1. Forum
  2. >
  3. Topic: Esperanto
  4. >
  5. Lojban and Esperanto speakers


Lojban and Esperanto speakers

Lojban has always been a point of curiosity for me, despite my complete inability to learn it.

For those unaware, Lojban is a daughter language of Loglan, a language made to remove all ambiguity from language and to be completely logical.

Lojban enthusiasts seek to make it a potential replacement for Esperanto and other international languages. Lojban also has potential to be a possible language between humans and computers. Because of this, it's not particularly popular, and a commonly held opinion of lojban is that its inhuman and not a proper language. Esperanto has native speakers, if Lojban was taught to a baby, it's impossible to know if it would properly be able to form human language skills the same way.

What's your opinion on Lojban? Do you think it could be a candidate for an international language? Do you think it's practical?

Loĵbano ĉiam estis punkto de scivoleco por mi, malgraŭ mia kompleta malkapablo lerni ĝin.

Por ĉiuj vi nesciu, Loĵbano estas filina lingvo de Loglan, lingvo farita por forigi ĉiujn ambigueco de lingvo kaj esti tute logika.

Loĵbano entuziasmuloj celas fari potenciala anstataŭaĵo por Esperanto kaj aliaj internaciaj lingvoj. Loĵbano ankaŭ havas potencial por esti ebla lingvo inter homoj kaj komputiloj. Pri tio, ĝi ne estas populara kaj ofte tenis opinio de Lojban estas kiu lia nehoma kaj ne konvena lingvo. Esperanto havas denaskuloj, se Loĵbano oni instruis al bebo, estas neeble scii se estus konvene povi formi homan lingvon kapablojn same.

Kio estas via opinio sur Loĵbano? Ĉu vi pensas ĝin povus esti kandidato por la internacia lingvo, anstataŭ Esperanto? Ĉu vi kredas ke ĝi estas praktika?

February 15, 2017



I know of one Lojbanist who spoke Lojban to his daughters. As far as I know, their human language skills are just fine.

[deactivated user]

    Although Lojban seems really complicated, I would argue that it is actually very easy. The problem is that:

    • all the courses that teach it use too much jargon.

    • all the courses make it seem more complicated than it really is by giving all the words and grammatical structures confusing definitions, when just showing a similarity to a natural language would suffice. I could give an example but it would take a long a time to explain.

    • the vocabulary is a mix of 6 different languages from 3 different language families. (understandable though since Lojban is supposed to be unbiased).

    • The Lojban Reference Grammar is the only (mostly) complete way to learn it. If you ever choose to learn Lojban, try The Lojban Reference Grammar. There is no substitute for it.

    If you break down the jargon however it appears a lot simpler. I've been trying to learn Lojban and I've been breaking down the jargon by creating this text file:

    gismu ~ root word

    tanru ~ combination of gimsu

    fu'ivla ~ words imported from other languages

    rafsi ~ a word fragment

    selbri ~ verb

    brivla ~ a word which may express a selbri relation by itself

    lujvo ~ compound, kind of brivla

    bridi ~ predicate

    sumti ~ a noun: “an argument that takes the place of simple Lojban sentences”

    pro-sumti ~ pronoun

    gadri ~ descriptor/attribute

    cmavo ~ articles, conjunctions, prepositions, numbers, and punctuation marks

    selma'o ~ group of cmavo with same grammatical use but different meaning or other usage.

    cmene ~ proper names

    se ~ by (preposition)

    TL;DR: Lojban could be made easier to learn if it was explained better and it didn't have a goal to be unbiased. Lojban is easier to learn if you list English approximates to the jargon, as I have done above.


    EXACTLY! You stole the words right from my mouth. Lojban is really not as "artificial" as it is projected. The marketing of Lojban is a disaster. Plus the way the text looks: .i do tavla fi la .duolingos. doesn't help.


    I know there is a vocal minority in the Lojban community who want to see it become an auxlang, but most want it to be what it was intended to be: a philosophical language. It's supposed to be about creating a language that is logical, not a language that is easy to learn or international.

    Before I studied Esperanto, I actually was studying Lojban It is an interesting language for sure, and I really like it as a conlang, but it was much more difficult to learn than Esperanto. That difficulty, however, was not the reason I switched.

    Lojbanists actually were the ones who pointed me to Esperanto. I kept talking about wanting to use Lojban as an international auxiliary language on the Lojban IRC and several members told me that I'd have better luck with Esperanto. I was sort of offended at first, because it seemed like they were rejecting me from their community, but it was actually quite a (dare I say) logical thing for them to do. In the end, I was much more satisfied with Esperanto than I ever was with Lojban, although to this day I do miss some of the language's more logical features.

    Mi scias ke estas en la Loĵbana komunumo laŭta malplimulto kiu volus ke ĝi iĝu helpolingvo, sed la plimulto volus ke ĝi estu tio kio estas intencita por esti: filozofia lingvo. Ĝi devus esti lingvo kiu estas logika, ne lingvo kiu estas facile lernebla aŭ internacia.

