The History of Zeno's Arguments on Motion: Phases in the Development of the Theory of Limits Author(s): Florian Cajori Source:

The American Mathematical Monthly, Vol. 22, No. 3 (Mar., 1915), pp. 77-82 Published by: Mathematical Association of America Stable URL: http://www.jstor.org/stable/2971890 . Accessed: 30/04/2013 15:54
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Mathematical Association of America is collaborating with JSTOR to digitize, preserve and extend access to The American Mathematical Monthly.

http://www.jstor.org

This content downloaded from 193.225.200.93 on Tue, 30 Apr 2013 15:54:59 PM All use subject to JSTOR Terms and Conditions

THE

AMERICAN MATHEMATICAL MONTHLY

VOLUMEXXII MARCH,1915 NUMBER3

THE HISTORY OF ZENO'S ARGUMENTS ON MOTION.

OF THE THEORY OF LIMITS. PHASES IN THE DEVELOPMENT

IV.
By FLORIAN CAJORI, Colorado College.

ST. VINCENT,GALILEO,HOBBES. 4. EARLY DISCUSSIONSOF LIMITS: GREGORY century Limits in the Fifteenth and SixteenthCenturies. With the fifteenth in philosophy, are found Greek These germs new mathematicalideas appear. century but they failed to develop duringthe dark centuries. In the fifteenth considered variability the German cardinal, Nicolaus Cusanus (1401-1465), withoutbeing able to apply it successfully;he advanced the notion of a limit, the notion of infinito the limit; he entertained thoughunable to pass correctly calculus.' He held that tesimalsbut was not able to use them in an infinitesimal rules developed for the finitelose their validity for the infinite-a statement whichlater thinkershave not always heeded sufficiently.A point movingwith infinite velocityin a circleis each momentin every position on the circle; hence it is at rest. During the century,or centuryand a half,afterCusanus, concepts of limits parts and processesinvolvingthe passing to the limitbegin to appear in different of Europe, like flowerson a fieldin early spring. Perhaps firstin time, in the developmentof ideas consideredby Cusanus, is Giovanno B. Benedetti, a disof Galileo, who brought out a publication in 1585 at tinguished forerunner Turin, Italy. As early as 1586, and again in 1608, Simon Stevin at Leyden exhibitedthe processofpassingto the limit.2 In 1604 the Italian mathematician, which contains a Luc Valerio, published at Rome a treatise,De centrogravitatis, remarkable approach to the modern idea of limits.3 In Galileo's celebrated
1 K. Lasswitz, und Leipzig,1. Bd., 1890,pp. 283,284,287, Hamburg derAtomistik, Geschichte 1912. Leipzigund Berlin, H. Weber, Festschr. See also Max Simon,"Cusanus als Mathematiker," pp. 298-337. Annales chezSimonStevin," deslimites de la m6thode 2 H. Bosmans, "Sur quelquesexemples T. 37, 1912-13,2. fascicule. de Bruxelles, scientifique de la societe 3 H. Bosmans,"Les d6monstrations chez Luc Valerio,"Annales par l'analyseinfinitesimale de Bruxelles, T. 37, 1912-13,2. fascicule;C. R. Wallner,"Ueber die de la Societescientifique mathematica, 3. F., Bd. IV,.1903,p. 250, des Grenzbegriffes," Bibliotheca Entstehung

77

This content downloaded from 193.225.200.93 on Tue, 30 Apr 2013 15:54:59 PM All use subject to JSTOR Terms and Conditions

