diagrammed Summary of Lojban Grammar: Difference between revisions

From Lojban
Jump to navigation Jump to search
No edit summary
 
m (Conversion script moved page Diagrammed Summary of Lojban Grammar to diagrammed Summary of Lojban Grammar: Converting page titles to lowercase)
 
(One intermediate revision by one other user not shown)
Line 1: Line 1:
originally placed at http://www.lojban.org/publications/brochures/lojex.txt
<pre>


      Diagrammed Summary of Lojban Grammar Forms with Example Sentences
        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
Permission granted to reproduce for no charge to the recipient, for purposes
                          of promotion of Lojban.
                              Updated April 1992
These pages give diagrammed examples of basic Lojban sentence structures.  The
most general pattern is covered first, followed by successive variations on
the basic components of the Lojban sentence.  There are many more capabilities
not covered in these examples.  A brief pronunciation guide and Lojban
glossary will be found at the end of this paper.
                      I. THE LOJBAN SENTENCE STRUCTURE
A Lojban sentence expresses a relationship (bridi), normally claiming that the
relationship holds (that it is 'true').  A bridi relationship consists of sev-
eral ideas or objects called arguments (sumti), which are related by a
predicate relation (selbri).  The following uses the Lojban terms bridi,
sumti, and selbri, because it is best to come to understand them independent
of the English associations of the corresponding words.
Some words used as sumti:
  mi    I/me/my, we/us/our
  do    you/your
  ti    this/this-here/this one's, these/these here/these ones'
  ta    that/that-there/that one's, those/those ones'
  tu    that yonder, those yonder
  zo'e  unspecified value (used when a sumti is unimportant or obvious)
  sumti are not specific as to number (singular or plural), nor gender
  (masculine/feminine/neutral).  Such distinctions can be optionally added.
  Names may be expressed as sumti, labelled with "la":
  la meris.  the one/ones named Mary
  la djan.    the one/ones named John
  (Other Lojban spelling versions are possible for names from other lan-
  guages.)
Some words used to indicate selbri relations:
  vecnu  x1 (seller) sells x2 (goods) to x3 (buyer) for x4 (price)
  tavla  x1 (talker) talks to x2 (audience) about x3 (topic) in language x4
  blari'o x1 (object/light source) is blue-green
  melbi  x1 (object/idea) is beautiful to x2 (observer) by standard x3
  We will describe these and other possible sumti and selbri in more detail
  below.
�Conventions:
  The following conventions will be used to show the structure of Lojban sen-
  tences in diagrams:
  - The selbri relation will be double-underlined.
  - The sumti arguments will be single-underlined.
  - Optional separator/terminator words are placed in square brackets.  They
  may be omitted if so bracketed. The general rule is that these may be omit-
  ted if and only if no grammatical ambiguity results.  Each such word serves
  as an end marker for particular structures, making the overall structure of
  the sentence very clear.
Basic structure of a Lojban sentence:
      sumti sumti ... sumti [cu] selbri sumti sumti ... sumti [vau]
      ----- -----    -----      ====== ----- -----    -----
  - Normally, there must be at least one sumti before the selbri.
  - Each selbri relation has a specifically defined place structure that de-
  fines the role of each sumti in the bridi relationship, based on its posi-
  tion in order.  In the examples above, that order was expressed by la-
  belling the positions x1, x2, x3, and x4.
  - "cu" acts as a separator after at least one preceding sumti to clearly
  mark the selbri.  As the diagram indicates, it may often be omitted.  There
  will be examples below.
  - "vau" goes at the end of the sentence, indicating that no more sumti will
  follow.  It usually may be omitted.
Sentence examples:
      mi      [cu] vecnu  ti            ta      zo'e    [vau]
      --            =====  --            --      ----
      seller-x1    sells  goods-sold-x2 buyer-x3 price-x4
      I            sell  this          to that  for some price.
      I sell this-thing/these-things to that-buyer/those-buyers.  (The price
      is obvious or unimportant.)
  - Both the "cu" and the "vau" are optional in this example and could be
  omitted:
  - When an unspecified sumti ("zo'e") is at the end of a sentence, it may be
  omitted.
  - Normally, there will be one sumti (the x1) before the selbri.  There may
  be more than one:
      mi        ti          [cu] vecnu ta      [vau]
      --        --                ===== --
      seller-x1 goods-sold-x2    sells buyer-x3
      I        this              sell  to that.
      Translates as stilted or poetic English:
      I, this thing, do sell to that buyer.
�  Usually, more than one sumti before the selbri will be for style or for em-
  phasis on the sumti that are out-of-place from their normal position.
  (Native speakers of languages other than English may prefer such orders.)
Observatives:
  If there is no sumti before the selbri, then it is understood that the x1
  sumti value is equivalent to "zo'e"; i.e. unimportant or obvious, and
  therefore omitted.  Any sumti after the selbri start counting from x2, x3,
  x4 ...:
      ta            [cu] melbi      [vau]
      --                =====
      object/idea-x1    is-beautiful    (to someone by some standard)
      That/Those        is/are beautiful.
      That is beautiful.  (or)  Those are beautiful.
  When the x1 is omitted:
                    melbi        [vau]
      ------------  =====
      (Unspecified) is-beautiful (to someone by some standard)
      melbi
      Beautiful!  (or)  It's beautiful!
  Omitting the x1 adds emphasis to the selbri relation, which has become
  first and foremost in the sentence.  This kind of sentence is termed an ob-
  servative, because it is usually stated by someone when they first observe
  or take note of the relation, and wish to quickly communicate it to someone
  else.  Commonly understood English observatives include "Smoke!" upon see-
  ing smoke or smelling the odor, or "Car!" to a person crossing the street
  who might be in danger.  Any Lojban selbri can be an observative if no
  sumti appear before the selbri.
  "cu" does not occur in an observative; "cu" is a separator, and there must
  be a sumti before the selbri that needs to be kept separate, for it to be
  used.  With no sumti preceding the selbri, "cu" is not permitted.
True/false (yes/no) questions (the word "xu"):
  xu question sentence (Is-it-true-that...):
      xu          mi [cu] vecnu ti  ta  [vau]
                  --      ===== --  --
      Is-it-true? I      sell  this that.
      Is-it-true-that I sell this to that?
  "xu" has a very unrestricted grammar, and is permitted virtually anywhere
  in a sentence.  At the beginning of the sentence, "xu" asks about the truth
  of the bridi relationship.  Elsewhere, in a sentence, "xu" attaches to the
  immediately preceding word (or the structure implied by that preceding
  word, when it is the marker for a structure).  Thus, also after the "vau"
  ending the sentence, "xu" asks about the entire bridi (the "vau" cannot be
  omitted if "xu" is to appear 'after' it).
�  "xu" appearing after a sumti questions whether the bridi relationship ex-
  pressed by the sentence is true, in particular for that sumti value:
      mi [cu] vecnu ti  xu          ta  [vau]
      --      ===== --  <-          --
      I      sell  this Is-it-true? that.
      Is-it-true-that I sell THIS (as opposed to something else) to that?
  Similarly, "xu" following "vecnu" in the above example would question the
  truth of the bridi relationship by specifically asking whether "sell" is a
  true relation between the sumti.
  We will discuss how to answer a "xu" true/false question below in the sec-
  tion on selbri.
Varying sumti order:
  There are ways to vary the order of sumti from the numerical order speci-
  fied by the place structure.  A sumti may be placed out of numerical order
  by labelling it in front with a tag indicating the actual place structure
  numerical position of the sumti.  The structure is thus of the form "fi
  sumti" (where the "fi" category word shows which of the existing sumti
  places is being used, by number:
  fa/1st sumti - x1      fe/2nd sumti - x2      fi/3rd sumti - x3
  fo/4th sumti - x4      fu/5th sumti - x5
  One reason for using these tags is to skip a place structure place without
  having to insert a "zo'e" for each skipped place:
      mi    [cu] tavla fo  la lojban. [vau]
      --        ===== .>  ----------
      talker    talk  x4= language
      I          talk  in-language Lojban.
      I talk in Lojban (to someone about some topic).
  which is equivalent to:
      mi    [cu] tavla zo'e          zo'e              la lojban. [vau]
      --        ===== ----          ----              ----------
      talker    talk  audience      topic            language
      I          talk  to unspecified about unspecified in Lojban.
      I talk in Lojban (to someone about some topic).
  After a "fi" tag sets the place number, any later sumti places continue the
  numbering consecutively:
      mi    [cu] tavla fi ta      la lojban. [vau]
      --        ===== .> --      ----------
      talker    talk  x3=topic  language-x4
      I          talk  about that in-language Lojban.
      I talk about that in Lojban (to someone unspecified).
�  Another reason to use "fi" tags is to change emphasis; listeners focus most
  closely on the sumti at the beginning of a sentence.
      fi la lojban. [cu] tavla fa mi    do        [vau]
      .> ----------      ===== .> --    --
      x3=topic          talk  x1=talker audience-x2
      About Lojban      talk  I        to you.
      It's Lojban, that I talk to you about (in an unspecified language).
  There are other ways of rearranging the sumti of a sentence that will be
  discussed below.
  Note that in all examples where a sumti is omitted, the Lojbanist remembers
  that there is an unspoken and unspecified value for each of the omitted
  place structure places.
  An observative can be formed by using "fa" to move the first (x1) sumti to
  the position after the selbri.
      melbi      fa ti  [vau]
      =====      .> --
      beautiful  x1=this
      Beautiful  this-is (to someone by some standard).
      Beautiful, it is!
Basic structure of an utterance:
  People don't always say just one sentence.  Lojban has a specific structure
  for talk or writing that is longer than one sentence.  The entirety of a
  given speech event or written text is called an utterance.
      sentence .i sentence .i sentence [...] .i sentence
      ni'o sentence .i sentence .i sentence .i sentence [...] .i sentence
      ni'o sentence .i sentence [...] .i sentence
      [fa'o]
  "ni'o" separates paragraphs (covering different topics of discussion).  In
  a long text or utterance, the topical structure of the text may be indi-
  cated by multiple "ni'o"s, with perhaps "ni'oni'oni'o" used to indicate a
  chapter, "ni'oni'o" to indicate a section, and a single "ni'o" to indicate
  a subtopic corresponding to a single English paragraph.
  ".i" separates sentences.  ".i" is sometimes compounded with words that
  modify the exact meaning (the semantics) of the sentence in the context of
  the utterance.  ("xu", discussed above, is one such word - it turns the
  sentence from a statement to a question of truth.)  When more than one per-
  son is talking, a new speaker will usually omit the ".i" even though she/he
  may be continuing on the same topic.  It is still OK for a new speaker to
  say the ".i" before continuing; indeed it is encouraged for maximum clarity
  (since it is possible the 2nd speaker might merely be adding words onto the
  end of the first speaker's sentence).  A good translation for ".i" is the
  "and" used in run-on sentences when people are talking informally:  "I did
  this, and then I did that, and ..., and ...".
  "fa'o" is an optional end-of-utterance marker, used primarily in computer
  input.  It is not needed in human speech.
�  You may now see why the "vau" at the end of the sentence can generally be
  omitted.  Since the following word will usually be an ".i" or a "ni'o"
  starting a new sentence or paragraph, there is no possibility of ambiguity.
  These separators prevent the sumti at the beginning of the next sentence
  from being mistaken as a trailing sumti of this sentence.
Punctuation:
  Note that Lojban has no mandatory punctuation marks.  Because Lojban speech
  EXACTLY matches the written text representing that speech, all
  'punctuation' that is used in English to show sentence structure, ques-
  tions, exclamations or tone of voice, and even quotations must be expressed
  in Lojban as actual words that are spoken.
  The apostrophe (') is a letter representing a sound in the language.
  The period (.) is an optional reminder to the reader that a pause (which
  may be very short) is required between sounds at that point.  These may be
  omitted, but are almost always printed.
  Capitalization indicates unusual stress in the pronunciation of a name from
  a language other than Lojban.  It is NOT used at the beginning of a sen-
  tence.
  Commas (,) indicate unusual syllable breaks in pronunciation of a name or
  other word.
  There are some optional conventions that allow certain punctuation symbols
  to appear to clarify printed text, making it easier to read.  These punctu-
  ation symbols ALWAYS appear associated with the printed word representing
  that punctuation symbol.  Thus a "xu" question may be marked with a ques-
  tion mark immediately after the "xu" (or immediately before the "xu", pos-
  sibly inverted, like the Spanish punctuation style).  Other questions may
  similarly be marked with a question mark after the word indicating the
  question - NOT at the end of the sentence.  There are words that may be as-
  sociated with exclamation points, start of quotation (represented by "<<")
  and end of quotation (">>").
                II. THE BASIC COMPONENTS (sumti AND selbri):
We now discuss the substructures of the basic components that make up a sen-
tence.  Any variety of selbri may be placed in a sentence, or another
substructure below that mentions 'selbri'.  Likewise, any variety of sumti may
be placed in a sentence, or another substructure below that mentions 'sumti'.
You may see that this can optionally lead to extremely complicated structures
nested within one another.  Lojban's unambiguous grammar allows even these
most complicated structures to be untangled in only one way.
                              A. SIMPLE sumti
sumti are not specific as to number (singular or plural), nor gender
(masculine/feminine/neutral).  Such distinctions can be optionally expressed
by being specific.
      Diagrammed Summary of Lojban Grammar Forms with Example Sentences
        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
Permission granted to reproduce for no charge to the recipient, for purposes
                          of promotion of Lojban.
                              Updated April 1992
�"pronoun" sumti:
  These include the single-word sumti given above:
      mi            I/we
      do            you
      ti            this/these
      ta            that/those
      tu            that/those-yonder
      zo'e          something unspecified (it's either obvious or unimportant)
  Some other words in this category include:
      ri            he/she/it (the-last-referenced-sumti)
      ko            imperative you
      ko'a          it/he/she/they (a specific value)
      di'u          it/this (the last sentence)
  There are many others, each with a particular meaning.  For example, there
  are 9 other words related to "ko'a".  Each may be used to represent a sepa-
  rate value of "it".  Since Lojban has no gender or number, these 10 words
  represent "he", "she", and "they" as well, and it becomes more clear why so
  many are needed.
  "zo'e" is a place-filler allowing you to skip over a sumti place in the or-
  dered place structure without specifying a value.  The speaker indicates
  that there is a value, but that it is not important to specify it, or that
  the speaker thinks it is obvious given the context.
      do [cu] tavla zo'e  ti        [vau]
      --      ===== ----  --
      You    talk        about this.
      You talk about this. (to someone, in some language)
  "ri" is a quick back-reference sumti.  It may have a new ad-hoc meaning ev-
  ery time it occurs.  The rules for counting back to 'the last sumti' in-
  clude some special cases that can't be covered in this summary, but in most
  simple sentences, the referent will be obvious.  There are two other back-
  referencing sumti of this type.
      mi [cu] tavla fi la lojban.  ri              [vau]
      --      ===== .> ----------  --
      I      talk  x3=about-Lojban in-it(Lojban)-x4.
      I talk about Lojban in Lojban (to someone unspecified.)
  "ko" is used to express commands.  A statement with "ko" can be interpreted
  by replacing the "ko" with "do", and then taking the result as a command to
  the listener to make the sentence true for himself/herself as "do":
      ko        [cu] tavla mi  [vau]
      --            ===== --
      You (imp.)    talk  to-me.
      Talk to me!
  is the command equivalent of:
      do        [cu] tavla mi  [vau]
      --            ===== --
      You            talk  to-me.
      You talk to me.
�  "ko" need not be in the first position, but rather can occur anywhere a
  sumti is allowed, leading to possible Lojban commands that are very unlike
  English commands:
      mi [cu] tavla ko                  [vau]
      --      ===== --
      I      talk  to-you (imperative).
      Let me talk (to you)!
  "ko" is even permitted to occur in more than one place in a sentence, al-
  lowing for meaning-rich commands like:
      ko        [cu] tavla ko                [vau]
      --            ===== --
      You (imp.)    talk  to-you (imperative).
      Talk to yourself!
  The English misses some of the meaning, since the Lojban expresses two com-
  mands at once: that the listener talk to himself, but also that the lis-
  tener allow himself to be talked-to (by himself).
Names ("la name"):
  Lojban names always end with a consonant followed by a mandatory pause
  (which may be very short).  No other Lojban word ends with a consonant.
  Thus names are easily recognized by both their form, and by being marked
  with "la".
      do [cu] tavla  la .an. ti        [vau]
      --      ======= ------- --
      You    talk-to Ann    about this.
      You talk to Ann about this.
question sumti ("ma"):
  "ma" indicates a question about the value of a sumti.  It is answered by
  'filling in the blank', replacing the "ma" with the intended sumti value.
  It can be translated as "Who?" or "What?" in most cases, but also serves
  for "When?", "Where?", and "Why? when used in sumti places that express
  time, location, or cause.
      ma  [cu] tavla do    [vau]
      --      ===== --
      __?      talks to-you.
      What/Who talks to you?
  is answerable by:
      mi [cu] tavla do [vau]
      --      ===== --
      I talk to you.
  Like "ko", "ma" can occur in any position where a sumti is allowed, not
  just in the first position.:
      do [cu] tavla ma [vau]
      --      ===== --
      You talk to what/whom?
�  "ma" can also appear in multiple sumti positions in one sentence, in effect
  asking several questions at once.
      ma [cu] tavla ma [vau]
      --      ===== --
      What/Who talks to what/whom?
  The two separate "ma" positions ask two separate questions, and can there-
  fore be answered with different values in each sumti place.
Description sumti ("le selbri [ku]"):
  "le" describes a sumti that the speaker has in mind by describing a bridi
  relationship that it forms the first (x1) sumti of.  This bridi is repre-
