[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] Re: Bayesian evidential?
There are already some cmavo that can be used for expressing the certainty of a claim:je'u - truly (tautological, e.g. all dogs are dogs)
ju'o - certainly (almost complete certainly)
la'a - probably
----cu'i maybe (where all of these cmavo suffixed with -cu'i essentially mean the same thing)
la'anai - probably not
ju'onai - certainly not
je'unai - falsely (statements which claim falsehoods, e.g. Germany borders China)
Although a system using a numeric certainty value would be interesting, it probably (see what I did there, ^^) wouldn't be used very much. However, if I were to agree with a selma'o, MAI would be it. You need a system that produces a free modifier, accepting a number *somewhere*. The advantage is of course that lojban numbers can be vague, where some of these numbers could correspond with the cmavo listed above.
On the other hand, making up experimental cmavo is very touchy, and I wouldn't recommend it under most circumstances, this one being one of such instances. The potential usefulness of such an experimental cmavo is definitely outweighed by the simple fact that adding new words for concepts that are generally expressible using some other method is a bad practice. Lojban needs some kind of stability, and adding cmavo left and right is not the way to do it. Personally, some cmavo are nice to have as hacks (ze'ei) and some are nice as shortcuts (zo'ei, ZOhOI, and ma'oi), but depending on to whom I'm talking, I tend to adapt my grammar, especially with new Lojbanists, whom I don't want to indoctrinate into unofficial grammars.
All in all, I advocate using [fi'o te kanpe] as suggested (isn't it ironic though that kanpe itself is an experimental ;) ) when it comes to indicating the certainty of a claim. However, should it arise to indicate the certainty of some specific component where using a TAG is not appropriate (these cases are rare; for all sumti, it is possible to use [pe TAG] and for all selbri [be TAG]) I would propose using one of the modifiers in my above list.
mu'o mi'e la tsani
On 30 January 2012 15:28, Michael Turniansky
<mturniansky@gmail.com> wrote:
I"d probably use fi'o te kanpe, as arpis mentioned, myself...
--gejyspa
On Mon, Jan 23, 2012 at 3:53 PM, D
<datapacrat@gmail.com> wrote:
On Jan 23, 11:50 am, Adam Lopresto <adamlopre...@gmail.com> wrote:
> On Mon, Jan 23, 2012 at 10:27 AM, D <
datapac...@gmail.com> wrote:
> > the whole idea is to
> > associate a numerical value with the sumti being tagged
> > Hm... {bei'e} comes reasonably close to containing most of the sounds
> > of "Bayes", and doesn't seem to be in use by any other experimental
> > cmavo.
> I'd suggest making it a MAI; the number goes in front of it, and the entire
> thing is a free-modifier.
That's so crazy that it just might work.
And since the evidentials themselves are free-modifiers, it should be
simple enough to pretend that {pabei'e} and its cousins, while built
as MAIs, should be placed the way evidentials are (ie, at the start of
an evidenced sentence, or after a specific evidenced word). It also
allows for a few entertaining constructions, such as {ma'ubei'e}
meaning "more than 50% confidence" or "preponderance of the evidence",
and {xobei'e} to ask how confident the listener is about something.
So, how does this sound as a proposal for an experimental cmavo?
{bei'e}
MAI
Built as a MAI, placed as an evidential. The number added to the front
is the decibels of logarithmic Bayesian probability the speaker
assigns to the word or phrase that the {bei'e} refers to. 0,
{nobei'e}, represents 50% confidence, higher numbers greater
confidence, lower numbers lesser confidence, as according to E.T.
Jaynes' standard description.
(There's probably some easier way to refer to how the decibels compare
to standard linear odds, but the best way I know of is simply to point
to a table, such as this one:)
decibels / Level of belief / Rough Odds / notes
-oo / 0% / 1:oo / complete disbelief, unachievable save for paradoxes
-6 / 20.0% / 1:4 /
-5 / 24.0% / 1:3 /
-4 / 28.5% / 2:5 / a reasonable doubt
-3 / 33.3% / 1:2 /
-2 / 38.7% / 2:3 / probable cause
-1 / 44.3% / 4:5 /
0 / 50.0% / 1:1 / neither belief nor disbelief; agnosticism
1 / 55.7% / 5:4 / preponderance of the evidence
2 / 61.3% / 3:2 /
3 / 66.6% / 2:1 / clear and convincing evidence
4 / 71.5% / 5:2 /
5 / 76.0% / 3:1 / beyond a reasonable doubt
6 / 80.0% / 4:1 /
7 / 83.3% / 5:1 /
8 / 86.3% / 6:1 /
9 / 88.8% / 8:1 /
10 / 90.9% / 10:1 / one nine
13 / 95.2% / 20:1 / lone studies with p=0.05
20 / 99.0% / 100:1 / two nines, lone studies with p=0.01
26 / 99.7% / 400:1 / confirmed studies with p=0.05
30 / 99.9% / 1,000:1 / three nines
40 / 99.99% / 10,000:1 / four nines, confirmed studies with p=0.01
42 / 99.993% / 16,000:1 / 4 standard deviations
50 / 99.999% / 100,000:1 / five nines
60 / 99.9999% / 1,000,000:1 / six nines
62 / 99.99994% / 1,500,000:1 / 5 standard deviations
87 / 99.9999998% / 500,000,000:1 / 6 standard deviations
116 / 99.9999999997% / 390,000,000,000:1 / 7 standard deviations
oo / 100% / oo:1 / complete certainty, unachievable save for
tautologies
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.