From Lojban
Jump to: navigation, search
banjubu'o / français / 日本語 / English
zantufa farvi

What is Zantufa?

  • Zantufa is one of parsers of texts in Lojban.
  • Zantufa is based on an unofficial but beautiful and simple grammar aiming at impartiality according to Guskant.

Expected characteristics of Zantufa

How is Zantufa related to the other variations of Lojban?

What is expected?

It is expected that people will leave the grammar explained by La Bangu and live in a good grammar. Good luck to Zantufa!


What Guskant desired is almost completed by Zantufa_1.

There rest only removing bugs and adding new cmavo of existing selma'o.

Please inform Guskant if you find any bugs or problems.

The principles of selection of grammatical elements of Zantufa are different from the tendency of Lojban IRC dialect explained by La Bangu.

Frankly speaking, Lojban IRC dialect is pursuit of usefulness for European languages speakers or computer geeks: the usefulness is not necessarily universal.

On the other hand, Zantufa pursues simplicity of the grammar itself, eliminates the unfairness and attaches importance to the conformity with logic. For that purpose, it sacrifices certain usefulness in certain cases.

Those principles originate from aesthetic values by Guskant. According to her, Lojban IRC dialect is a grotesque mass brought by reckless additions of "useful" functions to the old house (the official Lojban) that became somehow old-fashioned and incovenient: it lacks synthetic beauty. On the other hand, the structure of Zantufa is similar to that of traditional Japanese house: a room is not destined for a fixed usage but used for various purposes. Zantufa has not many but versatile grammatical elements for high expressiveness.

Here is a list of characteristics of Zantufa_1. However, Ilmentufa experimental or Altatufa may take some of the following characteristics into their own grammar, and then those parts will become not the characteristics of Zantufa only.

Most part of the temporary grammar suggested by Xorxes is inherited.
In other words, it has a little difference from Iocixes of peg of the temporary grammar (both became inaccessible) that inherited the whole temporary grammar.
Here is the detail.
The idea of removing capital letters as marks for stress is not adopted.
It is because Guskant cannot modify preproc.js so that it could take the place of grammar of capital letters.
Syllabic consonants are not modified.
It is because the modification would prevent from analysing many old Lojban texts.
However, Zantufa extreme might inherit it.
Words ke and ke'e still span over sumti or bridi-tail.
It is not much necessary, but it should be permitted for replacement of NUhI and NUhU by KE and KEhE.
Selma'o LAU, TEI and FOI are still in the grammar.
It is because selma'o TEI and FOI are useful for expressing Chinese characters in Chinese or Japanese method (部首).
For example, "tei remna bu sanli bu foi" has the same meaning as "zoizoi 位 zoi bu".
The similar method is frequent in spoken Chinese or Japanese, as well as in a Chinese input method.
Selma'o TEI and FOI are distinguished from BY in the grammar, and then it is easy to distinguish a Chinese character from others.
(Selma'o LAU is not much necessary for Chinese or Japanese, but kept for similarity to TEI in usage.)
Most part of MEX grammar proposal by Xorxes is inherited,
but the relations between operators and operands are modified for convenience.
System of conjunctions is based on the suggestion of Xorxes and that of Selpa'i,
but it is characteristic that Zantufa gives much expressiveness to forethought connectives. The semantics to profit from expressiveness is proposed.
Also, it gives importance to the statements between selma'o NU and KEI.
In other words, the statements between NU and KEI has a full logical property, for example
([CU {du'u «(¹[da de] zo'u¹) (¹da [CU {broda «de VAU»}]¹) (¹i je [de {CU «brode da VAU²)»}]¹)» kei} VAU] IAU).
This property is necessary from a point of view of the philosophy of Quine.
Selma'o FA of Zantufa is based on a suggestion by a Japanese Lojban group.
Briefly speaking, they suggested that we don't need to distinguish selma'o FA of sumti tag and selma'o SE of selbri/conjunction/tag tag.
If the whole idea were adopted, selma'o FA would take the place of selma'o SE,
for example "seva'u" would be replaced by "feva'u", and "segu ra gi ri" by "fegu ra gi ri".
However, Camxes parses them as ([{fe KU} {va'u KU}] VAU) and ([fe {gu ra gi ri}] VAU). If the whole idea were adopted, the resulting structure of many Lojban texts would be different from that of Camxes[3].
This is the reason for that Zantufa adopted only a part of the idea.
However, Zantufa extreme would adopt the whole idea.
(In that case, the frequent sound "f" is not very beautiful, and the sounds "fa fe fi fo fu" might be therefore replaced by "sa se si so su" or something other.)
"Non-possession, non-production, and non-introduction" of selma'o ZOhOI, MEhOI, DOhOI.
Those selma'o have function of quoting a series of non-Lojban symbols separated by spaces or periods in a series of characters, or by glottal stops or pauses in a series of phonemes; they are teached in La Bangu or Lojban Wave Lessons as "useful" (for European language speakers or computer geeks) cmavo for quoting one non-Lojban word. They are quite popular in the "Lojban" community, but rejected by Zantufa because of violation of the principles on the following points:
it is ugly as grammatical structure that the success of quotation depends on the properties of non-Lojban language of the quoted part, that one word is separated by spaces, periods, glottal stops or pauses;
it is unfair for the languages of the quoted parts that these selma'o cannot be used for Polynesian languages in which a word may include glottal stops, while they can quote even a full book in Chinese or Japanese in which spaces or periods are unnecessary.
The new cmavo are put in the existing selma'o as much as possible.
No selection among cmavo according to the usefulness is performed: all are accepted because adding whichever cmavo to existing selma'o does not affect the structural beauty of the grammar.
For example, new cmavo created by Curtis W Franks, so-called "kurtyvla", are mostly put in the existing selma'o. Most of kurtyvla are cmavo related to mekso, and many of them are accompagnied by suggestions of new selma'o. However, considering the usage, it is sufficient for the mekso structure of Zantufa to add them to existing selma'o.
i JOI and i (BAI) BO are conjunctions of sentences, not of fragments (terms, conjunctions, numerals etc.).
Fragments can be separated by i, but the fragments don't have truth value, and thus they are not object of conjunction in logic. If the fragments were treated on the same conjunction as sentences, that would be a violation of one of the principles of Zantufa, of attaching importance to the conformity with logic[4].
For the same reason, it is logically nonsense to permit conjunctions attached to selma'o NIhO, which separates a series of sentences and fragments. Because of that, Zantufa_1 or earlier do not permit conjunctions to NIhO.
However, under the condition that we are conscious that NIhO is not a logical element, we may not confuse the conjunction of NIhO with that of Logic. Zantufa_1.1 permits conjunctions to NIhO. The conjunctions of NIhO, just like the conjunctions in tanru, should be ignored in logical calculation. With this property, a significant part of lo se manci te makfa pe la .oz. is parsed by Zantufa_1.1: not all, because there are some parts that are discordant with the grammar of Zantufa, like "i ba bo connects fragments," "nu'i nu'u and ge gi are used together" etc. For the parser applicable to la Oz, see Maftufa.

