On 19 May 2011 03:35, Michael Turniansky <
mturniansky@gmail.com> wrote:
>> On 17 May 2011 13:11, Michael Turniansky <
mturniansky@gmail.com> wrote:
>> > I say you CAN say "is-flying-teapot" of a nothing.
>>
>> "Nothing is a flying teapot"? I take notice of how it's convertible
>> into an _expression_ that says of something: "Everything is not a flying
>> teapot". The same conversion is possible in Lojban:
>>
>> no da broda --> ro da na broda
>>
>
> Umm... actually, no, not under the CLL. (although xorxes disagrees
> with their definition of bridi negation). The following transforms are
> true:
> no da broda --> ro da na'e broda
> ro da na broda -> su'o da na'e broda
"no da broda" has the implicit bridi affirmer, "ja'a"; why would this
binary NA become a non-binary (scalar) NAhE rather than the binary
opposite ("na")? Consider the following non-metaphoric pair:
no tanjo cu [ja'a] glare --> ro tanjo cu na'e glare
The first one is commonsensically true: no tangent is literally hot.
In the second one, "na'e glare" is best understood to point to some
value on the scale between the two obvious extremes, "glare" and
"lenku", but excluding values on "glare", i.e. to mean some non-glare
temperature. It doesn't make sense that every tangent has such a
property, at all. Every tangent has no temperature; na glare, na
lenku. We shouldn't mix up NA with NAhE. And "no/ro" are most
correlative with "ja'a/na" of NA.
>> Besides, I wonder if the form of "no da broda" is as much common as
>> the form of "ro da na broda" among natural languages. Spanish "nada"
>> and French "rien" can each mean "anything" rather than "nothing"
>> depending on the verb's negativity.
>>
>> No veo nada.
>> Je ne vois rien.
>> ("I do not see anything." rather than "I see nothing".)
>>
>> Should these be translated as "mi viska no da" or "mi na viska ro da"?
>
> Either "mi viska no da" or "mi na viska su'o da". "mi na viska ro da"
> means it's not true that you see EVERYTHING, there are some things you
> don't see.
As far as what the natlang sentences literally mean, it is the case
that "it's not true that I see everything". "anything" in a negative
_expression_ is an idiomatic term for "everything"; both correspond to
"ro da". "mi na viska ro da" may be a literal translation of "I do not
see anything". If "mi" isn't completely blind, however, there must be
some things which "mi" can see. The problem is the use of unrestricted
"anything/everything" on the natlang part. The same with "nothing"; "I
see nothing" doesn't necessarily mean that there is nothing which I
can see at all.
My point, anyway, was that "nothing is/does ..." (saying of nothing)
may not be more basic and universal a form than "everything/anything
is/does ..." (saying of something). Every _expression_ with "nothing"
seems to be convertible into an _expression_ with "everything"; and when
"everything" is not entirely accurate as in "I do not see everything
(while not completely blind)", the same inaccuracy can be found in "I
see nothing (while not completely blind)". So I wonder if any
statement with "nothing" has any truth independent of its "everything"
counterpart. If it isn't so independent, we can't really say of
nothing in its own right (although we can talk about the concept of
the *set* which has nothing, i.e. the set which doesn't have
everything).
>> > But that page is not canon. That's xorxes' proposed extension of the
>> > grammar.
>>
>> What canonical or more-acceptable-than-xorlo sources support your
>> argument for "lo no broda"?
>
> The zasni gerna cenba vreji page is NOT xorlo. They are only xorxes'
> additional proposed expansions to the grammar.
I know. I meant the xorlo definition of "lo broda", to which I was
comparing your "lo no broda". You implied that the basis for "nonai"
is unsatisfactory because it's not canonical, so I wondered whether
your own reasoning for "lo no broda" could be considered any more
satisfactory by the same standard you implicitly invoked.
>> To the extent that "za'uno" too can mean non-integers like "pimu",
>> though, it too would have to be subjected to the said semantic
>> restriction when used with "mei". So "za'uno" wouldn't be functionally
>> different from "nonai".
>>
> a) you weren't restricting nonai use with mei. But again, saying something
> should mean less than does when used in certain ways is falling into natlang
> traps that we should be trying to avoid.
ro lo ki'o gerku cu ro roi cmoni -- Does the latter "ro" mean as many
as the former?
so'i lo ki'o gerku cu batci so'i prenu -- Does the latter "so'i" mean
as many as the former?
These inexact PA are context-dependent -- the number is determined
compositionally. Would be the same for "nonai".
Other examples of dynamic meaning include "le ninmu na ninmu". This
particular sumti from "ninmu" means less than what the brivla's
definition says it means, due to "le". And "le" doesn't so much
specify the meaning for which "ninmu" is changed, whereas "mei" may
specify the kind of numbers the preceding PA may be interpreted as. So
I don't see why "nonai mei" would be any more undesirable than "le
ninmu" in how the modified word means less than what it otherwise
does.
> b) This may be opening up another can of worms, but I'm not convinced
> that pimu shouldn't be allowed with mei. If I am speaking of a number of
> oranges, why can't I have a half of one? And from that universe of one half
> orange, I can talk about a smaller amount. pipa lo pimu najnimre cu fusra
"A half of one" states that "one" is cardinal to "a half", i.e. you
can't have "a half" without "one" of which to leave the other half. As
far as "mei" is concerned, "pipa", "pimu", "cipiso", etc. are
derivatives from elementary cardinal numbers (0, 1, 2, 3...). And
there is at least one logical reason why we shouldn't use such
derivatives as an inner quantifier, as a proper cardinality value of
something. We must be wary of the fallacy of division:
http://en.wikipedia.org/wiki/Fallacy_of_division
Consider this example:
lo pimu vinji
Can we safely say a 1/2 'airplane' is an airplane with vinji2 and
vinji3 (i.e. a full-fledged vinji1)? A part of something doesn't
necessarily inherit all properties of that whole. If lo broda lacks
certain properties that define lo brode, lo broda may not be
identified as lo brode, any more than it can be *associated* with lo
brode. We can avoid this issue by not messing up the cardinality &
reference of "lo vinji" itself and instead outer-quantifying it:
pimu lo vinji
pipa lo pimu lo vinji
Alternatively:
lo xadba be lo vinji
pipa lo xadba be lo vinji
(Note that "lo xadba be ...", unlike "pimu lo ...", can be pamei,
remei, cimei, etc. in its own right.)
>> In the context of cardinality, "greater than zero" basically means "at
>> least one", and "su'o" would thereby be neater than "za'uno". But I
>> wanted to avoid that line of positive expressions, because earlier
>> comments (especially by xorxes) suggested that "su'o" may not be
>> considered a default inner quantifier for "lo broda".
>>
>
> But we aren't talking about defaults. We are talking about
> explicitnesses... weren't we?
> So if you believe it has to be at least one, su'o would do fine (although it
>still won't inherently restrict it to integers)
We are talking about whether or not it's possible for "lo broda" to
have "no" as its inner quantifier (to be nomei). That possibility
isn't acknowledged by xorlo, according to which the referent of "lo
broda" (the x1 of selbri in general) is to be considered more than
nothing. This entails that any inner quantifier of "lo broda" be other
than "no" by default. There is this certain default sense that the
inner PA can't be "no", and that mathematically points to "za'uno" or
"su'o". Should either of these be the default inner quantifier, then?
I'm not sure, insofar as I respect the xorlo-proposer's opinion.
mu'o
--
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@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/lojban?hl=en.
--
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@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/lojban?hl=en.