From Lojban
Jump to navigation Jump to search
Frequently Asked Questions About Lojban.
Ask a question on Lojban
on the mailing list for beginners
Help! I don't speak English!

In other languages

la .lojban.
la .lojban.
Nedder düütsch
Nedder düütsch
Lengua Aquatica
Lengua Aquatica
Nuvola apps edu languages.svg
No hablo muy bien inglés...
Je ne comprends pas très bien l'anglais...
Ich verstehe Englisch nicht sehr gut....
Я не говорю по-английски...
зайдите на страницу Добро пожаловать!/Русский#Контакты
mi na se bangu lo glibau
i ki'u ma do .iu tcidu dei i do na nitcu i ki'u bo lo za'i .ui do pujeca se bangu lo jbobau


What is Lojban?

Lojban is a carefully constructed spoken, as well as written, language designed in the hope of removing a large portion of the ambiguity from human communication. Lojban has been built over seven decades by dozens of workers and hundreds of supporters.

Lojban is proposed to be used a potential tool for machine translation, as a knowledge representation tool and as a programming language.

Why learn Lojban?

Who is involved in the Lojban project?

Key people are listed on the Contacts page.

How many people are there in the Lojban community? How many can use Lojban?

As of September 2015, close to two thousand people from around the world have developed an interest in Lojban to the point of ordering a book or joined a Lojban List. There is an IRC text chat where around 8 experts in Lojban are present to help you get answers to your questions about the language. At any given time, there are at least 50 to 100 active participants. A number of them can hold a real-time conversation in the language.

Lojban is practiced in social networks like Facebook, Reddit, Google+,

What is LogFest / jbonunsla ?

LogFest (la jbonunsla in Lojban) is the annual gathering of the Logical Language Group. We are required by an LLG bylaw to have an annual meeting. Every year we have a Lojban Convention for any and all in the community who are willing to come. LogFest was traditionally held at Bob LeChevalier's house in Fairfax VA, a suburb of Washington DC, usually over a weekend in July or August. We have typically had around 20 people come for part or all of the weekend. Some of these people are among the more committed Lojbanists, and around half the attendees typically are at beginner skill levels. Usually several people travel a substantial distance in order to attend, and those people are often treated as a guests of honor and are given first consideration for choosing the activities on which we focus.

People at LogFest not necessarily speak only Lojban. In fact, at LogFest in Fairfax VA you can mostly find Lojbanists that live nearby.

Other than giving those who had to travel furthest preference in picking topics, LogFest is largely unstructured, in part because we don't know who or how many are coming until a couple of days before things start. LogFest is what the community chooses to make it. Because the official LLG meeting occurs during Logfest, some of the major decisions affecting the community tend to get made there.

In 2005 and 2006, the annual gatherings started being held at science fiction conventions.

The name "LogFest" is a holdover from Loglan. The word for it in Lojban is "jbonunsla".

What is the difference between the Reference Grammar and the Complete Lojban Language?


What is the best way to start learning Lojban?

We recommend the following:

  1. Read the Crash Course or Lojban Wave Lessons.
  2. Read through the Reference Grammar - reading for concepts, not detail.
  3. Create a cheat sheet with lists of cmavo (particles) you're likely to need.
  4. Read and write Lojban texts using the Reference Grammar and your cheat sheets for reference. Post your translated text to the Lojban List for feedback.
  5. If you get serious about it, use Memrise (register and search for courses in Lojban) to help you memorize vocabulary or use Anki (downloadable, Lojban decks are downloadable within the program).

More links: Learning.

If some concepts in any of the above resources seem unclear, post your questions to the

How can I look up words when I am translating from Lojban?

la sutysisku dictionary also supports showing translations for several words at once.

Sources of text to read?

There are many. You can find a complete list of texts in Lojban here. Take your pick from original or translated, short or long, prose or poetry, currently-being-worked-on or completed-and-reviewed.

The Lojban List will have discussion in Lojban from time to time.

Other Lojbanists publish Web pages, some with Lojban texts and some entirely in Lojban. Here are some interesting sites.

Stuff on paper can be ordered from The Logical Language Group however it is much better if you can retrieve stuff directly off the net.

What messages are appropriate for the Lojban List?

