Re: Kinds of knowledge was (RE: An elegant distinction (was Re: brz, or Plan B revisited (LONG)))

On 9/26/05, Peter Bleackley <Peter.Bleackley@...> wrote:

> contains inconsistencies. Obviously, incomplete systems of logic are
> preferable to inconsistent ones.
>
> Pete
>
Usually true, but not obviously so :)

There are certain situations that we might not want a consistent
system.  The main drawback of consistent systems (All of them?  I
think so) is that they're "explosive" -- as soon as a contradiction is
reached, all sentences suddenly become both true and false
simultaneously by quod contradictione quodlibet.

This is bad if you're designing a guard robot, say, because then some
clever adventurer will just walk up to it, say, "This statement is
false", and walk on by while the robot begins to smoke. ;)

Humans, not being consistent reasoners, have no such trouble; if we
come to a contradiction in one part of our belief system, we can just
ignore it.  A contradiction in one area of our beliefs does not
necessarily destroy beliefs in other areas.  The damage is "localized"
in that sense.

Systems like this, in which contradictions do not "explode", are
called paraconsistent.  (I forget if there's more that's necessary to
be paraconsistent than just this.)  Plain old inconsistent systems?
Yeah, not so good.  But paraconsistent systems are apparently pretty
useful.

--
Patrick Littell
PHIL101: W 6:00-8:50
Voice Mail: ext 744
Fall 05 Office Hours: W 5:00-6:00, by appointment