78

ZENO S ARGUMENTS ON MOTION

discourses on mechanics and falling bodies(1638)there are frequent instances of limits. In theNetherlands again,Gregory St. Vincent, whoseresearches have, untilrecently, hardly received the recognition theydeserve, was familiar with thewritings ofLuc Valerio, andhimself contributed toward laying thefoundations forthe infinitesimal calculus. Similar of a limit studies bearing on the concept are due to Andreas Tacquet ofAntwerp, and to John Wallisin his Arithmetica infinitorum, 1655,whowerebothfamiliar withthe Opusgeometricum ofGregory St. Vincent.' We proceednow to a special mention of Zeno's arguments. of discussions Benedetti, whomwe mentioned above, held that the flying arrow, thought of at a pointin its path,does not covera finite from distance, but it differs an arrowat restby possessing the attribute of velocity whichpersists even in an infinitesimal time and space.2 Direct reference to Zeno in a mannerwhich exhibits reckless following ofthegreatdialectician is found in Giuseppe Biancani of Bolognawho about 1615 sought to establish of two the incommensurability linesbytheconsideration thata supposed common measure couldnotbe applied to either line,becausethemeasure mustfirst be appliedto halfofit,and before thatto halfofthathalf, and so on to infinity, which is as impossible an operation as Zeno's "Dichotomy."3 in the of Galileo. Far more successful Speculations than earlierwriters of infinitesimals to application wereKeplerand Cavalieri, but moreimportant us at present are the speculations of Galileo. Galileoapproached the problem ofinfinite was not which aggregates witha keenness ofvisionand an originality equalled before the timeof Dedekindand GeorgCantor. Galileo'sdialogues witha discussion of divisibility and continuity and space.4 Salviati, of matter who in general is inconrepresents the author'sownideas, says,5"theinfinite ceivableto us, as is the last indivisible." Simplicio, is who in thesedialogues the spokesman ofAristotelian scholastic philosophy, remarks that "the infinity ofpoints on a longer linemustbe greater thantheinfinity ofpoints on a shorter one." Thencometheremarkable words ofSalviati:
to "These difficulties arisebecausewe withour finite mindsdiscusstheinfinite, attributing the latterproperties for is not justifiable; derivedfrom the finite and limited. This,however, the attributes great,small and equal are not applicableto the infinite, sinceone cannotspeak of greater, one can or equal infinities. smaller, . . . If now I ask how manysquaresare there, roothas a answer withtruth, just as manyas there are roots;forevery squarehas a root,6 every I see no escape, square,no squarehas morethanone root,no rootmorethan one square. exceptto say: the totality of numbers is infinite, of squaresis infinite, the totality the totality ofrootsis infinite; is neither themultitude ofsquaresis not less thanthe multitude ofnumbers, R. Wallner, loc.cit.,p. 257. K. Lasswitz, op. cit.,Vol. II, p. 17. ' J. C. Heilbronner, Historia matheseos universce, Lipsiam, 1742,p. 175. 4 See a German translation in Ostwald's Klassiker, No. 11, pp. 24-37,also No. 24, p. 17; an Englishtranslation of the parts bearingon aggregates is givenby E. Kasner in Bulletin Am. Math.Soc., Vol. XI, 1904-5,pp. 499-501. ' Ostwald's Klassiker, No. 11, p. 29. 6 Following the customof his time,Galileo considers only one root of a positivenumber, namely theprincipal root.
3

on mechanics,Discorsi e Dimostrazioni matematiche, 1638, opens the "firstday"

1 C.

This content downloaded from 193.225.200.93 on Tue, 30 Apr 2013 15:54:59 PM All use subject to JSTOR Terms and Conditions

ZENO'S ARGUMENTSON MOTION

79

the latterthe greater; and, finally, the attributes equal, greater, and less are not applicableto infinite but solelyto.finite quantities."

We shallsee thatGalileohas been curiously misinterpreted by somewriters, including Cauchy, as demonstrating herethatan actualinfinity has no existence. That there shouldbe as manysquaresas there are integers altogether was taken as absurd; hence the existenceof actual infinity was considered disproved. in theuse oftheinfinite in demonstrations Galileo'sskill is shown inthefollowing passageon falling bodies:1
"If the velocity wereproportional to the distancethrough whichit has fallenor is to fall, thenthosedistances wouldbe passedoverin equal times;thus, ifthevelocity withwhich a body overcomes four yardsis to be doublethevelocity withwhich thefirst twoyardswereovercome, thenthetimes neededforthesetwoprocesses wouldbe thesame; but four yardscan be overcome in the same timeas twoyardsonlyin the case of instantaneous motion; we see on the contrary that thebodyneedstimeto fall,and that it needsless timefora fallof twoyardsthanoffour yards;henceit is not truethatthevelocities to thedistance increase proportionally fallen."