Beginners are encouraged to post. Anything is appropriate as long as the title approximately reflects the content and you're not selling magazine subscriptions or mail-order brides (exception: even THAT is OK if the post is written in Lojban!). You can post on any subject in Lojban, or about Lojban in any language. You can post in any language you think people will understand.

Are there archives? WWW site? ftp site?

There are archives of some of the more recent e-mail list messages:

What is available in languages other than English?

There are brochures in Spanish, French, Esperanto, and Russian. Jorge and Jose have translated the gismu list into Spanish.

What software is available?

Up-to-date information is available at the lojban software tools page on the main site. Below are direct links to some of the software directly associated with the LLG. - is a great place where you can learn lojban words online. Just register on that website, try several courses and use the one that you like most - is a great flashcard app available for both desktop and mobile devices. It has flashcard decks for Lojban available for download - automatic grammar check of what you write in Lojban (which is unique among all languages!) with English glossing (i.e. translation). - Even better than the previous item. But without glossing. allow automatic creation of lujvo from combining words and splitting lujvo back to them

Prolog Semantic Analyzer

Random Sentence Generator - updated version in progress, the available version has somewhat obsolete grammar definitions

zmifanva - Lojban ↔ English Machine Translation Engine based on phrase-based statistical machine translation.


This part tends to be an "add as you go" page that answers questions that beginners and intermediates may ask about particular elements in the language that confuses them or they find to be "void." Such as the lack of grammatical gender.

While it is to be expected that these questions are vague and underinformed, and we should do our best to answer them regardless, that does not excuse not doing one's best to ask precise and informed questions. The more effort you put into the question, the more you'll get out of the answer.

Does Lojban get rids of synonyms? Does Lojban have polysemous (multiple meanings) words? Does Lojban have homophones?

Does Lojban have words that has the same meaning?

There are, but they are not many. Synonym words aren't a problem; what can be a problem are polysemous words (words having several distinct meanings) and homophones (two different words pronounced the same).

Lojban has no homophones, and predicate words have a single meaning each.

Lojban isn't designed to avoid them. lo cmavo (particle words) can have different functions in different contexts, but the grammar always make it unambiguous which function a cmavo has in a specific grammatical context.

Lojban predicates may have a vague meaning (that is, a broad range of meaning), but never have several unrelated meanings (ambiguous meaning, polysemy).

For example, lo danlu (animal) is more vague than lo cipni (birds), that in turn is more vague than lo jipci (chicken).

But that is not polysemy, just a matter of vagueness (degree of precision).

So Lojban does not prevent you to speak with vagueness if you want so (vagueness can be very helpful!)

(Glossary) In English descriptions of lojban, I see lojban words used, as well as lojban jargon. What do these mean?

Here are APPROXIMATE definitions.

Note: Words in ALL CAPS on the Lojban Mailing List often refer to Lojban parts of speech. the English-language phonetic translation of ' (single quote) is "h", so when this convention is used, "h" is used as the "capitalization" of '. So in English-language discussion, the lojban part of speech contaning the word la'e (the referent of a sumti) would be LAhE. <languages/>attitudinal or cnima'o = a Lojban interjection (corresponding to "Wow!" or "Eek!" in English, Lojban has many more of them more precise)

audiovisual isomorphism
Used to describe the fact that spoken and written Lojban have the same form
verb phrase
~= a predication
a verb, any word that can state a relationship among several objects or concepts, and thus be the core word in a Lojban bridi. Can be converted to noun be prefixing with lo or similar articles.
a predicate word
A little word showing structure rather than carrying content meaning.
Lojban names
Special word indicating how the speaker got their information
Borrowed word
Lojban prefix word for sumti
Basic 5-letter lojban root word
the basic gismu list is available here

ju'i lobypli (literally meaning "Attention, Lojban-users!", also known as JL) was a Lojban quarterly journal published by The Logical Language Group; 18 issues appeared between July 1986 (as Me la Uacintyn Loglytuan, the Washington D.C. Loglan journal) and June 1993.

Read more ...

le lojbo karni
(The Lojbanic Journal) (also known as LK). A Lojban newsletter intended to keep our lowest level of supporters informed as to what is going on (in hopes of inspiring greater activity). Also currently suspended.
The old name for fu'ivla
Logical Language Group (LLG)
Compound word

sumtcita = modal = preposition = adverb = tag

An attitudinal meaning "In my opinion"
place structure
The specified canonical order of bridi, so that you know who is doing what to whom.
Building block (or blocks) of lujvo (compound words)
verb relation, the part of a Lojban sentence that expresses the relationship between the various objects (sumti)
class of a particle in Lojban
A hypothetical borrowed word, which would not be legal because it could be interpreted as parts of other words in some contexts.

