/LMAR=0/XLINE=3/FONT#0=BASL30/FONT#1=BASI30/FONT#2=BASB30/FONT#10=BAXM30/FONT#11=ZERO30/FONT#3=STA200/FONT#4=NGB25
␈↓ ↓H␈↓␈↓βS␈↓∧ Artificial Intelligence Laboratory, STANFORD UNIVERSITY, Stanford, California 94305

␈↓ ↓H␈↓∧Telephone 415 497-4430␈↓ 	nFebruary 7, 1978 




␈↓ ↓H␈↓Professor Charles Akemann
␈↓ ↓H␈↓Department of Mathematics
␈↓ ↓H␈↓U. C. Santa Barbara
␈↓ ↓H␈↓Santa Barbara, CA 93106

␈↓ ↓H␈↓Dear Professor Akemann:

␈↓ ↓H␈↓        I␈α
feel␈α
∨attered␈αto␈α
be␈α
asked␈α
to␈αwrite␈α
a␈α
recommendation␈α
for␈αMartin␈α
Davis.␈α
 I␈α
won't␈αwrite␈α
about
␈↓ ↓H␈↓his␈α∂world-wide␈α⊂reputation␈α∂in␈α⊂mathematical␈α∂logic␈α∂and␈α⊂recursive␈α∂function␈α⊂theory␈α∂because␈α⊂he␈α∂has
␈↓ ↓H␈↓undoubtedly␈α
as␈α
references,␈α
people␈αwho␈α
also␈α
have␈α
substantial␈α
reputation␈αin␈α
that␈α
≡eld.␈α
 (For␈α
me␈αto
␈↓ ↓H␈↓write␈αabout␈α
his␈αrole␈αin␈α
settling␈αHilbert's␈α
tenth␈αproblem␈αcould␈α
only␈αserve␈α
to␈αshow␈αthat␈α
I␈αam␈α
not␈αa
␈↓ ↓H␈↓complete␈αilliterate␈α
in␈αthe␈αarea,␈α
but␈αcan't␈αadd␈α
anything␈αto␈αhis␈α
fame.)␈αI␈αcan,␈α
however,␈αwrite␈αabout␈α
his
␈↓ ↓H␈↓work connected with computation and arti≡cal intelligence.

␈↓ ↓H␈↓        Martin␈αDavis␈αwas␈αone␈αof␈αthe␈α≡rst␈αpeople,␈αand␈αperhaps␈αeven␈αthe␈αvery␈α≡rst␈αperson,␈αto␈αwrite␈αa
␈↓ ↓H␈↓computer␈α⊃program␈α∩for␈α⊃proving␈α∩sentences␈α⊃in␈α∩≡rst␈α⊃order␈α⊃logic,␈α∩and␈α⊃he␈α∩was␈α⊃one␈α∩of␈α⊃the␈α∩≡rst␈α⊃to
␈↓ ↓H␈↓encounter␈αthe␈α
di≠culties␈αthat␈α
interfere␈αwith␈α
making␈αa␈α
computer␈αprogram␈α
which␈αcan␈αmatch␈α
human
␈↓ ↓H␈↓performance␈αin␈α≡rst␈α
order␈αtheorem␈αproving.␈α
 While␈αhe␈αdidn't␈α
make␈αit␈αhis␈α
main␈α≡eld␈αof␈αactivity,␈α
he
␈↓ ↓H␈↓retuurns to it from time to time, always with a substantial contribution.

␈↓ ↓H␈↓        Recently,␈αpartly␈αin␈αcollaboration␈αwith␈αJack␈αSchwartz,␈αhe␈αhas␈αbegun␈αstudying␈αthe␈αproblem␈αof
␈↓ ↓H␈↓proving␈α
programs␈α
correct.␈α
 Their␈α
approach␈α
seems␈α
promising,␈α
but␈α
since␈α
Martin␈α
has␈α
just␈αreturned␈α
to
␈↓ ↓H␈↓NYU␈α⊃and␈α⊂Jack␈α⊃is␈α⊃on␈α⊂sabbatical,␈α⊃it␈α⊂probably␈α⊃hasn't␈α⊃advanced␈α⊂far␈α⊃into␈α⊃actual␈α⊂implementation.
␈↓ ↓H␈↓Should␈α
he␈α
move␈α
that␈α
work␈α
to␈α
Santa␈αBarbara,␈α
I␈α
believe␈α
it␈α
will␈α
greatly␈α
bene≡t␈α
both␈αcomputer␈α
science
␈↓ ↓H␈↓and mathematics.

␈↓ ↓H␈↓        Last␈αyear␈αMartin␈αspent␈αa␈αmonth␈αat␈αthe␈αStanford␈αArti≡cial␈αIntelligence␈αLaboratory␈αwhich␈αhe
␈↓ ↓H␈↓followed␈αwith␈αbiweekly␈αvisits.␈α His␈αexpertness␈αin␈αmathematical␈αlogic␈αwas␈αof␈αgreat␈αassistance␈αto␈αour
␈↓ ↓H␈↓theorem␈α⊃proving␈α⊃and␈α⊃proof␈α⊂checking␈α⊃work.␈α⊃ Besides␈α⊃that,␈α⊂he␈α⊃helped␈α⊃clarify␈α⊃the␈α⊂mathematical
␈↓ ↓H␈↓nature␈α∃of␈α∃a␈α∀proposed␈α∃method␈α∃for␈α∀making␈α∃computer␈α∃programs␈α∀"jump␈α∃to␈α∃conclusions"␈α∀from
␈↓ ↓H␈↓insu≠cient␈αdata,␈α
something␈αwe␈αhave␈α
become␈αconvinced␈αis␈α
necessary␈αif␈αarti≡cial␈α
intelligence␈αis␈αto␈α
be
␈↓ ↓H␈↓achieved.␈α
 Martin␈α
is␈α∞a␈α
delightful␈α
collaborator,␈α∞and␈α
our␈α
graduate␈α∞students␈α
gained␈α
much␈α∞from␈α
his
␈↓ ↓H␈↓presence,␈αshort␈α
as␈αit␈αwas.␈α
 Of␈αall␈αthe␈α
famous␈αmathematical␈α
logicians,␈αhe␈αand␈α
Dana␈αScott␈αhave␈α
been
␈↓ ↓H␈↓the most helpful to computer science.

␈↓ ↓H␈↓        The prospect of having him on the West Coast again is very attractive.
␈↓ ↓H␈↓αProfessor Charles Akemann␈↓ ¬\February 7, 1978␈↓ 
nPage 2␈↓

␈↓ ↓H␈↓Sincerely,



␈↓ ↓H␈↓John McCarthy
␈↓ ↓H␈↓Director
␈↓ ↓H␈↓Professor of Computer Science