There are 15 messages in this issue. Topics in this digest:
1. Celestial Laefêvëši From: Andrej �uc 2a. Re: If you only had 16 characters per day... (a language for Streetp From: Mechthild Czapp 2b. Re: If you only had 16 characters per day... (a language for Streetp From: taliesin the storyteller 2c. Re: If you only had 16 characters per day... (a language for Streetp From: taliesin the storyteller 3. More Logic & Semantics From: Logan Kearsley 4a. Re: In Defense of Monster Raving Loony Alignment From: Logan Kearsley 5a. Re: Allophony in Siye From: Anthony Miles 5b. Re: Allophony in Siye From: Roger Mills 5c. Re: Allophony in Siye From: Alex Fink 6a. Active/Stative Distinctions and Pronominal Prefixes From: Anthony Miles 6b. Re: Active/Stative Distinctions and Pronominal Prefixes From: Logan Kearsley 6c. Re: Active/Stative Distinctions and Pronominal Prefixes From: Jörg Rhiemeier 6d. Re: Active/Stative Distinctions and Pronominal Prefixes From: Alex Fink 7a. Re: Conjunction Curiosity From: Anthony Miles 7b. Re: Conjunction Curiosity From: Christophe Grandsire-Koevoets Messages ________________________________________________________________________ 1. Celestial Laefêvëši Posted by: "Andrej �uc" ashu...@gmail.com Date: Wed Aug 8, 2012 1:38 pm ((PDT)) Hey everyone! This my first post on the list, even though I've subscribed to it quite a while ago, and I have to say, I really like reading some of the interesting topics that have appeared here and I'm jsut sorry I didn't subscribe to the list before. :) In any case, I've decided to finally post here because after all the years I've spent on my main conlang, some of you may know it under the name Laefêvëši, I have now created (or rather, I'm creating) a new version of the language which I really like, for the first time, actually. The language so far known as Laefêvëši now goes under the name Classical Laefêvëši and the new version is called Ascended or Celestial Laefêvëši. The mean reason why I decided to redesign the language is because Classical Laefêvëši did not fit with the culture of its speakers anymore (the main problem were genders, but also a few other minor things, both grammar and vocabularywise). Celestial Laefêvëši doesn't have any genders, and it doesn't distinguish between words like mother and father, brother and sister, or daughter and son. But moving to my point now. I've made a short video, an introduction to Celestial Laefêvëši, and I would really like to hear your comments/thoughts and maybe get some feedback, regarding both the video and the language. Here's the link to the video: http://www.youtu.be/qPuujTQoiWw And the link to my wiki: http://olilowiki.jumpwiki.com (you can find here all my conlangs, although not everything is online and some things need some major revising, but I'm working on that) Thanks and any feedback/comments/thoughts/questions are really appreciated. :) Cheers, Andrej <http://www.youtu.be/qPuujTQoiWw>-- The future is predetermined by the character of those who shape it. Prihodnost vnaprej določajo karakterji tistih, ki jo oblikujejo. Messages in this topic (1) ________________________________________________________________________ ________________________________________________________________________ 2a. Re: If you only had 16 characters per day... (a language for Streetp Posted by: "Mechthild Czapp" rejista...@me.com Date: Wed Aug 8, 2012 1:45 pm ((PDT)) Yeah, a code would be easiest probably, but OTOH, I used Streetpass greetings for months to just chat with a neighbor who had a very different schedule than me and we did not just talk about game related things, as you can guess. So a limited set of expressions is IMHO a bit too limited. Am 08.08.2012 um 21:19 schrieb taliesin the storyteller <taliesin-conl...@nvg.org>: > On 2012-08-02 10:03, Mechthild Czapp wrote: >> I have yet another insane idea for a constructed language (after >> having stricken fear and confusion on the minds of the recent relay >> ring 2 with Neoquux): A language which is exclusively written and >> used to communicate on Streetpass for the Nintendo 3DS. Now what is >> Streetpass? It is a way how 3DSes can "see" each other wirelessly >> just by being near each other. Among the things which are possible >> is to exchange a 16 letter long greeting. >> >> Any suggestions for this? Is a number base of 360 feasible /../ > > I would have made an encoder and a decoder. There would be a webpage with a > "keyboard" and a "grammar" and example sentences. > > The keyboard would have symbols like "mii", the mii colors, the games that > have street pass like mario kart and some verbs, like "need", "meet", ways to > give a date (same year should be enough) etc. > > "Play MK library 5th of november" > "Play MK online 8 oclock" > "Need green mii" > > The symbols would be encoded for compactness and keyed in as the greeting. > Then the receiver would go to the same page and decode the message (which > would look like gobbledygook when encoded). Should probably start with a > version code so that it can be updated (new version) later on. > > It would spread by spreading the url in a street pass greeting, so the url > would need to be short. > > And of course my brain is already coming up with ways to use this to spam. > Meh, stop thinking about work, brain. > > > t., hopelessly practical and pragmatic, I know... Messages in this topic (13) ________________________________________________________________________ 2b. Re: If you only had 16 characters per day... (a language for Streetp Posted by: "taliesin the storyteller" taliesin-conl...