpleasent.


I was wondering if you could elaborate on that. And what do you think can be done about it?
 

karis

unread,
Nov 8, 2014, 11:34:43 PM11/8/14
to loj...@googlegroups.com
I'm sorry everyone, but I disagree that Lojban needs us to every increase the vocabulary. One of the highlights of the language its the small vocabulary and the fact that you can express anything as precisely or imprecisely as you want. Nothing says that every idea has to be a single short word. This certainly isn't the case in English. Why must Lojban have a single short word for traffic light when English doesn't? If it was so hard to express emotions and many other things in Lojban than how could there be beautiful poetry other literature written in or translated into it using just the basic vocabulary and rafsi? Look up the writings of Athlestan and various other materials from the 1990s. I even translated a poem I wrote in High School into it without anything but gismu. Not everything needs to be in the shortest form. If it did we would all use, "sad, mad,and glad" rather than "excited, overjoyed, pleased, frustrated, angry, furious, tearful, miserable, or down in the dumps."

.karis. (Karen Stein)

Gleki Arxokuna

unread,
Nov 9, 2014, 3:15:26 AM11/9/14
to loj...@googlegroups.com
I think there can be no single point of view here. If a person needs a single word then 'ey adds it. Since many people want immediate solutions then we must provide them either by placing them to tatoeba.org or to jbovlaste.

tatoeba.org is better since you can add any sentence there even if you dont need a single brivla for "traffic lights".


2014-11-09 7:34 GMT+03:00 karis <comca...@gmail.com>:
I'm sorry everyone, but I disagree that Lojban needs us to every increase the vocabulary. One of the highlights of the language its the small vocabulary and the fact that you can express anything as precisely or imprecisely as you want. Nothing says that every idea has to be a single short word. This certainly isn't the case in English. Why must Lojban have a single short word for traffic light when English doesn't? If it was so hard to express emotions and many other things in Lojban than how could there be beautiful poetry other literature written in or translated into it using just the basic vocabulary and rafsi? Look up the writings of Athlestan and various other materials from the 1990s. I even translated a poem I wrote in High School into it without anything but gismu. Not everything needs to be in the shortest form. If it did we would all use, "sad, mad,and glad" rather than "excited, overjoyed, pleased, frustrated, angry, furious, tearful, miserable, or down in the dumps."

.karis. (Karen Stein)

And Rosta

unread,
Nov 9, 2014, 6:51:41 AM11/9/14
to loj...@googlegroups.com


On 9 Nov 2014 04:34, "karis" <comca...@gmail.com> wrote:
>  Why must Lojban have a single short word for traffic light when English doesn't?

I don't think anybody is saying the the Lojban equivalent of the English word 'traffic light' must be shorter than 'traffic light'. They're saying that just as English does, Lojban needs a word that means 'traffic light'.

--And.

MorphemeAddict

unread,
Nov 9, 2014, 12:54:55 PM11/9/14
to loj...@googlegroups.com
Or at least a specific phrase that means 'traffic light', just as English uses not a single word, but a phrase. 

stevo

--

TR NS

unread,
Nov 9, 2014, 4:46:20 PM11/9/14
to loj...@googlegroups.com


On Sunday, November 9, 2014 12:54:55 PM UTC-5, stevo wrote:
Or at least a specific phrase that means 'traffic light', just as English uses not a single word, but a phrase. 

 
Maybe that's the kind of construct Lojban needs --a way to use two (or more) words together that have a specific (gestalt?) meaning rather then a metaphorical one. Yet I don't see how the grammar would allow it.

Andrew Browne

unread,
Nov 9, 2014, 6:13:43 PM11/9/14
to loj...@googlegroups.com
The phrase, "traffic light" could mean lights that illuminate the road and traffic (i.e. "street lights"), or all of the light pollution from street lights and lights on cars.
This phrase communicates the concept not because it describes the concept so well, but because the phrase "traffic light" is specifically associated with that thing.

I think an association between a specific phrase and a specific thing comes with more usage, however in many cases lojban can be more precisely descriptive instead.

lo ma'efle jitro sinxa

imo this is a better semantic description of a traffic light than "traffic light" is.
This lojban phrase is missing is a strong association between this specific phrase and the specific thing. Is that association still required with a more detailed description?

