Differenes between “ka” and “nu”

[Leo Beltran]

I’m now trying to learn abstractions (not as easy as I thought) an I can’t see the differences between “mi gleki lo nu tavla do” and “mi gleki lo ka tavla do.” For me they are the same. Do you agree?

[Gleki Arxokuna]

Very often {ka} is translated using infinitives in English:

1. {mi gleki lo ka tavla do} = I am glad to talk to you.

Compare:

2. {mi gleki lo nu tavla do} = I am glad that someone is talking to you (maybe that I talk to you but that isn't asserted).

We can rephrase the first example:

3. {mi gleki lo nu mi tavla do} = I am glad that I talk to you.

Obviously, the first and the third examples are almost the same in meaning.

2015-07-08 20:42 GMT+03:00 Leo Beltran <leobe...@mail.com>:

I’m now trying to learn abstractions (not as easy as I thought) an I can’t see the differences between “mi gleki lo nu tavla do” and “mi gleki lo ka tavla do.” For me they are the same. Do you agree?

--
You received this message because you are subscribed to the Google Groups "Lojban Beginners" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban-beginne...@googlegroups.com.
To post to this group, send email to lojban-b...@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban-beginners.
For more options, visit https://groups.google.com/d/optout.

[Leo Beltran]

I can see the differences now ^_^

[Erik Natanael Gustafsson]

I have myself a few difficulties understanding abstractions on a deeper level so I will ask some follow up questions:

Is ka used generally and nu for specific events? Can I rephrase the translations like this:

1. {mi gleki lo ka tavla do} = I get glad when talking to you (regardless of whether I am doing it right now).

2. {mi gleki lo nu tavla do} = I am glad because of this single event of talking to you.

or am I way off?

[selpa'i]

la'o me. Erik Natanael Gustafsson .me cusku di'e

> Is ka used generally and nu for specific events? Can I rephrase the
> translations like this:
>
> 1. {mi gleki lo ka tavla do} = I get glad when talking to you
> (regardless of whether I am doing it right now).
> 2. {mi gleki lo nu tavla do} = I am glad because of this single event of
> talking to you.

What you put in a sumti place generally doesn't affect what tense the
main bridi is in. {mi gleki} means I'm actually happy, no matter what
other sumti appear in the sentence (as long as no irrealis attitudinals
or {da'i} appears).

There aren't many places where more than one abstractor can be used
sensibly. Usually any given place is defined (or implicitly understood
by regular users of the language) to have exactly one type of sumti that
can be put in it. The idea that more than one abstractor type can be
used, or that an abstraction and a concrete sumti can both go in a place
(sometimes referred to as "polymorphism") did not find many supporters.

Even though I am one of the people who originally suggested {djica lo
ka}, I no longer use it, and now use {djica lo nu} and {.aidji lo ka}.
To me {djica} means "x1 wants x2 to happen", and a property cannot happen.

Similar things can be said about {gleki lo ka}. If {gleki} means "x1 is
happy that x2 is the case or that x2 happens" then {gleki lo ka}
automatically becomes questionable.

One reason why {lo ka} even crops up in these places is that people try
to avoid having to repeating sumti, the {ka} is supposed to auto-insert
the sumti for them, but this causes problems when using {ka} in places
where you don't want the {ce'u} to be "filled", where you don't want the
{ka}-property to be applied to any outside sumti. {nelci} is such an
example. {mi nelci lo ka limna} could mean "I like the property of
swimming", "I like swimmingness", without any indication or claim that
the speaker is swimming. The aforementioned practice of forcing {lo ka}
to supply its {ce'u} with an outside sumti removes this reading and
replaces it with "I like to swim", "I like it when I swim", but for a
big price. Moreover, unless you want {lo ka} to be ambiguous or
context-dependent, you need to define for each brivla how it interacts
with {lo ka} vs how it interacts with {lo nu} etc. It is easier and
cleaner for a language like Lojban to split the two meanings into two
separate brivla, one that uses {nu} and one that uses {ka}.

Now, your question:

> Is ka used generally and nu for specific events?