The slinku'i test (le valslinku'i cipra in Lojban) refers to the test for type 4 fu'ivla being morphologically valid. Words that fail the test are called slinku'i, as the word "slinku'i" fails the test. This rule applies mostly to type 4 fu'ivla, which have to be very careful not to cause any morphological conflicts, they become ambiguous if certain syllables come before them. For example, if you say *pa slinku'i, it becomes the lujvo paslinku'i ("robe-chain-caretaker").

noun/pronoun/personal name. An object or idea which may be related to others, that relationship being expressed in a Lojban bridi.
the expression filling an argument place in a bridi.
A compound verb, or in other words, a phrase formed of two or more brivla
  1. REDIRECT the Loglan Institute

How do you borrow words from other languages?

There are four ways to borrow words. Only the most common method is covered in Lojban Reference Grammar (Chapter 4, Section 7). Borrowed words are called fu'ivla, meaning approximately "copied words" (after all, "borrowing" implies we're going to give them back someday!).

The use of lujvo is not always appropriate for very concrete or specific terms (e. g. "brie," or "cobra"), or for jargon words specialized to a narrow field (e. g. "quark," "integral," or "iambic pentameter"). These words are in effect names for concepts, and the names were invented by speakers of another language. The vast majority of names for plants, animals, foods, and scientific terminology cannot be easily expressed as tanru. They thus must be "borrowed" (actually, taken) into Lojban from the original language, forming words called fu'ivla. The word must be Lojbanized into one of several permitted rafsi is then attached to the beginning of the Lojbanized form, usually using a vocalic consonant as "glue" to ensure that the resulting word doesn't fall apart. The rafsi categorizes or limits the meaning of the fu'ivla. Otherwise a word having several different jargon meanings in other languages (such as "integral"), would require a choice made as to which meaning should be assigned to the fu'ivla. fu'ivla, like other brivla, are not permitted to have more than one definition.

Summarizing the most common method to make fu'ivla:

  1. Lojbanize the word to be borrowed by the methods used for cmene.
  2. Convert all "y"s to some other vowel or to a vocalic consonant.
  3. Modify the ending to be a vowel, either by dropping a final consonant or by adding an extra vowel.
  4. Modify the beginning to be a consonant, either by adding an extra consonant or dropping an initial vowel.
  5. Choose a gismu (not a rafsi) that categorizes the fu'ivla into a "topic area." Replace the final vowel of the gismu with a vocalic "r".
  6. Prefix the modified fu'ivla.


English lojban
spaghetti cidjrspageti or djarspageti
maple tree tricrmeipli
maple sugar saktrmeipli
mathematical integral cmacrnintegra or cmacrntegra
brie cirlrbri
cobra sincrkobra
quark saskrkuarka
iambic pemcrniambo

A fu'ivla thus consists of three parts: the classifier, the glue, and the borrowed part

The quintessential example is djarspageti, meaning "spaghetti." dja is the classifier: it is the short form rafsi for cidja, meaning "food". "r" is the glue: it is necessary to keep the word from falling into two parts. spageti is the Lojbanized version of "spaghetti". The classifier is glued on the front for two reasons: it helps identify strange borrowings, and it prevents borrowings that happen to coincide with things that are already Lojban words. For example if you borrowed the word "spageti" directly, it could lead to ambiguity in a phrase like

ko bevri re spageti palta

which could mean "Bring two plates of spaghetti" or something like "Be a carrying-reptile and a plate made of this":

ko bevri respa ge ti palta

In a natural language there would be no doubt which of the two was meant, but Lojban is constructed so that you shouldn't need to understand the sentence to know where one word ends and the next begins.

Isn't it confusing that some rafsi are identical to cmavo?

