Hilbert S 24 Problems Assoc

Hilbert's Twenty-FourthProblem
Riidiger Thiele
1. INTRODUCTION. For geometers, Hilbert's influential work on the foundations

of geometry is important.For analysts, Hilbert's theory of integral equations is just as
important.But the address "MathematischeProbleme" [37] that David Hilbert (1862-
1943) delivered at the second InternationalCongress of Mathematicians (ICM) in Paris
has tremendous importance for all mathematicians. Moreover, a substantial part of
Hilbert's fame rests on this address from 1900 (the year after the American Mathe-
matical Society began to publish its Transactions). It was by the rapid publication of
Hilbert's paper [37] that the importance of the problems became quite clear, and it
was the American Mathematical Society that very quickly supplied English-language
readers with both a report on and a translation of Hilbert's address. (In Paris, the
United States and England were represented by seventeen and seven participants, re-
spectively.)
Indeed, this collection of twenty-three unsolved problems, in which Hilbert tried
"to lift the veil behind which the future lies hidden" [37, p. 437] has occupied much
attention since that time, with many mathematicians watching each contribution at-
tentively and directing their research accordingly. Hermann Weyl (1885-1955) once
remarked that "We mathematicians have often measured our progress by checking
which of Hilbert's questions had been settled in the meantime" [110, p. 525]. (See
also [31] and [115].)
Hilbert and his twenty-three problems have become proverbial. As a matter of
fact, however, because of time constraints Hilbert presented only ten of the prob-
lems at the Congress. Charlotte Angas Scott (1858-1931) reported on the Congress
and Hilbert's presentation of ten problems in the Bulletin of the American Mathemat-
ical Society [91]. The complete list of twenty-three problems only appeared in the
journal Gottinger Nachrichten in the fall of 1900 [37], and Mary Winston Newson
(1869-1959) translated the paper into English for the Bulletin in 1901 [37]. Already
by September 1900, George Bruce Halsted (1853-1922) had written in this MONTHLY
that Hilbert's beautiful paper on the problems of mathematics "is epoch-making for the
history of mathematics" [34, p. 188]. In his report on the InternationalCongress, Hal-
sted devoted about forty of the article's eighty lines to the problems. As to the actual
speech, no manuscript was preserved, nor was the text itself ever published.
Recently, Ivor Grattan-Guinness presented an interesting overview of Hilbert's
problems in the Notices of the American Mathematical Society, discussing the form
in which each of the twenty-three problems was published [30]. Yet, in dealing with
the celebrated problems from this viewpoint, he failed to mention the most interesting
problem of Hilbert's collection: the canceled twenty-fourth. Hilbert included it neither
in his address nor in any printed version, nor did he communicate it to his friends
Adolf Hurwitz (1859-1919) and Hermann Minkowski (1864-1909), who were proof-
readers of the paper submitted to the Gottinger Nachrichten and, more significantly,
were direct participantsin the developments surroundingHilbert's ICM lecture.
So, for a century now, the twenty-fourth problem has been a Sleeping Beauty. This
article will try to awaken it, thus giving the reader the chance to be the latter-day
Prince (or Princess) Charming who can take it home and solve it. This paper also aims
to convince the reader of the utility of the history of mathematics in the sense to which
January2003] HILBERT'STWENTY-FOURTHPROBLEM 1
ConstantinCaratheodory (1873-1950) referredin his speech at an MAA meetingin
1936 [10, p. 101]: "I will be glad if I have succeededin impressingthe idea thatit
is not only pleasantandentertainingto readat times the worksof the old mathemat-
ical authors,but thatthis may occasionallybe of use for the actualadvancementof
science."
2. THE CANCEIED PROBLEM. Let me prefacemy fairy tale "Lookingback-
ward,2003-1888" (i.e., fromthe presentback to the discoveryof a finiteideal basis
in invarianttheory)with somequestionsthatwill serveas a guidefor the remainderof
my investigation:
i. Why did Hilbertgive a talk on unsolvedproblemsand not on new resultsor
methodsin generaluse?
ii. How shouldthe twenty-fourthproblembe classifiedin relationto Hilbert'sfa-
mouscollectionof twenty-threeproblems?
iii. Why andhow did Hilbertdeal with the canceledtwenty-fourthproblemlater?
And whatpartdid this problemplay in Hilbert'slaterresearch?
iv. Finally,wheredid I findthe canceledproblem?
Let me beginby presentingtheproblemitself. The twenty-fourthproblembelongs
to the realmof foundationsof mathematics.In a nutshell,it asksfor the simplestproof
of any theorem.In his mathematicalnotebooks[38:3,pp. 25-26], Hilbertformulated
it as follows (author'stranslation):
The 24th problemin my Parislecturewas to be: Criteriaof simplicity,or proof

of the greatestsimplicityof certainproofs.Develop a theoryof the methodof
proofin mathematicsin general.Undera givenset of conditionstherecanbe but
one simplestproof. Quitegenerally,if thereare two proofsfor a theorem,you
mustkeep going untilyou havederivedeach fromthe other,or untilit becomes
quiteevidentwhatvariantconditions(andaids)havebeenusedin thetwo proofs.
Giventwo routes,it is not rightto takeeitherof these two or to look for a third;
it is necessaryto investigatethe arealying betweenthe two routes.Attemptsat
judgingthe simplicityof a proofarein my examinationof syzygiesandsyzygies
[Hilbertmisspelledthe word syzygies] betweensyzygies [see Hilbert[42, lec-
turesXXXII-XXXIX]].The use or the knowledgeof a syzygy simplifiesin an
essentialway a proofthata certainidentityis true.Becauseany processof ad-
dition [is] an applicationof the commutativelaw of additionetc. [andbecause]
this alwayscorrespondsto geometrictheoremsor logical conclusions,one can
count these [processes],and, for instance,in provingcertaintheoremsof ele-
mentarygeometry(thePythagorastheorem,[theorems]on remarkablepointsof
triangles),one canverywell decidewhichof theproofsis the simplest.[Author's
note:Partof the last sentenceis not only barelylegiblein Hilbert'snotebookbut
also grammatically incorrect.CorrectionsandinsertionsthatHilbertmadein this
entry show thathe wrote downthe problemin haste.]
In answerto the firstquestion,we beginwith a shortprehistoryof Hilbert'sfamous
speech.As preparation for treatingthe secondquestion,we nextpresentsomeremarks
on the natureof the proposedproblems.Hilbertintendedto buildup the whole science
of mathematicsfrom a system of axioms. However,before one formalizesand ax-
iomatizes,theremustbe some (meaningful)mathematicalsubstancethatcan be taken
for grantedwithoutfurtheranalysis[52, pp. 171, 190], [53, p. 65], [57, 7th ed., An-
hang 9, p. 288]. For Hilbert,the subjectof foundationsfell, generallyspeaking,into
2 (r) THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

two main branches: proof theory and metamathematics, related to formalization and
meaning, respectively [51, p. 152], [58:3, p. 179]. Proof theory was developed in order
to increase certainty and clarity in the axiomatic system, but in the end what really
mattered for Hilbert was meaning. The penultimate question will lead us finally to an
investigation of how Hilbert carried out his twofold research program on foundations
by incorporatinginto it both proof theory and metamathematics.
3. THE PREHISTORY. In the winter of 1899-1900 Hilbert, though but thirty-eight

years old one of the most respected German mathematicians of the day, was invited
to present one of the major addresses at the opening session of the forthcoming ICM
in Paris. Three years earlier, at the Zurich Congress, Henri Poincare (1854-1912) had
delivered the speech "Surles rapportsd'analyse pure et de la physique mathematique"
[On the relationships between pure analysis and mathematical physics]. Hilbert vacil-
lated between replying to Poincare and choosing another subject, for example, a col-
lection of important, open questions through which at the end of the century he could
try to sense the future of mathematics. In January,he asked for Minkowski's opinion.
His friend wrote in a letter of 5 January 1900: "Most alluring would be the attempt to
look into the future and compile a list of problems on which mathematicians should
test themselves during the coming century. With such a subject you could have people
talking about your lecture decades later" [71] (see also [87, p. 69]).
Minkowski was correct, of course, yet Hilbert continued to waiver. At the end of
March, he asked Hurwitz for his opinion on the matter.Finally, in the middle of July,
Hilbert surprisedboth Minkowski and Hurwitz with page proofs of the paper "Mathe-
matische Probleme" [37], an expanded version of his Paris talk written for publication
in the Gottinger Nachrichten. However, by that time Hilbert had already missed an
ICM deadline, so the program mailed to Congress participantsincluded an announce-
ment of neither a major lecture nor any other contribution by Hilbert himself. One
must bear in mind that at this juncture in his life Hilbert was extremely pressed for
time; he had preparedthe Paris address during a summer term in which he was obliged
to lecture ten hours a week. It is noteworthy that in one of these courses, on the theory
of surfaces, he developed the celebrated invariantintegral that would also play an es-
sential role in the twenty-third problem just a few weeks later [37, pp. 472-478] (see
as well [99, pp. 253-264]).
Both friends advised him to shorten the lecture. Hilbert agreed, presenting only ten
of the problems.
4. ON THE ROLE OF PROBLEMS. How should Hilbert's proposed problems be

characterized?Time pressures probably explain some of the points commented upon
by Grattan-Guinness(personal selection, mixture of distinct kinds of problems that are
only partially grouped, overlapping or missing problems) [30, pp. 756-757]. (Opinions
on this subject differ. See, for example, those of Pavel S. Alexandrov (1896-1982) in
his edition of the problems [37, German ed., p. 20].) Still, it was Hilbert's aim to
present "merely samples of problems" or, more precisely, problems that showed "how
rich, how manifold, and how extensive the mathematical science of to-day is" [37,
p. 478]. Moreover, it was Hilbert's conviction that a branch of science is full of life
only as long as it offers an abundance of problems: a lack of problems is a sign of
death [37, p. 438]. Eight of the twenty-three problems read more like research pro-
grams than problems as such; of the remaining fifteen problems, twelve have been
completely solved. Also, in a general sense one can regard the twenty-third problem
as a program that could not be implemented (the further development of the methods
of the calculus of variations) except from a certain vantage point as a (solved) problem
January2003] HILBERT'S TWENTY-FOURTHPROBLEM 3

(invariantintegralsin the calculusof variations).It is of interestto note thatHilbert's
notebooks[38:3, pp. 69-70] containan unpublishedentryfully anticipatingthe later
generalconceptsof sufficientconditions(field theories)in the calculusof variations
of AndreRoussel(1904-?), HermannBoerner(1906-1982), or Rolf K6ltzler(see [99,
pp. 387402]).
Using wordsthatrecallHilbert's,AndreWeil(1906-1998) once said,"Greatprob-
lems furnishthe daily breadon whichthe mathematician thrives"[105, p. 324]. As a
matterof fact,all of Hilbert'sproblemshaveservedup beautifulfood for thought.De-
spitetheirgreatimportance,however,we shouldnoton theiraccountelevateHilbertto
the statureof a prophetof futuremathematics,forhe himselfregardedsuchprophecies
as absolutelyimpossible.Indeed,quite typicalof him, he regardedthe impossibility
of propheciesas a veritableaxiom [38:3, insertedpages]. This convictionnotwith-
standing,Hilbertdaredto makesomepredictions.He consideredthe seventhproblem,
which deals with the irrationalityand transcendenceof certainnumbers,extraordi-
narilydifficultand did not expect a speedy solution,but such problemshave been
treatedsuccessfullysince the 1920s [8:1, pp. 241-268], [115, pp. 171-202]. On the
otherhand,he was quiterightin his prognosisthatFermat'sLastTheoremwouldbe
solved in the twentiethcentury[112]. In 1970, the year YurijMatiyasevichshowed
thatHilbert'stenthproblemis unsolvable[70] (see also [115, pp. 85-114]), J. Fang
wrote[18, p. 123]:"Thetenthproblem,like the eighth,is not likely to be solvedin the
nearfuture."The eighthproblem(the RiemannHypothesis),the problemthatHilbert
viewedas the mostimportantof mathematics(see [98, p. 14]), still showslittle sign of
yieldingto the intenseeffortsto settleit.
Toreturnto the subjectof Hilbert'scollectionof problemsas a whole,the historyof
mathematicsrecordsalmostnothingelse comparableto whatHilbertsingle-handedly
undertookbecause,to quote a letterof Hilbert'sstudentJohnVon Neumann(1903-
1957) to HendrikKloosterman(1900-1968), chairmanof the programcommitteeof
the 1954ICM:"Thetotalsubjectof mathematicsis clearlytoo broadfor anyone of us.
I do not thinkthatany mathematician since Gausshas coveredit uniformlyandfully,
even Hilbertdid not, and all of us are of considerablylesser width (quiteapartfrom
the questionof depth)thanHilbert"[85, p. 8]. Nevertheless,in 1900 Hilbert"dared
to chartout the most promisingavenuesfor researchin the twentiethcentury"[12].
(Duringthe MillenniumMeetingin Parisin May2000, the ClayMathematicsInstitute
(CMI)of Cambridge,Massachusetts,identifiedsevenMillenniumPrizeProblems,for
each of which it has put up a one million dollarprize for a solution.The Scientific
AdvisoryBoardof the CMIdeclaredthatthe problems"arenot intendedto shapethe
directionof mathematicsin the next century"[12].) When he receivedhis doctoral
degreein 1866, GeorgCantor(1845-1918) followedthe customof the dayby defend-
ing certaintheses thathe had advanced.(Hilbertwas four yearsold at the time.) The
thirdof thesereads[9, p. 8]: "Inmathematicsthe artof askingquestionsis morevalu-
able than solving problems."Indeed,it is preciselyby the identificationof concrete
problemsthatmathematicshas been able to andwill continueto develop.Thatis the
deeperreasonwhy Hilberttooktheriskof offeringa list of unsolvedproblems.Forthe
axiomatizationof a theoryone needsits completion.On the otherhand,for the devel-
opmentof a mathematicaltheory,one needs problems.In additionto the completion
of a theory,Hilbertinsistedon problemsandthereforeon the developmentof a theory.
In otherwords,Hilbertwas not at all the pureformalisthe is oftentakento be.
Hilbertalso explainedto his audiencewhat the natureof good problemsshould
be, his wordsechoingwhathe wroteelsewhere[38:1,p. 55]: "Theproblemsmustbe
difficultwhile being plain-not elementaryyet convoluted,becauseconfrontedwith
themwe wouldbe helpless,or we wouldneedsomeexertionof ourmemory,to bearall
4 ()? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