@nvg.org Date: Wed Aug 8, 2012 1:46 pm ((PDT)) On 2012-08-02 10:03, Mechthild Czapp wrote: > I have yet another insane idea for a constructed language (after > having stricken fear and confusion on the minds of the recent relay > ring 2 with Neoquux): A language which is exclusively written and > used to communicate on Streetpass for the Nintendo 3DS. Now what is > Streetpass? It is a way how 3DSes can "see" each other wirelessly > just by being near each other. Among the things which are possible > is to exchange a 16 letter long greeting. > > Any suggestions for this? Is a number base of 360 feasible /../ I would have made an encoder and a decoder. There would be a webpage with a "keyboard" and a "grammar" and example sentences. The keyboard would have symbols like "mii", the mii colors, the games that have street pass like mario kart and some verbs, like "need", "meet", ways to give a date (same year should be enough) etc. "Play MK library 5th of november" "Play MK online 8 oclock" "Need green mii" The symbols would be encoded for compactness and keyed in as the greeting. Then the receiver would go to the same page and decode the message (which would look like gobbledygook when encoded). Should probably start with a version code so that it can be updated (new version) later on. It would spread by spreading the url in a street pass greeting, so the url would need to be short. And of course my brain is already coming up with ways to use this to spam. Meh, stop thinking about work, brain. t., hopelessly practical and pragmatic, I know... Messages in this topic (13) ________________________________________________________________________ 2c. Re: If you only had 16 characters per day... (a language for Streetp Posted by: "taliesin the storyteller" taliesin-conl...@nvg.org Date: Wed Aug 8, 2012 11:43 pm ((PDT)) On 2012-08-08 22:44, Mechthild Czapp wrote: > Yeah, a code would be easiest probably, but OTOH, I used Streetpass > greetings for months to just chat with a neighbor who had a very > different schedule than me and we did not just talk about game > related things, as you can guess. So a limited set of expressions is > IMHO a bit too limited. Less chance of Nintendo trying to block it if it is limited. "Protect the children" and all that. HM Messages in this topic (13) ________________________________________________________________________ ________________________________________________________________________ 3. More Logic & Semantics Posted by: "Logan Kearsley" chronosur...@gmail.com Date: Wed Aug 8, 2012 4:42 pm ((PDT)) I've had a PDF about type-directed natural language syntax parsing sitting on my desktop for quite a while, and finally got around to reading it. I thought it would have something to do with coding selectional features of predicates as types to do POS tagging and binding disambiguation; turns out somebody just accidentally reinvented Categorial Grammar to produce a syntax parser (incidentally, they did it entirely in Haskell, which I found neat because I've been contemplating the grammar of Haskell as I try to figure out how to handle parentheses-less binding disambiguation in Palno; also incidentally, that means that what I thought it was going to be about is still a potential open area for new research). This reminded me of my prior complaints about how introducing types into Palno in order to allow unambiguous usage of higher-order predicates would be horribly unwieldy, as I realized that the complex types of lexemes in Categorial Grammar were exactly isomorphic to the complex recursive function types that would be required for resolving binding ambiguities with typed predicates in Palno. My review of Categorial Grammar has made me more pessimistic about the prospects of successfully implementing higher-order operations in a human-usable language; there is no provision for parametric types in Categorial Grammar, and clearly human brains can only handle explicit selectional types of finite complexity, which means there must be a fixed limit to the possible complexity of a type and thus a limit on the number of types available in a language (which may correspond to the creation of different lexical classes). And that means no nth-order operators. That's slightly disappointing, if true; in trying to create a minimalist language with applicative semantics based on predicate calculus, it's one thing to disallow something or to allow arbitrary quantities of it, but quite another to have to specify "up to exactly x fixed number". Of course, that being true depends on it actually being a cognitive linguistic universal that