You are on page 1of 31

The Development of Scientific Method in the School of Padua Author(s): John Herman Randall, Jr.

Source: Journal of the History of Ideas, Vol. 1, No. 2 (Apr., 1940), pp. 177-206 Published by: University of Pennsylvania Press Stable URL: http://www.jstor.org/stable/2707332 . Accessed: 11/02/2014 20:58
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.

University of Pennsylvania Press is collaborating with JSTOR to digitize, preserve and extend access to Journal of the History of Ideas.

http://www.jstor.org

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

THE DEVELOPMENT OF SCIENTIFIC METHOD IN THE SCHOOL OF PADUA


BY JOHN HERMAN RANDALL, JR.

century had so The Aristotelian science which the thirteenth eagerly worked into its Christian philosophy of life aimed at an understanding of nature divorced from power over things. But more and more men began to hold that century duringthe sixteenth science should be directed,not merelyto understandingand vision, but to a kind of understandingthat mightgive power, action, and of the practical arts. A leading intellectualenteran improvement prise of the time was the search for a fruitfulmethod that could serve thisnew aim to whichknowledgewas turning. Those thinkers whose energies were not wholly absorbed by the theological issues in termsof whichthe major battles were still being fought,concentask trated on this problem of method as the paramount scientific of the day. methodwas finally"discovIronically enough,when the fruitful in turned out to be the least novel of and it ered" proved practice, all the elements that went into the formationof the new science. After exploringmany a blind alley, men came to realize that one of thegreat medieval intellectualtraditionshad already made an excellent beginningat just the kind of practical and useful knowledge and fourteenth centuryschools, theynow wanted. In the thirteenth therehad been worked out the idea of an experimentally grounded and mathematicallyformulatedscience of nature, and since then muchhad been done in the way of actual achievement. In Leonardo in the Italian mathematicians and physicistsof the the penetrating, in Copernicus,Kepler, and Galileo, such a science sixteenth century, had indeed come of age. strands,each with Into this sciencethere enteredmany different its own history. And the powerful stimulusimparted during the sixteenthcenturyby the recovery of the techniques of the Greek mathematiciansis not to be minimized. But the conceptionof the nature of science, of its relation to the observationof fact, and of that was the methodby whichit mightbe achieved and formulated, handed on to his successorsby Galileo, was not the work of the new method. It appears rather as the culminaseekers after a fruitful of ten generationsof scientistsinquirefforts tion of the coioperative
177

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

178

JOHN

HERMAN

RANDALL,

JR.

ing into methodologicalproblems in the universities of northern Italy. For threecenturiesthe natural philosophersof the school of Padua, in fruitful commercewith the physicians of its medical faculty,devotedthemselves to criticizing and expandingthis conception and method,and to groundingit firmly in the careful analysis of experience. It left their hands with a refinement and precision of statementwhich the seventeenth centuryscientistswho used it did not surpass in all their careful investigationof method. In contrastwiththis cumulativeand organized elaborationof the theoryand methodof science,the many humanistseekers,revolting fromthe scholasticismof the Scotists with their technical "terminist" logic, seem to have displayed all the customaryignorance and futilityof intellectual revolutionaries,and to have proposed new methodsdistinguished chiefly by the noveltyof theirignorance. As might be expected, these servants of the word for the most part sought their new methodin language and in rhetoric,and tried to erect a "natural dialectic" on the basis of Cicero and Quintilian. Others like Bruno were fascinatedby the suggestionsof Lully for a universal language that might reveal all truth. And still others, emphasizingthe place of a knowledgeof nature in human wisdom, urged men to close theirbooks and observe the world. The humanistsmight seek the method of a new science in the rhetorician'sart of persuasion; a Vives or a Bacon, recognizingno useful knowledgein the investigations of the mathematiciansand astronomersof their day, mightcounsel experience and ever more experience. Their combinedonslaughthelped to shake men's faith in the complacent academic traditionalismof the schools, already sorely disturbedby the new literaryand theologicalmovements;it hardly contributed much guidance to those already busily engaged upon scientific problems. Both in its traditionalinsightsand in its novel guesses the imagination needed the discipline of a critical methodbefore there could be any significant observation of facts. The body of ideas whichin Galileo and Descartes dared to arrogate to itself the name of natural science, and which in Newton definitivelymade good that proud claim, had other and far deeper roots, stretching back throughand beyond the twelfth-century European appropriationof ancientlearning. History has fallen into error in accepting uncriticallythe estimate the pioneer thinkersof the sixteenthand seventeenth century made of theirownturning away fromtheheritageof the past. Their

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

179

consciousnessof freshdiscoveryand radical reorientationobscured in materials, methods,and even the countless bonds of continuity, achievements, unitingthemto their predecessorsin the late middle ages. In particular the fact that the seventeenth centuryscientists, in revoltagainst the humanists' appeal to the authorityof the past, preferredto put their trust in "natural reason" alone, and hence cared nothingfor historical continuity, has sadly misled our judgmentas to thefashionin whichtheirthought was generated. Taking themat theirownword,we have assumed that that cooperativecriticism and reconstruction of a well-organizedsystemof ideas, shaken from time to time by fresh insights which have had to be worked into the logical structure-that that process which has since the seventeenth of the procedureof scienbeen so characteristic century tificadvance, played no part in its earlier stages. In the present generationmuch has been broughtto light about the organized scientific traditionsof the later middle ages in which the sixteenth and seventeenthcentury pioneers carried on their work. But muchmore remains to be done. In particular,the fact that several of the most influential investigatorshave been French has focused attentionon the activities of the University of Paris, whilethefurther fact thatmanyof themhave been Catholic scholars has made themnot undulyappreciativeof theworkof the free-thinking and anti-clericalItalian schools. For its part, Italian scholarand ship has been attractedby the spectacularhumanisticmovement by the presumablymorenovel and originalliteraryPlatonism of the Florentines. As a result,thoughit is clear that the thoughtof the Italian universities formstheimmediate of the sixteenthbackground century scientificmovement that culminated in Galileo, its substantial achievement has as yet receivedalmost no study. The basic idea of an experimentallygrounded science of the mathematicalstructureof nature appeared as soon as Europeans began to explorethewisdomof the ancients. It developedwithinthe of thefirst general framework body of ancientmaterialsto be assimilated, the Augustinian philosophy of reason-itself the platonized outcomeof Hellenisticthought. It drew specifically upon the Arabic versions of Alexandrian science, though direct contact with the whole of Greekmathematics, and mechanicswas the last astronomy, to be established; Archimedeswas not knowntill the sixteenthcentury. But the idea of such a science,and much of its method and concepts, were in the possession of Europeans from the twelfth centuryon.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

180

JOHN HERMAN

RANDALL, JR.

Aristotle's logic, his theoryof science and method,was discovered in the Analyticsduringthe first half of the twelfth century;his basic concepts and principles of natural science were learned from the Physics in the second half. The comingof Aristole introduced a body of materials too impressive to be ignored. Thereafterfor centuriesthe Aristotelianphysical writingswere taken as the starting-point for all natural science,howeverfar men mighteventually depart fromthem; and the Aristoteliantheoryof science,however men mightinterpret it, remained dominanttill the time of Newton. From the beginningof the fourteenth century, however,there set in of the Aristotelian tradia persistentand searchingreconstruction whendirectedto thePhysics,led by gradual stages to the tion,which, mechanicaland mathematical problemsof theGalilean age, and when directed to the Logic led to the precise formulationof the method and structureof science acclaimed by all the seventeenth-century scientists. There were two main criticalmovements duringthe later middle ages. The Ockhamitesbegan in Oxford in the thirteenth century, and while persistingthere found a new strongholdduring the next hundredyears in theFaculty of Arts at Paris. The Latin Averroists began in Paris in the thirteenth century,and shiftedtheir seat to Padua early in thefourteenth. Both set out by expressinga secular and anti-clerical spirit,and by undertaking a destructive criticismof Thomism and Scotism, the thirteenth centurysyntheses of science and religion. But both soon advanced beyondmere criticismto the constructiveelaboration of natural science: they became the two great scientific schools of the later middle ages. The original work of the Ockhamitesbelongs to the fourteenthcentury,that of the Paduans, to the fifteenth and sixteenth. The formerwas done in and the logic of continuity dynamics,kinematics, and intensity;the latter,in methodology and in the furtherdevelopmentof dynamics. Both turnedfromthe earlier religious synthesesto the purely natural philosophyof Aristotlehimself;and bothdeveloped primarilyby a constructive criticism of theAristoteliantextsand doctrines. The Ockhamiteswere at firstthe more "progressive" and "modern"; they were interestedin the free developmentof the Aristotelian physics,and their works take the form of questions and problems suggested by Aristotle's analyses. The Averroists, though much more secular and anti-clerical, were originallymore conservativein their attitudetoward Aristotle and his interpreterAverroes: their

