From lojban+bncCOTEtqyUDhD51NjzBBoEyMhCpQ@googlegroups.com Sun Sep 18 11:20:24 2011 Received: from mail-pz0-f57.google.com ([209.85.210.57]) by chain.digitalkingdom.org with esmtp (Exim 4.72) (envelope-from ) id 1R5LyS-0005XZ-Lh; Sun, 18 Sep 2011 11:20:23 -0700 Received: by pzd13 with SMTP id 13sf2642357pzd.12 for ; Sun, 18 Sep 2011 11:20:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlegroups.com; s=beta; h=x-beenthere:received-spf:date:from:to:subject:message-id:references :mime-version:content-type:content-disposition:in-reply-to:x-pgp-key :x-pgp-keyid:x-cunselcu'a-valsi:user-agent:x-original-sender :x-original-authentication-results:reply-to:precedence:mailing-list :list-id:x-google-group-id:list-post:list-help:list-archive:sender :list-subscribe:list-unsubscribe; bh=q+0JomBNTHvnYZKNCoK+wmiI50F32NOqWHSstiggj5k=; b=gTbGNudMmsdwxC7KbfwxDV057KIG1XzFaCghKcwZjlsMTNpnVW9LXx0MJIlmawpIej nY2jaBqBGI/J5gdQubAYDWeI+6kUronFjq6ZfOt/kkNOOWqNTsx4jEdjsaNORyYKkczB 7ihJ1oPsLGB+wOO32fo4XKuIxRkEN8Ew7+Kek= Received: by 10.68.59.195 with SMTP id b3mr175415pbr.57.1316366969979; Sun, 18 Sep 2011 10:29:29 -0700 (PDT) X-BeenThere: lojban@googlegroups.com Received: by 10.68.33.194 with SMTP id t2ls17907230pbi.0.gmail; Sun, 18 Sep 2011 10:29:28 -0700 (PDT) Received: by 10.68.0.33 with SMTP id 1mr1273567pbb.4.1316366968811; Sun, 18 Sep 2011 10:29:28 -0700 (PDT) Received: by 10.68.0.33 with SMTP id 1mr1273566pbb.4.1316366968783; Sun, 18 Sep 2011 10:29:28 -0700 (PDT) Received: from sdf.lonestar.org (mx.sdf.org. [192.94.73.19]) by gmr-mx.google.com with ESMTPS id p6si15003032pbc.0.2011.09.18.10.29.28 (version=TLSv1/SSLv3 cipher=OTHER); Sun, 18 Sep 2011 10:29:28 -0700 (PDT) Received-SPF: pass (google.com: best guess record for domain of mbays@sdf.org designates 192.94.73.19 as permitted sender) client-ip=192.94.73.19; Received: from gonzales.homelinux.org (root@sverige.freeshell.org [192.94.73.4]) by sdf.lonestar.org (8.14.4/8.14.3) with ESMTP id p8IHTRok006589 for ; Sun, 18 Sep 2011 17:29:27 GMT Received: from martin by gonzales.homelinux.org with local (Exim 4.75) (envelope-from ) id 1R5LBH-0004Hp-46 for lojban@googlegroups.com; Sun, 18 Sep 2011 13:29:27 -0400 Date: Sun, 18 Sep 2011 13:29:27 -0400 From: Martin Bays To: lojban@googlegroups.com Subject: Re: [lojban] {zo'e} as close-scope existentially quantified plural variable Message-ID: <20110918172927.GA4310@gonzales> References: <1315774215.25455.YahooMailRC@web81304.mail.mud.yahoo.com> <20110912225255.GG28088@gonzales> <1315882379.97949.YahooMailRC@web81308.mail.mud.yahoo.com> <20110914041335.GO28088@gonzales> <55EFAEEE-10A9-4002-951E-7BD949DC29F7@yahoo.com> <20110914232007.GC6492@gonzales> <1316055853.22283.YahooMailRC@web81304.mail.mud.yahoo.com> <20110916000632.GD7274@gonzales> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="jI8keyz6grp/JLjh" Content-Disposition: inline In-Reply-To: X-PGP-Key: http://mbays.freeshell.org/pubkey.asc X-PGP-KeyId: B5FB2CD6 X-cunselcu'a-valsi: korbi User-Agent: Mutt/1.5.21 (2010-09-15) X-Original-Sender: mbays@sdf.org X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com: best guess record for domain of mbays@sdf.org designates 192.94.73.19 as permitted sender) smtp.mail=mbays@sdf.org Reply-To: lojban@googlegroups.com Precedence: list Mailing-list: list lojban@googlegroups.com; contact lojban+owners@googlegroups.com List-ID: X-Google-Group-Id: 1004133512417 List-Post: , List-Help: , List-Archive: Sender: lojban@googlegroups.com List-Subscribe: , List-Unsubscribe: , --jI8keyz6grp/JLjh Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable * Friday, 2011-09-16 at 12:23 -0400 - John E. Clifford : > [1] Sincei have doubts about what {zo'e} means, I'm not sure what {mii > na klama zo'e} means, but I suppose it must be either "I do not go > there", where "there" is some place obvious from context (the value of > the Skolem function for the argument mi) or "I don't go anywhere", if > it's a particular quantified variable, certainly not you second > choice. I'd allow also things like A: xu do pu klama su'o friko gugde B: mi pu na klama with the latter meaning that B hasn't been to any African country. > The problem with short-scope variables ( especially unstated > one) is just their recalcitrant behavior under negation and overt > quantifiers. {zo'e} presumably keeps it's referent under negation, > while a quantifier would change radically. But these behaviours seem to be exactly what we want from {zo'e}. What's more, I now think they actually are what we want from {lo}, and that we the definition of {lo} I'm suggesting setting {lo} =3D=3D {zo'e noi} really does explain many uses of {lo}. Let me expand on that slightly. To reiterate and amend, I'm suggesting that we understand {zo'e} as follows: * All omitted numbered places are filled with {zo'e} * By a "zo'e expression" I mean an instance of {zo'e} along with any attached relative clauses, e.g. {zo'e noi broda}. * Let us assume the Nirvana Conjecture: When interpreting lojban, other rules reduce to the problem of determining the truth value in a given possible world of a bridi whose sumti are all either elements of the universe or are {zo'e} expressions (or are anaphora to the latter, but let's ignore that). So reordering, we have selbri(c_1,...,c_n,zo'e_1,...,zo'e_m). * Interpret this as \exists (x_1,...,x_m) \in C. (selbri(c_1,...,c_n,x_1,...,x_m)) * C here is a glorked subset of the mth cartesian power of the universe; it depends on the current context, in particular on any quantified variables the current formula is in the scope of. * Importantly, C is required to be such that any (x_1,...,x_m)\in C satisfy all relative clauses in the zo'e expressions. * Note that {noi} and {poi} have the same effect for existential quantifiers, so {zo'e noi} is the same as {zo'e poi}. * Handling plurals: take our universe to contain pluralities as well as atoms, as discussed elsewhere