the assumptionsandconditionsin mind."In otherwords,the formulationof a problem
has to be shortandto the point.Hilberthimselfpossessedan uncannyabilityto make
thingssimple,to eliminatethe unnecessaryso thatthe necessarycouldbe recognized.
The last quotationin the previousparagraphis takennot from the Parisaddress
(see [37, p. 438]) butfroma parallelremarkin Hilbert'smathematical notebook,where
I also found the canceledtwenty-fourthproblem.To the best of my knowledge,the
twenty-fourthproblemhas remainedunpublisheduntilnow,andI do not knowof any
responsesto this problem,with the exceptionof Hilbert'sown. Surprisingly,although
Hilbertwas undertime pressurewhen he preparedthe Parisaddress,he made little
use of his notebookas a sourceof inspiration.The notebook(moreprecisely,the three
copybooks[38:1-3] in which Hilbertwrote down mathematicalremarks,questions,
and problemsfrom 1885 onward)is housedin the Universityof GottingenLibrary's
SpecialCollections(Handschriftenabteilung). Unfortunately, no entryin the notebook
is dated,which makesit difficultto attachan exact date to the entryconcerningthe
twenty-fourthproblem.On the basis of entries(datableto 1901) dealingwith results
in WernerBoy's dissertationandOttoBlumenthal'shabilitation[38:3,pp. 29, 33] that
appearonly a few pages afterthe one concerningthe twenty-fourthproblem[38:3,
pp. 25-26], it is very likely that Hilbertwrote the entryin which he mentionedthe
cancellationin 1901.
5. SIMPLICITY AND RIGOR. It is widely believed amongmathematiciansthat

simplicityis a reliableguidelinefor judging the beauty (see [84, sect. 4-5], [106,
chap. 6]) or elegance [113] of proofs, but like all aestheticprinciples,such a crite-
rionis highly subjective.Canone reallysay thatcertainmathematicalproofsaresim-
pler thanothers?In otherwords,the questionof what the simplestproof is depends
uponinterpretation, andinterpretation bringsnonmathematical conceptsinto play.As
a formalmeansof overcomingsuchsubjectiveaspectsof nonmathematical reasoning,
Hilbertdevelopeda prooftheorythatdeals with formulasandtheirdeduction.In this
framework,each mathematicalstatement(theorem)becomesa deducibleformula[54,
p. 137], [56, p. 489]. Onthe firstlevel, thistheoryis concernedwiththe set of provable
theorems,butin a broadersense,it also dealswiththe structureof proofs,for example,
with the conceptof the simplicityof a proof.In the emergingfieldslatercalledproof
theory and metamathematics[51, p. 152], Hilbertwantedas early as 1900 to have
a detailedinvestigationof the questionof simplicity(see also [42, lecture37]). This
furnishedthe rationalefor a twenty-fourthproblem.Still, an obviousquestionarises:
Is the formulathatis (by whatevermeans)assignedto the simplestproof of a given
theorema derivableformulain prooftheoryitself?
By 1899 geometricconceptshadalreadybeen formalizedandtheirrelativeconsis-
tency provedby Hilbert(by assumingthe consistencyof the theoryof real numbers;
see [43], [57]). But Hilbert'sultimategoal was more ambitious:to prove the con-
sistencyof mathematicsitself. The next step was to addressarithmetic(andthen, of
course,logic proper[44, p. 176]). Indeed,shortlyafterthe publicationof the Grund-
lagen der Geometrie[57] in 1899, Hilbertpresentedhis ideas on the foundationsof
arithmetic[43] in the same way thathe had treatedgeometry.Furthermore, lateredi-
tions of the Grundlagen[57] articulatedthese ideas [57, chap. 13]; from the third
throughthe seventheditionsof the Grundlagenthe arithmeticaxioms [43] appeared
as an appendix[57, Anhang6]. In the beginning,however,the necessarytaskof prov-
ing consistencyfor the proposedsystem of axioms was only indicated,not actually
executed.
This task,however,was broughtto the fore a few monthslater,as the secondprob-
lem in Hilbert'sParisaddress(see Kreisel [66, pp. 93-130]). This rapidprogressin
Hilbert'sfoundationof arithmetic(axiomatizationand the relatedquestionof con-
sistency) and logic is quite remarkable.Hilbertbelievedthat such investigationsof
formalizationcarriedout by an extensionof his axiomaticmethodwouldnot merely
be promisingbut wouldnecessarilybe successful,especiallyif pursuedby meansof
the reductionof proofsto an algebraiccalculusor "logicalarithmetic"[44], [45]. That
was still his belief in the 1920s ([48], [51]-[56]; see also [66]). Hilbertwantedto
achievehis objectiveof establishingthe consistencyof axiomatizedmathematicsby
proceedingfroma purelyfinitarystandpoint,i.e., he acceptedonly thosefactsthatcan
be expressedin a finitenumberof (elementary)symbolsandadmittedonly those op-
erationsthatcan be executedin a finitenumberof steps. (HaskellBrooksCurry,who
tookhis doctor'sdegreewithHilbertin 1929,translatedHilbert'stechnicalterm"finit"
as "finitary.")
In supportof the precedingstatement,let me quote from CharlotteAngas Scott's
interviewwith Hilbertfor herICMreport[91, pp. 67-68]: "Asto ouraim withregard
to any problem,theremustbe a definiteresultof some kind;it cannotbe laid aside
untilwe haveobtainedeithera satisfactorysolutionor a rigorousdemonstration of the
impossibilityof a solution.The mathematical rigorthatis essentialin the treatmentof
a problemdoes not requirecomplicateddemonstrations; it requiresonly thatthe result
be obtainedby a finite numberof logical steps from a finite numberof hypotheses
furnishedby the problemitself;in seekingthis rigorwe may findsimplicity."
In his speech,Hilbertdid not deal with provingthe simplicityof proofsfrom the
finitaryperspectiveas fully as he didin theinterviewwithScottjustcited.Hereis what
he saidto his Parisaudience[37, p. 441]: "Besidesit is an errorto believethatrigorin
the proofis the enemyof simplicity.On the contrarywe findit confirmed... thatthe
rigorousmethodis at the same time the simplerand the moreeasily comprehended.
The very effortfor rigorforces us to find out simplermethodsof proof."At the end
of herreport,Scottthankedall the speakersfor theirassistance.Thisacknowledgment
lends credenceto the suggestionthatHilberthad explainedhis idea to her in greater
detail.The firstpartof the quotation,the expressionof a belief in the solvabilityof
each well-posedproblem,mightflatlybe called the Hilbertaxiom(see [37, p. 445]).
In the secondpartof the quotation,however,the role of simplicity(fromthe finitary
pointof view) is clearlyemphasized,muchmoreclearlythanin the lecture.
The questionas to what constitutesa rigorousproof is a logical question.By ex-
aminingthe languagein which the proof is expressedone may ask: What are the
conditionsunderwhicha strictlogicaldeductionproceedingby a long chainof formal
inferencesandcalculationsandleadingfromlink to link by blindcalculationscan be
regardedas simple?As earlyas 1900,Halstedreportedthatin Hilbert'sopinion"math-
ematicalrigorwhichwe requiredoes not necessitatecomplicateddemonstrations; the
mostrigorousmethodis oftenthe simplestandtheeasiestto comprehend" [34, p. 189].
To Hilbert,the qualitiesof rigorandsimplicityarenot at odds with each otherbutgo
handin hand.In 1943,in Hilbert'sobituary,Weylechoedthatsentiment:"WithHilbert
rigorfiguresno longeras enemybutas promoterof simplicity"[109:4,p. 124].
Naturally,it has been an open questionever since Hilbert'stime how or to what
extentinvestigationsof simplicitymightbe carriedout,if theycanbe carriedout at all.
6. THE PLACE OF THE TWENTY-FOURTHPROBLEM IN PROOF THE-

ORY. Whydid Hilbertcancel the questionof simplicity,as formulatedin the twenty-
fourthproblem?We do not have the necessaryhistoricalsourcesto give a definitive
answerto this question,but we can make some conjecturesand presentevidencein
supportof them.Moreover,we can look at the progressof Hilbert'sprooftheoryand
metamathematics fromthe perspectiveof the canceledproblem.
6 (? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

I do not think that the main reason was time pressure; rather I believe it was the
status nascendi of proof theory. In Hilbert's actual speech, the three problems offoun-
dation (the first, second, and sixth problems in [37]) that he chose to include among
the ten presented were already, from his perspective, an acceptable representation of
this branch of mathematics, for in the printed version these three were not supple-
mented by additional problems of this genre. Moreover, one of the three problems on
foundations is concerned with the consistency of the axioms of arithmetic (the sec-
ond problem; see [66]), which later played a prominent role in Hilbert's program [115,
pp. 37-58]. In conjunction with the twenty-fourth problem, this shows that, even at the
time, Hilbert had at least a vague idea of metamathematics. Also, Hilbert had already
included a decision theory problem on his list: the tenth problem seeks an algorithm
to determine whether or not an arbitrarypolynomial with integer coefficients has an
integer root. (As mentioned earlier, in 1970 Matiyasevich established the nonexistence
of such an algorithm [70], [8:2, pp. 323-378], [115, pp. 85-114].)
In his 1917 Zurich talk "Axiomatisches Denken" [Axiomatic thinking], Hilbert sur-
veyed the role of axiomatization and tried to turn the attention of mathematicians to
the study of proofs (see [48]). He pointed out clearly that the problem of simplicity
is among the significant questions of foundations and is closely related to tasks faced
in the realm of proving consistency: "When we consider the matter [axiomatization of
logic] more closely we soon recognize that the question of consistency of the integers
and of sets is not one that stands alone, but that it belongs to a vast domain of difficult
epistemological questions which have a specifically mathematicaltint: for example (to
characterizethis domain of questions briefly) the problem of solvability in principle of
every mathematical question [which we have earlier called the "Hilbert axiom"], the
problem of the subsequent checkability of the results of a mathematical investigation,
the question of a criterion of the simplicity for mathematicalproofs, the question of the
relationship between content andformalism [Inhaltlichkeitund Formalismus] in math-
ematics and logic, and finally the problem of decidability of a mathematical question
by a finite number of operations. We cannot rest content with the axiomatization of
logic until all questions of this sort and their interconnections have been understood
and cleared up" ([48, p. 412], [58:3, p. 153], English translation in [17, p. 1113]; see
also [18, p. 195]). Furthermore,near the end of his life, Hilbert created an index for
his notebooks by inserting one extra page on which the problem of simplicity appears
among the few key words [38:3, inserted page]. Indeed, the problem did not let go of
him.
7. A LOOK AT HIIRERT'S FOUNDATIONS OF MATHEMATICS. From Hil-