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE SCHOOL

OF PADUA

181

works are characteristically on the texts. From 1400 commentaries on, however, they knew and taught all the Ockhamite departures from Aristotelian doctrine: Paul of Venice (*1429) is remarkably up-to-date,and his Summa Naturalis contains an exposition of all the ideas of the dynamicsof the Paris Ockhamitesand the Oxford logicians. The works of these fourteenth-century thinkers were printedin many editions so soon as the press reached Italy, all of them by 1490; and in the sixteenthcenturyit was primarily the Italians who advanced by successive stages to the formulationsof Galileo. About 1400,therefore, theinterestin thedevelopment of scientific ideas shiftsfromOckhamiteParis to the Padua Averroists. From the time of Paul of Venice to Cremonini (*1631) the Aristotelian physics and a nascent "Galilean" physicswere in definite and conscious opposition at Padua, and this critical conflictcontributed greatly to the workingout of the latter. Paul of Venice had been sent by his order to Oxford in 1390, where he remained for three years; he then taught for two more in Paris at the time of the last great Ockhamite,Pierre d'Ailly. He thus knew all the Ockhamite developmentsat firsthand, and explained them fully though critically in his encyclopedic writings. His successorat Padua, Cajetan of Thiene (*1465), was the most of the Averroists,and the most sympathetic radical scientifically to the Paris teachingson dynamics. He initiateda great controversy over the Calculations of Suisseth (Swineshead), in which all the argumentsfor a mathematicalas against a qualitative physics are in many ediexamined,so that the documentsof this controversy, worksprintedin Italy in the 1480's. The tions,were among the first fundamentalDe latitudinibusformarumof Nicholas of Oresme, in whichthe rule for uniformly accelerated motionfirstappears, came out in 1482, with a discussion by Blasius de Parma de Pelicanis; Albert of Saxony's Tractatus de proportionibus,arguing for a of qualities (already reportedin the Summa quantitativetreatment Naturalis of Paul ofVenice) also appeared in the same year; in 1496 it was reprintedwith the De intensioneet remissioneformarumof Walter Burleigh,a defenseof the logic of qualitativechange opposed to the spiritof Oresme,and witha fullreplyto Burleigh in behalf of quantitativeanalysis by the physicianJacopo da Forli. Among the of all thesedocuments, mostinteresting indicativeof a livelyconcern withwhat was to become the fundamentalscientific question,is the

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

182

JOHN HERMAN

RANDALL, JR.

bus of a Milanese physician, Johannes Tractatus de proportioni Marlianus (Pavia, 1482), which brings experimentalproof to bear on the quantitative side, describes the rolling of balls down an inclinedplane to measure theirvelocityand acceleration,and narrates withpendulums. experiments The question whetherthe operation of causes was to be formuor qualitatively (whetherthe "first accident" lated mathematically of substancewas to be taken as quantityor not-which happens to be also the way in whichKepler expressed his view that a cause is a mathematical law) was thusvigorouslydebated at Padua toward the and the notion of "cause" as a mathecentury, end of the fifteenth matically formulatedformal cause won many adherents. In the next centurytherebroke out another great controversyamong the Paduans as to whetherthe "cause" of natural motion was to be way of behaving soughtin a formor in a force,that is, in a definite way. Galileo joined those or in somethingthat acted in a definite who identified"cause" with a "force"; but since he also defined force in terms of its way of acting, his divergencewas not great. And towardsthe end of the same centurythereoccurredanotherdisfinalcauses had any place in natural philosophy. pute,as to whether The outcomeof these successive debates was to delimitthe conceptionof cause, and to make the Galilean positioninevitable. They are here mentionedto suggest certain other strands in the development whichthis study does not presume to set of Italian Aristotelianism forthin detail, and whichin particular illuminatethe change from a qualitative to a mathematicaltreatmentof natural operations. It has becomea recentfashion to view the whole "Renaissance, " and indeed the very "birth" of modernscience itself,as philosophically a turningfromthe Aristotle of the Schools to Platonism; and has been representedas domiof thefifteenth Italian thought century that the vigornated by that turning. But it must not be forgotten ous intellectuallife of the Italian universitiesremainedloyal to the century Aristoteliantradition. Now in most countriesthe fifteenth of the earlier philosophies,Scotism, saw the teachingand refinement with littlebasically new. But in northThomism,and Ockhamism, ern Italy, at Padua, Bologna, and Pavia, and to a lesser extent at Siena, Pisa, and the brilliantnew universityof Ferrara, Aristotelianism was stilla livingand growingbody of ideas. What Paris had what Oxfordand Paris togetherhad been in the thirteenth century, the centerin Padua became in the fifteenth: been in the fourteenth,

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

183

whichideas fromall Europe were combinedinto an organized and cumulative body of knowledge. A succession of great teachers carriedthatknowledgeto thepointwherein the nextcentury it could findfruitful marriage withthe new interestin the mathematicalsciences. In the Italian schools alone the emergingscience of nature did not mean a sharp break with reigningtheologicalinterests. To themit came rather as the natural outcome of a sustained and coof Aristotelianideas. If in the sixteenth operativecriticism century the more originalmindswere led to a formalbreak withthe Paduan we mustnot forgetthat even Galileo occupied a chair there teaching, from 1592 to 1610, and that in method and philosophy if not in physicshe remaineda typicalPaduan Aristotelian. That Italian Aristotelianism was thus able to lead the European schools in the fifteenth and sixteenthcenturieswas due to several circumstances,not the least of which was the settled commercial prosperitythe Italian cities had now achieved. They had long enjoyed and taughtin theiruniversitiesa thoroughly secular and anticlerical philosophyexpressive of the new culture of a this-worldly and commercialsociety. By 1400 that nice blend of Aristotelianscience and Christian faith which Thomas and Duns Scotus had constructedhad, in Italy at least, retreated into the monastic orders. At Padua, Bologna, and Pavia therereignedan Aristotelianism that made little attemptto accommodateitself to theological interests. And it is no accidentthat while the Church-controlled science of the North drove all those who felt the new currentsinto open rebellion against scienceitself,the anti-clericalscienceof the Italian universities could progress steadilyin self-criticism to the achievementof a Galileo. Fundamental also was the close alliance between the study of Aristotleand the study of medicine. At Paris the Faculty of Theology crownedthe Sorbonne; at Padua the Faculty of Arts led only to that of Medicine,and Aristotlewas theretaughtas a preparation, not for an ecclesiasticalcareer,but for the studyof medicine,with a his natural hisconsequentstrongemphasis on his physicalwritings, tory, and his scientificmethodology. A physician's Aristotle is bound to differfroma theologian's. The teachers wrote no theological works,no commentarieson the Sentences. They normally held medical degrees themselves; they applied Aristotle to medical problems,and to questions of method arising in medical science; him in the light of the best medical writersof the they interpreted Greek and Arabic traditions.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

184

JOHN HERMAN

RANDALL, JR.

Finally, the liberty of teaching and speculation guaranteed by Venice,theleading Italian anti-papal and anti-clericalstate,afterits acquisition of Padua in 1404, attractedthe best minds fromall over Italy, especially the philosophicalSoutherners. Padua remainedto schoolof Europe, the strongthedays of Galileo theleading scientific hold of the Aristotelianqualitative physics,and the trainer even of thosewho were to break withit. Cusanus, Purbach, Regiomontanus, Copernicus,as well as the Italians, all studied at Padua. If the concepts of a mathematicalphysics were arrived at by a of Aristotelianideas, the "new method," the logic and long criticism methodologytaken over and expressed by Galileo and destined to methodof the seventeenthcenturyphysicists, become the scientific century as contrastedwiththe manynoisy proposals of the sixteenth "buccinators" down to Francis Bacon, was even more clearly the of the Aristoteliantheory critical reconstruction result of a fruitful and fertilizedby the in Padua particular, undertaken at of science, methodological discussions of the commentatorson the medical writers. For threehundredyears, afterPietro d 'Abano broughtthe problemsto the fore, the Paduan medical teachers were driven by protheir texts, especially Galen, to a careful analysis of scientific on Galen, Jacopo da Forli (*1413), cedure. The great commentators wrotewidelyon the methodsof the Paris physicists, who incidentally and Hugo of Siena (*1439), gradually built up a detailed theoryof methodwhichthe Aristotelianscholars,themselvesholders scientific of medical degrees,incorporatedinto their version of the nature of science. It is possible to trace step by step in rather beautiful fashion the gradual elaboration of the Aristotelian method,in the light of the medical tradition,from its firstdiscussion in Pietro d 'Abano to its completedstatementin the logical controversiesof Zabarella, in which it reaches the formfamiliar in Galileo and the scientists. seventeenth-century This discussionof methodwas carried on in termsof the general in formalstructureof science and its relation to the subject-matter it As is such applicable and exemplified. discovered whichit is to be equally to all four of the Aristoteliankinds of cause; and hence it does not insist,any more than the extensivediscussions of the seventeenth century,that the principles of natural science be strictly mathematicalin characterand that its formalstructurebe stated in mathematicalterms. This mathematicalemphasis is an important elementin Galilean science. It too has a long and gradual develop-

