From cowan@ccil.org Sun May 04 08:47:45 2003 Return-Path: X-Sender: cowan@mercury.ccil.org X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-8_2_6_6); 4 May 2003 15:47:45 -0000 Received: (qmail 41162 invoked from network); 4 May 2003 15:47:44 -0000 Received: from unknown (66.218.66.216) by m15.grp.scd.yahoo.com with QMQP; 4 May 2003 15:47:44 -0000 Received: from unknown (HELO mercury.ccil.org) (192.190.237.100) by mta1.grp.scd.yahoo.com with SMTP; 4 May 2003 15:47:44 -0000 Received: from cowan by mercury.ccil.org with local (Exim 3.35 #1 (Debian)) id 19CLiL-0005YN-00; Sun, 04 May 2003 11:47:45 -0400 Date: Sun, 4 May 2003 11:47:45 -0400 To: oskar2379 Cc: lojban@yahoogroups.com Subject: Re: [lojban] Re: Some ideas/questions (long) Message-ID: <20030504154745.GX28808@ccil.org> References: <20030504030759.GQ28808@ccil.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.3.28i X-eGroups-From: John Cowan From: John Cowan X-Yahoo-Group-Post: member; u=212516 X-Yahoo-Profile: johnwcowan X-Yahoo-Message-Num: 19586 oskar2379 scripsit: > Sorry about posting this twice, I don't know where the other one > went. I've never used these boards before so I am kind of lost. No problem; it came to me only. > > It would also eliminate the difference between tanru and lujvo, an > > essential Loglan/Lojban concept. > > But why is this seperation so important? Because tanru are intentionally imprecise. They allow speakers and listeners to communicate intuitively. Lujvo are by intention as well-defined as gismu, which means they may be broad and vague, but not ambiguous. > Yes, but then again there are a lot of asymmetrical tanru (your book > has a whole section on it...there's pixra cukta (picture > book)...cinfo kerfa (lion mame)...and even fenki litki, the example I > used). Is this order used in apprehension that lojban learners will > be mostly English-speaking? Well, book readers had better be English-reading, or they won't get much out of the book. :-) The general intention with tanru is that the more important semantic component goes at the end. > The reason I said false is that there is nothing in fenki that > *causes* craziness, so I don't see how that meaning could be > extracted. It's all right for there to be missing information from tanru: the price of infinite precision is infinite verbosity. If we add an element for cause, the next question will be "But just *how* does the liquid cause craziness?". And if we pin that down, the question can be repeated indefinitely. > > You are reading too much into a casual example. Lojban does not classify > > relationships this way. > > You mean there are more specific classifications? My meaning was that Lojban doesn't really classify relations at all: the gismu list is a large and disorderly bag whose merit is that it blankets semantic space, not that it divides it up neatly. > > The point is that an object which doesn't have any contents, at least > > potentially, has no claim to be a kabri. > > But then, shouldn't kabri be a lujvo? It could be like my 'le jubme > klama' example, only it would be 'le kabri vasru' (again, using my > word order). It *could* be expressed as a lujvo. For that matter, "mamta" could be expressed as a lujvo too. The gismu are not intended to be a minimal basis set. > > Because the presence of the lid is what makes it a botpi rather than > > a kabri or perhaps a patxu or even something else. "Lid" can be > > identified with cork or stopper as well. > > So what about the x2 place of gacri? We could say 'le botpi se gacri' > to specify a jar that is covered. That works too. > If we applied this kind of method to the entire language, until every > word follows the system, we could potentially eliminate a lot of > words and thus have enough three-letter combinations for every word. Feel free to tackle the problem! (It's a big one.) > To explain that you weep because of an onion, could you not use the > physical causation ri'a? > And emotional response could use the motivational mu'i. Sure. > But if it *depends* on something, wouldn't the statement be > conditional? > I understand that the reaction also depends on x1, but x1 is what > commits the action so that's a given. Nothing "given" about it, or rather the dependency on both subject and environment is what is "given". In another worldview, the subject would be mostly irrelevant, and all the credit for the response would be due to the environment. > Right, so the state of x2 after it has been reflected is x4 (sorry if > my wording is confusing). But that's not a change in x2. The fact that you are or are not reflected in a mirror at some instant does not represent any alteration to your state, any more than the fact that I am or am not looking at you does. Your wording is not confusing, but it is (metaphysically) confused. > In other words, it refers to the part of x2 that was changed (the > location). In minra, the part of x2 that was changed is the image. A location (and an image) are not parts! > > You are not allowing for auto-locomotion. The earth klama, but it > does > > not benji (there is no agent pushing it around the sun, except > > metaphorically). > > Couldn't you just delete the x1 place of benji with zi'o? That would > eliminate the need for having two seperate predicates. We revel in having lots of separate predicates. Certainly using zi'o (a late addition to the language, historically) would enable many to be removed. > I always thought of it this way. An object reflecting something is > seperate from sending it to another location. Absolutely. When an object is sent somewhere, it is no longer where it was before. When an object is reflected, nothing has happened to it at all! > But then again, I may > have misunderstood x3. Is it sort of like 'by standard/frame of > referance'? It sort of messes up the system because all the other 'by > standards' are attributive like 'foolish' and 'small'. Not really. The paradigm case is that you (x3) look into a mirror (x1) and see the image (x4) of a cat (x2). The image of the cat is not the cat; exactly what the image is, depends on where you are. > 'x1 translates x2 to x3' could handle all cases if you specified, > somehow, the language of x2 (direct object) and the language of x3 > (later state). Such as, I translate this which is in Latin to this > which is in German. Is that too far of a stretch? That works, but at the sacrifice of convenience. You should read Chapter 12, Section 16 for some insight into the processes (not entirely consistent, to be sure) used to assign gismu place structures. > exceptions...*shudder* :) Lojban isn't a computer language. It is large and complicated, and nowhere more so than in the gismu. The place structures are the product of many hands writing and rewriting different sections of the list over a long period of time, and absolute consistency is not to be looked for. -- Knowledge studies others / Wisdom is self-known; John Cowan Muscle masters brothers / Self-mastery is bone; jcowan@reutershealth.com Content need never borrow / Ambition wanders blind; www.ccil.org/~cowan Vitality cleaves to the marrow / Leaving death behind. --Tao 33 (Bynner)