and as presented nicely in Chierchia98 section 2.1. * Handling kinds: also handled, if handled we want them to be, just by having them in our universe, as in beloved Chierchia98. Note that it is fairly easy to prove the Nirvana Conjecture for sufficiently restricted fragments of Lojban understood appropriately, in particular where we understand {lo} as {zo'e noi}. With this understanding of {zo'e}, is {lo} as {zo'e noi} actually right? I think it is. e.g. in the above exchange, it would allow B to answer A: xu do pu klama su'o friko gugde B: mi na klama lo gugde With this {zo'e} rule, that easily gets the intended interpretation, and I don't see any other non-horrific way it could. It also explains many of xorxes' uses of {lo} which I guess he would explain using Kinds. e.g. (from the gadri section page) {ca ro nu mi rere'u catlu lo skina kei mi cpacu lo so'i se cusku poi mi na cpacu ca lo pamoi} "Every time I see a movie for the second time I get all this dialogue that I missed the first time." Here, if we want to interpret {lo so'i se cusku} as referring to actual dialogue rather than a kind, there clearly has to be a sense in which it's within the scope of the {lo skina}. The {zo'e} approach handles this - the {zo'e} corresponding to the {lo se cusku} is within the scope of the {ro nu}, so the corresponding C is allowed to depend on the event, and hence film, in question. In this case, I would probably glork C as referring to all the pluralities of dialogue in the film in question which are so'imei and which the speaker missed (all of) the first time. It remains to handle generic readings of {lo}. As xorxes suggests, this can be done with no changes to the above semantics by using kinds: have certain places of certain predicates (in fact most) be such that when they take a kind, the predication is interpreted in every world as generic quantification over worlds and instances. But sadly this doesn't seem to handle e.g. {ca lo nicte lo cinfo cu kalte lo cidja}, which should be something like Gen (w, n:nicte_w, cin:cinfo_w) ( cabna_w(n) -> \exists cid:cidja_w(cid, cin). kalte_w(cin, cid) ) "For generic (contextually relevant) worlds, nights and lions, with the world cotemporaneous with the night, the lion hunts for something which is food to it." I don't see how any of the current understandings of {lo} could get that existential scoped within that generic... doi xorxes, if you're still listening: how do you get the right meaning there? > [3] Well, I probably don't understand xorxes either, at least partly > because I carry too many of his earlier schemes in mind and partly > because I haven't read either of the works you two keep citing > (indeed, I don't even know the name or source of either). You probably mean Carlson77 and Chierchia98 - see References below. > As for the problem that leads to the need for kinds or generics or > both), I haven't seen the evidence that these can't be handled by just > a bunch of things (admittedly, I don't know exactly how to do some > cases, but that seems just a temporary situation, not a principled > one)? In general, the hard cases seem to be about statistics of one > sort or another, "dodos on average weighed about 20kg", "dodos don't > exist anymore" and so on. Here the single thing needed is just the > bunch of all dodos. I think I see now the problem I have with this solution. Both xorxes and I, and I think most users of lojban, have implicitly been assuming that {lo} is extensional rather than intensional - where by this I hope (ju'oru'e) I am in agreement with formal semantics jargon in meaning that it is interpreted in the context of a possible world and gives an element (or elements, depending on the setup) of the universe, rather than being interpreted as a function from possible worlds to elements. So when we're interpreting a lojban sentence and hit {lo cipnrdodo}, we will always already have a possible world we need to interpret it with respect to, and its referents must satisfy {cipnrdodo} *in that world*. So to express that dodos are extinct using {lo cipnrdodo}, we either have to use the tense system to quantify over possible worlds, or we have to have elements which satisfy {cipnrdodo} in all worlds and are such that their properties correspond to the properties of dodos in other worlds. This is what kinds give us - the kind Dodo satisfies {cipnrdodo} in all worlds, and corresponds to {lo ka cipnrdodo}. Truth conditions for predications (in any world) involving Dodo can therefore involve dodos in other worlds. In particular, it holds in any world that extinct(Dodo). > [4] So, as an alternative, where do I learn about Haskell, whatever > that is? The internet. But I wouldn't advise it if all you want to do is understand my code... I'll make a literate version which explains what it's doing before I release it. References: [Chierchia98]: Chierchia, "Reference to kinds across languages", 1998, Natu= ral Language Semantics Volume 6 Number 4... of which a preprint was easily available the other day, but seems to have now been taken down, which really isn't funny. I'll mail you and any lojbanist who asks a copy of the preprint. [Carlson77]: Carlson, "A unified analysis of the english bare plural", 1977, Linguistics and Philosophy 1 (3). No preprint that I can find. > On Sep 15, 2011, at 20:06, Martin Bays wrote: >=20 > > * Wednesday, 2011-09-14 at 20:04 -0700 - John E Clifford : > >=20 > >> [1] So you would have the referent of {zo'e} be a Skolem function, > >> which returns a different referent for each value of {da}, which seems > >> to get the right result, the same effect could be done by either > >> taking the bunch of possible x2s and picking the right one from > >> context, including especially the choice of x1 > >=20 > > But we can see the difference between an inner existential and a Skolem > > function when we consider negation. Does > > {mi na klama} =3D=3D {na ku zo'u mi klama zo'e} > > (if you'll grant that equivalence for now) mean "it is not true that > > I go to any of [contextually relevant destinations]", or "for some > > contextually relevant destination D, I do not go to D"? I think the > > former, and I don't see how you'd get at it with Skolem functions. > >=20 > >> (remember, I am not > >> convinced that {zo'e] makes any sense at all or which of its several > >> senses actually works or is at work in a given case, so problems here > >> are not a major issue for me). C, then, is the range of that function > >> (which then is just the set of possible referents for {zo'e}). I just > >> don't see the gain in throwing in another quantifier, especially given > >> the problem with shifting that result. > >=20 > > What problem is that, sorry? > >=20 > >> [3] It is not clear that you have fewer superfluous entities than > >> xorxes, just different ones (or, at least, different ways of > >> expressing the move that seems to me to be direct). > >=20 > > To be clear, I have not actually made a proposal for the "ontology of > > lojban" which conflicts with xorxes'. I'm not entirely happy with > > xorxes', but that might be because I still don't really understand it. > >=20 > >> I would think that > >> the way to deal with bare plurals in English (etc.) would be to use > >> plurals (or Lesniewskian sets, if your are hung up that way). > >> I dson't see how talking about things other than bears, even if you > >> somehow get back to them eventually from these new critters, is at all > >> clarifying to a simple plural. I can say that bears shit in the woods > >> meaningfully and truthfully because, by and large, bears do shit in > >> the woods: at least a subjective statistic and maybe even backed up by > >> objective ones. What do generics add? Even if you go off into > >> subjunctives, it is still bears, not bearkind you are dealing with. > >> Bearkind don't shit anywhere. Or Mr. Bear. How is all that stuff > >> about the number of unicorns indicating the number of unicorn horns an > >> improvement on saying "Wherever there are unicorns there are unicorn > >> horns" just thusly? Note, that, if this means that because unicorns > >> are in the domain of discourse, then unicorns are also does not apply, > >> so the principle is, in fact, not valid at the meta level, just in > >> very worlds. Widespread creatures are numerous (collective property) > >> and occur in lots of places (disjunctive), rare creatures are few and > >> far between. No worlds, no tenses -- even spatial ones, just ordinary > >> predicates. =20 > >=20 > > As far as I can tell, the reason for invoking Kinds is that they do, in > > natural languages, satisfy predicates which can't sensibly be reduced to > > instances like this - e.g. "Bears' closest living relatives are > > the pinnipeds and musteloids", or "the dodo is extinct". In natural > > languages, the same noun phrases are used for generic references > > ("the dodo weighed around 20 kg"). Formal semanticists wanted a single > > entity which could handle both uses... hence kinds. > >=20 > > (This is just my ill-informed impression... probably there's more to it) > >=20 > >> [4] I'll be happy to see whatever you have; I can use a refresher. > >> [Later: looking at what you have sente suggests that you have gone > >> beyond the Montague grammar to parsers and the like in languages > >> I don't know. I would rather just see the Montague grammar for now, > >> not the stuff built on it yet.] > >=20 > > Yes, it's written in Haskell rather than maths. Converting from the code > > to standard logical notation would be in principle straightforward but > > in practice painful and time-consuming, so I have no plans to do that. > > Sorry. > >=20 > > Martin > >=20 > >> ----- Original Message ---- > >> From: Martin Bays > >> To: lojban@googlegroups.com > >> Sent: Wed, September 14, 2011 6:20:07 PM > >> Subject: Re: [lojban] {zo'e} as close-scope existentially quantified p= lural=20 > >> variable > >>=20 > >> * Wednesday, 2011-09-14 at 12:05 -0400 - John E. Clifford=20 > >> : > >>=20 > >>> [1] quantified variables are variables, they do not refer to anything > >>> but can be replaced under certain condition by referring expression. > >>> {zo'e} is officially a referring expression, hence not a variable; > >>> what it refers to is both vague and ambiguous (the obvious thing, it > >>> is unimportant what). To be sure, at some levels of analysis, it is > >>> useful to take referring expression as universal scope particular > >>> quantifiers, but that doesn't operate here. The thing about referring > >>> expressions is that they pass through quantifiers undisturbed, whereas > >>> bound variables, eve with tight scope, do not in general. > >>=20 > >> Yes, but we seem to want the latter. In {ro da klama zo'e}, the intend= ed > >> referents of {zo'e} will depend on the x1. xorxes would fix this with > >> kinds/generics; I'd prefer the straightforward solution of making the > >> zo'e scope within the universal quantifier. > >>=20 > >>> [...] > >>> tI am still not clear what C is. > >>=20 > >> Perhaps it would clarify if I re-express it in a purely singularist wa= y: > >>=20 > >> I'm suggesting we analyse {zo'e} as introducing, at tightest scope, an > >> existential quantifier "\exists x \in C", where the C is a contextually > >> determined set. So e.g. > >> {zo'e broda ro da} -> \forall y. \exists x\in C_y. broda(x,y) > >>=20 > >> I confused things slightly by working with a plural semantics. > >>=20 > >>> [2] Assuming "the sum of all chihuahuas" means a bunch of chihuahuas > >>> that gets them all in, how does this focus the referent of the > >>> pronoun to some chihuahuas in a way that just having chihuahuas in > >>> the domain does not? > >>=20 > >> Is this now clear? If context indicates that C is the set of all > >> chihuahuas, the