No, it isn't. In theory you can tell completely from the neighboring syllables whether something is a rafsi. This is how the computer is able to parse Lojban without understanding its meaning. For example the dei in bavlamdei ("tomorrow") is a rafsi for "day", not the cmavo dei, a special pronoun meaning "this sentence". We know which is which because Lojban words can't end in a consonant, so dei must be a part of bavlamdei.

bavLAM can't be a whole word. (No, it can't be a name, either. Names end with a consonant followed by a pause, written as a ".") In practice you can also use your knowledge of the meanings of the words to help with this. It is possible to think up a sentence like

la .bavLAM. dei cusku

"Bavlam says this sentence.", but it is not likely in practice if you don't know anyone named "Bavlam"!

Why do some texts use "h" instead of ' (single quote), or have parentheses, or other non-standard conventions?

Aesthetic reasons. And Rosta in particular thinks mohi looks better in print than mo'i (space motion) and sometimes uses it the hopes that he can influence the Lojbanic Community Links to accept this other spelling convention. This is purely a difference in spelling; they are pronounced the same, and should be considered different ways of writing the same "letter." This alternate spelling also is somewhat closer to the spelling of TLI Loglan. There is another similar spelling convention, which has never been used, designed to make Lojban look more familiar to potential converts from TLI. This alternate convention may be found in the Lojban Reference Grammar.

There are a number of other non-standard spelling conventions:

Lojban text is often wrapped in { } (curly brackets) when is quoted in anon-Lojban text.

And Rosta had a style in which ' (single quote) is omitted altogether where the vowels couldn't possibly be stuck together; for example, he'd write coe for co'e (unspecified bridi), since "oe" is not a legal combination. He has used "." (period) as in English, to end a sentence, rather than as a pause. He has capitalized the first word of the sentence.

La Xod uses a single capital letter for acronyms instead of the lojban letterals (i.e. A instead of .abu.).

There have been others. In general, they are all quite understandable by a competent lojbanic person, with the possible exception of And's removal of ' (single quote).

Why are there so many words for AND? Why not just let ".e" (and, between sumti) connect two bridi, bridi-tails (a predicate and a sumti), or anything else?

We use different connectives for different scopes. Doing so helps the listener keep track of what exactly the speaker wants connected. This is much more important in speech than in text, because in text you can re-read and ponder. If you have a multi-part nested sumti joined to another multi-part nested sumti, having clear indicators of scope may make the sentence understandable when otherwise it is not. It is therefore hoped that spoken Lojban and written Lojban can be similar in level of complexity. (You know, audiovisual isomorphism).

However, according to non-official dialects of Lojban described e.g. in The Crash Course in Lojban .e is a synonym of je. Likewise, instead of gi'e we can say vau .e or vau je. This is all stylistic variations but might be more intuitive for many people. Once again: this is used in a non-official dialect.

Is the 's' at the end of lojbanized names (such as 'noras.') silent?

No. 'noras.' is really pronounced with an 's' at the end.

Some languages can borrow names with virtually no changes, but Lojban is one of the languages that sometimes requires quite heavy modification of the names. Specifically, all Lojban words of the class cmene must always end in a consonant. Sometimes, this requires removing a vowel at the end, or adding a consonant. By convention, this is often 's', but it could be any consonant.

If names were not changed this way, some of them could be mistaken for other words, or sequences of words. In this case, were there no 's' at the end, 'nora' would read as 'no ra', ie. "none of them"! This distinction is as important in speech as in writing, so names, as any other Lojban word, must be pronounced exactly as written. There are no silent letters in Lojban.

Remember folks, Translation Is Hard. You can't translate anything you don't understand.
Why isn't there anything like his/her, and what can I use to get that effect?

Different languages have different concepts ofgrammatical gender. In Lojban, the closest thing to it is ra and ri for any gender, ra poi fetsi and ri poi fetsi for she and ra poi nakni and ri poi nakni for he. Is is also possible to use to assign lerfu variables (letters) to referents. See lerfu pro-sumti, and why ko'a sucks for more information on some Lojbanic ways to deal with anaphora.

I am translating from a language that the speaker refers to himself/herself using various different words, should I and how can I translate those to lojban? (For instance, Japanese; watashi, atashi, boku, ore, sessha, etc.)

This requires actually understanding the meaning of the words. In Japanese, contrary to popular belief, the words do not imply the gender of the speaker. Rather, they reflect the speaker's attitude and politeness about and towards the environment around him, and his perceived status in it. Combinations ofle'o, ga'i, .io and others suffice in this case.

