henri bergson, introduction to metaphysics

7/29/2019 Henri Bergson, Introduction to Metaphysics

1/11

William James: Pragmatism. Reprinted by permission of the publisher from Pragmatism: ANew Name for Some OldWays ofThinking, Frederick Burkhardt, Fredson Bowers, and lgnasK. Skrupskelis, eds., Cambridge, Mass.: Harvard University Press, Copyright 1975 by thePrestdenr and fellows of Harvard College.Henn Bergson: An lntrodudion to Metaphysics, translated by T. E. Hulme. Copyrtght 1985 by Macmillan Publishing Company. Copyright 1949, 19.55 Reprinred by permissionof Macmillan Pubhshing Company.john Dewey: Experience and l:.ducation. Copyright 1938, by Kappa Delta Pt, An InternationalHonorSociety in Educanon. Reprinted by permission of Kappa Delta Pi.Alfred North Whitehead: Science and the Modern World. Copyright 192.5 by The Macmillan Company. Copyright renewed 1953 by Evelyn Whitehead. Reprinted by arrangemenrwtthMacmillan Publishing Company.Bertrand Russell: The ProblemsofPhilosophy, reprinted by permission of Oxford UniversityPress.Marrin llcidegger: "What Is Metaphysics?" from Existence and Being, translated by R. F. C.Hull and Alan Crick. Reprinted by permission of Rcgnery-Gateway, Inc.Ludwig Wittgenstcin: Philosophical Investigations, translated by G. E. M. Anscom be. Copyright a ~ i l Blackwell & Mott, Led., 1958. Repnnted with permission of Basil Blackwell.Karl Barth: Texr of The Word of God and the Word of Man translated by Douglas I lorton.Copyright 192.8, 1956, 1957 by Douglas Horton. Reprinted by permission of Harper &Row, Publishers, Inc.

ITH E UNIVERSITY OF CHICAGO

The Great Books is published with the editona l adviceof he faculties of T e University ofChicago

No prr of h1s work may be rtproduced or u h ~ e d in any form or by any mrans,dcctromcor mechanical, tncluding photocopying, rrcording, or by any mforn1a11on~ c o r o g t and retrieval system, without permi.sion in wriung frorn he publisher.

First Edition 1952Second Edition 990Second Printing, 1991E n c y d o p ~ d i a Britannica, Inc.All Rights Reserved

Library of Congress Catalog Card Number: 90-80213International Standard Book Number: o-852.:1.95316Manufactured in the Umted States of Americ.t

General ContentsPHILOSOPHY

WILLIAM JAMESPragmatism .

HENRI BERGSONAn Introduction tO Metaphysics . 71Translated b y T. E. Hulme

JOHN DEWEYExperience and Education . . . 99

ALFRED NORTH WHITEHEADScience and the Modern World . . . . r35

BERTRAND RUSSELLThe Problems of Philosophy . . . . . . . 243

MARTIN HEIDEGGERWhat Is Metaphysics? . . . . . . . . . . 2.99Tra1zslated by R. F. C. Hull and Alan Crick

LUDWIG WITTGENSTEINPhtlosophicallnvcsugauons . 317Translated by G. E. M. Anscombe

RELIGION

KARL BARTH

The Word of God and the Word of Man . . 451Translated by Douglas Horton


2/11

An Introduction to MetaphysicsAcomparison of the definitions of metaphysicsnd rhe various concepts of the absolute leads:o the discovery p h i l o ~ o p h e r s , sp.tc0 ( rheir apparent dtvergenctes, agree m dtstinguishing two profoundly different ways ofknowing a thing. The first implies that we moveround the object; the second, that we enterinto it. The first depends on the point of viewat which we are placed and on the symbolsby whtch we express ourselves. The secondneither depends on a point of view nor relieson any symbol. Th e first kind of knowledgemay be said tO stop at the relative; the second,in those cases where it is possible, to attainthe absolute.Consider, fo r example, the movement of anobject in space. My perception of the motionwill vary with the point of view, moving or sta-tionary, from whic h I observe it. My expressionof it will vary with the systems of axes, or thepoints of reference, to which I relate it; that is,wtth the symbols by which I translate it. Forthis double reason I call such motion relative:m the one case, as m the other, I am placedoutside t:he object itself. But when I speak ofan absolute movement, I am attributmg tO themoving object an interior and, so ro speak,states of mind; I also imply that I am in sym-pathywith those stares, and that I insert myselfll1 t h ~ m by an effort of imagination. Then, ac-cordtng as the object is moving or stationary,according as it adopts one movement or an-

~ t h e r , ~ h a t I experience will vary. And what;xpenencc will depend neither on the pointVtew l may take up in regard to the object,Since I am inside the object itself, nor on the?mbols by which I may translate the motion,

~ n e e I have rejected all translations in order to;..ssess the original. In short, I shall no longerWhsp the movement from without, remainingas ,;:e l but from where it is, from within,C 5 Itself. 1shall possess an absolu te.tureOnstder, again, a character whose adven-rn sate related to me in a novel. The authorrn:Y multiply the traits of his hero's character,Y tnake him speak and act as much as he

pleases, but all this can never be equivalentro the s1mple and indivisible feeling which Ishould experience if 1were able for an instantto tdentify myself with rhe person of the herohimself. Ou t of that indivisible feeling, as froma spring, all the words, gestures, and actions ofrhc man would appear to me ro flow naturally.They would no longer be accidents which,added to the idea I had already formed of thecharacter, continually enriched that idea, with-out ever completing it. The character would begiven to me all at once, in its entirety, and thethousand incidents which manifest it, insteadof adding themselves to the idea and so enrich-ing it, would seem to me, on the contrary, todetach themselves from it, without, however,exhausting it or impoverishing its essence. Allthe things I am told about the man provideme with so many points of view from which Ican observe him. All the traits which describehim, and which can make him known to meonly by so many comparisons with persons orthings I know already, are signs by which he isexpressed more or less symbolically. Symbolsand points of view, therefore, place me outsidehim; they give me only wha t he has in commonwith others, and not what belongs to him andto htm alone. But that which is properly him-self, that which constitutes his essence, cannotbe perceived from without, being internal bydefinition, nor be expressed by symbols, be-ing incommensurable w ith everything else. De-scription, h istory, an d analysis leave me here inthe relative. Coincid ence with the person him-self would alo ne give me the absolute.

7 1

It is in this sense, and in this sense only, thatabsolute is synonymous with perfection. Wereall the photographs of a town, taken from allpossible points of view, ro go on indefinitelycompleting one another, they would never beequivalent to the solid town in which we walkabout. Were all the translations of a poeminto all possible languages to add togerhcr theirvarious shades of meaning and, correcting eachother by a kind of mutual retouching, ro givea more an d more faithful image of the poem


3/11

BERGSONthey translate, they would yet never succeedin rendering the inner meaning of the original.A representation taken from a ccrrrun point ofview, a translation made with certain symbols,wtll always remain imperfect in comparisonwith the object of which a view has been tak en,or which the symbols seek to express. But theabsolute, which is the object and not its representation, the onginal and not its translation, isperfect, by being perfectly what it is.

lr is doubtless for this reason that the ab-solute has often been idemified with the in-finite. Suppose that I w1shed to communicateto someone who did no t know Greek the extraordinarily simple impression that a passagein Homer makes upon me; I should first give atranslation of the lines, I should then commenton my translation, and then develop the commentary; in this way, by piling up explanationon explanation, I might approach nearer andnearer to what Iwanted to exp re ss; but I shouldnever quite reach it. When you raise your arm,you accomplish a movement of which youhave, from within, a simple perception; butfor me, watching it from the outside, yourarm passes through one point, then throughanother, and between these two there wiJI bestill other points; so that, if I began to count,the operation would go on forever. Viewedfrom the ins1de, then, an absolute is a simplething; but looked at from the outside, that isto say, relatively to other things, it becomes,in relation to these s1gns which express it, thegold coin for wh1ch we never seem able to fin-ish giving small change. Now, that which lendsitself at the same ttme both to an indivisibleapprehension and to an mexhaustible enu meration is, by the very definition of the word,an infinite.