Gregory St. Vincent. The mostimportant of Zeno givenat this discussion timeis thatby Gregory in his Opusgeometricum St. Vincent, quadraturc circuli et sectionum coni,published in 1647,but written at Antwerp apparently twentyfiveyears earlier. It is a massivevolumeof 1400 pages. Influenced in his geometrical researches by the medievalscholastic conceptof the continuum, a linedivided to which is notreduced to indivisible according elements repeatedly ofbeingsubdivided ad infinitum, as taught by theatomists, but admits Gregory St. Vincent tooka stepdifferent from thatofArchimedes.While, in hisproofs, a certain ofsmallness Archimedes until wasreached, kepton dividing, only degree to continue ad infinitum. St. Vincent thesubdivisions unlimited permitted Using a geometric series thatwastruly aninfinite section ingeometry heintroduced series.2
I A B
I

D E K

before Thismuch had beenaccomplished byat leastonewriter him,3 but,so faras to thestudy nowknown, he is thefirst to applytheinfinite geometric progression ofthe"Achilles." Taking AK hedivides itat B in a given a definite linesegment ratio,then he dividesBK in the same ratioat C, and so on. The segments AB, BC, CD, . . . forman infinitegeometricprogression. The points C, D, E . . . lie, all of them,betweenA and K; they approach K as near as we please, neverreach it. As Gregory but (in accordancewith scholastic philosophy) thefurther advance so to speak,against K is an obstacle, conceives thismatter, of the series of points A, B, C, . . . , similarto a rigid wall. "Terminus prolicetin infinitum ad quem nulla progressio gressionis est serieifinis, pertinget, ad eum accedere dato proprius sed quovisinteruallo continuetur; poterit." By thesegments thesegment "series"is meant AK, by "progressio," AB, AC, . . .
24, p. 17. of this T. 1, pp. 51-56,95-97; forour knowledge Opus geometricum, St. Vincent, Gregory mathein theBibliotheca account uponC. R. Wailner's entirely partofthebookwe are dependent 3. F., Vol. IV, 1903,pp. 251-255. matica, a See H. Wieleitner 3. F., Vol. 14, 1914,pp. 150-168. in Bibliotheca mathematica,
2

I Ostwald's No. Klassiker,

This content downloaded from 193.225.200.93 on Tue, 30 Apr 2013 15:54:59 PM All use subject to JSTOR Terms and Conditions

80

ZENO 'S ARGUMENTS ON MOTION

AK aequalem esse toti states his conclusionthus: "Dico magnitudinem Gregory rationis AB ad BC in progressionimagnitudinumcontinue proportionalium, concontinuatae;siue quod idem est, rationisAB ad BC in infinitum infinitum tinuatae terminumesse K." Consideringthe "Achilles" in this connection, with the sumtime definitely he associates this paradox on motionforthe first writer series. Moreover,GregorySt. Vincent is the first mation of an infinite known to us who states the exact time and place of overtakingthe tortoise. So far as we are able to ascertain,Gregorywas not troubled,in explainingthe "Achilles," by the fact that in his theory,the variable does not reachits limit. Nor, apparently,did this mattertrouble his readers. His mode of solvingthe problemappealed to many. We shall see that Leibniz makes special reference to it. Over a centuryafter Gregory'spublication,Saverien refersin his dictionary'to the "Achilles," "dont Gregoirede Saint Vincent a fait voir la faussete." Formeygave GregorySt. Vincent's explanationin the article "Mouvement" in Diderot's Encyclopedie(1754), later reprintedin the Encyclopedie methodique,and in 1800 translated at Padova into the Italian language. The does not allow the methodique of a limit as given in the Encyclopedie definition variableto surpass its limit but places no obstacle in the way of its reachingits limit. Descartes, De Morgan and Others. Descartes at one time discussed the "Achilles." His treatmentis much like that of GregorySt. Vincent. It is given in a letterof July,1646, to Clerselier.2 He lets Achilles,or in his place a horse,be, at the start,10 leagues behindthe tortoise,but movingten timesmore of the paradox he does not touch,for rapidlythan the latter. The real difficulty he says:
partie gardeque,si A la dixieme si on prend Asoudre, "L'Achillede Zenonne serapas difficile & encorela qui est une centi6me, de cettedixi6me, on adioutela dixi6me de quelque quantit6 toutesces & ainsiAl'infini, de la premiere, qui n'est qu'une milliesme dixi6me de cettederniere, toutesne composent infinies, r6ellement supos6es quoy qu'ellessoient dixi6mes jointesensemble, quantit6. . . Et la captionest de la premiere une neusi6me finie, foisqu' une quantit6 sqauvoir A cause qu'on infinie, partied'unelieue est une quantit6 en ce qu'on imagine que cetteneusi6me infinies." en des parties la divisepar son imagination