Gleki Arxokuna

unread,
Nov 10, 2014, 1:35:56 AM11/10/14
to loj...@googlegroups.com
"traffic lights" is one single concept in English, not just a combination of two words. Even in dictionaries it is usually shown in a separate node.

Pierre Abbat

unread,
Nov 11, 2014, 7:56:43 AM11/11/14
to loj...@googlegroups.com
On Sunday, November 09, 2014 15:13:43 Andrew Browne wrote:
> lo ma'efle jitro sinxa
>
> imo this is a better semantic description of a traffic light than "traffic
> light" is.
> This lojban phrase is missing is a strong association between this specific
> phrase and the specific thing. Is that association still required with a
> more detailed description?

"ma'efle jitro sinxa" and "jezystisni" can also describe the sign with "STOP"
on one side and "SLOW" on the other, which is flipped by who for historical
reasons is called a flagman. The latter can also describe a pedestrian signal.
The former can also describe a "YIELD" sign. And that's fine. Lojban doesn't
divide the world the way English does, nor English the way Hebrew does, or so
on.

MorphemeAddict

unread,
Nov 11, 2014, 9:18:27 AM11/11/14
to loj...@googlegroups.com
And the synonym "stop lights" is ambiguous, too, meaning either 'traffic lights' or 'brake lights' (maybe). 

stevo

And Rosta

unread,
Nov 11, 2014, 11:23:58 AM11/11/14
to loj...@googlegroups.com
> On Sunday, November 9, 2014 12:54:55 PM UTC-5, stevo wrote:
>
> Or at least a specific phrase that means 'traffic light', just as
> English uses not a single word, but a phrase.

Morphologically, _traffic light_ is probably a single word -- a compound stem --, since the main accent is on the 'traff' (TRAFFiclight), whereas in a phrase the accent is normally on the final word in the phrase (*traffic LIGHT).

_Traffic light_ is also a single word qua listeme. Listemes are gestalts, with some properties that aren't derived from their parts (if they have any parts).

Andrew Browne, On 09/11/2014 23:13:
> The phrase, "traffic light" could mean lights that illuminate the
> road and traffic (i.e. "street lights"), or all of the light
> pollution from street lights and lights on cars. This phrase
> communicates the concept not because it describes the concept so
> well, but because the phrase "traffic light" is specifically
> associated with that thing. I think an association between a specific
> phrase and a specific thing comes with more usage,

In natlangs, listemes develop diachronically through usage in the way you describe. In Lojban, listemes must (i) be single brivla and (ii) be formally listed in reference materials that define the language.

TR NS, On 09/11/2014 21:46:
> Maybe that's the kind of construct Lojban needs --a way to use two
> (or more) words together that have a specific (gestalt?) meaning
> rather then a metaphorical one. Yet I don't see how the grammar would
> allow it.

Something more than {zei}?

--And.

htffan...@gmail.com

unread,
Dec 6, 2014, 1:17:43 PM12/6/14
to loj...@googlegroups.com
1. It's designed to combine the words to from relations. Not a lot of words are needed because each added potential concept squares the number of existing ones. Any concept can be used with any other concept. All of your examples do exist in lojban, and are assumed to be from the perspective of the person saying it. Even so, they are terrible examples because they have no concrete definition even in English.
2. Those can be used with concepts. Traffic is such an abstract word it is actually a horrible example. None of those terms have anything in common. Traffic has multiple meanings and using it is like using Google Translate to speak professionally. You will never get there. The whole idea is to remove abstraction and you are bashing it for doing exactly that.
3. Hippopotomonstrososequapedaliaphobia. An actual word in English. What was that about long words again?
4. You memorized a lot more when you learned English. What was that about everyone being unable to learn new words?
5. Of course they aren't derived. It's not based on a single language. That would limit who speaks it.
6. You didn't find the parts because you were too strict in looking and didn't bother to look.

MorphemeAddict

unread,
Dec 7, 2014, 4:52:49 PM12/7/14
to loj...@googlegroups.com
The example offered for a long word in English is ridiculous. I doubt anyone uses it, or that it's more than a curiosity, and it's misspelled. Should be "hippopotomonstrosesquipedaliophobia
​".

stevo

--
Reply all
Reply to author
Forward
0 new messages