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

185

anidsixteenthcenturies,as has been mentat Padua in the fifteenth already indicated; and thesediscussionsinvolveboththe Oxfordlogic of quantityand theissues it raised, and the recoveryof Greekmathematical texts and techniques in the sixteenth century. But this seems to be another part of the story,best told in connectionwith the particular treatmentof particular natural questions. It would set of Aristotelian texts and problems, and a involve a different group of passages; it seems best to present the wholly different theoryin its own terms,and to postof methodological development pone this otherimportantstrand. These men were concernedwiththe discoveryand use of "princito Newfromthetwelfth century ples" in science. Like all thinkers as that as Aristotleunderstoodapxii, ton,theyunderstoodprincipium fromwhicha thingproceeds and has its originin any way whatever; in science a principiumis that which is in any way a specifically, of understanding. The particular character of "priority" source theoryderiving and origin,of whichboth Aristotleand the scientific from him distinguishedvarious kinds, depends upon the specific characterand contextof the problemunder consideration. The diskinds of order and method,therefore, betweenthe different tinctions runningthroughall the men here considered,are ipso facto distincdifferent kinds of "priority," that is, between differtions between1 ent meanings of principium. The question of methodwas raised, and the termsin whichit was by Pietro to be treated for three centurieswere clearly formulated, et praephilosophorum, d'Abano in his Conciliator differentiarum in 1310. In discussingthequestionwhether written cipue medicorum, medicineis a science he points out that "science" is used in two senses. infers inthemost which theconclusion sense is that through Science proper in like that sciencedefined and immediate, causes whichare proximate unthatwe knowa thing L. I, c. 2 [71b]; "We think Posteriora, Analytica and accidental manner, and not in a sophistical (simpliciter) qualifiedly of which(propter quam) the we know thecauseon account whenwe think thatit is thecause of thatfact,and thatit couldnotbe otherfactexists, "demonstration is gainedfrom quid" propter kindofscience wise"; andthis or why) or what Galen called "doctrinacomwherefore (demonstration senseofsciis a second wayofteaching). There positiva"(thecompositive since most us canbe saidtobe for proper; is alsoproper, andindeed encethat certain and knowable whatis more from forus thenatural wayis toproceed

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

186

JOHN HERMAN

RANDALL, JR.

way of teaching).1

forus to whatis more in theorder knowable ofnature:see thebeginning of thePhysica[184a]. When, in caseswhere effects inhere in theircausesaeto an essential cording order ofpriority, we arrive order by theopposite at thecausewe are seeking, and logically through proximate immediate middle an effect terms;or whenwe conclude from moregeneralcauses,omitting certain intermediate we acquireknowledge causes, by"demonstration quia" (demonstration resolutiva" that)orwhatis called"doctrina (theresolutive

Pietro is here making his central distinction between two kinds of scienceand demonstration in termsof the theoryof science developed in the Posterior Analytics. Science is definedas a demonstrative knowledgeof thingsthroughtheircauses; its instrument is the demonstrativesyllogism,which establishes the relations between causes and theireffects.2The problem of formingsuch syllogisms is the problemof discoveringcauses and defining themin such a way that theycan serve as the middle termsof demonstrations.3Pietro is thus distinguishing between two kinds of proof: that of effects throughcauses, and that of causes througheffects. The transformation of the demonstrative proof of causes into a method of discovery is precisely the achievementof the Paduan theoryof science. That change of contextis already suggested,in the above passage, in Pietro's identification of the two kinds of with the various ways of teaching. or doctrina. and demonstration
Op. cit., Diff.3, Prop. 1; (ed. Venice 1496). The two formsof "demonstrareferto An. Post. I, c. 13, whereAristotleis contrasttion quia" here distinguished ing scientific knowledgeof the wherefore(T' lOtTl) with scientific knowledgeof the formis illustrated that (To o'tL). The first by the proof,What does not twinkleis near, the planets do not twinkle,they are therefore near. "This syllogismproves not 'the wherefore' but only 'the that'; since theyare not near because they do not twinkle,but because they are near they do not twinkle." (An. Post. 78a.) The demonstration corresponding propter quid, followingthe causal order rather than the cause, makes the cause, "nearness,"the middle our "natural way" of discovering term: the planets are near, hence theydo not twinkle. The otherform of "demonstration quia" is the one where the middle term is more general than major or minor,and does not exhibitthe proximatecause. Aristotle'sillustrationis, What no wall can breathe. can breatheis an animal,no wall is an animal,therefore 2An. Post. 71b. 3 "In all our inquiries we are asking whetherthere is a middle term or what the middletermis: for the middleis preciselythe cause, and it is the cause we are seekingin all our inquiries. Thus, does the moon suffer eclipse) means,is thereor is therenot a cause producingeclipse of the moon? and when we have learnt that thereis, our next questionis, what thenis this cause?" (An. Post. 90a)
1

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

187

of order or dependence. These termsare drawn fromothersources than Aristotle, and they bring with them an interest that is methodological as well as purely logical. Pietro treats them at lengthin Difference8. The Tegni (Techne) or art of medicine of Galen commenceswith a prologue in whichthree doctrinaeor ways of teaching medical science are distinguished: by resolution,by composition, and by definition. In all thewaysofteaching (doctrinae) which follow a definite order there are three of procedure. One of them orders is thatwhich theway follows of conversion and resolution (dissolutio) ; in it you set up in yourmind thething at which and ofwhich a scientific youare aiming, youare seeking knowledge, as theend to be satisfied.Thenyou examine whatlies nearest to thatwithout to it,and nearest which cannot thething exist;norare you finished till you arriveat the principle whichsatisfies it. . . . The second of thefirst follows theway of comtposition, and is thecontrary way. In it at which youbegin with thething youhavearrived bythewayofresolution, and put themtogether and thenreturn to theverythings resolved, again untilyou arriveat thelast of them. (compone eas) in theirproperorder, And thethird theway of analysing thedefinition.4 follows The firsttwo doctrinaehad been identified by the Arabic commentator on Galen, Hali ('Ali ben 'Abbas, *994), with the two Aristhat whichproceeds fromeffects toteliankinds of demonstration, to causes, demonstrationquia (6rt), and that which proceeds from demonstrationpropter quid (6t6rt causes to effects, ).' This dividistinction sion, however,was naturallyconfusedwiththe threefold made by Averroes in the Prohemium to his commentaryon Arissimvliciter. or of cause and beina. totle's Phusica. into demonstration
Hali, translatedfrom Galieni principis medicorum Microtegnicum commento the Arabic by ConstantinusAfricanus; no date or place of printing,but prior to 1479; Texts I, II, III, IV. 5 Ibidem,Hali's comment are carriedout in these on Text III: "Demonstrations and demonstraby composition, quare is effected two doctrinae; but demonstration tion quia by resolution (dissolutio)." Cf. Conciliator,Diff. 8, Introd.: "As Hali propter quid, the resolutive by demonstration way is effected says, the compositive are to be assumed." quia; only these two kinds of demonstration by demonstration is given as An. Post. II, 1 (89b); Pietro The Aristoteliansource of the distinction explains it: "It is writtenin the beginningof the second book of the Posterior Analytics: that the kinds of question we ask are as many as the kinds of thing which we know. For we know only what we have already asked about. And the a thingis, and what it is; kinds of questionwe ask are four: two simple,whether namely,that (quia) it is, and wherefore(propter quid) or why and two composite, (quare) it is." (Conciliator,Diff.8, Prop. 3, ii.)
4

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

188

JOHN HERMAN

RANDALL, JR.

