/LMAR=0/XLINE=3/FONT#0=BAXL30/FONT#1=BAXM30/FONT#2=BAXB30/FONT#3=BAXS30/FONT#4=GACS25/FONT#5=CLAR30/FONT#8=SAIL25



␈↓ ↓H␈↓␈↓ ¬Research Proposal Submitted to


␈↓ ↓H␈↓␈↓ ∧∨␈↓αTHE NATIONAL SCIENCE FOUNDATION


␈↓ ↓H␈↓α␈↓ ε=␈↓for


␈↓ ↓H␈↓␈↓ ∧D␈↓αBasic Research in Arti␈↓␈↓βc␈↓␈↓αcial Intelligence


␈↓ ↓H␈↓α␈↓ ε@␈↓by


␈↓ ↓H␈↓␈↓ ¬iJohn McCarthy
␈↓ ↓H␈↓␈↓ ¬	Professor of Computer Science
␈↓ ↓H␈↓␈↓ ¬CPrincipal Investigator








␈↓ ↓H␈↓␈↓ ¬yOctober 1980





␈↓ ↓H␈↓␈↓ ¬Computer Science Department

␈↓ ↓H␈↓␈↓ ¬∩␈↓αSTANFORD UNIVERSITY

␈↓ ↓H␈↓α␈↓ ¬M␈↓Stanford, California



␈↓ ↓H␈↓¬␈↓ ↓KResearch Proposal Submitted to the National Science Foundation





␈↓ ↓H␈↓Proposed Amount ␈↓&␈↓λ$423,225␈↓␈↓)αβ. Proposed Starting Date ␈↓&␈↓λ1 July 1981␈↓␈↓)αβ. Proposed Duration ␈↓&␈↓λ3 years␈↓␈↓)αβ.



␈↓ ↓H␈↓Title ␈↓&␈↓λBasic Research in Artificial Intelligence␈↓␈↓)αβ



␈↓ ↓H␈↓Principal Investigator ␈↓&␈↓λJohn McCarthy␈↓␈↓)αβ␈↓ ε8Submitting Institution ␈↓&␈↓λStanford University␈↓␈↓)αβ
␈↓ ↓H␈↓  Soc. Sec. No. ␈↓&␈↓λ558-30-4793␈↓␈↓)αβ␈↓ ε8Department ␈↓&␈↓λ Computer Science    ␈↓␈↓)αβ
␈↓ ↓H␈↓␈↓ ε8Address ␈↓&␈↓λStanford, California 94305␈↓␈↓)αβ



␈↓ ↓H␈↓Make grant to ␈↓&␈↓λBoard of Trustees of the Leland Stanford Junior University␈↓␈↓)αβ



␈↓ ↓H␈↓Endorsements:

␈↓ ↓H␈↓␈↓ αXPrincipal Investigator␈↓ ¬XDepartment Head␈↓ λHInstitutional Admin. O␈↓λ␈↓β@␈↓λ␈↓icial

␈↓ ↓H␈↓Name␈↓ αX␈↓&␈↓λJohn McCarthy        ␈↓␈↓)αβ␈↓ ¬X␈↓&␈↓λEdward A. Feigenbaum ␈↓␈↓)αβ␈↓ λX␈↓&␈↓λ                     ␈↓␈↓)αβ


␈↓ ↓H␈↓Signature␈↓ αX␈↓&␈↓λ                     ␈↓␈↓)αβ␈↓ ¬X␈↓&␈↓λ                     ␈↓␈↓)αβ␈↓ λX␈↓&␈↓λ                     ␈↓␈↓)αβ

␈↓ ↓H␈↓Title␈↓ αX␈↓&␈↓λProfessor            ␈↓␈↓)αβ␈↓ ¬X␈↓&␈↓λProfessor & Chairman ␈↓␈↓)αβ␈↓ λX␈↓&␈↓λ                     ␈↓␈↓)αβ

␈↓ ↓H␈↓Telephone␈↓ αX␈↓&␈↓λ(415) 497-4430       ␈↓␈↓)αβ␈↓ ¬X␈↓&␈↓λ(415) 497-4079       ␈↓␈↓)αβ␈↓ λX␈↓&␈↓λ                     ␈↓␈↓)αβ

␈↓ ↓H␈↓Date␈↓ αX␈↓&␈↓λ                     ␈↓␈↓)αβ␈↓ ¬X␈↓&␈↓λ                     ␈↓␈↓)αβ␈↓ λX␈↓&␈↓λ                     ␈↓␈↓)αβ
␈↓ ↓H␈↓α␈↓ εP␈↓ L1