  sented by a selbri (e.g., one of the words described earlier in this pa-
  per).  Description sumti phrases have a terminator on the right side, "ku",
  which is omitted when no ambiguity results.
      mi [cu] tavla  le vecnu [ku] le blari'o [ku]          [vau]
      --      =====  ------------- ---------------
      I      talk-to the seller    about the blue-green-thing.
  "le vecnu" takes the selbri "vecnu", which has the "seller" in the x1
  place, and uses it in this sentence to describe a particular "seller" that
  the speaker has in mind (one that she probably expects the listener will
  also know about).  Similarly, the speaker has a particular blue-green thing
  in mind, which is described using "le" to mark "blari'o", a selbri whose
  first sumti is something blue-green.
  There are many variations on "le sumti [ku]" constructs, but to discuss
  them, we must first discuss the more complex structures of selbri.
                            B. selbri STRUCTURE
Though Lojban sentences often translate word-for-word into fairly clear En-
glish, selbri relations are actually quite unlike English.  The selbri "bajra"
expresses a relation of running.
      sumti  bajra    sumti      sumti      sumti
      -----  =====    -----      -----      -----
      runner [running] on-surface using-limbs with-gait
  In some sentence positions, "bajra" might be interpreted as the 'verb' "to
  run"; in other positions, "running".  In "le sumti [ku]", described in the
  preceding section, it represents the 'noun' interpretation of its x1 sumti
  place:  "runner".  (Some English words, like "cook", have similar proper-
  ties, but the analogy is weak.)
brivla:
  The simplest form of selbri is an individual word.  A word which may by it-
  self express a selbri relation is called a brivla.  The three types of
  brivla are gismu (root words), lujvo (compounds), and le'avla (borrowings
  from other languages).  All have identical grammar; they are allowed wher-
  ever "selbri" appear in these examples.
�  gismu:
      mi  [cu] klama ti          zo'e  zo'e  ta  [vau]
      --        ===== --          ----  ----  --
      Go-er    goes  destination origin route means.
      I go here (to this) using that means. (from somewhere via some route).
  lujvo:
      ta  [cu] blari'o [vau]
      --      =======
      That    is-blue-green.
  le'avla:
      ti  [cu] djarspageti [vau]
      --      ===========
      This    is-spaghetti.
  Some short words may serve as selbri, acting as variables that stand for
  another selbri.  The most commonly used of these is "go'i", which repre-
  sents the main bridi of the previous Lojban sentence, with any new sumti or
  other sentence features being expressed replacing the previously expressed
  ones.  Thus, in this context:
      ta  [cu] go'i                  [vau]
      --      ====
      That    too/same-as-last selbri.
      That (is spaghetti), too.
  The word "go'i", by itself, by repeating a bridi marked as a "xu"
  true/false question, repeats that bridi, thereby claiming it is true.
  Thus, in this sense only, "go'i" can mean "yes".  "xu" questions can also
  be answered "yes" by repeating the entire sentence in full, but "go'i" is
  much easier to say.
      xu ta [cu] blari'o [vau]
        --      =======
      Is-it-true-that that is-blue-green?
      go'i                  [vau]
      ====
      True.  (repeats "That is blue green.")
contradictory negation ("na selbri"):
  The negation particle "na" can occur at the beginning of any selbri.  It
  says that the relation claimed by the selbri does not hold (contradictory
  negation).  It may often be translated as "It is false that [English sen-
  tence]".
      mi [cu] na klama ti ta [vau]
      --      nn>===== -- --
      It is false that I go to this from that.
  The "na" is only permitted at the beginning of a complete selbri.  It is
  considered part of the selbri in other constructs in the language, but is
  restricted from other positions within a tanru (discussed below).
�  If the contradictory negation particle na precedes "go'i", the combination
  "na go'i" denies the relation claimed by "go'i".  Thus, after a "xu"
  true/false question, "na go'i" expresses the answer "False", or "No".
      mi [cu] ja'a go'i [vau]
      --      pppp>====
      It is TRUE that I do (go to this from that).
  If you were to use "go'i" after a sentence that contained a "na" contradic-
  tory negation, the negation would carry over to the repeated sentence.  Un-
  like English, "na go'i" would NOT form a double negative; it merely re-
  places the "na" by another "na" leaving the sentence unchanged.  Instead,
  you must cancel a negative by using the positive equivalent of "na"
  ("ja'a") to replace the previous sentence "na":
scalar negation ("na'e-word brivla"):
  "na" deals primarily with the truth or falsity of a bridi.  Lojban also
  supports a separate form of negation, called contrary or scalar negation.
  A scalar negation attaches tightly to the next brivla of the selbri, modi-
  fying the meaning of the word on some scale.  Scalar negation structures
  may appear anywhere where 'brivla' or 'selbri' is allowed.  Scalar negation
  words include "na'e" (other-than), "to'e" (absolute opposite-of), and
  "no'e" (neutral on the scale); "je'a" is a strong positive scale assertion,
  translating roughly as "certainly" or "indeed":
    |----------- POSITIVE ------------|------------ NEGATIVE -----------|
  je'a                              no'e            na'e            to'e
  Examples:
      mi [cu] melbi [vau]
      --      =====
      I am beautiful.
      mi [cu] na'e melbi [vau]
      --      ==========
      I am other-than beautiful.
      mi [cu] to'e melbi [vau]
      --      ==========
      I am ugly/opposite-of-beautiful.
      mi [cu] no'e melbi [vau]
      --      ==========
      I am plain/neutral on the beauty-ugliness scale.
      mi [cu] je'a melbi [vau]
      --      ==========
      I am indeed beautiful.
question selbri ("mo"):
  "mo", like its sumti relative "ma", is a fill-in-the-blank question.  It
  asks the respondent to provide a selbri that would be a true relation if
  inserted in place of the "mo":
      do [cu] mo [vau]
      --      ==
      You are-what/do-what?
�  "mo" may be used ANYWHERE a brivla or other selbri might.  Keep this in
  mind for later examples.  Unfortunately, by itself, "mo" is a very non-spe-
  cific question.  The response to the above question could be:
      mi [cu] melbi [vau]
      --      =====
      I am beautiful.
  or:
      mi [cu] tavla [vau]
      --      =====
      I talk.
  Clearly, "mo" requires some cooperation between the speaker and the respon-
  dent to ensure that the right question is being answered.  If context
  doesn't make the question specific enough, the speaker must ask the ques-
  tion more specifically using a more complex construction such as tanru
  (below).
  It is perfectly permissible for the respondent to fill in other unspecified
  places in responding to a "mo" question.  Thus, the respondent in the last
  example could have also specified an audience, a topic, and/or a language
  in the response.
Conversion ("se-word brivla"):
  "se" and others in its word-category modify a brivla used in a selbri by
  changing the order of the sumti that are attached, resulting a new selbri
  that expresses the same relation, but with different order of emphasis.
  "se" exchanges the first and second sumti places of the unmodified brivla.
  This reversal is called conversion.
  The bridi sentence:
      do      [cu] vecnu ta            mi      zo'e    [vau]
      --            ===== ----          --      ----
      seller-x1    sells thing-sold-x2 buyer-x3 price-x4
      You sell that to me.
  can be converted to:
      ta          [cu] se vecnu  do        mi      zo'e    [vau]
      --                ========  --        --      ----
      thing sold-x1    is-sold-by seller-x2 buyer-x3 price-x4
      That is sold by you to me.
  The effect is similar to what in English is called 'passive voice'.  In
  Lojban, however, a conversion is NOT 'passive':  the converted selbri has a
  place structure that is renumbered to reflect the place reversal, thus
  having effects when such a conversion is used in combination with other
  constructs (such as "fi" and "le selbri [ku]").
      Diagrammed Summary of Lojban Grammar Forms with Example Sentences
        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
Permission granted to reproduce for no charge to the recipient, for purposes
                          of promotion of Lojban.
                              Updated April 1992
�  The other simple relatives of "se" are "te" (switches 1st and 3rd places),
  "ve" (switches 1st and 4th places), and "xe" (switches 1st and 5th places).
  The effects may be seen on the 5-place gismu selbri "klama":
      GO-ER  goes  to destination  from origin  via route  using means
      -----  =====  --------------  -----------  ---------  -----------
        x1  klama        x2            x3          x4          x5
      DESTINATION  is-gone-to  BY GO-ER  from origin  via route  using means
      -----------  ==========  --------  -----------  ---------  -----------
          x1      se klama      x2          x3          x4          x5
      ORIGIN  is-gone-from  to destination  BY GO-ER  via route  using means
      ------  ============  --------------  --------  ---------  -----------
        x1    te klama            x2          x3        x4          x5
      ROUTE  is-gone-via  to destination  from origin  BY GO-ER  using means
      -----  ===========  --------------  -----------  --------  -----------
        x1  ve klama            x2            x3        x4          x5
      MEANS  is-used-to-go  to destination  from origin  via route  BY GO-ER
      -----  =============  --------------  -----------  ---------  --------
        x1  xe klama              x2            x3          x4        x5
tanru ("modifier-selbri modified-selbri"):
  tanru are compound selbri - constructions of multiple brivla/selbri com-
  ponents.  Each component might be a single word, or it might be a word
  modified by "se", or scalar negations (like "na'e").  tanru take selbri
  components (including other tanru) in pairs, with the first part modifying
  the second part.
  The kind of modification is vague:  tanru may act like an English ad-
  jective-noun (fast-runner), adverb/verb (quickly-run) or it may restrict a
  larger set (runner-shoes).  Context will generally indicate what is a
  plausible interpretation of a tanru.  But allow for creative interpreta-
  tion:  "runner-shoes" might be interpreted in some imaginative instances as
  "shoes that run by themselves".  In general, however, the meaning of a
  tanru is determined by the literal meaning of its components, and not by
  any connotations or figurative meanings (sutra tavla - "fast-talker" thus
  would not necessarily imply any trickery or deception, and a "jikca toldi"
  - "social butterfly" must always an insect with large brightly-colored
  wings, of the family lepidoptera).
  The place structure of a tanru is always that of the final brivla/component
  of the tanru.  Thus, the following has the place structure of "klama":
      mi [cu] sutra klama la meris. [vau]
      --      =========== ---------
      I quickly-go to Mary.
  With the conversion "se klama" in the final position, the place structure
  is that of "se klama": the x1 place is the destination, and the x2 place is
  the go-er:
      mi [cu] sutra se klama la meris. [vau]
      --      ============== ---------
      I quickly am-gone-to by Mary.
�  A similar example shows that there is more to conversion than merely
  switching places, though:
      la tam. [cu] melbi tavla la meris. [vau]
      -------      =========== ---------
      Tom beautifully-talks to Mary.  or  Tom is a beautiful-talker to Mary.
  has the place structure of "tavla", but note the two distinct inter-
  pretations.
  Now, using conversion, we can modify the place structure order:
      la meris. [cu] melbi se tavla la tam. [vau]
      ---------      ============== -------
      Mary is beautifully-talked-to by Tom.  or  Mary is a beautiful-audience
      for Tom.
  and we see that the modification has been changed so as to focus on Mary's
  role in the bridi relationship, leading to a different set of possible
  interpretations.
  Note that there is no place structure change if the modifying term is
  converted, and hence less drastic variation in possible meanings:
      la tam. [cu] tavla melbi la meris. [vau]
      -------      =========== ---------
      Tom is talkerly-beautiful to Mary.
      la tam. [cu] se tavla melbi la meris. [vau]
      -------      ============== ---------
      Tom is audiencely-beautiful to Mary.
  and we see that the manner in which Tom is seen as beautiful by Mary
  changes, but Tom is still the one perceived as beautiful, and Mary, the
  observer of beauty.
  Any selbri form can be used in either position of a tanru.  This allows
  more specific "mo" questions:
      do [cu] mo tavla mi [vau]
      --      ======== --
      You are _____(what?)-kind-of talker to me?
      do [cu] tavla mo mi [vau]
      --      ======== --
      You are talker-ly _____(what?) to me?
  It was stated above that you can use scalar negation ("na'e" and its
  equivalents) in tanru:
      do [cu] na'e sutra tavla [vau]
      --      ================
      You are an other-than-quick talker. (or)  You are a slow talker.
      do [cu] sutra na'e tavla [vau]
      --      ================
      You are quickly other-than-talking. (or)
      You are doing something other-than-talking, quickly.
�quantified selbri ("number moi"):
  Lojban numbers are expressed as strings of digits.  The basic digits are:
      pa  re  ci  vo  mu            xa  ze  bi  so  no      pi
      1    2    3    4    5            6    7    8    9    0      .
  number moi (usually seen combined into one word) (ordinal numbers):
      le tavla [ku] cu ci moi [vau]
      -------------    ======
      The talker is third.
  number mei (usually seen combined into one word) (cardinal numbers):
      le tavla [ku] cu ci mei [vau]
      -------------    ======
      The talkers are a-threesome.
  number si'e (usually seen combined into one word) (portional numbers):
      le blari'o [ku] cu pimu si'e [vau]
      ---------------    =========
      The blue-green (things) are a .5 portion (a half).
  number cu'o (usually seen combined into one word) (probability numbers):
      le blari'o [ku] cu pimu cu'o [vau]
      ---------------    =========
      The blue-green (occurrences) are a .5 probability (a half).
  Note the interpretation of x1 in the last example, which is a result of the
  place structure of probability numbers.  Each of these special kinds of
  selbri have other places besides the x1 sumti that appear in these
  examples.
  number selbri may also be used as part of a tanru:
      mi [cu] papa moi tavla do [vau]
      --      ============== --
      I am the 11th talker to you.
  The place structure again is that of the final component of the tanru.
Attaching internal sumti to a selbri:
  Each component of a tanru is not merely a single-word brivla, but a rep-
  resentation of an entire bridi relationship.  Lojban grammar allows the
  sumti that complete and define that bridi to be combined into the selbri.
  Combined sumti are called internal sumti.  We'll first show the structure
  of such a complex selbri component:
    brivla/selbri be sumti [bei sumti][bei sumti] ... [bei sumti] [be'o]
�  where the sumti attached with "be" is normally x2, and other sumti are
  optionally attached in numerical order (x3, x4, x5), each preceded by the
  marker "bei".  "be'o" is the end-marker for internal sumti, appearing after
  the last internal sumti for a brivla or other selbri.  Let's now look at
  one way that this construct is used.
tanru with internal sumti:
  Using the internal sumti structure, any of the components of a tanru can
  have its own sumti:
      ta [cu] tavla be do  bei le melbi ku [be'o] vecnu [vau]
      --    |    |  --| |    -----------      |
            |      -----  ----------------------|
              ====================================
      That is a talker-to-you-about-the-beautiful-thing(s) salesperson.
      (or, more simply)
      That's a salesperson who talks to you about beautiful things.
  In compound constructs such as this one, the normally optional elidable
  [omissible] right terminators may be mandatory to keep the sentence un-
  ambiguous.  Thus, in this last example, either the "ku" or the "be'o" must
  not be elided ["ku" was chosen arbitrarily].  Otherwise, "vecnu" is
  absorbed into the internal sumti:
      ta [cu] tavla be do  bei le melbi vecnu [ku] [be'o] [vau]
      --    |    |  --| |    -------------------      |
            |      -----  ------------------------------|
              ============================================
      That is a talker-to-you-about-the-beautiful-salesperson. (or)
      That one talks about beautiful salespeople to you.
  Obviously, a different statement.  In Lojban, you MUST be careful about
  properly including terminators when needed.  If in doubt, include the
  terminator; the statement CANNOT be ambiguous with the terminator present.
  In the last example, by omitting the elidable terminators "ku" and "be'o",
  we ended up with sumti attached to the selbri word in final position.  The
  latter sentence is thus identical in meaning to the same sentence expressed
  without internal sumti:
      ta [cu] tavla do le melbi vecnu [ku] [vau]
      --      ===== -- -------------------
      That is a talker-to-you-about-the-beautiful-salesperson. (or)
      That one talks about beautiful salespeople to you.
  Internal sumti can use any sumti construct, including the fa/fe/fi/fo/fu
  series to rearrange place orders:
      ta [cu] tavla be do  bei fo la lojban. [be'o] vecnu [vau]
      --    |    |  --| |    -------------      |
            |      -----  ------------------------|
              ======================================
      That is a talker-to-you-in-Lojban salesperson. (or, more simply)
      That's a salesperson who talks to you in Lojban.
�tanru inversion ("modified-selbri co modifier-selbri")
  We rephrased the English translations of the Lojban in the last two ex-
  amples in order to simplify the English structure and make the sentence
  more clear.  The same type of rearrangement is possible in Lojban.  The
  technique is called tanru inversion.  The modifier selbri is placed AFTER
  the modified one, with a "co" separating them:
          modified-selbri co modifier-selbri
  The "co" can often be translated as "of type".
  tanru inversion affects the interpretation of sumti that are in the sur-
  rounding bridi relationship in a peculiar manner.  The inversion causes a
  new brivla to be in the final position, and any following sumti are
  interpreted as being associated with that final brivla of the modifier-
  selbri.  The sumti PRECEDING the selbri still are associated with the final
  term of the modified-selbri, because that is the most essential relation
  that is being claimed by the sentence.
  One obvious advantage of tanru inversion is to simplify the apparent
  structure of a selbri.  As we have said, the selbri in the final position
  does not need to use internal sumti structures to attach its sumti.  The
  first of the examples above that use internal sumti becomes the simpler:
      ta [cu] vecnu co tavla do le melbi [ku] [vau]
      --      ============== -- -------------
      That is a seller of-type talker to you about the beautiful thing(s).
      (or)
      That's a salesperson who talks to you about beautiful things.
  "do" and "le melbi [ku]" are the x2 and x3 places of "tavla", while "ta"
  remains the x1 of the underlying modified-relation, which is "vecnu".
  The resulting Lojban now matches the colloquial English phrasing more