in whichas in mathematicsthe causes are firstboth for us and for the order of nature; demonstrationpropter quid, or of cause, in which,as in natural science, we start with what is firstin nature for us; and demonstration but not first of being,or of sign, in which to arrive at causes.6 The distinctionbetween we start with effects to causes and that fromcauses the two procedures,that fromeffects to effects,is thus Aristotelian. Averroes set the procedure of mathematicsofffrom the a priori procedure of demonstrationin physics; fromGalen, and also fromthe rhetoricalmethodof Cicero and Boethius,7came the terms resolutiveand compositiveto designate these two procedures. of doctrinaestill remained a distincFor Pietro, this distinction or compositive, kinds of of whichthe first, between two science, tion was alone science in the strictestsense; the second or resolutivewas science only because of "the weakness of the human mind," which in natural science has to start with experiencedeffectsto discover their causes or reasons why (propter quid). There are really but for the a priori methodof mathematics two kinds of demonstration, and that of physicsare but two species of the one kind,demonstration propter quid. The physicianJacopo da Forli (*1413), who occupied in turnthe chairs of medicine and of natural philosophy at Padua, followed betweenthe two doctrinae,comHali and Pietro in theirdistinction positive and resolutive.8 He added, however,a furtheranalysis of
6 "Averroes says that there are three kinds of demonstration: demonstration causae or propterquid; and demonstraor causae et esse; demonstration simpliciter, . . . Demonstration occurs when cause and effect simpliciter tion esse or significati but rarely occurs in natural are equally known, which happens in mathematics, science. Whence the Commentator [Averroes] says in the beginning of the Physics: 'If it happens that what is known to us is known by nature, then the in this science will be of cause and being. If it happens that the demonstrations thingsknownto us are not knownby nature, and are not prior in being but posin that science will be of signs (signorum) and not terior,then the demonstrations simpliciter. For in natural science the thingsknown to us are not demonstrations that is, by nature; the contraryobtains in mathethose that are knownsimpliciter, and are causes matics. The thingsthat are knownin the latter science simpliciter, prior in being,are knownto us."' (Conciliator,Diff.8, Introd.,Prop. IV.) 7 "Likewise Cicero and Boethius in theirTopics divided the power of definition whichis subdividedinto resolutiveand into two parts: one dealing with judgment, compositive." (Conciliator,Diff.8, Introd.) way, or path by whichthe teacher the manner, Haly understands 8 "By doctrina or definitive." (Jacobi compositive proceedsin teachingthe pupil, eitherresolutive, de Forlivio super Teqni Galeni,Padua, 1475; comm.Text I.)

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

189

the method of resolution which brings it closer to a procedure of investigation. though naturalor real,and logical. Real resolution, Resolutionis twofold, in many senses,is strictlythe separation and division of taken improperly parts. Logical resolutionis so called metaphoria thinginto its component cally. The metaphoris derived in this fashion: just as when something compositeis resolved,the parts are separated fromeach otherso that each is made, is leftby itselfin its simplebeing,so also when a logical resolution so that the a thing at firstunderstoodconfusedlyis understooddistinctly, grasped. Thus if when parts and causes touchingits essenceare distinctly you have a feveryou firstgrasp the concept of fever,you understandthe the feverinto its causes, feverin generaland confusedly. You thenresolve since any fevercomeseitherfromthe heatingof the humoror of the spirits or of the members;and again the heatingof the humoris eitherof the blood or of the phlegm,etc.; until you arrive at the specificand distinctcause and knowledgeof that fever.9

witha numberof examples drawn illustratesmost of his distinctions


from medicine.

This is a clear case of the method of medical diagnosis;

Jacopo

9 Ibid., comm.Text I. 10"By the way of resolutionwe are to understanddemonstration quia, that is, the knowledgeof an effect proceedingto a knowledgeof its causes; and conversely, we are to understanddemonstration by the way of composition propter quid. The firstis from the 'notion,'that is, from the knowledgeof the end, that is, of the effect. This comes fromthe 'dissolution,'that is, fromthe resolutionof the effect into its causes. The second way comes fromthe composition of what has been discovered,that is, of the causes discoveredby resolution. For those thingsthat are discoveredby resolutionin demonstration quia are afterwardsput and joined togetherin a demonstration propterquid, until we arrive at the immediatecause, and concludethe effect. This expositionis approved by the Conciliator,Diff.8, and by Gentilis,and by othermodernssince them." (Ibid., comm.Text I.) 11"As all demonstration is eitherthrougha cause or through an effect, doctrina, whichis the settingforthof what is demonstrable, can be carried on only in one of these two modes, that is, by resolutionor by composition." (Expositio Ugonis Senensis super libros Tegni Galieni, Venice, 1498; comm.Text I.)

seems to have crystallizedin his usage.'0 Hugo of Siena (*1439), teacher of medicineat Padua, Ferrara, and Parma, is still more concerned with methodology. Starting doctrinaas the settingforthof what is demonwithGalen, he defines strable (manifestatiodemonstrabilis);" it has two modes,resolution and composition, and in a complete science like physics or medicine it is impossible to use only one method; both are required,

The methodological sense of the traditional terms

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

190

JOHN HERMAN

RANDALL, JR.

quia, and in of causeswe use demonstration becausein theknowledge quid. It is propter we use demonstration of effects knowledge thescientific are necessary, as wellas thatbothoftheseprocedures opinion thecommon of manydefinitions.12 the explanation seek the cause, methodwill begin withthe effects, A proper scientific fromwhich it started by that cause. and then explain the effects
cause.

to effect from term or cause we proceed of themiddle In thediscovery


. .

morecomwhich is commonly an effect, from we proceed in thatdiscovery of the to a cause whichis simpler;and becauseby thisdiscovery posite, the cause,we say that demonstration the effect through cause we certify knowledge.... of resolutive pro pterquid and of cause is the foundation their cause through ofa science of effects But I myself see in thediscovery of a scientific and likewise in the discovery a doubleformof procedure, is the distheireffects.The one procedure of causesthrough knowledge of its forth of the middletermor cause,the otheris the setting covery ofdiscovery in thecase ofdemonor effects. And theprocess consequences forth the consewhilethatof setting causesis resolutive, stration through it is just theother In demonstration effects through is compositive. quences wayaround.'3 Thus Hugo refuses to separate the two procedures: in any science discovery or inventio and setting forth and in any demonstration, both enteras successive phases of the the consequencesor notificatio method to be employed. Discovery and proof are both essential momentsin all method. method had alThis notion of a "double process" in scientific by Urban the Averroistin his large commentary ready been set forth on thePhysics in 1334. Following Averroes,he distinguishesthree simpliciterof mathemodes of demonstration:the demonstration matics,in whichthe principlesdo not have to be sought after; demby which the onstrationof sign, proceedingfrom observed effects, physicistlearns the causes of natural things; and a thirdkind, theyare always though causeswhich, from whichproceed demonstrations and less known often posterior are and naturamn, known quoad more prior priorfor in in those things from to us. Thisoccurs naturalscience, which are posterior which causes, investigate their effects, we modes are us,whose resolution. method of of the and less known And is the way this to us.
TextI. 12Ibid., comm. 13 Ibid., comm.Text I.

because . Such a way of acquiring knowledgewe call resolutive,

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

191

But after we have investigated the causes, we demonstratethe effects through those causes; and this is the way of the method of composition. Thus physical demonstrationsfollow after mathematicaldemonstrations in certainty,because they are the most certain after those in
mathematics.'4

Paul of Venice (*1429) examines still more closely this double procedure in physical demonstrations, defending it against the charge of being what Aristotle called a circular proof. Scientific knowledgeof the cause depends on a knowledgeof the effect, just as scientific knowledgeof the effect depends on a knowledgeof the cause, since we knowthe cause throughthe effect beforewe know the effect throughthe cause. This is the principal rule in all investigation, that a scientific demands a prior knowledgeof their knowledgeof natural effects causes and principles.-This is not a circle, however:-In scientific procedurethereare threekinds of knowledge. The first is of the effect without any reasoning,called quia, that it is. The second is of the cause through knowledgeof that effect;it is likewise called quia. The third is of the effect throughthe cause; it is called propter quid. But the knowledgeof why (propter qtid) the effect is, is not the knowledgethat (quia) it is an effect. Therefore the knowledgeof the effect does not depend on itself,but upon somethingelse.15 The Commentatorrecognizes a double procedure in natural science. The firstis fromwhat is less known to nature to what is more known to nature,and is fromeffect to cause. The secondis fromwhat is moreknown to nature to what is less knownto nature,and is fromcause to effect.... Natural sciencebeginsbothfromthe causes and fromwhat is caused, but in different senses. It beginsfromthe causes "inclusively," that is, by knowing them; but fromthe things caused "exclusively," that is, by knowing by means of them. . . . There is thus a twofoldknowledgeof every cause, the one kind by the procedurequia and the otherby the procedurepropter quid. The second kind depends on the first, and the firstis the cause of the second; and thus the procedurequitais also the cause of the procedure propterquid.'6 During the fifteenthcentury attention was increasingly focused on this "double procedure" involved in scientific method. It came
14 Urbanus Averoysta philosophus Summus. . . commentorum omnium Averoys super librum Aristotelis de physico audito expositor,Venice, 1492; comm. Text II. 15 Summa philosophiaenaturalismagistri Pauli Veneti,Venice, 1503; I, cap. ix. 16 Expositio Pauli Veneti super octo libros Physicorum Aristotelis, dated 1409; Venice, 1499; I, comm.Text 2.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

