Lojban just got harder

« previous post | next post »

Matt Treyvaud forwarded this from the Lojban mailing list:

"Lojban changes to hanzi writing system" (4/1/15)

Some people complained that although the spelling in Lojban is very easy to grasp the grammar is not. So the committee for the development of Lojban (BPFK) decided to fix this issue and to make the spelling hard as well.  Especially for those people who are not familiar with hanzi (Chinese characters).

This is the new look of Lojban:

"我 爱 你" – new spelling.
"I love you" – meaning.
"mi prami do" – old spelling.

"者 猫 是 喝 者 乳" – new spelling.
"A cat drinks milk" – meaning.
"lo mlatu cu pinxe lo ladru" – old spelling.

Being familiar with Hanzi, I can make sense of the Hanzi writing of the first sentence, and I know what the second one is trying to say, but there must be some unusual grammatical rules in Lojban, because these are the meanings and parts of speech of the Hanzi:

者 nominalizer — "the one who / which"

猫 noun — "cat"

是 copula — "is"

喝 verb — "drink"

者 nominalizer — "the one who / which"*

乳 noun — "milk"


*I think this may be a mistake for 着 continuative particle — "-ing".

Hanzi are hard enough as is.  When they are wedded to Lojban grammar, the result is a daunting, maddening puzzle.

To avoid unnecessary grief and vexation, I should have posted this four days ago.


  1. Bob Sanders said,

    April 5, 2015 @ 12:08 pm

    I think they meant to use 着 for the durative particle, which is homophonous with 者. Wouldn't be the first time someone has failed to look up at the screen and hit the wrong selection key.

  2. Gleki said,

    April 5, 2015 @ 12:10 pm

    No, not a mistake. Literally, in Lojban we say "The one who is a cat drinks what is milk".

  3. ThomasH said,

    April 5, 2015 @ 12:15 pm

    Well, making it harder will help preserve English as a lingua franca, so we should cheer.

  4. Bruce Rusk said,

    April 5, 2015 @ 12:18 pm

    Surely they could have gone one better by following the Japanese Kanbun model of writing in Classical Chinese and adding annotations that allow one to reconstruct the Lojban syntax and pronunciation.

  5. John said,

    April 5, 2015 @ 12:24 pm

    It seemed to me that they're trying to make 者 stand in for an article of some sort; and sure enough, googling "lo lojban" gives me this page, which identifies lo and le as articles and tries to explain the rules for using them:


    Setting aside the weirdness of their choice of character, WHY would you choose to have articles in constructed language!? They're the hardest part of language to grasp if you come from a language that doesn't have them!

  6. Gleki said,

    April 5, 2015 @ 1:17 pm

    John, {lo} and {le} are only called articles. In fact {lo} is like English "-er". It just converts verbs to nouns.
    {sazri} = to drive.
    {lo sazri} = driver/driver.

    By default Lojban has no nouns. Everything is a verb. To get a noun you prefix the verb with {lo}.

    So choosing 者 was not a mistake. This is intentional.

    And yes, the original message was published on April 1.

  7. Nymiaidon said,

    April 5, 2015 @ 1:33 pm

    Did everyone catch that this was an April Fool's joke?

    @ThomasH: Lojban is not meant to be an auxlang, and it's not a common goal to make it "replace" other languages. On the other hand, your comment sounds rather dismissive.

    @John: You might be interested to hear that hardly anyone uses "le" anymore. Using "lo" is very much like saying a bare noun phrase in languages that don't have articles. There is no semantic difficulty there, it's only grammatically an article – more specifically it simply turns verbs into nouns, without any complicated article semantics.

  8. Paolo said,

    April 5, 2015 @ 1:40 pm

    Meanwhile in Germany Deutsche Grammatik wird radikal vereinfacht. ;)

  9. mudri said,

    April 5, 2015 @ 3:36 pm

    Bruce Rusk: someone has made a system that somewhat resembles modern Japanese. There's an example here: https://twitter.com/xsznix/status/554469268712980480. I believe that it's supposed to be read (in phonemic IPA) as: /la ʔʃogasʔ lo mi xanʃɛ sɛl.ʃiha ʃu mo/ (translated roughly as “What does Cogas [a Japanese Lojbanist] think of my handwriting?”). The interesting thing is the use of hiragana after kanji to disambiguate the form of content morphemes. Here, “寫は” signifies -ʃiha-, a combining form of {ʃiska} (which, I guess, would be written “寫か”). As far as I can tell, the combination of semantic character and final mora (which may be a lone consonant word-medially) is enough to uniquely specify a native content morpheme.