  closely, and the sentence is simpler since it does not require the complex
  marker system needed for internal sumti within the selbri.
  All of the forms of selbri listed in this section are governed by rules
  that form a hierarchy.  The most complex constructs are those using rules
  higher in the hierarchy.  In general, constructs built from higher rules
  cannot be used inside lower-rule constructs.  This hierarchy of rules is
  the primary reason why Lojban's grammar is unambiguous.
  Inversion of tanru uses rules which are highest in the hierarchy, thus
  allowing you to invert almost all other selbri constructs.  However, this
  means that you also CANNOT substitute a tanru inversion into very many
  other constructs within a selbri.
selbri grouping in tanru:
  tanru may be composed of more than two components, any of which may be more
  complex than the simple brivla and/or other selbri structures discussed
  above.  Lojban allows complex tanru structures to be unambiguously
  expressed, so that any such complex structure can be broken down into a
  series of modifier-modified pairs.  We present two of the variety of ways
  to express more complex groupings.
      Diagrammed Summary of Lojban Grammar Forms with Example Sentences
        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
Permission granted to reproduce for no charge to the recipient, for purposes
                          of promotion of Lojban.
                              Updated April 1992
�  In the absence of any grouping indications, components in tanru are pre-
  sumed grouped from the left:
      ti [cu] cmalu nanla ckule [vau]
      --      =================
      This    is-a small-boys school.  (This is a school for small boys.)
  But what if we want to group this selbri so as to talk about a "boys
  school" which is small.  One way is with tanru inversion:
      ti [cu] nanla ckule co cmalu [vau]
      --      ====================
      This    is-a boys school of-type small.  (This is a boys school which is
      small.)
  tanru inversion can work for many simple grouping problems.  However, since
  it changes the final brivla, it affects the interpretation of any sumti
  following the selbri.  There is a more general solution that does not
  affect which selbri is in final position, a choice that might be important
  because of the structural markers required.
  Any selbri, or any portion of a tanru that could stand alone as a selbri,
  may be surrounded with word-brackets "ke" (left) and "ke'e" (right) to
  indicate priority in grouping.  Normally, you will only use ke/ke'e
  grouping around strings of two or more selbri components, since the
  structure conveys no useful information around a single component.  At the
  end of the selbri and in other places where no ambiguity results, the
  "ke'e" terminator becomes optional (elidable):
      ti [cu] cmalu ke nanla ckule [ke'e] [vau]
      --      ===========================
      This    is-a small boys-school.
  A selbri structure surrounded by "ke" and "ke'e" has the same grammar as a
  single word brivla.  As a result, you can modify such structures with
  "na'e" and other scalar negation words, or with "se" and other conversion
  words:
      ti [cu] cmalu na'e ke nanla ckule [ke'e] [vau]
      --      ================================
      This    is-a small other-than-(boys-school).
abstraction selbri:
  The final selbri form we will cover in this paper isn't found very often in
  selbri in its most basic form.  However, it turns out to be one of the most
  important constructs in the language, showing up frequently as a part of
  more complex structures.  This selbri form is called sentence abstraction.
  The basic form is:
    nu-word complete-bridi-sentence [kei]
  where the bridi sentence inside the two markers can be a Lojban sentence of
  ANY type discussed in this paper, no matter how complex.  The word "nu"
  indicates an event abstraction, the most common kind of abstraction found
  in Lojban.
�      ti      [cu]  nu  mi    [cu] tavla do        [vau] [kei] [vau]
      --            |    --        ===== --                    |
                    |    talker    talks to-audience          |
                    |    I          talk  to you.              |
      --------      ===========================================
      event-x1      event-of "I talk to you."
      This is an event of my talking to you.
  The term 'event' should not be misconstrued.  In Lojban, it can refer to a
  momentary occurrence, or to a situation lasting hours, days, or even an
  indefinite period of time.  "nu" can stand for any of these kinds and dura-
  tions of events.  Other words may substitute for "nu" when you need to be
  specific and indicate particular event contours, such as a point event in
  time (mu'e) or a steady, unchanging state (za'i) of indefinite duration.
  There are other types of abstractions, as well, each indicated by words
  that substitute for "nu".  The most common of these are "ka" for a prop-
  erty/quality abstraction and "du'u" for a fact/assertion abstraction.
  These will be exemplified in the next section.  Most abstraction selbri
  have only one place (x1) which is the event, property, fact or other ab-
  stract 'thing' being described by the selbri.
  Abstractions, like other selbri, may be used in tanru; indeed, they are
  more common in tanru than alone:
      ti [cu] sutra bajra cukta [vau]
      --      =================
      This is-a fast-runner book.
  which might be a book about fast runners, while:
      ti [cu] sutra nu bajra kei cukta [vau]
      --      ========================
      This is-a fast-event(s)-of-running book.
  which is more likely a book about races, or a how-to book about running
  fast.  Note that the "kei" could not be elided in the last example, or the
  following would have resulted:
      ti [cu] sutra nu bajra cukta [kei] [vau]
      --      ====================
      This is-a fast event-of runner-book.
  and one imagines a very short-lived book about runners (it is the event
  that is fast, not the running or the book).
  Most abstractions that appear often in selbri like this, tend to be ab-
  breviated into a single-word selbri (a brivla) which embeds the "nu" into
  the concept.  Such compounds don't require a "kei", since the abstraction
  encompasses only the idea expressed within the word.  Lojban compound words
  (lujvo) are composed of combining forms of their component brivla (content
  words like "bajra") and cmavo (short structure words like "nu").
�  The rules for constructing lujvo are not difficult if you have a list of
  the combining forms (called rafsi); the rules are designed carefully to
  ensure that the pieces stay attached together and cannot be accidentally
  interpreted as separate words, since (as in the last example) the grammar
  of separate words may require added markers and terminators to obtain the
  meaning that you intend.  The lujvo for an event of running is formed from
  "nun", the rafsi combining form of "nu", and "bajra", which serves as its
  own combining form:
      sumti  nunbajra  sumti  sumti      sumti      sumti
      -----  ========  -----  -----      -----      -----
      event  [running] runner  on-surface using-limbs with-gait
      x1 is an event of x2 running on surface x3 using limbs x4, with gait x5
  Comparing this place structure with the one at the beginning of this
  section, you will see that x1 has been assigned to the event, while the
  remaining places are those of "bajra".  With this new lujvo, the next to
  the last example sentence:
      ti [cu] sutra nu bajra kei cukta [vau]
      --      ========================
      This is-a fast-events-of-running book.
  can use less-complicated structure:
      ti [cu] sutra nunbajra cukta [vau]
      --      ====================
      This is-a fast-(events-of-running) book.
                              III. COMPLEX sumti
More "le selbri [ku]" descriptions:
  Now that we have seen a variety of selbri forms, it perhaps becomes obvious
  that any of these selbri structures can be used in description sumti marked
  with "le".  Indeed, some of these structures, especially internal sumti and
  abstractions, are much more commonly found embedded in sumti, than in the
  selbri that defines the main relation of the sentence.
      ko [cu] tavla le sutra klama [ku] [vau]
      --      ===== -------------------
      You (imperative) talk to the quick-goer. (Talk to the quick-goer.)
      do [cu] klama le se tavla [ku] ta [vau]
      --      ===== ---------------- --
      You go to the one-talked-to from that.
  This last example shows the virtue of "se" conversion.  It has allowed us
  to mentally convert "tavla" to make its x2 place accessible by description
  in the sumti.
  Now for a tricky usage.  We use "go'i" to refer to the bridi of the last
  sentence.  Therefore "le go'i [ku]" refers to the first place of that bridi
  (in this case, the go-er "do").  If we want to refer to the second place of
  the last sentence bridi, the destination, we can mentally convert that
  sentence using "se", and "le se go'i [ku]" means the destination ("le se
  tavla [ku]").
�      le se go'i [ku] cu melbi [vau]
      ---------------    =====
      The destination is beautiful. (or)
      The one talked-to (the destination) is beautiful.
      le na'e melbi [ku] cu tavla [vau]
      ------------------    =====
      The other-than-beautiful one talks.
      mi [cu] tavla le mo klama [ku] ri [vau]
      --      ===== ---------------- --
      I talk to the what-kind-of go-er about itself?
      le re moi prenu [ku] cu tavla mi [vau]
      --------------------    ===== --
      The second person talks to me.
  It is IMPORTANT to remember and correctly use the elidable separator "cu"
  with description selbri.  If you misplace it or omit it (and its less-often
  used alternative "ku"), you will create some very strange tanru.
      le sutra [ku] cu vecnu cukta [vau]
      -------------    ===========
      The quick-one is a seller book.
      le sutra vecnu [ku] cu cukta [vau]
      -------------------    =====
      The quick seller is a book.
  And, if you omit the "cu", you get only a sumti:
      le sutra vecnu cukta [ku] [vau]
      -------------------------
      The quick-seller book.
sumti descriptions with internal sumti:
      le selbri be sumti bei sumti bei sumti ... bei sumti [be'o] [ku]
      |        |-------------------------------------------------|    |
      ----------------------------------------------------------------
  is all one sumti.
      le tavla  be la .an. bei le vecnu [ku] [be'o] [ku] cu klama [vau]
      |        |  -------    -------------      |    |  =====
      |          -----------------------------------    |
      --------------------------------------------------
      The talker to Ann    about the seller                goes.
  You can even fill in the places of an internal sumti:
      le tavla be le melbi be la .an. [be'o] [ku] [be'o] [ku] cu klama [vau]
      |      |  |      |  -------      |    |      |    |  =====
      |      |  |        -----------------    |      |    |
      |      |    ------------------------------      |    |
      |        -----------------------------------------    |
      ------------------------------------------------------
      The talker to the one who is beautiful to Ann goes..
      (The one talking to the one Ann thinks is beautiful, goes.)
�  Without the omitted terminators this looks a bit less wordy:
      le tavla be le melbi be la .an. cu klama
      The one talking to the one Ann thinks is beautiful, goes.
  Since a selbri can be a tanru with internal sumti, a sumti may be built on
  such a selbri, possibly even having internal sumti on BOTH components of a
  tanru (in a tanru embedded in a sumti, even the sumti attached to the final
  component must be attached with be/bei/be'o).:
      le melbi be la .an. [be'o] tavla be la mark. [be'o] [ku] cu vecnu ti
      |      |  --------      |    |  --------      |    |  ===== --
      |        -----------------      ------------------    |
      -------------------------------------------------------
      The beautiful-to-Ann talker-to-Mark sells this.
      (The one Ann thinks is beautiful who is talking to Mark, sells this.
  Here, either the "ku" or the "cu" is elidable before the main selbri (both
  "be'o"s are unconditionally elidable).  Most frequently, when there is a
  choice, the terminator that best communicates the sentence structure is
  chosen.  "cu", in this case, clearly separates the complex sumti from the
  selbri, and is preferred.  Often, a single "cu" may allow you to omit
  several elidable terminators that would otherwise be necessary.  This
  happens most frequently with abstraction selbri that are used in sumti
  descriptions.
abstraction sumti clauses:
  le nu sentence kei:
      mi [cu] djica le nu mi [cu] klama le zarci [ku] [vau] [kei] [ku] [vau]
      --      ===== | |  --      ===== -------------            |    |
                    |  ==========================================    |
                    -------------------------------------------------
      I      want  the-event-of: I go to the store.
      I want to go to the store.
  With the omitted terminators not printed, this sentence looks much shorter:
      mi djica le nu mi klama le zarci
      I want to go to the store.
  An even shorter form will typically appear in Lojban text.  "le nu" occurs
  so frequently in combination that it is often written as a single word;
  this isn't mandatory - cmavo compounds are always understood as meaning the
  same thing as the words written separately.  cmavo are generally written
  one word when they together equate to a concept that is written in other
  languages as one word.
  In addition, the "mi" inside the abstraction will often be omitted.  When a
  listener hears this sentence and realizes that the go-er wasn't specified
  (as well as the origin, the route and the means), the obvious value(s) will
  be assumed.  Leaving out the "mi" is exactly comparable to the difference
  between the two English sentences:
      I want to go to the store.    (and)    I want myself to go to the store.
      mi djica lenu klama le zarci
      I want to go to the store.
      Diagrammed Summary of Lojban Grammar Forms with Example Sentences
        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
Permission granted to reproduce for no charge to the recipient, for purposes
                          of promotion of Lojban.
                              Updated April 1992
�  If an abstraction is in the x1 position, "cu" allows four other elidable
  markers to be omitted.  For example in a conversion of the last example
  sentence:
      lenu mi [cu] klama le zarci [ku] [vau] [kei] [ku] cu se djica mi [vau]
      | |  --      ===== -------------            |    |  ======== --
      |  =========================================    |
      ------------------------------------------------
      The-event-of: (I go to the store) is desirable to me.
  the "cu" allows four other markers to be omitted.  Using "cu" is much
  easier for the listener, who thus knows in one word that the complex sumti
  is completely ended and the main selbri comes next.
  We promised to give examples of two of the other types of abstractions in
  this section.  These abstractions tend to be associated with specific
  places of particular brivla.
      la mark. [cu] ricfu  le ka melbi [vau] [kei] [ku] [vau]
      --------      =====  ----------------------------
      Mark is rich in the quality of x1 being beautiful to x2 by standard x3
      mi [cu] djuno le du'u la djan. [cu] te vecnu [vau] [kei] [ku] [vau]
      --      ===== -----------------------------------------------
      I      know  the-fact-that John is a buyer.
      I know that John buys (something).
  As with the earlier examples. these sentences will typically appear much
  shorter in print:
      la mark. ricfu leka melbi
      Mark is rich in beauty.
      mi djuno ledu'u la djan. te vecnu
      I know that John buys (something).
Quantified sumti ("le number [boi] selbri [ku]"):
  One way to quantify a sumti being described is to insert the number fol-
  lowed by a terminator "boi" that may be omitted when no ambiguity results
  (the usual case):
      le re [boi] tavla [ku] cu klama [vau]
      ----------------------    =====
      The two talkers go.
  "boi" may not be elided when a number selbri follows, since you wouldn't
  know where one number stops and the next begins.  (Lojban does not allow
  such boundaries to be expressed by contrastive stress as in English.):
      le reno boi remei [ku] cu klama [vau]
      ----------------------    =====
      The twenty twosomes go.
�Quantified selection from sumti ("number [boi] le [number] [boi] selbri
[ku]"):
  You can also put a number preceding the "le", to select from the set in-
  dicated by the description:
      re [boi] le tavla [ku] cu klama [vau]
      .......> -------------    =====
      Two of the (unspecified number of) talkers go.
      re [boi] le reno boi remei [ku] cu klama [vau]
      .......> ----------------------    =====
      Two of the twenty twosomes go.  (i.e. four altogether)
      le tavla be pa [boi] le ci [boi] bajra [ku] [be'o] [ku] cu melbi [vau]
      |      |  .......> ----------------------      |    |  =====
      |        -----------------------------------------    |
      ------------------------------------------------------
      The talker to one-of the three  runners                is-beautiful.
Indefinite description sumti ("lo selbri [ku]")
  If you wish to describe a sumti, but do not have a specific in mind, you
  can instead refer generically to something that meets the terms of the
  description selbri:
      lo tavla [ku] cu klama [vau]
      -------------    =====
      A talker goes.  (or)  Some talkers go.
  "lo" may be used interchangeably with "le" in the preceding examples, with
  an indefinite description as a result.
  Lojban allows you to omit the "lo" in:
    number [boi] [lo] [number] [boi] selbri [ku]
      re [boi] tavla [ku] cu klama [vau]
      -------------------    =====
      re tavla cu klama
      Two [of the unspecified number who are] talkers go.
"di'u" and "la'e di'u":
  In English, I might say "The school is beautiful", and you might reply
  "This pleases me."  How do you know what "this" refers to?  Lojban uses
  different expressions to convey the possible meanings of the English:
      le ckule [ku] cu melbi [vau]
      -------------    =====
      The school is beautiful.
�  The following three sentences all might translate as "This pleases me."
      ri [cu] pluka mi [vau]
      --      ===== --
      This (the school, the last expressed sumti) pleases me.
      di'u [cu] pluka mi [vau]
      ----      ===== --
      This (the last sentence) pleases me (perhaps because it is grammatical
      or sounds nice).
      la'e di'u [cu] pluka mi [vau]
      ---------      =====
      This (the referent of the last sentence; i.e. that the school is
      beautiful) pleases me.
  The last sentence is an example of using one sumti to point to or refer to
  another by inference.  "la'edi'u" is often written as a single word, and is
  used more often than "di'u" by itself.
Other sumti types:
  Lojban supports several other sumti types, more than we can discuss in a
  short paper.  These include bare numbers, several kinds of quoted text
  (single words, grammatical or possibly ungrammatical text, and non-Lojban
  text).
              IV. ATTACHMENTS TO sumti AND selbri AND sentences
                          A. ATTACHMENTS TO sumti
All structures in this section apply to sumti at the main level of a sentence,
as well as to sumti within substructures.
Adding a new sumti place to a bridi relationship ("modal + sumti"):
  time or location:  e.g. "pu"/before, "ba"/after, "ca"/simultaneous with,
  "vi"/at, "va"/near, "vu"/far from