zo'e means "whatever chihuahuas" - which is somewhere = in > >> between "whatever" (C=3Duniverse) and "Chichi" (C=3D{Chichi}). > >>=20 > >>> Sent from my iPad > >>> [3] What is more natural than having a bunch of things? What seems > >>> unnatural to ( Nominalist) me is having in addition a generic > >>> chihuahua (and maybe chihuahua kind besides). > >>=20 > >> I would prefer to do without them. Xorxes thinks this > >> impractical/undesirable. > >>=20 > >>> So, reading between the lines of your description of a kind, brodakind > >>> is pretty much brodaness, the function from possible wolds to the > >>> ( set of) brodas in that world (tense and the like merely deal with > >>> certain regular complexities in the system of worlds). But properties > >>> don't love people -- or hate them. > >>=20 > >> Yes. Chierchia (and, basically, Carlson) deal with this by > >> type-shifting. If you use a kind as an argument in a predicate which > >> expects mundane individuals, to make the types work out either an > >> existential or generic-universal quantifier over instances of the kind > >> is introduced (which depends on the sentence). They use this to explain > >> bare plurals in english, and corresponding phenomena in other natural > >> languages. > >>=20 > >>> {pavyseljirna} btw, vs is not legal medial. > >>=20 > >> (thanks) > >>=20 > >>> But talking about unicorns does not mean that unicorn horn have to be > >>> in the universe of discourse. Using metrology here means breaking up > >>> bunches of thing in the universe, not breaking up individual things in > >>> the universe. > >>=20 > >> I wasn't saying it was true, just an example of a statement we can make > >> using kinds (or properties). > >>=20 > >>> We can talk about widespread and rare creatures without tenses, even > >>> without location tenses. > >>=20 > >> Without using kinds? How? > >>=20 > >>> I'm not sure which notation you are using, but I suppose that "cup" is > >>> the avatar counting function for situations. > >>=20 > >> Sounds right. I took it from Chierchia - \cup K is the map from > >> situations to the sum of all instances of the kind K in the situation. > >>=20 > >>> [4] hideosities from Richard's classes 50 years or so ago but not > >>> successfully resolved so far as I can tell: tanru, compound predicates > >>> (without transformations), ditto arguments and argument sequences, > >>> "modals", UI, {du'u}, and on and on. > >>=20 > >> Yes... many of these I'd expect to just ignore for now, or implement > >> hackily and approximately. > >>=20 > >>> Can I see what you have so far; I haven't mucked in this midden for > >>> years and it would be interesting to see what progress there has been= =2E =20 > >>=20 > >> Little if any progress beyond previous attempts, I suspect. But I'll > >> send you my work-in-progress anyway. (In a separate mail; it isn't > >> anything like ready for public release) > >>=20 > >>> [...] > >>> Bunches operate intensionally like anything else (except, as a bow to > >>> xorxes, they aren't things). A bunch of broda can be anything from > >>> a single broda to all the broda ever on all possible worlds (including > >>> impossible situations). You do have to indicate, somehow, what your > >>> range is (and thus specify your universe of discourse a bit).=20 > >>=20 > >> Let me ponder this. > >>=20 > >> Martin > >>=20 > >>> On Sep 14, 2011, at 0:13, Martin Bays wrote: > >>>=20 > >>>> * Monday, 2011-09-12 at 19:52 -0700 - John E Clifford=20 > >> : > >>>>=20 > >>>>> [1] The new definition (I don't know how much it differs from the > >>>>> old) seems defective, since {zo'e} ought not be able to stand for a= ny > >>>>> variable, not just {no da}. {zo'e} is a referring expression, not > >>>>> a variable, although what it refers to may be different in each > >>>>> occurrence (or, more often, non-occurrence). > >>>>=20 > >>>> How does that differ from using a tight-scope existential quantifier? > >>>>=20 > >>>>> As for [zi'o} radically changing the meaning of a predicate, it doe= s, > >>>>> but often in the interest of making the meaning of the predication > >>>>> clearer by removing irrelevant considerations. Note, nothing seems = to > >>>>> say that {zi'o} can't or shouldn't be the reading of a blank. > >>>>=20 > >>>> Well, the common understanding seems to be that it can't. > >>>> (If the common understander is reading and disagrees, please speak) > >>>>=20 > >>>> e.g. > >>>> http://dag.github.com/cll/7/7/ > >>>> is reasonably clear in saying that omitted places should be consider= ed > >>>> filled with {zo'e}s. > >>>>=20 > >>>>> I'm not sure what C is (context?), > >>>>=20 > >>>> C is determined by context, hence the choice of letter. It's just > >>>> some things. > >>>>=20 > >>>>> but it is pretty clear that many {zo'e} do not refer to things in > >>>>> that, since they refer to irrelevancies, by saying they are irrelev= ant > >>>>> (I suppose that depends on how you set up your universe, but anythi= ng > >>>>> not mentioned can be dispensed with in almost any way of doing that= =2E) > >>>>=20 > >>>> Is this different from existentially quantifying over the universe? > >>>>=20 > >>>>> [2] I don't follow this at all. Why need the referent of {zo'e} be= in one=20 > >>>>> pragmatic range or the other depending on the size of the possibili= ties. =20 > >>> The=20 > >>>=20 > >>>>> crucial question seems to be about being grokked from context, whic= h seems=20 > >>>>> independent of the possible answers. [mi klama] has a nearly infin= ite=20 > >>> number of=20 > >>>=20 > >>>>> fill-ins for x2, yet in the given case is taken to be an "obvious" = case. Or=20 > >>> is=20 > >>>=20 > >>>>> that what C does, fine down the range? > >>>>=20 > >>>> Yes, that was the idea. > >>>>=20 > >>>>> And , if so, how would the sum all chihuahuas help do that, since t= hat > >>>>> seems a rather big range (whatever it is, btw). > >>>>=20 > >>>> This would handle the case of context indicating that the {zo'e} sho= uld > >>>> mean "some chihuahuas". > >>>>=20 > >>>>> [3] Try mine: {lo broda} refers to a bunch of broda (which bunch is > >>>>> contextually determined). This bunch is related to a predicate in = any > >>>>> of a variety of ways: collective, conjunctive, disjunctive, some > >>>>> intermediate forms in which subbunches are related in various > >>>>> collective ways (the individual cases being the bottom row of this), > >>>>> and various "statistical" cases (which, admittedly, take more work, > >>>>> most of which is yet to be done). Your different sorts of > >>>>> generics/kinds seem to me to be just different ways that a bunch can > >>>>> satisfy a predicate (the "tendency" sort being of the > >>>>> not-quite-worked-out sort, the others being of more familiar sorts). > >>>>=20 > >>>> Maybe (i.e. depending on what exactly you mean) this can handle kind= s, > >>>> but I don't think it would be very natural. > >>>>=20 > >>>> Following Chierchia98 once more, we can consider the kind Broda to be > >>>> a reification of the map from situations (which for simplicity, igno= ring > >>>> tense issues, we can identify these with possible worlds in Kripke > >>>> semantics) to the (sum of the) actual brodas in that situation. > >>>>=20 > >>>> So we can say things like "unicorns necessitate unicorn horns" and m= ean > >>>> that any situation with a unicorn also has a unicorn horn > >>>> \forall s. ( \cup(Unicorn)(s) !=3D 0 -> \cup(UnicornHorn)(s) !=3D 0 = ) . > >>>>=20 > >>>> (I know... unicorns again... sorry for lack of imagination) > >>>>=20 > >>>> (if we add time and space information to situations, we could also t= alk > >>>> about a kind being 'rare' or 'widespread') > >>>>=20 > >>>> Does this mean something about {lo pavseljirna} and {lo pavseljirna > >>>> jirna} where these refer to bunches? If the bunch refered to by {lo > >>>> pavseljirna} consists of all unicorns in all possible worlds, and co= mes > >>>> with the data as to which unicorns are in which possible world, then > >>>> yes. But that's a rather specific bunch, and a lot of information to= be > >>>> carrying around. > >>>>=20 > >>>> In reality, I think I'd prefer to translate the above sense of "unic= orns > >>>> necessitate unicorns horns" into lojban by something like > >>>> {lo ka pavseljirna cu za'e zatni'i lo ka pavseljirna jirna}. > >>>>=20 > >>>>> As for Mr. Broda, he has been around for at least thirty years, > >>>>> arising from a conflation of Quine and some social anthorpologist > >>>>> dealing with Trobriand Islanders, and has had more definitions and > >>>>> explanations that I can count up from memory, but basically it is > >>>>> something present wherever a broda is present doing whatever the br= oda > >>>>> does (i.e., a distributive predication of the bunch). > >>>>>=20 > >>>>>=20 > >>>>> [4] At a certain point, it became clear that all the various chats > >>>>> about brodas were talking about the same thing, so it seemed to fol= low > >>>>> that that should be the simple reference and all the messy details = go > >>>>> elsewhere. Since the details are about how this basic thing, lo > >>>>> broda, trlated to, the predicate involved, it seem that the place to > >>>>> put the info is where the two meet up. It is not clear exactly how= to > >>>>> do this, but having a different descriptor for eachcase, when what = is > >>>>> being referred to is always the same, seem bad logical form. =20 > >>>>>=20 > >>>>> Have you ever tried to go a Montague grammar for even a small part = of > >>>>> Lojban? The hideosities of some of the complex structures will blow > >>>>> your mind and you paradigms, although the very basic stuff is pretty > >>>>> straightforward. > >>>>=20 > >>>> Yes, but only the easy bits (those corresponding to FOL) thus far. I= got > >>>> stuck on handling {lo} and {zo'e}... > >>>>=20 > >>>> What kind of paradigm-blowing complexities do you have in mind? > >>>>=20 > >>>>> (By the way, using {zi'o} for "doesn't matter" blanks makes life a = lot > >>>>> easier). > >>>>>=20 > >>>>> [5] WTF is "vague ambiguity"? ambiguity between two vague concepts? > >>>>> Usually the two words contrast with one another. And how is polyse= my > >>>>> different from ambiguity? (term of art?)=20 > >>>>=20 > >>>> I meant "polysemy" to indicate a discrete ambiguity, and "vague > >>>> ambiguity" a continuous one. I think the first at least might be a t= erm > >>>> of art (but not my art). > >>>>=20 > >>>>> Of course, I take {lo broda} to refer to a bunch of broda, vaguely > >>>>> specified, perhaps. The connection with the predicate is then > >>>>> ambiguous (as matters now stand), since there are half-a-dozen > >>>>> possibilities at least, and several of them may be plausible in > >>>>> a given situation. > >>>>=20 > >>>> So I think the main thing I don't understand about your understandin= g of > >>>> {lo} is how these bunches work wrt intension, i.e. how they interact > >>>> with possible worlds and tense. > >>>>=20 > >>>>> [6] It seems the discussion arises from trying to get from {lo tcia= uau > >>>>> cu prami lo prenu} to something that will end up allowing an AE cla= im > >>>>> to be converted, salve vertitatem, into an EA claim. Most of the > >>>>> steps suggested seem implausible at best and none of them seem to t= ake > >>>>> account of the real features being employed, beyond the quantifiers, > >>>>> variously understood (no consideration for the type of connection to > >>>>> the predicates, for a main example). The moves to save parts of th= is > >>>>> seem just desperate.