Skip to content

Instantly share code, notes, and snippets.



Last active Nov 8, 2020
What would you like to do?
The Lojban I speak

The Lojban I speak

by la lalxu (Lynn)

This is a list of ways in which my dialect of Lojban differs from the dialect described in The Complete Lojban Language. Most people on the #lojban IRC channel seem to accept and use most of these changes, as well.

The sections marked with an asterisk are approved as part of “base Lojban” by the BPFK (through a voting process). The other sections are simply unofficial additions that people (including me) like enough to use in daily conversation, because they make the language nicer and simpler.



In CLL Lojban, cmevla and fu'ivla are allowed to contain consonant-glide-vowel sequences like in kuin (Queen). Because it wasn’t clear where it was a good idea to allow them, the BPFK banned them everywhere! Now you can’t have a glide immediately after a consonant anymore, and you need to insert a vowel, writing kuuín (Ku-ween).


The comma pseudo-letter seems to have disappeared, to be replaced by glides…? I dunno, maybe some people still use it, but for e.g. o,a I just write oua.



The “veridicality” distinction between lo and le has been replaced by a simpler system where lo is a completely general article. I use lo mlatu to mean “a cat/some cats/the cat(s)”, and lo'e mlatu to mean “cats in general/stereotypically”. I personally never use le and le'e.

Double-dotted cmevla*

It used to be the case that cmevla were not allowed to contain la, lai and doi. (This is because there were only dots at the end of cmevla starting with consonants, and there was otherwise no way to distinguish .i ko co'e la lojban. from a long cmevla .ikoco'elalojban.)

Now, this restriction is lifted, and you must put dots before and after cmevla, like so: mi'e la .kiiulum.

Cmevla–brivla merger

A popular change in unofficial Lojban grammars is to just make cmevla a special kind of brivla: .alis. has the place structure “x₁ is Alice”. la .alis. still means “Alice”. lo .vailplum. means “a Vileplume/some Vileplumes”. I adopt this change.

It used to be the case that la .lojban. mo meant “what about Lojban?” but la lalxu mo was parsed as a tanru. Also, la lalxu ku was a valid sumti, but la .alis. ku was an error!

Now, cmevla no longer get special grammar. Because they’re just a kind of brivla, they can be used in tanru, such as in lo .ford. karce cu melbi. And la .alis. ku is now a valid sumti. It's far easier.

Simple connectives (jecu)

I adopt solpahi's simplified connective system.

It replaces all the tricky connectives with a simpler system: gi'e becomes jecu, .e becomes je, and (non-tanru) forethought connectives become gaje X gi Y. (Use e.g. naje and jenai for negations. And you can use ja/jo/ju instead, or even joi and jo'u and stuff!)

Plural logic

I’m still learning this myself… but solpahi has explained it very well here, and it’s an important change. Predicate variables (like the X in “for all X…”) can now refer to multiple things at the same time. The meaning of pa broda is changed, too.

It has to do with distributivity: when you say “five people gathered and dined”, then no one of them did any gathering, but each of them dined individually. Lojban is bad at expressing this kind of concept, and stumbles through it with sets; solpahi’s system is far simpler and gets rid of those clunky sets. It’s a very interesting read, but it gets a little technical!


Omitting dots

It’s common to omit dots in written, spaced Lojban: after all, their positions are simple to restore. If a word starts with a vowel, you know there’s an implicit stop (dot) there, and if it ends in a consonant, you know it should be surrounded with stops (dots). Thus, .i .u'i doi la .kiiulum. can be safely written as i u'i doi la kiiulum.

Marking stress

In the CLL book, unusual stress is marked with capital letters: the name Nicole would be written .niKOL. or .nikOl. This is pretty ugly, so I prefer using an acute accent to mark stress: .nikól. (Other people seem to use grave accents; it doesn’t really matter which you pick.)


I use tildes to draw out my vowels for comic effect. .i da~rno is like “it’s so faaaar!” This isn’t super standard or anything.


