[07:52:14] *** Quits: proteus-guy (~proteus-g@cm-134-196-84-142.revip18.asianet.co.th) (Remote host closed the connection) [08:21:10] *** Joins: proteus-guy (~proteus-g@184.82.233.186) [11:39:09] *** Quits: proteus-guy (~proteus-g@184.82.233.186) (Remote host closed the connection) [12:05:15] this sentence really gets to the essential part of what has changed in the grammar ^_^ [12:05:15] TestLangEng> without some teacher whom they help always paint us married to ourselves there is something today [12:05:15] TestLangEng-OLD> without some teacher whom they help always paint us married to themselves there is something today [12:18:56] :D [12:20:02] it's tricky to make generalisations about agreement [12:20:15] so we've got a few reflexive functions [12:20:22] e.g. ReflVP : VPSlash -> VP [12:20:28] and ReflA2 : A2 -> AP [12:20:36] (which can be modified into VP with UseComp) [12:20:53] so then we've got the VP object as a table Agr => Str [12:23:20] but then you can have some more VP -> VPSlash transformations, e.g. with SlashV2V [12:23:40] like "help X " [12:23:53] then it makes sense that the reflexive agrees with X and not with the subject [12:24:03] "I help him like himself", not "I help him like myself" [12:24:43] but this breaks when you can have all kinds of VP -> VPSlash transformations; VPSlashPrep is the stupidest [12:25:02] it just attaches a Prep into a VP and expects another core argument, not adv [12:25:43] so then you'd get "like with X" [12:25:49] and that doesn't make all that much sense to agree with X [12:25:59] "I like himself with him" [12:27:19] I've tried a couple of different heuristics, like "if it has an infinitive, then agree with object" or "if it doesn't have a preposition, don't agree with object" but you can have legit core objects that take a preposition (or case) [12:27:37] so I guess we just need a boolean that says "has VPSlashPrep messed things up for this VP" [12:27:43] and agree with object if not [12:57:34] *** Joins: proteus-guy (~proteus-g@cm-134-196-84-196.revip18.asianet.co.th) [14:57:05] inariksit: two people working on catalan now [15:20:32] nice! [15:20:46] tell them that pronoms febles need some work [15:20:58] if you want to scare them :-D [15:21:16] (also tell them they don't need any of that bullshit in the mini resource :-P) [15:21:42] (also no need to tell them anything, you're an autonomous individual!) [18:06:24] daherb: these almost make sense [18:06:25] TestLangGer> ich helfe viele Würmer sich zu werfen [18:06:25] TestLangGer-OLD> ich helfe viele Würmer mich zu werfen [18:06:25] TestLangEng> I help many worms throw themselves [18:06:46] and how about these? [18:06:46] TestLangGer> es ist ich der etwas nicht schreiben will [18:06:47] TestLangGer-OLD> es ist mich den etwas nicht schreiben will [18:06:47] TestLangEng> it is me that something doesn't want to write [18:08:25] and this? [18:08:25] TestLangGer> es ist du die etwas nicht schreiben will [18:08:25] TestLangGer-OLD> es ist dich die etwas nicht schreiben will [18:08:26] TestLangEng> it is you that something doesn't want to write [19:04:17] *** Quits: daherb (~daherb@cse-principia.cse.chalmers.se) (Ping timeout: 276 seconds) [19:11:34] *** Joins: daherb (~daherb@cse-principia.cse.chalmers.se)