ce ki tau jau

From Lojban
Jump to navigation Jump to search

The ce-ki-tau-jau (and occasionally joined by du) dialect of Lojban has certain CV'V cmavo swapped with similar CV cmavo, as those CV'V are (far) more useful than the CV. The proposal can be generalized to a far wider class of proposed swaps. Thus, it is an umbrella-term. The word for this dialect in Lojban is " tcekitaujau "; in English it is often called "ce-ki-tau-jau" (possibly without the hyphens, using whitespaces instead), although this is ambiguous and perhaps "Cekitaujau" or, better, "Tcekitaujau" (or all lowercase) would be preferred.

In the following possibilities, each proposed swap (listed hereinafter) is independent of all others, with the exception that "X becomes Y" necessitates also that "Y becomes X". By "X becomes Y, and Y becomes X", it is meant that the old meaning of X is assigned as the new meaning of Y, while/and the old meaning of Y is assigned as the new meaning of X, and that only the new meanings are used (old meanings are deprecated and deleted from current dictionaries except as a historic note), at least in the case of cmavo. (For rafsi, "meaning" here means "gismu to which it is assigned, if any").

Primary proposal

In particular, it is suggested that:

  • ce'u becomes ce,
  • ke'a becomes ki,
  • tu'a becomes tau,
  • jo'u becomes jau (xorxes has proposed that joi [and zipcpi has proposed jei] should be used for jo'u; if that were the case, we wouldn't have to repurpose jau),
    • It might be nice to put all/many (basic) nonlogical connectives in jVV outfits (such that all jVV words are nonlogical connectives), particularly when considering certain proposals for the logical connectives.
    • We could develop a theme for NU to end with "-u".
    • Note that reassigning "jau" to something other than Cau will break the series in the non-decimal hexadecimal digits. Part of this series is already broken by assigning "rei" (E) to "xei" in that they are no longer alphabetically arranged in order from least to greatest in meaning.
  • du'u becomes du,
  • and, for each of the previous, vice-versa


Due to the proliferation of fancylojban and the general idea of the type system, ce'u and tu'a become very important cmavo, for maintaining precision and type-correctness, respectively. As for ke'a, it sees far more use than ki, and becomes especially useful with the proposal for new voi (assigned to poi'i for standard Lojban). jo'u has increased in usefulness due to the introduction of plural logic by xorlo.

Secondary possibilities

Other proposed swaps:

  • Replace rei/xei (one only) with sei or tei (one only) and then unassign the duplicate for "hex-E",
    • This would help preserve the series for non-decimal hexadecimal digits being in alphabetic order while representing monotonically increasing numbers.
      • Alternatively, rei/xei could be swapped with kei, mei, nei, pei, or (if it ever becomes a valid Lojban word) qei. The first two break certain vague similarities (many terminators begin with "k", many members of MOI are of form mVV). Or, vai could be swapped with Cai, where C is alphabetically strictly-later than the first letter of "hex-E"; presently the possibilities are sai, tai, vai (trivial), xai, zai; only the last option is available for "hex-E" being assigned to xei if we wish to maintain the alphabetic series and the diphthong series. Or, force vai to zai (by swapping) and either unassign rei (keeping only xei meaning "hex-E", which is rather mnemonic, incidentally), or swap rei/xei with vei while unassigning the duplicate for hex-E.
    • Something along these lines is actually constructive in maintaining/increasing the pattern-derived elegance in the vocabulary while also addressing the well-documented complaints concerning hex-E being originally assigned to rei. In any case, it would remove the duplicate, which is nice. Shifting later in the alphabet also reduces the chances of running into a commonly used (or, even, non-experimental) word, increasing back-compatibility in a practical but not theoretical sense. These characteristics are not common amongst the other individual proposals in this article.
    • See below concerning tei and to'e (modifiers of UI).
  • poi'i becomes voi (as mentioned previously),
  • su'o becomes su,
    • Alternatively, su'u becomes su, further reinforcing the theme for NU of often being of form consonant+u.
  • ka becomes ku,
    • See previous comment about the theme of NU
  • ni becomes xu
    • While this would disrupt the frequent usage of xu in older documents, it would again reinforce the earlier theme; additionally, quantitative cmavo very frequently begin with "x" in Lojban (see: xo. xo'e, etc.). There is no syllable cost. There may be an aesthetic benefit on both ends from this swap; moreover, some people may find "x" difficult or ugly to pronounce, making it annoying in such a common role as indicating a question. But having xo and xu so similar could be problematic (actually, it already could be).
  • bu'u becomes bu or zai,
  • ko'oi becomes koi,
  • si'au becomes si'u,
  • zu'ai becomes se'e,
  • mo'oi becomes ge (ce ki tau jau fully embraces the connective system simplification, making gaje a complete replacement for official ge),
  • moi'oi becomes gei (gei becomes ge'i?),
  • i'au becomes go,
  • zo'u becomes .u (the former-A-connectives may be a great space for "punctuation" cmavo due to similarity to .i . Similarly, perhaps ju'ei can be .e and zo'au (postnex) can be .a, though there may be better candidates for this precious cmavo-space),
  • lau'u becomes lau,
  • xu'u becomes po,
    • The very common combination of lo + NU to which xu'u is assigned could get po (terminator ku'au becomes po'e or nei) of selma'o PO, which directly turns bridi into sumti (equivalent to lo su'u). This cuts down Lojban's syllable count considerably.
    • The function of po/po'e is replaced by po'a, BAI of ponse; and possibly pesai and pecai for mere expression of strength/permanence of relation
  • fi'o becomes foi? (Did anyone want foi for something else?),
  • poi'a becomes pu'a, noi'a becomes nu'a, both of which are of selma'o NOIhA as described in this blog article,
  • and, for each of the previous options, vice-versa.
  • za'i changes its meaning to a modal for zvati. This add a symetry with tu'i/stuzi, and its more useful than its previous menaing.
  • ra'u is assigned to the old meaning of za'i, because the current meaning of ra'u can be achieved with NOIhA and perhaps a bridi-to-cnima'o converter.