bert's point of view, any part of mathematics can be representedby a deductive system
based on a finite set of axioms. Hilbert's idea was to axiomatize the branches of math-
ematics and then to investigate the consistency of their axioms, especially those of
arithmetic, which underlies all other branches. For this purpose, one must first for-
malize the system (by means of formal logic) in order to make rigorous derivations
possible.
In Hilbert's formalistic view, mathematics is to be replaced by mechanical deriva-
tions of formulas, without any reasoning concerning their specific content. Recall the
words of Griffith [32, p. 3]: "A mathematical proof is a formal and logical line of
reasoning that begins with a set of axioms and moves through logical steps to a con-
clusion .... A proof confirms truth for the mathematics." In such a formal system,
in proof theory, the subject of research is the mathematical proof itself [48, p. 413],
[58:3, p. 155]. To master this subject (in the object language), one must control the
field of proofs (in a metalanguage). However, since one cannot be universal and can-
not knoweverythingthatis to be knownof all proofs,accordingto Hilbert'sideasone
must developproof theorycompletelyfrom thefinitaryviewpointand its intuitively
convincingmethods(see [60:1,secs. 2c-e]). Withinthis kindof framework(finitism),
he hopedto be ableto reduceanybranchof mathematicsto a systemthatrestson more
solid ground.In any areaof mathematics,therearequestionsthatby theirvery nature
referto the infinite.This fact notwithstanding, Hilbertneverspecifiedpreciselywhat
the finitary,or for thatmatterwhatthe simplest,proofswere. Clearly,for Hilbert,it
had to be possible to convertany proof of a mathematicaltheoreminto a finite one.
Hilbertwas probablyled by certainexamples(for instance,syzygies in the algebraic
calculus[40], [42, lectures47-48]) to the ideaof finite(oreven simplest)proof,buthe
hadnot yet thoughtto extendtheseparticularresultsandto expressthemin a general
formas a problem.The idea was still too vague.Nevertheless,since Hilbert'stimethe
questionhas persisted:Whatis simplicity?
Threeyears afterthe Parisaddress,Hilbertgave a talk in Gottingenon the foun-
dationsof arithmeticin which he promisedto workout detailsof the axiomaticap-
proach.He triedto accomplishthisin his Heidelberglecture[44] at the 1904ICMand
in the course"LogischePrinzipiendes mathematischen Denkens"[Logicalprinciples
of mathematicalthought](see [45], [46]) thathe taughtafterthe Heidelbergcongress
at the Universityof Gottingen[81, pp. 92-94, 98-101]. Strictlyformalisticreasoning
cannotprovethe consistencyof an axiomaticsystem,becausethe necessaryformulas
for provingconsistencyare not derivablewithinthis system (G6del'ssecondincom-
pletenesstheorem[26]). In a differentsense,in his metamathematics, Hilbertintended
to establishsuch conceptsas consistencyby intuitivejustifications,i.e., by appealing
to self-evidentprinciplesof reasoningandby directmethodsinvolvingpurelyfinitary,
combinatorialinferences.Metamathematics mustrestrictitself to countingbeans, as
HansFreudenthal(1905-1990) sarcasticallycharacterized this attitude[20, p. 1056].
After 1904, Hilbert'sresearchtook an unexpectedturn,and a thirteen-yearbreak
from the circle of ideas underdiscussionhere ensued.These were the years during
whichHilbertwas developingthe theoryof integralequations(see the surveyof Ernst
Hellinger(1883-1950) in Hilbert[58:3, pp. 94-145] and [88, pp. 117-129]). In the
latteryearsof this period,WorldWarI took its toll on his abilityto conductresearch.
In the aforementioned lectureon axiomaticthinkingdeliveredin Zurichin 1917,how-
ever, Hilbertpicked up the problemof simplicityagain and presentedthe same ex-
ampleswe know from the twenty-fourthproblemin the notebook[38:3, pp. 25-26].
He later based an article,"AxiomatischesDenken"[48], on the Zurichtalk. Arend
Heyting(1898-1980) regardedthe discussionof simplicityin this articleas the most
important,becauseit demonstrated thatHilbertviewedproofsas mathematical objects
in themselves:"Init we perceivethe germ of the later 'ProofTheory'" [36, p. 36].
(See also [48, pp. 412-413], [58:3, p. 153].) As notedearlier,Hilbertheld this view
as earlyas 1900 (see [42, lecture37]). In the WinterTermof 1917, he returnedto it,
deliveringa course"Principiender Mathematik"[Principlesof mathematics],which
was workedout by PaulBerays (1888-1977) [47] andwhichfinallyled to the book
GrundlagenderMathematik[60]. Thebasicidea of Hilbert'sprooftheoryis to ensure
the validityof argumentsby reducingmathematicsto a finitenumberof rules of in-
ferencethatgovernthe manipulationof formulas.The set of derivableformulasis, of
course,infinite,butit is "generated" by a finitesubsetof formulas.Is suchan extended
set consistent?Orone mightask generally:Is this prooftheoryitself consistent?
In the 1920s, duringwhich quantumtheorywas rapidlyemerging,Hilbert,too,
was occupiedwith the physicaland mathematicalproblemsof quantummechanics.
Still, in this periodhe launcheda majoreffort,the so-calledHilbertprogram,to prove
that mathematicsis consistent (for the mathematicalstatements,not for Hilbert's
8 )? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

philosophical opinion, see the 1922 essay "Neubegriindung der Mathematik" [New
foundations of mathematics] [49]). The theory (metamathematics) that supported this
formalism was more or less completed by Hilbert in a 1922 talk, "Logische Grundla-
gen der Mathematik"[Logical foundations of mathematics], in Leipzig [51], although
a more readable version is his 1928 paper "Grundlagen der Mathematik" [53]. In
his 1925 lecture "Uber das Unendliche" [On the infinite] [52], Hilbert expressed his
philosophical thoughts in their mature form. Among other things, in an epistemolog-
ical aside, he mentioned that in mathematics one finds the philosophical concept of
"ideal propositions" (in Kant: regulative device [regulatives Prinzip]) as counterparts
of the real or "finitepropositions" (in Kant: constitutive device [konstitutivesPrinzip]).
Such ideal elements do not correspond to anything in reality but serve as regulators
in Kant's terminology "if ... one understands as an idea a concept of reason which
transcends all experience and by means of which the concrete is to be completed into
a totality" [52, p. 190]. The reliability of such ideal methods is to be established by
finitary means.
Thus, only in his early sixties did Hilbert truly proceed to create proof theory and
metamathematics.In this proof theory, he developed his formalistic standpointin detail
and dealt with the question of whether or not a formula can be deduced from the
axioms that define a system. Hilbert regarded any derivable formula as an "image"
of mental activity, for he was convinced that each mode of thought is accompanied
by such symbols [44, p. 176]; he expressly formulated this conviction in the "axiom
of thinking" [45, p. 119]. For Hilbert, both nature and human reflections on nature
were finite. Therefore, Hilbert's program rested on the 'finite.' Relying upon a purely
finite approach, in a system in which the axioms were regarded only as hypotheses
for a theory and not as self-evident mathematical truths (see [57, chap. 1, sec. 1],
[44], and [46, p. 141, 186]), Hilbert wanted to show that the essential mathematical
methods could never lead to contradictions [52, pp. 162, 164]. The kind of formal
axiomatic method regarded instrumentally by Hilbert is not in itself sufficient for the
foundations of mathematics; for such a purpose, the axiomatization must be extended
beyond its formal viewpoint. Meaning (inhaltliche Mathematik) is then introduced at
the metamathematicallevel, in the metalanguage. In the end, the details of his program
remained to be worked out (for the most refined presentation, see [51]).
Hilbert's aim was to secure meaning by using finite concepts. We mention Zer-
melo's credo as a conviction that is the antithesis of Hilbert's. The following artic-
ulation of Zermelo's leading ideas is taken from his five "theses about the infinite
in mathematics" (1921): "Each genuine mathematical proposition has an 'infinitary'
character,i.e. it... has to be viewed as a combination of infinitely many 'elementary'
sentences. The infinite is neither physically nor psychologically given to us in the real
world. It has to be comprehended and 'posited' as an idea in the Platonic sense. Since
infinitarypropositions can never be derived from finitary ones, also the 'axioms' of all
mathematical theories have to be infinitary and the 'consistency' of such a theory can
only be proved by exhibiting a corresponding consistent system of infinitely many ele-
mentary sentences" (Nachlass Zermelo, quoted in [15, pp. 148, 158]). For more about
Zermelo's prejudice against a finitary characterof genuine mathematical propositions,
see [29], [116]. In light of the fact that in his former G6ttingen days Zermelo was on
common ground with Hilbert [80, pp. 5, 118-122], it is indeed striking how widely
divergent Hilbert's and Zermelo's opinions on the foundation of mathematics became.
8. THE HIITRERT AXIOM: SOLVABILITY OF EVERY PROBLEM. Modem

formalism descends from Hilbert's theory but has evolved into something quite differ-
ent from it. Metamathematics, a term coined by Hilbert, has over the years come to be

regardedas a virtualsynonymfor prooftheory.(PaulBerays alreadyequatedthe two
in 1934 [60].) "Prooftheory"is now primarilya namefor the studyof formalmodels
of mathematicalsystems.Hilbert'soptimisticbelief, the Hilbertaxiom [37, p. 445]
thathe hadpreservedfromhis universitydays, is summedup by the quotation[38:3,
p. 95]: "Theproofof all proofs:thatit mustalwaysbe possibleto arriveat a proof."
This standsin remarkablecontrastto the fact thatin the last decadesof the nineteenth
century,especiallyduringthe time Hilbertwas a student,pessimismwas widespread.
In 1872, the well-knownphysiologistand physicistEmil Du Bois-Reymond(1818-
1896) delivereda famousspeechin Leipzig,"Uberdie GrenzenderNaturerkenntnis"
[Onthe limitationsof knowledgein the naturalsciences],thatwas widely read(eight
editionsof the talkhadappearedby 1898).Du Bois-Reymond,a brotherof the mathe-
maticianPaulDu Bois-Reymond(1831-1889), maintainedthatcertainproblemswere
unsolvable,amongthemthe naturesof matterand force or the originof motion.He
concludedthe lecturewith the thenoft-quotedagnosticcatchwords"Ignoramus et ig-
norabimus"[Weareignorantandwe shall(always)be ignorant].
Up untilthe end of his career,Hilbertcontinuedto rejectagainandagainthe "fool-
ish ignorabimus"of Du Bois-Reymondand his successors(for Hilbert'slast lecture,
deliveredin 1941, see [73, p. 71]). In his notebook,Hilbertphraseshis "Noscemus"
[Weshallknow]this way [38:3,p. 104]:"Thatthereis no ignorabimusin mathematics
can probablybe provedby my theoryof logical arithmetic."Elsewherehe contin-
ued [38:3, p. 98]: "Maybeit will turnout that thereis no sense in saying thereare
insolubleproblems."In K6nigsberg,on September8, 1930,in his famousspeech"Na-
turerkennen undLogik"[Naturalphilosophyandlogic] [55], Hilbertpointedout that
the positivistphilosopherAugusteComte(1798-1857) hadonce assertedthatscience
wouldneversucceedin ascertainingthe secretof the chemicalcompositionof thebod-
ies in the universe.Then,preparinghis optimisticconcludingwords,Hilbertwent on
to say:"Afew yearslaterthis problemwas solved"[55, p. 963] (see also [87, p. 196]).
Indeed,in the courseof timeHilbert'sgroup,the formalists,achievedlargepartsof
theirgoals in the foundationsof mathematics(WilhelmAckermann,JohnVon Neu-
mann,see [54, p. 137];PaulBerays [60:1,p. v]). It seemedthenthatthe proofof the
consistencyof numbertheorywas moreor less a matterof findingthe propermathe-
maticaltechnique.Consequently,atthe 1928ICMin Bologna,Hilbert'stalkconcerned
problemsof foundations[54, pp. 139-140]. In his program,Hilbertaddedto the old
problemof consistencythe new problemof the completenessof formalsystems(first
mentionedin [57, introduction],thenin the secondproblemof the Parisaddress[37],
and also in [45, chap. 1, sec. 3]). Hans Hahn(1879-1934) communicatedHilbert's
extendedprogramto the ViennaCircle,andin 1930 a youngAustrianmathematician
namedKurtGodel(1906-1978) demonstrated in his Ph.D. dissertation[24] thatfirst-
orderpredicatelogic is complete,i.e., everytruestatementin it can be derivedfrom
its axioms(see [93]).
Even the thingsone most confidentlyexpectsdo not alwayscome to pass. In one
of the great ironies in the historyof mathematics,it was coincidentwith Hilbert's
great 1930 speech in Konigsbergthat the same mathematicianGodel again entered
the scene, unnoticedby Hilbert;Godel was in Konigsbergattendinga philosophi-
cal congress [93]. Indeed,it was only in the discussionthat took place on Septem-
ber 7 (one day beforeHilbert'sfamousspeech)thatGodel made an offhandremark
on a workin progress,whichis now knownas Godel'sincompletenesstheorem([33,
pp. 147-148], see also [25]). Alreadyin November1930 the LeipzigjournalMonats-
heftefir Mathematikund Physikreceivedthe epoch-making,25-page article"Uber
formalunentscheidbare Satze"[Onformallyundecidablepropositions][26], in which
the 25-year-oldauthorprovedstrikingresultsin a way thatHilberthadnot anticipated.

