4 ms·

I have been learning Lojban off-and-on for a while. I think the core idea behind it is very interesting: sentences as logical predicates. The root word concept

I have been learning Lojban off-and-on for a while. I think the core idea behind it is very interesting: sentences as logical predicates. The root word concept is powerful. But I feel like most of the rest of it is too complex. They could have would up with Prolog. Instead they wound up with C++, where everything they could think of wound up in the language. That said, it's still big and inspiring. But a lot of talk about Lojban turns into talking about whether a sentence is valid or not. The software ecosystem around it has grown up a huge amount in the last few years.

BuuQu9hu 10y ago
I feel that you are mischaracterizing us. Anybody who wants to learn Lojban can start with https://mw.lojban.org/papri/la_karda https://mw.lojban.org/papri/la_karda and come to #ckule on Freenode IRC. It takes a weekend to learn surface syntax and 2 to 3 months to get conversant.
- fusiongyro 10y ago
  I apologize. I think there are different conceptions of complexity. Lojban is built on a rigorous definition of complexity and clearly much work has been put in to minimize it. My feeling from having spent time learning it is that the goals are interesting and clearly they are very close to achieving them, but they don't match up to my expectations for simplicity, from reading about linguistics and other constructed languages as a hobby. I am not trying to bag on Lojban. It's very interesting and worth spending time learning. I will definitely follow up on your resources—I have been reading la karda for the last week already actually. :)
kmicklas 10y ago
Lojban would have been infinitely more promising if it didn't force you to memorize a completely arbitrary argument template for each root and instead used standardized semantic role prepositions (destination, agent, reason, instrument, etc.).
- schoen 10y ago
  There is an argument about that here: https://mw.lojban.org/papri/Why_positional_case_system https://mw.lojban.org/papri/Why_positional_case_system I think your position is more persuasive, although I've never argued about this with an expert Lojbanist. I speculate that overall they feel that (1) there could be dozens or hundreds of different cases reflecting the role that a noun plays in a particular predicate relationship, (2) most don't apply at all to any given selbri [predicate relationship], (3) [an argument made in this page] some people's feeling of which case is appropriate for describing a particular relationship is culturally arbitrary, as they differ significantly from natural language to natural language, and (4) they don't want to force the use of prepositions, for concision.
  - fusiongyro 10y ago
    Even with "just" transitive verbs taking three arguments, it's easy to lose the plot in natural languages like English. The idea that it wouldn't happen in unmarked sentences with two more seems absurd to me. They found some natural relationships that happened to have five members and they had five vowels, so the whole FA series came from that. Also the SE series only seems speakable for se/te. Beyond that, I think it would be a head-scratcher in English. I also think even if people are getting fluent in Lojban, they're unlikely to be gracefully handling free word order scenarios where the selbri shows up not in position 2. Are there SOV-speakers producing SOV-ish shaped Lojban and everybody is getting along fine? If you look at the official cmavo list above, FA and SE account for five words each, so ten of the 1000 in the list. Of course, they're pretty productive as prefixes too, so more than that. But I think part of what makes it complex is that there are so many of everything... forethought and afterthought of all the connectives; differently typed connectives for combining sumti, selbri, bridi; the non-obligatory tense/aspect/mood system that accounts for probably at least 40 particles; etc. It's a neat language. There's a lot of neat stuff in it. I'm sure it's fun to be able to speak and understand it. I guess mainly I think it doesn't speak to your innate language facility's strengths, so I doubt it can be done realtime as a language, as specified. It can clearly be done with our reasoning faculties. Maybe that's enough.
    - 10y ago

BuuQu9hu 10y ago

I feel that you are mischaracterizing us. Anybody who wants to learn Lojban can start with https://mw.lojban.org/papri/la_karda https://mw.lojban.org/papri/la_karda and come to #ckule on Freenode IRC. It takes a weekend to learn surface syntax and 2 to 3 months to get conversant.

fusiongyro 10y ago
I apologize. I think there are different conceptions of complexity. Lojban is built on a rigorous definition of complexity and clearly much work has been put in to minimize it. My feeling from having spent time learning it is that the goals are interesting and clearly they are very close to achieving them, but they don't match up to my expectations for simplicity, from reading about linguistics and other constructed languages as a hobby. I am not trying to bag on Lojban. It's very interesting and worth spending time learning. I will definitely follow up on your resources—I have been reading la karda for the last week already actually. :)

kmicklas 10y ago

Lojban would have been infinitely more promising if it didn't force you to memorize a completely arbitrary argument template for each root and instead used standardized semantic role prepositions (destination, agent, reason, instrument, etc.).

schoen 10y ago
There is an argument about that here: https://mw.lojban.org/papri/Why_positional_case_system https://mw.lojban.org/papri/Why_positional_case_system I think your position is more persuasive, although I've never argued about this with an expert Lojbanist. I speculate that overall they feel that (1) there could be dozens or hundreds of different cases reflecting the role that a noun plays in a particular predicate relationship, (2) most don't apply at all to any given selbri [predicate relationship], (3) [an argument made in this page] some people's feeling of which case is appropriate for describing a particular relationship is culturally arbitrary, as they differ significantly from natural language to natural language, and (4) they don't want to force the use of prepositions, for concision.
- fusiongyro 10y ago
  Even with "just" transitive verbs taking three arguments, it's easy to lose the plot in natural languages like English. The idea that it wouldn't happen in unmarked sentences with two more seems absurd to me. They found some natural relationships that happened to have five members and they had five vowels, so the whole FA series came from that. Also the SE series only seems speakable for se/te. Beyond that, I think it would be a head-scratcher in English. I also think even if people are getting fluent in Lojban, they're unlikely to be gracefully handling free word order scenarios where the selbri shows up not in position 2. Are there SOV-speakers producing SOV-ish shaped Lojban and everybody is getting along fine? If you look at the official cmavo list above, FA and SE account for five words each, so ten of the 1000 in the list. Of course, they're pretty productive as prefixes too, so more than that. But I think part of what makes it complex is that there are so many of everything... forethought and afterthought of all the connectives; differently typed connectives for combining sumti, selbri, bridi; the non-obligatory tense/aspect/mood system that accounts for probably at least 40 particles; etc. It's a neat language. There's a lot of neat stuff in it. I'm sure it's fun to be able to speak and understand it. I guess mainly I think it doesn't speak to your innate language facility's strengths, so I doubt it can be done realtime as a language, as specified. It can clearly be done with our reasoning faculties. Maybe that's enough.
  - 10y ago