    A partial reference is available here: https://docs.google.com/spreadsheets/d/17V_Cir64IW0gu-JlynPwLjmukytSpOIl4b9K7MF5pN4/edit#gid=4835124. Gloss words are in Lojban, so it won't make much sense if you don't know Lojban vocabulary. The last sheet, though, gives a hiragana table (using standard Lojban orthography, which is not too different from IPA).

    John: It's difficult to conceive of a Lojban-like language without articles. Here are some options:

    – Do away with the idea of noun phrases. This is something I've tried myself, but it starts getting complicated when dealing with quantifiers. You essentially need double the quantifiers you would usually use depending on whether you're quantifier is working with respect to the “noun” or with respect to the “verb”. You have to distinguish between “All the dogs leap” and “All who leap are dogs”. Then dealing with polyadic predicates just gets confusing.

    – Have an explicit tanru-creation particle. If content words are juxtaposed in Lojban, they combine to make a tanru, representing a single predicate. For example, {xunre bolci} means “red ball”, where {xunre} means “red” and {bolci} means “ball”. However, this isn't enough. You would also need to fix word order to work out where the verb was. Lojban's flexible word order is seen as an advantage. It's common to see a mix of verb-first, verb-second and verb-third in actual Lojban, as well as sentences (and particularly clauses) with no nouns.

    – Have case markers to distinguish noun phrases from verb phrases. This is possible with FA tags (seeing place structure as a kind of case system), but this is essentially using case markers as articles, so not really an alternative.

    – Have a morphological distinction between nouns and verbs. This is fundamentally against Lojban's way of creating nouns (specified merely as something that fits a given place of a given predicate). Hence, this is not an option.

    Besides, choice of article does give some useful distinctions. For example, {lo prenu} refers to one or more people as individuals, whereas {loi prenu} refers to one or more groups of people, each group considered as a single entity. This is important when talking about carrying pianos, an activity generally done by a group of people working together, rather than multiple people each doing it alone.

    Finally, Lojban doesn't aim to be familiar or easy to learn. It is designed for simplicity, expressiveness and ease of (correct) use. The grammatical rules, as a consequence of simplicity, all have a sense of meaningfulness to them that make them easy to remember.

  10. Victor Mair said,

    April 6, 2015 @ 7:15 am

    If the Lojbanists want to make their writing really, really hard, not only should they switch to Chinese characters, they should interpret the characters as pictures of abstract ideas and logical processes ("symbols of abstract logic", or whatever), in line with the explanations of Westerners who encountered them before the development of rigorous philological approaches for dealing with morphosyllabic writing.

    Athanasius Kircher (1602-1680):


    …claimed that the Chinese were descended from the sons of Ham, that Confucius was Hermes Trismegistus/Moses and that the Chinese characters were abstracted hieroglyphs.

    In Kircher's system, ideograms were inferior to hieroglyphs because they referred to specific ideas rather than to mysterious complexes of ideas, while the signs of the Maya and Aztecs were yet lower pictograms which referred only to objects.


    Gottfried Wilhelm von Leibniz (1646-1716):


    Many characters well known in his day, including Egyptian hieroglyphics, Chinese characters, and the symbols of astronomy and chemistry, he deemed not real.[50] Instead, he proposed the creation of a characteristica universalis or "universal characteristic", built on an alphabet of human thought in which each fundamental concept would be represented by a unique "real" character:

    It is obvious that if we could find characters or signs suited for expressing all our thoughts as clearly and as exactly as arithmetic expresses numbers or geometry expresses lines, we could do in all matters insofar as they are subject to reasoning all that we can do in arithmetic and geometry. For all investigations which depend on reasoning would be carried out by transposing these characters and by a species of calculus.[51]



    Or take the Imagism of Florence Ayscough, Amy Lowell, Ezra Pound, and others, guided by Ernest Fenollosa (1853-1908), who were inspired by Chinese characters, but interpreted them in, shall we say, idiosyncratic ways.