To be specific, Godel gave negative answers to the remaining problems of foun-
dations posed by Hilbert at the 1928 ICM [54]. Over Hilbert's great expectations fell
the shadow of harsh reality represented by Godel's results: (axiomatic) mathematical
knowledge is always imperfect; we cannot prove all that is to be known. (In 1781,
Konigsberg's other great son Immanuel Kant (1724-1804) had stated this result philo-
sophically in The Critique of Pure Reason: "What the things-in-themselves [Dinge an
sich] may [finally] be I do not know, nor do I need to know, since a thing can never
come before me except in appearance" [Critique, version A, p. 276; trans. N. Kemp
Smith].) G6del's results confirmed Cantor's belief that there are no foundations of
mathematics without metaphysics, i.e., without infinite methods. And thus Hilbert's
prediction, the sketched but never completely established finitary program, met the
fate that the Bible so eloquently ordains: "The wind bloweth where it listeth, and thou
hearest the sound thereof, but canst not tell whence it cometh, and whither it goeth: so
is every one that is born of the spirit" (John 3:8).
If we wished to continue on this theme, we would soon encounter the names of
Ernst Zermelo (1871-1953), Thoralf Skolem (1887-1963), Adolf Fraenkel (1891-
1965), Rudolf Carnap (1891-1970), Alonzo Church (1903-1995), Stephen Cole
Kleene (1909-1994), Alan Turing (1912-1954), and others. That, however, would
be another story (see Jean-Yves Girard's contribution to [82:2, pp. 515-545]). Hans
Freudenthalprovided a rathersad postscript to the foregoing discussion [20, p. 1057]:
"At a closer look, 1931 is not the turning point but the startingpoint of foundations of
mathematics as it has developed since. But then Hilbert can hardly be counted among
the predecessors." Still, of those who insisted on the importance of mathematical
proofs themselves and regarded proofs as mathematical objects, Hilbert obviously
must be counted among the first.
Furthermore, Hilbert's work on metamathematics has greatly improved our un-
derstanding of the nature of mathematical reasoning. Despite the fact that Hilbert's
programwas largely discredited by Godel's theorems, Hilbert's ideas concerning foun-
dations are not without value for certain areas. In a letter to Constance Reid dated
March 1966, Godel himself underscored this point: "I would like to call your attention
to a frequently neglected point, namely the fact that Hilbert's scheme for the founda-
tion of mathematics remains highly interesting and important in spite of my negative
results. What has been proved is only that the specific epistemological objective which
Hilbert had in mind cannot be obtained .... However, viewing the situation from a
purely mathematical point of view, consistency proofs on the basis of suitably chosen
stronger metamathematical presuppositions (as have been given by GerhardGentzen
(1909-1945) and others) are just as interesting, and they lead to highly important in-
sights into the proof theoretic structures of mathematics .... As far as my negative
results are concerned, apartfrom the philosophical consequences mentioned before, I
would see their importance primarily in the fact that in many cases they make it pos-
sible to judge, or to guess, whether some specific part of Hilbert's program can be
carried through on the basis of given metamathematical presuppositions" ([28]; see
also [87, pp. 217-218]).
Indeed, there is a surprising amount of work that can be carried out along the lines
of the Hilbert program. In some sense, part of it was already done by Hilbert and his
collaborators,particularlyWilhelm Ackermann (1896-1962) and Paul Berays (1888-
1977). The first major variation of Hilbert's program was due to Gerhard Gentzen,
who in the 1934 paper "Die Widerspruchsfreiheitder Zahlentheorie" [21] established
the consistency of number theory (see [72]), which Ackermann did independently in
1940 [2]. In accordance with Godel's results, such investigations must resort to princi-
ples that lie outside pure number theory. For example, transfiniteinduction is used, but
apartfromthis "transcendental"element,the proofsarecarriedout completelywithin
the frameworkof numbertheory.(Noteworthyin thisconnectionis the factestablished
by ChurchthatG6del'sresultscannotbe obtainedby finitemeans[11].)
In the axiomaticapproach,the "tree"of all possiblemathematicalformulasgrow-
ing from axioms is not only extremelyexpansivebut, as Godel pointedout, even
disconnected.However,froma finitarystandpoint,therestrictionto "meaningfulques-
tions" (the Hilberttree) would lead to a "humanmathematics"(see the interview
withMikhaelGromovin [82:2,pp. 1213]).True,G6del'sresultsdictatecertainlimits
to Hilbert'sfoundationsof mathematics,butthereremainsmuchof valuein continu-
ing Hilbert-likeprograms.Togive one importantexample:nonstandard analysis,in the
spiritof AbrahamRobinson(1918-1974), is a new branchof mathematicsthatgrew
out of model theoryaround1960 andowes muchto the ideas of Hilbert.In addition,
the adventof computerscienceled to a rebirthof Hilbert'sprooftheory.
9. SIMPLICITYVERSUS COMPLEXITY.Hilbertwas not alonein his desirefor

maximalsimplicityin mathematical proofs.Thisissue wasof importance,forexample,
to the FrenchmathematicianEmileLemoine(1840-1912), who showedgreatinterest
in simplifyinggeometricconstructions.In 1888, Lemoinereducedall geometriccon-
structionsby rulerand compassto five basic operations.One of them, for instance,
was simply placingan end of the compassat a given point. Lemoinequantifiedthe
simplicity of a construction as the total number of times these five basic operations
wereusedin it. In this way,he was able assigna numericalvalueto the complexityof
a geometricconstruction[68].
It is probablethatin some analogousway Hilbertwantedto makeproofsin gen-
eral a measurableobjectof anothertheory("logicalarithmetic")in which only finite
methodswereacceptable.Thereductionof proofsto an algebraiccalculusby meansof
formallogic wouldallowone to decidewhichof two givenproofsis simplermerelyby
comparingthe numberof operationsinvolvedin each. In invarianttheory[1], Hilbert
hadalreadytoucheduponthe questionof how to expressrelationsin termsof a finite
basis (see also his 1897 lecture[42]). Fromsuch a viewpoint,it wouldseem possible
to arrangemathematicalproofs into stratacharacterizedby theirdegreeof simplic-
ity. Mathematicswouldthen assumethe appearanceof a neatlyorganizedwarehouse
in which formulaswould be stackedin hierarchicalorder:the lower the stratum,the
simplerthe proof.
As to the complexityof technicaldetailsforproofsthatsit in thismathematicalstor-
age facility,I wouldremindthe readerof fourwell-knownproblems.The proofof the
Burnsideconjectures(see [61, p. 106]) occupies aboutone thousandpages, while a
completeproof of Ramanujan'sconjecturesis estimatedto requireat least two thou-
sandpages. Nevertheless,such proofs,includingthose of Fermat'sLast Theoremby
AndrewWiles [97], [112] or of the "Four-colorProblem"by Appel and Haken[3],
are examplesof the kind of finitemathematicsthatHilberthopedto establishby his
program(see [101]). Wiles's proof can be carriedout by hand;by contrast,no proof
unaidedby technologyis yet knownfor the four-colorproblem.Of course,in gen-
eral,computer-aided proofshavetoo manycases for any humanbeing to check them
step by step.However,even for finiteproofscreatedsolely by the humanbrain,there
remainsHilbert'spracticalquestion[38:1,p. 53]: "whetherin mathematicsproblems
exist that cannotbe dealt with in a prescribedshorttime?"His example(see [38:2,
p. 1], [38:3, insertedindex];see also [48, p. 414] or [58:3, p. 155]):computethe nth
digit in the decimalexpansionof 7r,wheren is equalto 10 raisedto the 10thpower
to the 10th power, i.e., n = 11010. (In the theory of invariants one is faced with sim-
ilar questions.Hilbert'sfirstpaperssettledthe finitenessquestiononly in principle,
12 2? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

without any indication that we can actually calculate certain numbers; [42, lecture 37],
[74], [89].)
The solvability of this problem in finitely many operations is evident, but from
an epistemological viewpoint such computing tasks, which have been undertakenby
Yasumasa Kanada, Takahashi,Fabrice Bellard, the brothersGregory and David Chud-
novsky, and others and which have provided millions of digits of Xr (as of April 1999
the number was up to 68 billion), are problematic. Imagine a supercomputer of the
largest size possible, that of the universe, and imagine also that this computer has been
calculating since the Big Bang. On the basis of the standardmodel of cosmology it is
obvious that the number N of steps that this computer would have been able to exe-
cute must be finite. (It has been assumed that the total number of long-lived particles
(electrons and protons) in the universe is about 1080 (Sir Arthur Eddington, 1931).
This empirical assumption provides the maximal number of components of the central
processing unit. On the other hand, quantummechanics requires a minimal time for an
operation. Coupled with the elapsed time due to the cosmological standardmodel the
total number N of operations this computer would be able to carry out is somewhere
between 10120 and 10160;see [22, pp. 44-46], [104, pp. 64-65]). For the given n (or,
more generally, for any given natural number), one can choose a natural number g
for which ng > N. Even at the rate of one thousand operations for the calculation of
one digit of Jr, the "Jr-hunters"will likely remain below 1015for the foreseeable fu-
ture. The upshot of this discussion is that there is and will always be a limit to our
knowledge; in the end, it is the complexity of the required operations that determines
our access to the (presumed) realm of Platonic knowledge. (Incidentally, dealing with
the prime-number theorem G. H. Hardy (1877-1947) looked upon n34, the Skewes
number, as the largest number in mathematics of any practical significance.)
Therefore, the complexity of computations (or, more precisely, of algorithms) is of
the utmost importance (see the concise survey [90]). Suppose that a problem can be
solved by means of any of several algorithms. For various reasons, it might be de-
sirable to compare the complexities of the algorithms at hand. Such complexities are
formulated in terms of Turing machines. Among the classes of complexity there are
those of practical interest that can be solved by a deterministic and a nondeterminis-
tic Turing machine, respectively, whose time complexity is bounded by a polynomial
time function (P-problem and NP-problem, respectively). Obviously, the class of NP-
problems contains that of P-problems. In 1971, Stephen Cook formulated the P versus
NP problem (now among the Millennium Prize Problems [12]): N = NP? The strong
connections to computer science become obvious here. This rapidly increasing field of
research has developed naturallinks with the search for the most effective and shortest
algorithms, an endeavor with clear ties to Hilbert's vision of proof simplicity (see [66],
[83], [116]).
One might ask: How many proofs do mathematicians publish each year? A back-
of-the-envelope calculation yields a rough approximation:multiplying the number of
journals by the number of yearly issues by the number of papers per issue by the aver-
age number of theorems per paper, someone has arrivedat an estimated lower bound of
two hundred thousand theorems a year! Who could conceivably judge which of them
are established via the simplest possible proof? Nevertheless, we read in Hilbert's note-
book: "All our effort, investigation, and thinking is based on the belief that there can be
but one valid view" [38:3, p. 96]. That means also there must be a simplest proof. In a
more colloquial spirit, Hilbert added [38:3, inserted page]: "Apply always the strictest
proof! Philological-historic import must be wiped out. Given 15-inch guns, we don't
shoot with the crossbow." He resorted to the same metaphor a second time with a ref-
erence to the Franco-Prussian war (1870-1871) [38:2, p. 99]: "We did not go to war
against France with bows and arrows, although they too might have produced their
effect."
Regarding the diversity of proofs, I point to the history of the fundamental the-
orem of algebra as a celebrated example. There are two basic ideas: for an algebraic
proof, the approachespoused by LeonhardEuler (1707-1783), and for an analytic one,
the mode of attacked favored by Jean-Baptist Le Rond D'Alembert (1717-1783). Of
course, there is the research of Carl FriedrichGauss (1777-1855) that forms a link be-
tween the two. Despite his motto "Pauca, sed matura"[Few, but ripe], Gauss returned
to this subject many times and altogether gave four proofs of this importanttheorem.
For Gauss, the simpler and more elegant he could make the proof, the better.
Frequently, a mathematical theorem is regarded as "deep" if its proof is difficult.
The opposite of "deep" is "trivial,"a term suggesting that little or no proof is neces-
sary.Nevertheless, there is a constant movement in mathematics toward simplification,
toward finding ways of looking at a matter from an easier, more "trivial,"and hope-
fully more revealing vantage point. The simplicity of a proof depends on a multitude
of factors: the length of its presentation,the techniques used, one's familiarity with the
concepts involved, the proof's abstract generality, the novelty of ideas, and so forth.
We read these words in Hilbert's notebook [38:3, p. 101]: "Always endeavor to make a
proof with the least elementary means, for that way mastery of the subject comes best
to the fore (the opposite of Weierstrass and Kronecker and their imitators). 'Elemen-
tary' is the designation only for what is known and familiar."
As to the mathematician's temptation to undertake generalization for generaliza-
tion's sake, Hilbert declared [38:1, p. 45]: "The mathematicians' function should be to
simplify the intricate. Instead they do just the opposite, and complicate what is sim-
ple, and call it 'generalizing'. Even if a method or an elaboration achieves no more
than half, yet is two times simpler, I find that a great advantage." A simpler proof
makes us wiser; the simpler its premises and deductions are, the more convincing a
theory is. A proof is the most straightforwardway to justify mathematical reasoning.
"A mathematicalproof,"to quote G. H. Hardy,"should resemble a simple and clear-cut
constellation, not a scattered cluster in the Milky Way" [35, p. 113].
In contrast to such views, Solomon Feferman remarked [19, p. 20]: "A proof be-
comes a proof after the social act of 'accepting it as a proof.'" The social perspec-
tive thus becomes relevant to the informal concept of the simplest proof. From this
viewpoint complicated and long proofs, having features that might hamper or retard
acceptance, would fail the test of simplicity. A famous historical example is the work
of GirardDesargues (1591-1661) that was not well received in his time because De-
sargues invented too many "strange"new technical terms.
10. THE BACKGROUND OF THE TWENTY-FOURTH PROBLEM. Instead of