192

JOHN HERMAN

RANDALL, JR.

to be knownby the Averroisticterm "regress"; the dependenceof on theprior investigationand establishment all strictdemonstration of the appropriate principleswas stronglyemphasized,and the dewere carefullyexamined. The outstandtails of that establishment ing natural philosopher of the middle of the century,Cajetan of Thiene (*1465), repeats Paul's treatment.'7 The fullestaccount of these problemsis to be found in the commentaryof Agostino Nifo on the Physics (1506). After explaining the three kinds of demonstrationin Averroes's prohemium,and assigning to natural philosophythe two procedures,"the one from the effectto the discovery of the cause, the other from the cause '18 Nifo takes up at once the question discovered to the effect," whetherthis is a circular proof, and cites Philoponus, Themistius, and Averroes in defenseof such a regressus. are four kindsofknowlthatthere maintain (recentiores) Recent writers or observation; the thesenses, through kindis oftheeffect edge. The first which the effect, is of thecausethrough (inventio) second is thediscovery of the same cause called demonstration of sign; the thirdis knowledge from which there first bytheintellect, (negotiatio) an examination through to serveas the ofthecausethatit is fit suchan increased comes knowledge is a knowledge the fourth of simpliciter; middletermof a demonstration thatcause known so certainly as to propter quid,through thatsameeffect term. be a middle Since the second knowledge of the effectdifferswidely from its initial observation,this is no circle,but rather a "regress." Nifo then asks, what is this examination or negotiatio by the intellect? nor is it It is clearly neither a demonstrationnor a definition, induction. and division. For whenthe cause itself is composition This negotiatio and dividesuntilit knowsthe composes theintellect has been discovered, term term. For though causeand middle be ofa middle causein theform differ in their form thesamething, (ratio). For it is calledthecause they it be better known from than it,whether proceeds in as muchas theeffect in as muchas it is a definition. term or not. But it is a middle the effect of discovering the cause; negoto cause is thustheprocedure Fromeffect term and a definition. But thecause as a middle toward tiatiois directed
Gaietani de Thienis, Recollecte super octo libros Physicorum Aristotelis, Venice,1496; Lib. I, Quest. v.
17

Venice, 1552; I, com. Text 4.

18Augustini

Niphi

philosophi

suessani

expositio . . . de

Physico

auditu,

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

193

is discoveredonly throughcomposition and division,it is since a definition throughthem that the cause is discoveredin the formof a middle term, fromwhichwe can thenproceedto the effect.'9 Nifo later added a further recognitio in which he suggests another view of what he calls "demonstrative regress." It is customaryto treat at length the regress in physical demonstrations; I say "physical," because there is no regress in mathematics. In this difficulty the most recent writers (iuniores) conceive three kinds of regress. The firstis knowledgethat the knowledgein the demonstrative is true; and the effect effect is (quia), i.e., that the propositionsignifying this knowledge comes from the senses. For instance, that man has the capacity for science is knownby sense. The second kind of knowledgeis of the reason why (propter quid) what is observedby sense is so. Thus we considerthe reason why man has the capacity for science,and not the brute; and we say, because he has a rational soul. Thereforeof the effect, thereare two kinds of knowledge: the effect, or of the proposition signifying the one, that it is true, and this is clear to sense; the other,why it is so, and this is knownto us throughthe discoveryof the cause. Of the cause, or of the propositions the cause, thereis but one kind of knowlsignifying edge, and this is discovery(inventio), whichis nothingelse than that it is the cause, or that the propositionssignifying the cause are true. Hence these writersconceivethat throughthis knowledgewhich is the discovery of the cause, or that the propositions whichsignifythe cause are true,there is so, or why the conclusion which is learned the reason why the effect is true. Thus in the regressin physical demonstration signifies the effect while the of whichtwo are of the effect thereare threekinds of knowledge, third is the discoveryof the cause. When the last is related to the effect, is so; but whenit is related to the cause, it is it is the reason whythe effect is made throughthe effect. the fact that it is the cause. And this discovery

It is significant that Nifo cites his examples from the History of Animals, the most empirical of the Aristotelianwritings.
From this it is clear that there is no need of any negotiatioto render held; for the mere greater our knowledgeof the cause, as we formerly is so. Yet when knowledgethat it is the cause is the reason whythe effect of and the commentaries I more diligentlyconsiderthe words of Aristotle, Alexander and Themistius,of Philoponus and Simplicius, it seems to me the firstprocess,by that in the regress made in physical demonstrations which the discovery of the cause is put into syllogisticform,is a mere it the discoveryof the since through hypothetical(coniecturalis) syllogism,
'9 Ibidem.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

194

JOHN HERMAN

RANDALL, JR.

causeis syllogized in a merely conjectural fashion. But thesecond process, bywhich is syllogized thereason whytheeffect is so through thediscovered is demonstration cause, propter quid-not thatit makes us know stmpliciter, but conditionally (ex conditione), thatthatreallyis the cause,or provided provided thatthe propositions are truethatrepresent it to be the cause, case the scienceof natureis not a scienceat all. We mustsay thatthe scienceof natureis not a sciencesimpliciter, like mathematics. Yet it is a science propter quid,becausethe discovery of the cause,gainedthrough a conjectural is thereasonwhytheeffect is so. . . That somesyllogism, as thatan effect thing is a causecan neverbe so certain, exists;fortheexto the senses. That it is the cause remains istence of an effect is known evenifthatconjectural existence is better known thantheeffect conjectural, in theorderof knowledge itself propter quid. For if thediscovery of the thereason is so is alwaysknown. Hencein causeis assumed, whytheeffect thathe is notsetting forth Aristotle thetruecausesof theMeteors grants butonlyin so faras was possible forhim, natural and in conjectural effects, fashion.20 or hypothetical
Here, then, at the beginning of the sixteenth century we find plainly set forth a formulation of the structure of a science of hypothesis and demonstration, with the dependence of its first principles upon empirical investigation. This was the one element in the Aristotelian theory of science that had remained obscure. The Posterior Analytics had seemed to say that while the principles and causes in terms of which a given subject-matter might be understood were to be discovered through sense-experience, they were seen to be true by vovs, by sheer intellectual vision. The scholastic theologians, like Thomas and Duns Scotus, had been led by their Augustinian Platonism to emphasize this power of intellectus to recognize the truth of principles. It is significant that at no time do the Paduan medical Aristotelians attribute any such perceptive power to intellect. The method by which principles are arrived at is rather the guarantee of their validity; they are "dependent" on that method, and it is the "cause" of their explanatory power. Nifo has merely made explicit what is implicit in the long previous discussion. Each of the major logicians of the century added his I t to the rounding out of this conception. Achillini and Zimara both sharpen the distinction between mathematics, demonstrated a priori, and a science of nature which is demonstrated a vosteriort. even in the
20

and that nothingelse can be the cause.

. .

