Lojban Wave Lessons/Introduction

From Lojban
Jump to navigation Jump to search

Lesson 0: Sounds

While you are (hopefully) eager to get started on the inner workings of Lojban grammar, a short lesson on the sounds and writing conventions of the language is beneficial. Learning a language only by reading is hard, and it's not easier if your internal voice is mispronouncing it.

For more details on vowels and consonants sounds used in Lojban, click on the letters described. They are pointing to Wikipedia articles which describe the sound and usually have an audio record of it.


There are five proper vowels in Lojban and one almost-vowel. First the proper ones:

a as in "father" or "large"
e as in "get" or "gem"
i as in "machine" or "scream" (not as in "hit")
o as in "bold" or "more" — not as in "so" (this should be a 'pure' sound.)
u as in "rude" or "due" (not as in "but")
y as "a" in "Tina" (not as in "but")

These are pretty much the same as vowels in Italian or Spanish. The sixth (almost-)vowel, y, is called a "schwa" in the language trade, and is pronounced as "comma", "taken" or "surprise!". It's the sound that comes out when the mouth is completely relaxed.

Two vowels together are pronounced as one sound (and called a "diphthong"). Some examples are:

ai as in "high"
au as in "how"
ei as in "hey"
oi as in "boy"
ia as in "yacht"
ie as in "yes"
iu as in "you"
ua as in "wander"
ue as in "wet"
uo as in "woke"
ui as in "we"

i and u act as semivowels when they precede another vowel. Double vowels are rare. The only examples are ii, which is pronounced like English "ye" (as in Oh come all ye faithful) or Chinese "yi", and uu, pronounced like "woo".

Note that because of variation between English dialects, approximations like these will necessarily be wrong for some speakers. The surest way to learn the phonology of the language is to imitate those who already speak it!


There are seventeen consonants in Lojban and one almost-consonant. The Lojban consonants are the same as the English, except that Lojban doesn't use the letters H, Q or W. Most of the consonants are pronounced like in English, but there are some exceptions:

g always "g" as in "gum", never "g" as in "gem"
c "sh", as in "ship"
j as in "measure" or French "bonjour"
x as in the exclamation "Ach!", or in German "Bach", Spanish "Jose" or Arabic "Khaled"

The almost-consonant is the apostrophe This letter is pronounced like the English letter H, but is only used between two vowels to prevent them from running into each other. Thus ui is normally pronounced "we", but u'i is "oohee". Groups of cmavo are called selma'o and are written using capital letters; but unexpectedly the capital version of this letter is h not H (ko'a is KOhA).

The combination dj is like the j in "joke". The combination tc is like the ch in "chat".

The following sound as they do in English, with the exception that r is often rolled (but does not have to be):

b, d, f, k, l, m, n, p, r, s, t, v, z.

Letter names

Speaking of the letters, what are the names for them? For example, when reciting the alphabet in English the letter C is pronounced "see". This is rather different than the sound the letter makes when used in a word. How about Lojban? Well, consonants are straightforward: The name of a consonant letter is the sound of that letter, plus y. So the consonant letters of Lojban, "b, c, d, f, g ...", are called by cy dy fy gy in Lojban. The almost-consonant ' is called .y'y (pronounced like an agreeing uh-huh but without the stress).

Vowels are handled by following the vowel sound with the word bu, which signifies we're speaking about a symbol. So the vowels of Lojban are: .abu .ebu .ibu .obu .ubu and .ybu.

Lastly you should know that stress is placed on the second-to-last syllable in words with more than one syllable, ignoring syllables containing y, and that one-syllable words are not stressed.

Correct pronunciation

You don't have to be very precise about Lojban pronunciation, because the phonemes are distributed so that it is hard to mistake one sound for another. This means that rather than one 'correct' pronunciation, there is a range of acceptable pronunciation—the general principle is that anything is OK so long as it doesn't sound too much like something else. For example, Lojban r can be pronounced like the "R" in English, Scottish or French.