but not impossible as mysterious, Descartes looked upon the actually infinite or absurd. He seemed to accept it in the abstract,but deny it in the concrete. extractsfromGalileo) therewas At this time and even earlier(see the foregoing of the human mind and its consequentinabilityto contalk about the finitude ceive the infinite. This was ridiculedby De Morgan. He claimed that if the human mindis limited,we tacitlypostulate the "unknowable"; moreover,even thereis no morereason against its conceivingthe if the human mindwere finite, infinite than thereis fora mindto be blue in orderto conceive of a pair of blue eyes. Or, as De Morgan puts it in another place,-the argumentamounts to be fat." From Descartes to Hamilton, this,"who drivesfat oxen should himself
I Saverien, de mathematique et de physique, Paris, 1753,Art."Mouveuniversel Dictionnaire ment." 2 Oeuvres T. IV, pp. 445-447. de Descartes par CharlesAdamet Paul Tannery,

This content downloaded from 193.225.200.93 on Tue, 30 Apr 2013 15:54:59 PM All use subject to JSTOR Terms and Conditions

ZENO S ARGUMENTS ON MOTION

81

says De Morgan,' this doctrineis accepted by many minds. But its genesis is found,as we have stated, long beforeDescartes. explanationof "Achilles" comes A whollydifferent, but no more satisfactory fromanother Frenchman of that time,Pierre Gassendi, the physicist. In his if with Epicurus one assumes not points view Zeno's proofs need no refutation, in velocity of motion, seems to arise fromdifferences but atoms. A difficulty the more forin the same time that a body moves over the physicallyindivisible, rapid body must travel over several indivisibles. In his opinion this difficulty may perhaps be overcome by conceiving motion as discontinuous,and slower motionas a mixtureof rest and motion. To the senses motionwould still seem in accepting the existence continuous.2 To those who experienced difficulties an easy solutionby of indivisibleatoms,the capuchin,Casimirof Toulouse. offers that angels had extension,yet were physicallyindivisible.3 reminding It is worthyof note that John Dee, the famous astrologerwho wrote an elaborate mathematicalprefaceto Billingsley'sedition of Euclid (1570), departs fromthe contentionthat two lines containingthe same numberof parts must be of equal length. He says:
as thegreatest. As,a Lineofan can be dividedintoso manypartes "Our leastMagnitudes as may thediameter of thewholeworld, inchlong(with vs) maybe dividedintoas manypartes, from East to West: or any way extended."

Discussion of Thomas Hobbes. The earliest British writer,after Duns Thomas Hobbes is the philosopher, Zeno's arguments Scotus, to take up explicitly he wrote:4 In 1655 (1588-1679).
in thispropoof Zeno againstmotion, consisted the forceof that famousargument in number, thesame is infinite; which he without whatsoever may be dividedintoparts,infinite sition, is false. For to be dividedinto infinite parts,is to be true,yet nevertheless doubt,thought thata elsebut to be dividedintoas manypartsas anymanwill. But it is notnecessary nothing in number, or be infinite, becauseI can divideand subdivide it as lineshouldhave partsinfinite number is finite; becausehe that as I please; forhow manypartssoeverI make,yettheir often but leaves it to the without addinghow many,does not limitany number, says parts,simply, a linemay be dividedinfinitely; therefore we say commonly, of thehearer, which determination cannotbe truein any othersense."

