Lojban Language

A constructed human language (see ConLang), like EsperantoLanguage and KlingonLanguage, with the unique characteristic that it was designed to permit the full range of human language expression, yet be unambiguous and computer parseable.

In fact, here is the grammar in: The successor of LoglanLanguage.

Read all about it at http://www.lojban.org/.

See also CulturallyNeutralLanguage, LojbanPoem, LojbanicJars, RulesOfLojban, HowWouldLojbanEnableAi, MalGlico, LojbanistaniCulture and SomethingUnspecified.

Lojban has its own wiki: http://www.lojban.org/wiki/.

See LojbanPronunciation.



Lojban and Loglan

Lojban is not a dialect of Loglan, any more than Portuguese is a dialect of Spanish, but a RELEX of it. They are mutually unintelligible but related. All the root words have been remade from scratch, and the grammar has been overlaid with a huge amount of new refinements.

For example, Vizka la spat. in LoglanLanguage is something about seeing spot. In LojbanLanguage, it would be viska la spat. meaning zo'e viska la spat. or "(SomethingUnspecified) sees Spot". In Loglan, it means "(Hey you!) See Spot". If a lojban speaker were to try to carry on a conversation with a Loglan speaker, they would not get very far.

Comments from a lojbanist, whom was said i ku'i .a'u troci ("however, it would be interesting to try") :

.i ganai mi troci ga mi steba .i mi na djuno le prenu poi tavla fo la loglan. .ija'e mi na ka'e troci .i mi na djica troci
 xe fanva/translation:
If I try, then I am frustrated. I do not know any people who speak loglan. Therefore* I can't try. I don't want to try (presumably meant to be and I don't want to try?)

(*) Not exactly the same, but close enough for government work.

BTW, since you don't try, and ganai, gi is like a logical if, then, in lojban you have told us nothing. Use va'o instead if you know you don't try.

The above exchange demonstrates a weakness of Lojban, that it seems to demand exact logical definition of ideas, whereas most spoken languages have a little slack to them. In English, we knew what the AllYourBaseAreBelongToUs dialogue meant even though it was hideously malformed, and in most languages, including Esperanto, grammar errors only cause a problem when they create ambiguity or uncertainty. Sentences are not to be debugged like programs, and the above correction in particular sounded like an "expression result never used" compiler warning, when it's reasonable to assume here that the speaker was trying to imply the subjunctive mood. "I don't try, but if I did..."

I partially agree, but partially disagree. If in French you use "savoir" instead of "connaƮtre" you ought to be corrected. To the native English speaker it can be hard to tell the difference. Similarly if you use "oui" when you actually should use "si". Incorrect usage will not usually prevent communication, but it can sometimes create subtle (or not so subtle) errors. There are minefields in any language that differs significantly from your native language. French and English are really quite close, and yet there are many subtle traps and faux amis waiting for you. Lojban is very different from English, hence the potential for problems is much greater.

Also, lojban was specifically designed to have a precise logical sub-language specifically to see if learning it (lojban) changed the user's thought patterns. One should hardly complain when it finds evidence of that for which it seeks.

Lojban doesn't demand exact logical definition of ideas. The original writer could have used much looser versions of the connectives and it would have been both correct and expressive. The problem most likely arose because he (or she) tried to use connectives based on vague English glosses. That doesn't work because lojban is not a regloss of English. That's part of what you learn when you study lojban. It's also what you have to learn when you study Finnish, Russian, Mandarin or Estonian.


Which is better to learn? Which has more speakers?

Simply, Lojban is a tiny, free, living language, where Loglan is a tiny, proprietary, dead* language. If you'd ever like to speak Loglan with someone else, learn Lojban. If you're fond of dead languages, learn LoglanLanguage instead.

Lojban is increasing in number of speakers and people interested, and there is good reason to believe that that will continue. It is even quite possible that some day some people will use Lojban for day-to-day purposes; it would take a drastic and unforeseen turn of events for any of the above to happen to Loglan.

(*) "Living language" sounds like a tough claim - what standard leaves Lojban living but Latin dead? How's this standard: That's not a standard, just an assertion. Lojban may not be dead, but it isn't alive yet, either.

Lojban is embryonic. Latin is halfway decomposed. Lojban may still have a miscarriage, but nothing short of a miracle would bring LatinLanguage back as a common native language.

Historical 'dead' languages can be revived; this happened with Hebrew in the mid-20th century. However, the circumstances which led to this were most likely historically unique. -- JayOsako

Lojban is 'spoken' every day on IRC dynamically. All grammatically correct Lojban is gleaned from the channel and stored on the web. It's certainly alive, if very tiny, like Livonian. Is Latin spoken every day outside the Mass? Maybe by the Latin secretaries in the Vatican, to each other. -- komfoamonan
Sometimes I feel the need for a logical language like Lojban when naming variables or documenting code. The other day, I needed a word for "downness" or "inverse altitude". English doesn't seem to have any good ones. I considered using "altitude" with an inverted value, but the term seemed too specific and baggaged: fighter jets and stuff instantly came to mind.

The Lojban term, dizlo, on the other hand, is very simple and clear in meaning. In standard cartesian coordinates, it means "has a low y-coordinate". (Actually, "x1 is low or downward in frame of reference x2 compared to baseline x3".) Combined with the amount abstractor ni, you get the very concise ni dizlo, "amount of downness", which expresses just what I had in mind - no more, no less, leaving out all and any unnecessary information.

The English word for this concept is "depth".

But a hole is deep regardless of how downwards it is compared to a certain baseline (a hole being airmailed is still deep). "Depth" is ambiguous.

"Certain Baseline" has nothing to do with it, you're introducing "relative to x" as another statement. I think the original concept could simply be stated in English as "amount of depth". Choose a baseline - if you're going up, you're talking about "relative height" (or altitude); if you're going down, you're talking about "relative depth".

Okay, granted, "height" and "depth" are adequate terms. This was a rather stupid example. The point is that Lojban defines a long array of words with very precise semantics, and being a programmer, I appreciate precise semantics - especially when related to code.


This is off-topic but is this reminding anyone else of VaclavHavel?'s play The Memorandum? Haven't read it, but am familiar with Lojban. What about it?


Copious Lojban material is available at http://go.to/lojbanlinks (formerly http://www.pdmi.ras.ru/~sklyanin/lojban-links.html).

Also, an IRC channel has been set up on the Open Projects network: the #lojban channel.


How many Lojbanists does it take to change a broken lightbulb?

Two: one to decide what to change it into, and another to figure out what kind of bulb would emit broken light.
CategoryNaturalLanguage / CategoryNaturalLanguage (!) (What happened to CategorySpokenLanguage?)

EditText of this page (last edited December 8, 2008) or FindPage with title or text search