  11. Victor Mair said,

    April 6, 2015 @ 9:22 am

    From a colleague:

    John Wilkins' Essay Toward a Real Character and Philosophical Language also drew inspiration from Jesuit (?) descriptions of Chinese writing, if I recall correctly. It's been a while, but I think David Mungello has a chapter on this in Curious Land: Jesuit Accommodation and the Origins of Sinology.

  12. J.W. Brewer said,

    April 6, 2015 @ 10:18 am

    It certainly does seem a major flaw in the logban project that it should use so arbitrary and illogical a script as the Latin alphabet.

  13. Dan Lufkin said,

    April 6, 2015 @ 10:48 am

    If Lojbanists want to stay logical, why don't they adopt Hangul? The shapes of the letters are supposed to reflect the mouth-shape for that sound. That's about as objective an alphabet as you can get. I recall that one of the old names for Hangul was "alphabet you can learn in a morning."

  14. Charles N said,

    April 6, 2015 @ 12:05 pm

    Clearly, the logical script for logban has to be the Shavian alphabet. Then someone could translate Shaw's Androcles and the Lion into logban as a companion to the Penguin Shavian edition (alas, now out of print).

  15. Nick said,

    April 7, 2015 @ 2:39 am

    Dr. Mair,

    Sorry to be off topic here…but I'm dismayed to find few reviews or discussions of William Hannas' Writing on the Wall. I wonder if you could offer some suggestions why the silence is so loud that it's almost burst my ear drums. Besides David Moser's thoughtful review I discern something like conspiracy of neglect.

    Many thanks for any reply!


  16. Gleki said,

    April 7, 2015 @ 3:43 am

    J.W. Brewer, yes, Latin isn't nice. It was chosen for obvious reasons (to simplify entering texts).

    Dan, in fact there was a proposal to use Hangul: http://lojban.org/papri/Hangul

  17. David Marjanović said,

    April 7, 2015 @ 6:44 am

    I recall that one of the old names for Hangul was "alphabet you can learn in a morning."

    "The wise man plagues himself with Chinese characters all his life, and the stupid man can never learn them. The wise man learns Hangul within a single morning, and the stupid man in ten days."

  18. Victor Mair said,

    April 7, 2015 @ 7:14 am

    @David Marjanović

    That's a great quote! Do you have a source for it, or is it just "proverbial wisdom"?

  19. Victor Mair said,

    April 7, 2015 @ 7:17 am


    When Bill Hannas's Writing on the Wall was first published, a top editor at UPenn Press told me that copies were "flying off the shelves". But then came this:


    Some people have said that the New York Times "did a job on Hannas". They sure did bring out some big guns against him.

    I still think that Writing on the Wall is an important book. I assign it in my "Language, Script, and Society in China" course, and I recommend it to others who are interested in the Chinese writing system and its relationship to education, creativity, science and technology, and so on.

    A follow-up to one aspect of Writing on the Wall is Chinese Industrial Espionage: Technology Acquisition and Military Modernisation (Asian Security Studies) by William C. Hannas, James Mulvenon, and Anna B. Puglisi.


  20. Mark S. said,

    April 7, 2015 @ 9:23 am

    Re. The Writing on the Wall: a selection on Chinese writing is available on my site. Google Books also offers extensive selections.

    Also well worth reading is Hannas' Asia's Orthographic Dilemma. The chapter on the supposed appropriateness of Chinese characters to East Asian Languages is on my site. Again, Google Books has much of the book on offer.

  21. J. W. Brewer said,

    April 7, 2015 @ 9:35 am

    Why use Hangul for Lojban when you could use Tengwar? http://mw.lojban.org/papri/tengwar

  22. J. W. Brewer said,

    April 7, 2015 @ 11:30 am

    Hannas apparently makes the bold claim that hangul suffers from the same creativity-inhibiting defects as traditional Chinese characters (I'm just getting that from the intro to his book, since the substantive chapter is not previewable for free). So the lojbanists may need to wrestle with that critique. That claim seems not only bold (which I'm using as a euphemism for a more pejorative description), but unnecessary, since if we assume the plausibility of Hanas' overall thesis, he could quite plausibly assert that the hanza->hangul transition in Korea, especially among the elite, is too recent to have overcome the cultural legacy of the ancien regime.