␈↓ ↓H␈↓α␈↓ βGAbstract                                  ␈↓ εh␈↓the␈α⊃memory␈α⊃of␈α⊂a␈α⊃computer,␈α⊃and␈α⊃the␈α⊂heuristic
                                          ␈↓ εh␈↓part␈α∂concerns␈α∂procedures␈α∂for␈α∂deciding␈α∂what␈α∞to
␈↓ ↓H␈↓        This␈α
is␈α∞a␈α
request␈α
for␈α∞a␈α
grant␈α
of␈α∞xxx␈α
for   ␈↓ εh␈↓do␈α⊃on␈α⊃the␈α⊃basis␈α⊃of␈α⊃the␈α⊃necessary␈α⊃facts.␈α⊂ Most
␈↓ ↓H␈↓continued␈αsupport␈α
of␈αbasic␈αresearch␈α
in␈αarti␈↓βC␈↓cial␈↓ εh␈↓work␈α→in␈α~AI␈α→has␈α→concerned␈α~␈↓↓heuristics␈↓,␈α→and
␈↓ ↓H␈↓intelligence␈α⊂with␈α⊂emphasis␈α⊂on␈α⊂the␈α⊂structure␈α⊂of  ␈↓ εh␈↓computer␈α∩representations␈α⊃of␈α∩information␈α⊃have
␈↓ ↓H␈↓formal␈α∩reasoning,␈α∩epistemological␈α∪problems␈α∩of ␈↓ εh␈↓been␈α∩chosen␈α∩that␈α⊃are␈α∩capable␈α∩of␈α⊃representing
␈↓ ↓H␈↓arti␈↓βC␈↓cial␈α
intelligence␈α
and␈α
mathematical␈α
theory␈α
of␈↓ εh␈↓only␈α⊂a␈α∂part␈α⊂of␈α∂the␈α⊂information␈α∂that␈α⊂would␈α∂be
␈↓ ↓H␈↓computation.␈α≥ The␈α≥mathematical␈α≡theory␈α≥of      ␈↓ εh␈↓available␈αto␈αa␈αhuman.␈α The␈αmodes␈αof␈αreasoning
␈↓ ↓H␈↓computation␈α%supports␈α%the␈α%AI␈α&work␈α%by            ␈↓ εh␈↓used␈α~by␈α~present␈α~programs␈α~are␈α≠often␈α~even
␈↓ ↓H␈↓providing␈α∪tools␈α∀for␈α∪reasoning␈α∀about␈α∪complex    ␈↓ εh␈↓weaker than the representations themselves.
␈↓ ↓H␈↓strategies␈α∪and␈α∪showing␈α∩that␈α∪they␈α∪attain␈α∩their
␈↓ ↓H␈↓goals.␈α
 The␈α
line␈αof␈α
research␈α
being␈α
pursued␈αwas   ␈↓ εh␈↓        The␈α∞lines␈α∞of␈α∞research␈α∞we␈α∞plan␈α∞to␈α∞pursue
␈↓ ↓H␈↓already␈αoutlined␈αin␈αour␈α1977␈αproposal,␈αand␈αthis  ␈↓ εh␈↓are␈α$exempli␈↓↓␈↓βC␈↓↓␈↓ed␈α$in␈α$the␈α$attached␈α#papers
␈↓ ↓H␈↓proposal␈α∞incorporates␈α
much␈α∞material␈α∞from␈α
that  ␈↓ εh␈↓(McCarthy␈α#1979a,b,1980).␈α" Here␈α#we␈α"shall
␈↓ ↓H␈↓one.                                      ␈↓ εh␈↓explain␈α∞how␈α∞this␈α∞work␈α∞␈↓↓␈↓βC␈↓↓␈↓ts␈α∞together.␈α∞ Our␈α∞long
                                          ␈↓ εh␈↓range␈α∞goal␈α
is␈α∞a␈α
program␈α∞that␈α
can␈α∞be␈α∞told␈α
facts
␈↓ ↓H␈↓α␈↓ ↓⎇Epistemological Problems of Arti␈↓␈↓βc␈↓␈↓αcial     ␈↓ εh␈↓about␈αthe␈αworld␈αand␈αcan␈αuse␈αthem␈αe␈↓↓␈↓β@␈↓↓␈↓ectively␈αto
␈↓ ↓H␈↓ β1␈↓αIntelligence                              ␈↓ εh␈↓achieve␈αthe␈α
goals␈αit␈αis␈α
given.␈α Sometimes␈α
it␈αwill
                                          ␈↓ εh␈↓use␈α∂the␈α∂facts␈α∂directly␈α∂from␈α∂its␈α∂data␈α∂base␈α∞using
␈↓ ↓H␈↓        Arti␈↓βC␈↓cial␈α∞intelligence␈α∞has␈α∞proved␈α∞to␈α∞be␈α∞a␈↓ εh␈↓deductive␈α~and␈α~inductive␈α~processes␈α~like␈α~the
␈↓ ↓H␈↓di␈↓β@␈↓icult␈α∨branch␈α≡of␈α∨science.␈α∨ Some␈α≡people       ␈↓ εh␈↓deductive␈α∨processes␈α∨of␈α∨mathematical␈α≡logic.
␈↓ ↓H␈↓thought␈α
that␈α
human-level␈α
intelligence␈α∞could␈α
be ␈↓ εh␈↓However,␈α
we␈α
are␈α
already␈α
sure,␈α
(McCarthy␈α
1980),
␈↓ ↓H␈↓achieved␈α↔in␈α↔ten␈α_or␈α↔twenty␈α↔years,␈α_but␈α↔such        ␈↓ εh␈↓that␈α
conjectural␈α
processes␈α
will␈α
be␈α
needed␈α
that␈α
go
␈↓ ↓H␈↓estimates␈α∀were␈α∀based␈α∀on␈α∀the␈α∀di␈↓β@␈↓iculties␈α∀they    ␈↓ εh␈↓beyond␈α"deduction␈α!as␈α"presently␈α!conceived.
␈↓ ↓H␈↓could␈α≠see␈α≠when␈α≠they␈α≠made␈α≠the␈α≠optimistic         ␈↓ εh␈↓Sometimes␈α∀it␈α∀will␈α∀use␈α∀these␈α∀facts␈α∀to␈α∀compile
␈↓ ↓H␈↓predictions.␈α! Our␈α!opinion␈α!is␈α!that␈α!major        ␈↓ εh␈↓"expert␈α
programs"␈α
that␈α
use␈α
these␈α
facts␈α
in␈αa␈α
more
␈↓ ↓H␈↓scienti␈↓βC␈↓c␈α%discoveries␈α%are␈α%required␈α$before     ␈↓ εh␈↓e␈↓↓␈↓β@␈↓↓␈↓icient␈α∞way␈α∞than␈α∞simple␈α∂reasoning.␈α∞ However,
␈↓ ↓H␈↓human-level␈α&general␈α%intelligence␈α&can␈α%be       ␈↓ εh␈↓the␈α↔expert␈α↔programs␈α⊗will␈α↔have␈α↔to␈α↔defer␈α⊗to
␈↓ ↓H␈↓reached.␈α∂ Moreover,␈α⊂many␈α∂of␈α⊂these␈α∂discoveries  ␈↓ εh␈↓reasoning␈α∂when␈α⊂unexpected␈α∂use␈α∂of␈α⊂the␈α∂factual
␈↓ ↓H␈↓require␈α∃theoretical␈α⊗advances␈α∃and␈α⊗not␈α∃merely    ␈↓ εh␈↓data-base is required.
␈↓ ↓H␈↓extending␈α⊂current␈α⊂ideas␈α⊂for␈α⊂producing␈α∂"expert
␈↓ ↓H␈↓programs"␈α!to␈α new␈α!domains.␈α! The␈α recent          ␈↓ εh␈↓        We␈α∞do␈α∞not␈α
propose␈α∞to␈α∞implement␈α∞such␈α
a
␈↓ ↓H␈↓emphasis␈α!by␈α!ARPA␈α!and␈α!other␈α agencies            ␈↓ εh␈↓program␈α⊂immediately␈α⊃-␈α⊂maybe␈α⊂not␈α⊃within␈α⊂the
␈↓ ↓H␈↓sponsoring␈α-AI␈α-research␈α-on␈α,immediate           ␈↓ εh␈↓next␈α∃␈↓↓␈↓βC␈↓↓␈↓ve␈α∃years.␈α∃ This␈α∃is␈α∃because␈α∃its␈α∃success
␈↓ ↓H␈↓applications␈α
has␈α
resulted␈α
in␈α
a␈αserious␈α
imbalance.␈↓ εh␈↓depends␈α⊗on␈α∃successful␈α⊗formalization␈α⊗of␈α∃facts
␈↓ ↓H␈↓Deciding␈α⊂what␈α⊂the␈α⊂basic␈α⊂issues␈α⊂are␈α⊂is␈α∂di␈↓β@␈↓icult   ␈↓ εh␈↓about␈α⊂the␈α⊂world.␈α⊃ We␈α⊂have␈α⊂made␈α⊃progress␈α⊂in
␈↓ ↓H␈↓enough␈αwithout␈αhaving␈αto␈αformulate␈αeverything  ␈↓ εh␈↓this␈α
formalization,␈α
but␈α
we␈α
may␈α
be␈α
occupied␈α
with
␈↓ ↓H␈↓in␈α_terms␈α_of␈α_demonstration␈α_programs␈α→to␈α_be        ␈↓ εh␈↓it␈α
exclusively␈αfor␈α
the␈αforseeable␈α
future.␈α In␈α
short
␈↓ ↓H␈↓available␈α∂in␈α∂two␈α⊂years.␈α∂ While␈α∂there␈α⊂has␈α∂been    ␈↓ εh␈↓we␈α,will␈α+emphasize␈α,theoretical␈α+arti␈↓↓␈↓βC␈↓↓␈↓cial
␈↓ ↓H␈↓some␈α∩recent␈α⊃increase␈α∩in␈α⊃understanding␈α∩of␈α⊃the    ␈↓ εh␈↓intelligence␈α⊂except␈α⊂for␈α∂some␈α⊂work␈α⊂with␈α∂proof-
␈↓ ↓H␈↓need␈α⊃for␈α⊃basic␈α⊃research,␈α⊃especially␈α⊃since␈α⊃1978, ␈↓ εh␈↓checkers.
␈↓ ↓H␈↓the main interest is still on short term goals.
                                          ␈↓ εh␈↓        The␈α$main␈α#areas␈α$of␈α$our␈α#previous
␈↓ ↓H␈↓        Our␈α∂research␈α∂is␈α⊂based␈α∂on␈α∂the␈α⊂idea␈α∂that,  ␈↓ εh␈↓accomplishment␈α≡and␈α≥future␈α≡work␈α≡are␈α≥the
␈↓ ↓H␈↓for␈α∩many␈α∪purposes,␈α∩the␈α∩problems␈α∪of␈α∩arti␈↓βC␈↓cial    ␈↓ εh␈↓following (as now seen):
␈↓ ↓H␈↓intelligence␈α
can␈α
be␈α
separated␈α
into␈α
two␈α
parts␈α
-␈α
an
␈↓ ↓H␈↓epistemological␈α∞part␈α
and␈α∞a␈α
heuristic␈α∞part.␈α
 The ␈↓ εh␈↓1.␈αDevelopment␈αof␈α␈↓↓circumscription␈↓␈αas␈αa␈αmode␈αof
␈↓ ↓H␈↓␈↓↓epistemological␈↓␈α∃part␈α∃concerns␈α∃what␈α∃facts␈α∀and   ␈↓ εh␈↓reasoning␈α≡and␈α≡its␈α≡application␈α∨to␈α≡arti␈↓↓␈↓βC␈↓↓␈↓cial
␈↓ ↓H␈↓inference␈α_rules␈α_are␈α_available␈α_for␈α_solving␈α_a     ␈↓ εh␈↓intelligence.␈α∨ ␈↓↓Circumscription␈↓␈α formalizes␈α∨the
␈↓ ↓H␈↓problem␈α⊂and␈α⊂how␈α⊃they␈α⊂can␈α⊂be␈α⊃represented␈α⊂in       ␈↓ εh␈↓process␈α∞of␈α∂concluding␈α∞(often␈α∞incorrectly)␈α∂that␈α∞a
␈↓ ↓H␈↓α␈↓ εP␈↓ J2


