Re: "Friendly Premises"


dchris@xxxxxxxxx wrote:
>Acme Diagnostics wrote:
>> "Dan Christensen" <dchris@xxxxxxxxxxxxx> wrote:
>>><geraldrm@xxxxxxxxxxxxx> wrote in message
>>>> premises" -- who discusses their use, what is the correct terminology,
>>>> etc. I've read a great deal in books on logic for the layman but
>>>> haven't come across anything that corresponds to this issue.
>>>>
>>>> By "friendly premises" I mean premises that people posit (often
>>>> unstated) that are necessary to a theory they are building on
>>>> circumstantial evidence. That is, they examine circumstantial
>>>> evidence, construct a theory that pulls the circumstantial evidence
>>>> together plausibly, but the proposal demands certain factual premises.
>>>
>>>[snip]
>>>
>>>Just my two cents, but maybe you are confusing logic or mathematics with
>>>science. Logicians and mathematicians explore questions of "what if,"
>>>constructing useful or at least interesting theoretical systems of numbers,
>>>points, algebraic operators, etc. on paper.
>>
>> Ok, but your two cents worth seems most confused.

Sorry for that wording, btw. "Incorrect" would have been much
better.
>
>Quite possible.
>
>
>> That is not
>> what logic is. You are mainly describing mathematics here.
>>
>[snip]
>
>Does it matter? Assumptions or premises about physical systems
>("friendly" or otherwise) can be stated in the language of logic and
>mathematics. It is not, however, the job of logic or mathematics to
>verify such assumptions, only to draw to reasonable conclusions from
>them. That was my point.

I can't really be responsive because some of my relevant
comments appear to be missing. I don't think I can follow
the exchange well enough.

But I'd like to post two examples. I don't know if they relate to
that, perhaps they do, but I think they might at least be
helpful in some way to the original poster:

Case #1

Betty Hill claimed to have been abducted by a UFO. As the
story goes, sometime later under hypnosis she drew a star map
from the perspective of the supposed home planet. Then, as
the story goes, somebody put the map on the computer. Close,
but no cigar. One of the stars she drew wasn't there, though the
others matched up. But then, as the story continues, a few years
later, that star was discovered.

Big load of crap, of course. But let's just suppose that star
map, hypnosis, etc., and computer work really could be verified
to be legit, and they certainly could be if someone went to the
trouble. (No, nobody has gone to the trouble! <g>) Ok, so now you
have facts, i.e. premises.

You apply logic to the facts. The new star is predicted, but what
does that mean? Only logic can tell you what it means. The
abduction is proved, but not by scientific verification. That
only provided source facts, i.e. true premises for the argument.
In this case, you only have a conclusion, but no scientific
verification, for one of the most important discoveries in
history. That's one reason logic is important in these real-world
theories. One way of putting this is that a theory achieves
"logical significance." I.e. logic tells you that it is
nearly impossible not to be true - at least as reliable and
probably more so than a scientific verification.

Most people wouldn't believe it. They would require scientific
verification. But a practicing logician would believe it. They'd
have no choice because experience requires them to accept that
such simple logic always works, turns out to be true, can risk
your life on it (emergency workers and military do it, double-ace
fighter pilots and there you are talking about some very complex
logic in the sense of reasoning), etc. To get the best
information, then, you could legitimately ask in a logic group.

Case #2

Let's say there is a kook posting in a newsgroup (not here). Over
time, you learn things about this kook, including some personal
history and current real life circumstances. As you learn things,
you write down observations in the form of rough psychiatric
speculations and any trivia whether it seems relevant or not.

Then one day, you get an email to check out a
certain site which describes a particular psychiatric disorder.
Bingo! Immediately symptoms match up. Since you already wrote
down these symptoms, they were predictive, which is an absolute
requirement. When do you have a positive diagnosis? Well, you
need to use coincidental probability for that (or some similar
probability alternative). Also it would help if you were a
qualified psychiatrist (!), but this is just an example.

Probability is a branch of logic, i.e. the inductive branch.
Applied math for quantification, and then there is a lot of
logic having to do with independence of premises (probably
the single most important thing), possible effects from outside
the system (or theory), etc. There are just all kinds of logic
issues in problems of this nature, i.e. anything having to do
with applied probability or statistics, and having nothing to do
with the underlying logic theory of probability itself.

Now having that diagnosis, and your list of various information,
all information about the disorder might be predictive. For
instance, 95% of people with this disorder have been
incarcerated. Now knowing that, a piece of information with a
well-known prison name becomes coincidental. Is it independent?
What logical relationships might exist? The two pieces of
information work together to both increase the probability of the
diagnosis and the probability of incarceration. In this way,
logic can be used to discover "secret premises," or premises that
are made more probable or even discovered (because you are caused
to look or look at them in a new context) only because the
"theory" is so good.

So, yes, according to this example, a theory can certainly bear
on the reliability of premises or supply new premises just
because the theory is "good." But it would help for the OP
to post a more detailed problem definition.

Ok, so most of this logic is specific to the application, and you
could say that it doesn't belong in logic proper, but in the
logic of whatever application. But suppose you are in the
business of solving these types of problems, i.e. expert at the
generalized logic that can affect any such theories or systems?
Then you would be mainly a logician ready to apply your logic
knowledge to a wide range of applications. But don't quote me on
that. <g>

Larry
.

Relevant Pages

  • Re: Lottery facts for Sherry
    ... regardless of the mathematics involved. ... This is the basic law of probability that I learned in grade school. ... In the lottery this law is used to formulate the permutation formula ... Mathematically the odds say it only has a 50/50 chance but the previous ...
    (rec.gambling.lottery)
  • Re: End-Point Paradox ( Infinity Paradoxes rise again
    ... Dear Hagman,, you just repeated the premises of the problem,, not ... there exists a real dilemma in this part of mathematics and has to be ... The correctness of those premises make CALCULUS a true science, ... that we "people on earth" send them TVs in any quantity or quality ...
    (sci.math)
  • Re: Endurantists vs Perdurantism
    ... >> You are saying something about an infinite regress of causation and that is ... derived from the premises we have defended. ... belief have a relatively high probability of being true. ... Compatibilists think that it does not. ...
    (sci.logic)
  • Re: Einstein Unravelled; Effectuationism and Value
    ... >> So you accept some parts of what Einstein says and other depending on ... >> mathematics rests on no premises other than logic. ... the term 'conceptions' they do not rest on premises sacred or otherwise. ...
    (sci.physics.relativity)
  • Re: Quantum Gravity 293.0: Probable Causation/Influence (PI) in Prime Number Geometry/Factorizations
    ... Huang, Einstein had much insight, but I think that he underestimated ... but statisticians (who use probability as their main ... mathematics), but mostly without Probability or Statistics, as ... But devising a tool based on "existential indeterminacy" is very ...
    (sci.physics)
Quantcast