. But you object that in that

Ibid., comm.Text 4, Recognitio.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

195

second part of its procedure,fromcauses to effects, or propterquid. Zimara makes a new distinction between order,concernedwith the teaching and exposition of the subject-matter, and method, concerned with the discoveryand demonstrationof its principles and properties. The order of exposition,Achilliniholds, should follow the order of nature. Though,as a Hellenist, Simon Porta, pupil of Pomponazzi and Achillini,brushes aside the traditionallong discussions of method,he still accepts the "double method" of natural science,following Alexander as the best interpreter. Natural things can be considered either as theyhave been made and generated,or as theyare; in the latter investigationwe must proceed by dividing, fromeffects. He admits the "regress" as the procedureto be followed in physics: firstthere is the way of discovery, by whichalone principlesare learned fromthe investigationof experience; thenthe way of science and judgmentby which we explain that experience. Pomponazzi' s commentary on the Physics remains unpublished;21 in his printed works his treatmentof method is psychological ratherthan logical. His whole argumentfromthe necessity of sense-imagesin all knowingleads him to emphasize the intimate union of the particular and the universal,and the necessitynot only " withthe former of starting but also of returning to it by a " reditus. Like most of the later Paduans, he follows Alexander rather than Thomas and Averroes in maintainingthat the intellectmust know in order to abstract universals fromthem. particulars directly, the cogitative Since the humansoul graspsthe singularfirst through and thentheuniversal theintellect, it in the power, through contemplating thesense-image, thatis known it trulymakesa resamesingular through a conversion; and consequently since fromthe singular turn (reditumn), the same soul returns the intellect a sense-image known through through to thesamething.22 The two operations of the intellecthe takes to be compositionand disjunction. Bernardinus Tomitanus (*1576), holder of the chair in logic at Padua for years and teacher of Zabarella, carried the development a step farther. He was almost wholly preoccupied with questions of method. In physics both discovery and demonstrationproceed
A manuscriptat Arezzo is describedby Fiorentino in his Studi e Ritratti, pp. 63-79, but nothingtouchingon methodis quoted. animae,cap. xii, 3. cf. Apologia, I, cap. 3. 22De immortalitate
21

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

196

JOHN HERMAN

RANDALL, JR.

fromsigns,that is, fromparticular effects;fromuniversals to particulars is the way of explaining and stating what has been discovered in the contraryway. He vigorouslydefendsthe "regress" as themethodof natural science; it is the combination of demonstration quia and propter quid. And for the firsttime demonstrationquia is formallyidentified with "induction" as the way of inquiry (inthe quisitio). Without use of the method of induction,or of the "regress," there could be no science of nature at all for the Peripatetic. With the same simplicity and luciditywithwhichhe summedup the collectivewisdom of the Paduan school on all their other problems, Zabarella formulatedthe classic version of their teaching on method,in the terms and with the distinctionslater so fruitfully employed and consciously expressed by Galileo. Out of this long and patient critique of the Aristotelian theory of science there developed at last the method that was to issue in fresh triumphs over nature. Logic Zabarella regards,followingthe Greeks,not as a science,but purelyas an instrument. The wholetreatment of logicis aboutsecond but theseare our notions; and by our will can either own work, be or not be. They are therefore notnecessary but contingent, and hencedo not fall underscience, things, sincescience is onlyof necessary things.23 Logic is a tool, sought not for its own sake but for its utility in science. furthering intellectual stateof mind(habitus)or instruLogic is an instrumental from mental created thepractice discipline, by philosophers of philosophy, in the concepts of things, and makesthem which constructs second notions the truemaybe knownand disintoinstruments by whichin all things thefalse.24 from tinguished In this sense it is like the other instrumental discipline, grammar. But logic is really twofold,"one applied to things,and already put to use; the otherseparate fromthings"; and the formeris identical withscienceitself. For the sciencesare "nothing otherthan logical methodsput to use."25 For Zabarella, therefore, logic and method are interchangeable terms; and he criticizes the logicians, who ought to treat method
De Natura Logicae, L. I, G. iii. Ibid., I, C. xx. 25 De Methodis, L. I, c. i.
23 24

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

197

forleaving it ratherto the physicians,who in turn neglect carefully, Aristotle for their cherished Galen. "Method is an intellectual instrument producingknowledgeof the unknownfrom the known. . . .Method has the force of inference,and connects this with that."26 Method is therefore the same as the syllogism: "the definition of method does not differfrom the definition of the syllogism." The syllogismis, in fact,"the commongenus of all methods " 27 Since all necessaryconnection and logical instruments. is causal in character,all methodmust be eitherfromcause to effect or from effect to cause. There are hence only two possible methods,composition and resolution. For all scientific progress from theknown to theunknown is either from to cause. The former cause to effect or fromeffect is the demonstrative which method, thelattertheresolutive; there is no other procedure generfrom to ates a certain knowledge of things. For if we progress something therecannot be neither is thecause of the other, something else,of which henceno certain between themany essentialand necessary connection; can be can follow from knowledge thatprogress.It is thusclearthatthere
no scientific and the resolutive.. methodexcept the demonstrative

method from that monstrative is a syllogism generating science propositions and the causesof the conclusion. are necessary, better immediate, known, method is a syllogism consisting of necessary . . . Resolutive propositions, and effects thatare better known thatare posterior which leadsfrom things and causes.28 ofprior to thediscovery things Zabarella makes the Averroistic distinctionbetween the resolutive methodsuitable for natural science and the "analytic" method of mathematics. In the latter, both the principles and the conseand are co6rdinate,so that whether quences have the same certainty we startwiththe one or the otheris a merelytechnicalquestion. In natural science,however,we must start witheffects observedby the senses. of our mindand powersthe principles Since becauseof the weakness from which is to be madeare unknown to us, and sincewe demonstration theunknown, we are of necessity forced to resort to a cannot set out from whichis the resolutive method thatleads to kindof secondary procedure, are found we can demonstrate of principles, so thatoncethey thediscovery method them. Hencetheresolutive is a subordinate effects from thenatural
26

. De-

De Methodis,III, caps. ii, i. Ibid., L. III, iii. 28 Ibid., III, xvii xviii.
27

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

198

JOHN HERMAN

RANDALL, JR.
.

of things is a perfect science,whichis knowledge method monstrative is discovery method theircauses; but the end of the resolutive through that their effects we seekcausesfrom byresolution since rather thanscience, from their causes, notthatwe mayrest know theeffects we mayafterwards thatif in coming ofthecausesthemselves.. . It is certain in a knowledge of a knowledge in possession of all its prinwe werealready to anyscience wouldthere be superfluous.29 resolution ciples,
There are two kinds of resolutive method, two ways of discovery, which differgreatly in their power.

. procedure,and the servant of the demonstrative.

. The end of the de-

in the performance of its from which effects, The one is demonstration of and we employ it forthe discovery efficacious; function is exceedingly thosethingsthat are veryobscureand hidden. The otheris induction, forthediscovform of resolution, and is employed weaker which is a much but need onlyto be whichare hardlyunknown, ery of onlythosethings clearer. madea little
By induction we discover only those principles that are "known secundum naturamr," that is, that are sensible, in that their instances or singulars are perceived by sense. By demonstration a signo we can discover those principles that are "unknown secundum naturam'," that is, the instances of which are not sensible, and hence can only be inferred from their effects, like first matter or the prime mover. Zabarella is here following his teacher Tomitanus in making "induction" a form of the method of resolution. His distinction between the two kinds of resolution in terms of the two kinds of principle discovered is that between arriving at a formal and at an explanatory principle; it is analogous to that made in the next century between the laws of motion and the forces causing acceleration, between the mathematical principles of natural philosophy and the forces of inertia and gravitation-both of which Newton likewise "'deduced from phenomena. "

demonstration from is distinguished then, induction differences, By these posterior things ofgoing from method For eachis a resolutive from effects. to us. The onekind are presented ofprinciple to principles.But twokinds needsno logicalinstrument and hence "naturally" [i.e.,bysense], is known alonesuchprinciples can be madeknown. For induction, by which except takesits originfrom all our knowledge sense,nor can we knowanything it first unless we haveknown with ourminds by sense. Henceall principles
29

Ibid., III, xviii.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

199

. knowingitself,whichare said to be indemonstrable.

notsaid and are therefore to us by induction, ofthiskindare madeknown onlyare said to be proved, or proved;forthosethings to be demonstrated something else. But through whichare demonstrated strictly speaking, sense else; in a certain something through doesnotprovea thing induction is notdistinguished itself. For theuniversal through it reveals thatthing but onlyby reason. And sincethe itself, in thething from theparticular becauseit is said thanas a universal, known as a particular thing is better induction is thusa process and notas universal, to be sensible as particular in thataspect thesamething to thesamething-from from thesamething in thataspect ofthatsamething to theknowing obvious, in which it is more not onlyare theprinand hidden. Therefore in whichit is moreobscure of science or of but also theprinciples known by induction, ciplesofthings method, by the resolutive method, then,are discovered the demonstrative
a some by inductionalone, othersby demonstration
signo.30
.