It follows from this that an absolute couldonly be given in an intuition, whilst everythingelse falls within the province of analysis. Byintuition is meant the kind of intellectual sym-pathy by which one places oneself within anobject in order to coincide with what is uniquein it and consequently inexpressible. Analysis,on the contrary, is the operation which reducesthe object to elements already know n, tbat is,to elements common both to it and other objects. To analyze, therefore, is to express a thingas a function of something other than itself.All analysis is rhus a translation, a development mro symbols, a representation taken from

successive points of view from which weas many resemblances as possible betweennew object which we arc studying andwhich we believe we know already. In itsnally unsatisfied desire to embrace thearound which it is compelled to turn,mult1plieswithout end the number of irsof view in order to complete irs alwayscomplete representation, and ceaselesslyits symbols that it may perfect the alwaysperfect translation. It goes on, thereforeinfinity. But intuition, if intuition 1s poss1a simple act.

Now it is easy to see that the ord1function of positive science is analysis.science works, then, above all, withEven the most concrete of the naturalthose concerned with life, confine theto the visible form of living beings, theirand anatomical elements. They make rn r nn - isons between these forms, they reducemore complex to the more simple; inthey study the workings of life in what is,to speak, o nly its visual symbol. If thereany means of possessing a reality au,"'"'"instead of knowing it relatively, of placingself within it insread of looking at itoutside points of view, of having the intuinstead of making the analysis: in short, ofing it without any expression, tra1nstat10nsymbolic representation-metaphysics ismeans. Metaphysics, then, is the sciencedaims to dispense with symbols.There is one reality, at least, which we allfrom within, by intuition and not by analysis. It is our own personality in itsthrough rime-ourselfwhich endures. Wesympathize intellectually with nothi ng else,we certainly sympathize with our own selves.

When I direct my attcmion inward totemplate my own self (supposed for thement to be inactive), I perceive at first,a crust solidified on the surface, all theceptions which come to it from the m:ltel1!111world. These perceptions are clear, dijuxtaposed or juxtaposable one withthey rend ro group themselves into oNext, I notice the memories wh1ch mo re oradhere to these perception s and which serveinterpret them. These memories have beentached, as it were, from the depth of myaliry, drawn to rhc surface by the pelrce]ptll::l l'

AN INTRODU CTION TO METAPHYSICS 73hich r e s e m b l ~ them; t ~ e y rest on the surface"f mY mind Wlthout bemg absolutely myself.

~ s d y , I feel the stir of tendencies and motorh btrs-a crowd of vinual actions, more or essf i ~ n l y bound to these perceptions and memo-es All these clearly defined elements appear

rt 0 ;e distinct from me, the more distinct they:e from each other. Radiating, as they do,from within outwards, they form, collecuvely,the surface of a sphere which tends to growlargerand lose itself in the exterior world. But if1 raw myself in from the periphery towards thecenter, if search in the depth of my being thatwhich is most uniformly, most constantly, andmost enduringly myself, [ find an altogetherdilferent thing .There is, beneath these sharply cut crystals and this frozen surface, a con tinuous fluxwhich is not comparable to any flux I have everseen. There is a succession of stares, each ofwhich announces that which follows and contains that which precedes it. Theycan, properlyspeaking, only be said to form multiple stateswhen I have already passed them and turn backro observe their track. W hilst I was expcriencmg them they were so solidly organized, soprofoundly animated with a common life, that Icould not have said where any one of them fintshed or where another commenced. In realityno one of them begins or ends, but all extendinto each other.This inner life may be compared to the unrolling of a coil, for there is no living being whodoes not feel himself coming gradually to theend of his role; and to live is to grow old. ButIt may just as well be compared to a continualrolling up, like that of a thread on a ball, forour past follows us, it swells incessantly wi ththe present rhat it picks up on its way; andconsciousness means memory.

Bur actually it is neither an unrolling nora rolling up, for these two similes evoke theIdea of lines and surfaces whose parts are ho~ o g e n e o u s and superposable on one another.how, there are no two identical moments int e life of the same conscious being. Take rhe: ~ ~ p l e s t sensation, suppose it o n : ~ t a n r , absorbhit the entire personality: the consciousness: lch ~ i l l accompany rhis sensation cannotlltlllaln Identical with itself for two consecallve moments, because the second momentrnWays contains, over and above the first, rheemory that rhc first has bequeathed to ir.

A consciousness which could experience twoidentical moments would be a consciousnesswithout memory. It would die and be bornagain continually. In what other way could onerepresent unconsciousness?

It would be better, then, to use as a comparison the myriad-tinted spectrum, with itsinsensible gradations leading from one shadeto another. A current of feeling which passedalong the spectrum, assuming in tu rn the tintof each of 1ts shades, would experience a seriesof gradual changes, each of which would announce the one to follow and would sum upthose which preceded it. Yet even here the successive shades of the spectrum always remainexternal one to another. They are juxtaposed;they occupy space. But pure duration, on thecontrary, excludes all idea of juxtaposition, reciprocal externality, and extension.Let us, then, rather, imagine an infinitelysmall elastic body , contracted, if it were possible, to a mathematical point. Let this be drawnou t gradually in such a manner that from thepoint comes a constantly lengthening line. Letus fix our attention not on the line as a line, buton the action by which it is traced. Let us bearin mind that this action, in spite of its duration,is indivisible if accomplished without stopping,that if a stopping-point is inserted, we have rwoactions instead of one, that each of these separate actions is then the indtvisible operation ofwhich we speak, and that it is no t the movingaction itself which is divisible, but, rather, thestationary line it leaves behind it as its trackin space. Finally, let us free ourselves from thespace which underlies the movement in orderto consider only the movement itself, the actof tension or extension; in short, pure mobility.We shall have this time a more faithful image ofthe development of our self in duration.

Howeve r, even this image is incomplete, and,indeed, every comparison will be insufficient,because the unrolling of our duration resembles in some of its aspects the unity of anadvancing movement and in others the multiplicity of expanding states; and, clearly, nometaphor can express one of these two aspects without sacrificing the o r h e ~ . If I use rhccomparison of the spectrum with irs thousandshades, I have before me a thing already made,whilst duration is continually in the making. IfI think of an elastic which is being stretched,or or a spring which is extended or relaxed,


4/11

74 BERGSONr forget the richness of color, characteristicof duration that is lived, to see only the simple movement by which consciousness passesfrom one shade to another. The inner life is allthis at once: variety of qualities, continuity ofprogress, and unity of direction. Jr cannot berepresented by images.But it even less possible to represent it bycortcepts, that is by abstract, general, or simpleideas. It is true that no image can reproduceexactly the original feeling 1 have of the flowof my own conscious life. But it is not evennecessary that I should attempt to render it. Ua man is incapable of getting for himself the intuition of the constitutive duration of his ownbeing, nothing will evergive it to him, conceptsno more than images. Here the single aim ofthe philosopher should be to promote a certaineffort, which in most men is usually fetteredby habits of mind more useful to life. Now theimage has at least this advantage, that it keepsus in the concrete. No image can replace theintuition of duration, but many diverse images,borrowed from very different orders of things,may, by the convergence of their action, directconsciousness to the precise point where thereis a certain intuition to be seized. By choosing images as dissimilar as possible, we shallprevent any one of them from usurping theplace of the intuition it is intended to call up,since it would then be driven away at once byits rivals. By providing that, in spite of theirdifferences of aspect, they all require from themind the same kind of attention, and in somesort the same degree of tension, we shall gradually accustom consciousness to a particularand clearly-defined disposition-that preciselywhich it must adopt in order to appear to itself as it really is, without any veil. But, then,consciousness must at least consent to makethe effort. For it will have been shown nothing:It will simply have been 2Jaced in the attjtudeit must take up in order to make the desiredeffort, and so come by itself to the intuition.Concepts, on the contrary-especially if theyare simple-have the disadvantage of being inreality symbols substituted for the object theysymbolize, and demand no effort on our part.Examined closely, each of them, it would beseen, retains only that part of the object whichis common to it and to others, and expresses,still more than the image does, a comparisonbetween the object and others which resemble

it. Bur as the comparison has made manifestresemblance, as the resemblance is a prnr>A_ ..of the object, and as a propeny haspearance of being a part of the objectpossesses it, we easily persuade ourselvesby setting concept beside concept we areconstructing the whole of the object withpans, thus obtaining, so to speak, its '" ' ..........tual equivalent. ln this way we believecan form a faithful representation of uu'"l.n.by setting in line the concepts of unity,plicity, continuity, finite or infinite divisietc. There precisely is the illusion. Therethe danger. Just in so far as abstract ideasrender service co analysis, that is, to thetific study of the object in its relations coobjects, so far are they incapable ofintuition, that is, the metaphysicalof what is essential and unique in theFor, on the one hand, these concepts, laidby side, never actually give us more thanartificial reconstruction of the object, ofthey can only symbolize certain generala way, impersonal aspects; it is thereforeto believe that with them we can seize aof which they present to us the shadowAnd, on the other hand, besides rhethere is also a very serious danger. For thecept generalizesat the same time as it abstractThe concept can only symbolize aproperty by making it common to an inthings. It therefore always more or less dthe property by the extension it givesReplaced in the metaphysical object tobelongs, a prop erty coincides with theor at least molds itself on it, andsame outline. Extracted from the " ' ' "object, and presented in a concept, itindefinitely larger, and goes beyond theitself, since henceforth it has ro containalong with a number of other objects. Thusdifferent concepts that we form of theties of a thing inscribe round it so manyeach much too large and none of themit exactly. And yet, in the thing itself theerties coincided with the thing, and coiconsequently with one another. So that ifare bent on reconstructing the object withcepts, some artifice must be soughtthis coincidence of the object and its on>oe:n"'can be brought about. For example,choose one of the concepts and try,from it, to get round to the others. Bur we

