Return-Path: Received: from kantti.helsinki.fi by xiron.pc.helsinki.fi with smtp (Linux Smail3.1.28.1 #1) id m0rnc9y-00007ZC; Sun, 12 Mar 95 03:09 EET Received: from fiport.funet.fi (fiport.funet.fi [128.214.109.150]) by kantti.helsinki.fi (8.6.10+Emil1.1/8.6.5) with ESMTP id DAA21089 for ; Sun, 12 Mar 1995 03:09:30 +0200 Received: from SEARN.SUNET.SE (MAILER@SEARN) by FIPORT.FUNET.FI (PMDF V4.3-13 #2494) id <01HO13FT439C002ZAI@FIPORT.FUNET.FI>; Sun, 12 Mar 1995 01:05:31 +0200 (EET) Received: from SEARN.SUNET.SE (NJE origin LISTSERV@SEARN) by SEARN.SUNET.SE (LMail V1.2a/1.8a) with BSMTP id 1638; Sun, 12 Mar 1995 02:05:47 +0100 Date: Sat, 11 Mar 1995 20:14:26 -0500 (EST) From: jorge@PHYAST.PITT.EDU Subject: Re: selbri as sumti Sender: Lojban list To: Veijo Vilva Reply-to: jorge@PHYAST.PITT.EDU Message-id: <01HO13FT4FHU002ZAI@FIPORT.FUNET.FI> X-Envelope-to: veion@XIRON.PC.HELSINKI.FI Content-transfer-encoding: 7BIT X-To: lojban@cuvmb.cc.columbia.edu MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Length: 3133 Lines: 87 And: > > I was thinking of it as the time encompassed by all days of a given > > kind. > That's okay. I wonder how you'd do "two of the same days of the > week" (put clumsily) - i.e. 2 Mondays, or 2 Tuesdays. Maybe > > re djedi pe xohu pisuho pa jefydje Or simply: ca pa jefydje ca re djedi This is not a very general answer to your original question, though. And I still don't know what is the best definition for {jefydje}. > > But anyway, with your definition: > > mi klama le zarci ca ro jeftu ca lo mintu jefydje cmima > > {ca lo cmima be lo jefydje poi mintu lei drata}, if you don't want tanru. > > I don't see how {mintu lei drata} helps, but I do see how {cmima} helps. Well {lei drata} helps because it is a singular term. That is a way to avoid the prenex, by fixing the day that you mean with an in-minder. Without that, the cmima is selected independently for each week. > > mi klama le zarci ca ro cmima be pa mintu jefydje > > should work. The problem has been kicked to the x2 of mintu, > > and to make it explicit, we would need to use the prenex, but > > I believe context does make it clear. > > I'll go for > pa da poi jefydje zohu mi zarci klama ro cmima be da > or, > mi zarci klama ro cmima be xohu pa jefydje > or, > mi zarci klama ro luha xohu pa jefydje I believe you mean {ca ro ...} in all cases. I did think of xo'u but I didn't want to bring it up :) Yes, I think it would work but I still don't like xo'u. > Tho it might be objected that a category is not a set, I think > I would prefer to interpret {luha} as neutralizing the distinction, > and favour the last of the three versions. What is the difference between category and set? > > mi citka lo finpe poi cmima lo cizra > > mi citka lo cidja poi cmima lo cizra > > mi tcidu lo selcku poi cmima lo cizra > > Still not quite what I want. I want to say that a class that is > pisuho of the class of fish/food/books is strange. Change {lo cizra} to {lo cizra ke finpe klesi pagbu}. I don't think your problem arises in this case for the reason I gave before: there are only existential quantifiers, which commute without any problem. > > roldei: x1 is a quotidian activity (i.e. it belongs to that > > set of quotidian activities that you had defined.) > I will go for: > > ca le cabdei mi baho gasnu luha lo roldei gihe klesi be > ro lohi selzukte be mi > > - again, horribly messy. And not what you want. I'm not even sure it's grammatical. The {gi'e} falls outside of the {lo}. Besides, you didn't restrict it to the activities that you do today. You are saying that today you did all your quotidian activities, including those that don't happen today. I'd say: ca le cabdei mi ba'o gasnu ro lo mi ca roldjeke'u selzukte where {roldjeke'u} is "x1 is an event that recurs every day" (note that it is a krefu, not the same identical event but a repetition of an event.) The {selzukte} may even be superfluous. Here we are avoiding your problem simply by defining a lujvo like "quotidian" that already contains the quantifier. Jorge