=20 > >>>>>=20 > >>>>>=20 > >>>>> ----- Original Message ---- > >>>>> From: Martin Bays > >>>>> To: lojban@googlegroups.com > >>>>> Sent: Mon, September 12, 2011 5:52:55 PM > >>>>> Subject: Re: [lojban] {zo'e} as close-scope existentially quantifie= d plural=20 > >>=20 > >>>>> variable > >>>>>=20 > >>>>> * Sunday, 2011-09-11 at 13:50 -0700 - John E Clifford=20 > >>> : > >>>>>=20 > >>>>>> {zo'e} is a strange word. It is more often understood than used a= nd, > >>>>>> when used,=20 > >>>>>>=20 > >>>>>> has primarily a pragmatic function more than a semantic one. It is > >>>>>> one of the stock expansions of the space left by a missing argumen= t in > >>>>>> a bridi, along with pronouns (anaphoric or deictic), regular noun > >>>>>> phrases (to the same purposes as pronouns), {da} etc., and {zi'o}. > >>>>>=20 > >>>>> [1]I thought the modern convention was that {zi'o} isn't allowed (s= ee > >>>>> www.lojban.org/tiki/BPFK+Section%3A+Grammatical+Pro-sumti > >>>>> ). Certainly it makes things more complicated if it is! > >>>>>=20 > >>>>> If we don't allow {zi'o}, it seems that "somethings among C", where > >>>>> C is glorked from context, deals with all the expansions you mentio= n. > >>>>> i.e. {broda zo'e} -> EX X among C. broda(X) > >>>>>=20 > >>>>> (well... technically this doesn't handle {da}, if we consider that = to be > >>>>> a singular variable, but it's close to doing so) > >>>>>=20 > >>>>>> Its pragmatic function flows from the laws of quantity; it means "I > >>>>>> don't need to tell you what", either because you already know from > >>>>>> context or because it plays no role in the story being told. Each > >>>>>> occurrence of it is a constant, but separate occurrences are > >>>>>> independent (L3). =20 > >>>>>> In its "plays no role" role, ir does not imply that speaker knows = what its=20 > >>=20 > >>>>>> reference is, though its context-sensitive role does. > >>>>>=20 > >>>>> [2]I'd want to say that this difference is got at by the choice of = C. If > >>>>> C is something like the sum of all people, we effectively have the > >>>>> "plays no role" idea. If C is small, like {la .alis. joi la bob.} o= r the > >>>>> sum of all chihuahuas, we have the "you-know-what" interpretation. > >>>>>=20 > >>>>>> On the whole, it is an=20 > >>>>>> odd thing to have in a logical language: if something plays no rol= e, then=20 > >>>>>> eliminate the temptation to refer to it with {zi'o}. If it is obv= ious > >>>>>> from the context, then put it in in a minimal way. There ought no= t be > >>>>>> two (let alone half-a-dozen) elision transformations with such var= ied > >>>>>> meanings, without clear clues to choose among them. > >>>>>=20 > >>>>> Quite. > >>>>>=20 > >>>>>> In any case, it seems a weak base to build an explanation of {lo} = on. > >>>>>> To do this latest. it has to mean (as it does not obviously in the > >>>>>> original situation) "what I have in mind" -- something I may at=20 > >>>>>> some point indeed have to tell you. In addition, MB, at least, > >>>>>> seems to thing it should also mean some specified kind/generic: {= lo > >>>>>> broda} refers to brodakind, the generic broda or (as we used to sa= y) > >>>>>> Mr. Broda. > >>>>>=20 > >>>>> [3]MB doesn't really think that any more. MB is (belatedly) largely > >>>>> despairing of getting a neat and politically acceptable theory of {= lo}. > >>>>>=20 > >>>>> He also thinks we should be careful to separate generic brodas, whi= ch > >>>>> satisfy a predicate iff brodas tend to satisfy it, from kinds which= can > >>>>> have entirely different properties (like being widespread). > >>>>>=20 > >>>>> He never understood who Mr. Broda was meant to be. > >>>>>=20 > >>>>>> xorxes agrees that this is a possible reading of {lo broda} -- > >>>>>> without, that I can see, driving this back onto {zo'e}. All of th= is, > >>>>>> needless to say, takes place a long way from the realm of gaps in > >>>>>> a predicate place structure, and so it is hard to get a grasp on t= he > >>>>>> arguments. Starting from the original meaning, we get the=20 > >>>>>> following > >>>>>> 1E* Everybody is loved by some chihuahua (this seems to me, on the > >>>>>> basis of my experiences with chihuahuas, to be absurd, but anyhow) > >>>>>> 2E* Everybody is loved by the thing, which is a chihuahua. > >>>>>> 3E* The thing, which is a chihuahua, loves everybody > >>>>>> 4E* Some chihuahua loves everybody. =20 > >>>>>> 4>1, 1/4, 2=3D3, 2>1, 3>4, so 3>1, 2>4 and 1/2. 1/3, 4/2,3 > >>>>>> These remarks are essentially domain independent (ignoring domains > >>>>>> with only one chihuahua) and I don't see any reason why one domain > >>>>>> rather than another is to be chosen, unless it is to make an infer= ence > >>>>>> from 1 to 2 or 4 to 3 go through. =20 > >>>>>> But domains with only one chihuahua (assuming it has any at all) a= re highly=20 > >>>=20 > >>>>>> implausible. So, I don't get what is going on here (or, rather, I= do, but=20 > >>> am=20 > >>>=20 > >>>>>> damned if I will give it the satisfaction of actually saying it so= far). > >>>>>=20 > >>>>>> So, to go back to basics, {lo broda} refers to a bunch of broda (no > >>>>>> metaphysical commitments here, just a fac,on de parler that is eas= ier > >>>>>> to work with in some fine cases) and iis neutral with respect to h= ow > >>>>>> its referent is related to various predicates (including {broda}, = in > >>>>>> fact). Lojban lacks the means to say explicitly what that relatio= n is > >>>>>> in most cases, but, in any case, saying what that relation is is n= ot > >>>>>> a matter to be taken up by a gadri, but at the sumti-selbri interf= ace, > >>>>>> if at all. > >>>>>=20 > >>>>> [4] I'm not sure what you mean by all that. > >>>>>=20 > >>>>> But my starting point here is that the sumti-selbri interface shoul= d be > >>>>> simple - corresponding to elements and relations in a Kripke model = (or > >>>>> some more baroque structure along the same lines), as in Montague-s= tyle > >>>>> formal semantics. > >>>>>=20 > >>>>> Are you saying that you think this simply inappropriate for lojban? > >>>>>=20 > >>>>>> And, it needs to be noted, the choices are not limited to > >>>>>> (conjunctive) distribution and