AN INTRODUCTION TO METAPHYSICS 75soon discover that according as we startr oJTl one concept or another, th e meeting andfr JTlbination of the concepts will take place inalrogether different way. According as wea art for example, from unity or from mulri

s ~ i c i c y , we shall have to conceive differentlyphe multiple unity of duration. Everything will~ e p e n d on the weight we attribute to this orthat concept, and this weight will always bearbitrary, since the concept extracted from theobject has no weight, being only the shadow ofa body. In this way, as many different systemswill spring up as there are external points ofview from which the reality can be examined,or larger circles in which it can be enclosed.Simple concepts have, then, not only the inconvenience of dividing the concrete unity ofthe object into so many symbolical expressions; they also divide philosophy into distinctschools, each of which takes its seat, choosesits counters, and carries on with the others agame that will never end. Either metaphysics isonly this play of ideas, or else, if it is a seriousoccupation of the mind, if t is a science and notsimply an exercise, if must transcend conceptsin order to reach intuition. Certainly, conceptsare necessary to it, for all the other scienceswork as a rvle with concepts, and metaphysicscannot dispense with the other sciences. Butit is only truly itself when it goes beyond theconcept, or at least when it frees itself fromrigid and ready-made concepts in order to create a kind very different from those whichwe habitually use; I mean supple, mobile, andalmost fluid representations, always ready tom?l.d themselves on the f!eeting fQrms of int u ~ o n . We shall return later to this importantPOmt. Let it suffice us for the moment to haveshown that our duration can be presented toUS directly in an intuition, that it can be suggested to us indirectly by images, but that it cannever-if we confine the word concept to itsProper meaning-be enclosed in a conceptualrepresentation.d Let us try for an instant to consider oururarionas a multiplicity. It will then be necesto add that the terms of this multiplicity,

~ e a d of being distinct, as they are in . anyatder multiplicity, encroach on one another;t h a ~ while we can no doubt, by an effortel 10lagmarion, solidify duration once it hasc apsed, divide it into juxtaposed portion s and0 l ln t all these portions, yet this operation is

accomplished on the frozen memory of theduration, on the stationary trace which the mobility of duratio n leaves behind it, and not onthe duration itself. We must admit, therefore,that if there is a multiplicity here, it bears no re-semblance to any oth er multiplicity we know.Shall we say, then, that duration has unity?Doubtless, a continuity of elements which prolong themselves into on e anoth er participatesin unity as much as in multiplicity; but thismoving, changing, colored, living unity hashardly anything in common with the abstract,motionless, and empty unity which the conceptof pure unity circumscribes. Shall we concludefrom this that duration must be defined as unityand multiplicity at the same time? But s i n g u -larly enough, however much I manipulate thetwo concepts, portion them out, combine themlifferently, practice on them the most subtleoperations of mental chemistry, J never obtainanything which resembles the simple intuition .that I have of duration; while, on the contrary,when I replace myself in duration by an effortof intuition, J immediately perceive how it isunity, multiplicity, and many other things besides. These different concepts, then, were onlyso manystandpoints from which we could consider duration. Neither separated nor reunitedhave they made us penetrate into it.We do penetrate into it, however, and thatcan only be by an effort of intuition. In thissense, an inner, absolute knowledge of the duration of the self by the self is possible. Butif metaphysics here demands and can obtainan intuition, science has none the less needof an analysis. Now it is a confusion betweenthe function of analysis and that of intuitionwhich gives birth to the discussions betweenthe schools and the conflicts between systems.Psychology, in fact, proceeds like all theother sciences by analysis. lt resolves the self,which has been given to it at first in a simpleintuition, into sensations, feelings, ideas, etc.,which it studies separately. It substitutes, then,for the self a series of elements which formthe facts of psychology. But are these elementsreally parts? That is the whole question, and itis because it has been evaded that the problemof human personality has so often been statedin insoluble terms.It is incontestable that every psychical state,simply because it belongs to a person, reAectsthe whole of a personality. Every feeling, how-


5/11

BERGSONever simple it may be, contains virtually withinit the whole past and present of the being experiencing ir, and, consequently, can only beseparated and constituted into a "state" by aneffort of abstraction or of analysis. But it isno less incontes table that without this effort ofabstraction or analysis there would be no possible development of the science of psychology.What, th en, exactly, is the operation by which apsychologist detaches a mental state in order toerect it into a more or less independent entity?He begins by neglecting that special colorin g ofthe personality which cannot be expressed inknown and common terms. Then he endeavorsto isolate, in the person already thus simplified,some aspect which lends itself to an interesting inquiry. If he is considering inclination,for example, he will neglect the inexpressibleshade which colors it, and which makes theinclination mine and not yours; he wiU fix hisattention on the movement by which our personality leans toward a certain object: he willisolate this attitude, and it is this special aspectof the personality, this snaps hot of the mobilityof the inner life, this "diagram" of concreteinclination, that he will erect into an independent fact. There is in this something very likewhat an artis t passing through Paris does whenhe makes, for example, a sketch of a tower ofNotre Dame. The tower is inseparably unitedto the building, which is itself no less inseparably united to the ground, co its surroundings,to the whole of Paris, and so on. It is firstnecessary to detach it from aU these; only oneaspectof the whole is noted, that formed by thetower of Notre Dame. Moreover, the specialform of this cower isdue to the grouping of thestones of which it is composed; bur the artistdoes no t concern himself wirh these stones,he notes only the silhouette of the tower. Forthe real and internal organization of the thinghe substitutes; then, an external and schematicrepresentation. So that, on the whole, hissketch corresponds to an observation of theobject from a certain point of view and to thechoice of a certain means of representation.But exactly the same thing holds true of theoperation by which the psychologist extracts asingle mental state from rhe whole personality.This isolated psychical state is hardly anythingbut a sketch, the commencement of an artificial reconstruction; it is the whole consideredunder a certain elementary aspect in which we

are specially interested and which w e have cafully noted. It is not a part, bur an e l e m ehas not been obtained by a natural dismemmenr, but by analysis.

Now beneath aU the sketches he has madeParis the visitor will probably, by way o fmemo, write the word "Paris." And as he hireally seen Paris, he wiJI be able, with the hof the original intuition he had of the whole,place his sketches therein, and so join themtogether. But there is no way of performing rinverse operation; it is impossible, even withinfinite number of accurate sketches, and ewith the word ''Paris" which indicates that tmust be combined together, to get back tointuition that one has never had, and to oneself an impression of what Paris is likeone has never seen it. This is because wenQt dealing here with real parts, but with mnotes of the total impression. To take amore striking example, where the nota tionmore completely symbolic, suppose that Ishown, mixed together at ~ n d o m , the let t .which make up a poem I am ignorant of. Cthe letters were parts of the poem, I couJtattempt to reconstitute the poem with rhaltby trying the different possible arrangemen!ftas a child does with the pieces of a Chindtpuzzle. But I should never for a moment thi!fof attempt ing such a thing in this case, becautlthe letters are not component parts, but onltpartial expressions, which is quite a diiferellthing. That is why, if I know the poem, la t or dput each of the letters in its proper p lace anlfjoin them up without difficulty by a continli-ous connection, whilst the inverse operation IIimpossible. Even when I believe I am actualltattempting this inverse operation, even when lput the letters end to end, I begin by thinkingofsome plausible meaning. I thereby give m.fsdan intuition, and from this intuition I attell19lto redescend ro the elementary symbols whidwould reconstitute its expression. The vetfidea of reconstituting a thing by r a t i ~practiced on symbolic elemenrs alone imphe!such an absurdity that it would never occur tOanyone if they recollected that they were noldealing with fragments of the thing, but only,-'it were, with f ~ g m e n t s of irs symbol.Such is, however, the undertaking ofphilosophers who try to reConstruCtity with psychical states, whe ther theythemselves to those states alone, or w t t e t l ~ " ' ! i