No. The difference lies elsewhere. {ka} is related to {du'u}, but has a
{ce'u} place, making it an incomplete bridi, while {nu} is a physical
manifestation (actual or hypothetical) of a bridi. {ka} and {du'u} are
not physical.

mi'e la selpa'i mu'o

[guskant]

I think la selpa'i has perfectly responded to the question, but I have an objection about the English interpretation of {ka}, and a complement to the explanation about relation of {ka} and other abstractors.

Le lundi 13 juillet 2015 21:45:58 UTC, selpa'i a écrit :

One reason why {lo ka} even crops up in these places is that people try
to avoid having to repeating sumti, the {ka} is supposed to auto-insert
the sumti for them, but this causes problems when using {ka} in places
where you don't want the {ce'u} to be "filled", where you don't want the
{ka}-property to be applied to any outside sumti. {nelci} is such an
example. {mi nelci lo ka limna} could mean "I like the property of
swimming", "I like swimmingness", without any indication or claim that
the speaker is swimming.

{mi nelci lo ka limna} means rather "I like property of swimmer", not "swimming". {ka} (property) signifies a special case of {si'o} (idea, concept) in Lojban.

Both {lo si'o broda} and {lo ka broda} signify « signified » (meaning) of predicate {broda}. The difference is that a sentence in {ka}-clause has only one sumti place blank (denoted with {ce'u}), while a sentence in {si'o}-clause has all places blank. {si'o}-{ka}-{du'u} forms full set of predicate abstraction.

lo si'o (ce'u) limna (ce'u) : 

idea, concept of swimming ; idea, concept that is signified by a predicate "X swims in fluid Y."

lo ka (ce'u) limna (zo'e) : 

property of swimmer ; property that is signified by a predicate "X swims in some fluid," "X has property of swimmer in some fluid."

lo ka (zo'e) limna ce'u : 

property of swimming fluid ; property that is signified by a predicate "someone swims in fluid Y," "Y has property of fluid in which someone swims."

lo du'u (zo'e) limna (zo'e) : 

« signified » of a predicate "someone swims in some fluid."

lo se du'u (zo'e) limna (zo'e) : 

« signifying » (string of symbols) of a predicate "someone swims in some fluid."

mu'o

[Gleki Arxokuna]

2015-07-14 19:09 GMT+03:00 guskant <gusni...@gmail.com>:

"I like the property of
swimming", "I like swimmingness"

that would rather be {mi nelci lo si'o ka limna}

[Erik Natanael Gustafsson]

Okay. I'm slowly grasping it. Is what is written in the CLL on this subject still what is used by Lojbanists? Thank you for your very complete answers, I'll read a few more times :)

mu'o

[Pierre Abbat]

I don't know how closely everyone follows CLL on this. I think that some
selbri that take abstractions in a place can take multiple kinds. For
instance, "djuno" is listed as taking "du'u", but I'd use it with others as
well:

mi djuno lo du'u la .fred. tumymre
I know that Fred is a surveyor.

mi djuno lo pu'u tumymre
I know how to survey.

mi djuno lo li'i tumymre .i trombikula .i sisku lo korba'a .i co'e
I know what it's like to survey: chiggers, looking for boundary markers, and
so forth.

Pierre
--
When a barnacle settles down, its brain disintegrates.
Já não percebe nada, já não percebe nada.

[selpa'i]

la .piier. cu cusku di'e

> On Thursday, July 16, 2015 03:53:59 Erik Natanael Gustafsson wrote:
>> Okay. I'm slowly grasping it. Is what is written in the CLL on this subject
>> still what is used by Lojbanists? Thank you for your very complete answers,
>> I'll read a few more times :)
>
> I don't know how closely everyone follows CLL on this. I think that some
> selbri that take abstractions in a place can take multiple kinds. For
> instance, "djuno" is listed as taking "du'u", but I'd use it with others as
> well:
>
> mi djuno lo du'u la .fred. tumymre
> I know that Fred is a surveyor.
>
> mi djuno lo pu'u tumymre
> I know how to survey.
>
> mi djuno lo li'i tumymre .i trombikula .i sisku lo korba'a .i co'e
> I know what it's like to survey: chiggers, looking for boundary markers, and
> so forth.