collection, but include at least > >>>>>> disjunctive distribution > >>>>>=20 > >>>>> Why would you want to include that? > >>>>>=20 > >>>>>> and various statistical and quasi-statistical > >>>>>> modes. There does not appear to be any reason to think that [lo > >>>>>> broda] is ambiguous rather than vague (just what all did I have in > >>>>>> mind) > >>>>>=20 > >>>>> [5]I think that if we allow {lo broda} to be the Kind broda, this w= ould be > >>>>> polysemy rather than just vague ambiguity. > >>>>>=20 > >>>>> But it sounds like you don't want to? > >>>>>=20 > >>>>>> or that the arguments above, with {lo broda} in place of {zo'e > >>>>>> noi broda} would go through unmodified (in a sensible way, rather = than > >>>>>> dragging in an odd domain). > >>>>>=20 > >>>>> [6]I'm not sure what you mean here. > >>>>>=20 > >>>>> Martin > >>>>>=20 > >>>>>=20 > >>>>>> ----- Original Message ---- > >>>>>> From: Martin Bays > >>>>>> To: lojban@googlegroups.com > >>>>>> Sent: Sun, September 11, 2011 12:46:56 PM > >>>>>> Subject: Re: [lojban] {zo'e} as close-scope existentially quantifi= ed plural=20 > >>>=20 > >>>>>> variable > >>>>>>=20 > >>>>>> [notational note: I looked again at some of the formal semantics > >>>>>> literature, and it seems that 'kind' is preferred to 'generic' for= the > >>>>>> abstract individuals like 'chihuahuas' we've been talking about - = which > >>>>>> I think corresponds to what are called 'kinds' in e.g. Chierchia > >>>>>> "References to Kinds across Languages" 1998; 'generic' seems to be > >>>>>> reserved for the {lo'e} idea of "typical individuals". I use 'kind' > >>>>>> below.] > >>>>>>=20 > >>>>>> * Saturday, 2011-09-10 at 15:04 -0300 - Jorge Llamb=EDas=20 > >>> : > >>>>>>=20 > >>>>>>> On Sat, Sep 10, 2011 at 12:10 PM, Martin Bays wro= te: > >>>>>>>> * Saturday, 2011-09-10 at 10:43 -0300 - Jorge Llamb=EDas=20 > >>>>>>> : > >>>>>>>>>=20 > >>>>>>>>> 1L: ro prenu cu se prami su'o tciuaua > >>>>>>>>> 1E: Everyone is loved by some chihuahua. > >>>>>>>>>=20 > >>>>>>>>> 2L: ro prenu cu se prami zo'e noi tciuaua > >>>>>>>>> 2E: Everyone is loved by chihuauas. > >>>>>>>>>=20 > >>>>>>>>> 3L: zo'e noi tciuaua cu prami ro prenu > >>>>>>>>> 3E: Chihuahuas love everyone. > >>>>>>>>>=20 > >>>>>>>>> 4L: su'o tciuaua cu prami ro prenu > >>>>>>>>> 4E: Some chihuahua loves everyone. > >>>>>>>>>=20 > >>>>>>>>> We also have two domains of discourse: > >>>>>>>>>=20 > >>>>>>>>> D1 =3D {lo prenu ku xi pa, lo prenu ku xi re, lo prenu ku xi ci= , .., lo > >>>>>>>>> tciuaua ku xi pa, lo tciuaua ku xi re, ...} > >>>>>>>>> =3D {person_1, person_2, person_3, ...., chihuahua_1, chihua= hua_2,=20 > >>>>> ...} > >>>>>>>>>=20 > >>>>>>>>> D2 =3D {lo prenu ku xi pa, lo prenu ku xi re, lo prenu ku xi ci= , .., lo=20 > >>>>>>> tciuaua} > >>>>>>>>> =3D {person_1, person_2, person_3, ...., chihuahuas} > >>>>>>>>>=20 > >>>>>>>>> D1 and D2 are not the same domain. Sentences 1 and 4 "put us" in > >>>>>>>>> domain D1, while sentences 2 and 3 "put us" is domain D2. By th= at I > >>>>>>>>> mean that those are the natural domains in which to interpret t= hose > >>>>>>>>> sentences without any more context. Do we agree so far? > >>>>>>>>=20 > >>>>>>>> Not entirely. I think 1E-4E could just as well be interpreted in= the > >>>>>>>> union D12 of D1 and D2 - because a sentence involving "some chih= uahua" > >>>>>>>> can't have the generic "chihuahuas" as an witness, and although > >>>>>>>> (as in Carlson) a predication involving "chihuahuas" is ambiguous > >>>>>>>> between being about the generic and about its > >>>>>>>> manifestations/stages/whatever, that doesn't mean the domain of > >>>>>>>> discourse has to be different for different interpretations. > >>>>>>>=20 > >>>>>>> The English situation is additionally complicated by the > >>>>>>> singular/plural morphology. You say that "some chihuahua" can't h= ave > >>>>>>> chihuahuas as a witness, even when chihuahuas are in the domain of > >>>>>>> discourse. But why is that? Is it because only chihuahuas can be a > >>>>>>> witness, and chihuahuas are not chihuahuas? That can't be the rea= son > >>>>>>> because chihuahuas are indeed chihuahuas. I think it has to do wi= th > >>>>>>> something like the witness has to satisfy "...is a chihuahua", an= d not > >>>>>>> just "...are chihuahuas". So those two are different predicates in > >>>>>>> English, at least when the domain of discourse is D12. In Lojban = we > >>>>>>> have to make do with "tciuaua" for both. > >>>>>>=20 > >>>>>> The situation in English is rather strange. The singular does inde= ed > >>>>>> seem to refer specifically to individuals. Meanwhile the plural is > >>>>>> ambiguous between pluralities of individuals and pluralities of > >>>>>> strict subkinds - "some chihuahuas" can't be witnessed by the kind > >>>>>> 'chihuahuas', but it *can* be witnessed by 'black chihuahuas' or '= dead > >>>>>> chihuahuas'. > >>>>>>=20 > >>>>>> So it seems English does differentiate between kinds and mundanes, > >>>>>> but it confuses the two in plurals. > >>>>>>=20 > >>>>>> This does seem to really be a binary ambiguity, though. I've tested > >>>>>> a few native english speakers on the phrase "some dogs love everyo= ne; > >>>>>> indeed, chihuahuas do", and they report understanding the intentio= n, but > >>>>>> experiencing some surprise on reaching the second clause and havin= g to > >>>>>> re-evaluate the first clause to refer to kinds rather than individ= uals. > >>>>>>=20 > >>>>>> Further evidence for its binary nature: > >>>>>> *"many dogs love everyone; indeed Barney does, chihuahuas do..." > >>>>>> is, I think, semantically anomalous. > >>>>>>=20 > >>>>>> I don't see why we should import this ambiguity to lojban. Even ap= art > >>>>>> from all the problems it causes which it doesn't cause in english = ('I > >>>>>> hate dogs' doesn't imply 'I hate one or more dogs', even when > >>>>>> interpreted in the same domain of discourse), I