humans can't handle parametric types, which I think will be rather difficult to prove; just because they happen not to be accounted for in Categorial Grammar doesn't necessarily mean that they're impossible, and I have no idea how to even start analyzing natural languages with that idea in mind. In the meantime, I've been giving more thought to Palno's semantics without higher order operators; much of my new thought can be integrated into the interpretation of Palno's syntactic & morphological structures without having to actually change those structures, which is nice. The only major morphological change that's needed is the introduction of a bit of morphology to say "this is a logical predicate", i.e., asserts a truth value. Without that, we can interpret all predicates as returning some sort of nominal- usually, the Davidsonian event argument. Since Palno predicates are in fact, most of the time, no longer interpreted as actual logical predicates after this change, I am motivated to change the name of that particular lexical class; I am not sure if it would be more useful to call them "operators" or something like that as distinct from atoms, or to just group all Palno open-class words into a single class of roots, some of which happen to be 0-arity and return substantives and qualities and such. Now I also recently re-read "Predicates and Pronominal Arguments in Straits Salish" (Jelinek & Demers, 1994), since I've been toying with doing a Salish-inspired language anyway, and I noticed a rather odd claim: "Pronominal arguments are a necessary but not a sufficient condition for the lack of a noun/verb contrast; there are pronominal argument languages that have nouns [...]. But for a language to lack a noun/verb contrast, it must have only pronominal affixes and clitics in A-positions (i.e. argument positions). Otherwise, if each root heads its own clause, there would be an infinite regress in argument structure." with footnote: "We thank an anonymous reviewer for raising the question of whether there might be a language just like Straits Salish, except for having DetPs in A-positions. In such a language, the predicates on which the argumental DetP would be based would in turn have their own DetP argument structure, and so on ad infinitum." It seems to me they missed a rather obvious intermediate case; Salish apparently only allow 1-depth trees, where a language that *prohibited* pronominal arguments with no N/V distinction would clearly have infinite trees, which is not practical for communication; but one could still have a language with no nouns that allows *either* subordinate clauses *or* pronominals as arguments, with pronominals acting as terminals. I suspect that is what I shall end up doing if I get around to my Salish-inspired language, but in the meantime, an applicative language like Palno does nearly the same thing, with 0-arity roots serving as terminals; of course, one could argue that their ability to serve as syntactic terminals is a good argument for keeping a separate lexical class of atoms/nouns as distinct from operators/verbs, but that could be circumvented by introducing pronouns/variables as an explicit semantic class in Palno (which would be a much bigger change than what I've contemplated so far), with operators having arity of at least 1. This gets me onto another topic which has long bugged me about Palno's structure, which inspiration from Salish might actually help to fix. Relative clauses in Palno have so far been handled either by an internally-headed structure with morphology to indicate that an internal argument is returned/projected rather than the usual value for that predicate, or with a hacky adjunction-with-coreference thing. The first option is now made to fit much better into the rest of the language by my change in the interpretation of predicate/operator applications (the predicate morphology could be seen as just one case of a set of inflections that specify which argument of an operator to use as the syntactic projection / return value), but is still a little annoying because the syntactic structure does not exactly match my conception of the final interpreted semantic structure, and the second is just gross because there is no other occurrence of adjunction anywhere in Palno's grammar. What would be very nice is to have some way of encoding the logical proposition "x such that P1(x) & P2(x) & P3(x) ...", etc., without having to introduce the capacity to bind arbitrary variables in what is supposed to be a speakable language. Salish languages *have* to do something like that in order to achieve the implication of a nominal argument in a syntax that only allows pronominal arguments by ensuring that the same referent is bound to multiple argument positions in the main and subordinate-adjoined clauses. They do this by having strict rules about what pronouns in different clauses must or must not be coreferential with each other, which sometimes disallow certain logically possible sentence types (for which they have voice transformations to make up). Something similar could