␈↓ ↓H␈↓certain␈αcollection␈αof␈αfacts␈αis␈αall␈αthat␈αare␈αrelevant␈↓ εh␈↓given␈α∃opportunities␈α∃to␈α∃observe␈α∃and␈α∃compute
␈↓ ↓H␈↓for␈αsolving␈αa␈αproblem.␈α It␈αdoes␈αthis␈αby␈αallowing  ␈↓ εh␈↓and act.
␈↓ ↓H␈↓one␈αto␈α
formally␈αassume␈αthat␈α
the␈αentities␈αthat␈α
are
␈↓ ↓H␈↓generated␈α↔by␈α↔speci␈↓↓␈↓βC␈↓↓␈↓ed␈α⊗processes␈α↔are␈α↔all␈α⊗the     ␈↓ εh␈↓        Besides␈αthe␈αwork␈αon␈αarti␈↓↓␈↓βC␈↓↓␈↓cial␈αintelligence,
␈↓ ↓H␈↓entities␈αof␈αa␈αspeci␈↓↓␈↓βC␈↓↓␈↓ed␈αkind.␈α This␈αis␈α
common␈αin   ␈↓ εh␈↓McCarthy␈α∂has␈α⊂been␈α∂studying␈α⊂various␈α∂methods
␈↓ ↓H␈↓human␈α
reasoning␈αand,␈α
for␈αreasons␈α
described␈αin   ␈↓ εh␈↓for␈α∂proving␈α∂programs␈α∂correct.␈α∂ In␈α∂the␈α⊂last␈α∂few
␈↓ ↓H␈↓(McCarthy␈α∩1980),␈α∪cannot␈α∩be␈α∪accomplished␈α∩by     ␈↓ εh␈↓years,␈α∩an␈α∩approach␈α⊃called␈α∩Elephant␈α∩has␈α⊃been
␈↓ ↓H␈↓any form of deduction.                    ␈↓ εh␈↓developed␈α⊃that␈α⊃involves␈α⊃representing␈α⊃program
                                          ␈↓ εh␈↓variables␈α)as␈α)functions␈α)of␈α)time.␈α) This
␈↓ ↓H␈↓2.␈αTreating␈αconcepts␈αas␈αobjects.␈α This,␈αdescribed␈↓ εh␈↓complements␈α⊃the␈α⊃approach␈α⊃of␈α⊃(McCarthy␈α⊃and
␈↓ ↓H␈↓in␈α∩(McCarthy␈α∩1979a),␈α∩facilitates,␈α∩and␈α∪may␈α∩be    ␈↓ εh␈↓Cartwright 1979c).
␈↓ ↓H␈↓required␈αfor,␈αreasoning␈αabout␈αknowledge,␈αbelief,
␈↓ ↓H␈↓wants,␈α∪possibility␈α∪and␈α∪necessity.␈α∪ As␈α∪we␈α∩shall  ␈↓ εh␈↓α␈↓ εiRecent Advances and Changes in Direction of 
␈↓ ↓H␈↓explain␈αlater,␈αit␈αmay␈αbe␈αnecessary␈αto␈αreplace␈αthe ␈↓ εh␈↓ λK␈↓αthe Research 
␈↓ ↓H␈↓approach␈α⊃of␈α⊃that␈α⊃paper␈α⊃by␈α⊃one␈α⊃which␈α⊃makes
␈↓ ↓H␈↓distinctions␈α∀of␈α∪language␈α∀between␈α∀objects␈α∪and   ␈↓ εh␈↓        Since␈α_our␈α_1977␈α_proposal,␈α_interest␈α↔has
␈↓ ↓H␈↓and␈αconcepts␈αof␈αthem␈αonly␈αwhen␈αrequired.␈α Not    ␈↓ εh␈↓increased␈α→in␈α→non-monotonic␈α→reasoning.␈α_ We
␈↓ ↓H␈↓making␈α
the␈α
distinctions␈α
would␈α
be␈αaccomplished  ␈↓ εh␈↓organized␈α⊗a␈α↔mini-conference␈α⊗at␈α↔Stanford␈α⊗in
␈↓ ↓H␈↓by some kind of non-monotonic reasoning.  ␈↓ εh␈↓1978,␈αand␈αthat␈αled␈αto␈αa␈αspecial␈αissue␈αof␈α␈↓↓Arti␈↓␈↓βS␈↓␈↓↓cial
                                          ␈↓ εh␈↓↓Intelligence␈↓␈α
containing␈α∞updated␈α
versions␈α∞of␈α
the
␈↓ ↓H␈↓3.␈α≥The␈α≡current␈α≥biggest␈α≥gap␈α≡in␈α≥computer          ␈↓ εh␈↓papers␈α↔presented␈α↔at␈α↔the␈α↔conference␈α⊗together
␈↓ ↓H␈↓reasoning␈α→about␈α→the␈α→physical␈α→world␈α→is␈α→the       ␈↓ εh␈↓with␈αa␈αpaper␈αby␈αRaymond␈αReiter␈α(1980).␈α Drew
␈↓ ↓H␈↓complete␈α⊃lack␈α⊃of␈α⊃a␈α⊃system␈α⊃for␈α∩reasoning␈α⊃with     ␈↓ εh␈↓McDermott␈α⊃and␈α⊃Jon␈α⊂Doyle␈α⊃have␈α⊃attacked␈α⊂the
␈↓ ↓H␈↓partial␈α∞information␈α∞about␈α∞concurrent␈α∞processes.␈↓ εh␈↓problem␈α∞through␈α
what␈α∞they␈α∞call␈α
␈↓↓non-monotonic
␈↓ ↓H␈↓All␈α≠the␈α~current␈α≠problem␈α≠solving␈α~programs       ␈↓ εh␈↓↓logic␈↓␈α∀and␈α∀Raymond␈α∀Reiter␈α∀has␈α∀introduced␈α∪a
␈↓ ↓H␈↓assume␈α≥that␈α≤each␈α≥action␈α≤of␈α≥the␈α≤program          ␈↓ εh␈↓␈↓↓logic␈α∂of␈α∂defaults␈↓.␈α∂ Our␈α∂own␈α⊂approach␈α∂through
␈↓ ↓H␈↓produces␈α⊗a␈α⊗next␈α⊗state␈α⊗that␈α⊗depends␈α⊗on␈α⊗the        ␈↓ εh␈↓circumscription␈α∂has␈α∂been␈α∂further␈α∂developed␈α∂in
␈↓ ↓H␈↓current␈α⊃state,␈α⊃the␈α∩action,␈α⊃and␈α⊃sometimes␈α∩on␈α⊃a    ␈↓ εh␈↓the published version.
␈↓ ↓H␈↓random␈αvariable.␈α They␈αdon't␈αtake␈αinto␈αaccount
␈↓ ↓H␈↓continuous␈α↔processes␈α↔or␈α↔the␈α↔fact␈α↔that␈α↔other     ␈↓ εh␈↓        Participation␈α∃in␈α∃the␈α∃study␈α⊗on␈α∃arti␈↓↓␈↓βC␈↓↓␈↓cial
␈↓ ↓H␈↓actions␈α⊃and␈α⊃events␈α⊃may␈α⊃be␈α⊃taking␈α⊃place.␈α⊂ We      ␈↓ εh␈↓intelligence␈α
and␈α∞philosophy␈α
held␈α
in␈α∞1979-80␈α
at
␈↓ ↓H␈↓believe␈α∂that␈α∂circumscription␈α∂may␈α⊂be␈α∂important  ␈↓ εh␈↓the␈α∞Center␈α∞for␈α∞Advanced␈α∞Study␈α∞in␈α∞Behavioral
␈↓ ↓H␈↓in␈α⊃formalizing␈α∩what␈α⊃people␈α⊃know␈α∩about␈α⊃such      ␈↓ εh␈↓Sciences␈α∂has␈α∂produced␈α∞somewhat␈α∂of␈α∂change␈α∞in
␈↓ ↓H␈↓processes.␈α∪ Little␈α∪progress␈α∪has␈α∪been␈α∀made␈α∪by    ␈↓ εh␈↓viewpoint␈α
although␈αnot␈α
yet␈αto␈α
de␈↓↓␈↓βC␈↓↓␈↓nite␈αtechnical
␈↓ ↓H␈↓anyone on this problem in the last few years.␈↓ εh␈↓results.  Consider the following problem:

␈↓ ↓H␈↓4.␈α⊃We␈α⊂also␈α⊃plan␈α⊂some␈α⊃study␈α⊂of␈α⊃the␈α⊃theory␈α⊂of        ␈↓ εh␈↓        ␈↓↓Suppose␈α∞that␈α∞a␈α∞law␈α
making␈α∞it␈α∞a␈α∞crime␈α
to
␈↓ ↓H␈↓patterns,␈α⊗especially␈α↔higher␈α⊗order␈α↔patterns␈α⊗in  ␈↓ εh␈↓↓"attempt␈α↔to␈α↔bribe␈α↔a␈α↔public␈α↔o␈↓␈↓βP␈↓␈↓↓icial"␈α↔has␈α⊗been
␈↓ ↓H␈↓which␈α∞function␈α∞variables␈α
may␈α∞be␈α∞matched␈α
and     ␈↓ εh␈↓↓passed,␈α∂and␈α∂suppose␈α∂further␈α∂that␈α∂the␈α⊂law␈α∂has
␈↓ ↓H␈↓relations␈α⊃between␈α∩patterns␈α⊃-␈α⊃for␈α∩example,␈α⊃the   ␈↓ εh␈↓↓been␈α∃enforced␈α∀with␈α∃various␈α∃trials,␈α∀convictions
␈↓ ↓H␈↓relation␈α
between␈α
a␈α
pattern␈α
describing␈α
a␈α
type␈αof  ␈↓ εh␈↓↓and␈α⊃appeals␈α⊃for␈α∩20␈α⊃years␈α⊃before␈α∩the␈α⊃following
␈↓ ↓H␈↓three-dimensional␈αobject␈α
such␈αas␈α
a␈αperson␈α
or␈αa   ␈↓ εh␈↓↓questions arise.
␈↓ ↓H␈↓vehicle␈αand␈αits␈αpatterns␈αof␈αits␈αperception␈α-␈αsuch
␈↓ ↓H␈↓as␈α∪its␈α∪projection␈α∪on␈α∩a␈α∪retina␈α∪as␈α∪modi␈↓↓␈↓βC␈↓↓␈↓ed␈α∩by       ␈↓ εh␈↓↓        (1)␈α∂Person␈α∂A␈α∂attempts␈α∂to␈α∂bribe␈α∂B␈α∂to␈α∞help
␈↓ ↓H␈↓angle␈αof␈αvision,␈αlighting␈αand␈αocclusion␈αby␈αother ␈↓ εh␈↓↓him␈α∞get␈α
a␈α∞contract␈α
under␈α∞the␈α
impression␈α∞that␈α
B
␈↓ ↓H␈↓objects.                                  ␈↓ εh␈↓↓is␈α
a␈α
private␈α
consultant␈α
to␈α
a␈α
government␈αagency.
                                          ␈↓ εh␈↓↓In␈α
fact,␈α
B␈αis␈α
an␈α
o␈↓␈↓βP␈↓␈↓↓icial␈αof␈α
the␈α
agency.␈α
 A␈αo␈↓␈↓βP␈↓␈↓↓ers
␈↓ ↓H␈↓        In␈α↔all␈α⊗this␈α↔work,␈α⊗the␈α↔emphasis␈α↔is␈α⊗on     ␈↓ εh␈↓↓the␈α∞defense␈α∞that␈α∞since␈α∞he␈α∞didn't␈α∞know␈α∞B␈α∞was␈α
a
␈↓ ↓H␈↓representation␈α≡of␈α∨the␈α≡information␈α∨that␈α≡is      ␈↓ εh␈↓↓public␈αo␈↓␈↓βP␈↓␈↓↓icial,␈αhe␈αcouldn't␈αhave␈αbeen␈αattempting
␈↓ ↓H␈↓actually␈α∪available␈α∪to␈α∪a␈α∪person␈α∪or␈α∪robot␈α∪with     ␈↓ εh␈↓↓to bribe a public o␈↓␈↓βP␈↓␈↓↓icial.
␈↓ ↓H␈↓α␈↓ εP␈↓ J3


␈↓ ↓H␈↓↓        (2)␈α∪C␈α∀attempts␈α∪to␈α∪bribe␈α∀D,␈α∪but␈α∀D␈α∪had      ␈↓ εh␈↓presuming␈α≡the␈α≡existence␈α≡of␈α≡unicorns␈α≡and
␈↓ ↓H␈↓↓resigned␈α⊂his␈α⊂position␈α⊂as␈α⊂a␈α⊂public␈α⊃o␈↓␈↓βP␈↓␈↓↓icial␈α⊂before ␈↓ εh␈↓planned␈α⊂to␈α⊂resolve␈α⊂it.␈α⊂ The␈α⊂formalism␈α⊃of␈α⊂that
␈↓ ↓H␈↓↓the␈α⊂attempt␈α⊂was␈α⊂made.␈α⊂ The␈α⊂defense␈α⊃is␈α⊂o␈↓␈↓βP␈↓␈↓↓ered     ␈↓ εh␈↓paper␈α≡allowed␈α≡both␈α≡physical␈α∨entities␈α≡and
␈↓ ↓H␈↓↓that␈α∞attempting␈α∞to␈α∞bribe␈α∞someone␈α∞who␈α∞in␈α∂fact␈α∞is   ␈↓ εh␈↓concepts␈α∩of␈α⊃them␈α∩to␈α⊃be␈α∩individuals␈α⊃in␈α∩a␈α⊃␈↓↓␈↓βC␈↓↓␈↓rst
␈↓ ↓H␈↓↓not␈α∂a␈α∂public␈α∂o␈↓␈↓βP␈↓␈↓↓icial␈α∂isn't␈α∂attempting␈α∂to␈α⊂bribe␈α∂a  ␈↓ εh␈↓order␈α∃theory.␈α∃ All␈α∀statements␈α∃of␈α∃knowing␈α∀or
␈↓ ↓H␈↓↓public o␈↓␈↓βP␈↓␈↓↓icial.                           ␈↓ εh␈↓wanting␈α
or␈α
attempting␈α
take␈α
concepts␈α
of␈α
objects,
                                          ␈↓ εh␈↓and␈α≠the␈α≠␈↓↓de␈α≠re␈↓␈α≠and␈α≠␈↓↓de␈α≠dicto␈↓␈α≤versions␈α≠of
␈↓ ↓H␈↓↓        (3)␈αE␈α
lets␈αit␈α
be␈αknown␈α
around␈αtown␈αthat␈α
he ␈↓ εh␈↓"attempting␈α∪to␈α∪bribe␈α∪a␈α∪public␈α∪o␈↓↓␈↓β@␈↓↓␈↓icial"␈α∪would
␈↓ ↓H␈↓↓will␈αpay␈α$1,000␈αto␈αany␈αpublic␈αo␈↓␈↓βP␈↓␈↓↓icial␈αwho␈αcan␈α␈↓␈↓βS␈↓␈↓↓x   ␈↓ εh␈↓have␈α↔di␈↓↓␈↓β@␈↓↓␈↓erent␈α⊗forms,␈α↔and␈α⊗if␈α↔the␈α⊗legislature
␈↓ ↓H␈↓↓his␈αdrunk␈αdriving␈αconviction.␈α There␈αhappens␈αto ␈↓ εh␈↓wrote␈α∩the␈α⊃law␈α∩in␈α⊃the␈α∩language␈α∩of␈α⊃(McCarthy
␈↓ ↓H␈↓↓be␈α⊂no␈α⊂o␈↓␈↓βP␈↓␈↓↓icial␈α⊂who␈α⊃can␈α⊂␈↓␈↓βS␈↓␈↓↓x␈α⊂the␈α⊂conviction.␈α⊃ Is␈α⊂E     ␈↓ εh␈↓1979a),␈αthey␈αwould␈αautomatically␈αsay␈αone␈αor␈αthe
␈↓ ↓H␈↓↓guilty␈α∂of␈α∞attempting␈α∂to␈α∂bribe␈α∞a␈α∂public␈α∂o␈↓␈↓βP␈↓␈↓↓icial␈α∞if ␈↓ εh␈↓other␈α∀or␈α∃the␈α∀conjunction␈α∀or␈α∃the␈α∀disjunction.
␈↓ ↓H␈↓↓he␈αisn't␈αattempting␈αto␈αbribe␈αa␈αspeci␈↓␈↓βS␈↓␈↓↓c␈αperson?␈α Is␈↓ εh␈↓However,␈α⊂it␈α⊂now␈α⊂seems␈α⊂that␈α⊂philosophers␈α∂and
␈↓ ↓H␈↓↓it␈αrelevant␈αwhether␈αthere␈αis␈αa␈αperson␈αwho␈αcan␈αdo   ␈↓ εh␈↓lawyers␈α
invent␈α
new␈α
distinctions␈α
all␈α
the␈αtime,␈α
and
␈↓ ↓H␈↓↓what E wants done␈↓?                        ␈↓ εh␈↓we␈α∞certainly␈α
can't␈α∞revise␈α
the␈α∞foundation␈α∞of␈α
our
                                          ␈↓ εh␈↓language␈α⊃every␈α⊃time␈α⊃such␈α⊃a␈α∩distinction␈α⊃comes
␈↓ ↓H␈↓        Such␈α≠questions␈α≠are␈α≠familiar␈α≠to␈α≠both    ␈↓ εh␈↓along.␈α∃ Even␈α∃if␈α⊗we␈α∃could,␈α∃the␈α⊗question␈α∃still
␈↓ ↓H␈↓philosophers␈α≤and␈α≤lawyers.␈α≥ The␈α≤distinction    ␈↓ εh␈↓arises␈α⊂of␈α⊃what␈α⊂was␈α⊂the␈α⊃meaning␈α⊂of␈α⊃what␈α⊂was
␈↓ ↓H␈↓between␈α
(1)␈α
and␈α
(2)␈α
is␈αthe␈α
well␈α
known␈α
␈↓↓de␈α
re␈α-␈α
de     ␈↓ εh␈↓said before the distinction was made.
␈↓ ↓H␈↓↓dicto␈↓␈α≥distinction␈α≥between␈α≥interpretations␈α≥of
␈↓ ↓H␈↓"attempting␈α∪to␈α∪bribe␈α∪a␈α∪public␈α∀o␈↓↓␈↓β@␈↓↓␈↓icial".␈α∪ This   ␈↓ εh␈↓        While␈αit␈αis␈αpossible␈αthat␈αthe␈αformalism␈αof
␈↓ ↓H␈↓distinction␈α*and␈α)even␈α*more␈α)complicated         ␈↓ εh␈↓(McCarthy␈α∂1979a)␈α⊂could␈α∂be␈α∂extended␈α⊂to␈α∂cover
␈↓ ↓H␈↓distinctions␈α
have␈α
been␈αmade␈α
in␈α
court␈αdecisions. ␈↓ εh␈↓all␈α∩cases␈α∩of␈α∩interest,␈α∩this␈α∩now␈α∩seems␈α∩unlikely,
␈↓ ↓H␈↓A␈αfamous␈α
1879␈αcase␈αconcerns␈α
a␈αcow␈α
which␈αwas       ␈↓ εh␈↓and␈α↔it␈α↔is␈α↔interesting␈α↔to␈α↔explore␈α↔a␈α↔di␈↓↓␈↓β@␈↓↓␈↓erent
␈↓ ↓H␈↓bought␈α
on␈αthe␈α
basis␈α
that␈αthe␈α
seller␈α
thought␈αthe   ␈↓ εh␈↓approach.␈α≥ This␈α≤new␈α≥approach␈α≥has␈α≤other
␈↓ ↓H␈↓cow␈α∪was␈α∪infertile␈α∪and␈α∪the␈α∪buyer␈α∀thought␈α∪he       ␈↓ εh␈↓advantages␈α∩as␈α∩well.␈α∩ The␈α∩idea␈α∪is␈α∩complicated
␈↓ ↓H␈↓could␈α"make␈α!the␈α"cow␈α!breed.␈α" When␈α!he              ␈↓ εh␈↓and␈α∨not␈α≡well␈α∨developed,␈α≡but␈α∨here␈α∨is␈α≡a
␈↓ ↓H␈↓discovered␈α∞that␈α∞the␈α∞cow␈α∞was␈α∞actually␈α
pregnant,  ␈↓ εh␈↓preliminary version.
␈↓ ↓H␈↓the␈αseller␈α
refused␈αto␈αdeliver␈α
on␈αthe␈αgrounds␈α
that
␈↓ ↓H␈↓he␈α_had␈α_sold␈α_a␈α_barren␈α_cow.␈α_ The␈α_issue␈α_is           ␈↓ εh␈↓        We␈α∂still␈α∂use␈α∂a␈α∞␈↓↓␈↓βC␈↓↓␈↓rst␈α∂order␈α∂system,␈α∂but␈α∞we
␈↓ ↓H␈↓whether there was a meeting of minds.     ␈↓ εh␈↓have␈α⊃only␈α⊂one␈α⊃form␈α⊂of␈α⊃expression.␈α⊃ We␈α⊂have
                                          ␈↓ εh␈↓the␈α
default␈α
rule␈α
that␈α
equals␈α
may␈α
be␈αsubstituted
␈↓ ↓H␈↓        When␈αwe␈αtry␈αto␈αdesign␈αan␈αAI␈αsystem,␈αour   ␈↓ εh␈↓for␈αequals␈αunless␈αthere␈αis␈αa␈αreason␈αwhy␈αnot.␈α In
␈↓ ↓H␈↓interest␈α
in␈α
such␈αproblems␈α
is␈α
di␈↓↓␈↓β@␈↓↓␈↓erent␈α
from␈αthat  ␈↓ εh␈↓philosopher's␈αLatin,␈α␈↓↓Ceteris␈αparibus,␈αde␈αre␈α=␈αde
␈↓ ↓H␈↓expressed␈α∃by␈α∃either␈α∃philosophers␈α∃or␈α∀lawyers.   ␈↓ εh␈↓↓dicto␈↓.␈α≥ Our␈α≤hope␈α≥is␈α≤that␈α≥extending␈α≤non-
␈↓ ↓H␈↓They␈α
are␈αinterested␈α
in␈αmaking␈α
distinctions␈αthat ␈↓ εh␈↓monotonic␈α!reasoning␈α!to␈α!say␈α!that␈α!certain
␈↓ ↓H␈↓have␈α∞not␈α∞previously␈α∞been␈α∞made.␈α∂ The␈α∞primary     ␈↓ εh␈↓concepts␈α↔are␈α⊗not␈α↔ambiguous␈α⊗unless␈α↔there␈α⊗is
␈↓ ↓H␈↓AI␈α∂interest␈α⊂should␈α∂be␈α∂in␈α⊂the␈α∂state␈α∂of␈α⊂mind␈α∂of      ␈↓ εh␈↓reason␈α∪to␈α∩the␈α∪contrary␈α∩will␈α∪allow␈α∪AI␈α∩systems
␈↓ ↓H␈↓the␈α≥lawyer␈α≥or␈α≡judge␈α≥who␈α≥tries␈α≡cases␈α≥of           ␈↓ εh␈↓that behave more like people in that respect.
␈↓ ↓H␈↓attempting␈α
to␈α
bribe␈α∞a␈α
public␈α
o␈↓↓␈↓β@␈↓↓␈↓icial␈α∞for␈α
twenty
␈↓ ↓H␈↓years␈α∂without␈α∂ever␈α∂thinking␈α∂of␈α∂the␈α∂distinction. ␈↓ εh␈↓        The␈α↔philosophers␈α↔have␈α↔a␈α_slogan␈α↔that
␈↓ ↓H␈↓When␈α∪the␈α∪distinction␈α∪is␈α∪pointed␈α∪out,␈α∪he␈α∩can      ␈↓ εh␈↓doing␈α∞philosophy␈α∞should␈α∞not␈α∞depend␈α∞on␈α
doing
␈↓ ↓H␈↓understand␈αthat␈αthere␈αis␈αa␈α
distinction,␈αalthough␈↓ εh␈↓all␈α→science␈α→␈↓↓␈↓βC␈↓↓␈↓rst.␈α→ Thus␈α→understanding␈α→what
␈↓ ↓H␈↓he␈α
may␈α
not␈α
have␈α
any␈α
de␈↓↓␈↓βC␈↓↓␈↓nite␈α
way␈α
of␈α
resolving      ␈↓ εh␈↓people␈α∀mean␈α∃by␈α∀the␈α∃word␈α∀"␈↓↓␈↓βC␈↓↓␈↓sh"␈α∃should␈α∀not
␈↓ ↓H␈↓the ambiguity in the law.                 ␈↓ εh␈↓depend␈α~on␈α≠the␈α~philosopher␈α≠knowing␈α~what
                                          ␈↓ εh␈↓distinguishes␈α⊗␈↓↓␈↓βC␈↓↓␈↓sh␈α∃from␈α⊗other␈α⊗vertebrates.␈α∃ It
␈↓ ↓H␈↓        (McCarthy␈α∩1979a)␈α∩presents␈α∪a␈α∩formalism ␈↓ εh␈↓appears␈α∪that␈α∪we␈α∀need␈α∪another␈α∪slogan␈α∀to␈α∪the
␈↓ ↓H␈↓that␈α∂resolves␈α∂simple␈α∂forms␈α∂of␈α∂␈↓↓de␈α∂re␈α∂-␈α⊂de␈α∂dicto␈↓     ␈↓ εh␈↓e␈↓↓␈↓β@␈↓↓␈↓ect␈α∃that␈α∃doing␈α∃AI␈α∃should␈α∃not␈α∃depend␈α∃on
␈↓ ↓H␈↓distinction,␈α
and␈α
we␈αknew␈α
about␈α
the␈α
problem␈αof    ␈↓ εh␈↓doing␈α⊂all␈α⊂philosophy␈α⊂␈↓↓␈↓βC␈↓↓␈↓rst.␈α⊂ Thus␈α⊂it␈α⊂should␈α⊂be
␈↓ ↓H␈↓stating␈α∞that␈α∞John␈α∞is␈α∞seeking␈α∞a␈α∞unicorn␈α
without   ␈↓ εh␈↓possible␈αto␈αdesign␈αa␈αprogram␈αthat␈αcould␈αdiscuss
␈↓ ↓H␈↓α␈↓ εP␈↓ G4


␈↓ ↓H␈↓cases␈α∩of␈α∩attempting␈α∩to␈α∩bribe␈α∩a␈α∩public␈α∩o␈↓↓␈↓β@␈↓↓␈↓icial    ␈↓ εh␈↓rules␈α⊃of␈α⊂chess␈α⊃that␈α⊂black␈α⊃was␈α⊂in␈α⊃check.␈α⊂ This
␈↓ ↓H␈↓without␈α∃the␈α∃programmer␈α∃knowing␈α∃about␈α∀the       ␈↓ εh␈↓single␈α
use␈α
of␈α
semantic␈α
attachment␈α
saves␈αseveral
␈↓ ↓H␈↓ambiguous␈αcases.␈α
 The␈αprogram␈αhe␈α
creates,␈αlike  ␈↓ εh␈↓hundred steps over traditional formalizations.
␈↓ ↓H␈↓the␈α⊗programmer␈α⊗himself,␈α⊗should␈α⊗be␈α⊗able␈α⊗to
␈↓ ↓H␈↓understand␈α⊃the␈α⊃distinction␈α⊃when␈α⊃it␈α⊃is␈α⊂pointed   ␈↓ εh␈↓        Filman's␈α∂proof␈α∂is␈α∂still␈α∂much␈α⊂longer␈α∂than
␈↓ ↓H␈↓out␈α∀but␈α∀needn't␈α∪have␈α∀any␈α∀automatic␈α∀way␈α∪of        ␈↓ εh␈↓the␈α∞informal␈α∞proof,␈α∞showing␈α∞that␈α∞we␈α∞still␈α
don't
␈↓ ↓H␈↓resolving the doubtful cases.             ␈↓ εh␈↓fully␈α(understand␈α'how␈α(humans␈α'combine
                                          ␈↓ εh␈↓observation␈α∂with␈α∞deduction.␈α∂ Filman␈α∞is␈α∂now␈α∞at
␈↓ ↓H␈↓        We␈α have␈α some␈α examples␈α!of␈α using         ␈↓ εh␈↓the␈αUniversity␈α
of␈αIndiana,␈αbut␈α
we␈αare␈α
trying␈αto
␈↓ ↓H␈↓circumscription␈α→for␈α_this␈α→in␈α_a␈α→second␈α_order      ␈↓ εh␈↓develop␈α$a␈α#new␈α$approach␈α$to␈α#combining
␈↓ ↓H␈↓formalism.␈α⊂ Namely,␈α⊂a␈α⊂suitable␈α⊂circumscription␈↓ εh␈↓observation␈α≡and␈α≡deduction␈α∨before␈α≡starting
␈↓ ↓H␈↓formula␈α∂asserts␈α⊂that␈α∂a␈α∂function␈α⊂like␈α∂␈↓↓Telephone␈↓  ␈↓ εh␈↓another proof of this kind.
␈↓ ↓H␈↓of␈α(McCarthy␈α1979a)␈αis␈αextensional␈αunless␈αthere
␈↓ ↓H␈↓is␈α
a␈αreason␈α
why␈αnot.␈α
 However,␈αthis␈α
is␈α
may␈αnot     ␈↓ εh␈↓α␈↓ πuOrganization of the work 
␈↓ ↓H␈↓be␈α∪a␈α∪su␈↓↓␈↓β@␈↓↓␈↓iciently␈α∪general␈α∪way␈α∪of␈α∀treating␈α∪the
␈↓ ↓H␈↓problem.                                  ␈↓ εh␈↓        The␈α work␈α will␈α be␈α done␈α by␈α John
                                          ␈↓ εh␈↓McCarthy␈α→and␈α_interested␈α→graduate␈α_students.
␈↓ ↓H␈↓        McCarthy␈α↔and␈α↔interested␈α↔students␈α⊗will ␈↓ εh␈↓Graduate␈α∪students␈α∪help␈α∪with␈α∩implementations
␈↓ ↓H␈↓explore this problem in the next three years.␈↓ εh␈↓and␈α$pursue␈α$thesis␈α$research␈α%in␈α$arti␈↓↓␈↓βC␈↓↓␈↓cial
                                          ␈↓ εh␈↓intelligence␈α→(concentrating␈α→on␈α→epistemological
␈↓ ↓H␈↓        Much␈αof␈αthe␈αresearch␈αhas␈αinvolved␈αusing ␈↓ εh␈↓problems)␈α≥and␈α≥in␈α≥mathematical␈α≥theory␈α≥of
␈↓ ↓H␈↓our␈α≤interactive␈α≤proof-checker␈α≤and␈α≤theorem     ␈↓ εh␈↓computation.␈α⊂ The␈α∂group␈α⊂shares␈α⊂interests␈α∂with
␈↓ ↓H␈↓prover␈α∂for␈α∂␈↓↓␈↓βC␈↓↓␈↓rst␈α∂order␈α∂logic.␈α∂ Work␈α∂on␈α∂FOL␈α∞is      ␈↓ εh␈↓the␈α6separately␈α6supported␈α6groups␈α6in
␈↓ ↓H␈↓now␈α∪being␈α∪supported␈α∪under␈α∪a␈α∪di␈↓↓␈↓β@␈↓↓␈↓erent␈α∪NSF        ␈↓ εh␈↓mathematical␈α#theory␈α"of␈α#computation␈α"and
␈↓ ↓H␈↓grant.␈α⊗ Besides␈α∃FOL,␈α⊗Jussi␈α∃Ketonen␈α⊗is␈α∃now       ␈↓ εh␈↓theorem proving.
␈↓ ↓H␈↓developing␈α⊃from␈α⊃scratch␈α⊂a␈α⊃new␈α⊃proof␈α⊂checker
␈↓ ↓H␈↓caled␈α⊃EKL␈α∩incorporating␈α⊃the␈α⊃lessons␈α∩of␈α⊃FOL      ␈↓ εh␈↓α␈↓ λcFacilities 
␈↓ ↓H␈↓that␈α
will␈α
hopefully␈α
be␈α
be␈α
suited␈α
for␈α
checking␈α
the
␈↓ ↓H␈↓proofs␈α⊗of␈α⊗substantial␈α⊗mathematical␈α∃theorems.  ␈↓ εh␈↓        The␈α
project␈α
will␈α∞be␈α
part␈α
of␈α∞the␈α
Stanford
␈↓ ↓H␈↓Separate␈α⊗support␈α⊗is␈α⊗being␈α⊗solicited␈α⊗for␈α⊗that    ␈↓ εh␈↓University␈α≥Arti␈↓↓␈↓βC␈↓↓␈↓cial␈α≥Intelligence␈α≥Laboratory
␈↓ ↓H␈↓work,␈α∂but␈α∂McCarthy's␈α∂participation␈α∂in␈α∂helping  ␈↓ εh␈↓and␈α
will␈α
use␈α∞its␈α
computer␈α
facilities␈α∞which␈α
have
␈↓ ↓H␈↓plan and apply EKL is part of this proposal.␈↓ εh␈↓now␈α∃been␈α∃merged␈α∃with␈α∃the␈α∃facilities␈α⊗of␈α∃the
                                          ␈↓ εh␈↓Computer␈α∪Science␈α∩Department␈α∪for␈α∩accounting
␈↓ ↓H␈↓        The␈α use␈α of␈α proof-checkers␈α in␈α∨this      ␈↓ εh␈↓purposes.␈α
 The␈α
Laboratory␈α
is␈α
directed␈α∞by␈α
John
␈↓ ↓H␈↓research␈α≠is␈α≠exempli␈↓↓␈↓βC␈↓↓␈↓ed␈α≠by␈α≠Filman's␈α≠(1979)      ␈↓ εh␈↓McCarthy␈α⊂and␈α⊂has␈α∂been␈α⊂mainly␈α⊂supported␈α∂by
␈↓ ↓H␈↓thesis.␈α∀ Unlike␈α∪mathematical␈α∀reasoning,␈α∪much  ␈↓ εh␈↓ARPA␈α⊗in␈α∃the␈α⊗past,␈α∃but␈α⊗the␈α∃fraction␈α⊗of␈α∃its
␈↓ ↓H␈↓common␈α∀sense␈α∀reasoning␈α∀uses␈α∀observation␈α∀as     ␈↓ εh␈↓support provided by ARPA is diminishing.
␈↓ ↓H␈↓well␈α
as␈αexplicitly␈α
given␈αpremisses.␈α
 Filman␈αused
␈↓ ↓H␈↓FOL␈α⊃to␈α⊃show␈α⊂that␈α⊃the␈α⊃reasoning␈α⊃involved␈α⊂in       ␈↓ εh␈↓α␈↓ λ←Personnel 
␈↓ ↓H␈↓the␈α~solution␈α→of␈α~a␈α→hard␈α~retrospective␈α→chess
␈↓ ↓H␈↓problem␈α∂can␈α∂be␈α∂formalized␈α∂in␈α∂␈↓↓␈↓βC␈↓↓␈↓rst␈α⊂order␈α∂logic    ␈↓ εh␈↓α␈↓ πYBiography of John McCarthy 
␈↓ ↓H␈↓supplemented␈α
by␈α∞the␈α
ability␈α∞to␈α
observe␈α∞a␈α
chess
␈↓ ↓H␈↓board.␈α∞ the␈α∂semantic␈α∞attachment␈α∂mechanism␈α∞of   ␈↓ εh␈↓BORN:  September 4, 1927 in Boston,
␈↓ ↓H␈↓FOL␈α∪was␈α∪used␈α∪to␈α∩build␈α∪a␈α∪simulation␈α∪of␈α∩his         ␈↓ εh␈↓␈↓ π(Massachusetts
␈↓ ↓H␈↓chess␈α⊂world.␈α⊃ He␈α⊂could␈α⊃then␈α⊂use␈α⊃the␈α⊂semantic
␈↓ ↓H␈↓simpli␈↓↓␈↓βC␈↓↓␈↓cation␈α∞routines␈α∞of␈α∞FOL␈α∞to␈α∞answer␈α∞(in␈α∞a    ␈↓ εh␈↓EDUCATION:
␈↓ ↓H␈↓single␈α
step)␈α
questions␈α
like␈α
"is␈α
the␈α
black␈α∞king␈α
in ␈↓ εh␈↓B.S.  (Mathematics) California Institute of
␈↓ ↓H␈↓check␈α∩on␈α∩board␈α∩B"␈α∩by␈α∩looking␈α∩at␈α∪the␈α∩model         ␈↓ εh␈↓␈↓ π(Technology, 1948.
␈↓ ↓H␈↓rather␈α⊂than␈α⊂deducing␈α∂from␈α⊂axioms␈α⊂giving␈α∂the     ␈↓ εh␈↓Ph.D. (Mathematics) Princeton University,
␈↓ ↓H␈↓positions␈α∪of␈α∩the␈α∪pieces␈α∩and␈α∪others␈α∪about␈α∩the     ␈↓ εh␈↓␈↓ π(1951.
␈↓ ↓H␈↓α␈↓ εP␈↓ I5


␈↓ ↓H␈↓HONORS AND SOCIETIES:                     ␈↓ εh␈↓[3]  (with S. Boilen, E. Fredkin, J.C.R.
␈↓ ↓H␈↓American Mathematical Society,            ␈↓ εh␈↓␈↓ π(Licklider) "A Time-Sharing Debugging
␈↓ ↓H␈↓Association for Computing Machinery,      ␈↓ εh␈↓␈↓ π(System for a Small Computer", ␈↓↓Proc.
␈↓ ↓H␈↓Sigma Xi,                                 ␈↓ εh␈↓↓␈↓ π(AFIPS Conf.␈↓ (SJCC), Vol. 23, 1963.
␈↓ ↓H␈↓Sloan Fellow in Physical Science (1957-59),␈↓ εh␈↓[4]  (with F. Corbato, M. Daggett) "The
␈↓ ↓H␈↓ACM National Lecturer (1961),             ␈↓ εh␈↓␈↓ π(Linking Segment Subprogram Language
␈↓ ↓H␈↓Sigma Xi National Lecturer (1978),        ␈↓ εh␈↓␈↓ π(and Linking Loader Programming
␈↓ ↓H␈↓IEEE,                                     ␈↓ εh␈↓␈↓ π(Languages", ␈↓↓Comm. ACM␈↓, July 1963.
␈↓ ↓H␈↓A.M. Turing Award from Association for    ␈↓ εh␈↓[5]  "Problems in the Theory of Computation",
␈↓ ↓H␈↓␈↓ αλComputing Machinery (1971).               ␈↓ εh␈↓␈↓ π(␈↓↓Proc. IFIP Congress 1965␈↓.
                                          ␈↓ εh␈↓[6]  "Time-Sharing Computer Systems", in W.
␈↓ ↓H␈↓PROFESSIONAL EXPERIENCE:                  ␈↓ εh␈↓␈↓ π(Orr (ed.), ␈↓↓Conversational Computers␈↓, Wiley,
␈↓ ↓H␈↓Proctor Fellow, Princeton University (1950-51),␈↓ εh␈↓␈↓ π(1966.
␈↓ ↓H␈↓Higgins Research Instructor in Mathematics,␈↓ εh␈↓[7]  "A Formal Description of a Subset of
␈↓ ↓H␈↓␈↓ αλPrinceton University (1951-53),           ␈↓ εh␈↓␈↓ π(Algol", in T.  Steele (ed.), ␈↓↓Formal
␈↓ ↓H␈↓Acting Assistant Professor of Mathematics,␈↓ εh␈↓↓␈↓ π(Language Description Languages for
␈↓ ↓H␈↓␈↓ αλStanford University (1953-55),            ␈↓ εh␈↓↓␈↓ π(Computer Programming␈↓, North-Holland,
␈↓ ↓H␈↓Assistant Professor of Mathematics, Dartmouth␈↓ εh␈↓␈↓ π(Amsterdam, 1966.
␈↓ ↓H␈↓␈↓ αλCollege (1955-58),                        ␈↓ εh␈↓[8]  "Information", ␈↓↓Scienti␈↓␈↓βS␈↓␈↓↓c American␈↓,
␈↓ ↓H␈↓Assistant Professor of Communication Science,␈↓ εh␈↓␈↓ π(September 1966.
␈↓ ↓H␈↓␈↓ αλM.I.T. (1958-61),                         ␈↓ εh␈↓[9]  "Computer Control of a Hand and Eye", in
␈↓ ↓H␈↓Associate Professor of Communication Science,␈↓ εh␈↓␈↓ π(␈↓↓Proc.  Third All-Union Conference on
␈↓ ↓H␈↓␈↓ αλM.I.T. (1961-62),                         ␈↓ εh␈↓↓␈↓ π(Automatic Control (Technical Cybernetics)␈↓,
␈↓ ↓H␈↓Professor of Computer Science Stanford    ␈↓ εh␈↓␈↓ π(Nauka, Moscow, 1967 (Russian).
␈↓ ↓H␈↓␈↓ αλUniversity (1962 - present).              ␈↓ εh␈↓[10]  (with D. Brian, G. Feldman, and J. Allen)
                                          ␈↓ εh␈↓␈↓ π("THOR ␈↓↓␈↓βE␈↓↓␈↓ A Display Based Time-
␈↓ ↓H␈↓PROFESSIONAL RESPONSIBILITIES             ␈↓ εh␈↓␈↓ π(Sharing System", ␈↓↓Proc. AFIPS Conf.␈↓
␈↓ ↓H␈↓␈↓ αλAND SCIENTIFIC INTERESTS:                 ␈↓ εh␈↓␈↓ π((FJCC), Vol.  30, Thompson, Washington,
␈↓ ↓H␈↓With Marvin Minsky organized and directed ␈↓ εh␈↓␈↓ π(D.C., 1967.
␈↓ ↓H␈↓␈↓ αλthe Arti␈↓↓␈↓βC␈↓↓␈↓cial Intelligence Project at M.I.T.␈↓ εh␈↓[11]  (with James Painter) "Correctness of a
␈↓ ↓H␈↓Organized and directs Stanford Arti␈↓↓␈↓βC␈↓↓␈↓cial  ␈↓ εh␈↓␈↓ π(Compiler for Arithmetic Expressions",
␈↓ ↓H␈↓␈↓ αλIntelligence Laboratory.                  ␈↓ εh␈↓␈↓ π(Amer. Math. Soc., ␈↓↓Proc. Symposia in
␈↓ ↓H␈↓Originated the LISP programming system for␈↓ εh␈↓↓␈↓ π(Applied Math., Math. Aspects of
␈↓ ↓H␈↓␈↓ αλcomputing with symbolic expressions,      ␈↓ εh␈↓↓␈↓ π(Computer Science␈↓, New York, 1967.
␈↓ ↓H␈↓␈↓ αλparticipated in the development of the    ␈↓ εh␈↓[12]  "Programs with Common Sense", in
␈↓ ↓H␈↓␈↓ αλALGOL 58 and the ALGOL 60                 ␈↓ εh␈↓␈↓ π(Marvin Minsky (ed.), ␈↓↓Semantic Information
␈↓ ↓H␈↓␈↓ αλlanguages.                                ␈↓ εh␈↓↓␈↓ π(Processing␈↓, MIT Press, Cambridge, 1968.
␈↓ ↓H␈↓Present scienti␈↓↓␈↓βC␈↓↓␈↓c work is in the ␈↓↓␈↓βC␈↓↓␈↓elds of ␈↓ εh␈↓[13]  (with Lester Earnest, D. Raj. Reddy,
␈↓ ↓H␈↓␈↓ αλArti␈↓↓␈↓βC␈↓↓␈↓cial Intelligence, Computation with  ␈↓ εh␈↓␈↓ π(Pierre Vicens) "A Computer with Hands,
␈↓ ↓H␈↓␈↓ αλSymbolic Expressions, Mathematical        ␈↓ εh␈↓␈↓ π(Eyes, and Ears", ␈↓↓Proc. AFIPS Conf.␈↓
␈↓ ↓H␈↓␈↓ αλTheory of Computation, Time-Sharing       ␈↓ εh␈↓␈↓ π((FJCC), 1968.
␈↓ ↓H␈↓␈↓ αλcomputer systems.                         ␈↓ εh␈↓[14]  (with Patrick Hayes) "Some Philosophical
␈↓ ↓H␈↓SELECTED PUBLICATIONS:                    ␈↓ εh␈↓␈↓ π(Problems from the Standpoint of Arti␈↓↓␈↓βC␈↓↓␈↓cial
␈↓ ↓H␈↓[1]  "Towards a Mathematical Theory of    ␈↓ εh␈↓␈↓ π(Intelligence", in Donald Michie (ed.),
␈↓ ↓H␈↓␈↓ αλComputation", in ␈↓↓Proc. IFIP Congress 62␈↓,  ␈↓ εh␈↓␈↓ π(␈↓↓Machine Intelligence 4␈↓, American Elsevier,
␈↓ ↓H␈↓␈↓ αλNorth-Holland, Amsterdam, 1963.           ␈↓ εh␈↓␈↓ π(New York, 1969.
␈↓ ↓H␈↓[2]  "A Basis for a Mathematical Theory of␈↓ εh␈↓[15]  "The Home Information Terminal", ␈↓↓Man
␈↓ ↓H␈↓␈↓ αλComputation", in P. Bra␈↓↓␈↓β@␈↓↓␈↓ort and D.        ␈↓ εh␈↓↓␈↓ π(and Computer, Proc. Int. Conf.,
␈↓ ↓H␈↓␈↓ αλHershberg (eds.), ␈↓↓Computer Programming    ␈↓ εh␈↓↓␈↓ π(Bordeaux, 1970␈↓, S. Karger, New York,
␈↓ ↓H␈↓↓␈↓ αλand Formal Systems␈↓, North-Holland,        ␈↓ εh␈↓␈↓ π(1972.
␈↓ ↓H␈↓␈↓ αλAmsterdam, 1963.
␈↓ ↓H␈↓α␈↓ εP␈↓ I6


␈↓ ↓H␈↓[16]  "Mechanical Servants for Mankind," in␈↓ εh␈↓␈↓ π(for Philosophy and Computer Technology,
␈↓ ↓H␈↓␈↓ αλ␈↓↓Britannica Yearbook of Science and the    ␈↓ εh␈↓␈↓ π(New York, 1977.
␈↓ ↓H␈↓↓␈↓ αλFuture␈↓, 1973.                             ␈↓ εh␈↓[27]  "Epistemological Problems of Arti␈↓↓␈↓βC␈↓↓␈↓cial
␈↓ ↓H␈↓[17]  Book Review: "Arti␈↓↓␈↓βC␈↓↓␈↓cial Intelligence: A␈↓ εh␈↓␈↓ π(Intelligence", ␈↓↓Proceedings of the Fifth
␈↓ ↓H␈↓␈↓ αλGeneral Survey" by Sir James Lighthill, in␈↓ εh␈↓↓␈↓ π(International Joint Conference on Arti␈↓␈↓βS␈↓␈↓↓cial
␈↓ ↓H␈↓␈↓ αλ␈↓↓Arti␈↓␈↓βS␈↓␈↓↓cial Intelligence, Vol. 5, No. 3␈↓, Fall␈↓ εh␈↓↓␈↓ π(Intelligence␈↓, M.I.T., Cambridge, 1977.
␈↓ ↓H␈↓␈↓ αλ1974.                                     ␈↓ εh␈↓(Additional McCarthy papers are listed in the
␈↓ ↓H␈↓[18]  "Modeling Our Minds" in ␈↓↓Science Year␈↓ εh␈↓␈↓ π(references at the end of this proposal).
␈↓ ↓H␈↓↓␈↓ αλ1975␈↓, The World Book Science Annual,
␈↓ ↓H␈↓␈↓ αλField Enterprises Educational Corporation,
␈↓ ↓H␈↓␈↓ αλChicago, 1974.
␈↓ ↓H␈↓[19]  "The Home Information Terminal,"
␈↓ ↓H␈↓␈↓ αλinvited presentation, AAAS Annual
␈↓ ↓H␈↓␈↓ αλMeeting, Feb. 18-24, 1976, Boston.
␈↓ ↓H␈↓[20]  "An Unreasonable Book," a review of
␈↓ ↓H␈↓␈↓ αλ␈↓↓Computer Power and Human Reason␈↓, by
␈↓ ↓H␈↓␈↓ αλJoseph Weizenbaum (W.H. Freeman and
␈↓ ↓H␈↓␈↓ αλCo., San Francisco, 1976) in SIGART
␈↓ ↓H␈↓␈↓ αλNewsletter
␈↓ ↓H␈↓558, June 1976, also in ␈↓↓Creative Computing␈↓,
␈↓ ↓H␈↓␈↓ αλChestnut Hill, Massachusetts, 1976 and in
␈↓ ↓H␈↓␈↓ αλ"Three Reviews of J. Weizenbaum's
␈↓ ↓H␈↓␈↓ αλ␈↓↓Computer Power and Human Reason␈↓, (with
␈↓ ↓H␈↓␈↓ αλB. Buchanan and J. Lederberg), Stanford
␈↓ ↓H␈↓␈↓ αλArti␈↓↓␈↓βC␈↓↓␈↓cial Intelligence Laboratory Memo
␈↓ ↓H␈↓␈↓ αλ291, Computer Science Department,
␈↓ ↓H␈↓␈↓ αλStanford, November 1976.
␈↓ ↓H␈↓[21]  Review: ␈↓↓Computer Power and Human
␈↓ ↓H␈↓↓␈↓ αλReason␈↓, by Joseph Weizenbaum (W.H.
␈↓ ↓H␈↓␈↓ αλFreeman and Co., San Francisco, 1976) in
␈↓ ↓H␈↓␈↓ αλPhysics Today, 1977.
␈↓ ↓H␈↓[22]  "The Home Information Terminal" to
␈↓ ↓H␈↓␈↓ αλappear in The Grolier Encyclopedia, 1977,
␈↓ ↓H␈↓␈↓ αλalso to appear in ␈↓↓The International
␈↓ ↓H␈↓↓␈↓ αλYearBook and Statemen's Who's Who␈↓,
␈↓ ↓H␈↓␈↓ αλSurrey, England, 1977.
␈↓ ↓H␈↓[23]  "Dialnet and Home Computers" (with Les
␈↓ ↓H␈↓␈↓ αλEarnest), ␈↓↓Proceedings of the First West
␈↓ ↓H␈↓↓␈↓ αλCoast Computer Faire and Conference␈↓, San
␈↓ ↓H␈↓␈↓ αλFrancisco, April 1977.
␈↓ ↓H␈↓[24]  "On The Model Theory of Knowledge"
␈↓ ↓H␈↓␈↓ αλ(with M. Sato, S. Igarashi, and T.
␈↓ ↓H␈↓␈↓ αλHayashi), ␈↓↓Proceedings of the Fifth
␈↓ ↓H␈↓↓␈↓ αλInternational Joint Conference on Arti␈↓␈↓βS␈↓␈↓↓cial
␈↓ ↓H␈↓↓␈↓ αλIntelligence␈↓, M.I.T, Cambridge, 1977.
␈↓ ↓H␈↓[25]  "Another SAMEFRINGE", in SIGART
␈↓ ↓H␈↓␈↓ αλNewsletter No. 61, February 1977.
␈↓ ↓H␈↓[26]  "Ascribing Mental Qualities to Machines"
␈↓ ↓H␈↓␈↓ αλto appear in ␈↓↓Essays in Philosophy and
␈↓ ↓H␈↓↓␈↓ αλComputer Technology␈↓, National Symposium
␈↓ ↓H␈↓α␈↓ εP␈↓ J7


␈↓ ↓H␈↓α␈↓ ε!Budget 

␈↓ ↓H␈↓∧PERIOD COVERED: 3 Years: 1 June 1978 through 31 December 1981.

␈↓ ↓H␈↓∧Dates:                     6/1/78-5/31/79   6/1/79-5/31/80   6/1/80-5/31/81

␈↓ ↓H␈↓∧                                   Person-           Person-          Person-
␈↓ ↓H␈↓∧A. SALARIES AND WAGES              months            months           months

␈↓ ↓H␈↓∧   1. Senior Personnel:

␈↓ ↓H␈↓∧     a. John McCarthy,    24,257.    6.5    27,007.    6.5   29,168.    6.5
␈↓ ↓H␈↓∧       Professor
␈↓ ↓H␈↓∧       Summer 75%(2 mos.)
␈↓ ↓H␈↓∧       Acad. Yr. 50%

␈↓ ↓H␈↓∧   2. Other Personnel


␈↓ ↓H␈↓∧     a. Student Research 
␈↓ ↓H␈↓∧        Assistants
␈↓ ↓H␈↓∧        (50% Acad.Yr.;
␈↓ ↓H␈↓∧        100% Summer)

␈↓ ↓H␈↓∧       (1)                 7,155.    7.5     7,704.    7.5    8,320.    7.5

␈↓ ↓H␈↓∧       (2)                 7,155.    7.5     7,704.    7.5    8,320.    7.5

␈↓ ↓H␈↓∧     b. Support Personnel

␈↓ ↓H␈↓∧       (1) Sec'y (20%)     2,092.    2.4     2,259.    2.4    2,440.    2.4

␈↓ ↓H␈↓∧       (2) Sys.Prog.(15%)  2,937.    1.8     3,172.    1.8    3,426.    1.8
␈↓ ↓H␈↓∧                          _______           _______          _______

␈↓ ↓H␈↓∧   Total Salaries & Wages 43,596.           47,846.          51,674.

␈↓ ↓H␈↓∧B. STAFF BENEFITS         
␈↓ ↓H␈↓∧   9/1/77-8/31/78:19.0%          
␈↓ ↓H␈↓∧   9/1/78-8/31/79:20.3%
␈↓ ↓H␈↓∧   9/1/79-8/31/80;21.6%                                          
␈↓ ↓H␈↓∧   9/1/80-8/31/81;22.4%
␈↓ ↓H␈↓∧                           8,708.           10,179.          11,472.
␈↓ ↓H␈↓∧                          _______          ________         ________
␈↓ ↓H␈↓∧C. TOTAL SALARIES, WAGES,
␈↓ ↓H␈↓∧   AND STAFF BENEFITS     52,304.           58,025.          63,146.
␈↓ ↓H␈↓α␈↓ εP␈↓ I8


␈↓ ↓H␈↓∧D. PERMANENT EQUIPMENT     5,000                --               --
␈↓ ↓H␈↓∧   (2 Datamedia terminals)

␈↓ ↓H␈↓∧E. EXPENDABLE SUPPLIES     1,632.            1,730.           1,834.
␈↓ ↓H␈↓∧   & EQUIPMENT(e.g.,copying,
␈↓ ↓H␈↓∧   office supplies, postage,      
␈↓ ↓H␈↓∧   freight, consulting,
␈↓ ↓H␈↓∧   honoraria)

␈↓ ↓H␈↓∧F. TRAVEL                  1,840.            1,950.           2,067.
␈↓ ↓H␈↓∧   All Domestic Travel

␈↓ ↓H␈↓∧G. PUBLICATIONS            1,000.            1,060.           1,124.

␈↓ ↓H␈↓∧H. OTHER COSTS             6,640.            7,038.           7,460.
␈↓ ↓H␈↓∧   1.Communication  1,600.
␈↓ ↓H␈↓∧     (telephone)     
␈↓ ↓H␈↓∧   2. Computer      5,040.
␈↓ ↓H␈↓∧      Equip. Maint.
␈↓ ↓H␈↓∧                          _______          ________          _______  

␈↓ ↓H␈↓∧I. TOTAL COSTS             68,416.          69,803.          75,631.
␈↓ ↓H␈↓∧    (A through H)

␈↓ ↓H␈↓∧J. INDIRECT COSTS:58% of   36,781.          40,486.          43,866.
␈↓ ↓H␈↓∧   A through H, less D.   ________         ________         ________     


␈↓ ↓H␈↓∧K. TOTAL COSTS            105,197.         110,289.         119,497.

␈↓ ↓H␈↓∧L. THREE YEAR TOTAL                                         334,983.
␈↓ ↓H␈↓α␈↓ εP␈↓ J9


␈↓ ↓H␈↓α␈↓ β7References                                ␈↓ εh␈↓␈↓αMoore, Robert C.␈↓ (1980) Reasoning about
                                          ␈↓ εh␈↓␈↓ π_Knowledge and Action. SRI  Arti␈↓α␈↓βC␈↓α␈↓cial
␈↓ ↓H␈↓␈↓αBoyer, R.S., and Moore, J.S.␈↓ (1975) Proving␈↓ εh␈↓␈↓ π_Intelligence Center, Technical Note 191, SRI
␈↓ ↓H␈↓␈↓ ↓xTheorems About LISP Functions, JACM       ␈↓ εh␈↓␈↓ π_International, Menlo Park, California.
␈↓ ↓H␈↓␈↓ ↓xVol 22. No. 1 pp. 129-144. New York:
␈↓ ↓H␈↓␈↓ ↓xACM.                                      ␈↓ εh␈↓␈↓αSato, M.␈↓ (1977) A study of Kripke-type Models
                                          ␈↓ εh␈↓␈↓ π_for some Model Logics by Gentzen's
␈↓ ↓H␈↓␈↓αFilman, R.E.␈↓ (1979) The Interaction of    ␈↓ εh␈↓␈↓ π_Sequential Method, (to appear in Publ.
␈↓ ↓H␈↓␈↓ ↓xObservation and Inference.                ␈↓ εh␈↓␈↓ π_R.I.M.S., Kyoto University).
␈↓ ↓H␈↓␈↓ ↓x(Stanford AI Memo AIM-327) Stanford
␈↓ ↓H␈↓␈↓ ↓xUniversity.                               ␈↓ εh␈↓␈↓αWagner, Todd J.␈↓ (1977) Hardware Veri␈↓α␈↓βC␈↓α␈↓cation.
                                          ␈↓ εh␈↓␈↓ π_Ph.D. thesis, Stanford University.
␈↓ ↓H␈↓␈↓αMcCarthy, J. and Painter, J.␈↓ (1967) Correctness␈↓ εh␈↓␈↓ π_Available as <Arpanet:SAIL>
␈↓ ↓H␈↓␈↓ ↓xof a Compiler for Arithmetic Expressions. ␈↓ εh␈↓␈↓ π_THESIS.PUB[THE,TJW].
␈↓ ↓H␈↓␈↓ ↓x␈↓↓Proc. Symposia in Applied Math., Math.
␈↓ ↓H␈↓↓␈↓ ↓xAspects of Computer Science␈↓ New York:
␈↓ ↓H␈↓␈↓ ↓xAmer. Math. Soc..

␈↓ ↓H␈↓␈↓αMcCarthy, J. and Hayes, P.J.␈↓ (1969) Some
␈↓ ↓H␈↓␈↓ ↓xPhilosophical Problems from the
␈↓ ↓H␈↓␈↓ ↓xStandpoint of Arti␈↓α␈↓βC␈↓α␈↓cial Intelligence.
␈↓ ↓H␈↓␈↓ ↓x␈↓↓Machine Intelligence 4␈↓, pp. 463-502 (eds
␈↓ ↓H␈↓␈↓ ↓xMeltzer, B. and Michie, D.). Edinburgh:
␈↓ ↓H␈↓␈↓ ↓xEdinburgh University Press.

␈↓ ↓H␈↓␈↓αMcCarthy, J.␈↓ (1979a) First Order Theories of
␈↓ ↓H␈↓␈↓ ↓xIndividual Concepts and Propositions.
␈↓ ↓H␈↓␈↓ ↓x␈↓↓Machine Intelligence 9␈↓, (Ed., Michie, D.).
␈↓ ↓H␈↓␈↓ ↓xEdinburgh, Edinburgh University Press.

␈↓ ↓H␈↓␈↓αMcCarthy, J.␈↓ (1979b) Ascribing Mental
␈↓ ↓H␈↓␈↓ ↓xQualities to Machines.  ␈↓↓Philosophical
␈↓ ↓H␈↓↓␈↓ ↓xPerspectives in Arti␈↓␈↓βS␈↓␈↓↓cial Intelligence␈↓, (Ed.,
␈↓ ↓H␈↓␈↓ ↓xRingle, M.).  Harvester Press.

␈↓ ↓H␈↓␈↓αCartwright, Robert and McCarthy, John␈↓
␈↓ ↓H␈↓␈↓ ↓x(1979c) Recursive Programs as     Functions
␈↓ ↓H␈↓␈↓ ↓xin a First Order Theory. ␈↓↓Proceedings of the
␈↓ ↓H␈↓↓␈↓ ↓xInternational Conference on Mathematical
␈↓ ↓H␈↓↓␈↓ ↓xStudies of Information Processing␈↓, Kyoto,
␈↓ ↓H␈↓␈↓ ↓xJapan.

␈↓ ↓H␈↓␈↓αMcCarthy, J.␈↓ (1980) Circumscription - A Form
␈↓ ↓H␈↓␈↓ ↓xof Non-Monotonic Reasoning.  ␈↓↓Arti␈↓␈↓βS␈↓␈↓↓cial
␈↓ ↓H␈↓↓␈↓ ↓xIntelligence␈↓, Volume l3, Numbers l, 2, April.

␈↓ ↓H␈↓␈↓αMcCarthy, J., Sato, M., Hayashi, T. and
␈↓ ↓H␈↓α␈↓ ↓xIgarashi, S.␈↓ (1977e) On the Model Theory
␈↓ ↓H␈↓␈↓ ↓xof Knowledge.  Presented at ␈↓↓IJCAI-1977␈↓;
␈↓ ↓H␈↓␈↓ ↓xto appear in the ␈↓↓SIGART Newsletter␈↓.