A new guide that needs guidance

10 views
Skip to first unread message

Muhammad Nael

unread,
Nov 1, 2011, 11:13:32 AM11/1/11
to loj...@googlegroups.com
First, greeting to all, those new, old and anything in between,
For those of you who have known me for some time -less than one-, you remember that I first came here for a school project... It didn't go well and that pretty much it. Move along.

You may know that we have a 4-day feast starting next week... During that time, I'll devote a rather big chunk for Lojban. I've already read the lessons for beginners and the reference grammar. While I'm reasonably good at the grammar I know much less vocab. I can't get into any of the technical discussions here but I hope that by the week after the next, I would be able to.

I've, several times, voiced my 'opinion' that Lojban needs more refining. The problem is that I really can't help but feel that something needs improvement, I have to witness it to believe it. Still, following the right way of thinking, I'm giving more time to the existing Lojban before I could start any new conlanging project; make no mistake it's coming.

So, I obviously need help here. I'll keep this thread updated with any technical difficulties that might arise. You help me solve it and I'll present you, hopefully, at the end of the next week with a new guide to learning Lojban that is more suitable for young audience, let's not forget that I'm one of them!

Thank you all for enduring this monolog!

MN

Muhammad Nael

unread,
Nov 3, 2011, 3:34:13 AM11/3/11
to loj...@googlegroups.com
Also, when I post a question, say how many times (approximately) you have been asked about it...
1. The names of lerfu in Lojban are 'very' ambiguous... The only way to differentiate between 'k' and 'q' is to use something normally used for a vowel...
EX: In languages that use the Arabic alphabet;
/g/ written as ج is called /gi:m/
/Ʒ/ written as ج is called /Ʒi:m/
/Ʒ/ written as چ is called /dƷeh/
/Ʒ/ written as ژ is called /Ʒeh/
/g/ written as گ is called /ga:f/
/tʃ/ written as چ is called /tʃeh/

Alex Rozenshteyn

unread,
Nov 3, 2011, 9:24:29 AM11/3/11
to loj...@googlegroups.com
I'm not sure what you mean by the lerfu names being ambiguous.

{ky} and {ky bu} sound different to me, and {bu} is not used only (or primarily) for vowels (mu'a denpa bu)

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban/-/HP-3teIDF94J.

To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.



--
          Alex R

M. Nael

unread,
Nov 3, 2011, 11:12:30 AM11/3/11
to loj...@googlegroups.com
For me till now, 'bu' is a rafsi (right?) that means a 'letteral'... What I meant is, why not have a sort of cmene to mean specific lerfu? especially those of other languages? I can't imagine using la for that...

2. Could Lojban be used as a primarily SOV language? Somewhere between chapter 2 and 3 of Lojban for beginners, the writers says that you can 'promote' several sumti to preceed the selbri; does this mean it could act as a complete SOV lang.?

Jonathan Jones

unread,
Nov 3, 2011, 12:12:25 PM11/3/11
to loj...@googlegroups.com
On Thu, Nov 3, 2011 at 9:12 AM, M. Nael <muhamm...@gmail.com> wrote:
For me till now, 'bu' is a rafsi (right?) that means a 'letteral'...

No. It's a cmavo. rafsi are the combining forms of words used to make lujvo.
 
What I meant is, why not have a sort of cmene to mean specific lerfu?

We do. The name of Q is "kybu".
 
especially those of other languages? I can't imagine using la for that...

2. Could Lojban be used as a primarily SOV language? Somewhere between chapter 2 and 3 of Lojban for beginners, the writers says that you can 'promote' several sumti to preceed the selbri; does this mean it could act as a complete SOV lang.?

Yes, it is possible to say/write Lojban bridi in the such a way as to act like it's an SOV language. Or an OSV one, or a VSO one, or any other word order. But Lojban doesn't have /verbs/, so it would only be /appearing/ to be such-it wouldn't actually be.

--
mu'o mi'e .aionys.