AN INTRODUCTION TO METAPHYSICS 77heY add a kind of thread for the purpose ofjning the states together. B oth empiricists and10rionalists are victims of the same fallacy. Bothf hem mistake partial notations for real parts,0hus confusing the point of view of analysis and1 f ntuition,of science and of metaphysics.0 fh e empiricists say quite rightly that psy-nological analysis discovers nothing more in'ersonality than psychical stares. Such is, infact, the function, and .the very de.finition ofanalysis. The psychologiSt has nothing else todo but analyze personality, that is, to note certain states; at the most he may put the label"ego" on these states in saying they are "statesof rhe ego," just as the artist writes the word"Paris" on each of his sketches. On the level atwhich the psychologist places himself, and onwhich he must place himself, the "ego" is only asign by which the primitive, and moreover veryconfused, intuition which has furnished thepsychologist wi th his subject-marrer is recalled;it is only a word, and the great error here lies inbelieving that while remaining on he same levelwe can find behind the word a thing. Such hasbeen the error of those philosophers who havenot beenable to resign themselves to being onlypsychologists in psychology, Taine and StuartMill, for example. Psychologists in the meth odthey apply, they have remained metaphysiciansin the object they set before themselves. The ydesire an intuition, and by a strange inconsistency they seck this intuition io analysis whichis the very negation of it. They look for theego, and they claim to find it in psychical stares,rbough this diversity of States has itself onlybecn obtained, and cou ld only be obtained, bytransportingoneself outside the ego alrogether,so as to make a series of sketches, notes, andmore or less symbolic and schematic diagrams.T h u ~ , however much they place the states sideby s ~ d e , multiplying points of contact and explonng the intervals, che ego always escapesbbem, so that they finish by seeing in ir nothingUta vain phantom. We might as well denythe Wad had a meaning, on the ground\a t_we had looke d in vain for that meaning int e Intervals between the letters of which it iscomposed.of Philosophical empiricism is born here, then,1 a confusion between the point of view ofnt 0 . u ~ t t o n and that of analysis. Seeking for rhec Jglnal in the transladon, where naturally itannot be, it denies the existenceof rhe original

on the ground that it is not found in the translation. It Leads of necessity ro negations; buton examining the matter closely, we perceivethat these negations simply mean that analysisis not intuition, which is self-evident. From theoriginal, and, o ne must add , very indistinct intuition which gives positive science its material,science passes immediately ro analysis, whichmultiplies to infinity its observations of thismaterial from outside points of view. It sooncomes to believe that by purring together allthese diagrams it can reconstitute the o bject itself. No wonder, then, tbat it sees this object flybefore it, like a child that would like to makea solid plaything ou t of the shadows outlinedalong th e wall!But rationalism is rhe dupe of the same illusion. 1t starts out from the same confusion asempiricism, and remains equally powerless toreacb the inner self. Like empiricism, it considers psychical states as so many fragmentsdetached from an ego that binds them together.Like empiricism, it tries to join these fragmentsrogether in order to re-create the unity of theself. Like empiricism, finally, it sees this unityof the self, in the continually renewed effort itmakes co clasp it, steal away indefinitely like ap h a n t o m . But whilst empiricism, weary of thestruggle, ends by declaring that there is nothingelse bur the multiplicity of psychical states, rationalism persists in affirming rhe unity of theperson. It is true that, seeking this unity onthe level of the psychical states themselves, andobliged, besides, to put down to the accountof these states all the qualities and determinations that it finds by analysis (since analysisby its very definition leads always to states),nothing is left to it, for the unity of personality,but something purely negative, the absence ofall determination. The psychical states havingnecessarily in this analysis taken artd kept forthemselves everything that can serve as matter, the "unity of the ego" can never be morethan a form without content. It will be absolutely indeterminate and absolutely void. Tothese detached psychical states, w these shadows of the ego, the sum of which was forthe empiricists the equivalent of the self, rationalism, in order ro reconstitute personality,adds something still more unreal, the void inwhich these shadows move-a place for shadows, one might say. How could this "form,''which is in truh formless, serve LO character-


6/11

BERGSONizc a hvmg, active, concrete personality, or todisringuish Peter from Paul? Is it asronishingthat rhe philosophers who have isolated this"form" of personalary should, then, find ir insufficient to characterize a definate person, a ndthat they should be gradually led ro make theirempty ego a kind of bottomless receptacle,which belongs no more to Peter than ro Paul,and in which there is room, according ro ourpreference, for entire humanity, for God, orfor existence in general? I see in this matteronly one difference between empincism andrationalasm. The former, seeking the uniry ofthe ego in the gaps, as it were, between thepsychical stares, as led ro fill the gaps witb otberstates, and so on indefinarely, so that the ego,compressed an a constantly narrowing interval,tends roward:. zero, as analysas is pushed fartherand fanher; whilst rationalism, making the egorhe place where mental states arc lodged, isconfronted with an empty space which we haveno reason to limit here rather than there, whichgoes beyond each of rhc successive boundariesthat we try to assign to it, which constantlygrows larger, and which tends to lose itself nolonger in zero, bur in the infinite.

Th e distance, then, b etween a so-called "empiricism" like that of Taine and the mosrtransccndenral speculations of certain Germanpantheasts as very much less than is generallysupposed. The method as analogous in bothcases; it consists in reasonang about rhe ele-ments of a translataon as if they were parts ofthe original. But a true empiricism is that whichproposes to get as near to rhc origin::ll itself a spossable, to search deeply into its life, and so,by a kind of intellectual auscultation, ro feelthe throbbings of its soul; and this true empiricism is the true metaphysics. It is true thatthe task an extremely difficult one, fo r noneof the rendy-made conceptions which thoughtemploys in its daily operations can be of anyusc. Nothing is more easy than to say thar theego is multiplicity, or thot it is unity, or that itis the synrhesis of both. Unity and multiplicityare here representations that we have no needto cur out on the model of the object; theyare found ready-made, and have only to bechosen from a heap. They arc stock-size clotheswhich do just as well for Peter as for Paul, forthey set off the form of neither. But an empiricism worchy of the name, an empiricism whichworks only w measure, i!> obhged for each new

object that it studies to make an ..vo._, .. . . ...fresh effort. lr curs ou t for the object a conwhich is appropriate to that objectconcept which can as yer hardly be called acept, since it applies to thas one thing. Itnot proceed by combining current ideasuniry and multiplicity; bur it leads us, oncontrary, to a simple, unique r l ' r l r t > ~ : P n r ~which, however once formed, enables usunderstand easily how it as that we can placein rhe frames uniry, multipliciry, etc., all mlarger than itself. In short, philosophy rhusfined does nor consist in the choice ofconcepts, and in raking sides with a school, bllt(in the search for a un ique anruition from wh1cbwe can descend wirh equal ease ro differeQtl.conceprs, because we are placed above the divj.;;sions of the schools.