Two things to be careful of, though, are pronouncing Lojban i and u like Standard British English "hit" and "but" (Northern English "but" is fine!). This is because non-Lojban vowels, particularly these two, are used to separate consonants by people who find them hard to say. For example, if you have problems spitting out the zd in zdani (house), you can say zɪdani — where the ɪ is very short, but the final i has to be long.

Writing Lojban

As you have already seen, Lojban uses the Latin alphabet, though various Lojbanists have suggested different, usually self-designed ones. Furthermore, Lojban almost always uses lower-case letters. Capital letters are only used to mark stress in proper names, but people tend to avoid them even in names.

Apart from the letters, some punctuation is used:

A full stop (period) is a glottal stop (the consonant in the middle of "uh-oh") or a short pause. The rules of Lojban make it easier for one word to run into another when the second word begins with a vowel; so any word starting with a vowel conventionally has a full stop placed in front of it. Full stops are not usually used to end sentences.

Commas are rare in Lojban. They are only used to emphasize boundaries between syllables, and adding or removing commas never changes a word's pronunciation or meaning.

The following are found writing styles of different Lojbanists, but they are not conventional:

Spaces are usually used between words. They are mandatory between some words (more on that in lesson thirteen). Double or triple space is sometimes used before the beginning of new sentences. This is to clearly mark sentence shift visually. This might compensate for lack of capital letters which are used for the same purpose in English.

In òrder to visuàlly reprèsent the stress on the penultìmate syllàble, and bècause they find it visuàlly plèasing, some pèople use grave or acute àccents òver the vòwel of those syllàbles.

Some people borrow other punctuation marks from English, even though they are not canon, and Lojban is equipped with actual words which should compensate for any punctuation one might want to use. Nonetheless, question marks, for example, clearly marks a sentence as a question and is much easier to catch with the eye than any word is, and so some Lojbanists use them. Quotation marks, parenthesis and exclamations marks can be used similarly. While this is not ungrammatical, since that doesn't interfere with the sentences, some people think exotic punctuation creates an unwanted difference between written and spoken Lojban, generally a big no-no.

Lojbanized foreign proper names (cmevla)

The following names are Lojbanized - their sounds are transcribed into Lojban and their ending sound have been changed to a consonant. The final consonant is necessary, because that's how foreign names are differentiated from Lojban words. Again, more on that in lesson thirteen.

Exercise 1

Where are these places?

  1. .nuIORK.
  2. .romas.
  3. .xavanas.
  4. .kardif.
  5. .beidjin.
  6. .ANkaras.
  7. .ALbekerkis.
  8. .vankuver.
  9. .keiptaun.
  10. .taibeis.
  11. .bon.
  12. .delis.
  13. .nis.
  14. .atinas.
  15. .lidz.
  16. .xelsinkis.


  1. New York: USA
  2. Rome: Italy
  3. Havana: Cuba
  4. Cardiff: Wales (The Welsh for "Cardiff" is "Caerdydd", which would Lojbanise to something like kairdyd..)
  5. Beijing: China
  6. Ankara: Turkey
  7. Albequerque: New Mexico, USA
  8. Vancouver: Canada
  9. Cape Town: South Africa
  10. Taipei: Taiwan (note b, not p. Although actually, the b in Pinyin is pronounced as a p... But this isn't meant to be a course on Mandarin!)
  11. Bonn: Germany
  12. Delhi: India (The Hindi for "Delhi" is "Dillî", which would give dilis. or dili'is..)
  13. Nice: France
  14. Athens: Greece ("Athina" in Greek)
  15. Leeds: England
  16. Helsinki: Finland

Exercise 2

Lojbanise the following names.

There are usually alternative spellings for names, either because people pronounce the originals differently, or because the exact sound doesn't exist in Lojban, so you need to choose between two Lojban letters. This doesn't matter, so long as everyone knows who or where you're talking about.