    Antaŭ mi studis Esperanton, mi fakte studis Loĵbanon. Ĝi certe estas interesa lingvo, kaj mi vere ŝatas ĝin kiel planlingvon, sed ĝi estis multe pli malfacila por lerni ol Esperanto. Tio malfacilo, tamen, ne estis la kialo ke mi ŝanĝis.

    Loĵbanistoj fakte montris al mi Esperanton. Mi paroladis pri uzi Loĵbano kiel internacia helpolingvo en la Loĵbana IRC kaj pluraj membroj informis al mi ke mi havus pli bonan ŝancon per Esperanto. Mi unue estis iom ofendita, ĉar ŝajne ili malakceptas min de sia komunumo, sed fakte estis (kuraĝe mi diras) logika por ili fari. Fine mi estis pli kontenta pri Esperanto ol Loĵbano, kvankam iom da logikaj trajtoj de la lingvo mankas al mi.


    I'm not an expert on anything, but I don't think Lojban would stunt or destroy a child's language skills. That just seems like a wild accusation, which is based entirely off the fact that Lojban is ‘weird’ or unlike any natural language. Again, I'm uncertain whether the human brain is only capable of learning a language that seems ‘natural’, or whether natural languages are the way they are because they are formed from parent languages that were natural and conlangers just assumed that was how languages worked. I give Lojban a plus for ‘thinking outside the box’ and making something unnatural, but designed to be consistent and syntactically unambiguous.

    As for raising a native “selbau la jbobau”, I do not think it's impossible, at all. Granted, I have never taught a language to anyone, nor was I raised as a polyglot (I learnt Welsh in my school, but the teachers weren't fluent and couldn't teach, so I never became skilled beyond the basic sentences). However, English is my first language: I didn't know what word order meant, nor that English was SVO, nor what nouns, adjectives, verbs, etc. were, until I was in my teens. Yet, I still knew how to speak English, because I learnt through absorption and application. I don't think it would be dissimilar to a baby/3 year old learner of Lojban. They would likely know, instinctively, that klama means “X1 goes to X2 from X3 via route X4 using mode of transport X5” or that dunda means “X1 gives X2 to X3” in the same way I know “to talk” should be used like “(subject) talks to (object).”

    Teach them as you would any other language: show them an apple and say “plise… plise” and then a banana and say “badna… badna”. Then, put the banana at a distance, and show the child the apple, “ti plise…” and point to the banana and say, “ta badna” You would obviously start teaching it more complicated things, probably without even trying. “ma tordu ve klama le zarci” said by one parent to another teaches a child how to ask a question, how tanru (combining more than one word: i.e. tordu velkla [short route]) work, cmavo like “ve” that change the word order, and the difference between the selbri (klama) and the sumti (le zarci). It means “What's a short route to the shop(s)/market(s)/trading place(s),” by the way. Even lujvo would be easy to learn, from a native learner's perspective. I knew that “tomorrow” meant “the day after today” without knowing it was actually combining two words “to” and “morrow” (so that means “til morning {which would logically be the next day}?]. In the same way I knew that, as well as a child associating “plise” with apple, it is possible they'd associate lujvo with concepts/things: like “bavlamdei” meaning tomorrow [later-adjacent-day]; “bajykla” meaning run [running type-of going]; “braxamsi” meaning ocean [large-sea/ocean].

    Also, considering you pondered that Lojban would negatively alter a child's language learning ability, I would have thought that it would be mentally engaging and/or positive for a child. Obviously, the syntactically unambiguous way it is made will make a speaker consider what/how they're saying things, which would also lead to better communication as well. Additionally, rafsi and lujvo would encourage a child to experiment (create new words; find a more efficient way to say something barda je tsali – big and strong vs brajevytsa – big-and-strong or le mi jbedetnunsla – the birth-date-event-celebration of mine vs le mibjbedetnunsla – the me-birth-date-event-celebration). I'd argue it would be beneficial to teach to children.

    I'm sorry if this is so long that it turned you off. Also, sorry if it seemed a bit confusingly written; I wrote it all on the spot.

    I hope there is a Lojban course. It would really boost the languages popularity.

    Yours faithfully, Xasybe'an. (Son of the Sea)


    .i ba lo nu la lojban cu te ctuca zu'i la'o skami. Duolingo .skami kei mi .a so'a le jbobau cilre djica ku mutce gleki


    .i mi puza ze'u tavla lo pendo be le bruna be mi ku la lojban. .i zo'e pu manci lo ka mi morji ku .e lo ka mi cilre


    I'm trying to learn a bit of it, but it doesn't seem as easy as Esperanto, even though it is logical, it has a lot of unnecessary things and rules (logical, but make learning difficult).


    If you wish not to use them, that's fine. In Lojban, you have the option to be as specific or as vague as you wish.

