From the Logical Languages Wiki
Jump to navigation Jump to search

This page might be incomplete — you are advised to refer to the article Voksigid on the lojban.org wiki.

Voksigid (loglang)
created in: 1991–1992
by: Bruce R. Gilson et al.

Voksigid, a “predicate language,” was a group project led by Bruce R. Gilson during the years 1991–1992. Voksigid was inspired by Lojban, but aimed for greater naturalism and ease of use. It is not clear whether formal logic played any role in the semantics. Very little documentation on the language survives, most of it in a few articles apparently written by Gilson on Frathwiki.[1][2][3][4]

History of the project

Bruce R. Gilson gave the following synopsis of Voksigid.

Voksigid was an attempt to construct a predicate language of a different type from those which had gone before. The first predicate language (Loglan, developed by James Cooke Brown), and its descendant Lojban, developed by Robert LeChevalier, both used word order to mark the various places in the predication. I felt that remembering which position meant which role in the predication might be beyond easy memorization for most people. In October 1991 I made a proposal to the Conlang group putting forward some ideas for a language that I envisioned. I assembled a development committee following that post, and for several months we worked on the language, which we named Voksigid. The language has a syntax which was (as mentioned in the proposal letter) somewhat influenced by Japanese (but reversed; Japanese is verb-last and postpositional, Voksigid is verb-first and prepositional), and a vocabulary based mostly on European language roots.

After some time, I became aware that many of the ideas we were using in the construction of Voksigid resembled the theory of case grammar propounded by Charles J. Fillmore […]. Had we been working longer on this project, perhaps there would have been more use of Fillmore’s ideas […].

The project was formally ended in a message I submitted as a final report to the Conlang group on July 23, 1992.[5]

Gilson’s final report attributes Voksigid’s failing momentum to extrinsic factors, rather than any deficit in the language itself. After several co-developers had to go on hiatus, the project stalled. According to Gilson, the flat organizational structure of the project became an impediment, since votes on design choices were often tied.[6]

Phonology and orthography

Voksigid has a deliberately average phonology, suggesting that it is aimed to have the global accessibility of an auxlang.

Consonant phonemes
Labial Alveolar Palatal Velar Glottal
Plosive p b t d k g
Fricative f v s (z) (t)ʃ (d)ʒ h ~ x
Nasal m n
Lateral l
Rhotic (r)
Semivowel w j

There is no contrast between /s/ and /z/ or between /l/ and /r/, but their “official” pronunciations are distinct, and words are still written with all of the letters ⟨s z l r⟩. This is not the case with other pairs like [t͡ʃ] and [ʃ] or [x] and [h], which are fully interchangeable.

/r/ can be any rhotic, including guttural R. /n/ assimilates to [ŋ] before /k/ and /g/.[2] Both /v/ and /w/ are phonemes, unlike in some languages with similar “worldlang” phonologies.

The vowels also allow a range of pronunciation.

Vowel phonemes
Front Central Back
Close i ~ ɪ u ~ ʊ
Mid e ~ ɛ o ~ ɔ
Open a ~ ɑ

There are three diphthongs, /ai̯/, /au̯/ and /oi̯/.

The alphabet is as follows.[2]

The Voksigid alphabet
Grapheme a b c d e f g h i j k l m n o p r s t u v w y z
Phoneme a ~ ɑ b t͡ʃ ~ ʃ d e ~ ɛ f g h ~ x i ~ ɪ d͡ʒ ~ ʒ k l m n o ~ ɔ p r s t u ~ ʊ v w j z

(Notable differences from the IPA are highlighted.)

Morphology and syntax

The parsing morphology may be self-segregating by means of stress, word-shape and lexical exclusivity. However, this cannot be verified from the available information.

There are three parts of speech: predicates (a.k.a. verbs, but including equivalents of nouns and other content words), tags (prepositions) and particles. Particles include structure words such as conjunctions and ending markers, and aspectual and modal markers. The syntax is strongly head-initial, with verb-first word order, modifiers to the right of their heads, and prepositions. However, the language predominantly uses suffixes.

Suffixes are used to nominalize predicates, and the resulting nouns can serve as modifiers when they follow other nouns. Other suffixes turn tags, particles or strings of particles into predicates. Suffixes are of the form consonant diphthong, whereas root predicates cannot end in diphthongs.

Two prefixes are mentioned in surviving documentation: mas and fem, both used to specify natural gender. The coda consonants of these assimilate to a following consonant in voicing and place, respectively.[3]

There is a rigid and minimalist structure to sentences.

All sentences begin with a verb, possibly preceded by aspectual or modal particles. All other elements of the predication are expressed by the use of prepositional phrases. These include the elements that would in English be expressed by subject and direct and indirect object, as well as adverbial phrases. No distinction is made between these phrases, and if more than one phrase is used, they may be placed in any order, depending on the speaker/ writer’s desire to emphasize one or another.[4]

Relative, coordinating and subordinate clauses are introduced with a conjunction. Relative clauses use a relative pronoun, lau. Alternatively, relative phrases can be used. These are described as follows:

Another equivalent of a relative clause in other languages is a phrase constructed with a special case tag governing a nominalization of the verb which was the clause main verb. The nominalization indicates which case the restricted phrase is supposed to fit. When necessary (in the rare instances where the restricted phrase is not going to be an argument of the clause verb) or desired for stylistic reasons, nominalize with cen (abstraction) and use lau as a pronoun to represent the restricted phrase, with its proper case tag. This will be designated a quasi-relative phrase.[4]

There are four ending markers, equivalent to Lojban’s terminators. These are necessary for syntactic unambiguity. One marker terminates sentences, including “all unclosed phrases and clauses”; one terminates clauses, which are begun by conjunctions; one terminates phrases, begun by tags; and one terminates modifiers so that subsequent modifiers modify the head word only, rather than the modifier.[4]


  1. Anonymous editor “BRG” (probably Gilson). 2009. “Voksigid.” Frathwiki. Accessed from https://www.frathwiki.com/Voksigid (19 June 2021).
  2. 2.0 2.1 2.2 Anonymous editor “BRG” (probably Gilson). 2009. “Voksigid orthography.” Frathwiki. Accessed from https://www.frathwiki.com/Voksigid_orthography (19 June 2021).
  3. 3.0 3.1 Anonymous editor “BRG” (probably Gilson). 2009. “Voksigid morphology.” Frathwiki. Accessed from https://www.frathwiki.com/Voksigid_morphology (19 June 2021).
  4. 4.0 4.1 4.2 4.3 Anonymous editor “BRG” (probably Gilson). 2009. “Voksigid syntax.” Frathwiki. Accessed from https://www.frathwiki.com/Voksigid_syntax (19 June 2021).
  5. Gilson, Bruce R. 2009. Voksigid. Archived at https://web.archive.org/web/20170427022212/http://viewsoflanguage.host56.com/voksigid/ (last accessed 19 June 2021).
  6. La Gleki. 2014. ”Voksigid.“ Lojban.org. Accessed from https://mw.lojban.org/papri/Voksigid (19 June 2021).