  1. John
  2. Melissa
  3. Amanda
  4. Matthew
  5. Michael
  6. David Bowie
  7. Jane Austen
  8. William Shakespeare
  9. Sigourney Weaver
  10. Richard Nixon
  11. Istanbul
  12. Madrid
  13. Tokyo
  14. San Salvador


  1. .djon. (or .djan. with some accents)
  2. .melisys.
  3. .amandys. (again, depending on your accent, the final y may be a, the initial a may be y, and the middle a may be e.)
  4. .matiius.
  5. .maikyl. or .maikl. , depending on how you say it.
  6. .deivyd.bauis. or .bouis.
  7. .djein.ostin.
  8. .uiliiam.cekspir.
  9. .sigornis.uivyr. or .sygornis.uivyr.
  10. .ritcyrd.niksyn.
  11. .istanBUL. with English stress, .IStanbul. with American, .istanbul. with Turkish. Lojbanists generally prefer to base cmevla on local pronunciation, but this is not an absolute rule.
  12. .maDRID.
  13. .tokiios.
  14. .san.salvaDOR. (with Spanish stress)

Lesson 1: Bridi, jufra, sumti and selbri

Bridi is the most central unit of Lojban utterances. The concept is very close to what we call a proposition in English. A bridi is a claim that some objects stand in a relation to each other, or that an object has some property. This stands in contrast to jufra, which are merely Lojban utterances, which can be bridi or anything else being said. The difference between a bridi and a jufra is that a jufra does not necessarily state anything, while a bridi does. Thus, a bridi might be true or false, while not all jufra can be said to be such.

To have some examples (in English, to begin with), Mozart was the greatest musician of all time is a bridi, because it makes a claim with a truth value, and it involves an object, Mozart, and a property, being the greatest musician of all time. On the contrary, Ow! My toe! is not a bridi, since it does not involve a relation, and thus does not state anything. Both, though, are jufra.

Try to identify the bridi among these English jufra:

  1. I hate it when you do that.
  2. Woah, that looks delicious!
  3. Geez, not again.
  4. No, I own three cars
  5. Nineteen minutes past eight.
  6. This Saturday, yes.

Answer: 1, 2 and 4 are bridi. The rest contain no relation or claim of a property.

Put in Lojban terms, a bridi consists of one selbri, and one or more sumti. The selbri is the relation or claim about the object, and the sumti are the objects which are in a relation. Note that object is not a perfect translation of sumti, since sumti can refer to not just physical objects, but can also purely abstract things like "The idea of warfare". A better translation would be something like "subject, direct or indirect object" for sumti, and main verb for selbri, though, as we will see, this is not optimal either.

We can now write the first important lesson down:

bridi = selbri + one or more sumti

Put another way, a bridi states that some sumti do/are something explained by a selbri.

Identify the sumti and selbri equivalents in these English jufra:

I will pick up my daughters with my car.

Answer: selbri: pick up (with). sumti: I, my daughters, my car

He bought 5 new shirts from Mark for just two hundred euro!

Answer: selbri: bought (from) (for) sumti: He, 5 new shirts, Mark and two hundred euros

Since these concepts are so fundamental to Lojban, let's have a third example:

So far, the EPA has done nothing about the amount of sulphur dioxide.

Answer: selbri: has done (about) sumti: The EPA, nothing and the amount of sulphur dioxide

Now try begin making Lojban bridi. For this you will need to use some words, which can act as selbri:

dunda = x1 gives x2 to x3 (without payment)
pelxu = x1 is yellow
zdani = x1 is a home of x2

Notice that these words meaning give, yellow and home would be considered a verb, an adjective and a noun in English. In Lojban, there are no such categories and no such distinction. dunda can be translated gives (verb), is a giver (noun), is giving (adjective) as well as to an adverb form. They all act as selbri, and are used in the same way.

As well as a few words, which can act as sumti:

mi = "I" or "we" – the one or those who are speaking
ti = "this" – a close thing or event nearby which can be pointed to by the speaker
do = "you" – the one or those who are being spoken to

See the strange translations of the selbri above - especially the x1, x2 and x3? Those are called sumti places. They are places where sumti can go to fill a bridi. Filling a sumti in a place states that the sumti fits in that place. The second place of dunda, for example, x2, is the thing being given. The third is the object which receives the thing. Notice also that the translation of dunda has the word to in it. This is because, while this word is needed in English to signify the receiver, the receiver is in the third sumti place of dunda. So when you fill the third sumti place of dunda, the sumti you fill in is always the receiver, and you don't need an equivalent to the word to!