  There are many more of these, and some specialized rules for compounding
  them.  These are discussed in a more advanced paper.
      mi [cu] tavla pu le nu do [cu] tavla [vau] [kei] [ku] [vau]
      --      ===== .> ------------------------------------
      mi tavla pu lenu do tavla
      I talk before the-act-of you talk.  (I talk before you do.)
  You must think carefully about what you mean with these constructions:
      mi [cu] tavla pu do [vau]
      --      ===== .> --
      I talk before you.  (I talked before "you" even existed.)
�  modal tag: e.g. "secau" (without ...), "mu'i" (motivational because ...),
  "du'o" (according to ...)
      do [cu] klama secau mi [vau]
      --      ===== ....> --
      You go without me.
modal questions ("modal + ma"):
  Express common English questions using time, location, and modal tag words
  combined with "ma":
      ca ma do [cu] tavla [vau]
      --            =====
      Simultaneous-with ____(what?), you talk?  (When do you talk?)
      vi ma do [cu] tavla [vau]
      --            =====
      At ____(what?), you talk?  (Where do you talk?)
      mu'i ma do [cu] tavla [vau]
      --              =====
      Motivationally because ____(what?), you talk?  (Why do you talk?)
      (Lojban has other "because" modal tags for asking a variety of different
      "why?" questions.)
sumti relative phrases ("sumti pe modifier-sumti [ge'u]"):
  A sumti may be identified more exactly by attaching a relative phrase,
  another sumti, that in some way restricts the possible set of things being
  referred to:
      le cukta [ku] pe le vecnu [ku] [ge'u] cu blari'o [vau]
      ------------- <  -------------      |  =======
                    <......................
      The book of the salesperson is blue-green.
  "pe", the basic marker of a restrictive relative phrase, is vague as to the
  exact nature of the relationship between the original sumti and the sumti
  that is identifying.  It is thus similar to the loosest English possessive,
  as in "my chair/the chair of mine", which may be used for a chair that you
  sit in but which is owned by someone else.
  Successively tighter degrees of association/possession are indicated by
  "po" (alienable possession) and "po'e" (inalienable possession):
      le cukta [ku] po mi [ge'u]
      ------------- <  --      |
      The book of mine. (Even if you are holding it, it is still my book.  But
      I also could give it to you such that it is no longer my book.)
      le birka [ku] po'e mi [ge'u]
      ------------- <    --      |
      The arm of mine (It is still MY arm; it cannot be given away.)
�  "po'u" identifies a sumti by giving another IDENTITY: a sumti that could
  equivalently replace the original:
      la djan. po'u  le vecnu [ku] [ge'u] cu klama [vau]
      -------- <      -------------      |  =====
      John    who-is the seller              goes.
      John-the-saleman goes.
  "goi", used like "pe", defines "ko'a" and other variable sumti for use
  throughout a text without repeating:
      le melbi tavla [ku] goi ko'a [ge'u] [cu] vecnu [vau]
      ------------------- <  ----      |    =====
      The beautiful talker (hereinafter "ko'a") sells.
      .i ko'a djica lenu do klama ko'a [vau] [kei] [ku] [vau]
        ---- ===== -----------------------------------
      He/she (the beautiful talker) wants the event of you going to him/her.
  Lojban has words identical in grammar to "pe" that provide non-restrictive
  (incidental) information about a sumti.  "ne" is the non-restrictive
  (incidental) equivalent of "pe":
  This construct may be combined with the modal construct discussed just
  previously to identify a sumti:
      la djan. ne pu la mark. [ge'u] [cu] melbi tavla [vau]
      -------- <  >. --------      |    ===========
      John, who was (incidentally) before Mark, is a beautiful-talker.
  The contrast between "ne" (incidental) and "pe" (identifying) is shown by
  giving the same sentence with "pe":
      la djan. pe pu la mark. [ge'u] [cu] melbi tavla [vau]
      -------- <  >. --------      |    ===========
      (The) John who was before Mark is a beautiful-talker.
  Certain modal tags, in fact, are designed primarily for use in relative
  phrases, rather than to attach additional sumti structures to the main
  bridi relationship as described earlier.  One of these is "semau" ("more
  than ..."):
      la djan. ne semau la mark. [ge'u] [cu] tavla mi [vau]
      -------- <  >.... --------      |    ===== --
      John, (incidentally) more than Mark, talks to me.
      (John talks to me more than Mark does.)
      la djan. [cu] tavla mi ne semau la mark. [ge'u] [vau]
      --------      ===== -- <  >.... --------      |
      John talks to me, (incidentally) more than Mark.
      (John talks to me more than he does to Mark.)
�  Without linking a "semau" modal sumti to another sumti with "ne" or "pe",
  it is hard to understand:
      ???la djan. [cu] tavla mi semau la mark. [vau]
        --------      ===== -- >.... --------
      John talks to me, more than Mark. (but Mark is tied neither to John nor
      me, but to the talking).
      (John's talking to me) is more than (Mark).
  Comparing a talking relationship to a person is nonsense.
sumti relative clauses ("sumti poi sentence ku'o"):
  "pe" phrases are limited to what can be expressed in a single sumti.  When
  you need to include more complete information about a sumti, Lojban
  provides for relative clauses.  A restrictive relative clause marker,
  "poi", marks a following complete bridi as providing detailed identifying
  information about the sumti by providing a relationship that the sumti fits
  into.
  The placeholder for the sumti being identified is "ke'a", which is merely
  another in the set of single-word sumti.  "ke'a" is often left out if it is
  contextually obvious where it would go (especially when "ke'a" would go in
  the x1 position immediately after "poi", or in the first available
  unspecified place if that x1 is provided in the clause).  A relative clause
  is terminated with the marker "ku'o" which may be omitted if no ambiguity
  will result.  It is very rare that "ku'o" needs to be expressed.
      le ckule [ku] poi  mi [cu] klama ke'a [vau] [ku'o] cu blari'o [vau]
      ------------- <    --      ===== ----            |  =======
                    <....................................
      The school    which I      go-to it                  is-blue-green.
      The school I go to is blue-green.)
  Note that "ke'a" refers to the school.
      le bajra [ku] poi [cu] tavla [vau] [ku'o] cu vecnu [vau]
      ------------- <        =====            |  =====
                    <..........................
      The runner who talks is a seller.
  There is also a non-restrictive relative clause marker, "noi", for in-
  cidental information about a sumti.
tensed sumti ("le time/location/modal-tag + selbri [ku]")
  A sumti may also have a time or location or modal tag in front of a de-
  scription selbri:
      le pu bajra [ku] [cu] tavla [vau]
      ----------------      =====
      The earlier/former/past runner talked/talks. (Since Lojban tense is
      optional, we don't know when.)
      le vi bajra [ku] [cu] tavla [vau]
      ----------------      =====
      The here runner talks.  (This runner talks.)
      Diagrammed Summary of Lojban Grammar Forms with Example Sentences
        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
Permission granted to reproduce for no charge to the recipient, for purposes
                          of promotion of Lojban.
                              Updated April 1992
�short possessive sumti ("le possessor-sumti selbri [ku]")
  As a comparison with the last example, note that a description sumti may
  have a sumti after the "le", resulting in a short form of the loose "pe"
  possessive.
  English "this" might be represented by "vi" or by "ti", depending on
  context.
      le ti bajra [ku] [cu] tavla [vau]
      ----------------      =====
      The this-one's runner talks.  (This one's runner talks.)
  A more complete structure of a description sumti is "[number] le [number]
  [sumti] [modal] selbri [ku]".
                          B. ATTACHMENTS TO selbri
tensed or adverbial bridi relationships:
  Immediately after a "cu" before a selbri, you can have a modal (such a
  modal being there may make a "cu" unnecessary, since modals cannot be
  absorbed into tanru).  Such modals serve as an equivalent to English tenses
  and adverb functions.  In Lojban, tense is completely optional.  If
  unspecified, tense is picked up from context.
      do [cu] vu vecnu zo'e [vau]
      --      >. =====
      You yonder sell something-unspecified.
      le vi tavla [ku] [cu] ba klama [vau]
      ----------------      >. =====
      The here talker will go.  (This talker will go.)
      do [cu] mu'i tavla mi [vau]
      --      >... ===== --
      You motivatedly talk to me.
                        C. ATTACHMENTS TO sentences
A variety of constructs may occur anywhere in a sentence, operating inde-
pendent of the primary grammar of the sentence.  These constructs generally
have minimal internal grammar.  Like "xu", when not at the beginning of a
sentence, they indicate emphasis on the word or construct they immediately
follow.
Indicators:
  Indicators include a variety of expressions conveyed in English through
  interjections or tone of voice.  Lojban supports a enormous range of
  emotional expression through specific words and compounds.  Indicators may
  be modified for intensity, or classified by sphere (social, mental,
  emotional, physical, sexual, spiritual).
      .ie mi [cu] klama [vau]
          --      =====
      Agreement! I go. (Yep! I'll go.)
�      .ei mi [cu] klama [vau]
          --      =====
      Obligation! I go. (I should go.)
      mi [cu] klama le ckule .ui [ku] [vau]
      --      ===== --------(  )---
      I go to the school (and I am happy because it is the school I'm going
      to).
Discursives:
  Discursives allow free expression of certain metalinguistic comments
  (comments about the text).  Use of discursives allows clear separation of
  these metalinguistic features from the underlying statements and logical
  structure.  By comparison, the English words "but" and "also", which
  discursively indicate contrast or added weight-of-example, are logically
  equivalent to "and" without that discursive content.  The average English
  speaker does not think about, and may not even realize, the contrary-
  seeming idea that "but" basically means "and".
      mi [cu] klama [vau] .i do [cu] stali [vau]
      --      =====          --      =====
      I go.  You stay.
      mi [cu] klama [vau] .i ji'a do [cu] stali [vau]
      --      =====              --      =====
      I go.  In addition, you stay.  (added weight)
      mi [cu] klama [vau] .i ku'i do [cu] stali [vau]
      --      =====              --      =====
      I go.  However, you stay.  (contrast)
Evidentials:
  Evidentials indicate the speaker's relationship to the statement,
  specifically communicating how the speaker came to make the statement.
  These include "za'a" (I directly observe the relationship), "pe'i" (I opine
  that the relationship holds) "ru'a" (I postulate), and others.  Many
  Amerind languages use these type of words.
      pe'i do [cu] melbi [vau]
          --      =====
      I opine!  You are beautiful.
      za'a do [cu] melbi [vau]
          --      =====
      I directly observe!  You are beautiful.
  Some expressions overlap between categories, having value as indicators as
  well as evidential import (".ia", belief, faith), or attitudi-
  nal/metalinguistic import ("ga'i", hauteur, relative high-rank; cf.
  Japanese).
      .e'a ga'i ko'a [cu] citka lo titnanba [ku] [vau]
                ----      ===== ----------------
  Permission!/I permit! Hauteur! They eat cake.  (Let them eat cake!)
�                            V. LOGICAL CONNECTIVES
Lojban expresses logical connectives so as to unambiguously indicate logical
scope (what is being connected), as well as to isolate various non-logical
features from the logical ones (as noted above for discursives).  Logical
connectives may be expressed in forethought forms (both ... and ...) or
afterthought  forms (..., and ...).  The form of logical connectives directly
indicates their scope and the associated truth table for the two connected
terms.  For example, a bare vowel joins sumti (".a"/alternation, a and/or b, a
OR b; ".e"/conjunction, a AND b).  "na" and "nai" may be used to negate con-
nected terms ("na.a"/conditional, not a OR b, if a then b; "anai"/a or not b,
a if b).
Parallel structures are shown in the following abbreviated table:
    sumti              sumti forethought                  sentences
OR  a and/or b        .a      ga sumti gi sumti          .i ja
AND  a and b            .e      ge sumti gi sumti          .i je
XOR  a or b, not both  .onai    go sumti ginai sumti      .i jonai
->  if a then b        na.a    ganai sumti gi sumti      .i naja
  (a only if b)
<-  a if b            .anai    ga sumti ginai sumti      .i janai
    tanru              tanru forethought                  compound bridi
OR  a and/or b        ja      gu'a sumti gi sumti        gi'a
AND  a and b            je      gu'e sumti gi sumti        gi'e
XOR  a or b, not both  jonai    gu'o sumti ginai sumti    gi'onai
->  if a then b        naja    gu'anai sumti gi sumti    nagi'a
  (a only if b)
<-  a if b            janai    gu'a sumti ginai sumti    gi'anai
      mi [cu] citka ti .a ta [vau]
      --      ===== --(++)--
      I eat this and/or that.
      mi [cu] citka ti .e ta [vau]
      --      ===== --(++)--
      I eat this and that.
      mi [cu] citka ti na.a ta [vau]
      --      ===== --(++++)--
      I eat this only if that.
      mi [cu] citka ti .anai ta [vau]
      --      ===== --(+++++)--
      I eat this if that.
      ganai mi gi do [cu] klama [vau]
      +++++)--(++)--      =====
      If me, then you, go.
      mi [cu] klama [vau]  .i je  do [cu] klama [vau]
      |--      =====      |      |--      =====      |
      |___________________|(+++++)|___________________|
      I go AND you go.
      mi [cu] klama tu [vau]  gi'e  citka ti [vau] [vau]
      --    |===== --      |      |===== --      |
      --    |______________|(++++)|______________|
      I go to that-yonder AND eat this/these.
�Lojban also supports NON-logical connectives, such as mixed connection and set
membership (ordered/un-ordered):
      mi [cu] citka le blari'o joi pelxu [ku] [vau]
      --      ===== ----------(+++)----------
      I eat the blue-green-mixed-with-yellow (stuff).
      mi [cu] citka le blari'o ku joi le pelxu [ku] [vau]
      --      ===== -------------(+++)-------------
      I eat the blue-green (stuff) mixed-with the yellow (stuff).
  (The "ku" before the "joi" in the latter example CANNOT be left out due to
  the ambiguity resolution rules.)
                      VI. VERY BRIEF PRONUNCIATION GUIDE
              (covers only features occurring in these examples)
Vowels
a as in father, never as in sofa
e as in bet
i s in machine
o as in cocoa
u as in flute
Vowel pairs without apostrophes
ai  as /igh/ in "high"
au  as /ow/ in "cow"
ei  as in "weight"
ia  as the German "Ja"
ie  as /ye/ in "yep"
oi  as /oy/ in "toy"
ui  as the word "we"
Consonants as in English except
c sh as in "shoe"
j z as in "azure"
x hard ch of Scottish "loch" (breathe through a 'k')
g always hard, as in "go", NOT as in "germ"
s always as in "floss", NOT as in "flows"
A period indicates a mandatory pause, which may be very short.  An apostrophe
(') between vowels splits them into 2 syllables with no sound from your vocal
cords (usually like a brief 'h' sound) between.  There must be no pause, and
no 'voiced' sound: "e'o" should be a smooth transition from "e" to "o" that
doesn't sound like /ayo/ of "mayo".
Stress longer words (brivla) penultimately (the next to the last syllable).
      klama    /KLAH,mah/    blari'o  /blah,REE,ho/
� VII. BRIEF GLOSSARY OF LOJBAN WORDS USED IN THESE EXAMPLES
.a        and/or: sumti            ma      sumti question
.anai    only if: sumti            mei      cardinal # selbri
ba      after/future time        mi        I; me; we; us
be      internal sumti link      mo        selbri question
be'o    end internal sumti        moi      ordinal # selbri
bei      more internal sumti      mu        5
bi      8                        mu'e      abstract.: point event
boi      end-of-number            mu'i      modal: motiv. cause
ca      same time/present        na        contradict. negation
ci      3                        na'e      scalar other-than
co      tanru inversion          na.a      only if: sumti
cu      selbri separator          nago'i    last bridi false; No!
cu'o    probability selbri        ne        incidental rel. phrase
di'u    last sentence            ni'o      new topic
do      sumti: you                ni'oni'o  new section
du'o    modal: according to...    no        0; none
du'u    abstraction: fact        no'e      scalar neutral
.e        AND: sumti                noi      incidental rel. clause
.e'a      indicator: permission    nu        abstraction: event
.ei      indicator                pa        1
fa      x1 sumti next            papa      11
fa'o    end-of-text              pe        identifying phrase
fe      x2 sumti next            pe'i      evidential: I opine!
fi      x3 sumti next            pi        .
fo      x4 sumti next            pimu      .5
fu      x5 sumti next            po        alienable possession
ga      and/or: forethought      po'e      inalienable possess.
ga'i    indicator: high rank      po'u      identity phrase
ganai    if: forethought          poi      identifying clause
ge'u    end relative phrase      pu        before/past tense
gi      forethought medial        re        2
gi'e    and: bridi                remei    twosome, pair
go'i    last bridi; too; yes!    reno      20
goi      sumti definition          ri        the last sumti
.i        sentence link            ru'a      evidential: I assume!
.ia      indicator: belief/faith  se        x1/x2 conversion
.ie      indicator: agreement      secau    modal tag: without...
ja'a    it is true                semau    modal tag: more than
je      and: tanru, sentences    si'e      portion # selbri
je'a    scalar positive          so        9
ji'a    discursive: also          ta        that-there; those
joi      mixed-with 'and'          te        x1/x3 conversion
ka      abstraction: quality      ti        this-here; these
ke      tanru grouping            to'e      scalar opposite
ke'a    relative clause sumti    tu        that/those yonder
ke'e    end tanru grouping      .ui        indicator: happiness
kei      end abstraction          va        location: near
ko      imperative You!          vau      end of sumti list
ko'a    he/she/it - assigned      ve        x1/x4 conversion
ku      end description sumti    vi        location: here/at
ku'i    discursive: but          vo        4
ku'o    end relative clause      vu        location: far from
la      name follows              xa        6
la'e    the referent of          xe        x1/x5 conversion
la'edi'u bridi of last sentence    xu        Is it true?
le      description sumti; the    za'a      evidential: observed
ledu'u  the fact that ...        za'i      abstraction: state
leka    the quality of ...        ze        7
lenu    the event of ...          zo'e      unspecified sumti
lo      description sumti; a
�.an.      name: Ann
lojban.  name: Lojban
mark.    name: Mark
meris.    name: Mary
tam.      name: Tom
brivla actually translate to a complete sentence indicating the place structures)
bajra
birka
blari'o      x1 is blue-green
bridi
brivla
citka
ckule
cmalu
cmavo
cukta
djarspageti
djica
djuno
gismu
jikca
klama
le'avla
lujvo
melbi
nanla
nunbajra    x1 event of x2 runs on surface x3, limbs x4, gait x5
pelxu
pluka
prenu
rafsi
ricfu
selbri
stali
sumti
sutra
tanru
tavla
titnanba    x1 is sweet-bread/cake
toldi
vecnu
zarci
</pre>