proving mathematical theorems, we can examine deductive systems themselves and
explore their properties. In this setting we investigate proofs themselves as mathemat-
ical objects. In order to be able to manipulate objects, however, we must first learn to
distinguish them. This process begins by attributingcertain properties (e.g., simplicity,
complexity, shortness) to proofs as distinguishing markers.
It is evident that the question of simplicity is connected with the internal structureof
such theories, extending all the way back to the choice of axioms and their particular
formulation. In other words, the simplicity of axiom systems is also involved in the
twenty-fourth problem. Here again, one is confronted by subjective viewpoints that
lead into controversial areas of the foundations of mathematics.
However, we ought to remember that in no way is the equivalence of axiom sys-
tems affected by such requirements of efficiency, nor do we intend to suggest that
14 () THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

well-constructedaxiom systemswith "shortestaxioms"(whateverthat might mean)
will automaticallygive rise to the shortestor simplestproofs (see [95]). Although
some researchhas been done with a view towardsimplifyingcertainaxiomaticsys-
tems (aboveall by Lesniiewski(1886-1936) andhis school), the problemhas not yet
been fully discussed.(I am indebtedto FredRickeyfor bringingthese developments
to my attention.)
The finalgoal of presentinga branchof mathematicsis to expressit as a formalized
system.However,the mathematicalresearchthatprecedesthe creationof suchformal
systemshas need of its problems(for research)andexamples(for representation). As
we emphasizedearlier,importantproblemsappearedto Hilbert"as the life nerveof
mathematics"[109:4,p. 123]. It is of consequencethathe alwaysbegana projectwith
clearbut extremelysimple examplesand thatvery specialresultsled him eventually
to generalideas.As a case in point,he usuallystartedlectureson ordinarydifferential
equationswith a thoroughinvestigationof the elementarybutinstructiveexamples
y"(x) = O, y"(x) + y(x) = 0
[87, p. 104] (see also [41, p. 1 (insertion)]).However,"Thedetailedwork will not

receivethe highestsacredfireunlessthe look is turnedto generalityandto the under-
standingof foundations"[38:3,insertion].
So, it is not surprisingthatthe twenty-fourthproblemconsistsof a generalprogram
thatis illustratedby two specificexamples.ForHilbert,the simplestmathematicalop-
erationwas addition,so he used this operationin one of the examplesto motivatethe
twenty-fourthproblem.To each geometricor logical process,therecorrespondedfor
Hilbertan "addingtogether."In this way, he was able to formalizecertaingeomet-
ric conceptsarithmetically,especiallycalculationswith straightlines (Streckenrech-
nung) [57, sec. 15, 28]. Hilbertwent into some detailaboutthis subjectin his address
at the HeidelbergICM[44]. He also investigatedthe possibilityof correspondences of
this kindfor geometricconstructions."Thegeometricalfiguresaregraphicformulas,"
he said in his Paristalk [37, p. 443]. Thereis no doubtthatcertainconstructionsor
proofs rely completelyon finite processes,i.e., their truthcan be provedin a finite
numberof steps.And it was exactlythis finitarypointof view thatHilbertadvocated.
"Simplicity... is simplicityof ideas,not simplicityof a mechanicalsortthatcanbe
measuredby countingequationsor symbols,"declaredthe Nobel Laureatephysicist
Steven Weinberg[106, p. 107]. He went on to say: "Anysymmetryprincipleis at
the same time a principleof simplicity"[106, p. 110]. Weinberg'srejectionof mere
countingdoes not clash with Hilbert'sfinite notion of simplicity;rather,it supports
Hilbert'sbeliefbecauseHilbertwas deeplyconvincedthatproofsas shownin Riemann
(1826-1866) arebetterachievedthroughideasthanthroughlong calculations(thiswas
Hilbert's"Riemannprinciple").Weinberg'sstatementthat "Wedemanda simplicity
andrigidityin ourprinciplesbeforewe arewillingto takethemseriously"[106,p. 118]
coincides exactly with Hilbert'sintentionto use the simplestpossible, yet rigorous
concept.
In the end, Hilbert'saim was to justify classical mathematicsby finite methods.
"Topreservethe simpleformalrules of Aristotelianlogic [in light of finitemethods]
we must supplementthe finitarystatementswith ideal statements,"Hilberthad de-
claredin a lecture"Onthe infinite"deliveredin Minster on June4, 1925 [52, p. 174]
(see also [87, p. 177]). However,such "adjoint"[adjungierte]ideal statements(for
example,the existenceof the infinite,of the continuum,and of ideals in algebra)ob-
viously departfrom the finite viewpoint.Nevertheless,such extendedfinite systems
are indispensablefor the developmentof mathematics.But such an extensionof a fi-
nite domainthatwas takento consistof meaningfultruepropositionsof mathematics
and theirjustifyingproofs (termed"realmathematics" by Hilbert)is legitimate,pro-
vided this extension(termed"ideal"by Hilbert)does not cause any contradictions.
Suchaccommodation was strictlyrejectedby LeopoldKronecker(1823-1891), andin
the beginningHilbertrefusedto acceptKronecker'sviews [38:1,p. 53, 79, 91]. From
Hilbert'spoint of view any branchof mathematicsdealingwith ideal conceptscould
be accepted,as long as therewas a proofthatsuchan extendedsystemwas consistent.
Hilbert'sjustificationrestedon a divisionof mathematicsinto two parts:the real
mathematicsto be regulatedand the ideal mathematicsservingas regulator.There-
fore, in Hilbert'sreasoning,the questionof how to establishthe reliabilityof the ideal
methods(theregulators)by finitemeansis given greatweight,andthe consistencyof
extendedsystemsprovesto be an item of centralimportance.Hilbertattemptedto se-
curethe ideal (i.e., infinitary)partsof mathematicsby formalizingthem,thencalling
for a proof of theirreliability.He believedthatthis could be done by finitarymeans
alone [52, pp. 170-171]. In pursuingthis objective(i.e., the justificationof infinite
deductions),HilbertfinallyapproachedKronecker'sfinitismto a certainextent:once
metamathematics was takento be a weakpartof arithmetic,it closely correspondedto
finitarymathematicsa la Kronecker[56, p. 487].
11. SYZYGIES AS A POSSIBLE PARADIGMFOR SIMPLICITY. It may be
very hardto definethe simplicityof proofs,but we neverthelessrecognize"simplic-
ity"in proofswhen we see it. Differentcriteriafor assessingsimplicity(for example,
prooflength[27], totalnumberof symbolsin the proof,the absenceof certainterms,
the numberof basic operations)may becomerelevantas the contextchangesso that
differentfacetsof simplicitymaybe emphasizedby somecriteria,downplayedby oth-
ers (see [75:6,chap.7]). In an interestinghistoricalremarkon simplicityin geometry,
Rene Descartes(1596-1650) discussedin his La Geometrie(1637) the questionof
when one curveis simplerthananother(see [6, chap.25]). Keepingall these factors
in mind,whatfeaturesof simplicitycan we defineprecisely?
The introductionof measuresof simplicityfor proofsis, to be sure,a delicatebusi-
ness. As mentionedearlier,Hilbertwas guidedby his investigationsof invariants[1],
in which he madeuse of special algebraicobjectsknownas syzygies([74, pp. 163-
183] gives a comprehensivesurveyof the old theory).Geometricfacts that are in-
dependentof the coordinatesystem-in otherwords,invariants-can frequentlybe
expressedthroughthe requirementthatsome relatedalgebraicconditionbe satisfied.
Forexample,in Euclideangeometry,the relevantinvariantsareembodiedin quantities
thatare not alteredby geometrictransformations such as rotations,dilations,andre-
flections.In analyticalterms,the invariantsin questionareinvariantsof tensorsor, to
employthe terminologyof Hilbert'stime,invariantsof an n-aryformof degreem un-
derlineartransformations ([39], [42]; see also [74], [79], [89]). Oldpapersin invariant
theorytypically consisted of massesof endlessalgorithmiccomputations,whereasby
viewing invariantsin a broader frameworkHilbertprovedhis theoremsin a few pages,
almostwithoutcalculations(see [40], [74], [79], [89]).
In algebra,the term "syzygy"is used to signify a relationship.Startingwith a
polynomialring R = K[xl, x2 ..., xn] over a field K, Hilberthad shown-we use
modem terminology(see [61, chap. 8, sec. 4], [111, p. 251])-that the set of all
invariantsi forman idealI of R andanyinvarianti canbe representedby afinitebasis
il, i2, .. ., ik so that all invariantsare integral rational functions of the generators (ba-
sic invariants).However,at the sametime these basic invariantsare not algebraically
independent:theybelongto the zero-setsof certainpolynomialrelations,the syzygies
of the precedingparagraph.The collectionof syzygies is closed underthe operations

of addition and multiplication. Moreover, syzygies constitute an ideal in the relevant
polynomial ring. The syzygy ideal itself has a finite basis, the elements of which are
not always algebraically independent. Thus, one obtains second-order syzygies, and
so on.
In the theory it is proved that the foregoing cascade of syzygies comes to a halt
in a finite basis whose members are algebraically independent after at most k steps,
where k is the number of invariants of the basis [13, chap. 3], [74], [79]. (In his pa-
pers [1], Hilbert actually proved a stop in "at most k + 1" steps; see also [39], [42,
lecture 47].) Hilbert extended this result by establishing that any ideal I in a polyno-
mial ring R = K[xl, x2, ..., Xn] over a field K (or over a ring of integers) is finitely
generated, a fact now known as Hilbert's Basis Theorem [1], [42, lecture 35], [58:2,
no. 16, pp. 199-270] (for generalizations, see [88], [8:2, pp. 431-444], [61, pp. 387,
391], and [92]).
Accordingly, this special part of algebra can be placed in the kind of finitary frame-
work we have been discussing. As to the reduction of proofs to an algebraic calculus,
it may well be that what Hilbert had in mind was to generalize the situation described
by the basis theorem (see [48, p. 413], [17, p. 1113], and [18, p. 196]). In a lecture
"Theorie der algebraischen Invariantennebst Anwendungen auf Geometrie" [Theory
of algebraic invariantstogether with applications to geometry] [42] that Hilbert deliv-
ered in 1897 when his research in invarianttheory had been completed, he remarked:
"Witheach mathematicaltheorem, three things are to be distinguished. First, one needs
to settle the basic question of whether the theorem is valid .... Second, one can ask
whether there is any way to determine how many operations are needed at most to carry
out the assertion of the theorem. Kronecker has particularly emphasized the question
of whether one can carry it out in a finite number of steps" [42, lecture 37]. Consis-
tent with this statement is the following quotation taken from the same collection of
lectures: "It can also happen that a given invarianthas several different symbolic rep-
resentations. When making symbolic calculations, one of course chooses the simplest
one" [42, lecture 31] (for the symbolic representationsee [89, p. 22]).
Moreover, Hilbert's theory of algebraic number fields was partly foreshadowed by
certain finite investigations of so-called algebraic modular systems (Modulsysteme)
by Kronecker [13, p. 147]. Hilbert might have felt that these investigations could serve
as an example of a finite theory [56, p. 487]. In view of the twenty-fourth problem, it
could well be that Hilbert's general outline had its roots in the structureof the afore-
mentioned systems.
12. CONCLUSIONS. His biographer Otto Blumenthal (1876-1944) stressed that