That personality has uniry cannot be den iedtbut such an affirmation teaches one nothi11fabout the extraordinary nature of the partieular unity presented by personality. Tha t OIWself is multiple I also agree, but then it must btunderstood that it is a multiplicity which ha'nothing in common with any other multiplicity. What is really important for philosophy isto know exactly what unity, what multiplicity,and what reality superior both tO abstract unityand multiplicity the multiple unity of the sdfactually is. Now philosophy will know this onlywhen it recovers possession of the samp le intuition of the self by the self. Then, according tothe direction ir chooses for irs descent from thdsummit, it will arrive at unity or mulnphcacyor ar any one of the concepts by whach we rrtto define the movang life of the ~ e l f . But nominglingof rhese concepts would give anythtnfwhich ar all resembles rhe self that endures.If we arc shown a solid cone, we see withou t any difficulty how it narrows towards thesummit and tends ro be lost in a m a t h e m a n ~point, and also how it enlarges in the directionof he base into an indefinitely increasing circltBur neither the point nor rhe circle, nor thtjuxtaposition of the two on a plane, wouldgive us the least idea of a cone. The sarntthing holds true of the unity and multiplicitYof mental life, and of rhc zero and rhc inhnice towar ds which empiricism and rationa lisf11conduct personality.Concepts, as we shall show elsewhere, generally go together an couplesand represen tcontraries. There is hardly any concrete

AN INTRODUCTION TO METAPHYSICS 79hich cannot be observed from two opposstandpoints, which cannor consequentlytO d d ..e subsume un er rwo antagonasuc concepts.J-{ence a thesis and an antithesis whtch we endeavor in vain to reconcile logically, for the

eJY simple reason rhat ir is impossable, wath\ oncepts and obser varions taken from o utside~ a n t s of view, ro make a thing. Bur from theobject, seized by intuition, we pass easaly in01311y cases to the two conrrary concepts; andas in chat way thes.is and anrithesis can be seenro spring from realtry, we grasp at the same ttmehow ir is that rhe two are opposed and howthey are reconciled.It is rrue that to accomplish this, it is necessary to proceed by a reversal of the usualwork of the intellect. Thi,rking usually consistsin passing from concepts to things, and notfrom things ro concepts. To know a realiry,in rhe usual sense of the word "know," isto take ready-made concepts, to portion rhemout and to mix them together until a practicalequivalent of the reality is obtained. But it mustbe remembered that the normal work of theintellect is far from being disinterested. We donot aim generally at knowledge for the sakeof knowledge, but in order to rake sides, todraw profit-in short, ro satisfyan interest. Wemquire up to what point rhe object we seek toknow is this or that, to what known class itbelongs, and what kind of action, bearing, orarritude it should suggest to us. These differentpossible actions and attitudes are so many con-ceptual directions of our thought, determinedonce for all; it remains only ro follow them: inthat precisely consists the application of concepts to things. To try to fit a concept on anobject is simply to ask what we can do withthe object, and what it can do for us. To label?n object wirh a certain concept is to marktn precise terms the kind of action or attitudethe object should suggest to us. All knowledge,~ ~ o p e ; t y so called, is then oriented in a certain

~ e c u o n , or taken from a certain point of view.. as true that our interest is often complex. This18 why it happens that our knowledge of theobject may face several successive directons and may be taken from various points ofv,ew. It is this which constitutes, in the usualmea .h ntng of the terms, a "broad" and "compre-~ n s t v e " knowledge of the ob1ecr; the ObJeCt asu en brought nor under one single concept, butnder several in which it suppo!>ed ro "par-

ticipate." How does it participate in all theseconcepts at the same time? This is a questionwhich does not concern our practical actionand about which we need not trouble. It is,therefore, natural and legitimate in daily lifeto proceed by the juxtaposition and portioningou t of concepts; no philosophical difficulry willarise from this pro cedure, since by a tacit agreement we shall abstain from philosophizing.But to carry rhis modus operandi into philosophy, to pass here also from concepts to therhing, to use in order to obtain a disinterestedknowledge of an object (that this time we desirero grasp as it is in itself) a manner of knowinginsptred by a determinate interest, consistingby definition in an externally-taken view of theobject, is to go against the end that we havechosen, to condemn philosophy ro an eternalskirmishing between the schools and ro installcontradiction in the very heart of the objectand of the method. Either there is no phi losophy possible, and all knowledge of things is apractical knowledge aimed at the profit to bedrawn from them, or else philosophy consistsin placang oneself within th e obj ect itself by aneffort of intuition.

Bur in order to understand the nature of thisintuition, in order to fix with precision whereintuition ends and where analysis begins, it isnecessary to rerum to what was said earlierabout the flux of duration.It will be nored that an essential characteristic of the concepts and diagrams to whichanalysis leads is that, while being considered,they remain stationary. I isolate from the totalicy of interior life that psychical entity whichI call a simple sensation. So long as I srudyit, I suppose that it remains constant. lf I noticed any change in it, I should say that it wasnot a single sensation but several successivesensations, and I should then transfer to eachof these successive sensations the immutabilitythat I first attributed to rhe total sensation.In any case l can, by pushing the analysis farenough, always manage to arrive at elementswhich I agree to consider immutable. Th ere,and th ere only, shall I find the solid basis of operations which science needs for its own pro perdevelopment.But, then, I cannot escape the objection thatthere is no stare of mind, however simple,which does not change every moment, sincethere is no consciousness without memory, and


7/11

So BERGSONno continuation of a state without the adrution, to the present feeling, of che memoryof past moments. lr is th b wh1ch constitutesduration. Inner duration rs the con t inuous lifeof a memory which prolongs the past into thepresent, the present e1ther containing within itin a distmct form the ceaselessly growin g imageof rhe pasr, or, more probably, showing by itscontinual change of quality the hen vier and stillheavier load we drag behind us as we growolder. Without this survival {lf the past imo rhepresent there would be no dumtion, bu t onlyinstantaneity.Probably if I am thus accused of takrng themental state out of duratron by the mere factthat I analyze it, I shall reply, " Is nor eachof these elementary psychical states, to whichmy analysis leads, itself a state which occupiestime? My analysis," J shall say, "does indeedresolve the inner life into !>tares, each of whichis homogeneous with itself; only, since the homogeneity extends over a dcfimtc number ofminutes or of seconds, the elementary psychical state docs not cease ro endure, although itdoes not change."But, in saying that, J fail to sec that thedefinite number of mmutes and of seconds,which I am attributing here to the elementarypsychical state, has simply the value of a signintended ro remind me that the s y ~ . : h i c a l state,supposed homogeneous, is in rcahty a statewhich changes and endures. The state, taken10 itself, is a perpetual becoming. I have extracted from rhis becoming a certain averageof quality, which I have supposed invariable; Ihave in this way constituted a St


8/11

moreover, than t h i ~ o method of proce(jurc, solong as we arc concerned only with a practical knowledge of reality. Knowledge, in so faras ir is directed to practical matren, has onlyto enumerate the principal possible attitudesof the thing towards us, as well as ou r bestpossibleattitude towards it. Therein lies the ordinary function of ready-made concepts, thosestations with which we mark out rhe path ofbecoming. Burro seek to penetrate with theminto the inmost nature of things, is ro : ~ p p l y rothe mobility of the real a method created in order to give stationary points of observation onir. Jt is to forger rhar, if metaphysics is possible,it can only be a laborious, and even painful,effort to remount the natural slope of rhe workof thought, in order to place oneself Jirectly,by a kind of intellectual expansion, within thething studied: in short, a passage from realiryto concepts and no longer from concepts rorealiry. Is it astonishing that, like children tryingto catch smoke by closing their hands, philosophers so often see the object they would graspfly before them? It is in this way that many ofthe quarrels between the schools are perpetuated, each of them reproaching the others withhaving allowed the real to slip away.

But if metaphysics is to proceed by intuition,if intuition has the mobility of duration as itsobject, and if duration is of a psychical n ature,shall we no t be confining the philosopher to theexclusive contemplation of himself? Will notphilosophy come to consist in watching oneselfmerely live, "as a sleepy shepherd watches thewater flow"?1To talk in this way would be toreturn to the error which, since the beginningof this study, we have not ceased to point our.It would be to misconceive the singular natureof duration, and at rhe same time the essentiallyactive, I might almost say violent, charncter ofmetaphysical intuition. h w ould be failing to secthat the method we speakof alone perm rs us togo beyond ideali!>m, as well as realism, to affirmthe existence of objects inferior and superior(though in a certam sense interior) to us, romake them coexist together without difficulty,and to dissipate gradually the obscurilies thatanalysis accumulates round these great problems. Without entering here upon the study ofthese different points, let us confine ourselves

Rolla, Alfrt'd de Musse1.