Latest revision as of 08:45, 30 June 2014

originally placed at http://www.lojban.org/publications/brochures/lojex.txt


      Diagrammed Summary of Lojban Grammar Forms with Example Sentences

        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
 Permission granted to reproduce for no charge to the recipient, for purposes
                           of promotion of Lojban.
                              Updated April 1992

These pages give diagrammed examples of basic Lojban sentence structures.  The
most general pattern is covered first, followed by successive variations on
the basic components of the Lojban sentence.  There are many more capabilities
not covered in these examples.  A brief pronunciation guide and Lojban
glossary will be found at the end of this paper.


                       I. THE LOJBAN SENTENCE STRUCTURE

A Lojban sentence expresses a relationship (bridi), normally claiming that the
relationship holds (that it is 'true').  A bridi relationship consists of sev-
eral ideas or objects called arguments (sumti), which are related by a
predicate relation (selbri).  The following uses the Lojban terms bridi,
sumti, and selbri, because it is best to come to understand them independent
of the English associations of the corresponding words.

Some words used as sumti:

   mi     I/me/my, we/us/our
   do     you/your
   ti     this/this-here/this one's, these/these here/these ones'
   ta     that/that-there/that one's, those/those ones'
   tu     that yonder, those yonder
   zo'e   unspecified value (used when a sumti is unimportant or obvious)

   sumti are not specific as to number (singular or plural), nor gender
   (masculine/feminine/neutral).  Such distinctions can be optionally added.

   Names may be expressed as sumti, labelled with "la":

   la meris.   the one/ones named Mary
   la djan.    the one/ones named John

   (Other Lojban spelling versions are possible for names from other lan-
   guages.)

Some words used to indicate selbri relations:

   vecnu   x1 (seller) sells x2 (goods) to x3 (buyer) for x4 (price)
   tavla   x1 (talker) talks to x2 (audience) about x3 (topic) in language x4
   blari'o x1 (object/light source) is blue-green
   melbi   x1 (object/idea) is beautiful to x2 (observer) by standard x3

   We will describe these and other possible sumti and selbri in more detail
   below.







�Conventions:

   The following conventions will be used to show the structure of Lojban sen-
   tences in diagrams:

   - The selbri relation will be double-underlined.

   - The sumti arguments will be single-underlined.

   - Optional separator/terminator words are placed in square brackets.  They
   may be omitted if so bracketed. The general rule is that these may be omit-
   ted if and only if no grammatical ambiguity results.  Each such word serves
   as an end marker for particular structures, making the overall structure of
   the sentence very clear.

Basic structure of a Lojban sentence:

      sumti sumti ... sumti [cu] selbri sumti sumti ... sumti [vau]
      ----- -----     -----      ====== ----- -----     -----

   - Normally, there must be at least one sumti before the selbri.

   - Each selbri relation has a specifically defined place structure that de-
   fines the role of each sumti in the bridi relationship, based on its posi-
   tion in order.  In the examples above, that order was expressed by la-
   belling the positions x1, x2, x3, and x4.

   - "cu" acts as a separator after at least one preceding sumti to clearly
   mark the selbri.  As the diagram indicates, it may often be omitted.  There
   will be examples below.

   - "vau" goes at the end of the sentence, indicating that no more sumti will
   follow.  It usually may be omitted.

Sentence examples:

      mi       [cu] vecnu  ti            ta       zo'e    [vau]
      --            =====  --            --       ----
      seller-x1     sells  goods-sold-x2 buyer-x3 price-x4
      I             sell   this          to that  for some price.
      I sell this-thing/these-things to that-buyer/those-buyers.  (The price
      is obvious or unimportant.)

   - Both the "cu" and the "vau" are optional in this example and could be
   omitted:

   - When an unspecified sumti ("zo'e") is at the end of a sentence, it may be
   omitted.

   - Normally, there will be one sumti (the x1) before the selbri.  There may
   be more than one:

      mi        ti           [cu] vecnu ta      [vau]
      --        --                ===== --
      seller-x1 goods-sold-x2     sells buyer-x3
      I         this              sell  to that.
      Translates as stilted or poetic English:
      I, this thing, do sell to that buyer.


�   Usually, more than one sumti before the selbri will be for style or for em-
   phasis on the sumti that are out-of-place from their normal position.
   (Native speakers of languages other than English may prefer such orders.)

Observatives:

   If there is no sumti before the selbri, then it is understood that the x1
   sumti value is equivalent to "zo'e"; i.e. unimportant or obvious, and
   therefore omitted.  Any sumti after the selbri start counting from x2, x3,
   x4 ...:

      ta            [cu] melbi       [vau]
      --                 =====
      object/idea-x1     is-beautiful     (to someone by some standard)
      That/Those         is/are beautiful.
      That is beautiful.  (or)  Those are beautiful.

   When the x1 is omitted:

                    melbi        [vau]
      ------------  =====
      (Unspecified) is-beautiful (to someone by some standard)
      melbi
      Beautiful!  (or)  It's beautiful!

   Omitting the x1 adds emphasis to the selbri relation, which has become
   first and foremost in the sentence.  This kind of sentence is termed an ob-
   servative, because it is usually stated by someone when they first observe
   or take note of the relation, and wish to quickly communicate it to someone
   else.  Commonly understood English observatives include "Smoke!" upon see-
   ing smoke or smelling the odor, or "Car!" to a person crossing the street
   who might be in danger.  Any Lojban selbri can be an observative if no
   sumti appear before the selbri.

   "cu" does not occur in an observative; "cu" is a separator, and there must
   be a sumti before the selbri that needs to be kept separate, for it to be
   used.  With no sumti preceding the selbri, "cu" is not permitted.

True/false (yes/no) questions (the word "xu"):

   xu question sentence (Is-it-true-that...):

      xu          mi [cu] vecnu ti   ta  [vau]
                  --      ===== --   --
      Is-it-true? I       sell  this that.
      Is-it-true-that I sell this to that?

   "xu" has a very unrestricted grammar, and is permitted virtually anywhere
   in a sentence.  At the beginning of the sentence, "xu" asks about the truth
   of the bridi relationship.  Elsewhere, in a sentence, "xu" attaches to the
   immediately preceding word (or the structure implied by that preceding
   word, when it is the marker for a structure).  Thus, also after the "vau"
   ending the sentence, "xu" asks about the entire bridi (the "vau" cannot be
   omitted if "xu" is to appear 'after' it).






�   "xu" appearing after a sumti questions whether the bridi relationship ex-
   pressed by the sentence is true, in particular for that sumti value:

      mi [cu] vecnu ti   xu          ta   [vau]
      --      ===== --   <-          --
      I       sell  this Is-it-true? that.
      Is-it-true-that I sell THIS (as opposed to something else) to that?

   Similarly, "xu" following "vecnu" in the above example would question the
   truth of the bridi relationship by specifically asking whether "sell" is a
   true relation between the sumti.

   We will discuss how to answer a "xu" true/false question below in the sec-
   tion on selbri.

Varying sumti order:

   There are ways to vary the order of sumti from the numerical order speci-
   fied by the place structure.  A sumti may be placed out of numerical order
   by labelling it in front with a tag indicating the actual place structure
   numerical position of the sumti.  The structure is thus of the form "fi
   sumti" (where the "fi" category word shows which of the existing sumti
   places is being used, by number:

   fa/1st sumti - x1       fe/2nd sumti - x2       fi/3rd sumti - x3
   fo/4th sumti - x4       fu/5th sumti - x5

   One reason for using these tags is to skip a place structure place without
   having to insert a "zo'e" for each skipped place:

      mi    [cu] tavla fo  la lojban. [vau]
      --         ===== .>  ----------
      talker     talk  x4= language
      I          talk  in-language Lojban.
      I talk in Lojban (to someone about some topic).

   which is equivalent to:

      mi    [cu] tavla zo'e           zo'e              la lojban. [vau]
      --         ===== ----           ----              ----------
      talker     talk  audience       topic             language
      I          talk  to unspecified about unspecified in Lojban.
      I talk in Lojban (to someone about some topic).

   After a "fi" tag sets the place number, any later sumti places continue the
   numbering consecutively:

      mi    [cu] tavla fi ta      la lojban. [vau]
      --         ===== .> --      ----------
      talker     talk  x3=topic   language-x4
      I          talk  about that in-language Lojban.
      I talk about that in Lojban (to someone unspecified).








�   Another reason to use "fi" tags is to change emphasis; listeners focus most
   closely on the sumti at the beginning of a sentence.

      fi la lojban. [cu] tavla fa mi     do         [vau]
      .> ----------      ===== .> --     --
      x3=topic           talk  x1=talker audience-x2
      About Lojban       talk  I         to you.
      It's Lojban, that I talk to you about (in an unspecified language).

   There are other ways of rearranging the sumti of a sentence that will be
   discussed below.

   Note that in all examples where a sumti is omitted, the Lojbanist remembers
   that there is an unspoken and unspecified value for each of the omitted
   place structure places.

   An observative can be formed by using "fa" to move the first (x1) sumti to
   the position after the selbri.

      melbi      fa ti   [vau]
      =====      .> --
      beautiful  x1=this
      Beautiful  this-is (to someone by some standard).
      Beautiful, it is!

Basic structure of an utterance:

   People don't always say just one sentence.  Lojban has a specific structure
   for talk or writing that is longer than one sentence.  The entirety of a
   given speech event or written text is called an utterance.

      sentence .i sentence .i sentence [...] .i sentence
      ni'o sentence .i sentence .i sentence .i sentence [...] .i sentence
      ni'o sentence .i sentence [...] .i sentence
      [fa'o]

   "ni'o" separates paragraphs (covering different topics of discussion).  In
   a long text or utterance, the topical structure of the text may be indi-
   cated by multiple "ni'o"s, with perhaps "ni'oni'oni'o" used to indicate a
   chapter, "ni'oni'o" to indicate a section, and a single "ni'o" to indicate
   a subtopic corresponding to a single English paragraph.

   ".i" separates sentences.  ".i" is sometimes compounded with words that
   modify the exact meaning (the semantics) of the sentence in the context of
   the utterance.  ("xu", discussed above, is one such word - it turns the
   sentence from a statement to a question of truth.)  When more than one per-
   son is talking, a new speaker will usually omit the ".i" even though she/he
   may be continuing on the same topic.  It is still OK for a new speaker to
   say the ".i" before continuing; indeed it is encouraged for maximum clarity
   (since it is possible the 2nd speaker might merely be adding words onto the
   end of the first speaker's sentence).  A good translation for ".i" is the
   "and" used in run-on sentences when people are talking informally:  "I did
   this, and then I did that, and ..., and ...".

   "fa'o" is an optional end-of-utterance marker, used primarily in computer
   input.  It is not needed in human speech.




�   You may now see why the "vau" at the end of the sentence can generally be
   omitted.  Since the following word will usually be an ".i" or a "ni'o"
   starting a new sentence or paragraph, there is no possibility of ambiguity.
   These separators prevent the sumti at the beginning of the next sentence
   from being mistaken as a trailing sumti of this sentence.

Punctuation:

   Note that Lojban has no mandatory punctuation marks.  Because Lojban speech
   EXACTLY matches the written text representing that speech, all
   'punctuation' that is used in English to show sentence structure, ques-
   tions, exclamations or tone of voice, and even quotations must be expressed
   in Lojban as actual words that are spoken.

   The apostrophe (') is a letter representing a sound in the language.

   The period (.) is an optional reminder to the reader that a pause (which
   may be very short) is required between sounds at that point.  These may be
   omitted, but are almost always printed.

   Capitalization indicates unusual stress in the pronunciation of a name from
   a language other than Lojban.  It is NOT used at the beginning of a sen-
   tence.

   Commas (,) indicate unusual syllable breaks in pronunciation of a name or
   other word.

   There are some optional conventions that allow certain punctuation symbols
   to appear to clarify printed text, making it easier to read.  These punctu-
   ation symbols ALWAYS appear associated with the printed word representing
   that punctuation symbol.  Thus a "xu" question may be marked with a ques-
   tion mark immediately after the "xu" (or immediately before the "xu", pos-
   sibly inverted, like the Spanish punctuation style).  Other questions may
   similarly be marked with a question mark after the word indicating the
   question - NOT at the end of the sentence.  There are words that may be as-
   sociated with exclamation points, start of quotation (represented by "<<")
   and end of quotation (">>").


                 II. THE BASIC COMPONENTS (sumti AND selbri):

We now discuss the substructures of the basic components that make up a sen-
tence.  Any variety of selbri may be placed in a sentence, or another
substructure below that mentions 'selbri'.  Likewise, any variety of sumti may
be placed in a sentence, or another substructure below that mentions 'sumti'.
You may see that this can optionally lead to extremely complicated structures
nested within one another.  Lojban's unambiguous grammar allows even these
most complicated structures to be untangled in only one way.


                               A. SIMPLE sumti

sumti are not specific as to number (singular or plural), nor gender
(masculine/feminine/neutral).  Such distinctions can be optionally expressed
by being specific.






      Diagrammed Summary of Lojban Grammar Forms with Example Sentences

        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
 Permission granted to reproduce for no charge to the recipient, for purposes
                           of promotion of Lojban.
                              Updated April 1992

�"pronoun" sumti:

   These include the single-word sumti given above:

      mi            I/we
      do            you
      ti            this/these
      ta            that/those
      tu            that/those-yonder
      zo'e          something unspecified (it's either obvious or unimportant)

   Some other words in this category include:

      ri            he/she/it (the-last-referenced-sumti)
      ko            imperative you
      ko'a          it/he/she/they (a specific value)
      di'u          it/this (the last sentence)

   There are many others, each with a particular meaning.  For example, there
   are 9 other words related to "ko'a".  Each may be used to represent a sepa-
   rate value of "it".  Since Lojban has no gender or number, these 10 words
   represent "he", "she", and "they" as well, and it becomes more clear why so
   many are needed.

   "zo'e" is a place-filler allowing you to skip over a sumti place in the or-
   dered place structure without specifying a value.  The speaker indicates
   that there is a value, but that it is not important to specify it, or that
   the speaker thinks it is obvious given the context.

      do [cu] tavla zo'e   ti        [vau]
      --      ===== ----   --
      You     talk         about this.
      You talk about this. (to someone, in some language)

   "ri" is a quick back-reference sumti.  It may have a new ad-hoc meaning ev-
   ery time it occurs.  The rules for counting back to 'the last sumti' in-
   clude some special cases that can't be covered in this summary, but in most
   simple sentences, the referent will be obvious.  There are two other back-
   referencing sumti of this type.

      mi [cu] tavla fi la lojban.   ri              [vau]
      --      ===== .> ----------   --
      I       talk  x3=about-Lojban in-it(Lojban)-x4.
      I talk about Lojban in Lojban (to someone unspecified.)

   "ko" is used to express commands.  A statement with "ko" can be interpreted
   by replacing the "ko" with "do", and then taking the result as a command to
   the listener to make the sentence true for himself/herself as "do":

      ko        [cu] tavla mi   [vau]
      --             ===== --
      You (imp.)     talk  to-me.
      Talk to me!

   is the command equivalent of:

      do        [cu] tavla mi   [vau]
      --             ===== --
      You            talk  to-me.
      You talk to me.

�   "ko" need not be in the first position, but rather can occur anywhere a
   sumti is allowed, leading to possible Lojban commands that are very unlike
   English commands:

      mi [cu] tavla ko                  [vau]
      --      ===== --
      I       talk  to-you (imperative).
      Let me talk (to you)!

   "ko" is even permitted to occur in more than one place in a sentence, al-
   lowing for meaning-rich commands like:

      ko        [cu] tavla ko                 [vau]
      --             ===== --
      You (imp.)     talk  to-you (imperative).
      Talk to yourself!

   The English misses some of the meaning, since the Lojban expresses two com-
   mands at once: that the listener talk to himself, but also that the lis-
   tener allow himself to be talked-to (by himself).

Names ("la name"):

   Lojban names always end with a consonant followed by a mandatory pause
   (which may be very short).  No other Lojban word ends with a consonant.
   Thus names are easily recognized by both their form, and by being marked
   with "la".

      do [cu] tavla   la .an. ti        [vau]
      --      ======= ------- --
      You     talk-to Ann     about this.
      You talk to Ann about this.

question sumti ("ma"):

   "ma" indicates a question about the value of a sumti.  It is answered by
   'filling in the blank', replacing the "ma" with the intended sumti value.
   It can be translated as "Who?" or "What?" in most cases, but also serves
   for "When?", "Where?", and "Why? when used in sumti places that express
   time, location, or cause.

      ma  [cu] tavla do    [vau]
      --       ===== --
      __?      talks to-you.
      What/Who talks to you?

   is answerable by:

      mi [cu] tavla do [vau]
      --      ===== --
      I talk to you.

   Like "ko", "ma" can occur in any position where a sumti is allowed, not
   just in the first position.:

      do [cu] tavla ma [vau]
      --      ===== --
      You talk to what/whom?

�   "ma" can also appear in multiple sumti positions in one sentence, in effect
   asking several questions at once.

      ma [cu] tavla ma [vau]
      --      ===== --
      What/Who talks to what/whom?

   The two separate "ma" positions ask two separate questions, and can there-
   fore be answered with different values in each sumti place.

Description sumti ("le selbri [ku]"):

   "le" describes a sumti that the speaker has in mind by describing a bridi
   relationship that it forms the first (x1) sumti of.  This bridi is repre-
   sented by a selbri (e.g., one of the words described earlier in this pa-
   per).  Description sumti phrases have a terminator on the right side, "ku",
   which is omitted when no ambiguity results.

      mi [cu] tavla   le vecnu [ku] le blari'o [ku]           [vau]
      --      =====   ------------- ---------------
      I       talk-to the seller    about the blue-green-thing.

