on the baseline conformance imperative

From Lojban
Jump to navigation Jump to search

Original text

Consider these scenarios.

  • Scenario A. String S1 is not well-formed Lojban. Knowledge of Lojban can be used to interpret S1 in a principled and coherent way. S1 is excluded by the grammar for no principled reason. You use S1.
  • Scenario B. String S2 is well-formed Lojban, with official meaning M1. You use S2 to communicate not M1 but M2. You get understood without problem. This happens often.
  • Scenario C. String S3 is well-formed Lojban, with no definite official meaning. You use S3 to communicate meaning M3. The correspondence between S3 and M3 is inconsistent with the general patterns and principles governing correspondences between strings and meanings. You get understood without problem. This happens often.

If we think of there being a law, "Thou shalt conform to the baseline", then, I would argue, Scenario A violates the letter but not the spirit of the law, while Scenarios B and C violate the spirit but not the letter.

In Scenario A the content of the baseline is not being subverted. It is being fruitfully expanded. Nothing licit according to the baseline is rendered illicit or has its meaning changed. You are not subverting the baseline but are merely using a revised version of the grammar that simply renders licit what was formerly for no good reason illicit.

In Scenario B it becomes more likely that force of habit will overrule the baseline, to the point that anyone wanting to be understood would have to accept that S2 corresponds to M2 and not to M1.

In Scenario C it becomes more likely that internal inconsistencies, irregularities and ambiguities enter the language/grammar/langue, thus subverting guiding principles of its design.

One may perfectly well opine that the Loglan/Lojban experiment is more interesting if free rein is given to Scenarios B & C. But I take that as an argument that the experiment is more interesting if users do not respect the baseline.

--And Rosta


Commentary & discussion

Consider these scenarios.

  • Scenario A. String S1 is not well-formed Lojban. Knowledge of Lojban can be used to interpret S1 in a principled and coherent way. S1 is excluded by the grammar for no principled reason. You use S1.
    • A lot of these "no prinicipled reasons" are actually fully coherent if you bothered to listen to them. ka'enai AhA+NAI being the great example -- Cowan has explained why it was deliberately excluded (and it was a good decision imho; I think it wouldn't have been a bad idea to exclude PU+NAI also, but they survived). --mi'e .djorden.
    • Clearly you are thinking of some scenario other than Scenario A. In Scenario A, S1 is excluded by the grammar for no principled reason. Let's describe Scenario D:
      • Scenario D. String S1 is not well-formed Lojban. Knowledge of Lojban can be used to interpret S1 in a principled and coherent way. S1 is excluded by the grammar for a principled reason. You use S1.
    • Here the baseline violation depends on the nature of principled reason, chiefly whether the principled reason serves to avert undesirable knock-on consequences on the rest of the grammar.
  • Scenario B. String S2 is well-formed Lojban, with official meaning M1. You use S2 to communicate not M1 but M2. You get understood without problem. This happens often.
  • Scenario C. String S3 is well-formed Lojban, with no definite official meaning. You use S3 to communicate meaning M3. The correspondence between S3 and M3 is inconsistent with the general patterns and principles governing correspondences between strings and meanings. You get understood without problem. This happens often.

If we think of there being a law, "Thou shalt conform to the baseline", then, I would argue, Scenario A violates the letter but not the spirit of the law, while Scenarios B and C violate the spirit but not the letter.

  • I don't believe that you honestly believe that that is somehow the spirit of "conform to the baseline". In fact, I don't see why the spirit of such a statement has to be discussed, it is really quite clear.
  • I think you're confusing the spirit and the letter. At any rate, I am quite sincere, and I gave my reasons in the following paragraph. Of the reasons I can think of for why there should be a baseline, none are subverted by Scenario A: no aspect of the existing grammar or usage is altered or invalidated; everything still means what it always did. I believe that the baseline has two main functions. Firstly, it is a piece of PR to assuage people's fears that they will invest time in learning something that will become obsolete. Secondly, it serves to ensure that as far as possible a text written in 'correct' Lojban yesterday means the same thing as the same text written in 'correct' Lojban tomorrow, and that there is some objective and common standard for assessing what a text actually means, regardless of who has written it. And Rosta

In Scenario A the content of the baseline is not being subverted. It is being fruitfully expanded. Nothing licit according to the baseline is rendered illicit or has its meaning changed. You are not subverting the baseline but are merely using a revised version of the grammar that simply renders licit what was formerly for no good reason illicit.