Hilbert was a man of problems [58:3, p. 405]. This meant that Hilbert's startingpoints
were always simple, but importantproblems. I believe that this way to create a theory
will salvage essential parts of his proof theory and metamathematics.Indeed, the meth-
ods of proof theory are now playing, not surprisingly, a significant role in computer
science. Moreover, despite Godel's results that reveal the goal of the original Hilbert
program to be unattainable, a modified Hilbert program did lead to the development
of proof theory, metamathematics, and decision theory (or, as it is sometimes called,
computability theory).
In the end, neither Hilbert nor his staunchest adversary on foundational issues,
Luitzen Egbert Brouwer (1881-1966), felt any more obliged than most mathemati-
cians today to adhere to restrictive philosophical doctrines in their "everyday"math-
ematical research, despite the caveat of Godel's triumph. A striking example is found
in Brouwer's topological research-at least until 1917-in which the use of geometric

intuition plays a vital role (as it does in every contemporarymathematician's geomet-
rical paper) [64, pp. 145, 148-156].
Hilbert believed in the ultimate efficacy of (finite) mathematics. His writings, lec-
tures, letters, and discussions display the full conviction that well-posed mathematical
problems are always questions with meaningful answers in the same way that ques-
tions about physical reality have answers. As to parallels between nature and thought,
Hilbert stated in his mathematical notebook: "Between thought [Denken] and event
[Geschehen] there is no fundamental and no quantitativedifference. This explains the
pre-established harmony [between thought and reality] and the fact that simple exper-
imental laws generate ever simpler theories" [38:3, p. 95] (see also [56, p. 485]). We
have mentioned earlier that Hilbert believed thought and nature to be finite. Later on,
Hilbert became convinced of this metaphysical principle: there is a realm beyond phe-
nomena, and the universe is governed in such a way that a maximum of simplicity and
perfection is realized (compare also Einstein's view in [78]).
The idea of the infinite had stirredmen's emotion like no other subject ([52, p. 163];
see also [87, p. 176]). Hilbert broached the old question of the limits of human abil-
ity to handle the infinite. Can we grasp the mathematical infinite in finite terms? Is
it possible for a special problem or, beyond that, for a whole theory? Or is it a con-
tradiction in itself? Suppose that in mathematics we could eliminate ideal concep-
tions (for example, the continuum), i.e., express them in finite form. Would it then be
possible to exploit this finiteness to establish a framework for gauging the simplicity
of proofs? Or, if we insist upon remaining in the realm of the finite, could we then
carry out all the necessary mathematicalreasoning? Can the limited human brain even
begin to grasp in finite terms all the theorems in mathematics that stand in need of
proof?
How could this great mathematicianever have believed that such dreams would be
realized [20, p. 1056]? Does it appear that after Hilbert's discovery of the extremely
general finiteness principle upon which his proof of the basis theorem was based,
Hilbert was overly optimistic about finiteness results in other algebraic and even in
foundational contexts? Was Hilbert's unwavering belief in the power of thought just
naive? Was it indebted to the widespread belief in irresistible progress that prevailed at
the turn of the twentieth century, part of the Zeitgeist? David Mumford, Fields medal-
ist in 1974, remarkedin a paper "The Finite Generation of Subrings Such As Rings of
Invariants"dealing with Hilbert's fourteenth problem [8:2, pp. 431-432]: "However
my belief is that it [Hilbert's belief in finiteness] was not at all a blind alley: that on the
one hand its failure reveals some very significant and far-reachingsubtleties in the cat-
egory of varieties .... In fact, my guess is that it was Hilbert's idea to take a question
that heretofore had been considered only in the narrowcontext of invarianttheory and
thrustit out into a much broadercontext where it invited geometric analysis and where
its success or failure had to have far-reachingalgebro-geometric significance." We can
add this to it: Hilbert's philosophical ideas led him to believe that through an extension
of these finite properties he had first become aware of in invarianttheory it could be
possible to establish the foundations of mathematics by finitary means. The price one
normally pays for aiming impossibly high but, like Sisyphus, repeatedly falling short
of one's objective is despair. Yet, even after Godel's results, which set up impassable
roadblocks to the achievement of Hilbert's goal, the aging Hilbert was full of hope that
these foundational objectives were attainablein a modified sense. From our viewpoint,
however, a distinction is to be made between the programHilbert intended to carry out
and its realizable part [94].
We extract another quotation from Hilbert's Paris lecture [37, p. 444]: "Occasion-
ally it happens that we seek the solution under insufficient presuppositions or in an
18 1() THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

incorrect sense, and for this reason do not succeed. The problem then arises: to show
the impossibility of the solution underthe given hypotheses."What does simplest proof
mean in this context-the simplest counterexample? If we consider the last quotation
in relation to the second problem on the Hilbert list, the Continuum Problem, then
a remark of Godel, who had profound insights into the problem, is most interesting.
I quote from a letter of Godel to Constance Reid [28, 4 June 1969], [28, 22 March
1969] (see also [87, p. 218]): "It is frequently overlooked that, disregardingquestions
of detail, one quite importantgeneral idea of his has proved perfectly correct, namely
that the Continuum Problem will require for its solution entirely new methods de-
riving from the foundation of mathematics. This, in particular,would seem to imply
(although Hilbert did not say so explicitly) that the Continuum Hypothesis is undecid-
able from the usual axioms of set theory."Evidently, the open-minded Hilbert was not
so misguided by preconceived notions to think that something he wished to be true had
to be true (as is sometime maintained; see [20, p. 1057]).
Despite the fact that problems form the basis of mathematics and determine its
progress, for Hilbert they were not the be-all and end-all. Mathematics is more than
a collection of isolated problems. It is only the mathematical method that prepares
the ground for mathematics, highlights the basic ideas, and finally makes mathematics
more than a hodgepodge of problems. On the other hand, Hilbert was fully aware of
the vital part problems play [37, p. 444]: "He who seeks for methods without having a
definite problem in mind seeks for the most part in vain."
Mathematics is not unlimited in scope, true, but for all its limitations it does con-
tinue to make progress. Despite an ever-growing diversity of mathematical areas and
a rapidly expanding body of mathematical knowledge, the simplification of proofs by
axiomatic methods has made mathematics as a whole a more efficient and unified en-
terprise. "Axiomatics is the rhythm that makes music of the method, the magic wand
that directs all individual efforts to a common goal" [38:3, p. 93]. From the begin-
ning, for Hilbert such a common goal was the unity of mathematics ("the science of
mathematics ... is an indivisible whole, an organism whose ability to survive rests
on the connection between its parts" (translation by Weyl [109:4, p. 123]) in which
axiomatics, proof theory, and metamathematics are but distinct parts.
No one will dispute the fact that Hilbert's spirit and influence have played important
roles in mathematics. Although a naive assumption that progress is inevitable (in math-
ematics or otherwise) no longer prevails and, with the passage of time, Hilbert's once
highly acclaimed work on the foundations of mathematics is looked upon more criti-
cally, we should respect the past-not play the tempting role of "a backward-looking
prophet" (Friedrich Wilhelm Schelling, 1775-1806) and demean the past by making
unfair comparisons with the present.
Hermann Weyl's oft-cited poetic remark sums up the impact of Hilbert's contribu-
tions [108:4, p. 132]: "I seem to hear in them from afar the sweet flute of the Pied Piper
that Hilbert was, seducing so many rats to follow him into the deep river of mathemat-
ics." By introducing his problems, Hilbert himself stoked the fires of our mathematical
enthusiasm. In a lecture, he once pointed out to the audience that in mathematics we
cannot use and furthermoredo not need such lame excuses as "the ladder is too short"
or "the experiments are too expensive" [50, p. 24]. On another occasion, he stirred
his listeners with the words: "We hear within us the perpetual call: There is the prob-
lem. Seek its solution. You can find it by pure reason, for in mathematics there is no
ignorabimus" [37, p. 445].
That is what Hilbert believed. Clearly, because of Godel's results, there are some ob-
jections to Hilbert's profession of faith in mathematics. Nevertheless, mathematicians
will forever find inspiration in the optimistic tone sounded in the famous line from his
Parisaddress[37, p. 445] thatwas engravedon his tombstone(see [87, p. 220]):
"Wirmiissenwissen,wir werdenwissen"[Wemustknow,we shallknow].
ACKNOWLEDGMENTS. This paper is adaptedfrom a lecture delivered at the AMS-MAA meeting held in
New Orleans on January 13, 2001. It is a pleasure to thank David Zitarelli for his invitation to speak at the
special session and for his encouragementto publish this paper.I extend my thanks to the GermanAcademy
of Natural Sciences, Leopoldina (Halle), who supportedmy research (LPD 1996). Furthermore,I thank the
Niedersachsische Staats- und UniversitatsbibliothekGottingen, Handschriftenabteilung,the MathematicalIn-
stitute of the University of Gottingen,the Archive of the University of Freiburg,and the Institutefor Advanced
Study, Princeton,for the permission to quote from various unpublisheddocuments in their possession.
REFERENCES
1. M. Ackerman,Hilbert's InvariantTheoryPapers, Math Sci Press, Brookline, MA, 1970.

2. W. Ackermann,Zur Widerspruchsfreiheitder Zahlentheorie,Math. Annal. 117 (1940) 162-194.
3. K. Appel and W. Haken, Every planarmap is four-colorable,Bull. Amer.Math. Soc. 82 (1976) 711-712.
4. P. Berays, Uber Hilbert's Gedanken zur Grundlegung der Arithmetik, Jahresber. Deutsch. Math.
Verein.31 (1922) 65-85.
5. E. Borger,E. Gradel,and Yu. Gurevich,The Classical Decision Problem, Springer-Verlag,Berlin, 1997;
2nd ed., 2001.
6. H. Bos, RedefiningGeometrical Exactness, Springer-Verlag,Berlin, 2001.
7. N. Bourbaki,Foundationsof mathematicsfor the working mathematician,J. Symbolic Logic 14 (1949)
1-8.
8. F. E. Browder, ed., MathematicalDevelopmentsArising from Hilbert Problems, Proceedings of Sym-
posia in Pure Mathematics,vol. 28 (two parts), American MathematicalSociety, Providence, 1976.
9. G. Cantor,Briefe, H. Meschkowski and N. Nilson, eds., Springer-Verlag,Berlin, 1991.
10. C. Caratheodory,The beginning of researchin the calculus of variations,Osiris 3 (1937) 224-240; also
in Gesammelte mathematischeSchriften, vol. 2., Beck, Miinchen, 1955, pp. 93-107. (I quote the last
edition.)
11. A. Church,A note on the Entscheidungsproblem,J. SymbolicLogic 1 (1936) 40-41, 101-102.
12. The Clay Mathematics Institute, Millennium Prize Problems, announced May 24, 2000, College de
France,Paris. http://www.claymath.org/prize-problems/html.
13. L. Corry,ModernAlgebra and the Rise of MathematicalStructures,Birkhauser,Basel, 1996.
14. H. Curry,Foundationsof MathematicalLogic, McGraw-Hill,New York, 1963.
15. D. van Dalen and H.-D. Ebbinghaus,Zermelo and the Skolem paradox,Bull. of SymbolicLogic 6 (2000)
145-161.
16. J. L. Drost, A shorterproof of the Ramanujancongruence modulo 5, this MONTHLY 104 (1997) 963-
964.
17. W. Ewald, ed., From Kant to Hilbert: A Source Book in the Foundation of Logic, 2 vols., Clarendon
Press, Oxford, 1996.
18. J. Fang, Hilbert. Towardsa Philosophy of ModernMathematics,II, Paideai, Hauppauge, 1970.
19. S. Feferman, What does logic have to tell us about mathematicalproofs?, Math. Intelligencer 2 (1)
(1979) 20-24.
20. H. Freudenthal,David Hilbert, in Biographical Dictionary of Mathematicians,C. Gillespie, ed., Scrib-
ner's, New York, 1991, pp. 1052-1058; published earlier in the Dictionary of Scientific Biography,
Scribner's,New York, 1970.
21. G. Gentzen, The Collected Papers of Gentzen, M. Szab6, ed., North Holland, Amsterdam, 1969 (in-
cludes Die Widerspruchsfreiheitder Zahlentheorie,Math. Annal. 112 (1936) 493-565).
22. A. Gierer,Der physikalische Grundlegungsversuchin der Biologie und das psychophysische Problem,
Ratio 1 (1970) 40-54; English version in the English edition of this journal: The physical foundations
of biology and the psychophysic problem, Ratio 12 (1970) 47-64.
23. , Godel meets Camap: a prototypicaldiscourse on science and religion, Zygon 32 (1997) 207-
217.
24. K. Godel, Die Vollstandigkeitder Axiome des logischen Funktionskalkiils,Monatsheftefur Mathematik
und Physik 37 (1930) 349-360 (G6del's Dissertation, 1929); also with English transl. in Collected
Works,S. Fefermanet al., eds., vol. 1, Oxford University Press, Oxford, 1986, pp. 60-101.
20 ? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