I freely use commas, dashes, exclamation marks, and cutesy emoticons throughout my Lojban writing. Some people don’t like that, shrug. I almost always write to … toi as (to … toi), and lu … li'u as lu «…» li'u, for clarity, which many others do as well.


Gismu fixes

  • The gismu list uses “is a quantity of/contains/is made of” where modern Lojbanists mentally substitute “is”. So djacu just means “x₁ is water”.
  • xruti has a popular redefinition: x₁ returns to state x₂ (ka) from state x₃ (ka). The old definition is then equivalent to xrugau.
  • dukse is weird. People mostly seem to use this definition instead: x₁ is too x₂ (ka) for x₃ (nu) to happen.
  • The x₂ place of stagi makes no sense, and the IRC consensus is “pretend it doesn’t exist”.
  • cinse is too overloaded. Use lingeni for gender, firca for flirting/courtship, gletu for intercourse/mating, glotu for having sex, xrotu for affection. (cinse’s meaning of “sexual orientation” seems to be the obvious candidate to keep; then words like mitcinse (gay) and noncinse (asexual) still make sense. A proposed redefinition by solpahi is: “x₁ exhibits sexual orientation x₂ (ka)”.)
  • The cultural gismu are sort of shunned, as they violate cultural neutrality (only some cultures get them and others don’t). Prefer ISO-code words like gugdefuru and bancumunu.
  • God gismu are a mess. I’ll shut up now

Experimental words

Experimental cmavo that I use often

  • solpahi’s selma'o NOIhA for adverbs: noi'a, poi'a, soi'a, soi, noi'o'a, poi'o'a.
  • Emotions: mi'au (meow, :3 attitudinal), mau'u (mmmmm~…), ue'i (excitement), au'u (“aww, cuuuute”!), uu'i (schadenfreude), uau (wow!), bu'a'a (evil laugh), xy'y (hmmmm…)
  • Emotion-over-time: xu'o. (ii = scared; iixu'o = becoming more scared over time, iixu'osai = becoming rapidly more scared over time… This one is actually my own invention!)
  • Vocatives: di'ai (well-wish), fau'u (forgive me!)
  • Discursives: si'au (it seems), moi'i (I remember this fact), vei'i (I recall experiencing this), fu'au (luckily), xa'i (roleplaying/“in asterisks”)
  • Tenses: xa'o (already, yet; counterpart to za'o), (PA) de'ei (on the nth day from today — so “node'ei” is today, “pade'ei” is tomorrow, “ni'ucide'ei” is three days ago)
  • Tags: xau (xamgu modal / with beneficiary…), ka'ai (kansa modal / with companion…)
  • ko'oi (imperative marker for things other than “you!”: ko equals do ko'oi)
  • Making silly sounds: sa'ei marks the next cmevla as an onomatopoeia (sa'ei bam!!), ki'ai marks the next cmevla as a nonce attitudinal (ki'ai pfff…)
  • (lo'ai X) (sa'ai Y) le'ai: “oops, I meant Y instead of X” attitudinal. Either part in parentheses can be dropped and left up to context. Useful with le'aipei at the end too (“did you mean to write…?”)

Experimental gismu that I use often

Word Definition
kibro x₁ pertains to the Internet
laldo x₁ (person) is old to x₂
tende x₁ tends to do x₂
tsuku x₁ arrives at x₂

Gismu for non-binary people

Word Definition
nunmu x₁ is a non-binary person
paznu x₁ is a non-binary child of x₂
pentu x₁ is a non-binary sibling of x₂
preri x₁ is a non-binary parent of x₂
vepre x₁ is an enby (young non-binary person)

Cute, extremely important words I defined

Word Definition
tsundere x₁ is tsundere towards x₂
vrejai x₁ and x₂ cuddle
altaisakta x₁ is a marshmallow
satyjdu x₁ is a pudding/jello (sugary gelatin dessert) containing x₂

This comment has been minimized.

Copy link

@snan snan commented Aug 7, 2020

The paragraph on glides isn't clear on what's CLL lojban, what's BPFK lojban, and what's your lojban. What's your take on what's allowed and what's disallowed?
And what's the sitch on the whole lo / le thing, that always used to trip me up bitd

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.