From nobody@digitalkingdom.org Wed Aug 13 11:37:18 2008 Received: with ECARTIS (v1.0.0; list lojban-beginners); Wed, 13 Aug 2008 11:37:18 -0700 (PDT) Received: from nobody by chain.digitalkingdom.org with local (Exim 4.69) (envelope-from ) id 1KTLDg-0000V6-UD for lojban-beginners-real@lojban.org; Wed, 13 Aug 2008 11:37:18 -0700 Received: from yx-out-1718.google.com ([74.125.44.154]) by chain.digitalkingdom.org with esmtp (Exim 4.69) (envelope-from ) id 1KTLDb-0000U9-W2 for lojban-beginners@lojban.org; Wed, 13 Aug 2008 11:37:16 -0700 Received: by yx-out-1718.google.com with SMTP id 4so60982yxp.46 for ; Wed, 13 Aug 2008 11:37:09 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:in-reply-to:mime-version:content-type:references; bh=R5e97pFvIDq+CGeyIXsZki3tTWfdST6A1H2pNeaWos8=; b=E8bBakeGRxReWgR0y8rs4yFPtBizag2YkqmxM0Cc8vX8urYUHr2BwdP1qDXy5MUYrl 3rpKJZqPTunnjyO9sp6nxv7W36b7JOPa8tX2urBXxyD7m4jPYINLUTUT37yZWoDaa704 /c9c0B7moT+6qdGQ6bLXS4JpVvB0mwgNEXPpA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:in-reply-to:mime-version :content-type:references; b=PX1qjureJmQB6G7EQlPI+pu+jfnztHSvL0FVZ27qxHwL1pBjO6MYpToSAFiqaNaR08 4w3e9qisF6aDEmbaQIzlK/rU+sszbih7zEYY9lzFmUv9e2e3dKl3geF0GvSrDAQcyvZJ fEXFipJR63eJYh80ALEsrQhHyZK+DWzm4aYZU= Received: by 10.142.240.9 with SMTP id n9mr66826wfh.136.1218652628388; Wed, 13 Aug 2008 11:37:08 -0700 (PDT) Received: by 10.142.147.16 with HTTP; Wed, 13 Aug 2008 11:37:08 -0700 (PDT) Message-ID: Date: Wed, 13 Aug 2008 14:37:08 -0400 From: "Brett Williams" To: lojban-beginners@lojban.org Subject: [lojban-beginners] Re: go'i / ri In-Reply-To: <44788.74123.qm@web30408.mail.mud.yahoo.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_67243_1850811.1218652628383" References: <44788.74123.qm@web30408.mail.mud.yahoo.com> X-Spam-Score: 0.0 X-Spam-Score-Int: 0 X-Spam-Bar: / X-archive-position: 819 X-ecartis-version: Ecartis v1.0.0 Sender: lojban-beginners-bounce@lojban.org Errors-to: lojban-beginners-bounce@lojban.org X-original-sender: mungojelly@gmail.com Precedence: bulk Reply-to: lojban-beginners@lojban.org X-list: lojban-beginners ------=_Part_67243_1850811.1218652628383 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On 8/12/08, Tom Gysel wrote: > > coi rodo, > > Why does 'go'i' always have 'le' in front of it, while 'ri' has not? > I cannot see a difference. "ri" is a pro-sumti. Pro-sumti have some interesting characteristics: - They consist themselves of sumti. Saying them inserts into the present bridi a particular referent. This isn't exactly the case with "lo" and friends-- once you've said "lo", you've *started* to put a sumti into the bridi, but you have some more words to say before you're finished. "mi citka ko'a", I eat ko'a, sentence finished-- "ko'a" by itself fills the x2 space. "mi citka lo", I eat a..., an unfinished sentence, awaiting an internal selbri to finish the "lo" construction. - They terminate selbri. This is an invisible feature, a silent implication, that's worth taking a second look at. Selbri can go for many words without terminating, for instance any number of brivla in a row like "xekri xekri xekri xekri xekri" (I wrote a pome that's 37 xekri in a row) is a simple left-grouped tanru. Since a pro-sumti adds a sumti to the bridi, it implicitly says that any such selbri craziness is finished. For instance, look at how a pro-sumti can stand in for "ku" or "cu" in one of my favorite sentence shapes: "ko'a lo broda ko'e brodi" == "[fa] ko'a [fe] lo broda [ku] [fi] ko'e [cu] brodi" "mi lo sampu do cusku" I say something simple to you. "mi le mi zdani ri cadzu klama" I walkingly go to my house from itself. "go'i" is a pro-selbri. Pro-selbri also have interesting characteristics: - They consist themselves of selbri. They can serve the root purpose of a selbri, which is to be at the center of the main bridi of a sentence saying what its relationship is. For instance: "cusku lo sampu .i mi go'i fi ko .i ko go'i fi mi" -- "Saying something simple. I do it to you. You do it to me." Here the pro-selbri "go'i" repeats or stands for the previous relationship, and allows that relationship to again be the relationship of the present bridi. (Note that arguments already defined-- x2 = "lo sampu"-- are implicitly kept the same unless replaced with something else. And note further that the second "go'i" repeats the first "go'i"'s relationship, referring to the original bridi only indirectly!) - They can participate in tanru. It was not obvious to me at all at first that words like "mo" and "go'i" and "co'e" can all participate in tanru, but now that I grok it, it's quite useful. You also need to be aware of it, because tanru's combinatory tendency makes them selbri-hungry, and you must defend your selbri from them by proper termination (such as with pro-sumti)! Consider: "mi tatpi" (I'm tired) and then the response: "mi milxe go'i" (I'm mildly so). "go'i" forms a tanru with "milxe". Consider: "lo daspo co'e cu .ii klama" A destruction something is (eek!) coming. "daspo" forms a tanru with "co'e", and the "cu" is needed to keep the whole thing from just meaning "a destruction type-of something type-of goer". I love forming tanru questions with "mo", like: "do mo .a'u zukte", You're what kind (I'm interested) of doer? ("mo" forms a tanru with "zukte", oblivious to the UI between.) - They can be used in deeper grammatical structures which take a selbri, of which there are several (sei & fi'o come to mind), but by far the most common, of course, are the gadri, such as lo le la lo'e le'e loi lei, etc. "lo co'e" is a lot like "zo'e". "lo mo" is a lot like "ma", and "lo mo dacti" is a lot like "ma poi dacti". And of course it's very common to see "lo go'i", or even "lo se go'i" and "lo te go'i". The function of "lo" on "go'i" is exactly identical to the function of "lo" on a gismu or any other kind of selbri, so it's good to watch it carefully, because it might just help you pin down how this process works in general. So let's say that we've just said "mi citka lo plise", and now we're examining the tool of "go'i" in that context. The relationship referred to by "go'i" at this moment isn't eating in general, or my eating or eating apples, but the whole thing, my apple eating. We could say next "go'i lo nanba", go'i some bread, and it would mean "I eat bread", because I would still be in there as the x1 of the go'i relationship, not having been replaced with anything. All selbri (save perhaps Sir Co'e) come to us weighted, having some particular roles already assigned to their places. The x1 of plise is already confined to a tiny corner of the world, this particular kind of fruit, and from there we can refine it only to which particular apple in which situation. When the x1 of the previous bridi was "mi", the x1 of "go'i" is similarly but even more specifically constrained: It must be, specifically, me. So-- and this is a very Lojbanic thought so don't worry if it doesn't immediately make perfect sense-- in exactly the same way that "lo plise ku" describes a referent which is one of those possible plise x1s, an apple, if the last sentence had "mi" as its x1, then "lo go'i ku" describes a referent which is one of the possible x1s of that relationship, of which there is only one possible, me. On the other hand, we could also use "go'i" to refer to and refine a previous sumti: "ci gerku cu klama .i pa go'i cu xekri" -- "Three dogs are coming. One of them is black." Note that the referent of "pa [lo] go'i [ku]" must be one of the three dogs mentioned in the first bridi, not some other dog; "go'i" takes along the particular referents, not just their characteristics. We can also transform "go'i" in the same way we can transform other selbri, such as with "se". "mi citka lo plise .i lo se go'i cu kukte" -- "I eat an apple. It's delicious." (note that the "cu" is obligatory to prevent the tanru "se go'i kukte") You can refer to parts of the bridi which were left blank, and they'll have their original semantic range of motion, constrained only by the other arguments, so: "mi cusku lo pluja .i lo te go'i cu se cfipu" -- "I said something complicated. The person I said it to was confused." The "te go'i" is the "te cusku", the "fi" place, of the previous bridi, which was left blank, so it could be anyone, but it's definitely the person I said something complicated to. I'm a bit tired of writing at the moment! So I won't explain "lo nu go'i" as much as I meant to. But suffice it to say that it's no different than using "nu" with any other selbri. There's a whole bridi inside, which consists of "go'i" (just like you could just say "go'i" on its own). Here's an example that might help clarify, putting some more arguments into the "nu"ed bridis: "mi citka .i lo'e nu go'i lo plise kei .e lo'e nu go'i lo nanba cu .ui sai se gleki mi" -- I eat. Doing so to an apple and doing so to some bread both (wee!) make me happy. .i ki'e do > je'e mu'o mi'e se ckiku ------=_Part_67243_1850811.1218652628383 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline

On 8/12/08, Tom Gysel <to_mu1975@= yahoo.com> wrote:
<= /table>

 
"ri" is a pro-= sumti.  Pro-sumti have some interesting characteri= stics:

- They consist themselves of sumti.  Saying them inserts into the = present bridi a particular referent.  This isn't exactly the case = with "lo" and friends-- once you've said "lo", you&= #39;ve *started* to put a sumti into the bridi, but you have some more word= s to say before you're finished.   "mi citka ko'a&qu= ot;, I eat ko'a, sentence finished-- "ko'a" by itself fil= ls the x2 space.  "mi citka lo", I eat a..., an unfinished s= entence, awaiting an internal selbri to finish the "lo" construct= ion.

- They terminate selbri.  This is an invisible feature, a silent i= mplication, that's worth taking a second look at.  Selbri can go f= or many words without terminating, for instance any number of brivla in a r= ow like "xekri xekri xekri xekri xekri" (I wrote a pome that'= s 37 xekri in a row) is a simple left-grouped tanru.  Since a pro-sumt= i adds a sumti to the bridi, it implicitly says that any such selbri crazin= ess is finished.  For instance, look at how a pro-sumti can stand in f= or "ku" or "cu" in one of my favorite sentence shapes: = "ko'a lo broda ko'e brodi" =3D=3D "[fa] ko'a [fe= ] lo broda [ku] [fi] ko'e [cu] brodi"  "mi lo sampu do c= usku" I say something simple to you.  &q= uot;mi le mi zdani ri cadzu klama" I walkingly go = to my house from itself.


"go'i" is a pro-selbri.  Pro-selbri also have in= teresting characteristics:

- They consist themselves of selbri. &nb= sp;They can serve the root purpose of a selbri, which is to be at the cente= r of the main bridi of a sentence saying what its relationship is.  Fo= r instance:  "cusku lo sampu .i mi go'i fi ko .i ko go'i = fi mi" -- "Saying something simple.  I do it to you.  Y= ou do it to me."  Here the pro-selbri "go'i" repeat= s or stands for the previous relationship, and allows that relationship to = again be the relationship of the present bridi.  (Note that arguments = already defined-- x2 =3D "lo sampu"-- are implicitly kept the sam= e unless replaced with something else.  And note further that the seco= nd "go'i" repeats the first "go'i"'s relati= onship, referring to the original bridi only indirectly!)

- They can participate in tanru.  It was not obvious to me at all = at first that words like "mo" and "go'i" and "= co'e" can all participate in tanru, but now that I grok it, it'= ;s quite useful.  You also need to be aware of it, because tanru's= combinatory tendency makes them selbri-hungry, and you must defend your se= lbri from them by proper termination (such as with pro-sumti)!  Consid= er: "mi tatpi" (I'm tired) and then the response: "mi mi= lxe go'i" (I'm mildly so).  "go'i" forms a = tanru with "milxe".  Consider: "lo daspo co'e cu .i= i klama" A destruction something is (eek!) coming.  "d= aspo" forms a tanru with "co'e", and=  the "cu" is needed to keep the who= le thing from just meaning "a destruction&nbs= p;type-of something type-of goer".  I love&n= bsp;forming tanru questions with "mo", like: "= ;do mo .a'u zukte", You're what kind (I'm i= nterested) of doer?  ("mo" forms a = tanru with "zukte", oblivious to the UI between.)=  