25. , Nachtragzu der Diskussion zur Grundlegungder Mathematikam Sonntag, dem 7. Sept. 1930
[Supplementto the discussion [331on the K6nigsbergMeeting, September7, 1930], Erkenntnis2 (1930)
149-151.
26. , Uber formal unentscheidbare Satze der Principia Mathematica und verwandte Systeme I,
Monatsheftefir Mathematikund Physik 38 (1) (1931) 173-198; English translation,On formally un-
decidable propositions of PrincipiaMathematicaand related systems, I, in The Undecidable, M. Davis,
ed., Raven Press, Hewlett, NY, 1965, pp. 5-38; furtherEnglish edition by R. B. Braithwaite, Dover,
New York, 1992; with English translationin Collected Works,S. Feferman et al., eds., vol. 1, Oxford
University Press, 1986, pp. 144-195.
27. , Uber die Lange von Beweisen, Erg. Math. Koll. 7 (1934) 23-24.
28. , Letters to Constance Reid from March 22, 1966, and June 25, 1969; letters in the possession
of Mrs. Reid, not yet included in G6del's Collected Works,S. Fefermanet al., eds., but partlyquoted in
Reid [87]. I quote with the kind permission of the Institutefor Advanced Study, Princeton,NJ.
29. , Letterto ErnstZermelo from October 12, 1931, University of Freiburgi.Br., Universitatsarchiv,
C 129/36.
30. I. Grattan-Guinness,A sideways look at Hilbert's twenty-threeproblems of 1900, Notices Amer.Math.
Soc. 47 (2000) 752-757; correctionsby G. H. Moore and response of Grattan-Guinnessin Lettersto the
Editor,Notices 48 (2001) 167.
31. J. Gray, The Hilbert Challenge, Oxford University Press, Oxford, 2000.
32. P. A. Griffith,Mathematicsat the turnof the millennium, this MONTHLY107 (2000) 1-14.
33. H. Hahn et al., Diskussion zur Grundlegungder Mathematikam Sonntag, dem 7. Sept. 1930 [Discus-
sion on the second Meeting Erkenntnislehrein K6nigsberg, September5-7, 1930], Erkenntnis2 (1930)
135-149. Note both the Vorbemerkungof the editors on the page before the content and Godel's sup-
plement [25].
34. G. B. Halsted, The InternationalCongress of Mathematicians,this MONTHLY 7 (1900) 188-189.
35. G. H. Hardy, A Mathematician's Apology, Cambridge University Press, Cambridge, 1940; reprinted
with a forewordby C. P. Snow, 1967. (I quote the enlarged 1967 edition.)
36. A. Heyting, Mathematische Grundlagenforschung. Intuitionismus. Beweistheorie, Ergebnisse der
Mathematikund ihrerGrenzgebiete, Band 3, Heft 4, Springer-Verlag,Berlin, 1934.
37. D. Hilbert, Mathematische Probleme, lecture held at the Paris ICM 1900, Nachrichten von der
Koniglichen Gesellschaft der Wissenschaftenzu Gottingen (1900) 253-297; revised version in Archiv
der Mathematikund Physik 1 (1901) 44-63, 213-237, and [58:3, pp. 290-329]; English translation(by
M. W. Newson) in Bull. Amer.Math. Soc. 8 (1901) 437-479; reprintedin the Bulletin 37 (2000) 407-
436; also in Browder [8:1, pp. 1-34], Yandell [115, pp. 324-389], and Gray [31, pp. 240-282], selected
parts in Ewald [17:2, pp. 1096-1105] and Reid [87, chap. 9]; French translationsby L. Laugel (with
corrections and additions) in Compte rendu du deuxieme congres international des mathematiciens
tenu a Paris du 6 au 12 aout 1900, E. Duporcq, ed., Gauthier-Villars,Paris, 1902, pp. 58-114; (frag-
ments) L'Enseignement Mathematique 2 (1900) 349-354 and Revue Generale des Sciences Pures
et Appliquees 12 (1901) 168-174; annotated Russian edition, Problemy Gilberta, P. S. Alexandrov,
ed., Nauka, Moscow, 1969; its German edition Die Hilbertschen Probleme, Akademische Verlagsge-
sellschaft, Leipzig, 1976, 2nd ed. 1979, reprint 1998 (both the Russian and German editions give the
revised version of [58:3]). (I quote Newson's translation.)
38. , MathematischeNotizbiicher,3 notebooks, Niedersachsische Staats-und Universititsbibliothek
Gottingen, Handschriftenabteilung,Cod. Ms. D. Hilbert 600:1-3.
39. , Invariantentheorie,lecture notes from Winter 1886 by Hilbert, Niedersiichsische Staats- und
UniversitatsbibliothekGottingen, Handschriftenabteilung,Cod. Ms. D. Hilbert 521.
40. , Uber die Theorie der algebraischen Invarianten,submittedto the InternationalMathematical
Congress in Chicago 1893, in Mathematical Papers Published by the American MathematicalSociety,
vol. 1, E. H. Moore et al., eds., Macmillan, New York, 1896, pp. 116-124.
41. , Gewohnliche Differentialgleichungen, lecture notes from Summer 1896, Niedersachsische
Staats- und UniversitatsbibliothekGottingen, Handschriftenabteilung,Cod. Ms. D. Hilbert 546.
42. , Theorie der algebraischen Invariantennebst Anwendungenauf Geometrie, lecture notes from
Summer 1897 preparedby Sophus Marxsen, Libraryof the MathematicalInstitute of the University of
Gottingen; English translationTheory of Algebraic Invariants, R. C. Laubenbacherand B. Sturmfels,
eds. [using a differentcopy from the MathematicsLibraryof Cornell University], CambridgeUniversity
Press, Cambridge, 1993. (I quote Laubenbacher'stranslation.)
43. , Ulberden Zahlbegriff,Jahresber.Deutsch. Math. Verein.8 (1900) 180-184; also in [57] from
the 3rd to 7th editions, pp. 256-262 (3rd. ed., 1909), pp. 237-242 (4th to 6th ed., 1913, 1922, and 1923),
pp. 241-246 (7th ed., 1930); English translation(by W. Ewald) [17:2, pp. 1089-1095].
44. , Uber die Grundlagender Logik und der Arithmetik,in Verhandlungendes 3. Internationalen
Mathematiker-Kongressesin Heidelberg 1904, A. Krazer,ed., Teubner-Verlag,Leipzig, 1905, pp. 174-
185; also in [57] from the 3rd to the 7th editions, pp. 263-279 (3rd ed., 1909), pp. 243-258 (4th to
6th ed., 1913, 1922, and 1923), pp. 247-261 (7th ed. 1930); English translations(by G. B. Halsted)
The Monist 15 (1905) 338-352 and (by W. Woodward) in From Frege to Godel. A Source Book in
Mathematical Logic 1879-1931, J. van Heijenoort, ed., HarvardUniversity Press, Cambridge, 1967,
pp. 130-138; 2nd ed., 1971; French translationL'EnseignementMathematique7 (1905) 89-103.
45. , Logische Prinzipien des mathematischenDenkens, lecture notes from Summer 1905 prepared
by E. Hellinger, Libraryof the MathematicalInstituteof the University of Gottingen.
46. , Logische Prinzipien des mathematischenDenkens, lecture notes from Summer 1905 prepared
by M. Born, Niedersachsische Staats- und UniversitatsbibliothekGottingen, Handschriftenabteilung,
Cod. Ms. D. Hilbert558a.
47. , Principien der Mathematik,lecture notes from Winter 1917 preparedby P. Berays, Library
of the MathematicalInstituteof the University of Gottingen.
48. , Axiomatisches Denken, Math.Annal. 78 (1918) 405-415; also in [58:3, pp. 146-156]; English
translations (by Ewald) [17:2, pp. 1105-1115] and (by Fang) [18, pp. 187-198]; French translation
L'EnseignementMathematique20 (1918) 122-136.
49. , Neubegriindungder Mathematik,Erste Mitteilung, Abhandlungenaus dem Mathematischen
Seminar der UniversitdtHamburg 1 (1922) 157-177; also in [58:3, pp. 157-177]; English translation
(by W. Ewald) [17:2, pp. 1115-1134].
50. , Wissen und mathematischesDenken, lecture notes from Winter 1922 preparedby W. Acker-
mann;revised reprint,C.-F. Bbdigheimer,ed., MathematischesInstitutGottingen, 1988.
51. , Die logischen Grundlagender Mathematik,Math. Annal. 88 (1923) 151-165; also in [58:3,
pp. 178-191]; English translation(by W. Ewald) [17:2, pp. 1134-1148].
52. , Uber das Unendliche, Math. Annal. 95 (1926) 161-190; also in [57, 7th ed., pp. 262-288],
shortened version in Jahresber Deutsch. Math. Verein.36 (1927) 201-215; English translation (by
S. Bauer-Mengelberg)in From Frege to Godel. A Source Book in Mathematical Logic 1879-1931,
J. van Heijenoort,ed., HarvardUniversity Press, Cambridge, 1967, pp. 367-392; 2nd ed., 1971; French
translation(by A. Weil), Sur l'infinie, Acta Math. 48 (1926) 91-122.
53. , Die Grundlagen der Mathematik, [Zweite Mitteilung], Abhandlungen aus dem Mathema-
tischen Seminar der Universitat Hamburg6 (1928) 65-85; followed by Diskussionsbemerkungenzu
dem zweiten HilbertschenVortragby H. Weyl, pp. 86-88, and Zusatz zu HilbertsVortragby P. Bemays,
pp. 89-95; shortenedversion in [57,7th ed., pp. 289-312]; English translation(by S. Bauer-Mengelberg
and D. F0llesdal) in From Frege to Godel. A Source Book in Mathematical Logic 1879-1931, J. van
Heijenoort,ed., HarvardUniversity Press, Cambridge, 1967, pp. 464-479; 2nd ed., 1971.
54. , Probleme der Grundlegungder Mathematik,in Atti del Congresso Internazionale dei Mate-
matici, Bologna 1928, Tomo I, Zanichelli, Bologna, 1929, pp. 135-141; also in Math. Ann. 102 (1930)
1-9, [57,7th ed., pp. 313-323], and Hilbert. Gedenkband,K. Reidemeister,ed., Springer-Verlag,Berlin
1971, pp. 9-19.
55. , Naturerkennenund Logik, Naturwissenschaften(1930), 959-963; also in Gesammelte Ab-
handlungen [58:3, pp. 378-387]; English translations(by W. Ewald) in [17:2, pp. 1157-1165]; selected
partsin Reid [87, chap. 12]; English translationof the correspondingradio broadcastin Vinnikov [102].
56. , Die Grundlegung der elementaren Zahlenlehre, Math. Annal. 104 (1931) 485-494; short-
ened version in the GesammelteAbhandlungen[58:3, pp. 192-195]; English translation(by W. Ewald)
in [17:2, pp. 1148-1157].
57. , Grundlagen der Geometrie, Teubner-Verlag,Leipzig, 1899; from the 8th amended ed. by
P. Berays with supplements, Stuttgart,1956-1987; 14th ed., with contributionsby H. Kiechle et al.,
M. Toepell, ed., Stuttgart, 1999; first English translation The Foundations of Geometry by E. J.
Townsend, Open Court, Chicago, 1902; new translationby L. Unger, Open Court,La Salle, 1972. [The
first Germanedition of the Grundlagenwas a contributionto the Festschriftzur Enthullungdes Gauss-
WeberDenkmals in Gottingen, later editions were published separately.]
58. , GesammelteAbhandlungen,3 vols., Springer-Verlag,Berlin, 1932-1935; 2nd ed., 1970.
59. D. Hilbert and W. Ackermann, Grundziige der theoretischen Logik, Springer-Verlag,Berlin, 1928;
6th ed., 1972; English translationof the second edition Principles of Mathematical Logic, by L. M.
Hammond et al., Chelsea, New York, 1950; Chinese translationScience Press, Beijing, 1950; Russian
translationNauka, Moscow, 1979.
60. D. Hilbert and P. Berays, Grundlagender Mathematik,2 vols., Springer-Verlag,Berlin, 1934-1939;
2nd ed., 1968-70; Russian translationNauka, Moscow, 1979-1982.
61. T. Hungerford,Algebra, 8th ed., Springer-Verlag,New York, 1974.
62. H. N. Jahnke, Hilbert, Weyl und die Philosophie der Mathematik,Math. Semesterberichte37 (1990)
157-179.
63. T. Koetsier,Hilberts 24ste probleem, Nieuw Archiefvoor Wiskunde5 (2) (2001) 65-67. [The canceled
24th problem is reproducedin facsimile.]

