From Lojban
Revision as of 07:48, 2 August 2014 by Gleki (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
Spoken in: n.a.
Timeline/Universe: n.a.
Total speakers: n.a.
Genealogical classification: a priori experimental language
Basic word order: VSO (sort of)
Morphological type: agglutinating, polysynthetic
Morphosyntactic alignment: logical language
Created by:
Jörg Rhiemeier 2005-2010

X-1 (for 'eXperimental language #1') was a loglang that had only one open word class, the predicate words fulfilling the fuctions of nouns, verbs, adjectives etc. in natlangs, and a self-segregating morphology.

X-1 is based on a 2005 discussion in the CONLANG mailing list about an article by Jeff Prothero titled "Near-optimal Loglan syntax" and incorporates ideas from Ray Brown and others. It was formerly named brz (a name coined by Ray Brown) but I dropped that name because it was meaningless and not even a morphologically correct expression in the language: b would be a uniliteral morpheme, and rz a fragment of a triliteral one - hence, brz would be garbage.

X-1 owes a lot to other people:

  • to Jeff Prothero, the self-segregation system based on morpheme lengths indicated by the number of initial consecutive '1' bits;
  • to Raymond A. Brown, the phonology;
  • to veritosproject, the variable-based syntax.


X-1 is fundamentally a language of bits. An X-1 utterance is, at least underlyingly, a bit stream. Each morpheme consists of one or more bit quartets. For purpose of writing it in a more human-readably way, the language uses the following 16 letters, each representing one of the 16 possible bit quartets:

j g l z ñ d µ b p m t n s r k h

How is this pronounced? The letters are mapped onto a system of seven consonants (/p t k s l m n/) and four vowels (/i E O u/) by the following rules:

Each letter has a consonantal value followed by a vowel. The vowels are inserted according to an automatic rule that is described below.

The letters are realized thus:

Bits Letter Pronunciation
0000 j zero followed by a front vowel
0001 g [k] followed by a back vowel
0010 l [l] followed by a front vowel
0011 z [s] followed by a back vowel
0100 ñ [n] followed by a front vowel
0101 d [t] followed by a back vowel
0110 µ [m] followed by a front vowel
0111 b [p] followed by a back vowel
1000 p [p] followed by a front vowel
1001 m [m] followed by a back vowel
1010 t [t] followed by a front vowel
1011 n [n] followed by a back vowel
1100 s [s] followed by a front vowel
1101 r [l] followed by a back vowel
1110 k [k] followed by a front vowel
1111 h zero followed by a back vowel

When looking closer at this chart, you will notice some regularities. The second half contains the same consonant values as the first half, in reverse order. In fact, a bit pattern and its one's complement (i.e., what you get when you flip all the bits) have the same consonant value. The frontness is indicated by the last bit of the literal: 0 gives a front vowel, 1 a back vowel.

The consonant values of the first half of the chart are not assigned arbitrarily. The odds are obstruents, the evens are sonorants. The systematic becomes clear in the following chart:

0000 zero 0001 /k/
0010 /l/ 0011 /s/
0100 /n/ 0101 /t/
0110 /m/ 0111 /p/

There are four vowels, namely /E/, /i/, /O/ and /u/. Whether the vowel is high (/i/, /u/) or low (/E/, /O/) is indicated by the first bit of the following literal. A 0 gives a high vowel, a 1 a low vowel. If there is no literal following, the vowel is high. (Hint: nothing counts as zero.)

For example, dt is pronounced [tOti] because the bit pattern is 0101 1010. Both literals have the consonantal value /t/. The LSB of d is 1: back vowel. The MSB of t is 1: low vowel. The low back vowel is /O/. The LSB of t is 0: front vowel. There is no following literal: high vowel. The high front vowel is /i/.


Morphology of X-1 is self-segregating. The length of a morpheme (in quartets) is indicated by the number of consecutive 1s at the begin of the morpheme, plus one. (This is a slight modification of the rule in Jeff Prothero's Plan B: I count the bits that come first in the bit stream, while Prothero counts least significant bits. But position values of bits do not matter in this scheme, only the bits themselves.) So, the morpheme length can be told by the first bits, or the first letter:

Letter Bits       Morpheme length (in quartets)
j 0000 1
g 0001 1
l 0010 1
z 0011 1
ñ 0100 1
d 0101 1
µ 0110 1
b 0111 1
p 1000 2
m 1001 2
t 1010 2
n 1011 2
s 1100 3
r 1101 3
k 1110 4
h 1111 5+

If the first letter of the morpheme is h , the sequence of consecutive 1s extends to the next bit quartet. For example, a morpheme beginning with ht is six bit quartets long. This way, you can have infinitely many morphemes.

Morphemes with at least three quartets are predicate words, which are the only open lexical class of X-1, taking the functions of nouns, adjectives and verbs. (Yes, nouns are predicate symbols, too. Think about it.) Biliteral morphemes (2 quartets) are connectives, and uniliterals (1 quartet) are variables (except j, which is a scope delimiter, indicating that variables in following clauses are not coreferent with variables in preceding clauses).

The arity (or valency, i. e. the number of arguments) of a predicate word is indicated by its length. The arity is always the length (in quartets) of the predicate word minus 2. Thus, 3-quartet predicates are unary, 4-quartet predicates are binary, etc.


A sentence consists of a sequence of clauses, which consist of a predicate word followed by one or more arguments. Each predicate word has a fixed number of arguments (see above). Arguments can be proper names or variables.



fox x; brown x; quick x; dog y; lazy y; jump.over x y;

The quick brown fox jumps over the lazy dog.