[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lojban] Re: Reasoning by analogy


On 2021-01-06 16:13, scope845hlang343jbo@icebubble.org wrote:

Jacob Thomas Errington <jake@mail.jerrington.me> writes:


You can put fi'o tags in more places than that. Try these in a parser.

.i do jamfu cadzu cilre kakne .iseni'ibo fi'o simsa la'e di'u do xance
cadzu cilre kakne

Hm.  I hadn't thought of using a {fi'o} tag on a sumti of selma'o KOhA.


.i do jamfu co'e .i fi'o simsa bo do xance co'e

I don't think that's grammatical.  jufra can only be connected by simple
tags, not full-fledged sumti tcita.  Otherwise, I could simply say:

   do ka'e cilre fi lonu do cadzu fi lo jamfu
   .i fi'o simsa la'edi'u fe'u je semu'i bo
   do ka'e cilre fu lo xego'i fi lonu do cadzu fi lo xance

That would be nice and clean (and could further be simplified by use of
{go'i}, et al.), but the grammar doesn't allow it. .uu


What I wrote, {.i fi'o broda bo} is grammatical. Try it in a parser!

http://camxes.lojban.org/?text=.i%20fi%27o%20broda%20bo
You wrote something slightly different: {.i fi'o broda ko'a bo}. That's indeed ungrammatical. 


The usual strategy to interpret a {fi'o} clause is to rearrange to
make its selbri the top-level selbri. For example, I would interpret
{mi fi'o simsa do se bangu lo lojbo} as

   mi do simsa lo ka lo lojbo cu bangu
   + a claim that {mi se bangu lo lojbo}

No, I don't think that's how modal tags are interpreted.  At least
according to CLL, a modal tag adds an additional place to the underlying
selbri, expressing something that fills the x1 place of the {BAI} or
{FIhO} construct.  Your interpretation of that sentence would properly
be expressed: {lo me mi be fi'o simsa do me'u cu se bangu lo lojbo}.
Adding a modal tag to a sumti is not equivalent to using that tag in a
non-logical connection between sentences.
It turns out that in the decades since the CLL was published that the language has continued to evolve and that interpretations have been refined. The CLL interpretation for fi'o falls a bit flat because it doesn't explain the connection between this new place and the old selbri, whereas the 'new' interpretation is for the most part compatible with the old one (gives essentially the same interpretations) while being more precise. 
What strikes me about that is that it's saying that the similarity
between the possibility of learning to walk on your hands and on your
feet is implied by the possibility of learning to walk on your
feet.

Ah, yes, you have a point, there.  Implication of similarity wasn't the
meaning I'd intended.  My mistake.  This could be fixed using prophor:

     fi'o simsa la'edi'e fe'u do ka'e cilre fi lonu do cadzu fi lo jamfu
     .iseni'ibo do ka'e cilre fu lo xego'i fi lonu do cadzu fi lo xance


An analogy is not an implication. It's an observed similarly that is
used to make inferences. So the order of the bridi operators seems a
bit backwards in the fi'o examples.

Exactly!  That's the whole idea of reasoning by analogy: A and B both
have property X, and A has property Y, therefore (by analogy), B has
property Y.  As an example: Ducks and geese both have webbed feet.
Ducks have feathers.  Therefore, geese have feathers.  It's not logical.
It's reasoning by analogy.  It's illogical.  Yet it's quintessentially
human, and very common in natural language usage.


.i lo xance lo jamfu cu simsa lo ka kakne co cadzu fi ce'u kei .e ja'e
bo lo ka makau xe cilre co cadzu fi ce'u
Hands and feet are similar in that one can walk on them, and therefore
similar in what way one can learn to walk on them.

I think you're using {co} incorrectly.  The place structure of {broda co
brode} is that of {broda}, not that of {brode}.  Using {ni'i} instead of
{ja'e}, I might express your example as:

   .i lo xance lo jamfu cu simsa lo ka kakne cadzu fi ce'u kei
   .e ni'i bo lo ka cilre be fu makau cadzu fi ce'u

Or, using {co}:

   .i lo xance lo jamfu cu simsa lo ka cadzu co kakne fi ce'u kei
   .e ni'i bo lo ka cadzu co cilre be fu makau fi ce'u


No, I'm using {co} correctly. It's just a tricky beast.
{.i ko'a ... broda co brode fo'a ...} has the sumti {ko'a ...} filling places of {broda} starting at x1, and has the sumti {fo'a ...} filling the places of {brode} starting at x2. The upshot is that one can use {co} to elide abstractors using tanru. 

.i mi djica co cliva ti

I forget the exact CLL reference, but this is indeed in the CLL.


   lo najnimre lo plise cu simsa lo ka farvi bu'u lo tricu kei .e ja'e
bo lo ka makau tadji co kurji

I think you mean {be}, not {bu'u}:

   lo najnimre lo plise cu simsa lo ka farvi be lo tricu kei .e
   ja'e ja seni'i bo lo ka makau kurji tadji


Sure

.i mi'e la tsani mu'o

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lojban/562e4432-9329-49e3-42d8-cce39c4f27fe%40mail.jerrington.me.