To say a bridi, you simply say the x1 sumti first, then the selbri, then any other sumti.

Usual bridi: (x1 sumti) (selbri) (x2 sumti) (x3 sumti) (x4 sumti) (x5 sumti) (and so on)

The order can be played around with, but for now, we stick with the usual form. To say I give this to you you just say mi dunda ti do, with the three sumti at the right places.

So, how would you say This is a home of me?

Answer: ti zdani mi

Try a few more in order to get the idea of a place structure sink in.

You give this to me?

Answer: do dunda ti mi

And translate ti pelxu

Answer: This is yellow.

Quite easy once you get the hang of it, right?

Multiple bridi after each other are separated by .i This is the Lojban equivalent of full stop, but it usually goes before bridi instead of after them. It's often left out before the first bridi, though, as in all these examples:

.i = Sentence separator. Separates any two jufra (and therefore also bridi).

ti zdani mi .i ti pelxu This is a home to me. This is yellow.

Before you move on to the next lesson, I recommend that you take a break for at least seven minutes to let the information sink in.

Lesson 2: Skipping around with FA and zo'e

Most selbri have from one to five sumti places, but some have more. Here is a selbri with four sumti places:

vecnu = x1 sells x2 to x3 for price x4

If I want to say I sell this, it would be too much to have to fill the sumti places x3 and x4, which specify who I sell the thing to, and for what price. Luckily, I don't need to. sumti places can be filled with zo'e. zo'e indicates to us that the value of the sumti place is unspecified because it's unimportant or can be determined from context.

zo'e = something; fills a sumti place with something, but does not state what.

So to say I sell to you, I could say

mi vecnu zo'e do zo'e
I sell something to you for some price.

How would you say: That's a home (for somebody)?

Answer: ti zdani zo'e

How about (someone) gives this to (someone)?

Answer: zo'e dunda ti zo'e

Still, filling out three zo'e just to say that a thing is being sold takes time. Therefore you don't need to write all the zo'e in a bridi. The rule simply is that if you leave out any sumti, they will be considered as if they contained zo'e. If the bridi begins with a selbri, the x1 is presumed to be omitted and it becomes zo'e.

Try it out. What's Lojban for I sell?

Answer: mi vecnu

And what does zdani mi mean?

Answer: Something is a home of me or just I have a home.

As mentioned earlier, the form doesn't have to be {x1 sumti} {selbri} {x2 sumti} {x3 sumti} (ect.) In fact, you can place the selbri anywhere you want, just not at the beginning of the bridi. If you do that, the x1 is considered left out and filled with zo'e instead. So the following three jufra are all the exactly same bridi:

  • mi dunda ti do
  • mi ti dunda do
  • mi ti do dunda

Sometimes this is used for poetic effect. You sell yourself could be do do vecnu, which sounds better than do vecnu do. Or it can be used for clarity if the selbri is very long and therefore better be left at the end of the bridi.

There are also several ways to play around with the order of the sumti inside the bridi. The easiest one is by using the words fa, fe, fi, fo and fu.

fa = Tags the following sumti as filling x1
fe = Tags the following sumti as filling x2
fi = Tags the following sumti as filling x3
fo = Tags the following sumti as filling x4
fu = Tags the following sumti as filling x5

Notice that the vowels are the five vowels in the Lojban alphabet in order. Using one of these words marks that the next sumti will fill the x1, x2, x3, x4 and x5 respectively. The next sumti after that will be presumed to fill a slot one greater than the previous. To use an example:

dunda fa do fe ti doGiving by you of this thing to you. fa marks the x1, the giver, which is you. fe marks the thing being given, the x2. Sumti counting then continues from fe, meaning that the last sumti fills x3, the object receiving.