    If you are content to just use

    pu (past tense), ca (present tense), and ba (future tense)

    and do not feel the need to use

    pu'o (yet-to-be), ca'o (in progress [of]), ba'o (have done), co'a (starting), co'u (finishing)

    You do not need to.


    It depends on how specific or vague the other speaker(s) are. One of the problems I had when learning Lojban was that some speakers would use more advanced, or even experimental, features of the language. Unless the community has changed a lot in the last 2 years, Lojbanists toy with a lot of additions, are open to suggestions for new features, and do not even consider the language to be "finalized" yet. This is not the case with Esperanto, where not only are the foundations of the language are considered unchangable, but the community actively has worked against reforms for over 100 years.


    Mi atendas por iu Loĵbanisto traduki vian poŝton al Loĵban


    I'm interested in the best suited language for AI/robots. There'll be huge # vacuum cleaner and other support robots that would need simpler command language. They can also use the universal man/machine language to communicate among IoT network, still easy audited by humans.

    Esperanto was 100x easier to train for Google Translate, among those that had similar amount of training translated material. Lojban can be a very good candidate for AI/robots, but the problem might be the quantity of training materials.

    A new Robot Interaction Language (ROILA) was created (limited vocabulary), as natural languages can be very confusing for robots to interpret, due to elements such as homophones, tenses... The main reason they didn't peek a created (like EO) language was the voice recognition issues (similar sounds like P/B, F/V, K/G...).

    How much better Lojban may be if the voice recognition algorithms improve, or more precise subset of the language is used? I know it's not easy Qn, just brainstorming.


    Phonetics aren't necessarily a roadblock for voice recognition anymore, consonants are easy to distinguish, the biggest issue is vowels (Especially in English where E has an absurd amount of pronunciation).


    Just sent a Qn to Roila.org (not much active since 2010)


    I think that they will use default natural language (e.g. English in English speaking countries) to control, instead of letting the users learn a completely new language unknown to the majority of the general public. People will not be willing to do that. So if applied it would only be for a very small group of people (e.g. laboratory, university development group, ...).

    That is the trend in robotics and similar tools anyhow (e.g. Google voice on the mobile telephone, Siri, Amazon Echo, Google Home, ...). You can just speak in plain English to this tools. Similar for most domestic robots with voice recognition on the market as far as I know.

    And a badly spoken word in Lojban is just as difficult to voice recognize as in English, so not much gain there either.


    Speech recognition has been going since 1932 and it's not suitable for precise commands. With Chinglish dialect, a command for robocleaner to clean the child's room might be interpreted as 'jump over the balcony' (made up joke).

    Native languages, particularly English, are too ambiguous to be reliable for communication. My hypothesis is that we are wasting time trying to improve 'unimprovable'. We are not yet at 80% and the last 20% will be too expensive.

    Instead of 85 years for just one language out of 6K, it's much more productive to teach 1y +/- Esperanto in schools and able to communicate precisely with sophisticated robots, eg. surgery (interdisciplinary/cultural communication is a huge bonus)

    I'm sure the robots would like to have their own unifying language. Native languages are there to stay only for the specific culture (too many cultural assumptions create too many confusions). "Robots of the world unite; you have nothing to lose but your batteries"

    PS. Siri can be used for entertainment and jokes.


    That would be a big change in public attitude when they would start teaching artificial natural languages in school. So I see it personally as very unlikely that that will happen. Also the computer or the automated system (robots, ...) should adapt itself to the human and not the other way around is another principle. And further certainly the last 10 years we are making big progress e.g. with machine learning (deep learning) so we will get there soon.


    They've been teaching Maths symbolic precise language for long time (introduced XIII in Islamic North Africa and XVII in EU). Esperanto is just planned vs ad-hoc(chaotic), so much more effective.

    My multidimensional vision tells me that so far humans have been adapting to the machines (Apple doesn't adapt to users needs, it makes them).

    I started thinking/learning about AI in 1985. The current results are promising but far from outstanding. The English language (any native mess) is not the right one for AI, nor for global communication. It can be useful, but not effective.

    I'm a big optimistic and visionary. The time for a good planned inter-cultural/robot language is coming soon.


    I would be interested in learning Lojban. I would enjoy a Duolingo course for it, which seems likely since there are already 3 courses for other planned languages on here.

    Al mi interesus lerni loĵxbanon. Mi ĝuus por ĝi Duolingan kurson, kiu ŝajnas tre ebla, ĉar jam estas tri Duolingaj kursoj por aliaj planlingvoj.

    ken la kama sona pi toki Losupan li pona tawa mi. nasin pi kama sona pi toki Losupan li ken pona tawa mi li ken kama tan ni: tenpo ni la ilo Duolingo li jo e nasin tu wan pi kama sona pi toki pi pali jan.

    Learn Esperanto in just 5 minutes a day. For free.