Sapir-Whorfism: an approach to Lojban

tis an ill wind that blows no minds...

In this approach, the goal of Lojban learning is to conflict with one's preexisting linguistically-imposed conceptual assumptions, expanding perception and gaining the ability to conceive existence differently. Thus, are emphasized

  • Bellyfeeling Lojban, approaching level 1 fluency, thinking in Lojban, reading Lojban without a mental voice singing along an English translation.
  • Leaving the quirks and oddities of Lojban alone, for they may result in interesting conceptual knots (transcendentalism).
  • Preferring Lojban constructions that are very different from English to ones that are semantically unambiguous.
  • Preferring interesting, "aha!", metaphorical lujvo to ones that can be simply decoded easily.

I caricature the opposite approach, which may or may not be hardlinerism but is certainly marked by a disdain for SW effects in preference for "logical" usage, seljvajvo, Gismu Deep Structure, and an overriding emphasis on semantic unambiguity (in addition to and distinct from grammatic unambiguity!) as a "plastic soul" dorky notion that spoken language can be significantly purified. Apart from resulting from excessive youthful adoration of Star Trek Vulcans and Ayn Rand, it is philosophically corrupt but can be cured with the sting of the harsh rod of Logical Positivism, whereby the student realizes that all spoken languages are hopelessly sloppy and can never be coerced into anything resembling a formal system. Give up your dream, kids: the vast majority of sentences contain ill-defined concepts and unprovable relationships (as distinct from those provable but unproven), and this includes Lojban and even Objectivist Scripture. If you want a formal system, look to mathematics. If Lojban could help you, jboske would be generating theorems instead of endless, muddled, circular debates, year after grinding year.

Activity from these folks tends to have at its core a call for simple mapping of Lojban concepts into easily understood English, which is the polar opposite of an SW approach, obviously characterized by an allergy to English concepts. Thus it could be argued that only the SW way keeps Lojban from turning into a substitution cipher for English.

Grammatical unambiguity has been provided us by the language. Until we have a community of native speakers, and while everyone translates Lojban sentences into English in their heads, we will have no idea at all as to Lojban's semantic unambiguity relative to that of English, which is a function of the language's universe of discourse and not the language in isolation.

Gismu Deep Structure can be argued as a SW way to approach Lojban. However, the G places of lojban (where G is the number of all distinct tergi'u) are defined in fairly simple English terms, so Gismu Deep Structure limits us to a subset of basic English concepts, whereas breaking out of that closed world Gismu Deep Structure assumption and out of seljvajvo, allow us to create lujvo places that correspond to no easy English terminology; our only true way out of the substitution cipher problem.

Considering the prominence of SW testing in LLG propaganda, a newbie might be surprised to learn that this is an extreme minority view, with me (xod) as the only Lojbanist that would certainly sign on to this entire agenda. Michael Helsem and Bob LeChevalier have made noises in this direction, however, and at least are not implacably opposed to SWism. There is some indication that the older, Loglan generations were less opposed to SWism, but that the late 90s and on have seen an influx of young geeks who often have little interest in SWism.

Debate below. --xod

I think you misunderstand the whole point of looking at Lojban from
a Sapir-Whorf perspective. The idea is to attempt to experience the
effects of a language which washes its hands of natural-like things
such as lame metaphorical lujvo (which, btw, is a seperate category
than non-seljvajvo lujvo (which I accept as long as they aren't
metaphorical)), and which more rigorously defines its grammar and
its semantics. To me, Sapir-Whorf means Hardliner, otherwise
you're just testing <insert natlang> with a different name. As far
as I'm concerned, the above just describes a pseudo-subclass of
Naturalism (which is in my view directly opposite to attempting to
discover Sapir-Whorf effects, as it seeks to make Lojban more
natlang-like). --mi'e .djorden.

You don't seem to have understood my piece. --xod

  • But I think I did. I think it has nothing to do with Sapir-Whorf. Obviously you'll maintain that it does, so how about you add a paragraph which explains how you get the idea that it's somehow interesting from a Sapir-Whorf perspective to use figurative lujvo (which will undoubtably import figurative semantics from the source language-A prime example of this is litkycku for wiki). mi'e .djorden.