   "le vecnu" takes the selbri "vecnu", which has the "seller" in the x1
   place, and uses it in this sentence to describe a particular "seller" that
   the speaker has in mind (one that she probably expects the listener will
   also know about).  Similarly, the speaker has a particular blue-green thing
   in mind, which is described using "le" to mark "blari'o", a selbri whose
   first sumti is something blue-green.

   There are many variations on "le sumti [ku]" constructs, but to discuss
   them, we must first discuss the more complex structures of selbri.


                             B. selbri STRUCTURE

Though Lojban sentences often translate word-for-word into fairly clear En-
glish, selbri relations are actually quite unlike English.  The selbri "bajra"
expresses a relation of running.

      sumti  bajra     sumti      sumti       sumti
      -----  =====     -----      -----       -----
      runner [running] on-surface using-limbs with-gait

   In some sentence positions, "bajra" might be interpreted as the 'verb' "to
   run"; in other positions, "running".  In "le sumti [ku]", described in the
   preceding section, it represents the 'noun' interpretation of its x1 sumti
   place:  "runner".  (Some English words, like "cook", have similar proper-
   ties, but the analogy is weak.)

brivla:

   The simplest form of selbri is an individual word.  A word which may by it-
   self express a selbri relation is called a brivla.  The three types of
   brivla are gismu (root words), lujvo (compounds), and le'avla (borrowings
   from other languages).  All have identical grammar; they are allowed wher-
   ever "selbri" appear in these examples.




�   gismu:

      mi   [cu] klama ti          zo'e   zo'e  ta   [vau]
      --        ===== --          ----   ----  --
      Go-er     goes  destination origin route means.
      I go here (to this) using that means. (from somewhere via some route).

   lujvo:

      ta  [cu] blari'o [vau]
      --       =======
      That    is-blue-green.

   le'avla:

      ti  [cu] djarspageti [vau]
      --       ===========
      This     is-spaghetti.

   Some short words may serve as selbri, acting as variables that stand for
   another selbri.  The most commonly used of these is "go'i", which repre-
   sents the main bridi of the previous Lojban sentence, with any new sumti or
   other sentence features being expressed replacing the previously expressed
   ones.  Thus, in this context:

      ta  [cu] go'i                   [vau]
      --       ====
      That     too/same-as-last selbri.
      That (is spaghetti), too.

   The word "go'i", by itself, by repeating a bridi marked as a "xu"
   true/false question, repeats that bridi, thereby claiming it is true.
   Thus, in this sense only, "go'i" can mean "yes".  "xu" questions can also
   be answered "yes" by repeating the entire sentence in full, but "go'i" is
   much easier to say.

      xu ta [cu] blari'o [vau]
         --      =======
      Is-it-true-that that is-blue-green?

      go'i                   [vau]
      ====
      True.  (repeats "That is blue green.")

contradictory negation ("na selbri"):

   The negation particle "na" can occur at the beginning of any selbri.  It
   says that the relation claimed by the selbri does not hold (contradictory
   negation).  It may often be translated as "It is false that [English sen-
   tence]".

      mi [cu] na klama ti ta [vau]
      --      nn>===== -- --
      It is false that I go to this from that.

   The "na" is only permitted at the beginning of a complete selbri.  It is
   considered part of the selbri in other constructs in the language, but is
   restricted from other positions within a tanru (discussed below).


�   If the contradictory negation particle na precedes "go'i", the combination
   "na go'i" denies the relation claimed by "go'i".  Thus, after a "xu"
   true/false question, "na go'i" expresses the answer "False", or "No".

      mi [cu] ja'a go'i [vau]
      --      pppp>====
      It is TRUE that I do (go to this from that).

   If you were to use "go'i" after a sentence that contained a "na" contradic-
   tory negation, the negation would carry over to the repeated sentence.  Un-
   like English, "na go'i" would NOT form a double negative; it merely re-
   places the "na" by another "na" leaving the sentence unchanged.  Instead,
   you must cancel a negative by using the positive equivalent of "na"
   ("ja'a") to replace the previous sentence "na":

scalar negation ("na'e-word brivla"):

   "na" deals primarily with the truth or falsity of a bridi.  Lojban also
   supports a separate form of negation, called contrary or scalar negation.
   A scalar negation attaches tightly to the next brivla of the selbri, modi-
   fying the meaning of the word on some scale.  Scalar negation structures
   may appear anywhere where 'brivla' or 'selbri' is allowed.  Scalar negation
   words include "na'e" (other-than), "to'e" (absolute opposite-of), and
   "no'e" (neutral on the scale); "je'a" is a strong positive scale assertion,
   translating roughly as "certainly" or "indeed":

    |----------- POSITIVE ------------|------------ NEGATIVE -----------|
   je'a                              no'e             na'e             to'e

   Examples:
      mi [cu] melbi [vau]
      --      =====
      I am beautiful.

      mi [cu] na'e melbi [vau]
      --      ==========
      I am other-than beautiful.

      mi [cu] to'e melbi [vau]
      --      ==========
      I am ugly/opposite-of-beautiful.

      mi [cu] no'e melbi [vau]
      --      ==========
      I am plain/neutral on the beauty-ugliness scale.

      mi [cu] je'a melbi [vau]
      --      ==========
      I am indeed beautiful.

question selbri ("mo"):

   "mo", like its sumti relative "ma", is a fill-in-the-blank question.  It
   asks the respondent to provide a selbri that would be a true relation if
   inserted in place of the "mo":

      do [cu] mo [vau]
      --      ==
      You are-what/do-what?

�   "mo" may be used ANYWHERE a brivla or other selbri might.  Keep this in
   mind for later examples.  Unfortunately, by itself, "mo" is a very non-spe-
   cific question.  The response to the above question could be:

      mi [cu] melbi [vau]
      --      =====
      I am beautiful.

   or:

      mi [cu] tavla [vau]
      --      =====
      I talk.

   Clearly, "mo" requires some cooperation between the speaker and the respon-
   dent to ensure that the right question is being answered.  If context
   doesn't make the question specific enough, the speaker must ask the ques-
   tion more specifically using a more complex construction such as tanru
   (below).

   It is perfectly permissible for the respondent to fill in other unspecified
   places in responding to a "mo" question.  Thus, the respondent in the last
   example could have also specified an audience, a topic, and/or a language
   in the response.

Conversion ("se-word brivla"):

   "se" and others in its word-category modify a brivla used in a selbri by
   changing the order of the sumti that are attached, resulting a new selbri
   that expresses the same relation, but with different order of emphasis.
   "se" exchanges the first and second sumti places of the unmodified brivla.
   This reversal is called conversion.

   The bridi sentence:

      do       [cu] vecnu ta            mi       zo'e    [vau]
      --            ===== ----          --       ----
      seller-x1     sells thing-sold-x2 buyer-x3 price-x4
      You sell that to me.

   can be converted to:

      ta           [cu] se vecnu   do        mi       zo'e    [vau]
      --                ========   --        --       ----
      thing sold-x1     is-sold-by seller-x2 buyer-x3 price-x4
      That is sold by you to me.

   The effect is similar to what in English is called 'passive voice'.  In
   Lojban, however, a conversion is NOT 'passive':  the converted selbri has a
   place structure that is renumbered to reflect the place reversal, thus
   having effects when such a conversion is used in combination with other
   constructs (such as "fi" and "le selbri [ku]").









      Diagrammed Summary of Lojban Grammar Forms with Example Sentences

        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
 Permission granted to reproduce for no charge to the recipient, for purposes
                           of promotion of Lojban.
                              Updated April 1992

�   The other simple relatives of "se" are "te" (switches 1st and 3rd places),
   "ve" (switches 1st and 4th places), and "xe" (switches 1st and 5th places).
   The effects may be seen on the 5-place gismu selbri "klama":

      GO-ER  goes   to destination  from origin  via route  using means
      -----  =====  --------------  -----------  ---------  -----------
        x1   klama        x2             x3          x4          x5

      DESTINATION  is-gone-to  BY GO-ER  from origin  via route  using means
      -----------  ==========  --------  -----------  ---------  -----------
           x1      se klama       x2          x3          x4          x5

      ORIGIN  is-gone-from  to destination  BY GO-ER  via route  using means
      ------  ============  --------------  --------  ---------  -----------
        x1    te klama             x2          x3         x4          x5

      ROUTE  is-gone-via  to destination  from origin  BY GO-ER  using means
      -----  ===========  --------------  -----------  --------  -----------
        x1   ve klama            x2            x3         x4          x5

      MEANS  is-used-to-go  to destination  from origin  via route  BY GO-ER
      -----  =============  --------------  -----------  ---------  --------
        x1   xe klama              x2            x3          x4        x5

tanru ("modifier-selbri modified-selbri"):

   tanru are compound selbri - constructions of multiple brivla/selbri com-
   ponents.  Each component might be a single word, or it might be a word
   modified by "se", or scalar negations (like "na'e").  tanru take selbri
   components (including other tanru) in pairs, with the first part modifying
   the second part.

   The kind of modification is vague:  tanru may act like an English ad-
   jective-noun (fast-runner), adverb/verb (quickly-run) or it may restrict a
   larger set (runner-shoes).  Context will generally indicate what is a
   plausible interpretation of a tanru.  But allow for creative interpreta-
   tion:  "runner-shoes" might be interpreted in some imaginative instances as
   "shoes that run by themselves".  In general, however, the meaning of a
   tanru is determined by the literal meaning of its components, and not by
   any connotations or figurative meanings (sutra tavla - "fast-talker" thus
   would not necessarily imply any trickery or deception, and a "jikca toldi"
   - "social butterfly" must always an insect with large brightly-colored
   wings, of the family lepidoptera).

   The place structure of a tanru is always that of the final brivla/component
   of the tanru.  Thus, the following has the place structure of "klama":

      mi [cu] sutra klama la meris. [vau]
      --      =========== ---------
      I quickly-go to Mary.

   With the conversion "se klama" in the final position, the place structure
   is that of "se klama": the x1 place is the destination, and the x2 place is
   the go-er:

      mi [cu] sutra se klama la meris. [vau]
      --      ============== ---------
      I quickly am-gone-to by Mary.


�   A similar example shows that there is more to conversion than merely
   switching places, though:

      la tam. [cu] melbi tavla la meris. [vau]
      -------      =========== ---------
      Tom beautifully-talks to Mary.  or  Tom is a beautiful-talker to Mary.

   has the place structure of "tavla", but note the two distinct inter-
   pretations.

   Now, using conversion, we can modify the place structure order:

      la meris. [cu] melbi se tavla la tam. [vau]
      ---------      ============== -------
      Mary is beautifully-talked-to by Tom.  or  Mary is a beautiful-audience
      for Tom.

   and we see that the modification has been changed so as to focus on Mary's
   role in the bridi relationship, leading to a different set of possible
   interpretations.

   Note that there is no place structure change if the modifying term is
   converted, and hence less drastic variation in possible meanings:

      la tam. [cu] tavla melbi la meris. [vau]
      -------      =========== ---------
      Tom is talkerly-beautiful to Mary.

      la tam. [cu] se tavla melbi la meris. [vau]
      -------      ============== ---------
      Tom is audiencely-beautiful to Mary.

   and we see that the manner in which Tom is seen as beautiful by Mary
   changes, but Tom is still the one perceived as beautiful, and Mary, the
   observer of beauty.

   Any selbri form can be used in either position of a tanru.  This allows
   more specific "mo" questions:

      do [cu] mo tavla mi [vau]
      --      ======== --
      You are _____(what?)-kind-of talker to me?

      do [cu] tavla mo mi [vau]
      --      ======== --
      You are talker-ly _____(what?) to me?

   It was stated above that you can use scalar negation ("na'e" and its
   equivalents) in tanru:

      do [cu] na'e sutra tavla [vau]
      --      ================
      You are an other-than-quick talker. (or)  You are a slow talker.

      do [cu] sutra na'e tavla [vau]
      --      ================
      You are quickly other-than-talking. (or)
      You are doing something other-than-talking, quickly.


�quantified selbri ("number moi"):

   Lojban numbers are expressed as strings of digits.  The basic digits are:

      pa   re   ci   vo   mu            xa   ze   bi   so   no      pi
      1    2    3    4    5             6    7    8    9    0       .

   number moi (usually seen combined into one word) (ordinal numbers):

      le tavla [ku] cu ci moi [vau]
      -------------    ======
      The talker is third.

   number mei (usually seen combined into one word) (cardinal numbers):

      le tavla [ku] cu ci mei [vau]
      -------------    ======
      The talkers are a-threesome.

   number si'e (usually seen combined into one word) (portional numbers):

      le blari'o [ku] cu pimu si'e [vau]
      ---------------    =========
      The blue-green (things) are a .5 portion (a half).

   number cu'o (usually seen combined into one word) (probability numbers):

      le blari'o [ku] cu pimu cu'o [vau]
      ---------------    =========
      The blue-green (occurrences) are a .5 probability (a half).

   Note the interpretation of x1 in the last example, which is a result of the
   place structure of probability numbers.  Each of these special kinds of
   selbri have other places besides the x1 sumti that appear in these
   examples.

   number selbri may also be used as part of a tanru:

      mi [cu] papa moi tavla do [vau]
      --      ============== --
      I am the 11th talker to you.

   The place structure again is that of the final component of the tanru.

Attaching internal sumti to a selbri:

   Each component of a tanru is not merely a single-word brivla, but a rep-
   resentation of an entire bridi relationship.  Lojban grammar allows the
   sumti that complete and define that bridi to be combined into the selbri.
   Combined sumti are called internal sumti.  We'll first show the structure
   of such a complex selbri component:

     brivla/selbri be sumti [bei sumti][bei sumti] ... [bei sumti] [be'o]



�   where the sumti attached with "be" is normally x2, and other sumti are
   optionally attached in numerical order (x3, x4, x5), each preceded by the
   marker "bei".  "be'o" is the end-marker for internal sumti, appearing after
   the last internal sumti for a brivla or other selbri.  Let's now look at
   one way that this construct is used.

tanru with internal sumti:

   Using the internal sumti structure, any of the components of a tanru can
   have its own sumti:

      ta [cu] tavla be do   bei le melbi ku [be'o] vecnu [vau]
      --     |     |   --| |    -----------       |
             |      -----   ----------------------|
              ====================================
      That is a talker-to-you-about-the-beautiful-thing(s) salesperson.
      (or, more simply)
      That's a salesperson who talks to you about beautiful things.

   In compound constructs such as this one, the normally optional elidable
   [omissible] right terminators may be mandatory to keep the sentence un-
   ambiguous.  Thus, in this last example, either the "ku" or the "be'o" must
   not be elided ["ku" was chosen arbitrarily].  Otherwise, "vecnu" is
   absorbed into the internal sumti:

      ta [cu] tavla be do   bei le melbi vecnu [ku] [be'o] [vau]
      --     |     |   --| |    -------------------       |
             |      -----   ------------------------------|
              ============================================
      That is a talker-to-you-about-the-beautiful-salesperson. (or)
      That one talks about beautiful salespeople to you.

   Obviously, a different statement.  In Lojban, you MUST be careful about
   properly including terminators when needed.  If in doubt, include the
   terminator; the statement CANNOT be ambiguous with the terminator present.

   In the last example, by omitting the elidable terminators "ku" and "be'o",
   we ended up with sumti attached to the selbri word in final position.  The
   latter sentence is thus identical in meaning to the same sentence expressed
   without internal sumti:

      ta [cu] tavla do le melbi vecnu [ku] [vau]
      --      ===== -- -------------------
      That is a talker-to-you-about-the-beautiful-salesperson. (or)
      That one talks about beautiful salespeople to you.

   Internal sumti can use any sumti construct, including the fa/fe/fi/fo/fu
   series to rearrange place orders:

      ta [cu] tavla be do   bei fo la lojban. [be'o] vecnu [vau]
      --     |     |   --| |    -------------       |
             |      -----   ------------------------|
              ======================================
      That is a talker-to-you-in-Lojban salesperson. (or, more simply)
      That's a salesperson who talks to you in Lojban.





�tanru inversion ("modified-selbri co modifier-selbri")

   We rephrased the English translations of the Lojban in the last two ex-
   amples in order to simplify the English structure and make the sentence
   more clear.  The same type of rearrangement is possible in Lojban.  The
   technique is called tanru inversion.  The modifier selbri is placed AFTER
   the modified one, with a "co" separating them:

          modified-selbri co modifier-selbri
   The "co" can often be translated as "of type".

   tanru inversion affects the interpretation of sumti that are in the sur-
   rounding bridi relationship in a peculiar manner.  The inversion causes a
   new brivla to be in the final position, and any following sumti are
   interpreted as being associated with that final brivla of the modifier-
   selbri.  The sumti PRECEDING the selbri still are associated with the final
   term of the modified-selbri, because that is the most essential relation
   that is being claimed by the sentence.

   One obvious advantage of tanru inversion is to simplify the apparent
   structure of a selbri.  As we have said, the selbri in the final position
   does not need to use internal sumti structures to attach its sumti.  The
   first of the examples above that use internal sumti becomes the simpler:

      ta [cu] vecnu co tavla do le melbi [ku] [vau]
      --      ============== -- -------------
      That is a seller of-type talker to you about the beautiful thing(s).
      (or)
      That's a salesperson who talks to you about beautiful things.

   "do" and "le melbi [ku]" are the x2 and x3 places of "tavla", while "ta"
   remains the x1 of the underlying modified-relation, which is "vecnu".

   The resulting Lojban now matches the colloquial English phrasing more
   closely, and the sentence is simpler since it does not require the complex
   marker system needed for internal sumti within the selbri.

   All of the forms of selbri listed in this section are governed by rules
   that form a hierarchy.  The most complex constructs are those using rules
   higher in the hierarchy.  In general, constructs built from higher rules
   cannot be used inside lower-rule constructs.  This hierarchy of rules is
   the primary reason why Lojban's grammar is unambiguous.