Attempt to translate the following three sentences:

mi vecnu fo ti fe do

Answer: I sell, for the price of this, you. or I sell you for the price of this (probably pointing to a bunch of money)

zdani fe ti

Answer: This has a home. Here, the fe is redundant.

vecnu zo'e mi ti fa do

Answer: You sell something to me for this price

Lesson 3: tanru and lo

In this lesson, you will become familiar with the concept of a tanru. A tanru is formed when a selbri is put in front of another selbri, modifying its meaning. A tanru is itself a selbri, and can combine with other selbri or tanru to form more complex tanru. Thus zdani vecnu is a tanru, as well as pelxu zdani vecnu, which is made from the tanru pelxu zdani and the single brivla word vecnu. To understand the concept of tanru, consider the English noun combination lemon tree. If you didn't know what a lemon tree was, but had heard about both lemons and trees, you would not be able to deduce what a lemon tree was. Perhaps a lemon-colored tree, or a tree shaped like a lemon, or a tree whose bark tastes like lemon. The only things you could know for sure would be that it would be a tree, and it would be lemon-like in some way.

A tanru is closely analogous to this. It cannot be said exactly what a zdani vecnu is, but it can be said that it is definitely a vecnu, and that it's zdani-like in some way. And it could be zdani-like in any way. In theory, no matter how silly or absurd the connection to zdani was, it could still truly be a zdani vecnu. However, it must actually be a vecnu in the ordinary sense in order for the tanru to apply. You could gloss zdani vecnu as home seller, or even better but worse sounding a home-type-of seller. The place structure of a tanru is always that of the rightmost selbri. It's also said that the left selbri modifies the right selbri.

"Really?", you'd ask, skeptically, "It doesn't matter how silly the connection to the left word in a tanru is, it's still true? So I could call all sellers for zdani vecnu and then make up some silly excuse why I think it's zdani-like?"

Well yes, but then you'd be a dick. Or at least you'd be intentionally misleading. In general, you should use a tanru when it's obvious how the left word relates to the right.

Attempt to translate the following: ti pelxu zdani do

Answer: That is a yellow home for you Again, we don't know in which way it's yellow. Probably it's painted yellow, but we don't know for sure.

mi vecnu dunda

Answer: I sell-like give. What can that mean? No idea. It certainly doesn't mean that you sold something, since, by definition of dunda, there can be no payment involved. It has to be a giveaway, but be sell-like in some aspect.

And now for something completely different. What if I wanted to say I sold to a German?

dotco = x1 is German/reflects German culture in aspect x2

I can't say mi vecnu zo'e dotco because that would leave two selbri in a bridi, which is not permitted. I could say mi dotco vecnu but that would be unnecessary vague - I could sell in a German way. Likewise, if I want to say I am friends with an American, what should I say?

pendo = x1 is a friend of x2
merko = x1 is American/reflect US culture in aspect x2

Again, the obvious would be to say mi pendo merko, but that would form a tanru, meaning I am friend-like American, which is wrong. What we really want to is to take the selbri merko and transform it into a sumti so it can be used in the selbri pendo. This is done by the two words lo and ku.

lo = generic begin convert selbri to sumti word. Extracts x1 of selbri to use as sumti.
ku = end convert selbri to sumti process.

You simply place a selbri between these two words, and it takes anything that can fill the x1 of that selbri and turns it into a sumti.

So for instance, the things that can fill zdani's x1 are only things which are homes of somebody. So lo zdani ku means a home or some homes for somebody. Similarly, if I say that something is pelxu, it means it's yellow. So lo pelxu ku refers to something yellow.

Now you have the necessary grammar to be able to say I am friends with an American. How?

Answer: mi pendo lo merko ku

There is a good reason why the ku is necessary. Try to translate A German sells this to me

Answer: lo dotco ku vecnu ti mi If you leave out the ku, you do not get a bridi, but simply three sumti. Since lo…ku cannot convert bridi, the ti is forced outside the sumti, the lo-construct is forced to close and it simply becomes the three sumti of lo dotco vecnu {ku}, ti and mi.