What on earth does a source language have to do with lujvo? Even in the example you cite there is none: "liquid book" is hardly an English idiom that's being unfairly mapped into Lojban. --xod

  • The metaphorical use of 'liquid' is derived from existing concepts we have in other languages. 'litki' does not mean every English meaning of 'liquid', it only means the state of matter, so using it to form lujvo with this sort of meaning consists in adopting metaphorical meanings of the keyword in le rarna bangu pe le lujvo finti. Which is why I consider 'litkycku' malrarbau. --mi'e .djorden.

The point is to try to avoid unthinking adoption of English metaphors, not to abolish metaphor altogether. You fellows are trying to reproduce the semantics of a computer language here, aren't you? But all throughout the CLL and the gi'uste are little notes about metaphorical usages, so your extremism is an innovation, which I think never occurred to the people who created Loglan or Lojban. Anyway, it's not really enough of a criticism to protest that English might have used the phrase "liquid book", and therefore that bans the use of "litki" in a Lojban metaphor. Given that metaphor in Lojban has a long history and was sanctioned by the designers, must we forbid any usage which is conceivably used by any natlang, anywhere? Are we forced to do a metaphor patent search (check me out with the metaphors!) before we construct one? Is it really uncalled-for and malglico to take the sloshing-around aspect of liquid and use it for a metaphor? While you may validly complain about the use of white as a symbol for purity, for instance, I think the fact that liquids have no definite shape is a universal observation by all species that experience liquid. We must be allowed to create fresh, new metaphors, as Helsem did with likcku, even if there is a small chance that some of them will accidentally resemble non-Lojban constructions. --xod

These allusions to your opponents being computer geeks, and trying to bring in programming language semantics (which doesn't even make any sense to me, as an accusation), serve no purpose except to try and paint people in a poor light. --JayKominek

Yes, I openly admitted that I was using "caricature" and not an fair and objective description by any means. There are other wiki pages to paint the "other side" in a positive light. But as far as importing the semantics of computer languages, perhaps I didn't word it optimally, but there are no metaphors in those languages and that seems to be the ideal state of affairs for certain people who voice objections to SWism, who I think are trying to drive Lojban as close as possible to a formal system. --xod.

the vast majority of sentences contain ill-defined concepts and unprovable relationships (as distinct from those provable but unproven), and this includes Lojban and even Objectivist Scripture

Leaving aside Objectivist Scripture, just because we may not be able to succeed in completely formalizing every aspect of Lojban does not mean that we cannot make progress toward a more formal definition of the semantics, and does not mean that the attempt is not worthwhile. Formal semantics has gone a long way toward formalizing many aspects of natural language, and there's no reason why Lojban shouldn't take advantage of that. In fact, formal logic was developed to partially formalize natural language, so going by your argument, we shouldn't have a problem saying "la djan. .e la meris. cu penmi" for "John and Mary met" or "ro da prami da" for "there is something that everyone loves". Additional formal structures or formal explanations of existing structures are just continuing the tradition which is built into the risna of Lojban — its logical basis. I don't have a problem with figurative lojban, and I think it has a purpose, but I think non-figurative lojban has at least as important a role to play as figurative lojban, if not more so. I think that there is room for two registers in Lojban.

At least in one sense, only the non-figurative camp is truly trying to explore the Sapir-Whorf implications of Lojban. Every human culture uses metaphor, so in order truly expand your mind in a way that cannot be done within the framework of any existing language, you have to learn to think literally.

Activity from these folks tends to have at its core a call for simple mapping of Lojban concepts into easily understood English, which is the polar opposite of an SW approach, obviously characterized by an allergy to English concepts. Thus it could be argued that only the SW way keeps Lojban from turning into a substitution cipher for English.

Nick has noted that just because something can be explained in English (even explained simply in English) does not mean that it can be easily expressed in English, with the classic example being ce'u. ce'u is a relatively easy concept to explain in English, and one which resulted out of jboske discussions, but try translating lo'e jboselbau cu sarxe do le ka ce'u jinvi makau le jbobau. At best it's stilted.

