Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10 5/3/83; site umcp-cs.UUCP
Path: utzoo!linus!security!genrad!decvax!harpo!seismo!rlgvax!cvl!umcp-cs!dr_who
From: dr_who@umcp-cs.UUCP
Newsgroups: net.philosophy
Subject: contextual knowledge
Message-ID: <1578@umcp-cs.UUCP>
Date: Sun, 7-Aug-83 02:58:13 EDT
Article-I.D.: umcp-cs.1578
Posted: Sun Aug  7 02:58:13 1983
Date-Received: Sun, 7-Aug-83 13:17:26 EDT
Organization: Univ. of Maryland, Computer Science Dept.
Lines: 17

Response to Tom Craver on Larry Cipriani's objection:

Although I'm not sure Tom and I really disagree on this, I do sympathize
with Larry's claim that nothing is gained by adding context to knowledge.
Tom claims that it helps us to justify actions against the question, "are
you sure you're right?"  It is a complex question (because it involves the
question of what it means to call an action right), but I think that a
contextual *definition* of knowledge can't help us justify actions.  I think
that an action is right if it was done for the right goals, and the agent
acted on the best evidence available to him about how to achieve the goals,
regardless of whether he *knew* what the best way to achieve the goals were.
We need not claim knowledge to claim justified action.

I still think that the role proposed for knowledge-defined-contextually can
be fulfilled adequately by the concept "best justified belief."

--Paul Torek, U of MD College Park