a single ;let, but an indefinite s c r i ~ s of c ~ ; : doubtless of the same kind, bur each of a vel)particular species, and how th1s diveniry of ac'corre!>pondsto all tbe degrees of being.If I seck ro analyze duratiOn-that is, to reosolve it mto ready-made concepts-! am co tn.pelled, by the very nature of the concepts aJldof analysis, ro take two opposing views of du.ration ht general, with which I then attemPtto reconstruct it. This combination, which wiJJhave, moreover, something miraculous abouti t-since one does not understand how tWocontranes would ever meet e ; : ~ c h other-ca..presen t neither a diversity of degrees nor a van.cry of forms; like all miracles, it is or 1t is not. Ishall have to say, for example, that there is, 011the one hand, a multiplicity of successive statesof consciousness, and, on the other, a unil)which binds them together. Duration will bethe "synrhesis"of this unity and this mult1pliQiry, a mysterious oper ation which rakes place Gdarkness, and m regard to which, I repeat, onedocs nor see how it would ; : ~ d m i t of shades orof degrees. In this hypothesis there is, and cqonly be, one single duration, that in which ourown consciousness habitually works. To eJGpress it more clearly-if we consider duranoounder the simple aspect of a movement acco,.plishmg itself in space, and we seek to reduceto concepts movement considered as represen-tative of time, we shall have, on the one hand.as great a number of points on the trajecto'1as we m ; : ~ y desire, and, on the other hand, allabstract unity which holds them together as athread holds together the pearls of a necklace.Between this abstract multiplicity and this ~stract unity, the combination, when once it hasbeen posited as possible, is something unique.which will no more admit of shades than doesthe addition of given numbers in arithmetic.But if, instead of professing w analyze duration(i.e., at bottom, to make a synthesis of it widtconcepts), we at o nce place ourselves in it by aaeffort of intuition, we have the feelmg of a cer-tain very determinate tension, in which rhe de-termination itself appears as a cho ice bctweellan infinity of possible durations. Henceforward we can picture to ourselves as many dll"rations as we wish, all very different from eadtother, although each of them, on being reducedto concepts-that is, observed cxtemaJiy fr(lllltwo opposmg points of view-always comes ill

1 e 01any and the one.1 L.et us express the sam_e idea with m o _ r ~ pre-. 011 If I consider durauon as a muluphc1ty. oflSI h lflenrs bound to each other by a un1tywh1c

l f l ~ < through them like a thread, then, howeveroOt> .t>hort the chosen duranon may be, these mo-) ents are unlimited in number. I can suppose~ e l f l as close together as I please; there will' 1ways be between these mathematical points3ther mathematical points, and so on to infi01ty. Looked at from the pc:>int of view oflflultipliciry, then, duranon d1smtegrates mto awder of moments, none of which endures,being an instantaneity. If, on the otherhand , I consider the unity which binds the mo-111ents together, this cannor endure either, sinceby hypothesis e_verything that changing,everything t hat tS really durable m the duratton,has been pur to the account of the multiplicityof moments. As I probe more deeply into itsessence, this unity will appear to me as some1mmobile substratum of that which is moving,as some intemporal essence of time; it IS thisthat I sha ll call eternity; an eternity of death,since it is nothing else than the movementemptied of the mobility which made its life.Closely examined, the opinions of the opposingschools on the subject of duration would beseen to differ solely in this, that they attributea capital importance to one or the other ofthese rwo concepts. Some adhere to the pointof view of the multiple; they set up as concretereality the distinct moments of a rime whichthey have reduced to powder; the unity whichenables us to call the grains a powder they holdto be much more artificial. Other s, on the contrary, set up the unity of duration as concretereality. They place themselves in the eternal.Bur as thei r eternity remains, notwithstanding,abstract, since it is empty, being the eternity of

~ c o n c e p t which, by hypothesis, excludes fromItself the opposing concept, one does not seeh?w this eternity would permit of an indefinite number of moments coexisting in it. Inthe first hypotheSIS we have a world resting on~ o t h i n g , which must end and begin again ofIts own accord at each instant. In the secondWe have an infinity of abstract eternity, about111hich also it is just as difficult to understandhhy it docs no t remain enveloped in itself andb W It allows things to coexist with it. But inth cases, and wh1chever of the two meta-

~ p p e a r s , o{as a mixture of rwo abstractions, which admitof neither degrees nor shades. In one system asin the other, there is only one untque duration,which carries everything with i r -a bot om ess,bankless river, which flows without assignableforce in a direction which coul d not be defined.Even then we can call it only a river, and theriver only flows, because reality obtains fromthe two doctrines this concession, profiting bya moment of perplexity in th eir logic. As soonas they recover from this perplexity, they freezethis flux eith er into an immense solid sheet, orinto an infinity of crystallized needles, alwaysinto a thing which necessarily partakes of theimmobility of 3 pointofview.It is quite otherwise if we place ourselvesfrom the first, by an effort of intuition, in theconcrete flow of duration. Certainly, we shallthen find no logical reason for positing multipleand diverse duratio ns. Strictly, there might wellbe no other duranon than our own, as, forexample, there might be no other color in theworld but orange. Bur just as a consciousnessbased on color, which sympathized internallywith orange instead of perceiving it externally,would feel itself held between red and yellow, would even perhaps suspect beyond thislast color a complete spectrum into which thecontinuity from red to yellow might expandnaturally, so the intuition of our duration, farfrom leaving us suspended in the void as pureanalysis wou ld do, brings us into contact witha whole continuity of durations which we musttry to follow, whether downwards or upwards;in both cases we can extend ourselves indefinitely by an increasingly violent effort, in bothcases we transcend ourselves. In the first weadvance towards a more and more attenuatedduration, the pulsations of which, being morerapid than ours, and dividing our simple sensation, dilute its quality into quantity; at the Limitwould be pure homogeneity, that pure repeti-tion by which we define materiality. Advancingin the other direction, we approach a durationwhich strains, contracts, and intensifies itselfmore and more; at the limit would be eternity.No longer conceptual eternity, which is aneternity of death, bur an eternity of ife. A livingand th erefore still moving eternity in which ourown particular duration would be mcluded asthe vibrations are in light; an eterniry which


9/11

BERGSONwould be the concentration of all duration, asmateriality is its dispersion. Between these twoextreme limits intuition moves, and this movement is the very essence of metaphysics.There can be no question of following herethe various stages of this movement. But having presented a general view of the methodand made a first applicanon of it, it may no tbe amiss to formulate, as precisely as we can,the principles on wh1ch it rests. Most of thefollowing propOSitions have already received inthis essay some degree of proof. We hope todemonstrate them more completely when wecome ro deal with other problems.

I. There 1s a reality that is external and yetgiven immediately to the mind. Common senseis right on this point, as against the idealism andrealism of the philosophers.ll. This reality is mobility. No t things made,bur things in the making, not self-maintainingstates, but only changing states, exisr. Rest isnever more than apparenr, or , rather, relative.The consciousness we have of our ow n self inits continual flux introduces us to the interiorof a reality, on the model of which we mustrepresent other realities. All reality, therefore, istendency, if we agree to mean by tendency anincipient change of direction.

Ill. Ou r mind, which seeks for solid pointsof support, hal> for irs main function in the ordinary course of life that of representing statesand things. It rakes, at long intervals, almostmstantaneous views of the undivided mobilityof the real. lr thus obtains sensations and ideas.In th1s way u substitutes for the continuousthe disconnnuous, fo r motion stability, for tendency in process of change, fixed points marking a direo.:tion of change 3nd tendency. Thissubstitution is necessary to common sense, tolanguage, to practical life, and even, in a certainsense, which we shall endeavor ro determine, topositive science. Ou r i11tellect, when it followsits natural bent, proceeds, on the one hand, bysolid perceptiOIIS, and, the other, by stableconceptions. lr starts from the immobile, andonly conceives and expresses movement as afunction of immobility. It t3kes up its position in ready-made concepts, :1nd endeavors tocatch in rhcm, as in a net, something of thereality wh1ch passes. This is certainly no t donein order to obtain an internal and meraphysicaJ knowledge of the real, bur simply in order

to utilize the real, each concept (as alsosensation} being a practical question whichactivity puts to reality and to which rea)replies, as must be d one in business, by a Yesa No. But, in doing that, it lets that which isvery essence esc:1pe from the real.IV. Th e inherent difficulties ofthe anrinomies which it gives rise to,contradictions into which It falls, theinto antagoniStiC schools, and the iopposition between systems arc largelysuit of ou rapplymg, to the disinterestededge of the real, processes which we genemploy for practical ends. They arise fromfact that we place ourselves in the omrnno\illl'in order to lie in wait for the movingas it passes, instead of replacing o u r ~ c l v ethe moving thing itself, in order towith it the immobile positions. They ariseou r professing to reconstruct realtv-,Nh,is tendency and consequentlypercepts and concepts whose funmake it stationary. With stoppages, hnw, .v . - .numerous they may be, we shall never maktmobility; whereas, if mobility is given, weeby means of diminution, obtain from it btthought as many stoppages 3S we desire. IJother words, it 1s clear that fixed concepts mlljbe extracted by our thought (rom mobile e a l ~bu t there are no means of reconstructing tiJI,mobility of the real w1th fixed concepts. Dot"matism, however, in so far as it has beenbuilder of systems, has always attempted thil:reconstruction.V. In this It was bound to fail. It is on dlifimpotence and on this impotence on ly tldtthe skeptical, idealist, critic3l doctrines reall1dwell: in fact, all doctrines tha t deny to our in-telligence the power of attaining the absolutt.-But because we fail to reconstruct the livir$reality with stiff and ready-made concepts, irdoes nor follow that we cannot grasp it insom'other way. The demonstrtrtions which hlllllbeen given of the relativity of our knowleds'are therefore tai1zted with an original vice; th'%imply, like the dogmatism they attack, that (Ill>;knowledgemust necessarily start from conceiJI',with fixed outlines, in order to clasp with tht*the realitywhich flows.