. The principlesof

The originalityof Zabarella, and of the whole developmentof is thus to set offa "scientific experiwhich he is the culmination, ence" from mere ordinary observation,the accidental or planless collection of particular cases. The weakness of the logic of the Schoolmen had lain preciselyin their acceptance of firstprinciples established by mere common observation. In contrast,Zabarella, and with him the whole new science, insisted that experience must firstbe analyzed carefully to discover the precise "principle" or cause of the observed effects,the universal structureinvolved in them. After this analytic way of discoveryhas been pursued, we are then in a position to demonstratedeductivelyhow facts follow fromthisprincipleor cause: we can pursue theway of truth. Scientificmethod,that is, proceeds from the rigorous analysis of a few selected instances or illustrations to a general principle,and then and ordered body goes fromthat principleback to the systematized of facts, to the science itself formallyexpressed. Zabarella calls of the resolutiveand the compositivemethods; this the combination and such were precisely the procedure and the terms of Galileo.3' The presuppositionof this method is of course that there exists an under examination, of intelligible structurein the subject-matter which the particular cases observed by the senses are instances; plain. Zabarella makes this perfectly
30
31

risolutivo" of Galileo on the joint use of the "rmetodo For a typicalstatement see his Opere, Ed. Nazionale, IV, 520. For a beautiand the "metodocompositivo," see the account of his firstdisciple,Benedetto Castelli, in Galileo, ful illustration, Opere,Ed. Naz., XVII, 160ff. See also Opere, VII, 75, and VIII, 212.

Ibid.,III, xix.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

200

JOHN HERMAN

RANDALL, JR.

subject-matter, on in a necessary can be carried induction Demonstrative connection witheach other. Hence it thathave an essential and in things sinceaftercertainof them intoaccount, does not take all the particulars connection, the essential notices our mindstraightway havebeenexamined at onceto bring particulars proceeds theremaining and thendisregarding that the same the universal. For it knowsthat it is necessary together in therest.32 be embodied relations should No clearer statementcould be made of the procedure of the sevenscientists. teenth-century This double method,and particularlythe analysis of instalnces, Zabarella considersmore fullyin his littlework De Regressu. and whentheyare convertible cause and effect is between The regress we must alwaysset known tous thanthecause. For since theeffect is better demonstrate from theknown known to us, we first whatis better out from from thecauseso (regredimur) cause,and thenreturn effect theunknown whyit thereason thatwe mayknow to be demonstrated, known to theeffect is so.33 Zabarella does not bringin "nature," like Aristotle; for it is a logical and not a metaphysicalquestion he is considering. "Both demonstrationsare made by us and for us ourselves,not for nature."34 In our way of discoverywe are followingthe order of knowledge, not that of things. Like Nifo, Zabarella finds four stages in the procedure that is the regress. First we observe the particular effect. Secondly, we resolve the complex fact into its component parts and conditions. Thirdly,we examine this supposed or hypotheticalcause by a "mental consideration" to clarifyit and to find fromthat its essential elements. Finally, we demonstratethe effect cause. which is from Whenthefirst has beencompleted, stageoftheprocedure from the latterto the effect, theremnust to cause,before we return effect intermediate we maybe led to a intervene a third process(labor) by which of that cause whichso far has been knownonlycondistinct knowledge
32 De Regressu, c. 4. Cf. Galileo: "The knowledgeof a single fact acquired throughthe discovery of its causes prepares the mind to ascertain and understand otherfacts without need of recourseto experiments, preciselyas in the presentcase, whereby argumentation alone the Author proves with certainty that the maximum range occurswhenthe elevationis 45?. He thus demonstrates what has neverbeen observedin experience." (Opere, Ed. Naz., VIII, 296.) 33 De Regressu,c. 1. 34

Ibid.,c. 2.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN

THE

SCHOOL

OF PADUA

201

fusedly. Some men [Nifo] knowingthis to be necessaryhave called it a negotiatioof the intellect. We can call it a "mental examination" of the cause itself,or a "mental consideration." For afterwe have hit upon that cause, we begin to considerit, so that we may also understandwhat it is. -Zabarella then proceeds to a furtheranalysis of this "mental consideration."-But what this mental considerationmay be, and how it is accomplished,I have seen explained by nobody. For thoughsome say that this intermediate negotiatioof the intellectdoes play a part, still theyhave not shownhow it leads us to a distinctknowledgeof the cause, and what is the precise force of this negotiatio.. . . There are, I judge, two things that help us to knowthe cause distinctly. One is the knowledgethat it is, which prepares us to discover what it is. For when we form some hypothesis about the matter(in re aliquid praenoscimus)we are able to search out and discoversomething at all, we shall else in it; wherewe formno hypothesis never discover anything.. . . Hence when we find that cause to be suggested,we are in a positionto seek out and discoverwhat it is. The other help, withoutwhich this firstwould not suffice, is the comparisonof the cause discoveredwiththe effect not indeed throughwhichit was discovered, with the full knowledgethat this is the cause and that the effect, but just this thingwiththat. Thus it comesabout that we are led graducomparing ally to the knowledgeof the conditionsof that thing; and when one of the conditionshas been discoveredwe are helped to the discoveryof another, until we finallyknow this to be the cause of that effect. The regressthus consists necessarily of three parts. The firstis a "demonstrationthat" (quod), by which we are led froma confusedknowledgeof the effect to a confused knowledgeof the cause. The second is this "mental consideration," by which froma confusedknowledgeof the cause we acquire a distinct knowledgeof it. The third is demonstration in the strictestsense (potissinma), by whichwe are at lengthled fromthe cause distinctly known to the distinctknowledgeof the effect.35

These threephases may indeed occur simultaneously:theyare logically ratherthan psychologically distinct. "The end of the regress is a distinctscience of effects, whichis called science propter quid." The theoryof science set forthin the Analytics is a theory of proof,not a theoryof discovery. Here withinthe school of Paduan Aristotelians,therehas been workedout what was so sorely needed, a logic of investigationand inquiry. No longer are the firstprinciples of natural science taken as indemonstrable and self-evident: they have become hypothesesresting upon the facts they serve to explain. If Zabarella did not followup the suggestionof Nifo that all natural science thereforeremains conjectural and hvyothetical.
35

Ibid., c. 5.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

202

JOHN HERMAN

RANDALL, JR.

it was because he believed that an examination of particular instances would reveal an intelligiblestructurepresent in them; and this was precisely the faith that inspired seventeenth-century
science. Principia essendi are not propositionsbut things, nor are they of necessity knownbeforehand. But oftentheyare unknown;and theycan be had thoughnot a priori. For if theythemselves demonstrated a posteriori, priorprinciplestheywould not be principles.36 But these principles of existence are no mere conveniences of knowledge; they belong to their subject-matter, and are part of the

intelligiblestructureof the world.

if we consider from effects, Propositions accepted in demonstrations are no less necessary, themin themselves, less per se, or less essential,than the propositionsof strict demonstration. But if we consider our minds, they are not so clearly knownby us to be necessaryas the propositionsof in them; strictdemonstration.Still we recognizea certainkind of necessity for that sylloif not so much as is really there,at least so much as suffices gismto deservethe name and nature of demonstration.37 It is not surprising that Galileo should so often sound like For he arrived in Padua in 1592, while the echoes of Zabarella. the great controversies over method between Zabarella on the one hand and Francesco Piccolomini and his disciple Petrella on the other, fought in the 1580's, were still resounding-controversies of which a witness has recorded: The school of logic at Padua was divided into two sects,those who were partisansof Zabarella, and thosewho were partisansof Petrella, and a multitude seemedto stand on eitherside. Afterthis most exalted and famous so useful and fruitfulto all studentsof logic, both published controversy, on the PosteriorAnalytics; than which commentaries, though commentaries in the commonjudgment and containdivergent teaching, theyare different, of learned men nothingcan be found moreexquisite or more clear.38 In these two controversies, the points at issue were relatively minor in comparison with the bulk of agreement on method. Piccolomini, an older man, holder of the first chair of natural philosophy since 1564. had come to Padua from Siena, bringing a certain
36 37

c. iv. De tribuspraecognitis, c. x. De speciebus demonstrationis, 38 A. RiGcoboni, De gymnasiopatavino, Padua, 1598; IV, e. xi.