  • Some changes in the dialect are not mere swaps, but are instead geared towards repurposing little cmavo. One such change is to perhaps use lau instead of lau'u to represent lo su'u go'i or la'e di'u. All of those see far more usage in the corpus than lau with its current definition.


Proposed swaps regarding attitudinal/scalar modifiers:

  • cu'i becomes cei(?),
  • ru'e becomes rei(?),
    • Criticism: Let xei be hex digit fourteen; we don't need two monosyllabic cmavo for it
  • ne'e becomes tei(??),
    • (highly speculative proposal; details need to be worked out) Attitudinals should be refactored so that nai is merely a simple negator like na'e, while tei takes the polar-negation sense of to'e. For example, .uinai would just signify "not happy / other-than-happy", while .uitei would signify "sadness".
  • and, for each of the previous proposals, vice-versa.

Being a basic building block of predicate logic, the existential quantifier is a good candidate for a monosyllabic cmavo form. As for bu'u, it is the spatial equivalent of ca, which has one syllable less. Both tenses are equally important.

rafsi swaps

Reassignment of rafsi has also been considered. For example:

  • In standard Lojban the rafsi of te does not parallel those of the other members of SE (se, ve and xe), so it has been proposed to use the rafsi -tel- instead.
  • Ditto for -nol- (representing no'e), to parallel -nal- (for na'e) and -tol- (for to'e).
  • -se'i- is proposed to stand for sefsi, not sevzi (jbovlaste already contains lujvo using the new assignment).

Activation of this mode

To mark usage of the ce-ki-tau-jau dialect in an utterance/text, one may/should use jo'au tcekitaujau. It is not clear how many of the previously mentioned individual proposals (which exact standard) are included by this referent.

In order to clear up this vagueness, some version names have been proposed; for example, see Proposal: version names for Tcekitaujau dialects.

Implementation process if universally adopted

This section concerns how the LLG would go about implementing this proposal throughout all of its official texts if it becomes the sole official version of the language (meaning that it is the default dialect in all up-to-date texts and conversations).

The update, even if every individual proposal is ratified, is relatively small; so, it could he handled expediently, to its credit - and benefit. We would update the version number of the language or give this update mode a name. Then we find all non-historic material (particularly learning material and wordlists) and perform a find-and-replace swap on the words, which should be automated easily - just replace all instances of a given word X (possibly surrounded by a single whitespace in order to avoid lujvo, but there can be issues with this) with another word XYLOPHONE which does not appear in the text (choose a non-Lojbanic garbage string surrounded by whitespaces as appropriate), replace the other word Y with the first one (X) throughout the document, then replace the garbage string XYLOPHONE with the second word (Y) throughout the same; repeat this for every pair of words to be swapped according to the standard. Afterward, mark or update the (new) version number on these documents which underwent the swap procedure and republish. Mark all documents or conversations which are not updated as historic (obeying the previous version's conventions), possibly with their own version number. In this task, the most difficult and time-consuming part would be in finding and sorting the documents.


The main flaw, according to many (most?) criticisms, is that the benefit of changing to follow most (or even ~all) individual proposals is relatively (or in some estimations, very) small. Thus, any accompanying cost can easily dominate.

  • This proposal destroys back-compatibility. However, it does not actually fix anything that is broken, it just shines up some perfectly good bits of the language a bit (at the level of vocabulary). So, no new functionality is included nor is any old functionality really improved except superficially. You know what they say about fixing things that ain't broken...
    • Relatedly, it forces relearning and re-accustoming. Other proposals add new features which need to be learned the first time, but this forces people to learn things that they already know for no new functionality.
  • Discussion and approval of this proposal would detract from other efforts, including fixing or expanding the language in other ways. The same goes for updating all material so as to conform. Any positive amount of discussion or effort at this level would represent a significant proportion of all discussion or effort; we just cannot practically dedicate our (human) resources to this task without being morally obligated to wonder about how it would be better spent.
  • The only benefit derived directly from this proposal is having shorter words for some "more common" meanings, and perhaps gaining some aesthetic beauty as well. Note that current amount of usage could be biased in some ways and may not actually reveal general usefulness.
  • Parallels and series in the vocabulary could be broken.
  • The scope of the proposal is not clear. Some of its individual proposals seem unrelated to the original intent and are included as extras because they do not really fit anywhere else. At the least, each change would have to be adopted individually - make sure to include the "vice-versa"'s!
  • There is still some disagreement within some individual proposals.
  • Some individual proposals (such as those somehow concerning the connectives) are predicated on the adoption of other (big) proposals.
  • Etymological connections could be obscured.
  • Some proposals reduce the explicit nature of Lojban (such as by reducing "la'e di'u" or "lo nu" into single words each), which obfuscates some of the underlying, carefully-crafted structure. This may be highly displeasing to some users, especially due to the introduced lack of symmetry in the vocabulary and grammar solely for the sake of saving syllables in certain (but not necessarily all) pre-existing texts.

External links

  • [1] Mailing list discussion sparked by a "joke" or "test" which announced the (untrue) adoption of Tcekitaujau by the BPFK.