You always have to be careful with jufra like lo zdani ku pelxu. If the ku is left out the conversion process does not end, and it simply becomes one sumti, made from the tanru zdani pelxu and then converted with lo.

Lesson 4: Attitudinals

Another concept which can be unfamiliar to English speakers is that of attitudinals. Attitudinals are words that express emotions directly. They turned out to be incredibly awesome and useful. They all have a so-called free grammar, which means that they can appear almost anywhere within bridi without disrupting the bridi's grammar or any grammatical constructs.

In Lojban grammar, an attitudinal applies to the previous word. If that previous word is a word which begins a construct (like .i or lo), it applies to the entire construct. Likewise, if the attitudinal follows a word which ends a construct like ku, it applies to the ended construct.

Let's have two attitudinals to make some examples:

ui = attitudinal: simple pure emotion: happiness - unhappiness
za'a = attitudinal: evidential: I directly observe

Note that in the definition of ui, there are listed two emotions, happiness and unhappiness. This means that ui is defined as happiness, while its negation, means unhappiness. Negation might be the wrong word here. Technically, the other definition of ui is another construct, ui nai. Most of the time, the second definition of attitudinals - the ones suffixed with nai - really is the negation of the bare attitudinal. Other times, not so much.

nai = misc. negation - attached to attitudinals, it changes the meaning into the attitudinal's "negation"

And some more selbri, just for the heck of it:

citka = x1 eats x2
plise = x1 is an apple of strain/type x2

The sentence do citka lo plise ku ui, means You eat an apple, yay! (especially expressing that it is the apple that the speaker is happy about, not the eating, or the fact that it was you.) In the sentence do za'a citka lo plise ku, the speaker directly observes that it is indeed the you, who eats an apple as opposed to someone else.

If an attitudinal is placed at the beginning of the bridi, it is understood to apply to an explicit or implicit .i, thus applying to the entire bridi:

ui za'a do dunda lo plise ku miYay, I observe that you give an/some apple to me!

mi vecnu ui nai lo zdani ku I sell (which sucks!) a home.

Try it out with a few examples. First, though, here are some more attitudinals:

.u'u = attitudinal: simple pure emotion: guilt - remorselessness - innocence.
.oi = attitudinal: complex pure emotion: complaint - pleasure.
iu = attitudinal: miscellaneous pure emotion: love - hate.

Look at that, a word with three emotions in the definition! The middle one is accessed by suffixinng with cu'i. It's considered the midpoint of the emotion.

cu'i = attitudinal midpoint scalar: attach to attitudinal to change the meaning to the "midpoint" of the emotion.

Try saying I give something to a German, who I love

Answer: mi dunda fi lo dotco ku iu or zo'e instead of fi

Now Aah, I eat a yellow apple

Answer: .oi nai mi citka lo pelxu plise ku

Let's have another attitudinal of a different kind to illustrate something peculiar:

.ei = attitudinal: complex propositional emotion: obligation - freedom.

So, quite easy: I have to give the apple away is mi dunda .ei lo plise ku, right? It is, actually! When you think about it, that's weird. Why is it that all the other attitudinals we have seen so far expresses the speaker's feeling about the bridi, but this one actually changes what the bridi means? Surely, by saying I have to give the apple away, we say nothing about whether the apple actually is being given away. If I had used ui, however, I would actually have stated that I gave the apple away, and that I was happy about it. What's that all about?

This issue, exactly how attitudinals change the conditions on which a bridi is true, is a subject of a minor debate. The official, textbook rule, which probably won't be changed, is that there is a distinction between pure emotions and propositional emotions. Only propostional emotions can change the truth conditions, while pure emotions cannot. In order to express a propositional emotional attitudinal without changing the truth value of the bridi, you can just separate it from the bridi with .i. There is also a word for explicitly conserving or changing the truth conditions of a bridi:

da'i = attitudinal: discursive: supposing - in fact

Saying da'i in a bridi changes the truth conditions to hypothetical, which is the default using propositional attitudinals. Saying da'i nai changes the truth condition to the normal, which is default using pure attitudinals.