would think it > >>>>>> unlojbanic to have a binary ambiguity - especially one which can't= be > >>>>>> straightforwardly and clearly disambiguated. > >>>>>>=20 > >>>>>>>> You seem to be saying that D12 is an intrinsically unnatural dom= ain for > >>>>>>>> lojban. That seems to be a difference from english. > >>>>>>>=20 > >>>>>>> I'm saying a domain like D12 needs extra work. For example, inste= ad of > >>>>>>> a single predicate "tciuaua" we need two predicates, to go with t= he > >>>>>>> English "... is a chihuahua" and "... are chihuahuas", which we w= ould > >>>>>>> have to use instead of the plain "tciuaua" to properly restrict t= he > >>>>>>> quantifier. > >>>>>>=20 > >>>>>> Yes, I think something like this might be the solution. We allow d= omains > >>>>>> like D12 as a matter of course, and have a new predicate correspon= ding > >>>>>> to the kind sense of "... are chihuahuas" - i.e. meaning "is > >>>>>> a (non-strict) subkind of the kind 'chihuahuas'". It and {tciuaua}= should=20 > >>> be > >>>>>> mutually exclusive. > >>>>>>=20 > >>>>>> If we have a way of getting explicitly at the kind 'tciuauas', we = can > >>>>>> then use {klesi} to get at subkinds. > >>>>>>=20 > >>>>>> So we need something corresponding to Chierchia's down operator. T= his > >>>>>> shouldn't be {lo'e}, because that's about genericity. I'm wondering > >>>>>> whether it could indeed be {lo} - using Chierchia's type-shifting = (which > >>>>>> is similar to Carlson's quantification over stages) to get back to > >>>>>> existential quantification over instances. The crucial change from= what > >>>>>> you've been suggesting would be that although {lo broda cu broda} = would > >>>>>> hold, and {lo broda} would be referring to an individual in our > >>>>>> universe, it would *not* follow that this individual satisfies the= x1 of > >>>>>> {broda} in the usual sense - rather, {lo broda cu broda} would tra= nsform > >>>>>> by type-shifting to {su'o da poi [instance-of] lo broda cu broda} = (where > >>>>>> {me} might or might not work for [instance-of]). > >>>>>>=20 > >>>>>> I'll read some more of Chierchia and see if I can come up with > >>>>>> a proposal along these lines which would satisfy us both (and hope= fully > >>>>>> everyone else). > >>>>>>=20 > >>>>>>>> (In fact, I *would* like to claim that 1L logically implies 2L, = because > >>>>>>>> I would still like to analyse {zo'e} (but not {lo}) as in the su= bject > >>>>>>>> line of this thread. But that's beside the point.) > >>>>>>>=20 > >>>>>>> So you would like to claim > >>>>>>>=20 > >>>>>>> D1 |=3D 1L > >>>>>>> implies D1 |=3D 2L > >>>>>>>=20 > >>>>>>> and that D1 is a natural/preferred domain for 1L. > >>>>>>>=20 > >>>>>>> If that's how you want to analyse "zo'e", you still have to accou= nt > >>>>>>> for the "obvious" as opposed to the "irrelevant" sense of "zo'e".= As > >>>>>>> in: > >>>>>>>=20 > >>>>>>> - xu do djica lo nu klama lo zarci > >>>>>>> - u'u mi na kakne lo nu klama .i ei mi klama lo drata > >>>>>>> "Do you want to come to the market?" > >>>>>>> "Sorry, I can't go. I have to go somewhere else." > >>>>>>>=20 > >>>>>>> That should not come out as "I can't go anywhere." > >>>>>>=20 > >>>>>> Hmm. Yes, it isn't a simple existential quantifier. But how about = just > >>>>>> having the domain of the existential quantification be contextually > >>>>>> determined - i.e. the quantifier is "for some xs such that context > >>>>>> suggests I would likely be talking about xs here". The domain > >>>>>> 'everything' would always be plausible; other domains like {the ma= rket > >>>>>> we just mentioned} or 'all markets' would be plausible in certain > >>>>>> contexts. > >>>>>>=20 > >>>>>> Martin > >>>>>>=20 > >>>>>> --=20 > >>>>>> You received this message because you are subscribed to the Google= Groups=20 > >>>>>> "lojban" group. > >>>>>> To post to this group, send email to lojban@googlegroups.com. > >>>>>> To unsubscribe from this group, send email to=20 > >>>>>> lojban+unsubscribe@googlegroups.com. > >>>>>> For more options, visit this group at=20 > >>>>>> http://groups.google.com/group/lojban?hl=3Den. > >>>>>>=20 > >>>>>=20 > >>>>> --=20 > >>>>> You received this message because you are subscribed to the Google = Groups=20 > >>> "lojban" group. > >>>>> To post to this group, send email to lojban@googlegroups.com. > >>>>> To unsubscribe from this group, send email to=20 > >>> lojban+unsubscribe@googlegroups.com. > >>>>> For more options, visit this group at=20 > >>> http://groups.google.com/group/lojban?hl=3Den. > >>>>>=20 > >>>=20 > >>> --=20 > >>> You received this message because you are subscribed to the Google Gr= oups=20 > >>> "lojban" group. > >>> To post to this group, send email to lojban@googlegroups.com. > >>> To unsubscribe from this group, send email to=20 > >>> lojban+unsubscribe@googlegroups.com. > >>> For more options, visit this group at=20 > >>> http://groups.google.com/group/lojban?hl=3Den. > >>>=20 > >>=20 > >> --=20 > >> You received this message because you are subscribed to the Google Gro= ups "lojban" group. > >> To post to this group, send email to lojban@googlegroups.com. > >> To unsubscribe from this group, send email to lojban+unsubscribe@googl= egroups.com. > >> For more options, visit this group at http://groups.google.com/group/l= ojban?hl=3Den. > >>=20 >=20 > --=20 > You received this message because you are subscribed to the Google Groups= "lojban" group. > To post to this group, send email to lojban@googlegroups.com. > To unsubscribe from this group, send email to lojban+unsubscribe@googlegr= oups.com. > For more options, visit this group at http://groups.google.com/group/lojb= an?hl=3Den. >=20 --jI8keyz6grp/JLjh Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (GNU/Linux) iEYEARECAAYFAk52KnYACgkQULC7OLX7LNZM7QCfS5g+VPtFpHNQrPdK24V4QSlZ a6YAoN/a2XygggZoy1xz2gdI/sEFogvl =FyP9 -----END PGP SIGNATURE----- --jI8keyz6grp/JLjh--