V I. Bur the truth is that our intelligencefollow the opposite method. It can placewithin the mobile reality, and adopt irslessly changing d1rection; 111 short, c:1n grasp

AN INTRODUCTION TO METAPHYSICSbY means of that intellectual sympathy whichwe call iocuirion. This is extremely difficult.fbe mmd has to do violence to itself, toeverse the direction of the operation by wh1chrt habirually thinks, has perpetually to revise,rather to recast, all its categories. But inrhis way it wilJ attain ro fluid conce pts, capableof following reality in all 1ts smuos1ries andof adopting the very movemenr of the inwardlife of thmgs. Only thus will a progress1ve philosophy be built up, freed from the disputeswhich arise between the various schools, andable ro solve its problems naturally, bcc:1use itwill be released from the artificial expression inrerrns of which such problems are posted. Tophilosophize, therefore, is to invert the habitualdirection of he work of hought.VII. This inversion has never been practicedin a methodical manner; but a profoundlyconsidered history of human thought wouldshow that we owe to it all that is greateslin rhe sciences, as well as all that is pennancnt in metaphysics. The most powerful ofthe methods of investigation at the disposalof tbe human mind, the infinitesimal calculus,originated from this very inversion. Modernmathematics is precisely an effort to substitutethe being made for the ready-made, to followthe generation of magnitudes, to grasp motionno longer from without and in irs displayedresult, but from within and in its tendency tochange; in short, to adopt the mobile continu-Ity of the ourhnes of things. It s true that it isconfined to the outline, being only the sc1enceof magnitudes. It is true also that it has onlybeen able to achieve its marvelous 3pplic3tionsthe invention of certain symbols, and thatthe intuition of which we have just spokenhes at the origin of invention, it is the symbolalone which is concerned in the application_But metaphysics, which :1ims at no application,and usually musl abstain from convertIng intuition into symbols. Liberated from theobligation of working for practically useful re

~ u l t ~ , it will indefinitely enlarge the domain ofIts .tnvestigarions. What it may lose in corn

~ a n s o n with science in utility and exactitude,tt Will regain in range and extension. Though

~ a r h e m a t i c s is only the science of magnitudes,0 1ugh mathematical processes are 3pplicable11 Y to quantities, it must no t be forgotten that

( U ~ n t i t y is always quality m a nascent state;t s, we might say, the limitmg case of equal-

iry. It is natural, then, that metaphysics should:1dopt the generative idea of our mathematicsin order to extend it to all qualities; that is,to reality in general. It will not, by doing this,in any way be moving cowards universal mathematics, that chimera of modem philosophy.On the contrary, the farther it goes, the moreunrrans13rable into symbols will be the objectsit encounters. But it will at least have begunby getnng into contact with the continuity andmob1lity of the real, just where this contactcan be most marvelously utilized. It will havecontemplated itself in a mirror which reAectsan image of itself, much shrunken, no doubt,bm for that reason very luminous. It will haveseen with great er clearness what the mathematical processes borrow from concrete reality,and it will continue in the direction of concrete realiry, and not in that of mathematicalprocesses. Having then discounted beforehandwhat is too modest, and at the same rime tooambitious, in the following formula, we maysay that the object of metaphysics is to performqualitative differentiatiotls and integrations.VIII. The reason why this object has beenlost sght of, and why science itself has beenmistaken in the origin of the p r o c e s ~ e s it employs, is that intuition, once attained, mustfind a mode of expression and of applicationwhich conforms to the habits of our thought,and one which furnishes us, in the shape ofwell-defined concepts, with the solid pointsof support which we so greatly need. In thatlies the condition of what we call exactitudeand precision, and also the condition of theunltm1ted extension of a general method toparticular cases. Now this extension and thiswork of logical improvement can be conti nuedfor centuries, whilst rhe act which creates themethod lasts but fo r a moment. That is why weso often take the logical equipment of sciencefor science itself, forgetting the metaphysicalintuition from which all the rest has sprung.From the overlooking of this intuition proceeds oil that has been said by philosophers andby men of science themselves about the "relativity" of scientific knowledge. What is relativeis the symbolic knowledge by pre-existing concepts, which proceeds from the fixed to themoving, and no t the intuitive ktrowledge whichinstalls itself n that which is moving and adoptsthe very life of things. Th1s intuition attainsthe absolute.


10/11

86 BERGSONScience and metaphysics therefore come together in inruirion. A truly intuitive philosophywould realize the much-desired umon of science and metaphysics. While ir would make

of metaphysics a positive science-that is, aprogressive and indefinitely perfectible one-it would at the same time lead the positive sciences, properly so called, to become consciousof their true scope, often far greater than theyimagine. It would put more science into metaphysics, and more metaphysics imo science.lt wou ld result in restoring rhe continuity berween the intuitions wh ich the various scienceshave obtained here and there in the course oftheir history, and which they have obtainedonly by strokes of genius.IX. That there arc not two different ways ofknowing things fundamentally, that the varioussciences have their root in metaphys1cs, is whatthe ancient philosophers generally thought.Their error did not lie there. It consisted intheir being always dominated by the be lief, sonatural to the human mind, that a variationcan only be the expression and devcl opmen tof what is invariable. Whence it followed thataction was an enfeebled contemplation, duracion a deceptive a nd shifting mage of mmob1leeternity, the Soul a fall from the Idea. Th ewhole of the philosophy wh1ch begins withPlato and culminates in Plotmus is rhe development of a principle which may be formulatedrhus: "There is more in the immutable than inthe moving, and we pass from the stable co theunstable by a mere diminution." Now it is theconrrary which is true.

Modern science dates from the day whenmobility was sec up as an independent reality. Itdares from rhe day when Galileo, setting a ballrolling down an inclined plane, firmly resolvedto study this movement from top ro bottom fo ritself, in itself, instead of seckmg ItS principlein rhe concepts of high and low, two immobilities by which Aristotle believed he couldadequately explain the mo bility. And chis is noran isolated fact in the history of science. Several of rhe great discoveries, of those at leastwhich have transformed the positive sciencesor which have created new ones, have heen somany soundings in the deprhs of pure duration.The more living the reality touched, the deeperwas the sounding.But the lead-line sunk to the sea bottombnng!> up a Auid mass wh1ch the sun's heat

quickly dries into solid andg r a i n ~ of sand. And the intuition of uu.-.mlon_.when it is exposed to the rays of thederstanding, in like manner quickly turns fixed, disnnct, and immobile concepts. Inliving mobthry of things rhe understandingbent on marking real or vinual srations, itdepanures and arrivals; for this is all thatcerns the thought of man in so f ar as it is simhuman. It is more rhan human to graspis happening in the interval. But phtiosopl1can only be an effon to transcend the hcondition.Men of science have fixed their attenmainly on the concepts with which rheymarked out rhe pathway of ntuition.Thethey laid stress on these residualwhich have turned into symbols, the moreattributed a symbolic character to every ki ndscience. And the more they believed in thebolic character of science, the more diddeed make science symbolical. Graduallyhave blorted ou r all difference, in positiveence, between the natural and the artifici:4between the clara of immediate intuition, andthe enormous work of analysis which the understandingpursues round intuition. Thus rhcyhave prepared the way fo r a docrnne wh1chaffirms the relativity of all our knowledge.Bur metaphysics has also labored to thesameenJ.