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE

SCHOOL

OF PADUA

203

Platonism with his Aristotelianism; but though he defends some Platonic positions,on method he is as advanced as Zabarella, and as near to the ideas of the seventeenth-century scientists. He agrees on the central importance of resolution as the way of discovery. But against his demand that metaphysicsmust furnishthe and the frame of referencein all science, and that starting-point the scientistmust imitate the fixed structureof nature, Zabarella maintains the independenceand self-sufficiency of natural science, and indeed of each particular subject-matter, making the end of knowledge and inquiry a human thing,and directingthe sciences towardhuman goals and aims. He defendsthe knowledgegiven by inductionas perfectin its own kind, the constitution of man being what it is; it is no mere substitute for something that mightbe better gained in a more perfectway. And against Piccolomini 's Platonic conceptionof a natural order of perfections which science must follow, he maintains a purely immanentconception of natural ends: the perfectfunctioning of each kind of thing in the universe is its only end,and each subject-matter is to be understoodin termsof its own principles. Indeed, in his criticism of Platonic notions of teleologyZabarella went far along the path the radical graduate of Padua, Telesio, was following. Zabarella's version of the Aristotelianlogic, thoughinterpreted and coloredin termsof each of the threegreat theoriesof knowledge inherited and reconstructed by the seventeenth-century and thinkers, thoughreceivingin practice wide variations of emphasis on its several parts,remainedthe methodand ideal of sciencefor all "natural philosophers"I until the fresh criticisms of Locke and Berkeley. For though the language is diverse, the whole great literature on methodthat fillsthe scientific writingof the seventeenth centuryis at bottoma series of footnotesto the Organon of Aristotle. Indeed, the more fullythe record of late medieval and Renaissance thought is studied, the clearer it becomes that the most daring departures from Aristotelian science were carried on within the Aristotelian framework, and by means of a criticalreflection on the Aristotelian texts-however various the sources of the ideas that fertilizedthat criticism. The "father"I of modernscience,in fact,turns out to be none other than the Master of themthat know. With Zabarella the Paduan school had reached its culmination and done its work. His single successor,Cesare Cremonini(*1631), went still farther in an appeal to experience. His Tractatus de

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

204

JOHN HERMAN

RANDALL., JR.

paedia (1596) sounds like the solemnwarningof the great tradition of Aristotelian rational empiricismto the triumphantmathematicians.

Paedia is the powerof judgingrightly aboutthe manner of teaching and learning, founded on logic,withthe opportune intervention of experithegeneral method of all procedures.As Paedia is themother and nurse, so is method thedaughter and child. I add, "with theintervention of experience," because, though one be instructed by geniusor by logic,unless he be alsoexperienced in thevery thing in which he is to judge,he willthere no judgment.I say"opportune," exercise thesame or appropriate, because In mathematics, manner of experience for the is not foundeverywhere. of principles to employ induction confirming it is sufficient based on the observation of whatis in the materials mathematics is abstracted; whence in thatfield thetruth oftheprinciples is immediately evident. But in the naturalsciences suchobservation is not so obvious a way of gainingprinnoris thecollection of principles so easy. There ciples, by its employment from a zealousapplication a laborious is indeed required attention, procured keen at notwithout are arrived to things;and evenwithit theprinciples for the not only natural thought.Moreover, thisexperience is necessary foralmost if he is to arrive at first it is requisite scientist, principles; every in moralsin much of science. For experience manner is likewise required and even in divinity, sincewe do not ascendto those the same fashion, abstract causeswithout a manifold and laborious attention to their effects.39 And so he counseled ever closer attentionto the way of discovery, to the careful and painstakinganalysis of experience,to the method of resolution, withinwhichhe included as phases both inductionand demonstration a posteriort. There was but one element lacking in Zabarella's formulation of method: he did not insist that the principles of natural science be mathematical,and indeed drew his illustrations largely from Aristotle 's biological subject-matter. Though he had studied mathematicsunder Catena and Barocius, and was accounted expert in optics and in astronomy, these studies failed to leave any fundamental impress on his thought. The gradual emergenceof mathematics into the dominantposition it held in the seventeenthcentury is due to its cultivationby a small group of men working on the peripheryof the main intellectualmovementsof the sixteenthcenturv. There is a conventionalview that this shiftto mathematical
39 Tractatus de paedia, c. ii; in Explanatio proemii librorumAristotelis de Physico auditu; Padua, 1596.

ence....

Its intrinsicfunetionis to understand,dispose, and constitute

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

SCIENTIFIC

METHOD

IN THE SCHOOL

OF PADUA

205

interests was powerfullyfurtheredby the Renaissance revival of derived from Proclus, from Platonism and its number mysticism, Cabala. In the Germanies from the and the Pythagorean tradition, this has some basis in fact,and Kepler may stand as its consummato findany support for the view that attribtion. But it is difficult utes the great achievements of the Italians in mathematics and of Neoplatonism. On the one hand the mechanics to the influence interestin mathematics, Italian Platonists had almost no scientific and their"numbers" led themat once to the mazes of theologyand theosophy. And on the other, with rare exceptions the Italian down throughGalileo, when theypossessed a philomathematicians sophical interestat all, were not Platonists but Aristoteliansin their of its relations to physics,and of the proper view of mathematics, methodof natural knowledge. What theyfoundin the ancientsand what they worked upon themselveswas no mathematicalvision of the world, but effective techniques and practical problems of procedure and discovery. What they constructedas "new sciences" it remained for Descartes to interpretin the light of the tradition of AugustinianPlatonism.40 the Humanists can fairly claim to Indeed, the one contribution have made to the rise of modern science was to send men to the study of the original ancient sources in mathematics. In reestablishing connectionwith the mathematicsand mechanics of the HellenisticAge, the appeal to the ancients introducedArchimedesand Hero, as well as Apollonius, Pappus and Diophantus. The mathematical methodsof analysis and synthesisof Archimedes,of whom Tartaglia published the firstLatin edition in 1543, were the one Ockhamites nor the elementwhich neither the fourteenth-century Paduans possessed. From them the mathemasixteenth-century ticians took theirstart,and carried the day for the quantitativeside of the Paduan discussion,to which referencehas been made above. With this mathematicalemphasis added to the logical methodology of Zabarella, there stands completed the "new method" for which men had been so eagerly seeking. By the analysis of the mathematicalrelations involved in a typical "effect" or phenomenon we arrive at its formal structureor "principle." From that consequences,whichwe findillustrated principlewe deduce further in experience. Science is a body of mathematical and confirmed
40 See E. W. Strong,Procedures and Metaphysics, a Study in the Philosophy 17th Centuries. Science in the 16th anzd of Mathematical-Physical

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions

206

JOHN HERMAN

RANDALL, JR.

the principlesof whichare discovered by the resodemonstrations, lutionof selectedinstancesin experience. This is the methodcalled by Euclid and Archimedes a combination of "analysis" and "synthesis," and by the Paduans and Galileo, "resolution" and composition."'" It is traditionaland Aristotelianin regarding the structureof science as dialectical and deductive,and in seeing all verification and demonstration as inclusionwithin a logical system of ideas. It has altered the schemeof the medieval Aristoteliansin makingthe principlesof demonstration mathematicalin character; it has added the insistencethat the and to the scholasticempiricism way of discovery is not mere observation and generalization,not mere abstractionfromcommonexperience, but a careful and precise mathematicalanalysis of a scientific experience-what the medical traditionof Padua called "resolution" and what Archimedescalled "analysis." And to that experiencedemonstration must return,in a "regress," for confirmation, and the guarantee of the illustration, existence of the deduced consequences. But the return to experience is not for the sake of certain proof: for throughoutthe seventeenthcenturyit is almost impossible to findany natural scientist maintainingthat a mere fact can prove any certain truth. Columbia University

41 The precise formof the combination of the metodoresolutivoand the metodo compositivoGalileo most frequentlyemployed he called the "argomento'ex suppositione,"' and describedmost explicitlyin his letterto Carcavi in 1637 (Opere, Ed. Naz., XVII, 90); it is clearly illustratedin his discussionof naturallyaccelerated motionin the Two New Sciences (Eng. tr. by Crew and De Salvio, 161). The scientistbegins with a "hypothetical assumption,"a mathematicalhypothesisthat does not come immediately from the observationand measurementof facts, but rather from an analysis of the mathematicalrelations involved in a given effect. The propertiesthat mustfollow are then deductively demonstrated. Thirdly,cases or illustrations of that effect are analysed to discoverhow far those propertiesare in themand confirmed really exemplified by them. The demonstration Galileo calls the "compositivemethod"; the term "resolutivemethod" he applies both to the initial mathematical analysis of the problem,and to the confirming analysis of the examples. See E. W. Strong, Procedures and Metaphysics,ch. 6; E. A. Burtt, MetaphysicalFoundations of Modern Physical Science, p. 70 ff.; F. Wieser, Galilei als Philosoph,pp. 51-68).

This content downloaded from 141.213.236.110 on Tue, 11 Feb 2014 20:58:46 PM All use subject to JSTOR Terms and Conditions