  • The baseline provides a means for deciding whether ANY and EVERY utterance is syntactically correct Lojban. It does not merely say "these utterances are valid, and everything else is iffy" it says "these are valid, and everything else is invalid". There are not utterances whose state is illdefined. Their semantics, sure, but not the syntax.
  • This is trivially true for any explicit rule system. The grammar defines the set of well-formed sentences, and everything else in the universe and noosphere is not a well-formed sentence. But I contend that writing in a language other than Lojban is not contrary to the spirit of the baseline, and that neither is writing in a language that is not Lojban, but that can be understood on the basis of knowledge of Lojban. And Rosta

In Scenario B it becomes more likely that force of habit will overrule the baseline, to the point that anyone wanting to be understood would have to accept that S2 corresponds to M2 and not to M1.

  • Not if further baseline-compliant teaching materials would be written that would deal with the places that are problematic.
  • Indeed not, if further baseline-compliant teaching materials tend to eradicate Scenario B. But there is a strong body of opinion that argues that usage itself should be the major force in defining the language.

In Scenario C it becomes more likely that internal inconsistencies, irregularities and ambiguities enter the language/grammar/langue, thus subverting guiding principles of its design.

One may perfectly well opine that the Loglan/Lojban experiment is more interesting if free rein is given to Scenarios B & C. But I take that as an argument that the experiment is more interesting if users do not respect the baseline.

  • I would agree with this once there were users in the true sense. In the meantime the only way to 'bootstrap' things to a situation where it can actually be treated as a language is to force things to have a "correct" meaning. That said; if we ever get to a state of real usage, even then there would need to be some amount of restraint to keep things parsable, etc, however this sort of restraint would likely happen naturally (people don't generally go around proposing massive changes to languages people actually speak, because constructs have well defined meanings with the full force of a speaking population to prevent excessively rapid change. Jordan
    • I see you are a Fundamentalist but not a Naturalist. So we are not so far apart. And Rosta
  • (Unfortunately some of these would-be language mungers are attracted to lojban, but I digress]. --mi'e .djorden.
    • It is hardly surprising that language-mungers turn to a language created by mung(e)ing. Why be interested in Lojban? Because it offers things natlangs don't. For people such as xod and maikl, it offers new expressive and conceptual possibilities. For jboskepre of past and present, it offers logical explicitness and nonambiguity. It is natural for either group to want the language to meet their needs, especially in those respects that natlangs don't satisfy. And Rosta

Your scenarios are flawed.