Ho w could the masters of modern philoso.phy, who have been renovators of science atwell as of metaphysics, have had no sense ofthe moving continuity of reality? How couldthey have abstained from placing themse lvesin what we call concrete duration? They hawtdone so to a grearer extent than they wertaware; above all, much more than they sai d. Ifwe endeavor to link together, by a conrinuoutconnecuon, rhe Intuitions about wh ich systenlfhave become organized, we find, together withother convergent and divergent lines, one vetYdeterminate direction of thought and of feeling. What is this latent thought? Ho w shall wtexpress the feeling? To borrow once morelanguage of rhe Platonists, we will say-depnding the words of their psychological sense, a.ogiving the name of Idea to a cenain t t h ~down into easy intelligibility, and tha t of soulto a cenam longing after the restlessnesslife-that an mvisible current causesphilosophy to place the Soul above the

AN INTRODUCTION TO METAPHYSICSthus rends, like modern science, and evenIt re so than modern science, ro advance in an

roopsite direction ro ancient rhought.opBut thiS metaphysics, like thiS science, basfolded its deeper life in a rich tissue of sym:,ls forgetting sometimes that, while scienceneeds b ~ l s fo r its a n a l y t i ~ a l ~ e v e l o p m e n t ,he rnain obJeCt of metaphysiCS s to do away1 'th symbols. Here, again, the understandingp11rsued its work of fixing, dividing, andeconsrructing. lt has pursued chis, it is true,~ n d e r a rather different form. Without insist-10g on a point which we propose to developelsewhere, it is enough here to say rhat theunderstanding, whose fu nction it is ro operateon srable clements, may look for stability eitherin relations or in things. In so far as it works onconceptS of relations, It culminates in scientificsymbol ism. 1n so far as it works on conceptsof things, it culminates in metaphysicalsymbolisrn. But in both cases the arrangement comesfrorn the understanding. Hence, it wou ld fa inbelieve itself independent. Rather than recognize at once what it owes to an intuition of thedepths of reality, ir prefers exposing itself tothe danger that irs whole work may be lookedupon as nothing but an artificial arrangementof symbols. So rhat if we were to hold on tothe letter of what metaphysicians and scientistssay, and also to the material aspect of what theydo, we might believe that the metaphysicianshave dug a deep tunnel beneath reality, that rhescientists have thrown an elegant bridge over it.but that the moving stream of things passes between these two artificial constructions withour touching them.One of rhe principal anifices of the Kantian

c ~ t i c i s m consisted in taking the metaphysiCian and the scientist lircrally, forcing bothmetaphysics and sc1ence to the extreme limitof symbolism ro which they could go, andto ~ h i c h , moreover, they make their way ofthetr ow n accord as soon as the understandingclaims an independence full of perils. Havingonce overlooked the ties that bind science andmetaphysics to intellectual intuition, Kant hasno difficulty in showing that our science iswh.ol!y relative, and our metaphysics entirely

~ r t 6 c i a l . Since he has exaggerated the indcpenh ~ ~ o f the understandmg in borh cases, since' laS. relieved both metaphysics and scienceas .the Intellectual intuition wh1ch served themInward ballast, science w1th 1ts relations pre-

sents to him no more than a film of form, andmetaphysics, with its things, no more than afilm of matter. Is it surprising that the first,then, reveals tO him only frames packed withinframes, and the second only phantoms chasingphantoms?He has srruck such relling blow s at our science and ou r metaphysics that they have noteven yet quite recovered from their bewilderment. Our mind would readily resign itself toseeing in science a knowledge that is whollyrelative, and in metaphysics a speculation that

is entirely empty. It seems to us, even at thispresent date, that the Kantian criticism apphcs to all metaphysics and to all science. Inreality, 1t applies more especially to the philosophy of the ancients, as also to the formitself borrowed from the ancients-in whichthe moderns have most often left their thought.It is valid against a metaphysics which claimsto give us a single and completed system ofthings, against a science professing to be asingle system of relations; in short, against ascience and metaphysics presenting themselveswith the architectural simplicity of the Piatome rhcory of ideas or of a Greek remple.If metaphysics claims to be made up of concepts which were ours before its advent, if itconsists in an ingenious arrangement of preexisting ideas which we utilize as building material for an edifice, if, in short, it is anyrhingelse but the constant expansion of ou r mmd,the ever-renewed effon ro transcend our actualideas and perhaps also ou r elementary logic,it is but too evident that, Like all rhe worksof pure understanding, it becomes artificial.And if science is wholly and entirely a workof analysis or of conceptual representation, ifexperience is only ro serve therein as a verification for "clear ideas," if, instead of startingfrom multiple and diverse intuition-which insert themselves in rhe particular movement ofeach reality, but do not always dovetail intoeach ocher-it professes to be a vast mathematics, a single and closed-in system of relations,imprisoning the whole of reality in a networkprepared in advance- it becomes a knowledgepurely relative to human understanding. If welook carefully into the Critique of Pure Reason,we see that science for Kant did indeed meanthts kind of universal mathemattcs, and metaphySICS this practically unalter ed Platomsm. Intruth, the dream of a universal mathematics


11/11

88 BERGSONis itself but a survival of Platonism. Universalmathematics is what the world of ideas becomes when we suppose that the Idea consistsin a relationor in a law,and no longer in a thing.Kant took this dream of a few modern philosophers for a reality; more than this, he believedthat aU scientific knowledge was only a detached fragment of, or rather a steppingstoneto, universal mathematics. Hence the main taskof the Critique was to lay the foundation ofthis mathematics-that is, to determine whatthe intellect must be, and what the object, inorder that an Ullinterrupted mathematics maybind them together. And of necessity, i allpossible experience can be made to enter thusinto the rigid and already formed framework ofour understanding, it is {unless we assume a preestablished harmony) because our understanding itself organizes nature, and finds it self againtherein as in a mirror. Hence the possibilityof science, which owes all its efficacy to itsrelativity, and the impossibility of metaphysics,since the latter finds nothing more to do thanro parody with phantoms of things the workof conceptual arrangement which science practices seriously on relations. Briefly, the wholeCritique of Pure Reason ends in establishingthat Platonism, illegitimate i f deas are tllings,becomes legitimate if Ideas are relations, an dthat the ready-made idea, once brought down inthis way from heaven to earth, is in fact, as Platoheld, the common basis alike of hought and ofnature. But the whole of the Critique of PureReason also rests on this postulate, that ourintellect is incapable of aftything bu t Platonizing-that is, of pouring all possible experienceinto pre-existingmolds.

On this the whole question depends. ff scientific knowledge is indeed what Kant supposed,then there is one simple science, preformedand even preformulated in nature, as Aristotlebelieved; great discoveries, then, serve only toilluminate, poin t by poim, the already drawnline of chis logic, immanent in things, just as onthe night of a fete we light up one by one therows of a s ~ j e c s which already outline the shapeof some building. And if metaphysical knowledge is really what Kant supposed, it is reducedto a choice between two attitudes of the mindbefore aU the great problems, both equally possible; its manifestations are so many arbitraryand always ephemeral choices between tw o solutions, virtually formulated from all eternity:

it lives and dies by antinomies. But the truthat modem science does not present thislinear simplicity, nor does modem E ' t " n h , ~ these irreducible oppositions.

Modern science is neither one nor simple.rests, I freely admit, on ideas which in thewe find clear; but these ideas have 5-'.,, . 411become dear through the use made ofthey owe most of their clearness to thewhich the facts, and the applications tothey led, have by reflection shed on r n r r n . --clearness of a concept being scarcely am1thi1,.more at bottom than the certainty, atrained, of manipulating the co ncept proAt its origin, more than one of these rnnr......cmust have appeared obscure, not easilyoncilable wirh the concepts already adtnintinto science, and indeed very near thederline of absurdity. This means thatdoes not proceed by an orderlytogether of concepts predestined to fitother exactly. True and fruitful ideasmany close contacts with currents ofwhich do not necessarily converge on thepoint. However, rhe concepts in whichlodge themselves manage somehow, by ruo:Jm ..off each other's corners, to settle downenough together.

On the other hand, modern metaphysicsnot made up of solutions so radical thatcan culminate in irreducible oppositions.would be so, oo doubt, if there were noof acceptingat the same time and on thelevel the thesis and the antithesis of thetinomies. But philosophy consists preciselythis, that by an effort of intuition oneoneself within that concrete reality, ofthe Critique takes from without the twoposed views, thesis and antithesis. I couldimagine how black and white interpenetratehad never seen gray; but o nce I have seenl easily understand how it can be consl,ac: r""!'lfrom two points of view, that of whiteof black. Doctrines which have a certainof ntuition escape the Kantiancriticismin so far as they are intuitive; and thesetrines are the whole of metaphysics, nrtwt

henri bergson, introduction to metaphysics

Documents