- They can be used in deeper grammatical structures which take a selbri= , of which there are several (sei & fi'o come t= o mind), but by far the most common, of course, are the gadri, such as= lo le la lo'e le'e loi lei, etc.  &q= uot;lo co'e" is a lot like "zo'e".  "= lo mo" is a lot like "ma", and &quo= t;lo mo dacti" is a lot like "ma&nb= sp;poi dacti".  And of course it's very common to see &q= uot;lo go'i", or even "lo se go'i" and "lo te g= o'i".  The function of "lo" on "= ;go'i" is exactly identical to the function of= "lo" on a gismu or any other kind of&nb= sp;selbri, so it's good to watch it carefully,=  because it might just help you pin down how&= nbsp;this process works in general. 

So let's say that we've just said &quo= t;mi citka lo plise", and now we're exami= ning the tool of "go'i" in that cont= ext.  The relationship referred to by "go'i&qu= ot; at this moment isn't eating in general, or my&n= bsp;eating or eating apples, but the whole thing, = my apple eating.  We could say next &quo= t;go'i lo nanba", go'i some bread, and it&= nbsp;would mean "I eat bread", because I = ;would still be in there as the x1 of&nb= sp;the go'i relationship, not having been replaced&= nbsp;with anything. 

All selbri (save perhaps Sir Co'e) come to us w= eighted, having some particular roles already assi= gned to their places.  The x1 of plise i= s already confined to a tiny corner of the wo= rld, this particular kind of fruit, and from there=  we can refine it only to which particular&nb= sp;apple in which situation.  When the x1&nbs= p;of the previous bridi was "mi", the x1= of "go'i" is similarly but even more spe= cifically constrained: It must be, specifically, m= e.  So-- and this is a very Lojbanic&nbs= p;thought so don't worry if it doesn'= t immediately make perfect sense-- in exactly the&= nbsp;same way that "lo plise ku" describes a = referent which is one of those possible plise x1s,=  an apple, if the last sentence had "mi&= quot; as its x1, then "lo go'i ku" descri= bes a referent which is one of the possi= ble x1s of that relationship, of which there = is only one possible, me. 

On the other hand, we could also use= "go'i" to refer to and refine a previous=  sumti:  "ci gerku cu klama .i pa g= o'i cu xekri" -- "Three dogs are coming.&= nbsp; One of them is black."  Note = that the referent of "pa [lo] go'i [ku]"&= nbsp;must be one of the three dogs menti= oned in the first bridi, not some other dog;&= nbsp;"go'i" takes along the particular&nb= sp;referents, not just their characteristics. 

We can also transform "go'i" in the same way we can trans= form other selbri, such as with "se".  "mi citka lo pli= se .i lo se go'i cu kukte" -- "I eat an apple.  It's= delicious."  (note that the "cu" is obligatory to prev= ent the tanru "se go'i kukte")  You can refer to parts o= f the bridi which were left blank, and they'll have their original sema= ntic range of motion, constrained only by the other arguments, so: "mi= cusku lo pluja .i lo te go'i cu se cfipu" -- "I said&nb= sp;something complicated.  The person I said = it to was confused."  The "te go'i&q= uot; is the "te cusku", the "fi"&nb= sp;place, of the previous bridi, which was&nb= sp;left blank, so it could be anyone, but&nbs= p;it's definitely the person I said something = complicated to.

I'm a bit tired of writing at the moment!  So I won't expl= ain "lo nu go'i" as much as I meant to.  But suffice it = to say that it's no different than using "nu" with any other = selbri.  There's a whole bridi inside, which consists of "go&= #39;i" (just like you could just say "go'i" on its own).=  Here's an example that might help clarify, putting some more arg= uments into the "nu"ed bridis:  "mi citka .i lo'e n= u go'i lo plise kei .e lo'e nu go'i lo nanba cu .ui sai s= e gleki mi" -- I eat.  Doing so to an apple and doing so to some = bread both (wee!) make me happy.




coi rod= o,

Why does 'go'i' always have 'le' in front of it, wh= ile 'ri' has not?
I cannot see a difference.
.i ki&#= 39;e do

je'e

mu'o = mi'e se ckiku

------=_Part_67243_1850811.1218652628383--