The last two are raising. {mi djuno lo pu'u broda} does not mean "I know
how to broda" but "I know the process of broda" (which sounds fine in
*English* but isn't in Lojban. Really it means "I know that the process
of broda is true" and is similar to {mi djuno la .djan.}).

The correct way is {mi djuno lo du'u ma kau pu'u broda} or {mi djuno lo
du'u ma kau tadji lo ka broda}. The same goes for {li'i}, it should be
{mi djuno lo du'u ma kau li'i broda}.

[sphen...@gmail.com]

 It is easier and 
cleaner for a language like Lojban to split the two meanings into two 
separate brivla, one that uses {nu} and one that uses {ka}.

Yes, I have seen {kaidji} and made {kaitcu} as well, but I'm not sure this is a sustainable route; we'd then need new brivla for every brivla where a {ka} belonging to x1 is useful as a substitution of {nu}, such as {gleki}, {nelci}, etc...

[selpa'i]

la zipcpi cu cusku di'e

> It is easier and
> cleaner for a language like Lojban to split the two meanings into two
> separate brivla, one that uses {nu} and one that uses {ka}.
>
> Yes, I have seen {kaidji} and made {kaitcu} as well,

(I already use nitcu2 as a property)

> but I'm not sure
> this is a sustainable route; we'd then need new brivla for every brivla
> where a {ka} belonging to x1 is useful as a substitution of {nu}, such
> as {gleki}, {nelci}, etc...

Would we really, though? Is it necessary to have ka-variants of
everything? Would you, hypothetically, use {mi sruma lo ka ce'u bilma}
for "I assume [myself] to be sick"? Or what about {mi kanpe lo ka ce'u
ba jinga}? Would it be too much to say {mi kanpe lo nu mi ba jinga}?

The {ce'u} actually doesn't save us that much trouble compared to using
one of the usual back-referencing mechanisms. Logically speaking, the
reason why {ka} is used in places like nitcu2 or troci2 is not in order
to not have to repeat the x1, but because it avoids sumti raising and
makes the predicates much easier to interpret and define. It is
primarily a semantic concern, not one of convenience; convenience is
only a lucky by-product of it.

[Andrew]

.u'i
I thought I understood the difference between {ka} and {nu}. Now I'm not so sure.

Abstractions are complicated.

[la gleki]

On Friday, July 24, 2015 at 12:30:26 AM UTC+3, selpa'i wrote:

la zipcpi cu cusku di'e
>       It is easier and
>     cleaner for a language like Lojban to split the two meanings into two
>     separate brivla, one that uses {nu} and one that uses {ka}.
>
> Yes, I have seen {kaidji} and made {kaitcu} as well,

(I already use nitcu2 as a property)

> but I'm not sure
> this is a sustainable route; we'd then need new brivla for every brivla
> where a {ka} belonging to x1 is useful as a substitution of {nu}, such
> as {gleki}, {nelci}, etc...

Would we really, though? Is it necessary to have ka-variants of
everything? Would you, hypothetically, use {mi sruma lo ka ce'u bilma}
for "I assume [myself] to be sick"? Or what about {mi kanpe lo ka ce'u
ba jinga}? Would it be too much to say {mi kanpe lo nu mi ba jinga}?

Yes, given that {mi} can be a long noun with nouns inside, and the clause with {kanpe} can be inside another clause so not {ri}, not {vo'a}. 

And {lo no'a cu}, a lengthy construct would be the only safe option (compare short English infinitive "to").

The {ce'u} actually doesn't save us that much trouble compared to using
one of the usual back-referencing mechanisms. Logically speaking, the
reason why {ka} is used in places like nitcu2 or troci2 is not in order
to not have to repeat the x1, but because it avoids sumti raising and
makes the predicates much easier to interpret and define.

So that's it. We need definitions that define these relations.

For others: such definitions now exist only in one dictionary. 

Up to now such usage was intuitive and/or the one users got used to.

It is
primarily a semantic concern, not one of convenience; convenience is
only a lucky by-product of it.

It is also a concern of mastering the language, which this thread was started with.