So, what's two ways of saying I give the apple away? (and feel obligation about it)

Answer: mi dunda lo plise ku .i .ei and mi dunda da'i nai .ei lo plise ku

The feeling of an attitudinal can be subscribed to someone else using dai. Usually in ordinary speech, the attitudinal is subscribed to the listener, but it doesn't have to be so. Also, because the word is glossed empathy (feeling others emotions), some Lojbanists mistakenly think that the speaker must share the emotion being subscribed to others.

dai = attitudinal modifier: empathy (subscribes attitudinal to someone else, unspecified)

Example: .u'i .oi dai citka ti - Ha ha, this was eaten! That must have hurt!

.u'i = attitudinal: simple pure emotion: amusement - weariness

What often used phrase could .oi .u'i dai mean?

Answer: Ouch, very funny.

And another one to test your knowledge: Try to translate He was sorry he sold a home (remembering, that tense is implied and need not be specified. Also, he could be obvious from context)

Answer: u'u dai vecnu lo zdani ku

Lastly, the intensity of an attitudinal can be specified using certain words. These can be used after an attitudinal, or an attitudinal with nai or cu'i suffixed. It's less clear what happens when you attach it to other words, like a selbri, but it's mostly understood as intensifying or weakening the selbri in some unspecified way:

Modifying word Intensity
cai Extreme
sai Strong
(none) Unspecified (medium)
ru'e Weak

What emotion is expressed using .u'i nai sai ?

Answer: Strong weariness

And how would you express that you are mildly remorseless?

Answer: .u'u cu'i ru'e

Lesson 5: Reordering places with SE

Before we venture into the territory of more complex constructs, you should learn another mechanism for reordering the sumti of a selbri. This, as we will show, is very useful for making description-like sumti (the kind of sumti with lo).

Consider the sentence I eat a gift, which might be appropriate if that gift is an apple. To translate this, it would seem natural to look up a selbri meaning gift before continuing. However, if one looks carefully at the definition of dunda, x1 gives x2 to x3, one realizes that the x2 of dunda is something given – a gift.

So, to express that sentence, we can't say mi citka lo dunda ku, because lo dunda ku would be the x1 of dunda, which is a donor of a gift. Cannibalism aside, we don't want to say that. What we want is a way to extract the x2 of a selbri.

This is one example where it is useful to use the word se. What se does is to modify a selbri such that the x1 and x2 of that selbri trade places. The construct of se + selbri is on its own considered one selbri. Let's try with an ordinary sentence:

ti se fanva mi = mi fanva ti
This is translated by me (= I translate this). [literally]
fanva = x1 translates x2 to language x3 from language x4 with result of translation x5

Often, but not always, bridi with se-constructs are translated to sentences with the passive voice, since the x1 is often the object taking action.

se has its own family of words. All of them swap a different place with the x1.

se swap x1 and x2
te swap x1 and x3
ve swap x1 and x4
xe swap x1 and x5

Note that s, t, v, and x are consecutive consonants in the Lojban alphabet.

So: Using this knowledge, what would ti xe fanva ti mean?

Answer: This is a translation of this (or fanva ti fu ti)

se and its family can of course be combined with fa and its family. The result can be very confusing indeed, if you wish to make it so:

klama = x1 travels/goes to x2 from x3 via x4 using x5 as transportation device

fo lo zdani ku te klama fe do ti fa mi = mi te klama do ti lo zdani ku and since te exchanges x1 and x3: = ti klama do mi lo zdani ku = This travels to you from me via a home.

Of course, no one would make such a sentence except to confuse people, or cruelly to test their understanding of Lojban grammar.

And thus, we have come to the point where we can say I eat a gift.. Simply exchange the sumti places of dunda to get the gift to be x1, then extract this new x1 with lo...ku. So, how would you say it?

One (possible) answer: mi citka lo se dunda ku

This shows one of the many uses for se and its family.