Classes of Zantufa

These are the classes of Zantufa: Zantufa, Maltufa, Cekitaus and Maftufa.


Zantufa parses mainly the expressions of Guskant.


Maltufa, which Guskant began creating for people damning Zantufa[5], was modified in order to parse good old Lojban texts, for example la teris po'u lo tirxu cu vitke zi'o le barda tcadu.

Parsers Ilmentufa experimental and Ilmentufa camxes of the date 2015-08-10 UTC cannot parse it.
(Parsers Camxes and Jbofi'e parse it well because they have not yet adopted the new rule.)
(Similarly, Zantufa parses the new version la .teris. ku noi tigra cu stuvi'e lo barda tcadu, while Ilmentufa experimental, Ilmentufa camxes, Camxes and Jbofi'e (of the date 2015-08-10) cannot parse it[6].)


Zantufa and Maltufa have some sorts of Cekitaus that parse many Cekitaus texts.

For example, ka càtra lo vèrba is parsed by Zantufa_1 cekitaujoibus (it is very long, so divide the text into two parts to parse).

Most part of the idea of Cekitaus is adopted by Zantufa. However, for the importance of the selma'o GA, Zantufa cannot adopt exchanges (mo'oi ge) (i'au go) (zo'u go) (moi'oi gei ge'i) etc.

If you have another idea of exchange of cmavo, inform Guskant who can create a new sort of Zantufa or Maltufa of Cekitaus for you!


Maftufa parses lo se manci te makfa pe la oz.


  1. New voi is finally in la cekitaus, not in Zantufa principal, because word poi'i takes the place.
  2. The source of Ilmentufa was changed to what cannot be put under the power of the dictator after Guskant revealed to the members of lojban mailing list the detail of dictatorship of Gleki. Gleki explained however that the reason was unstability of "vrici" server, not remorse of his own dictatorship. Here is the evidence that is an extract of the archives of IRC #lojban channel:
    --- Day changed Tue Feb 02 2016
    12:45 < gleki> Ilmen: lo se benji be mi cu srana i sa'u la camxes bu'u la vrici zo'u na mutce zabna i la vrici ta'e co'u akti
  3. For constructing the same structure as ([fe {gu ra gi ri}] VAU) of Camxes under the condition that selma'o FA is replaced by selma'o SE, use cmavo ke: ([fe {ke «gu ra gi ri¹) GIhI» KEhE}] VAU) (cmavo ke for sumti is useful here, too). Even if selma'o FA is replaced by selma'o SE, the expressiveness of the language will not decline.
  4. There is also a parser that gives a word co'e elidible to each fragment (Altatufa of the date 2015-09-04 UTC). This parser changes a fragment into a sentence, and it allows us connecting a fragment and a sentence with a conjunction. It changes a fragment into a sentence, in other words, a certain truth value is given to each utterance. That is to say, we cannot speak anything without truth value unless a SU structure erases the parsed part (though Altatufa of the date 2015-09-04 UTC cannot even parse a SU structure). Taking away the liberty of speaking something meaningful without truth value, it is a folly!
  5. Most of Lojban IRC members and frequent Lojban speakers (who are the creator of La Bangu and the speakers of the Lojban IRC dialect that is explained by La Bangu) damn much Zantufa. The opinions of them are so illogical that Guskant will not sincerely respond. Zantufa is simply one of the parsers of texts in Lojban, and it forces nobody to do anything. Frequent abuse on Zantufa is similar to the situation that American English speakers heap abuse on Indian English. Despicable.
  6. Like this, Zantufa parses many lojban texts that cannot be parsed by Ilmentufa experimental. However, people who damn Zantufa say often that Zantufa is not of Lojban. This fact shows egocentrism of the "Lojban" community, saying "our dialects are modern dialects of Lojban, yours is not Lojban." If Zantufa that parses la teris were not of Lojban, what is Ilmentufa experimental that cannot parse it?

List of pages of Zantufa

In Lojban:

In French:

In Japanese:

In English: