/FONT#0=NGR25/FONT#1=BDI25
␈↓␈↓↓␈↓
␈↓ ↓H
Draft for comments: This paper is in the file LIGHT.RE3[2,JMC]@SAIL
␈↓ ↓H
It was solicited as a review by the AI Journal.
␈↓ ↓H

␈↓ ↓H
Artificial  Intelligence: A General  Survey by
␈↓ ↓H
Professor  Sir  James Lighthill,  FRS, in  ␈↓↓Artificial  Intelligence: a
␈↓ ↓H
paper symposium␈↓, Science Research Council 1973
␈↓ ↓H

␈↓ ↓H
␈↓ α_Professor␈αLighthill␈αof␈αCambridge␈αUniversity␈αis␈αa␈αfamous␈αhydrodynamicist␈αwith␈α
a␈α
recent␈α
interest␈α
in
␈↓ ↓Happlications␈αto␈αbiology.␈αHis␈αreview␈αof␈αartificial␈αintelligence␈αwas␈α
at␈α
the␈α
request␈α
of␈α
Brian␈α
Flowers,␈α
head␈α
of
␈↓ ↓Hthe␈α∞Science␈α∞Research␈α∞Council␈α∞of␈α∞Great␈α∞Britain,␈α∞the␈α
main␈α
funding␈α
body␈α
for␈α
British␈α
university␈α
scientific
␈↓ ↓Hresearch.␈αIts␈αpurpose␈αwas␈αto␈αhelp␈αthe␈αScience␈αResearch␈αCouncil␈αdecide␈αrequests␈α
for␈α
support␈α
of␈α
work␈α
in
␈↓ ↓HAI.␈α∂Lighthill␈α∂claims␈α∞no␈α∞previous␈α∞acquaintance␈α∞with␈α∞the␈α∞field,␈α∞but␈α∞refers␈α∞to␈α∞a␈α∞large␈α∞number␈α∞of␈α∞authors
␈↓ ↓Hwhose␈α
works␈α
he␈α
consulted␈α
though␈α
not␈α
to␈α
any␈α
specific␈α
papers.

␈↓ α_Unfortunately,␈α∂workers␈α∞in␈α∞artificial␈α∞intelligence␈α∞lose␈α∞intellectual␈α∞contact␈α∞with␈α∞Professor␈α∞Lighthill
␈↓ ↓Halmost␈α⊂immediately,␈α⊂because␈α⊂he␈α⊂defines␈α⊂the␈α⊂field␈α⊂in␈α⊂such␈α⊂a␈α⊂way␈α⊂as␈α⊂to␈α⊂exclude␈α⊂our␈α⊂most␈α⊂important
␈↓ ↓Hresearch␈αgoals.␈αHe␈αdoes␈αthis␈αby␈αclassifying␈αwork␈αin␈αartificial␈αintelligence␈αinto␈αthree␈αcategories␈αA,␈αB␈αand
␈↓ ↓HC.␈α⊃A␈α⊃stands␈α⊃for␈α⊃␈↓↓advanced␈α⊂automation␈↓␈α⊂which␈α⊂he␈α⊂likes,␈α⊂C␈α⊂stands␈α⊂for␈α⊂␈↓↓connections␈↓␈α⊂to␈α⊂psychology␈α⊂and
␈↓ ↓Hneurophysiology␈α⊂which␈α⊂he␈α⊂also␈α∂likes,␈α∂and␈α∂B␈α∂stands␈α∂for␈α∂␈↓↓bridge␈↓␈α∂between␈α∂the␈α∂other␈α∂two␈α∂and␈α∂also␈α∂for
␈↓ ↓H␈↓↓building␈αrobots␈↓␈αboth␈αof␈α
which␈α
he␈α
doesn't␈α
like.␈α
The␈α
term␈α
␈↓↓robot␈↓␈α
is␈α
defined␈α
in␈α
an␈α
idiosyncratic␈α
way␈α
to␈α
refer
␈↓ ↓Hto␈αany␈αcomputer␈αprogram␈αor␈αdevice␈αwhich␈αis␈αneither␈αaimed␈αat␈αan␈αapplication␈αnor␈αintended␈αto␈αmodel␈αthe
␈↓ ↓Hbrain.␈α∂He␈α∂then␈α∂states␈α∂that␈α∂activities␈α∂in␈α∂B␈α∂can␈α∞be␈α∞justified␈α∞only␈α∞in␈α∞so␈α∞far␈α∞as␈α∞they␈α∞make␈α∞a␈α∞connection
␈↓ ↓Hbetween␈α
A␈α
and␈α
C.

␈↓ α_Lighthill's␈α⊂ABC␈α⊂classification␈α⊂seems␈α⊂to␈α⊂exclude␈α⊂the␈α⊂possibility␈α⊂that␈α⊂there␈α⊂can␈α⊂be␈α⊂a␈α∂science␈α∂of
␈↓ ↓Hintelligent␈α
behavior␈α
that␈α
can␈α
be␈α
studied␈α
apart␈α
from␈α
applications␈α
and␈α
apart␈α
from␈α
biology.␈αHowever,␈αfor
␈↓ ↓Halmost␈α∂all␈α∂workers␈α∂in␈α∂the␈α∂field,␈α∂the␈α∂whole␈α∂idea␈α∂of␈α∂artificial␈α∂intelligence␈α∂is␈α∂that␈α∞the␈α∞relation␈α∞between
␈↓ ↓Hproblems␈α
and␈α
problem␈α
solving␈α
methods␈αand␈αthe␈αrelation␈αbetween␈αsituations␈αand␈αthe␈αbehavior␈αthat␈αwill
␈↓ ↓Hachieve␈α
goals␈α
can␈α
be␈α
studied␈α
by␈α
theory␈α
and␈α
computer␈α
experiment␈α
as␈α
an␈α
independent␈α
subject.

␈↓ α_He␈αmakes␈αno␈αargument␈αfor␈αhis␈αclassification,␈αand␈αgives␈αno␈αhint␈αthat␈αanyone␈αmay␈αthink␈αdifferently.
␈↓ ↓HThis␈α∞is␈α∞somewhat␈α∞puzzling␈α∞since␈α∞a␈α∞number␈α∞of␈α
the␈α
documents␈α
submitted␈α
by␈α
British␈α
AI␈α
workers␈α
for␈α
his
␈↓ ↓Hconsideration␈α⊂are␈α∂quite␈α∂explicit␈α∂about␈α∂the␈α∂point.␈α∂Perhaps␈α∂ignoring␈α∂this␈α∂claim␈α∂plays␈α∂a␈α∂tactical␈α∂role␈α∂in
␈↓ ↓Hjustifying␈αhis␈αproposal␈αthat␈αbasic␈αexperimental␈αresearch␈αin␈αAI␈αbe␈α
abandoned,␈α
because␈α
if␈α
AI␈α
research␈α
has
␈↓ ↓Hscientific␈αproblems␈αof␈αits␈αown,␈α
they␈α
should␈α
be␈α
pursued␈α
even␈α
though␈α
the␈α
level␈α
of␈α
funding␈α
may␈α
depend␈α
on
␈↓ ↓Hthe␈αprospects␈αfor␈αresults␈αat␈αthe␈αpresent␈αlevel␈αof␈αknowledge␈αand␈αtalent.␈αWhereas␈α
if␈α
the␈α
research␈α
is␈α
only
␈↓ ↓Ha␈αmeans␈α
toward␈α
solving␈α
some␈α
other␈α
scientific␈α
or␈α
practical␈α
problems,␈α
then␈α
the␈α
subject␈α
may␈α
be␈α
abandoned
␈↓ ↓Hif␈α
there␈α
are␈α
more␈α
promising␈α
ways␈α
of␈α
solving␈α
the␈α
other␈α
problems.

␈↓ α_Having␈αignored␈αthe␈α
possibility␈α
that␈α
AI␈α
has␈α
goals␈α
of␈α
its␈α
own,␈α
Lighthill␈α
goes␈α
on␈α
to␈α
document␈α
his␈α
claim
␈↓ ↓Hthat␈αit␈αhas␈αnot␈αcontributed␈αto␈αapplications␈αor␈αto␈αpsychology␈αand␈αphysiology.␈αHe␈αexaggerates␈αa␈αbit␈αhere,
␈↓ ↓Hit␈α∂seems␈α∂worthwhile␈α∂to␈α∂spend␈α∂some␈α∂effort␈α∂disputing␈α∂his␈α∞claims␈α∞that␈α∞AI␈α∞has␈α∞not␈α∞contributed␈α∞to␈α∞these
␈↓ ↓Hother␈α
subjects.

␈↓ α_In␈α∞my␈α∞opinion,␈α∞AI's␈α∞contribution␈α∞to␈α∞practical␈α∞applications␈α∞has␈α
been␈α
significant␈α
but␈α
so␈α
far␈α
mostly
␈↓ ↓Hperipheral␈αto␈αthe␈αcentral␈αideas␈αand␈αproblems␈αof␈αAI.␈αThus␈αthe␈αLISP␈αlanguage␈αfor␈αsymbolic␈α
computing␈α
was
␈↓ ↓Hdeveloped␈α∞for␈α∞AI␈α∞use,␈α∞but␈α
has␈α
had␈α
applications␈α
to␈α
symbolic␈α
computations␈α
in␈α
other␈α
areas,␈α
e.g.␈α
physics.
␈↓ ↓HMoreover,␈αsome␈αideas␈αfrom␈αLISP␈αsuch␈α
as␈α
conditional␈α
expressions␈α
and␈α
recursive␈α
function␈α
definitions␈α
have
␈↓ ↓Hbeen␈αused␈αin␈αother␈αprogramming␈α
languages.␈α
However,␈α
the␈α
ideas␈α
that␈α
have␈α
been␈α
applied␈α
elsewhere␈α
don't
␈↓ ↓Hhave␈αa␈αspecifically␈αAI␈αcharacter␈αand␈αmight␈αhave␈αbeen␈αbut␈αweren't␈αdeveloped␈αwithout␈αAI␈αin␈αmind.␈αOther
␈↓ ↓Hexamples␈αinclude␈αtime-sharing,␈αthe␈αfirst␈αproposals␈αfor␈αwhich␈αhad␈αAI␈αmotivations␈αand␈αsome␈α
techniques␈α
of
␈↓ ↓Hpicture␈α∞processing␈α∞that␈α
were␈α
first␈α
developed␈α
in␈α
AI␈α
laboratories␈α
and␈α
have␈α
been␈α
used␈α
elsewhere.␈α
Even
␈↓ ↓Hthe␈α∞current␈α∞work␈α∞in␈α∞automatic␈α∞assembly␈α∞using␈α∞vision␈α∞might␈α∞have␈α∞been␈α∞developed␈α∞without␈α∞AI␈α
in␈α
mind.
␈↓ ↓HHowever,␈α∪the␈α∪Dendral␈α∪work␈α∩has␈α∩always␈α∩had␈α∩a␈α∩specifically␈α∩AI␈α∩character,␈α∩and␈α∩many␈α∩of␈α∩the␈α∩recent
␈↓ ↓Hdevelopments␈α
in␈α
programming␈α
such␈α
as␈α
PLANNER␈α
and␈α
CONNIVER␈α
have␈α
an␈α
AI␈α
motivation.

␈↓ α_AI's␈α∞contributions␈α∞to␈α∞neurophysiology␈α∞have␈α∞been␈α∞small␈α∞and␈α∞mostly␈α∞of␈α∞a␈α∞negative␈α∞character,␈α∞i.e.
␈↓ ↓Hshowing␈αthat␈αcertain␈αmechanisms␈αthat␈α
neurophysiologists␈α
propose␈α
are␈α
not␈α
well␈α
defined␈α
or␈α
inadequate␈α
to
␈↓ ↓Hcarry␈αout␈αthe␈αbehavior␈αthey␈αare␈αsupposed␈αto␈αaccount␈αfor.␈αI␈αhave␈αin␈αmind␈αHebb's␈αproposals␈αin␈αhis␈αbook
␈↓ ↓H␈↓↓The␈αOrganization␈αof␈αBehavior␈↓.␈αNo-one␈αtoday␈αwould␈αbelieve␈αthat␈αthe␈αgaps␈αin␈αthose␈αideas␈αcould␈αbe␈αfilled
␈↓ ↓Hwithout␈α
adding␈α
something␈αmuch␈αlarger␈αthan␈αthe␈αoriginal␈αwork.␈αMoreover,␈αthe␈αlast␈α20␈αyears␈αexperience
␈↓ ↓Hin␈α
programming␈α
machines␈α
to␈α
learn␈α
and␈α
solve␈αproblems␈αmakes␈αit␈αimplausible␈αthat␈αcell␈αassemblies␈α␈↓↓per␈αse␈↓
␈↓ ↓Hwould␈α∂learn␈α∂much␈α∂without␈α∂putting␈α∂in␈α∞some␈α∞additional␈α∞organization,␈α∞and␈α∞physiologists␈α∞today␈α∞would␈α∞be
␈↓ ↓Hunlikely␈α
to␈α
propose␈α
such␈α
a␈αtheory.␈αHowever,␈αmerely␈αshowing␈αthat␈αsome␈αthings␈αare␈αunlikely␈αto␈αwork␈αis
␈↓ ↓Hnot␈α
a␈α
␈↓↓positive␈↓␈α
contribution.␈α
I␈α
think␈α
there␈α
will␈α
be␈α
more␈α
interaction␈α
between␈α
AI␈α
and␈αneurophysiology␈αas
␈↓ ↓Hsoon␈α∞as␈α
the␈α
neurophysiologists␈α
are␈α
in␈α
a␈α
position␈α
to␈α
compare␈α
information␈α
processing␈α
models␈α
of␈α
higher
␈↓ ↓Hlevel␈αfunctions␈αwith␈αphysiological␈αdata.␈α
There␈α
is␈α
little␈α
contact␈α
at␈α
the␈α
nerve␈α
cell␈α
level,␈α
because,␈α
as␈α
Minsky
␈↓ ↓Hshowed␈αin␈αhis␈αPhD␈α
dissertation␈α
in␈α
1954,␈α
almost␈α
any␈α
of␈α
the␈α
proposed␈α
models␈α
of␈α
the␈α
neuron␈α
is␈α
a␈α
universal
␈↓ ↓Hcomputing␈α⊂element,␈α⊂so␈α⊂that␈α⊂there␈α⊂is␈α⊂no␈α⊂connection␈α⊂between␈α⊂the␈α⊂structure␈α⊂of␈α∂the␈α∂neuron␈α∂and␈α∂what
␈↓ ↓Hhigher␈α
level␈α
processes␈α
are␈α
possible.

␈↓ α_On␈αthe␈αother␈αhand,␈αthe␈αeffects␈αof␈αartificial␈αintelligence␈αresearch␈αon␈αpsychology␈αhave␈αbeen␈αlarger
␈↓ ↓Has␈α∂attested␈α∂by␈α∂various␈α∂psychologists.␈α∂First␈α∂of␈α∂all,␈α∂psychologists␈α∂have␈α∞begun␈α∞to␈α∞use␈α∞models␈α∞in␈α∞which
␈↓ ↓Hcomplex␈αinternal␈αdata␈αstructures␈αthat␈αcannot␈αbe␈αobserved␈α
directly␈α
are␈α
attributed␈α
to␈α
animals␈α
and␈α
people.
␈↓ ↓HPsychologists␈αhave␈αcome␈αto␈αuse␈αthese␈αmodels,␈αbecause␈αthey␈αexhibit␈αbehavior␈αthat␈αcannot␈αbe␈αexhibited
␈↓ ↓Hby␈αmodels␈αconforming␈αto␈αthe␈αtenets␈αof␈αbehaviorism␈αwhich␈αessentially␈αallows␈αonly␈αconnections␈αbetween
␈↓ ↓Hexternally␈α∃observable␈α∃variables.␈α∃Information␈α∃processing␈α∀models␈α∀in␈α∀psychology␈α∀have␈α∀also␈α∀induced
␈↓ ↓Hdissatisfaction␈α→with␈α→psychoanalytic␈α_and␈α_related␈α_theories␈α_of␈α_emotional␈α_behavior.␈α_Namely,␈α_these
␈↓ ↓Hinformation␈α⊂processing␈α⊂models␈α⊂of␈α∂emotional␈α∂states␈α∂can␈α∂yield␈α∂predictions␈α∂that␈α∂can␈α∂be␈α∂compared␈α∂with
␈↓ ↓Hexperiment␈αor␈αexperience␈αin␈αa␈αmore␈αdefinite␈αway␈αthan␈αcan␈αthe␈αvague␈αmodels␈αof␈αpsychoanalysis␈αand␈αits
␈↓ ↓Hoffspring.

␈↓ α_Contributions␈α∂of␈α∂AI␈α∂to␈α∂psychology␈α∂are␈α∂further␈α∞discussed␈α∞in␈α∞the␈α∞paper␈α∞␈↓↓Some␈α∞Comments␈α∞on␈α∞the
␈↓ ↓HLighthill␈α∂Report␈↓␈α∂by␈α∂N.␈α∂S.␈α∂Sutherland␈α∞which␈α∞was␈α∞included␈α∞in␈α∞the␈α∞same␈α∞book␈α∞with␈α∞the␈α∞Lighthill␈α∞report
␈↓ ↓Hitself.

␈↓ α_Systematic␈α
comment␈α
on␈α
the␈α
main␈α
section,␈α
entitled␈α
␈↓↓Past␈α
Disappointments␈↓␈α
is␈α
difficult␈α
because␈α
of␈α
the
␈↓ ↓Hstrange␈α
way␈α
the␈α
subject␈α
is␈α
divided␈α
up␈α
but␈α
here␈α
are␈α
some␈α
remarks:

␈↓ α_1.␈α∃Automatic␈α∃landing␈α∃systems␈α∃for␈α∃airplanes␈α∃are␈α∃offered␈α∃as␈α∃a␈α∃field␈α∃in␈α∀which␈α∀conventional
␈↓ ↓Hengineering␈α∂techniques␈α∂have␈α∂been␈α∂more␈α∂successful␈α∂than␈α∂AI␈α∂methods.␈α∞Indeed,␈α∞no-one␈α∞would␈α∞advocate
␈↓ ↓Happlying␈αthe␈αscene␈αanalysis␈αor␈αtree␈αsearch␈αtechniques␈αdeveloped␈αin␈αAI␈αresearch␈αto␈αautomatic␈αlanding␈α
in
␈↓ ↓Hthe␈α⊃context␈α⊂in␈α⊂which␈α⊂automatic␈α⊂landing␈α⊂has␈α⊂been␈α⊂developed.␈α⊂Namely,␈α⊂radio␈α⊂signals␈α⊂are␈α⊂available␈α⊂to
␈↓ ↓Hdetermine␈α∂the␈α∂precise␈α∂position␈α∂of␈α∂the␈α∂airplane␈α∞in␈α∞relation␈α∞to␈α∞a␈α∞straight␈α∞runway␈α∞which␈α∞is␈α∞guaranteed
␈↓ ↓Hclear␈αof␈αinterfering␈αobjects.␈αAI␈αtechniques␈αwould␈αbe␈αnecessary␈αto␈αmake␈αa␈αsystem␈αcapable␈αof␈αlanding␈αon
␈↓ ↓Han␈α⊂unprepared␈α⊂dirt␈α⊂strip␈α⊂with␈α⊂no␈α⊂radio␈α⊂aids␈α∂which␈α∂had␈α∂to␈α∂be␈α∂located␈α∂and␈α∂distinguished␈α∂from␈α∂roads
␈↓ ↓Hvisually␈α
and␈α
which␈α
might␈α
have␈α
cows␈α
or␈α
potholes␈α
or␈α
muddy␈α
places␈α
on␈α
it.␈α
The␈α
problem␈α
of␈αautomatically
␈↓ ↓Hdriving␈α
an␈α
automobile␈α
in␈α
an␈αuncontrolled␈αenvironment␈αis␈αeven␈αmore␈αdifficult␈αand␈αwill␈αdefinitely␈αrequire
␈↓ ↓HAI␈α
techniques,␈α
which,␈α
however,␈α
are␈α
not␈α
nearly␈α
ready␈α
for␈α
a␈α
full␈α
solution␈α
of␈α
such␈α
a␈α
difficult␈α
problem.

␈↓ α_2.␈αLighthill␈αis␈α
disappointed␈α
that␈α
detailed␈α
knowledge␈α
of␈α
subject␈α
matter␈α
has␈α
to␈α
be␈α
put␈α
in␈α
if␈α
programs
␈↓ ↓Hare␈α∂to␈α∂be␈α∂successful␈α∂in␈α∂theorem␈α∂proving,␈α∂interpreting␈α∂mass␈α∂spectra,␈α∂and␈α∂game␈α∂playing.␈α∞He␈α∞uses␈α∞the
␈↓ ↓Hword␈α
␈↓↓heuristics␈↓␈α
in␈α
a␈α
non-standard␈α
way␈α
for␈α
this.␈α
He␈α
misses␈α
the␈α
fact␈αthat␈αthere␈αare␈αgreat␈αdifficulties␈αin
␈↓ ↓Hfinding␈αways␈αof␈αrepresenting␈αknowledge␈αof␈αthe␈αworld␈αin␈αcomputer␈αprograms␈αand␈αmuch␈αAI␈αresearch␈αand
␈↓ ↓Hinternal␈α∩controversy␈α∩are␈α∩directed␈α∩to␈α∩this␈α∩problem.␈α∩Moreover,␈α∩most␈α⊃AI␈α⊃researchers␈α⊃feel␈α⊃that␈α⊃more
␈↓ ↓Hprogress␈α
on␈α
this␈α
␈↓↓representation␈α
problem␈↓␈α
is␈α
essential␈α
before␈α
substantial␈α
progress␈α
can␈α
be␈α
made␈α
on␈α
the
␈↓ ↓Hproblem␈α∃of␈α∃automatic␈α∃acquisition␈α∃of␈α∃knowledge.␈α∃Of␈α∃course,␈α∃missing␈α∀these␈α∀particular␈α∀points␈α∀is␈α∀a
␈↓ ↓Hconsequence␈α
of␈α
missing␈α
the␈α
existence␈α
of␈α
the␈α
AI␈α
problem␈α
as␈α
distinct␈αfrom␈αautomation␈αand␈αstudy␈αof␈αthe
␈↓ ↓Hcentral␈α
nervous␈α
system.

␈↓ α_3.␈α
A␈α
further␈α
disappointment␈α
is␈α
that␈α
chess␈α
playing␈α
programs␈α
have␈αonly␈αreached␈αan␈α"experienced
␈↓ ↓Hamateur"␈αlevel␈αof␈αplay.␈αWell,␈αif␈αprograms␈αcan't␈αdo␈αbetter␈α
than␈α
that␈α
by␈α
1978,␈α
I␈α
shall␈α
lose␈α
250␈α
pounds␈α
and
␈↓ ↓Hwill␈αbe␈αdisappointed␈αtoo␈αthough␈αnot␈αextremely␈αsurprised.␈αThe␈αpresent␈αlevel␈αof␈αcomputer␈α
chess␈α
is␈α
based
␈↓ ↓Hon␈α⊂the␈α⊂incorporation␈α⊂of␈α⊂certain␈α∂intellectual␈α∂mechanisms␈α∂in␈α∂the␈α∂programs.␈α∂Some␈α∂improvement␈α∂can␈α∂be
␈↓ ↓Hmade␈αby␈αfurther␈αrefinement␈αof␈αthe␈αheuristics␈αin␈αthe␈αprograms,␈αbut␈αprobably␈αmaster␈αlevel␈αchess␈αawaits
␈↓ ↓Hthe␈αability␈αto␈αput␈αgeneral␈αconfiguration␈αpatterns␈αinto␈αthe␈αprograms␈αin␈αan␈αeasy␈αand␈αflexible␈αway.␈αI␈αdon't
␈↓ ↓Hsee␈α
how␈α
to␈α
set␈α
a␈α
date␈α
by␈α
which␈αthis␈αproblem␈αmust␈αbe␈αsolved␈αin␈αorder␈αto␈αavoid␈αdisappointment␈αin␈αthe
␈↓ ↓Hfield␈α
of␈α
artificial␈α
intelligence␈α
as␈α
a␈α
whole.

␈↓ ↓HDid␈α
We␈α
Deserve␈α
It?

␈↓ α_Lighthill␈α
had␈αhis␈αshot␈αat␈αAI␈αand␈αmissed,␈αbut␈αthis␈αdoesn't␈αprove␈αthat␈αeverything␈αin␈αAI␈αis␈αok.␈αIn␈αmy
␈↓ ↓Hopinion,␈α∪present␈α∪AI␈α∪research␈α∪suffers␈α∪from␈α∪some␈α∩major␈α∩deficiencies␈α∩apart␈α∩from␈α∩the␈α∩fact␈α∩that␈α∩any
␈↓ ↓Hscientists␈α
would␈α
achieve␈α
more␈α
if␈α
they␈α
were␈α
smarter␈α
and␈α
worked␈α
harder.

␈↓ α_1.␈α∞Much␈α∞work␈α∞in␈α∞AI␈α∞has␈α∞the␈α∞"look␈α
ma,␈α
no␈α
hands"␈α
disease.␈α
Someone␈α
programs␈α
a␈α
computer␈α
to␈α
do
␈↓ ↓Hsomething␈αno␈αcomputer␈αhas␈αdone␈αbefore␈αand␈αwrites␈αa␈αpaper␈αpointing␈αout␈αthat␈αthe␈αcomputer␈αdid␈αit.␈αThe
␈↓ ↓Hpaper␈α
is␈α
not␈α
directed␈α
to␈α
the␈α
identification␈α
and␈α
study␈α
of␈α
intellectual␈α
mechanisms␈αand␈αoften␈αcontains␈αno
␈↓ ↓Hcoherent␈αaccount␈αof␈αhow␈αthe␈αprogram␈αworks␈α
at␈α
all.␈α
As␈α
an␈α
example,␈α
consider␈α
that␈α
the␈α
SIGART␈α
Newsletter
␈↓ ↓Hprints␈α
the␈α
scores␈α
of␈α
the␈α
games␈α
in␈α
the␈α
ACM␈αComputer␈αChess␈αTournament␈αjust␈αas␈αthough␈αthe␈αprograms
␈↓ ↓Hwere␈α
human␈α
players␈α
and␈α
their␈α
innards␈α
were␈α
inaccessible.␈α
We␈α
need␈α
to␈α
know␈α
why␈αone␈αprogram␈αmissed
␈↓ ↓Hthe␈α∞right␈α∞move␈α∞in␈α∞a␈α∞position␈α∞-␈α
what␈α
was␈α
it␈α
thinking␈α
about␈α
all␈α
that␈α
time?␈α
We␈α
also␈α
need␈α
an␈α
analysis␈α
of
␈↓ ↓Hwhat␈αclass␈αof␈αpositions␈αthe␈αparticular␈αone␈αbelonged␈αto␈αand␈αhow␈αa␈αfuture␈αprogram␈αmight␈αrecognize␈αthis
␈↓ ↓Hclass␈α
and␈α
play␈α
better.

␈↓ α_2.␈α∩Every␈α⊃now␈α⊃and␈α⊃then,␈α⊃some␈α⊃AI␈α⊃scientist␈α⊃gets␈α⊃an␈α⊃idea␈α⊃for␈α⊃a␈α⊃general␈α⊃scheme␈α⊃of␈α⊃intelligent
␈↓ ↓Hbehavior␈αthat␈αcan␈αbe␈αapplied␈αto␈αany␈αproblem␈α
provided␈α
the␈α
machine␈α
is␈α
given␈α
the␈α
specific␈α
knowledge␈α
that
␈↓ ↓Ha␈α
human␈α
has␈α
about␈α
the␈α
domain.␈α
Examples␈α
of␈α
this␈α
have␈α
included␈α
the␈α
GPS␈α
formalism,␈αa␈αsimple␈αpredicate
␈↓ ↓Hcalculus␈αformalism,␈αand␈αmore␈αrecently␈αthe␈αPLANNER␈αformalism␈αand␈αperhaps␈α
the␈α
current␈α
Carnegie-Mellon
␈↓ ↓Hproduction␈α⊃formalism.␈α⊃In␈α⊃the␈α⊃first␈α⊃and␈α⊃third␈α⊃cases,␈α⊃the␈α⊃belief␈α⊃that␈α⊂any␈α⊂problem␈α⊂solving␈α⊂ability␈α⊂and
␈↓ ↓Hknowledge␈α∩could␈α∩be␈α∩fitted␈α∩into␈α∩the␈α∩formalisms␈α⊃led␈α⊃to␈α⊃published␈α⊃predictions␈α⊃that␈α⊃computers␈α⊃would
␈↓ ↓Hachieve␈αcertain␈α
levels␈α
of␈α
performance␈α
in␈α
certain␈α
time␈α
scales.␈α
If␈α
the␈α
inventors␈α
of␈α
the␈α
formalisms␈α
had␈α
been
␈↓ ↓Hright␈α⊃about␈α⊃them,␈α⊃the␈α⊃goals␈α⊃might␈α⊃have␈α⊃been␈α⊃achieved,␈α⊃but␈α⊃regrettably␈α⊂they␈α⊂were␈α⊂mistaken.␈α⊂Such
␈↓ ↓Hgeneral␈α⊃purpose␈α⊃formalisms␈α⊃will␈α⊃be␈α⊂invented␈α⊂from␈α⊂time␈α⊂to␈α⊂time,␈α⊂and,␈α⊂most␈α⊂likely,␈α⊂one␈α⊂of␈α⊂them␈α⊂will
␈↓ ↓Heventually␈αprove␈αadequate.␈αHowever,␈αit␈αwould␈αbe␈αa␈αgreat␈αrelief␈αto␈αthe␈αrest␈αof␈αthe␈αworkers␈αin␈αAI␈αif␈αthe
␈↓ ↓Hinventors␈α⊂of␈α⊂new␈α⊂general␈α⊂formalisms␈α⊂would␈α⊂express␈α⊂their␈α⊂hopes␈α⊂in␈α⊂a␈α⊂more␈α⊂guarded␈α∂form␈α∂than␈α∂has
␈↓ ↓Hsometimes␈α
been␈α
the␈α
case.

␈↓ α_3.␈α
At␈αpresent,␈αthere␈αdoes␈αnot␈αexist␈αa␈αcomprehensive␈αgeneral␈αreview␈αof␈αAI␈αthat␈αdiscusses␈αall␈αthe
␈↓ ↓Hmain␈αapproaches␈αand␈αachievements␈αand␈αissues.␈αMost␈αlikely,␈αthis␈αis␈αnot␈αmerely␈αbecause␈αthe␈αfield␈αdoesn't
␈↓ ↓Hhave␈αa␈αfirst␈αrate␈αreviewer␈αat␈αpresent,␈αbut␈αbecause␈αthe␈αfield␈αis␈αconfused␈αabout␈αwhat␈αthese␈αapproaches
␈↓ ↓Hand␈α
achievements␈α
and␈α
issues␈α
are.␈αThe␈αproduction␈αof␈αsuch␈αa␈αreview␈αwill␈αtherefore␈αbe␈αa␈αmajor␈αcreative
␈↓ ↓Hwork␈α
and␈α
not␈α
merely␈α
a␈α
work␈α
of␈α
scholarship.

␈↓ ∧λJohn␈α
McCarthy␈α
-␈α
9␈α
March␈α
1974