with indefinite. He takes an agnostic is synonymous WithHobbes, infinite attitudetoward problemsof infinity:
numberis infinite, it is to be understood as if it were "But whenno moreis said thanthis, name. . . And,therefore, thatwhichis commonly said, thisnamenumberis an indefinite said, is not to be so understood, as if there might be that space and timemay be dividedinfinitely, to be takenin thissense,whatever is dividedis divided or eternal division;but rather any infinite him that demonstrates thus? into such parts as may again be divided. . . . Who can commend of days, or othermeasures number of time,preceded 'If the worldbe eternal, thenan infinite ofAbraham thebirth ofIsaac; and therefore thebirth ofAbraham. But thebirth one preceded

and on the Sign of Equality,"in Trans. of the Cambridge IA. De Morgan,"On Infinity; 1871 [readMay 16, 1864]. Philosoph. Society, Vol. XI, p. 157,Cambridge, 2 Gassendi, Operaomnia,1658,I, p. 300a. An. I, p. 239; Lasswitz,op. cit.,Vol. II, p. 150. 3 Lasswitz, op. cit.,Vol. II, p. 494. Vol. I, London,1839,pp. 63, 64, 413. Hobbesrefers 4The English Works ofThomas Hobbes, also in his Latin works. See ThomaeHobbes,Operaphilosophica, to Zeno's arguments Vol. V, Londini,1845,pp. 207-213.

This content downloaded from 193.225.200.93 on Tue, 30 Apr 2013 15:54:59 PM All use subject to JSTOR Terms and Conditions

82

A GENERAL FORMULA FOR THE VALUATION OF SECURITIES

or one eternalthan another than anotherinfinite, infinite is greater eternal;which,'he says, is like his,whofromthis, thatthenumber 'is absurd.' This demonstration of evennumbers is thatthereare as manyevennumbers as thereare numbers infinite, wouldconclude simply, that are as manyas all the even and odd together. They whichin this is to say, the even numbers from theworld, do theynot by the same meanstake away eternity manner take away eternity of theworld?. . . And the menthatreasonthus absurdly from thecreator are not idiots, but, whichmakes this absurdity unpardonable, geometri, as, and such as take upon themto be but severejudgesof other men'sdemonstrations." judges,impertinent,

The referenceto odd and even numbers doubtless arose fromhis contact with Galilean thought. While sojourningon the Continent,he had gone to see the duplicationof the Galileo, then a prisoner. Hobbes thoughthe had effected cube and the squaring of the circle. On this matter he became involved in a with the algebraist,JohnWallis. The aged Hobbes was no heated controversy matchagainst youngWallis on mathematicalquestions. When the mathematical worksof Wallis were being broughtout, Wallis refusedto allow his controversial matter against Hobbes to be incorporated in them.' Whether the whole is greater than a part was an issue touched upon during this dispute. Hobbes is but the ambitionof school boys." said to Wallis: "All this arguingof infinities It cannot be said that Hobbes made any real contribution to a deeper understanding of the "Achilles" or any of Zeno's other argumentson motion. His in number, objection to the dictum,"whatever may be divided into parts infinite is no new contribution;Aristotlehad advanced that far. the same is infinite," How Achilles caught the turtle is beyond comprehension through our sensual imagination;Hobbes nowhereexplains this inability. However, he does touch withWallis. Hobbes chargedthat upon the concept of a limitin his controversy ofthe professors are " void ofsense"; one ofthoseprinciples some ofthe principles being, "that a quantity may grow less and less eternally,so as at last to be equal to anotherquantity; or, whichis all one, that there is a last in eternity."2

A GENERAL FORMULA FOR THE VALUATION OF SECURITIES.3

By JAMES W. GLOVER, University of Michigan.

The object of this paper is to derive a formulafor the valuation of a very general type of securities. The securityis redeemedin r equal installmentsat intervalsof t years, the first redemptionbeing made afterf years. The annual rate of dividendis g payable in m installments, and the securityis purchasedto realize the investora nominal rate of interest j with frequencyof conversionm.
1 A full between Hobbesand Wallisis givenin CroomRobertson's accountofthe controversy Hobbes, pp. 167-185. 2 TheEnglishWorks ofThomas Hobbes, Vol. 7, p. 186. 3Read beforethe Chicago Section of the AmericanMathematicalSociety,April,1912. in the theory interest employed of compound Those unfamiliar withthe notation and functions Institute ofActuaries, PartI, byRalphTodhunter;TheMathematical mayconsult Text-Book ofthe Bulletin No. 136, of theDepartment of Agriculture, Theory of Investment, by ErnestB. Skinner; on Highway Bonds,by LaurenceI. Hewes and JamesW. Glover.

This content downloaded from 193.225.200.93 on Tue, 30 Apr 2013 15:54:59 PM All use subject to JSTOR Terms and Conditions