  • I don't see why. They're both intrinsically possible and consistent with what you say below. And Rosta
  1. The utterance in question can be malformed, or not. This can always be decided for Lojban. If its in question, you're simply not dealing with Lojban.
  2. There is or there isn't an official meaning of the utterance.
    1. You do or don't intend to say what the official meaning of the utterance is.
      • No dispute about any of this. And Rosta agree, and what you say is compatible with each Scenario.
  1. You're producing invalid Lojban.
    • Neither of the other attributes are relevent. It is invalid. End of story.
  1. Official Meaningness:
    • There is an official meaning.
      1. You ignore it intentionally.
        • You're a dork.
      1. You don't get it right, accidentally.
        • If you make an attempt to correct yourself when the problem is pointed out, good for you.
        • If you don't attempt to fix your Lojban, you're a dork.
          • So those people who use dikyjvo when the lujvo isn't dikni (cyclical, periodic) or call something jbofi'e that doesn't finpe are dorks? (On the other hand, both examples are lujvo, which have one defined meaning that is not necessarily guessable from its constituents AIUI.) mi'e .filip.
          • Sorry, but nitcion must agree with my fellow fundamentalist (Jay, or Jordan?) To wilfully ignore a community norm even when one knows what the norm is, and to do so deliberately rather than out of neglect, is humpty-dumptying. Let's not pretend Lojban is anywhere near naturally evolving yet. dikyjvo is a mistake, and I am being a dork when I forget we now agree it is seljvajvo instead. jbofi'e is a cmene, and thus exempt. (I would certainly look at anyone saying lo jbofi'e rather than la jbofi'e as flirting with dorkhood.) Same kinda sorta goes for mriste -- nitcion
      1. You get it right.
        • Good for you.
    • There isn't an official meaning.
      • Filling in these spots is the sole redeeming feature of the jboske.
        • Again And Rosta basically agree with all this, except for the following:
        1. Your last sentence, of course.
        2. There is not a clear-cut dichotomy between ignoring official meaning and making an attempt to correct yourself; there are intermediate degrees of scrupulousness. I feel that the prevailing levels of scrupulousness are unhappily low, but scrupulousness adds enormously to the effort of producing and comprehending Lojban text, often with zero or even negative communicative benefits.
        3. I don't want to call people dorks just because they ignore official meaning. It's their prerogative, and some are patently not dorks. Furthermore, as I said in the previous para, the price of avoiding 'dorkiness' is a high one -- much effort, and it might even make communication less effective. So, they're not dorks; they're just acting against the spirit of the baseline. And Rosta
        4. nitcion don't disagree with the last sentence. This is a problem, therefore.
        5. Increasing scrupulousness is a worthy aim, even if it is ultimately unrealistic. (So is the entire Lojban venture, after all.) Lojban may be slowly turning into a language of self-flaggelators ("oh, if only I would remember the scope of quantifiers better!"), but the scrupulousness is instructive in what most of us want to learn from Lojban. Some portion of the Lojban community has a responsibility to encourage people towards such scrupulousness. I occasionally tried in my lessons, though not necessarily consistently enough. I am undecided as to whether the jboskeists are successfully doing this job, or even whether they should be expected to.
        6. Baseline means baseline policing; else it is meaningless. If you believe in a baseline, you believe in calling violators of the baseline dorks. This is a shift from the old "let usage decide" days, but then again, shit happens. Some dorkiness may well boil down to being a human being, and being plain unable to recall all the bits and pieces of the grammar. I plead human dorkiness for the routine misparsings of my Lojban, for example (e.g. in the Lojban Anthem, .i doi BRIVLA... cu... , which I thought would work (cu closes off sumti) but doesn't (cu actually separates sumti, and doi does not introduce a sumti). I don't think that quite lets me off the hook; and I believe there is a place for an ideal, Standard Lojban, even if noone actually speaks it. Nick
          1. I personally try not to correct people's usage; I try to figure out what they meant and respond to that, because in general I feel those of us who claim to seek fluency should demonstrate that by aiming for discussions that do not become meta-discussions about lojban grammar. --la xod
        7. And Rosta'm fed up with defending jboske. Intellectual enquiry shouldn't have to be defended. So I will try to just get on with it and not be diverted into defending it.
        8. Nobody's more fanatical for getting quantifier scope right etc. etc. than I am, so I don't want to exaggerate our differences. But do we really want to say that if you join the Lojban community you must necessarily consent to having your usage policed and to risk being called a dork? That sort of attitude seems more appropriate for a specific school or subcommunity of speakers who agree to police one another and to be called dorks.
        9. As I argue above, the functions of the baseline include (a) ensuring that there is a standard language that exists independently of speakers and usage, (b) allaying learners' fears about wasting effort on learning things that might become obsolete. When the baseline was declared it was not with the expressed intention that usage itself would be policed for baseline conformity.
        10. Borrowing an analogy from Lojbab, I'd rather see us as in a liberal society where the different political parties are loyal to the Crown or Constitution, and where nonextremists don't endeavour to enlist the police in enforce their own ideology on others... But in such a society, you & me wd probably be in the same party. And Rosta
          1. But different factions :-) Jokes aside, I have concrete reasons to be worried about this, as you know, with the upcoming Dictionary (which is not amenable to the Civil Society model, since it's not a "live and let live" project, but an "as many concrete answers as we can force people to agree on" venture.) And I will be doing my damnedest to keep everyone both focussed and effective. It may well fail; but I have a duty for now to antagonise both major parties... :-1/2 nitcion
          • The concerted effort to document Lojban beyond the "25%" is a different matter either from jboske in general or usage in general. Whether or not we agree about the value of jboske and the proprieties of usage, I imagine we'll be fairly close in agreement about what needs to be done for the Dictionary. This takes us onto the topic of the Dictionary and away from the topic of the baseline conformance imperative, so can move to another page... --And Rosta

Commentary and Discussion are all good, but the document you lot have produced is rather taxing to read as most of the additions lack attributions. Shy or lazy? --la xod

  • Attributions are there, but I'll make them explicit. -- n
    • They are there but implicit? Is this is joke? Are you Nick? --la xod
    • Yes, I'm Nick. By implicit, I meant the "Nick Nicholas" trick; I've added explicit "foo" where easily recoverable. There are a few attributions completely missing, but I'm assuming they're Jordan rather than Jay.
      • Ah, I see. The letter "i" being so thin, it's hard for me to see it's a different color. --la xod
      • I think they are from Jay: I'm pretty sure I marked all of my comments on this page: --mi'e .djorden.