be done with Palno, avoiding disallowed sentence types by providing a sufficiently rich set of pronouns with agreement features to handle the proper coindexing. A closed class of pronouns would not be powerful enough to produce all of the logical semantic structures that could be expressed in predicate calculus with arbitrary variable binding, but should be able to nicely handle all of the practical communicative needs of a human language. Without the matrix/subordinate distinction available in Salish, a pronoun with no agreement morphology could be seen as a variable binding which is then referenced in a further series of conjoined clauses containing agreeing pronouns until shadowed by another non-agreeing pronoun indicating a new binding. Exactly what kind of agreement morphology to use is something I shall have to think about. That would somewhat complicate the grammar and make it no longer 100% purely applicative, but the better structural match to semantics might just be worth it. With all of those changes implemented, Palno could still be used with the purely applicative postfix grammar (still the simplest grammar for any language that I know of), or slightly more complicated with conjunctions and reduction of repeated arguments as it was before, or at yet a third level with explicit coreferencing, which I expect to be the most easily human-usable and still capable of fitting on a postcard (for real, not like Esperanto is supposed to). No, back to thinking about how to manage higher-order operators.... -l. Messages in this topic (1) ________________________________________________________________________ ________________________________________________________________________ 4a. Re: In Defense of Monster Raving Loony Alignment Posted by: "Logan Kearsley" chronosur...@gmail.com Date: Wed Aug 8, 2012 4:49 pm ((PDT)) On 1 August 2012 15:13, Logan Kearsley <chronosur...@gmail.com> wrote: [...] > On 1 August 2012 06:17, Alex Fink <000...@gmail.com> wrote: >> Oh, and I meant to mention that when suggesting this phrasal verb origin I >> was also somewhat thinking of languages with verbal templates like the >> Athabaskan, where there are many "themes" formed on any given stem, and a >> theme can include any of various sorts of calcified stuff between the object >> and the subject marker. Some of this stuff is calcified directionals, I >> think, some calcified shape-classifiers for the object, some calcified >> adverbials of other sorts... do that with my "phrasal verb" proposal and >> it's even more flexible. > > Aha, object classifiers could help with the variability. Especially if > semantic shifts result in what were originally real classifiers with > strict semantic selectional requirements on the object end up causing > the selectional restrictions to erode away. Might have to do some > research on Athabaskan etymology to come up with more possible > derivational routes for multipartite verbs. I think I have found yet another route, potentially natlang-attested, by which one could accomplish Monster Raving Loony Alignment. Salish languages typically disallow two 3rd-person nominal arguments adjoined to the same clause. Those which do are assumed to do so only due to contact with English or other non-Salish languages, and there is ambiguity in that case over which adjoined nominal is associated with which argument position (which is probably why the situation is usually disallowed). So, a heavy contact situation that results in those kinds of clauses becoming exceedingly common in a pidginization/creolization scenario might provide another nice starting place for going off in all kinds of weird directions with different ambiguity resolution strategies (of course, the most obvious and most likely is that you just start using word order, but especially with all of the nominalization and subordination morphology that Salish languages provide, I'm sure a more interesting resolution can be found). -l. Messages in this topic (10) ________________________________________________________________________ ________________________________________________________________________ 5a. Re: Allophony in Siye Posted by: "Anthony Miles" mamercu...@gmail.com Date: Wed Aug 8, 2012 8:43 pm ((PDT)) My initial model for Siye allophony was Japanese, but I didn't want [p] to become [h], so I did think of French. Perhaps one of the dialects other than Standard or Far Western uses [h]? When I was beginning to study linguistics on my own, the change from "cambiare" to "changer" baffled me! Just out of curiousity, what does happen when the percentage of plosives dips too low? Is there a natlang that illustrates such a change? Messages in this topic (8) ________________________________________________________________________ 5b. Re: Allophony in Siye Posted by: "Roger Mills" romi...@yahoo.com Date: Thu Aug 9, 2012 8:53 am ((PDT)) --- On Wed, 8/8/12, Anthony Miles <mamercu...