In the case of actual gender, there is no attitudinal. However, consider, is the speaker actually making a point of the gender? Or are they simply following the rules of the language, in the most expedient fashion possible? We suggest you do the same (in either case: if the speaker needs to state something about someone's gender, then by all means do so explicitly. But you can use words like nakni and fetsi and add incidental information with noiclauses.

Can I write le mibrirni/doirirni <unrelated bridi> to mean "I"/"You listeners" with "I am/You listeners are a/as-a parent" as some incidental information, like when you refer to someone in other languages?

(Related to the previous question but not entirely.)

No. mibrirni is a lujvo and, whatever it means, it functions like any other brivla, so le mibrirni is a sumti that refers to that which I describe as a mibrirni (a "me-parent", either my parent(s) or myself as a parent maybe) and it will fill a place in a bridi like any other sumti.

doirirni = doi rirni (two separate words even if you don't write a space between them) is a vocative you can use to address your parent(s).

Grammar and morphology issues aside, do you mean something like "Now I know all you loving parents are hardworking decent people..."? Or something like that? You can use doi vocatives to indicate whom you're talking to. doi rirni means "I'm addressing you, parent(s)," but on the other side, if the audience knows who they are, it therefore means that you, the audience, are parents. You can do something similar with mi'e, but Mark Shoulson doesn't really see how to fit that in naturally. I don't think I understand the context of the question in this matter. Information incidental to something that's already being mentioned is done well with noi.

Why does no one like my lujvo?

Making bad lujvo is easy. Making good, understandable lujvo is subtle. It requires a nontrivial grasp of Lojban, and a solid idea of what lujvo you're trying to make. In particular, you have to understand that you're translating a concept, not a single word. Hot tip: think about how you'd use it in a sentence as the selbri, not in part of a sumti.

  • This question was so affecting that I couldn't help weeping while reading it -inego
Definitions of attitudinals and discursives (The selma'o UI layout)
  • Question:
    definition of .a'o is attitudinal: hope - despair. What does it mean?
  • Answer:
    it means that .a'o means hope, and .a'onai means despair.
    It's showing you the opposite/negative as well for clarity.


What parts of the language are well worked out, and which parts are in flux?

As of 1997, the language design has been baselined for a minimum of 5 years after the publication of the three books: the reference grammar, the dictionary, and the textbook. Since the latter two are not near publication, this means that no changes to the language will even be considered until at least 2006. This baseline is non-negotiable, even if we enter into discussions with The Loglan Institute to reunite the Loglan community now that JCB has passed on.

The phonology, orthography, and morphology have been essentially stable since 1988, except for a slight change in what counts as a legal fu'ivla. The 'gismu list has been stable since 1988, except that about 25 gismu have been added and 2 gismu had minor changes up until 1994, but since then only clarifications of confusing wording have been made to the official baseline gismu list, and they are now considered baselined. The rafsi have been unchanged since 1993, when around 20% were changed in a final tuning before baselining them. Much existing text has not been updated after the rafsi change, so texts dated before 1993 may be confusing. The grammar has been basically stable since 1993, was under careful documented control after that date, and frozen with the publication of The Complete Lojban Language in 1997. The cmavo list has been baselined since 1997, though new compounds may be defined through usage. The only area not frozen is the addition of new words to the lexicon through borrowing (lujvo). Those two productive areas are open-ended; a primary limitation on the language definition is the inability to record and define new coinings as fast as they are invented.

The basic semantics of the language are stable. There are still ongoing disputes about "how to say it best in Lojban"; we expect these to continue indefinitely. As a matter of policy, changes even to Lojban non-baselined usages, if they would require people to re-learn things, are resisted vigorously. Almost anything that practical people (i.e. beginners) would actually use is well worked out, debugged and stable. Points of controversy include highly technical philosophical issues such as whether empty sets are or are not excluded as the candidate referent set of a sumti, or whether current grammar is adequate to represent all possible forms of indirect questions. The answers are very important to the language definition (no joke), but people have been speaking natural languages for years without knowing the answers, so don't worry about Lojban. And you don't have to know what lambda calculus is. The major point of controversy that is significant to beginners is a feeling in some quarters that the policies used to design the place (argument, gismu (basic predicate words) were not exactly optimal. Sometimes the semantics of the less-used places are at issue. Officially, we'll go with what we have to preserve the investment people have made in learning the language, and after the baseline period is over, those who are actually speaking the language will be allowed to discuss changes in Lojban. There are no plans to ever seriously discuss changes to the language other than in Lojban.