   Inversion of tanru uses rules which are highest in the hierarchy, thus
   allowing you to invert almost all other selbri constructs.  However, this
   means that you also CANNOT substitute a tanru inversion into very many
   other constructs within a selbri.

selbri grouping in tanru:

   tanru may be composed of more than two components, any of which may be more
   complex than the simple brivla and/or other selbri structures discussed
   above.  Lojban allows complex tanru structures to be unambiguously
   expressed, so that any such complex structure can be broken down into a
   series of modifier-modified pairs.  We present two of the variety of ways
   to express more complex groupings.





      Diagrammed Summary of Lojban Grammar Forms with Example Sentences

        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
 Permission granted to reproduce for no charge to the recipient, for purposes
                           of promotion of Lojban.
                              Updated April 1992

�   In the absence of any grouping indications, components in tanru are pre-
   sumed grouped from the left:

      ti [cu] cmalu nanla ckule [vau]
      --      =================
      This    is-a small-boys school.  (This is a school for small boys.)

   But what if we want to group this selbri so as to talk about a "boys
   school" which is small.  One way is with tanru inversion:

      ti [cu] nanla ckule co cmalu [vau]
      --      ====================
      This    is-a boys school of-type small.  (This is a boys school which is
      small.)

   tanru inversion can work for many simple grouping problems.  However, since
   it changes the final brivla, it affects the interpretation of any sumti
   following the selbri.  There is a more general solution that does not
   affect which selbri is in final position, a choice that might be important
   because of the structural markers required.

   Any selbri, or any portion of a tanru that could stand alone as a selbri,
   may be surrounded with word-brackets "ke" (left) and "ke'e" (right) to
   indicate priority in grouping.  Normally, you will only use ke/ke'e
   grouping around strings of two or more selbri components, since the
   structure conveys no useful information around a single component.  At the
   end of the selbri and in other places where no ambiguity results, the
   "ke'e" terminator becomes optional (elidable):

      ti [cu] cmalu ke nanla ckule [ke'e] [vau]
      --      ===========================
      This    is-a small boys-school.

   A selbri structure surrounded by "ke" and "ke'e" has the same grammar as a
   single word brivla.  As a result, you can modify such structures with
   "na'e" and other scalar negation words, or with "se" and other conversion
   words:

      ti [cu] cmalu na'e ke nanla ckule [ke'e] [vau]
      --      ================================
      This    is-a small other-than-(boys-school).

abstraction selbri:

   The final selbri form we will cover in this paper isn't found very often in
   selbri in its most basic form.  However, it turns out to be one of the most
   important constructs in the language, showing up frequently as a part of
   more complex structures.  This selbri form is called sentence abstraction.
   The basic form is:

     nu-word complete-bridi-sentence [kei]

   where the bridi sentence inside the two markers can be a Lojban sentence of
   ANY type discussed in this paper, no matter how complex.  The word "nu"
   indicates an event abstraction, the most common kind of abstraction found
   in Lojban.




�      ti       [cu]  nu  mi    [cu] tavla do         [vau] [kei] [vau]
      --            |    --         ===== --                    |
                    |    talker     talks to-audience           |
                    |    I          talk  to you.               |
      --------       ===========================================
      event-x1       event-of "I talk to you."
      This is an event of my talking to you.

   The term 'event' should not be misconstrued.  In Lojban, it can refer to a
   momentary occurrence, or to a situation lasting hours, days, or even an
   indefinite period of time.  "nu" can stand for any of these kinds and dura-
   tions of events.  Other words may substitute for "nu" when you need to be
   specific and indicate particular event contours, such as a point event in
   time (mu'e) or a steady, unchanging state (za'i) of indefinite duration.

   There are other types of abstractions, as well, each indicated by words
   that substitute for "nu".  The most common of these are "ka" for a prop-
   erty/quality abstraction and "du'u" for a fact/assertion abstraction.
   These will be exemplified in the next section.  Most abstraction selbri
   have only one place (x1) which is the event, property, fact or other ab-
   stract 'thing' being described by the selbri.

   Abstractions, like other selbri, may be used in tanru; indeed, they are
   more common in tanru than alone:

      ti [cu] sutra bajra cukta [vau]
      --      =================
      This is-a fast-runner book.

   which might be a book about fast runners, while:

      ti [cu] sutra nu bajra kei cukta [vau]
      --      ========================
      This is-a fast-event(s)-of-running book.

   which is more likely a book about races, or a how-to book about running
   fast.  Note that the "kei" could not be elided in the last example, or the
   following would have resulted:

      ti [cu] sutra nu bajra cukta [kei] [vau]
      --      ====================
      This is-a fast event-of runner-book.

   and one imagines a very short-lived book about runners (it is the event
   that is fast, not the running or the book).

   Most abstractions that appear often in selbri like this, tend to be ab-
   breviated into a single-word selbri (a brivla) which embeds the "nu" into
   the concept.  Such compounds don't require a "kei", since the abstraction
   encompasses only the idea expressed within the word.  Lojban compound words
   (lujvo) are composed of combining forms of their component brivla (content
   words like "bajra") and cmavo (short structure words like "nu").








�   The rules for constructing lujvo are not difficult if you have a list of
   the combining forms (called rafsi); the rules are designed carefully to
   ensure that the pieces stay attached together and cannot be accidentally
   interpreted as separate words, since (as in the last example) the grammar
   of separate words may require added markers and terminators to obtain the
   meaning that you intend.  The lujvo for an event of running is formed from
   "nun", the rafsi combining form of "nu", and "bajra", which serves as its
   own combining form:

      sumti   nunbajra  sumti   sumti      sumti       sumti
      -----   ========  -----   -----      -----       -----
      event   [running] runner  on-surface using-limbs with-gait
      x1 is an event of x2 running on surface x3 using limbs x4, with gait x5

   Comparing this place structure with the one at the beginning of this
   section, you will see that x1 has been assigned to the event, while the
   remaining places are those of "bajra".  With this new lujvo, the next to
   the last example sentence:

      ti [cu] sutra nu bajra kei cukta [vau]
      --      ========================
      This is-a fast-events-of-running book.

   can use less-complicated structure:

      ti [cu] sutra nunbajra cukta [vau]
      --      ====================
      This is-a fast-(events-of-running) book.


                              III. COMPLEX sumti

More "le selbri [ku]" descriptions:

   Now that we have seen a variety of selbri forms, it perhaps becomes obvious
   that any of these selbri structures can be used in description sumti marked
   with "le".  Indeed, some of these structures, especially internal sumti and
   abstractions, are much more commonly found embedded in sumti, than in the
   selbri that defines the main relation of the sentence.

      ko [cu] tavla le sutra klama [ku] [vau]
      --      ===== -------------------
      You (imperative) talk to the quick-goer. (Talk to the quick-goer.)

      do [cu] klama le se tavla [ku] ta [vau]
      --      ===== ---------------- --
      You go to the one-talked-to from that.

   This last example shows the virtue of "se" conversion.  It has allowed us
   to mentally convert "tavla" to make its x2 place accessible by description
   in the sumti.

   Now for a tricky usage.  We use "go'i" to refer to the bridi of the last
   sentence.  Therefore "le go'i [ku]" refers to the first place of that bridi
   (in this case, the go-er "do").  If we want to refer to the second place of
   the last sentence bridi, the destination, we can mentally convert that
   sentence using "se", and "le se go'i [ku]" means the destination ("le se
   tavla [ku]").


�      le se go'i [ku] cu melbi [vau]
      ---------------    =====
      The destination is beautiful. (or)
      The one talked-to (the destination) is beautiful.

      le na'e melbi [ku] cu tavla [vau]
      ------------------    =====
      The other-than-beautiful one talks.

      mi [cu] tavla le mo klama [ku] ri [vau]
      --      ===== ---------------- --
      I talk to the what-kind-of go-er about itself?

      le re moi prenu [ku] cu tavla mi [vau]
      --------------------    ===== --
      The second person talks to me.

   It is IMPORTANT to remember and correctly use the elidable separator "cu"
   with description selbri.  If you misplace it or omit it (and its less-often
   used alternative "ku"), you will create some very strange tanru.

      le sutra [ku] cu vecnu cukta [vau]
      -------------    ===========
      The quick-one is a seller book.

      le sutra vecnu [ku] cu cukta [vau]
      -------------------    =====
      The quick seller is a book.

   And, if you omit the "cu", you get only a sumti:

      le sutra vecnu cukta [ku] [vau]
      -------------------------
      The quick-seller book.

sumti descriptions with internal sumti:

      le selbri be sumti bei sumti bei sumti ... bei sumti [be'o] [ku]
      |        |-------------------------------------------------|    |
       ----------------------------------------------------------------

   is all one sumti.

      le tavla   be la .an. bei le vecnu [ku] [be'o] [ku] cu klama [vau]
      |         |   -------     -------------       |    |   =====
      |           -----------------------------------    |
       --------------------------------------------------
      The talker to Ann     about the seller                 goes.

   You can even fill in the places of an internal sumti:

      le tavla be le melbi be la .an. [be'o] [ku] [be'o] [ku] cu klama [vau]
      |       |   |       |   -------       |    |      |    |   =====
      |       |   |        -----------------     |      |    |
      |       |    ------------------------------       |    |
      |        -----------------------------------------     |
       ------------------------------------------------------
      The talker to the one who is beautiful to Ann goes..
      (The one talking to the one Ann thinks is beautiful, goes.)

�   Without the omitted terminators this looks a bit less wordy:

      le tavla be le melbi be la .an. cu klama
      The one talking to the one Ann thinks is beautiful, goes.

   Since a selbri can be a tanru with internal sumti, a sumti may be built on
   such a selbri, possibly even having internal sumti on BOTH components of a
   tanru (in a tanru embedded in a sumti, even the sumti attached to the final
   component must be attached with be/bei/be'o).:

      le melbi be la .an. [be'o] tavla be la mark. [be'o] [ku] cu vecnu ti
      |       |   --------      |     |   --------       |    |   ===== --
      |        -----------------       ------------------     |
       -------------------------------------------------------
      The beautiful-to-Ann talker-to-Mark sells this.
      (The one Ann thinks is beautiful who is talking to Mark, sells this.

   Here, either the "ku" or the "cu" is elidable before the main selbri (both
   "be'o"s are unconditionally elidable).  Most frequently, when there is a
   choice, the terminator that best communicates the sentence structure is
   chosen.  "cu", in this case, clearly separates the complex sumti from the
   selbri, and is preferred.  Often, a single "cu" may allow you to omit
   several elidable terminators that would otherwise be necessary.  This
   happens most frequently with abstraction selbri that are used in sumti
   descriptions.

abstraction sumti clauses:

   le nu sentence kei:

      mi [cu] djica le nu mi [cu] klama le zarci [ku] [vau] [kei] [ku] [vau]
      --      ===== | |   --      ===== -------------            |    |
                    |  ==========================================     |
                     -------------------------------------------------
      I       want  the-event-of: I go to the store.
      I want to go to the store.

   With the omitted terminators not printed, this sentence looks much shorter:

      mi djica le nu mi klama le zarci
      I want to go to the store.

   An even shorter form will typically appear in Lojban text.  "le nu" occurs
   so frequently in combination that it is often written as a single word;
   this isn't mandatory - cmavo compounds are always understood as meaning the
   same thing as the words written separately.  cmavo are generally written
   one word when they together equate to a concept that is written in other
   languages as one word.

   In addition, the "mi" inside the abstraction will often be omitted.  When a
   listener hears this sentence and realizes that the go-er wasn't specified
   (as well as the origin, the route and the means), the obvious value(s) will
   be assumed.  Leaving out the "mi" is exactly comparable to the difference
   between the two English sentences:

      I want to go to the store.    (and)    I want myself to go to the store.

      mi djica lenu klama le zarci
      I want to go to the store.


      Diagrammed Summary of Lojban Grammar Forms with Example Sentences

        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
 Permission granted to reproduce for no charge to the recipient, for purposes
                           of promotion of Lojban.
                              Updated April 1992

�   If an abstraction is in the x1 position, "cu" allows four other elidable
   markers to be omitted.  For example in a conversion of the last example
   sentence:

      lenu mi [cu] klama le zarci [ku] [vau] [kei] [ku] cu se djica mi [vau]
      | |  --      ===== -------------            |    |   ======== --
      |  =========================================     |
       ------------------------------------------------
      The-event-of: (I go to the store) is desirable to me.

   the "cu" allows four other markers to be omitted.  Using "cu" is much
   easier for the listener, who thus knows in one word that the complex sumti
   is completely ended and the main selbri comes next.

   We promised to give examples of two of the other types of abstractions in
   this section.  These abstractions tend to be associated with specific
   places of particular brivla.

      la mark. [cu] ricfu   le ka melbi [vau] [kei] [ku] [vau]
      --------      =====   ----------------------------
      Mark is rich in the quality of x1 being beautiful to x2 by standard x3

      mi [cu] djuno le du'u la djan. [cu] te vecnu [vau] [kei] [ku] [vau]
      --      ===== -----------------------------------------------
      I       know  the-fact-that John is a buyer.
      I know that John buys (something).

   As with the earlier examples. these sentences will typically appear much
   shorter in print:

      la mark. ricfu leka melbi
      Mark is rich in beauty.

      mi djuno ledu'u la djan. te vecnu
      I know that John buys (something).

Quantified sumti ("le number [boi] selbri [ku]"):

   One way to quantify a sumti being described is to insert the number fol-
   lowed by a terminator "boi" that may be omitted when no ambiguity results
   (the usual case):

      le re [boi] tavla [ku] cu klama [vau]
      ----------------------    =====
      The two talkers go.

   "boi" may not be elided when a number selbri follows, since you wouldn't
   know where one number stops and the next begins.  (Lojban does not allow
   such boundaries to be expressed by contrastive stress as in English.):

      le reno boi remei [ku] cu klama [vau]
      ----------------------    =====
      The twenty twosomes go.







�Quantified selection from sumti ("number [boi] le [number] [boi] selbri
[ku]"):

   You can also put a number preceding the "le", to select from the set in-
   dicated by the description:

      re [boi] le tavla [ku] cu klama [vau]
      .......> -------------    =====

      Two of the (unspecified number of) talkers go.

      re [boi] le reno boi remei [ku] cu klama [vau]
      .......> ----------------------    =====
      Two of the twenty twosomes go.  (i.e. four altogether)

      le tavla be pa [boi] le ci [boi] bajra [ku] [be'o] [ku] cu melbi [vau]
      |       |   .......> ----------------------       |    |   =====
      |        -----------------------------------------     |
       ------------------------------------------------------
      The talker to one-of the three   runners                is-beautiful.

Indefinite description sumti ("lo selbri [ku]")

   If you wish to describe a sumti, but do not have a specific in mind, you
   can instead refer generically to something that meets the terms of the
   description selbri:

      lo tavla [ku] cu klama [vau]
      -------------    =====
      A talker goes.  (or)  Some talkers go.

   "lo" may be used interchangeably with "le" in the preceding examples, with
   an indefinite description as a result.

   Lojban allows you to omit the "lo" in:

     number [boi] [lo] [number] [boi] selbri [ku]

      re [boi] tavla [ku] cu klama [vau]
      -------------------    =====
      re tavla cu klama
      Two [of the unspecified number who are] talkers go.

"di'u" and "la'e di'u":

   In English, I might say "The school is beautiful", and you might reply
   "This pleases me."  How do you know what "this" refers to?  Lojban uses
   different expressions to convey the possible meanings of the English:

      le ckule [ku] cu melbi [vau]
      -------------    =====
      The school is beautiful.








�   The following three sentences all might translate as "This pleases me."

      ri [cu] pluka mi [vau]
      --      ===== --
      This (the school, the last expressed sumti) pleases me.

      di'u [cu] pluka mi [vau]
      ----      ===== --
      This (the last sentence) pleases me (perhaps because it is grammatical
      or sounds nice).

      la'e di'u [cu] pluka mi [vau]
      ---------      =====
      This (the referent of the last sentence; i.e. that the school is
      beautiful) pleases me.

   The last sentence is an example of using one sumti to point to or refer to
   another by inference.  "la'edi'u" is often written as a single word, and is
   used more often than "di'u" by itself.

Other sumti types:

   Lojban supports several other sumti types, more than we can discuss in a
   short paper.  These include bare numbers, several kinds of quoted text
   (single words, grammatical or possibly ungrammatical text, and non-Lojban
   text).


              IV. ATTACHMENTS TO sumti AND selbri AND sentences

                           A. ATTACHMENTS TO sumti

All structures in this section apply to sumti at the main level of a sentence,
as well as to sumti within substructures.

Adding a new sumti place to a bridi relationship ("modal + sumti"):

   time or location:  e.g. "pu"/before, "ba"/after, "ca"/simultaneous with,
   "vi"/at, "va"/near, "vu"/far from

   There are many more of these, and some specialized rules for compounding
   them.  These are discussed in a more advanced paper.

      mi [cu] tavla pu le nu do [cu] tavla [vau] [kei] [ku] [vau]
      --      ===== .> ------------------------------------
      mi tavla pu lenu do tavla
      I talk before the-act-of you talk.  (I talk before you do.)

   You must think carefully about what you mean with these constructions:

      mi [cu] tavla pu do [vau]
      --      ===== .> --
      I talk before you.  (I talked before "you" even existed.)







�   modal tag: e.g. "secau" (without ...), "mu'i" (motivational because ...),
   "du'o" (according to ...)

      do [cu] klama secau mi [vau]
      --      ===== ....> --
      You go without me.

modal questions ("modal + ma"):

   Express common English questions using time, location, and modal tag words
   combined with "ma":

      ca ma do [cu] tavla [vau]
      --            =====
      Simultaneous-with ____(what?), you talk?  (When do you talk?)

      vi ma do [cu] tavla [vau]
      --            =====
      At ____(what?), you talk?  (Where do you talk?)

      mu'i ma do [cu] tavla [vau]
      --              =====
      Motivationally because ____(what?), you talk?  (Why do you talk?)