@gmail.com> wrote: My initial model for Siye allophony was Japanese, but I didn't want [p] to become [h], so I did think of French. Perhaps one of the dialects other than Standard or Far Western uses [h]? When I was beginning to study linguistics on my own, the change from "cambiare" to "changer" baffled me! ------------------------------------------------------------------------------------------------- I've wondered about that too... Is it possible that "changer" descends from something other than *cambiare? Is "cambiare" actually the Vulgar Latin form, or was it something else in Classical? And I seem to recall reading somewhere that "cambiare"is ult. a Greek loan (?) Christophe will probably know about the French history of "changer"; Ray Brown will undoubtedly be able to answer my questions regarding "cambiare"..... It just seems difficult to see how the presumed cluster **...-[nasal]+bj-...ended up as nasal+[Z]. Could **-bj- somehow have gone through a **[g] stage??? IMO both are high unnatural, no? Are there any other French exs. of the same/similar change?? Messages in this topic (8) ________________________________________________________________________ 5c. Re: Allophony in Siye Posted by: "Alex Fink" 000...@gmail.com Date: Thu Aug 9, 2012 9:26 am ((PDT)) On Wed, 8 Aug 2012 23:43:07 -0400, Anthony Miles <mamercu...@gmail.com> wrote: >Just out of curiousity, what does happen when the percentage of plosives dips >too low? Is there a natlang that illustrates such a change? Fortition, I imagine. But I can't really think of a good example... I wonder if something like the Germanic relapse of *bh dh gh > [B D G] into [b d g] counts. It's complete to different degrees in the various langs: e.g. many of the daughters (but not e.g. OHG) kept them fricated intervocalically; Dutch didn't harden [G]; except after nasals, no-one hardened [G_w] which > [w]. (And Greek did nothing like this. There were more PIE stops of the *t than the *d series, weren't there?) On Thu, 9 Aug 2012 08:52:59 -0700, Roger Mills <romi...@yahoo.com> wrote: >It just seems difficult to see how the presumed cluster >**...-[nasal]+bj-...ended up as nasal+[Z]. Could **-bj- somehow have gone >through a **[g] stage??? IMO both are high unnatural, no? Are there any other >French exs. of the same/similar change?? Yes, many. Fr. _rage_ < Lat. _rabies_, OF _cage_ < Lat. _cavea_, subjunctive forms _sache_ etc. of _savoir_ < forms _sapiat_ etc. of Lat. _sapiō_, ... I'll also call attention to Siswati again, which had a palatalisation change affecting labials but *leaving velars untouched*, and only partially affecting alveolars: http://books.google.com/books?id=IV2qnBMI3FgC&lpg=PA138&ots=3h3grghEx8&pg=PA138#v=onepage&q&f=false Alex Messages in this topic (8) ________________________________________________________________________ ________________________________________________________________________ 6a. Active/Stative Distinctions and Pronominal Prefixes Posted by: "Anthony Miles" mamercu...@gmail.com Date: Wed Aug 8, 2012 9:28 pm ((PDT)) In addition to Siye, I have a proto-conlang that is still a sytactical skeleton. This proto-conlang (which will call C080812 hereafter) is, among other things, predominately nominative-accusative and uses preposed particles derived from prepositions to mark case. C080812 has the following basic syntax for common nouns (i.e., anything that does not involve the first or second person in the most agent-like role (T= theme, R = recipient, D = donor): VS/VPA/VTRD "3-eat ACC mouse NOM cat" "The cat ate the mouse" "3-give ACC gift DAT girl NOM man" "The man gave the gift to the girl" The case of S in VS varies on an active/stative basis. "3-Walk NOM man" "The man walked" "3-Sleep DAT man" "The man slept" The problem, however, which I now bring before the court of CONLANG-L, is that the first and second pronouns are prefixed to the verb. Thus, whereas the sentences "3-Walk NOM man" and "3-Sleep DAT man" are active and stative (the presumption is that stative verbs still benefit the subject, which is why it is dative rather than accusative), technically "1-walk" "I walked" and "1-sleep" "I slept" do not contrast grammatically. This seems wrong, but I'm not sure. One solution, of course, is to not distinguish active and stative in the 1st and 2nd persons. Another is to forsake the prefixed pronouns altogether, and make all verb forms 3rd person. The third option is to keep "1-walk" and "1-sleep" but expand them to "1-walk NOM 1" and "1-sleep DAT 1". Messages in this topic (4) ________________________________________________________________________ 6b. Re: Active/Stative Distinctions and Pronominal Prefixes Posted by: "Logan Kearsley" chronosur...@gmail.com Date: Wed Aug 8, 2012 9:39 pm ((PDT)) On 8 August 2012 22:28, Anthony Miles <mamercu...