Return-Path: Received: (qmail 13329 invoked from network); 1 Sep 2000 21:43:50 -0000 Received: from unknown (10.1.10.142) by m2.onelist.org with QMQP; 1 Sep 2000 21:43:50 -0000 Received: from unknown (HELO pi.meson.org) (209.191.39.185) by mta3 with SMTP; 1 Sep 2000 21:43:49 -0000 Received: (qmail 11684 invoked by uid 1000); 1 Sep 2000 21:39:25 -0000 Date: 1 Sep 2000 21:39:25 -0000 Message-ID: <20000901213925.11683.qmail@pi.meson.org> To: lojban@egroups.com In-reply-to: (message from Invent Yourself on Fri, 1 Sep 2000 17:08:42 -0400 (EDT)) Subject: Re: [lojban] How many? References: From: "Mark E. Shoulson" X-Yahoo-Message-Num: 4195 Content-Length: 3205 Lines: 81 >From: Invent Yourself >Date: Fri, 1 Sep 2000 17:08:42 -0400 (EDT) > > >On 1 Sep 2000, Mark E. Shoulson wrote: > >> >From: Pierre Abbat >> >Date: Fri, 1 Sep 2000 14:02:09 -0400 >> > >> > >> >If "How old are you" is {do nanca li xo}, would "how many species of fish are >> >there" be {ta finpe li xo} or {ta finpe xo da}? >> >> No... {ta finpe li xo} means "What number is the fish-species of that >> thing?" ... or something like that; it doesn't make sense, so it's hard to >> translate. The x2 place of finpe isn't meant for a number. > >Is Lojban a strongly-typed language? Actually, yes. At least to some extent. That was part of the big hassle with sumti-raising: some gismu wanted events/abstractions, and not simple things. There are cases where you need individuals, cases where you need masses, and cases where you need sets. In fact, the place structure of {Nmoi} (for some well-defined number N) is: x1 is a *mass* formed from the *set* x2 of N members, one or more of which is/are x3. It actually converts between masses and sets, while supplying a number. And we have lu'i and lu'o to convert to and between sets and masses also. There IS typing in Lojban, and numbers are a type. If I saw a large "4" in the air, it would NOT be correct to say {mi viska li vo}; that's a use-mention problem. (I'd need {lu'e li vo} or {me'o vo} or some such). >If we put a number in finpe x2, that number is associated with >fish-species. How many different ways are there to associate numbers with >fish-species? If species are numbered according to some scheme, that's one >way. In general though, numbers count distinct things. Therefore a number >there should be assumed to be counting distinct species. However to make >that clear, I can see the sense of using mei. If you put a number in finpe x2, that number is not *associated* with fish-species. You are claiming that it IS a fish-species (or more than one). That's what goes in finpe2: the fish-species, not something associated with it. >> The trouble is that {ta} refers to some *particular* thing/s you're >> indicating in some way, not "all the ones out there." >> >> For "how many fish are there?" Mmm... How about: That should be "fish-species" >> >> lo'i se finpe cu xomei > > >ro finpe xomei There's a tanru there I'm not sure you intended to make; this is a bare sumti: {all fish-(howmany-somes)}. Every one of the groups of fish of size (howmany?). You presumably wanted {ro finpe cu xomei}. But that means "All fish (considered as individuals, I think) are a how-many-some?" It *might* mean "How many fish are there in the world?" (as my mistyped translation above says), but I think the {xomei} needs a mass, not individuals. That would be {loi finpe} or {piro loi finpe}. Which of course means I blew it above as well; it should be {lo'i se finpe cu se xomei}, or maybe {loi se finpe cu xomei}---but that latter doesn't sound right; I keep worrying about the outer quantifier on masses; that problem doesn't happen with sets, I think. Simpler still would be just: xo da se finpe (or {xo da finpe} for "how many fish are there in the universe?") ~mark