.i.e'ucai ko cmima lo pilno be denpa bu .i doi.luk. mi patfu do zo'o
(Come to the Dot Side! Luke, I am your father. :D )

M. Nael

unread,
Nov 3, 2011, 3:07:25 PM11/3/11
to loj...@googlegroups.com
We do. The name of Q is "kybu".
I meant more distinctive names, as per the example I mentioned for Arabic ج
 
Yes, it is possible to say/write Lojban bridi in the such a way as to act like it's an SOV language. Or an OSV one, or a VSO one, or any other word order. But Lojban doesn't have /verbs/, so it would only be /appearing/ to be such-it wouldn't actually be.
I know Lojban's word is virtually unconstrained (or really, haven't reached that level yet). But I meant if it could act as a real SOV language, like Latin; without place tags or swithcers... eg. {zo'e zo'e zo'e rupnu} instead of {zo'e rupnu zo'e zo'e} or maybe {zo'e zo'e zo'e cu rupnu}?

Jonathan Jones

unread,
Nov 3, 2011, 3:25:17 PM11/3/11
to loj...@googlegroups.com
2011/11/3 M. Nael <muhamm...@gmail.com>

We do. The name of Q is "kybu".
I meant more distinctive names, as per the example I mentioned for Arabic ج

I don't know what you mean by distinctive. Each name is associated with exactly one thing. .uibu, for example, is the name of ":)". You can't use .uibu to refer to anything else.
 
Yes, it is possible to say/write Lojban bridi in the such a way as to act like it's an SOV language. Or an OSV one, or a VSO one, or any other word order. But Lojban doesn't have /verbs/, so it would only be /appearing/ to be such-it wouldn't actually be.
I know Lojban's word is virtually unconstrained (or really, haven't reached that level yet). But I meant if it could act as a real SOV language, like Latin; without place tags or swithcers... eg. {zo'e zo'e zo'e rupnu} instead of {zo'e rupnu zo'e zo'e} or maybe {zo'e zo'e zo'e cu rupnu}?

The sumti in a bridi, other than the first, can appear either before or after the selbri.

{lo mlatu cu kalte lo cipni lonu kelci}, {lo mlatu lo cipni cu kalte lonu kelci}, and {lo mlatu lo cipni lonu kelci cu kalte} are all the same bridi. The convention is for only the first to be before the selbri, likely because the sooner the listener hears the selbri, the sooner he knows what the sumti relations are.

{lo mlatu lo cipni lonu kelci cu klama} is a completely different bridi, but no one listening would know that until the very last word is said.

M. Nael

unread,
Nov 3, 2011, 3:57:09 PM11/3/11
to loj...@googlegroups.com
The sumti in a bridi, other than the first, can appear either before or after the selbri.
{lo mlatu cu kalte lo cipni lonu kelci}, {lo mlatu lo cipni cu kalte lonu kelci}, and {lo mlatu lo cipni lonu kelci cu kalte} are all the same bridi. The convention is for only the first to be before the selbri, likely because the sooner the listener hears the selbri, the sooner he knows what the sumti relations are.
{lo mlatu lo cipni lonu kelci cu klama} is a completely different bridi, but no one listening would know that until the very last word is said.
That settles the SOV part...
 
 I don't know what you mean by distinctive. Each name is associated with exactly one thing. .uibu, for example, is the name of ":)". You can't use .uibu to refer to anything else.
"uibu" is distinctive enough because 'ui' is distinctive... but I can't distinguish py. and by. Ignoring any technicalities; no aspiration or anything else. In English (Bi) and (Pi) are difficult to tell apart, while in Arabic (Zey=z) as a letter can't be mistaken for (Dhal=th in then).
_OFF

John E Clifford

unread,
Nov 3, 2011, 5:27:19 PM11/3/11
to loj...@googlegroups.com
"ui" is distinctive in Lojban. "pe" and "be" (and "zed" and "theta") are in English.  The fact that your language doesn't have the distinctions involved may make English -- or Lojban -- a bit harder for you but doesn't make them less distinctive in their own languages.  So, {ky} and {kybu} are quite distinct in Lojban, as distinct as {cy} and {ebu}.  Since this is obvious, I fear I am missing your point here.

From: M. Nael <muhamm...@gmail.com>
To: loj...@googlegroups.com
Sent: Thu, November 3, 2011 2:57:09 PM
Subject: Re: [lojban] Re: A new guide that needs guidance
--
You received this message because you are subscribed to the Google Groups "lojban" group.