@gmail.com> wrote: > In addition to Siye, I have a proto-conlang that is still a sytactical > skeleton. This proto-conlang (which will call C080812 hereafter) is, among > other things, predominately nominative-accusative and uses preposed particles > derived from prepositions to mark case. C080812 has the following basic > syntax for common nouns (i.e., anything that does not involve the first or > second person in the most agent-like role (T= theme, R = recipient, D = > donor): > VS/VPA/VTRD > "3-eat ACC mouse NOM cat" "The cat ate the mouse" > "3-give ACC gift DAT girl NOM man" "The man gave the gift to the girl" > The case of S in VS varies on an active/stative basis. > "3-Walk NOM man" "The man walked" > "3-Sleep DAT man" "The man slept" > The problem, however, which I now bring before the court of CONLANG-L, is > that the first and second pronouns are prefixed to the verb. Thus, whereas > the sentences "3-Walk NOM man" and "3-Sleep DAT man" are active and stative > (the presumption is that stative verbs still benefit the subject, which is > why it is dative rather than accusative), technically "1-walk" "I walked" and > "1-sleep" "I slept" do not contrast grammatically. This seems wrong, but I'm > not sure. One solution, of course, is to not distinguish active and stative > in the 1st and 2nd persons. Another is to forsake the prefixed pronouns > altogether, and make all verb forms 3rd person. The third option is to keep > "1-walk" and "1-sleep" but expand them to "1-walk NOM 1" and "1-sleep DAT 1". A fourth is to have different case forms for the 1st & 2nd prefixes. A fifth might be to disallow 1st & 2nd pronouns as subjects of statives and require some kind of voice-changing operation to license them. -l. Messages in this topic (4) ________________________________________________________________________ 6c. Re: Active/Stative Distinctions and Pronominal Prefixes Posted by: "Jörg Rhiemeier" joerg_rhieme...@web.de Date: Thu Aug 9, 2012 7:52 am ((PDT)) Hallo conlangers! On Thursday 09 August 2012 06:28:42 Anthony Miles wrote: > In addition to Siye, I have a proto-conlang that is still a sytactical > skeleton. This proto-conlang (which will call C080812 hereafter) is, among > other things, predominately nominative-accusative and uses preposed > particles derived from prepositions to mark case. C080812 has the > following basic syntax for common nouns (i.e., anything that does not > involve the first or second person in the most agent-like role (T= theme, > R = recipient, D = donor): VS/VPA/VTRD > "3-eat ACC mouse NOM cat" "The cat ate the mouse" > "3-give ACC gift DAT girl NOM man" "The man gave the gift to the girl" > The case of S in VS varies on an active/stative basis. > "3-Walk NOM man" "The man walked" > "3-Sleep DAT man" "The man slept" > The problem, however, which I now bring before the court of CONLANG-L, is > that the first and second pronouns are prefixed to the verb. Thus, whereas > the sentences "3-Walk NOM man" and "3-Sleep DAT man" are active and > stative (the presumption is that stative verbs still benefit the subject, > which is why it is dative rather than accusative), technically "1-walk" "I > walked" and "1-sleep" "I slept" do not contrast grammatically. This seems > wrong, but I'm not sure. One solution, of course, is to not distinguish > active and stative in the 1st and 2nd persons. Another is to forsake the > prefixed pronouns altogether, and make all verb forms 3rd person. The > third option is to keep "1-walk" and "1-sleep" but expand them to "1-walk > NOM 1" and "1-sleep DAT 1". My own active-stative conlang, Old Albic, uses pronominal suffixes, and it has two distinct sets for agents and patients, such that "I walked" and "I slept" use different suffixes. This is, as far as I know, the usual state of affairs in such languages. -- ... brought to you by the Weeping Elf http://www.joerg-rhiemeier.de/Conlang/index.html "Bêsel asa Éam, a Éam atha cvanthal a cvanth atha Éamal." - SiM 1:1 Messages in this topic (4) ________________________________________________________________________ 6d. Re: Active/Stative Distinctions and Pronominal Prefixes Posted by: "Alex Fink" 000...@gmail.com Date: Thu Aug 9, 2012 8:22 am ((PDT)) On Thu, 9 Aug 2012 00:28:42 -0400, Anthony Miles <mamercu...@gmail.com> wrote: >In addition to Siye, I have a proto-conlang that is still a sytactical >skeleton. This proto-conlang (which will call C080812 hereafter) is, among >other things, predominately nominative-accusative and uses preposed particles >derived from prepositions to mark case. C080812 has the following basic syntax >for common nouns (i.e., anything that does not involve the first or second >person in the most agent-like role (T= theme, R = recipient, D = donor): >VS/VPA/VTRD >"3-eat ACC mouse NOM cat" "The cat ate the mouse" >"3-give ACC gift DAT girl NOM man" "The man gave the gift to the girl" >The case of S in VS varies on an active/stative basis. >"3-Walk NOM man" "The man walked" >"3-Sleep DAT man" "The man slept" >The problem, however, which I now bring before the court of CONLANG-L, is that >the first and second pronouns are prefixed to the verb. Thus, whereas the >sentences "3-Walk NOM man" and "3-Sleep DAT man" are active and stative (the >presumption is that stative verbs still benefit the subject, which is why it >is dative rather than accusative), technically "1-walk" "I walked" and >"1-sleep" "I slept" do not contrast grammatically. This seems wrong, but I'm >not sure. As Joerg says, it's an uncommon way about things. But perhaps your verbal