What are the most current revisions of each part of the language descriptions?

The versions found on the Lojban web site are definitive, except that the printed version of The Lojban Reference Grammar (which is called The Complete Lojban Language) takes precedence over the HTML version. Working drafts of books in preparation are maintained on lojbab's home computers, and may be slightly updated from the on-line versions.

What projects are being worked on? When will they be done?

The Complete Lojban Language is of course complete and published.

For a current list of ongoing projects, see the Official LLG Projects page, the Official LLG Committees page, and the wiki ongoing projects page.


Projects are all being done by volunteers, and therefore will be done when people finish them. We've promised dates in the past and invariably been wrong. The priorities are for the dictionary and the introductory materials book, with the latter more likely to come out earlier than the former. Publication of books is severely hampered by finances (anyone with money is welcome to donate!), and the lack of ability to publish in the short term has tended to hurt motivation and productivity of those working on those publications.

What can I do to help?

A few possibilities:

  • Study Lojban vocabulary using Memrise or Anki
  • Try expressing yourself in Lojban
  • Invite others to join a live IRC chat in or about Lojban
  • Keep a diary in Lojban
  • Check out the Projects and the Committees pages for new and ongoing projects
  • Send money to the LLG using PayPal


How was the default selbri determined? (There does not appear to be any rhyme or reason for the order of gismu).

They went through a lot of revisions; it is something in-between planning and evolution. There is a considerable amount of system there, but it is ill-documented.

The nice thing to know is that you don't have to memorize the place structures. Just using the language, you will come to have a feel for which places are present and in what order.

Most people who take the time to become familiar with the language find that with practice, they begin to develop a sense of what the place structures ought to be. Conveniently, their sense usually matches up with the definition.

How did the gismu get made: discussion, etymology examples?

There is a full set of etymologies on the file server. There is some explanation available, including samples (broken link), and a more verbose description in the Level 0 package and at the Lojban Etymology Information page.

The etymologies in 6 languages and the scoring for each language are given, in order Mandarin Chinese/English/Hindi/Spanish/Russian/Arabic. A 0 score means that the language made no contribution to the word, and thus its etymological keyword did not matter. The languages were weighted. While new gismu are not being made, lojbab updates the weighting based on estimated numbers of first and second language speakers every couple of years. The original and 1999 weights are available online.

What is the difference between Loglan and Lojban? How is Loglan-82 related?

James Cooke Brown came up with the idea of Loglan in the 1940s and starting inventing the language around 1955. It has been evolving ever since. In 1982-4 or so there was a political disagreement and the Loglan community fell apart. A couple of years later, Bob LeChevalier (lojbab), then working with JCB, attempted to resurrect the community. This led to a falling out based on JCB's proprietary intellectual property claims on the language. The community split into two efforts, with JCB at the head of of the Loglan effort, The Loglan Institute (TLI), and (lojbab) at the head of the Lojban effort, the Logical Language Group (LLG). The latter is Lojban, which LLG (backed up by a court decision) considers a subcategory of Loglan. Sadly, JCB passed away in 2000. It is unclear if the Loglan organization has leadership capable of sustaining the effort. A more lengthy discussion of the issues can be found in the article The Loglan-Lojban Dispute, and in ju'i lobypli (Attention! Lojbanic Users).

lojbab makes occasional efforts to spark discussion that would lead to reunification of the efforts behind Lojban, although no consideration will be given to abandoning the Lojban baseline.

Loglan-82 is a completely unrelated computer language developed in Poland by people unaware of the existence of the Loglan/Lojban project.

Why do older Lojban texts (before 1993) not make much sense sometimes?

Lojban has been quite stable since 1992-3, when the last significant changes were made (since then, most "changes" have been additions to the language which did not make older text invalid). The last changes to the gismu list were made in 1992, and in 1993 the rafsi in Lojban word-making. There is a translation table from the older rafsi to the final set.


I'm not quite sure I understand the written descriptions of how the vowels sound.

Here is a link to vowel sound file in Ogg Vorbis format.

You may also find useful information on the FAQ-like Resources page.