Pierre Abbat

unread,
Nov 3, 2011, 7:36:58 PM11/3/11
to loj...@googlegroups.com
On Thursday 03 November 2011 12:12:25 Jonathan Jones wrote:
> Yes, it is possible to say/write Lojban bridi in the such a way as to act
> like it's an SOV language. Or an OSV one, or a VSO one, or any other word
> order. But Lojban doesn't have /verbs/, so it would only be /appearing/ to
> be such-it wouldn't actually be.

I say, rather, that it *does* have verbs, but the verb covers what in other
languages would be adjectives or nouns or adverbs. It works equally well as
SVO or SOV, until you make compound predicates or things like that. The other
orders need FA or SE.

Pierre
--
lo ponse be lo mruli po'o cu ga'ezga roda lo ka dinko

M. Nael

unread,
Nov 4, 2011, 4:28:24 AM11/4/11
to loj...@googlegroups.com
-It looks like I went too deep here so I'll round it up...
Why doesn't lojban have something even remotely like NATO's alphabet? Not long names for lerfu but just different enough. I know how pronounce a quite large part of the IPA main chart but don't ask me to do strange vowels... ie. I can say py. and by. But I can't differentiate them. So why not something like (byb. pyp.) for example?

-There are natlangs that use verbs as adjectives, called stative verbs if I recall. Compound predicates, I haven't reached that yet and I'm scanning the lessons bit by bit.

›› 3. I noticed (mi'e) and (mi'afra) which look like a (mi) with things attached... Should this mean that whenever there's an apostrophe for the sound /h/, the syllables before and after the apostrophe should be counted as one word? ie. /h/ could never be at the beginning or the end of a word?

.arpis.

unread,
Nov 4, 2011, 9:07:25 AM11/4/11
to loj...@googlegroups.com
-It looks like I went too deep here so I'll round it up...
Why doesn't lojban have something even remotely like NATO's alphabet?
Not long names for lerfu but just different enough. I know how pronounce a quite large part of the IPA main chart but don't ask me to do strange vowels... ie. I can say py. and by. But I can't differentiate them. So why not something like (byb. pyp.) for example?
They don't differ in the vowels but by consonants; if you can't tell a "b" from a "p", how would adding more of them solve the problem>

-There are natlangs that use verbs as adjectives, called stative verbs if I recall. Compound predicates, I haven't reached that yet and I'm scanning the lessons bit by bit.

›› 3. I noticed (mi'e) and (mi'afra) which look like a (mi) with things attached... Should this mean that whenever there's an apostrophe for the sound /h/, the syllables before and after the apostrophe should be counted as one word? ie. /h/ could never be at the beginning or the end of a word?
Correct; {y'y} is always in the middle of a word, between two vowels.

Personally, I skimmed through L4B quickly, and got most of my lojban knowledge from the CLL, but I know that's not for everyone.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.



--
mu'o mi'e .arpis.

Michael Turniansky

unread,
Nov 8, 2011, 7:55:13 AM11/8/11
to loj...@googlegroups.com


2011/11/3 Jonathan Jones <eye...@gmail.com>
2011/11/3 M. Nael <muhamm...@gmail.com>

We do. The name of Q is "kybu".
I meant more distinctive names, as per the example I mentioned for Arabic ج
I don't know what you mean by distinctive. Each name is associated with exactly one thing. .uibu, for example, is the name of ":)". You can't use .uibu to refer to anything else.
 
  Well, that's not exactly true.  Depends on context.  For example ky bu is q, ga'e ky bu is Q, je'o ky bu is  ק and presumably jo'o ky bu is ق (forgive me if I have that wrong, Muhammad)
 
      --gejyspa
 

Jonathan Jones

unread,
Nov 8, 2011, 1:36:57 PM11/8/11
to loj...@googlegroups.com
Every one of those examples is different, gejyspa, and illustrates what I said perfectly. {je'o ky. bu} is not {ky bu}.

Michael Turniansky

unread,
Nov 8, 2011, 4:36:14 PM11/8/11
to loj...@googlegroups.com
No, they ALL are....the je'o, ga'e, jo'o, etc. simply specify the context of what follows.  The fact that we normally default to a lo'a context is simply convention. I could have had any number of other lerfu, PA, heck, even whole paragraphs after using "je'o" before using the "kybu" and it would meant "ק", and not "q"
                           --gejyspa


 
--
You received this message because you are subscribed to the Google Groups "lojban" group.