      (Lojban has other "because" modal tags for asking a variety of different
      "why?" questions.)

sumti relative phrases ("sumti pe modifier-sumti [ge'u]"):

   A sumti may be identified more exactly by attaching a relative phrase,
   another sumti, that in some way restricts the possible set of things being
   referred to:

      le cukta [ku] pe le vecnu [ku] [ge'u] cu blari'o [vau]
      ------------- <  -------------       |   =======
                    <......................
      The book of the salesperson is blue-green.

   "pe", the basic marker of a restrictive relative phrase, is vague as to the
   exact nature of the relationship between the original sumti and the sumti
   that is identifying.  It is thus similar to the loosest English possessive,
   as in "my chair/the chair of mine", which may be used for a chair that you
   sit in but which is owned by someone else.

   Successively tighter degrees of association/possession are indicated by
   "po" (alienable possession) and "po'e" (inalienable possession):

      le cukta [ku] po mi [ge'u]
      ------------- <  --       |
      The book of mine. (Even if you are holding it, it is still my book.  But
      I also could give it to you such that it is no longer my book.)

      le birka [ku] po'e mi [ge'u]
      ------------- <    --       |
      The arm of mine (It is still MY arm; it cannot be given away.)





�   "po'u" identifies a sumti by giving another IDENTITY: a sumti that could
   equivalently replace the original:

      la djan. po'u   le vecnu [ku] [ge'u] cu klama [vau]
      -------- <      -------------       |   =====
      John     who-is the seller              goes.
      John-the-saleman goes.

   "goi", used like "pe", defines "ko'a" and other variable sumti for use
   throughout a text without repeating:

      le melbi tavla [ku] goi ko'a [ge'u] [cu] vecnu [vau]
      ------------------- <   ----       |     =====
      The beautiful talker (hereinafter "ko'a") sells.

      .i ko'a djica lenu do klama ko'a [vau] [kei] [ku] [vau]
         ---- ===== -----------------------------------
      He/she (the beautiful talker) wants the event of you going to him/her.

   Lojban has words identical in grammar to "pe" that provide non-restrictive
   (incidental) information about a sumti.  "ne" is the non-restrictive
   (incidental) equivalent of "pe":

   This construct may be combined with the modal construct discussed just
   previously to identify a sumti:

      la djan. ne pu la mark. [ge'u] [cu] melbi tavla [vau]
      -------- <  >. --------       |     ===========
      John, who was (incidentally) before Mark, is a beautiful-talker.

   The contrast between "ne" (incidental) and "pe" (identifying) is shown by
   giving the same sentence with "pe":

      la djan. pe pu la mark. [ge'u] [cu] melbi tavla [vau]
      -------- <  >. --------       |     ===========
      (The) John who was before Mark is a beautiful-talker.

   Certain modal tags, in fact, are designed primarily for use in relative
   phrases, rather than to attach additional sumti structures to the main
   bridi relationship as described earlier.  One of these is "semau" ("more
   than ..."):

      la djan. ne semau la mark. [ge'u] [cu] tavla mi [vau]
      -------- <  >.... --------       |     ===== --
      John, (incidentally) more than Mark, talks to me.
      (John talks to me more than Mark does.)

      la djan. [cu] tavla mi ne semau la mark. [ge'u] [vau]
      --------      ===== -- <  >.... --------       |
      John talks to me, (incidentally) more than Mark.
      (John talks to me more than he does to Mark.)









�   Without linking a "semau" modal sumti to another sumti with "ne" or "pe",
   it is hard to understand:

      ???la djan. [cu] tavla mi semau la mark. [vau]
         --------      ===== -- >.... --------
      John talks to me, more than Mark. (but Mark is tied neither to John nor
      me, but to the talking).
      (John's talking to me) is more than (Mark).

   Comparing a talking relationship to a person is nonsense.

sumti relative clauses ("sumti poi sentence ku'o"):

   "pe" phrases are limited to what can be expressed in a single sumti.  When
   you need to include more complete information about a sumti, Lojban
   provides for relative clauses.  A restrictive relative clause marker,
   "poi", marks a following complete bridi as providing detailed identifying
   information about the sumti by providing a relationship that the sumti fits
   into.

   The placeholder for the sumti being identified is "ke'a", which is merely
   another in the set of single-word sumti.  "ke'a" is often left out if it is
   contextually obvious where it would go (especially when "ke'a" would go in
   the x1 position immediately after "poi", or in the first available
   unspecified place if that x1 is provided in the clause).  A relative clause
   is terminated with the marker "ku'o" which may be omitted if no ambiguity
   will result.  It is very rare that "ku'o" needs to be expressed.

      le ckule [ku] poi   mi [cu] klama ke'a [vau] [ku'o] cu blari'o [vau]
      ------------- <     --      ===== ----             |   =======
                    <....................................
      The school    which I       go-to it                   is-blue-green.
      The school I go to is blue-green.)

   Note that "ke'a" refers to the school.

      le bajra [ku] poi [cu] tavla [vau] [ku'o] cu vecnu [vau]
      ------------- <        =====             |   =====
                    <..........................
      The runner who talks is a seller.

   There is also a non-restrictive relative clause marker, "noi", for in-
   cidental information about a sumti.

tensed sumti ("le time/location/modal-tag + selbri [ku]")

   A sumti may also have a time or location or modal tag in front of a de-
   scription selbri:

      le pu bajra [ku] [cu] tavla [vau]
      ----------------      =====
      The earlier/former/past runner talked/talks. (Since Lojban tense is
      optional, we don't know when.)

      le vi bajra [ku] [cu] tavla [vau]
      ----------------      =====
      The here runner talks.  (This runner talks.)




      Diagrammed Summary of Lojban Grammar Forms with Example Sentences

        Copyright 1990, 1991, 1992 - The Logical Language Group, Inc.
        2904 Beau Lane Fairfax VA 22031-1303 USA  Phone: 703-385-0273
 Permission granted to reproduce for no charge to the recipient, for purposes
                           of promotion of Lojban.
                              Updated April 1992

�short possessive sumti ("le possessor-sumti selbri [ku]")

   As a comparison with the last example, note that a description sumti may
   have a sumti after the "le", resulting in a short form of the loose "pe"
   possessive.

   English "this" might be represented by "vi" or by "ti", depending on
   context.

      le ti bajra [ku] [cu] tavla [vau]
      ----------------      =====
      The this-one's runner talks.  (This one's runner talks.)

   A more complete structure of a description sumti is "[number] le [number]
   [sumti] [modal] selbri [ku]".


                           B. ATTACHMENTS TO selbri

tensed or adverbial bridi relationships:

   Immediately after a "cu" before a selbri, you can have a modal (such a
   modal being there may make a "cu" unnecessary, since modals cannot be
   absorbed into tanru).  Such modals serve as an equivalent to English tenses
   and adverb functions.  In Lojban, tense is completely optional.  If
   unspecified, tense is picked up from context.

      do [cu] vu vecnu zo'e [vau]
      --      >. =====
      You yonder sell something-unspecified.

      le vi tavla [ku] [cu] ba klama [vau]
      ----------------      >. =====
      The here talker will go.  (This talker will go.)

      do [cu] mu'i tavla mi [vau]
      --      >... ===== --
      You motivatedly talk to me.


                         C. ATTACHMENTS TO sentences

A variety of constructs may occur anywhere in a sentence, operating inde-
pendent of the primary grammar of the sentence.  These constructs generally
have minimal internal grammar.  Like "xu", when not at the beginning of a
sentence, they indicate emphasis on the word or construct they immediately
follow.

Indicators:

   Indicators include a variety of expressions conveyed in English through
   interjections or tone of voice.  Lojban supports a enormous range of
   emotional expression through specific words and compounds.  Indicators may
   be modified for intensity, or classified by sphere (social, mental,
   emotional, physical, sexual, spiritual).

      .ie mi [cu] klama [vau]
          --      =====
      Agreement! I go. (Yep! I'll go.)

�      .ei mi [cu] klama [vau]
          --      =====
      Obligation! I go. (I should go.)

      mi [cu] klama le ckule .ui [ku] [vau]
      --      ===== --------(   )---
      I go to the school (and I am happy because it is the school I'm going
      to).

Discursives:

   Discursives allow free expression of certain metalinguistic comments
   (comments about the text).  Use of discursives allows clear separation of
   these metalinguistic features from the underlying statements and logical
   structure.  By comparison, the English words "but" and "also", which
   discursively indicate contrast or added weight-of-example, are logically
   equivalent to "and" without that discursive content.  The average English
   speaker does not think about, and may not even realize, the contrary-
   seeming idea that "but" basically means "and".

      mi [cu] klama [vau] .i do [cu] stali [vau]
      --      =====          --      =====
      I go.  You stay.

      mi [cu] klama [vau] .i ji'a do [cu] stali [vau]
      --      =====               --      =====
      I go.  In addition, you stay.  (added weight)

      mi [cu] klama [vau] .i ku'i do [cu] stali [vau]
      --      =====               --      =====
      I go.  However, you stay.  (contrast)

Evidentials:

   Evidentials indicate the speaker's relationship to the statement,
   specifically communicating how the speaker came to make the statement.
   These include "za'a" (I directly observe the relationship), "pe'i" (I opine
   that the relationship holds) "ru'a" (I postulate), and others.  Many
   Amerind languages use these type of words.

      pe'i do [cu] melbi [vau]
           --      =====
      I opine!  You are beautiful.

      za'a do [cu] melbi [vau]
           --      =====
      I directly observe!  You are beautiful.

   Some expressions overlap between categories, having value as indicators as
   well as evidential import (".ia", belief, faith), or attitudi-
   nal/metalinguistic import ("ga'i", hauteur, relative high-rank; cf.
   Japanese).

      .e'a ga'i ko'a [cu] citka lo titnanba [ku] [vau]
                ----      ===== ----------------
   Permission!/I permit! Hauteur! They eat cake.  (Let them eat cake!)




�                            V. LOGICAL CONNECTIVES

Lojban expresses logical connectives so as to unambiguously indicate logical
scope (what is being connected), as well as to isolate various non-logical
features from the logical ones (as noted above for discursives).  Logical
connectives may be expressed in forethought forms (both ... and ...) or
afterthought  forms (..., and ...).  The form of logical connectives directly
indicates their scope and the associated truth table for the two connected
terms.  For example, a bare vowel joins sumti (".a"/alternation, a and/or b, a
OR b; ".e"/conjunction, a AND b).  "na" and "nai" may be used to negate con-
nected terms ("na.a"/conditional, not a OR b, if a then b; "anai"/a or not b,
a if b).

Parallel structures are shown in the following abbreviated table:

     sumti              sumti forethought                   sentences
OR   a and/or b         .a       ga sumti gi sumti          .i ja
AND  a and b            .e       ge sumti gi sumti          .i je
XOR  a or b, not both   .onai    go sumti ginai sumti       .i jonai
->   if a then b        na.a     ganai sumti gi sumti       .i naja
  (a only if b)
<-   a if b             .anai    ga sumti ginai sumti       .i janai

     tanru              tanru forethought                   compound bridi
OR   a and/or b         ja       gu'a sumti gi sumti        gi'a
AND  a and b            je       gu'e sumti gi sumti        gi'e
XOR  a or b, not both   jonai    gu'o sumti ginai sumti     gi'onai
->   if a then b        naja     gu'anai sumti gi sumti     nagi'a
  (a only if b)
<-   a if b             janai    gu'a sumti ginai sumti     gi'anai

      mi [cu] citka ti .a ta [vau]
      --      ===== --(++)--
      I eat this and/or that.

      mi [cu] citka ti .e ta [vau]
      --      ===== --(++)--
      I eat this and that.

      mi [cu] citka ti na.a ta [vau]
      --      ===== --(++++)--
      I eat this only if that.

      mi [cu] citka ti .anai ta [vau]
      --      ===== --(+++++)--
      I eat this if that.

      ganai mi gi do [cu] klama [vau]
      +++++)--(++)--      =====
      If me, then you, go.

       mi [cu] klama [vau]  .i je  do [cu] klama [vau]
      |--      =====      |       |--      =====      |
      |___________________|(+++++)|___________________|
      I go AND you go.

      mi [cu] klama tu [vau]  gi'e  citka ti [vau] [vau]
      --     |===== --      |      |===== --      |
      --     |______________|(++++)|______________|
      I go to that-yonder AND eat this/these.

�Lojban also supports NON-logical connectives, such as mixed connection and set
membership (ordered/un-ordered):

      mi [cu] citka le blari'o joi pelxu [ku] [vau]
      --      ===== ----------(+++)----------
      I eat the blue-green-mixed-with-yellow (stuff).

      mi [cu] citka le blari'o ku joi le pelxu [ku] [vau]
      --      ===== -------------(+++)-------------
      I eat the blue-green (stuff) mixed-with the yellow (stuff).

   (The "ku" before the "joi" in the latter example CANNOT be left out due to
   the ambiguity resolution rules.)


                      VI. VERY BRIEF PRONUNCIATION GUIDE
              (covers only features occurring in these examples)

Vowels
a as in father, never as in sofa
e as in bet
i s in machine
o as in cocoa
u as in flute

Vowel pairs without apostrophes
ai   as /igh/ in "high"
au   as /ow/ in "cow"
ei   as in "weight"
ia   as the German "Ja"
ie   as /ye/ in "yep"
oi   as /oy/ in "toy"
ui   as the word "we"

Consonants as in English except
c sh as in "shoe"
j z as in "azure"
x hard ch of Scottish "loch" (breathe through a 'k')
g always hard, as in "go", NOT as in "germ"
s always as in "floss", NOT as in "flows"

A period indicates a mandatory pause, which may be very short.  An apostrophe
(') between vowels splits them into 2 syllables with no sound from your vocal
cords (usually like a brief 'h' sound) between.  There must be no pause, and
no 'voiced' sound: "e'o" should be a smooth transition from "e" to "o" that
doesn't sound like /ayo/ of "mayo".

Stress longer words (brivla) penultimately (the next to the last syllable).

      klama    /KLAH,mah/     blari'o   /blah,REE,ho/









� VII. BRIEF GLOSSARY OF LOJBAN WORDS USED IN THESE EXAMPLES

.a        and/or: sumti             ma       sumti question
.anai     only if: sumti            mei      cardinal # selbri
 ba       after/future time         mi        I; me; we; us
 be       internal sumti link       mo        selbri question
 be'o     end internal sumti        moi       ordinal # selbri
 bei      more internal sumti       mu        5
 bi       8                         mu'e      abstract.: point event
 boi      end-of-number             mu'i      modal: motiv. cause
 ca       same time/present         na        contradict. negation
 ci       3                         na'e      scalar other-than
 co       tanru inversion           na.a      only if: sumti
 cu       selbri separator          nago'i    last bridi false; No!
 cu'o     probability selbri        ne        incidental rel. phrase
 di'u     last sentence             ni'o      new topic
 do       sumti: you                ni'oni'o  new section
 du'o     modal: according to...    no        0; none
 du'u     abstraction: fact         no'e      scalar neutral
.e        AND: sumti                noi       incidental rel. clause
.e'a      indicator: permission     nu        abstraction: event
.ei       indicator                 pa        1
 fa       x1 sumti next             papa      11
 fa'o     end-of-text               pe        identifying phrase
 fe       x2 sumti next             pe'i      evidential: I opine!
 fi       x3 sumti next             pi        .
 fo       x4 sumti next             pimu      .5
 fu       x5 sumti next             po        alienable possession
 ga       and/or: forethought       po'e      inalienable possess.
 ga'i     indicator: high rank      po'u      identity phrase
 ganai    if: forethought           poi       identifying clause
 ge'u     end relative phrase       pu        before/past tense
 gi       forethought medial        re        2
 gi'e     and: bridi                remei     twosome, pair
 go'i     last bridi; too; yes!     reno      20
 goi      sumti definition          ri        the last sumti
.i        sentence link             ru'a      evidential: I assume!
.ia       indicator: belief/faith   se        x1/x2 conversion
.ie       indicator: agreement      secau     modal tag: without...
 ja'a     it is true                semau     modal tag: more than
 je       and: tanru, sentences     si'e      portion # selbri
 je'a     scalar positive           so        9
 ji'a     discursive: also          ta        that-there; those
 joi      mixed-with 'and'          te        x1/x3 conversion
 ka       abstraction: quality      ti        this-here; these
 ke       tanru grouping            to'e      scalar opposite
 ke'a     relative clause sumti     tu        that/those yonder
 ke'e     end tanru grouping       .ui        indicator: happiness
 kei      end abstraction           va        location: near
 ko       imperative You!           vau       end of sumti list
 ko'a     he/she/it - assigned      ve        x1/x4 conversion
 ku       end description sumti     vi        location: here/at
 ku'i     discursive: but           vo        4
 ku'o     end relative clause       vu        location: far from
 la       name follows              xa        6
 la'e     the referent of           xe        x1/x5 conversion
 la'edi'u bridi of last sentence    xu        Is it true?
 le       description sumti; the    za'a      evidential: observed
 ledu'u   the fact that ...         za'i      abstraction: state
 leka     the quality of ...        ze        7
 lenu     the event of ...          zo'e      unspecified sumti
 lo       description sumti; a

�.an.       name: Ann
 lojban.   name: Lojban
 mark.     name: Mark
 meris.    name: Mary
 tam.      name: Tom

brivla actually translate to a complete sentence indicating the place structures)

bajra
birka
blari'o      x1 is blue-green
bridi
brivla
citka
ckule
cmalu
cmavo
cukta
djarspageti
djica
djuno
gismu
jikca
klama
le'avla
lujvo
melbi
nanla
nunbajra     x1 event of x2 runs on surface x3, limbs x4, gait x5
pelxu
pluka
prenu
rafsi
ricfu
selbri
stali
sumti
sutra
tanru
tavla
titnanba     x1 is sweet-bread/cake
toldi
vecnu
zarci