    On the other hand, if Hannas' deep claim is that the Latin alphabet promotes innovation and creativity precisely *because* of its arbitrariness and high level of abstraction (which could equally well be called its lack of any sort of internal logic), the logbanists may have more to worry about than the status of hangul.

  23. Jason said,

    April 7, 2015 @ 7:42 pm

    @Victor Mair

    The quote is from education minister Joeng In Ji's postscript to the Hunmin Jeongeum Haerye, the introductory manual to Hangul, published in 1443. Wikipedia gives this citation: Hunmin Jeongeum Haerye, postface of Jeong Inji, p. 27a, translation from Gari K. Ledyard, The Korean Language Reform of 1446.

  24. Victor Mair said,

    April 7, 2015 @ 9:45 pm

    Thanks very much, Jason. I've read Gari Ledyard's great dissertation and Joeng In Ji's postscript to the Hunmin Jeongeum Haerye, but it was so long ago that I had forgotten this marvelous, and thoroughly modern sounding, quotation

  25. Victor Mair said,

    April 7, 2015 @ 9:53 pm

    The New York Times made up for the 2003 broadside aimed at Writing on the Wall by publishing at least these three positive reviews of Hannas' work (and I think there may be others, but I can't find them right now) around a decade later:




  26. Eidolon said,

    April 8, 2015 @ 6:06 pm

    The trouble with talking about creativity is – even now – that there is no testable measure of creativity. A lot of people would consider the Japanese – who have long used Chinese characters – to be rather creative, for example, and judging by the amount of Nobel prizes they've managed to accrue, there's a valid argument there. But then there's the other side that refuses to accept it, and who believe the Japanese are, similar to the Chinese, only great at adapting existing technologies and making small improvements. The same argument has been made for Koreans, despite hangul.

    In my opinion, the jury is still out on what constitutes creativity and to what degree it is affected by orthography.

  27. Victor Mair said,

    April 8, 2015 @ 11:12 pm


    Can you imagine what a "testable measure of creativity" might be like?

    You say that, in your opinion "the jury is still out on what constitutes creativity and to what degree it is affected by orthography." Can you imagine that there will ever be a time when the jury is not out, when somebody will figure out "what constitutes creativity and to what degree it is affected by orthography"?

  28. Eidolon said,

    April 9, 2015 @ 5:54 pm

    "Can you imagine what a "testable measure of creativity" might be like?"

    The Torrance Tests of Creative Thinking beat me to that, though I understand there are various methodological issues with the test that await improvement. All in all, I am not too hopeful of a simple pen-and-paper test for creativity, because my own opinion about creativity is that it is highly contextual, domain/environment-driven, and even subjective. One of my most impressionable memories from graduate school is hearing fellows – and even professors – complain about the "low-hanging fruits" already being picked by the great and the famous, which is simply another way of saying that our standard for creativity gets harder over time; imagination is not an endless fountain after all.

    "Can you imagine that there will ever be a time when the jury is not out, when somebody will figure out "what constitutes creativity and to what degree it is affected by orthography"?"

    Strictly speaking? I think it's going to take a long time, and that the answer is going to come through neuroscience. Construct an experiment in which identical twins are taught English and Chinese orthography, respectively, and monitor their neural developments over a period. That's the best bet for finding out what orthography ultimately affects, and whether what it affects has to do with the regions of the brain that we use for creative work.

  29. Vanya said,

    April 10, 2015 @ 1:51 am

    If I understand Hannas correctly he is not just claiming that Chinese orthography is a problem, he is claiming that syllabaries (i.e. hangul or katakana) are also inferior to alphabets and restrict one's ability to make abstract connections between sound and symbol, thus harming creativity and fostering resistance to alien ideas. That seems to me far more radical than the usual complaints that Chinese orthography is inefficient and wasteful.

  30. Rodger C said,

    April 10, 2015 @ 7:07 am

    Since when is Hangul a syllabary? It's an alphabet chunked into syllables.

RSS feed for comments on this post