In addition, if you make no attempt to understand English semantics formally, you are nearly certainly doomed to unconsciously transfering your English semantics over to Lojban. You may succeed in a couple places to come up with something new, but language is a huge system, and without systematically analyzing it and what the possibilities are, you're unlikely to change it significantly.

your extremism is an innovation

Yes, that's true, but that doesn't mean that it isn't worthwhile. Lojban is an innovation too.


I meant that this is an innovation in the history of Lojban. Your points are interesting and attractive; but like I alluded to above, there's no experiential evidence that Lojban is closer to a formal system than English, looking at the divergence of discussions held in it, even when the topics are restricted to small areas. If after all these decades we have no indication of progress towards a formal system, then to keep moving toward that goal is an act of faith. Discussions that would be converging deductively instead are petering out in confusing heaps of chaotic exhaustion. --xod

I meant that Lojban is an innovation in the history of Loglan. All good things were innovations at one time. I'm not sure what you mean by "experiential evidence", but I do think that jboske has made significant progress on understanding a great many issues, such as: ce'u, details of nu (such as what Nick just realized), details about how masses work, lo'e/le'e, tu'o, ka'e/ca'a, makau, me+sumti+moi, roi as sumti tcita, fa'a, tu'o, how scoping works between individual sumti, etc. I have revised how I use the language on numerous occasions as a result of jboske discussions. Of course, not all of these issues have been completely finalized, but progress has been made towards formalizing them, and some of them are satisfactorily well defined.

I think that you're misunderstanding the intent of the jboske enterprise. Jboske does not attempt to turn Lojban into a formal system in the mathematical sense, with axioms and inference rules and theorems, but rather to formalize the definitions of various language constructs. Also, there is not just one Right Way to do jboske, and so there's no reason to expect that everyone will agree about everything. This is, ultimately, mostly about how to define things, and there are certainly many ways to define something formally. Discussions may sometimes lead to a consensus, but even if they do not, the issues are often better understood at the end when they peter out.

No matter how much jboske progresses, in any sentence or sentence fragment, there will always be some part of the semantics which has not been formalized and must be glorked intuitively, and it may be advantageous to leave some parts of the language informal, but as jboske progresses, the amount of semantics that must necessarily be glorked can be reduced. I think one can draw an analogy between scientific inquiry and jboske: scientific inquiry is constantly increasing its understanding of the natural world, but there will always be new things to discover.

Furthermore, Lojban is only interesting in the Sapir-Whorf sense because of its logic and formalness. The original intent of Loglan was to see how a language which had a bit of its semantics formalized in logic could show Sapir-Whorf effects. Gismu Deep Structure, seljvajvo and further formalizations of other things serve to augment this and, if the Loglandian Sapir-Whorf hypothesis is correct, can only be expected to sharpen the Sapir-Whorf effects. If all I wanted was a language which was very different from English in a Sapir-Whorf way, I would skip Lojban, with its rules about scope for every sumti (actually for every term), intensional contexts for sisku, etc., ad infinitum, and go straight to Klingon or Quechua. --Adam

I believe that every new Lojbanist since 1998 is under 27 and knows perl! — xod
But its OK if they're 30-something and know Java? --anon. coward

  • What do you mean by "OK"? Clearly my having written this page places me out of the category of the people who I claim object to this page. Or am I wrong for noticing patterns in who has been attracted to Lojban recently? --xod
    • beats me, why on earth is it relevent to your argument? maybe you could make a RecentTrendsInLojbanists page, rather than tossing things willy nilly into an otherwise reasonable discussion as though they support your position.

      • Then again, I suppose you could stick to the substance of the discussion instead of letting yourself be distracted by style. --xod

(Our brief but vivid tsouris over hexadecimalism can be viewed as a naive attempt to take this point of view seriously. After all, we DO have words for A, B, C, D, E, & F (10 through 15)...)

Created by AndRosta. Last Modification: Monday 22 of September, 2003 22:14:28 GMT by AndRosta.