agreement markers predate the onset of the active/stative system. >One solution, of course, is to not distinguish active and stative in the 1st >and 2nd persons. That works. >Another is to forsake the prefixed pronouns altogether, and make all verb >forms 3rd person. That is, forsake person marking on the verb -- they're hardly 3rd person if they don't contrast with anything. Eh, you could, but it feels cutting off your nose to spite your face-ish. >The third option is to keep "1-walk" and "1-sleep" but expand them to "1-walk >NOM 1" and "1-sleep DAT 1". If you want to do something like this, I'd bear in mind that languages like getting away with omitting things. Rather than contrast "1-V NOM 1" and "1-V DAT 1", I'd build in an assumption that first and second person subjects default to being active (they're high up the animacy hierarchy so active is the likely default), and contrast active "1-V" and stative "1-V DAT 1". Done that way, this might be my favourite of your suggestions. And Logan's idea of not allowing SAPs of subjects as statives is fun, if you're willing to have things like "I sleep" be a tad morphosyntactically ungainly to express. (And it could evolve interestingly.) Alex Messages in this topic (4) ________________________________________________________________________ ________________________________________________________________________ 7a. Re: Conjunction Curiosity Posted by: "Anthony Miles" mamercu...@gmail.com Date: Wed Aug 8, 2012 9:54 pm ((PDT)) The part of my brain that I have allotted to things Siye is screaming "That ain't right!". All of the vocabulary items exist in Siye, but the already complex phrase would require this further consideration: is the relationship of the priest and the house alienable or inalienable. If alienable, the phrase "the priest of the house of our lady" would be "kutum sili laye sili le-me-me-ne" ("laye sili-me" means "lady"; if not "le-me-me-me". I think a Siye speaker would have to rephrase it as "laye sili le-me-me kutum sili-me". In this case, if the lady were ruler, it would become "laye sili le-me-me-me kumayam lusili me-pu kutum sili-me", where the "me" of "lusili me" is "this", not the possessive marker. If the priest were the ruler, then it would be "laye sili-me-me-me kutum sili-me kumayam lusili me-pu". Of course a better solution would be to make a relative clause "kutum sili laye sili le-me-me lusili me ekupumame-ne" "The priest of the house of our lady who rules this city" or "laye sili le-me-me-me kutum sili ekupumame-ne" "The priest of the house of our lady, who (i.e. the priest) rules this city. Suffixaufnahme is a real pain to work with! Messages in this topic (17) ________________________________________________________________________ 7b. Re: Conjunction Curiosity Posted by: "Christophe Grandsire-Koevoets" tsela...@gmail.com Date: Thu Aug 9, 2012 4:24 am ((PDT)) On 1 August 2012 23:23, David Brumbley <davidbrumb...@gmail.com> wrote: > I ran across this proverb recently and thought it would make for a > quick and fun translation exercise. I came across a few constructions > I hadn't yet been forced to address in Hsassiens, and I'm curious as > to whether other con- or nat-langs behave similarly in one aspect. > > I've been using two versions of the coordinating conjunction in > Hsassiens without really thinking too much about it. In the case of > one subject linked to two different verbs, the conjunction is 'sin.' > When two or more nouns, adjectives or adverbs are being listed, the > conjunction is 'zem.' So, to HsassiEnglish an example, "He is walking > SIN talking at the same time," but "Tommy ZEM Rebecca are fighting > again." Just curious if other languages distinguish in the same way, > and if so, how. > > > A very interesting thread, and an interesting proverb. Looking at how to translate it in Moten prompted me to look even deeper into how coordination is handled in that language. As I've described in this blog post: http://christophoronomicon.blogspot.nl/2012/02/moten-part-vii-particles.html, Moten has a bunch of coordinating clitics, which function somewhat but not exactly like coordinating conjunctions. The relevant ones here are _opa_ (described in the post) and _de_ (a particle I discovered a few weeks ago that helped me solve a years-old headache). When used to coordinate noun phrases, they can both be translated by "and", but with a different connotation: - _opa_ has a connotation of "also", and indicates that the two noun phrases refer to separate entities. E.g.: _mjan opa badej_: the cat and the dog (notice how the definite infix -e- only appears on the last noun yet both are definite. This is a strict syntactic rule in Moten: when noun phrases are coordinated, only the last one takes marks of case, number and definition, and those extend to all coordinated phrases in meaning); - _de_ has a connotation of "that is" and indicates that the two noun phrases refer to a single entity. E.g.: _olnesif de vajagzif_: expert and student (refers here to a single person who is considered both an expert and a student, for some reason :) ). _de_ is also used wherever English uses appositions to refer to one entity with more than one noun, including with titles. E.g.: _plisif de Beatliksi_: Queen Beatrix (could also be _Beatliksi de plisejf_, since _de_ is commutative :) . The definite infix -e- reappears in this word order because the last noun is a common noun rather than a proper noun). This was the afore-mentioned headache (in Moten apposition has a different function, so I couldn't use it for those cases). They are similar to A. de Mek's _wa_ and _wu_, although I developed them independently (my _de_ is actually influenced by the Wardwesân particle _ab_, although their uses are not exactly the same). They can also both be used to coordinate verbs, but only in the sense of your _sin_, and very strictly so: verbs coordinated using a coordinating clitic not only share exactly the same arguments (*all* of them, i.e. not only the subject, so you can't use a coordinating clitic for a sentence like "he left the bar and went home", but you can for a sentence like "he took a piece of bread and buttered it" -- and you don't need the resumptive "it" when doing so! --), but also the same tense, aspect, mood and voice! (so you cannot use a coordinating clitic for a sentence like "he did it before and will do it again") The reason is similar to the reason why coordinated noun phrases share the same case, number and definition: only the last coordinated element takes the morphological markings, and their meaning extend to all coordinated elements. When coordinating verbs, _opa_ and _de_ keep their connotations: - _opa_ indicates that the coordinated verbs correspond to different actions. E.g.: _bdan pe|laz opa eze|s ige_: I can see and hear you. - _de_ indicates that the coordinated verbs correspond to a single action, i.e. the second one is meant as a rephrasing or clarification of the first one. E.g.: _gobvuda|n vajaguz de oknestuluz ito_: I know about you, that's to say I've read about you ("to know" here is translated as "to have learned", so the second verb can also be in the perfect). While they can be used to coordinate noun phrases and verbs, the coordinating clitics *cannot* be used to coordinate clauses. In fact there is no such thing as clause-level coordination in Moten. However, this doesn't mean that they can't be used at clause-level at all. In fact they can (like all clitics, when they are put in front of the auxiliary verb, their meaning encompasses the whole clause). It's just that when used that way, they lose their coordinating function, and become more like clause-level adverbs (they can do that at the phrase level too, by the way). In that case, _opa_ becomes equivalent to "also" or "moreover" (indicating the clause is additional, separate information), while _de_ is more like "in other words" or "that's to say" (indicating that the clause is a rephrasing or clarification of the previous one). Now onto the proverb itself! > > He who knows not and knows not he knows not > He is a fool. Shun him. > He who knows not and knows he knows not > He is a student. Teach him. > He who knows and knows not he knows > He is asleep. Wake him. > He who knows and knows he knows > He is wise. Follow him. > OK, my problem is not so much vocabulary (although I do miss a couple of vocabulary items to translate it), but the constructions themselves. You see, we have here coordinated relative subclauses, and as I've explained about Moten doesn't have clause-level coordination. And you can't use verbal phrase-level coordination here as the different parts have different objects. I could possibly have a nominal completed by two separate relative subclauses corresponding to the two elements of the coordination, but I'm wondering whether it's possible (Moten is strictly head-last, with relative subclauses always in front of their heads, and I'm not sure separating a relative subclause from its head with *another* relative subclause is pragmatically possible). Before deciding on how something works syntactically in Moten, I usually check first how Basque and Japanese do it (in terms of syntax, Moten is quite close to both those languages), but I've been unable to find a translation of this proverb in either language. Has anyone got one? It'd help me greatly to figure out how Moten will handle it! All in all, great food for thought! -- Christophe Grandsire-Koevoets. http://christophoronomicon.blogspot.com/ http://www.christophoronomicon.nl/ Messages in this topic (17) ------------------------------------------------------------------------ Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/conlang/ <*> Your email settings: Digest Email | Traditional <*> To change settings online go to: http://groups.yahoo.com/group/conlang/join (Yahoo! ID required) <*> To change settings via email: conlang-nor...@yahoogroups.com conlang-fullfeatu...@yahoogroups.com <*> To unsubscribe from this group, send an email to: conlang-unsubscr...@yahoogroups.com <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/ ------------------------------------------------------------------------