Sebastian Fröjd

unread,
Nov 9, 2011, 4:00:22 AM11/9/11
to loj...@googlegroups.com
I understand that most jbopli today are using the Roman alphabet, and that's fine, but not really cultural neutral. So I don't understand why {lo'a} is defined as "shift letterals to Lojban (Roman) alphabet." I think every alphabet should be equally lojbanic, or? If lojban would have a default alphabet I think it should be Tengwar, since that's the most cultural neutral (and also phonetical) alphabet. Anyway, I'm already using tengwar sometimes, so I really think it would be nice to have a BY1 for tengwar also. Maybe {te'i}?
mu'o mi'e jongausib

2011/11/8 Michael Turniansky <mturn...@gmail.com>

Jonathan Jones

unread,
Nov 9, 2011, 5:00:05 AM11/9/11
to loj...@googlegroups.com
I disagree on both counts.

The roman characters are used by a lot of different languages/cultures, for one, and for two, not all writing systems are equally suitable for writing Lojban. I did an orthography for Japanese, and it required a lot of custom rules, as Japanese has no way of writing any ending consonant other than "n", and no way of writing consonant clusters other than doubles, or nC. Chinese is probablly even less suitable. Some are possibly equally suitable- I believe Russian and Arabic scripts work without too much fuss, but not all systems are equally Lojbanic. It's quite possible I'm wrong, but as far as I am aware, Roman script is the only writing system that is multi-linguistic- that is, used by more than one language.

On the second point, Tengwar is not the most culturally neutral. It was invented by an English-speaking writer as the script of a fictional race, used to write said race's fictional language. While I'm sure any LoTR fans would love to write Lojban using it, and I know that some people actually have, it is by no means neutral.

Just as with the "cultural gismu" this can really only be settled by an all or nothing approach. Either we use a system that is used by "all"- or at least the closest thing to "all"- the script which is the most widely used on this planet, or we use a system that is used by "none"- or at least used by no-one outside of Lojban, such larlermorna (and I'm pretty sure I got the name wrong).

Finally, I'm going to quote a wonderfully appropriate cliche- "If it ain't broke, don't fix it." Lojban's default- or maybe I should say official- orthography has been Latin script literally since it's birth. And if we count Loglan, since /before/ it's birth, as Loglan's official script - and probably only allowed one- is also Latin. From a purely functional standpoint, it has absolutely no problems. All Lojbanists are perfectly capable of writing and reading Lojban using it. And for those of us who's first language uses a different writing system, yes, it does mean learning another one- but for many foreign languages, this would be the case /anyway/. I'm learning Japanese right now- I have to learn a whole slew of new characters. If I decided to learn Jewish (Yiddish?), Arabic, Russian, etc., those require me to learn a new system as well. It kind of comes with the territory.

Pierre Abbat

unread,
Nov 9, 2011, 7:38:48 AM11/9/11
to loj...@googlegroups.com
On Wednesday 09 November 2011 05:00:05 Jonathan Jones wrote:
> I disagree on both counts.
>
> The roman characters are used by a lot of different languages/cultures, for
> one, and for two, not all writing systems are equally suitable for writing
> Lojban. I did an orthography for Japanese, and it required a lot of custom
> rules, as Japanese has no way of writing any ending consonant other than
> "n", and no way of writing consonant clusters other than doubles, or nC.
> Chinese is probablly even less suitable. Some are possibly equally
> suitable- I believe Russian and Arabic scripts work without too much fuss,
> but not all systems are equally Lojbanic. It's quite possible I'm wrong,
> but as far as I am aware, Roman script is the only writing system that is
> multi-linguistic- that is, used by more than one language.

Latin and Cyrillic letters are both used by a wide variety of languages, and
both are suitable for Lojban. Other multilanguage scripts are Arabic (used
for some Indo-Iranian languages and formerly for Turkish, though the Latin
alphabet suits Turkish better), Greek (used for Bactrian, Phrygian, and an
Albanian dialect), Hebrew (used for Yiddish and Ladino), kana (used for
Ainu), and Chinese (used for Japanese and sometimes Korean, along with
phonetic characters).

Pierre
--
.i toljundi do .ibabo mi'afra tu'a do
.ibabo damba do .ibabo do jinga
.icu'u la ma'atman.

Reply all
Reply to author
Forward
0 new messages