64. T. Koetsier and J. van Mill, General topology, in particulardimension theory, in the Netherlands, in
Handbookofthe History of General Topology,vol. 1, C. E. Aull and R. Lowen, eds., Kluwer,Dordrecht,
1997, pp. 135-180.
65. G. Kreisel and J.-L. Krivine, Elements de logique mathematique.Theorie des modeles, Dunond, Paris,
1967; English translationElements of MathematicalLogic (Model Theory),North Holland, Amsterdam,
1967.
66. G. Kreisel, What have we learntfrom Hilbert's second problem?, in Browder [8:1, pp. 93-130].
67. C. W. H. Lam, How reliable Is a Computer-BasedProof?, Math. Intelligencer 12 (1) (1990) 8-12.
68. E. M. H. Lemoine, La Geometrographieou l'Art des ConstructionsGeometriques,Naud, Paris, 1902.
69. U. Majer, Hilbert's programto axiomatize physics (in analogy to geometry), in History of Philosophy
and Science, M. Heidelbergerand F. Stadler,eds., Kluwer, Dordrecht,2002, pp. 213-224.
70. Yu. Matiyasevich, Diophantine enumerable sets, Doklady Akademii Nauk SSSR 191 (1970) 279-282
[Russian];English translationin Sov. Math Doklady 11 (1970) 354-357.
71. H. Minkowski, Briefe an David Hilbert, L. Riidenberg,ed., Springer-Verlag,Berlin, 1973. [The 105 let-
ters are preservedin the Niedersachsische Staats- und UniversitatsbibliothekGottingen, Handschriften-
abteilung, Nachlass Hilbert, Cod. Ms. D. Hilbert 258.]
72. E. Menzler-Trott,Gentzens Problem (including an essay on Gentzen's proof theory by J. von Plato),
Birkhauser-Verlag,Basel, 2001. [A shortenedEnglish translationis in preparation.]
73. H. Meschkowski, Richtigkeitund Wahrheit,BibliographischesInstitut,Mannheim, 1978.
74. F Meyer, Bericht iiber den gegenwartigen Stand der Invariantentheorie,Jahresber. Deutsch. Math.
Verein.1 (1890-91) 81-292.
75. G. H. Miiller, Q-Bibliographyof MathematicalLogic, 6 vols., Springer-Verlag,Berlin, 1987.
76. R. Murawski,Recursive Functions and Metamathematics.Problems of Completenessand Decidability,
Godel's Theorems,Kluwer, Dordrecht, 1999.
77. J. von Neumann, Zur HilbertschenBeweistheorie, Math. Zeitschrift26 (1927) 1-46.
78. J. D. Norton, 'Nature is the realisation of the simplest conceivable mathematicalideas': Einstein and
the canon of mathematicalsimplicity, Studies in the History of Moder Physics 31 (2000) 135-170.
79. K. V. H. Parshall,The one-hundredthanniversaryof the death of invarianttheory?, Math. Intelligencer
12 (4) (1990) 10-16.
80. V. Peckhaus, Hilbertprogrammund KritischePhilosophie, Vandenhoeck& Ruprecht,Gottingen, 1990.
81. , Hilbert's axiomatic programmeand philosophy, in TheHistory of ModernMathematics,vol. 3,
E. Knobloch and D. Rowe, eds., Academic Press, Boston, 1994, pp. 90-112.
82. J.-P. Pier, ed., Development of Mathematics, 1900-1950 and 1950-2000, 2 vols., Birkhauser,Basel,
1994 and 2000.
83. P. Pudlak, The lengths of proofs, in Handbook of Proof Theory, S. R. Buss, ed., Elsevier, Amsterdam,
1998, pp. 547-637.
84. S. Rebsdorfand H. Kragh,eds., EdwardArthurMilne-The relationsof mathematicsto science, Studies
in History and Philosophy of Modern Physics 33 (2002) 51-64.
85. M. R6dei, "Unsolved Problems in Mathematics":John von Neumann's addressto the ICM Amsterdam,
1954, Math. Intelligencer 21 (4) (1999) 7-12.
86. K. Reich, The American contributionto the theory of differentialinvariants,1900-1916, in TheAttrac-
tion of Gravitation,J. Earmanet al., eds., Birkhauser,Boston, 1993, pp. 225-247.
87. C. Reid, Hilbert, Springer-Verlag,New York, 1970; reprintedCopernicus,New York, 1996.
88. M. von Renteln, Zur Situation der Analysis um die Jahrhundertwende,in Vorlesungenzum Gedenken
an Felix Hausdorff, E. Eichhor and E. J. Thiele, eds., Heldermann,Berlin, 1994, pp. 107-130.
89. G.-C. Rota, Two turningpoints in invarianttheory,Math. Intelligencer 21 (1) (1999) 20-27.
90. B. Scarpellini, Komplexitatstheorie,Wissenschaftsmagazinder Universitdt Basel "uni nova" 87 (6)
(2000) 51-52.
91. C. A. Scott, The InternationalCongress of Mathematiciansin Paris, Bull. Amer Math. Soc. 7 (1900)
57-79.
92. J.-P. Serre, Algebre Locale, Ultilicites, Lecture Notes in Mathematics,no. 11, Springer-Verlag,Berlin,
1965.
93. K. Sigmund, Hans Hahn and the foundational debate, in The Foundational Debate, W. Depauli-
Schimanovic et al., eds., Kluwer, Dordrecht, 1995, pp. 235-245.
94. S. G. Simpson, Partialrealizations of Hilbert's program,J. Symb.Logic 53 (1988) 349-363.
95. B. Sobociiski, On well constructed axiom systems, in Polskie TowarzystwoNaukowe na Obctyznie
[Polish Society of Art and Sciences, Abroad], Yearbookfor 1955-56, London, 1956, pp. 1-12.
96. M. Stoltzner, How metaphysical is "Deepening the foundations"?Hahn and Frank on Hilbert's ax-
iomatic method, in History of Philosophy of Science, F. Stadler,ed., Kluwer, Dordrecht,2002, pp. 245-
262.
97. R. Taylor and A. Wiles, Ring-theoretic properties of certain Hecke algebras, Ann. Math. 141 (1995)
532-572.
98. R. Thiele, Hilbert and his twenty-four problems, in Mathematics at the Dawn of a Millennium, Pro-
ceedings of the Canadian Society for History and Philosphy of Mathematics, Hamilton, Ont., 2000,
vol. 13, M. Kinyon, ed., pp. 1-22. [Some documents including the canceled twenty-fourthproblem are
reproducedin facsimile.]
99. , Von der Beroullischen Brachistochrone zum Kalibratorkonzept. Untersuchungen zur
Geschichte der Feldtheorie bei einfachen Variationsproblemen,Habilitationsschrift,Universitat Ham-
burg, FachbereichMathematik,2001; Brepols, Turnhout,2003 (to appear).
100. , David Hilbert and French Mathematics,in Proceedings of the XXXI.International Congress
of History of Science, Mexico City, Mexico, 2001 (to appear).
101. D. Velleman, Fermat'slast theorem and Hilbert's program,Math. Intelligencer 19 (1) (1997) 64-67.
102. V. Vinnikov,We shall know: Hilbert's apology, Math. Intelligencer 21 (1) (1999) 42-46.
103. G. Vollmer, Geloste, ungeloste und unlosbare Probleme, in Fortschritt und Gesellschaft, E.-L. Win-
nacker,ed., Hirzel-VerlagStuttgart,1993, pp. 79-97; also in G. Vollmer, Wissenschaftstheorieim Ein-
satz, Stuttgart,Hirzel-Verlag,1993, pp. 183-210.
104. , Denkzeuge, in MannheimerForum90/91, E. P. Fischer, ed., Piper,Munchen, 1991, pp. 15-78.
105. A. Weil, The futureof mathematics,in Great Currentsof MathematicalThought,vol. 1, F. Le Lionnais,
ed. (translatedfrom the Frenchby A. Dresden), Dover, New York, 1971, pp. 320-336.
106. S. Weinberg,Dreams of a Final Theory,Vintage, London, 1993.
107. H. Weyl, Randbemerkungenzu Hauptproblemender Mathematik,Math. Zeitschrift20 (1924) 131-150.
108. , David Hilbert and his mathematicalwork, Bull. Amer Math. Soc. 50 (1944) 612-654.
109. , Obituary:D. Hilbert 1862-1943, American Philosophical Society YearBook (1944) 387-395;
also in GesammelteAbhandlungen,Springer-Verlag,Berlin, 1968, pp. 121-129.
110. , A half-centuryof mathematics,this MONTHLY58 (1951) 523-553; also in GesammelteAb-
handlungen,Springer-Verlag,Berlin, 1968, pp. 464-494.
111. , The Classical Groups. Their Invariants and Representations,2nd. ed., Princeton University
Press, Princeton, 1946.
112. A. Wiles, Modularelliptic curves and Fermat'slast theorem,Ann. Math. 141 (1995) 443-531.
113. L. Wos, Automatingthe search for elegant proofs, J. AutomatedReasoning 21 (g) (1998) 135-175.
114. L. Wos and R. Thiele, Hilbert'snew problem, Universityof L6dz,Bull. of the Section of Logic 30 (2001)
165-175.
115. B. H. Yandell, The Honors Class. Hilbert's Problems and Their Solvers, A. K. Peters, Natick, MA,
2002.
116. E. Zermelo, Lettersto K. Godel from September21, 1931, and October29, 1931, Universityof Freiburg
i. Br., Universitatsarchiv,C 129/36; reprintedin I. Grattan-Guinness,In memoriam Kurt G6del: His
1931 correspondencewith Zermelo on his incompletabilitytheorem,Historia Math. 6 (1979) 294-304;
see also the additionof J. W. Dawson, Completing the G6del-Zermelocorrespondence,Historia Math.
12 (1985) 66-70.
RUDIGER THIELE is a historianof mathematicswho studied both mathematicsand physics, receiving his
Ph.D. in mathematicsfrom Martin Luther University of Halle-Wittenberg(Germany) and his habilitation in
the history of physical sciences from University of Hamburg.In 1996 he was awarded the Forder Prize of
the German Academy of Natural Sciences, Leopoldina. Since 1986 he has held a position as Privatdozent
in Mathematics at the University of Leipzig. He has been a visiting professor and research scholar at the
Universities of Bonn, Mainz, Muinster,andTorontoand is vice presidentof the Euler Society in the USA. While
having diverse researchinterests in the history of mathematics,his main field of researchis the history of the
calculus of variations.His publicationsinclude a biographyof LeonhardEuler and a booklet on mathematical
proofs. Recently he edited the Festschrift "Mathesis"in honor of M. Schramm. His other interests include
philosophy and, above all, music.
Karl-Sudhoff-Institut,Universityof Leipzig, Augustusplatz10-11, D-04109 Leipzig, Germany
thieler@medizin.uni-leipzig.de

Hilbert S 24 Problems Assoc

Uploaded by

Copyright:

Available Formats

Hilbert S 24 Problems Assoc

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hilbert S 24 Problems Assoc

Uploaded by

Copyright:

Available Formats

Hilbert's Twenty-FourthProblem

1. INTRODUCTION. For geometers, Hilbert's influential work on the foundations

The 24th problemin my Parislecturewas to be: Criteriaof simplicity,or proof

2 (r) THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

3. THE PREHISTORY. In the winter of 1899-1900 Hilbert, though but thirty-eight

4. ON THE ROLE OF PROBLEMS. How should Hilbert's proposed problems be

January2003] HILBERT'S TWENTY-FOURTHPROBLEM 3

4 ()? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

5. SIMPLICITY AND RIGOR. It is widely believed amongmathematiciansthat

6. THE PLACE OF THE TWENTY-FOURTHPROBLEM IN PROOF THE-

6 (? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

7. A LOOK AT HIIRERT'S FOUNDATIONS OF MATHEMATICS. From Hil-

8 )? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

8. THE HIITRERT AXIOM: SOLVABILITY OF EVERY PROBLEM. Modem

January2003] HILBERT'S TWENTY-FOURTHPROBLEM 9

10 (? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

9. SIMPLICITYVERSUS COMPLEXITY.Hilbertwas not alonein his desirefor

12 2? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

10. THE BACKGROUND OF THE TWENTY-FOURTH PROBLEM. Instead of

14 () THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

y"(x) = O, y"(x) + y(x) = 0

[87, p. 104] (see also [41, p. 1 (insertion)]).However,"Thedetailedwork will not

16 (? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

12. CONCLUSIONS. His biographer Otto Blumenthal (1876-1944) stressed that

January2003] HILBERT'S TWENTY-FOURTHPROBLEM 17

18 1() THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

"Wirmiissenwissen,wir werdenwissen"[Wemustknow,we shallknow].

1. M. Ackerman,Hilbert's InvariantTheoryPapers, Math Sci Press, Brookline, MA, 1970.

20 ? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

22 (? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

24 (? THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 110

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.