René Wellek - The Term and Concept of Symbolism in Literary History

The Term and Concept of Symbolism in Literary History Author(s): René Wellek Reviewed work(s): Source: New Literary History, Vol. 1, No. 2, A Symposium on Periods (Winter, 1970), pp. 249270 Published by: The Johns Hopkins University Press Stable URL: http://www.jstor.org/stable/468631 . Accessed: 15/08/2012 01:38 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 [email protected].

.

The Johns Hopkins University Press is collaborating with JSTOR to digitize, preserve and extend access to New Literary History.

http://www.jstor.org

The Term and Concept of Symbolism in LiteraryHistory Rene Wellek

is so vast that it cannot even be sketchedwithinthe limits of this paper. The word goes back to ancient Greece and had, there, a complex historywhich has not, I suspect, been traced adequately in the only historyof the term,Max Schlesinger's Geschichtedes Symbols,published in 1912.x What I want to discuss is something much more specific: not even symbol and symbolismin literaturebut the term and concept of symbolismas a period in literaryhistory.It can, I suggest,be convenientlyused as a general term for the literaturein all Western countriesfollowingthe decline of 19th-century realismand naturalism and precedingthe rise of the new avant-gardemovements:futurism, expressionism,surrealism,existentialism,or whateverelse. How has it come about? Can such a use be justified? We must distinguish among differentproblems: the historyof the word need not be identical with the historyof the concept as we might today formulateit. We must ask, on the one hand, what the contemporariesmeant by it, who called himselfa "symbolist"or who wanted to be included in a movementcalled "symbolism,"and on the other hand, what modern scholarshipmightdecide about who is to be included and what characteristics of the period seem decisive.In speaking of "symbolism"as a period-termlocated in historywe must also think of its situation in space. Literary termsmost frequently radiate fromone centerbut do so unevenly;theyseem to stop at the frontiersof some countriesor cross them and languish thereor, surprisingly,flourishmore vigorouslyon a new soil. A geographyof literarytermsis needed which mightattemptto account for the spread and distributionof termsby examining rival termsor accidents of biographyor simply the total situation of a literature. There seems to be a widespreadagreementthat the literaryhistory i

Berlin, 1912.

250

NEW LITERARY

HISTORY

of the centuriessince the end of the Middle Ages can be divided into five successive periods: Renaissance, Baroque, Classicism, Romanticism and Realism. Among these termsBaroque is a comparativenewcomer which has not been adopted everywhere, though thereseems a clear need of a name forthe stylethatreactedagainstthe Renaissance but preceded Classicism.2There is, however,far less agreementas to what termshould be applied to the literaturethat followed the end of the dominance of Realism in the 188os and '9os. The term "Modernism" and its variants such as the German "Die Moderne"3 have been used but have the obvious disadvantagethat theycan be applied to any contemporaryart. Particularlyin English,the term"modern" has preserveditsearlymeaningof a contrastto classicalantiquityor is used for everythingthat occurredsince the Middle Ages. The Cambridge Modern History is an obvious example. The attemptsto discriminatebetweenthe "modern"period now belongingto the past and the "contemporaneous" seem forced, at least terminologically. "Modo," afterall, means "now." "Modernism"used so broadly as to include all avant-gardeart obscures the break between the symbolist movementssuch as futurism,surrealism, period and all post-symbolist it is used as a catchall for everything In etc. the East existentialism, and alienated: it has become a as disapproved decadent,formalistic, term the set glories of Socialist realism. against pejorative The older termswere appealed to at the turnof the centuryby theoristsand slogan writers,who eitherbelieved that thesetermsare applicable to all literatureor consciouslythoughtof themselvesas reviving the styleof an older period. Some spoke of a new "classicism,"particularlyin France,assumingthat all good art mustbe classical. Croce shares this view. Those who felt a kinship with the Romantic Age, mainly in Germany,spoke of "Neuromantik"appealing to Friedrich Schlegel'sdictum that all poetryis romantic.Realism also assertedits claim, mainly in Marxist contexts,in which all art is considered "realistic" or at least "a reflectionof reality."I need only allude to Georg Lukaics'srecentAesthetik,in which this thesisis repeated with obsessive urgency.I have counted the phrase "Widerspiegelungder Wirklichkeit"in the firstvolume; it appears 1,o32 times. I was too lazy or bored to count it in volume 2. All these monismsendanger meaningfulschemesof literaryperiodization.Nor can one be satisfied with a dichotomysuch as FritzStrich's"Klassik und Romantik"which See my papers "The Concept of Baroque in Literary Scholarship" (1945) and "Postscript" (1962) in Concepts of Criticism (New Haven, 1963), pp. 69-127. 3 Eugen Wolff, Die jiingste Literaturstr6mungund das Prinzip der Moderne (Berlin, 1887), seems the source of this form. In 1884 Arno Holz urges "Modern sei der Poet,/ Modern von Scheitel bis zur Sohle." 2

THE TERM AND CONCEPT OF SYMBOLISM

251

leads away fromperiod concepts into a universal typology,a simple division of the world into sheep and goats. For many years I have argued the advantage of a multiple schemeof periods as it permitsa varietyof criteria.The one criterion"realism" would divide all art into realistic and non-realisticart and thus would allow only one approving adjective: "real" or some variant such as "true" or "lifelike." A multiple scheme comes much closer to the actual varietyof the process of history.Period must be conceived neither as some essence which has to be intuited as a Platonic idea nor as a mere arbitrarylinguistic label. It should be understood as a "regulative idea," as a systemof norms, conventionsand values which can be traced in its rise, spread and decline, in competitionwith preceding and followingnorms,conventionsand values.4 "Symbolism"seems the obvious termfor the dominant stylewhich followed nineteenth-century realism. It was propounded in Edmund Wilson's Axel's Castle (1931) and is assumed as a matterof course in Maurice Bowra's Heritage of Symbolism(1943) . We must beware,of course, of confusingthis historicalformwith age-old symbolism,or with the view that all art is symbolic,as language is a systemof symbols. Symbolismin the sense of a use of symbolsin literatureis clearly omnipresentin literatureof many styles,periods and civilizations. Symbolsare all-pervasivein medieval literatureand even the classics of realism- Tolstoy and Flaubert,Balzac and Dickens - use symbols, often prominently.I am myselfguiltyof arguing for the crucial role of symbol in any definitionof Romanticismand I have writtenat lengthon the long German debate fromGoethe to FriedrichTheodor Vischer about the meaning of the term"symbol" and its contrastto the term "allegory."5 For our purposesI want to focus on the fortunesof the concept as a term,firstfora school, then as a movement,and finallyas a period. The term "symbolisme"as the designationfor a group of poets was firstproposed by Jean Mor6as, the French poet of Greek extraction. In 1885 he was disturbedby a journalisticattack on the decadentsin which he was named togetherwith Mallarm6. He protested: "The so-called decadents seek the pure Concept and the eternal Symbolin theirart,beforeanythingelse." With some contemptforthe mania of 4 See my "Periods and Movements in Literary History," in English Institute Annual, r94o (New York, 1941), pp. 73-93, and the chapter "Literary History" in my and Austin Warren's Theory of Literature (New York, 1949) 5 See my paper "The Concept of Romanticism in Literary History" (1949), in Concepts of Criticism (New Haven, 1963), pp. 128-99,and the passages on symbol and allegory in A History of Modern Criticism,4 volumes (New Haven, 1955-65), e.g., I, 21o-11; II, 41-42, 76, 174-75; III, 221-22.

252

NEW LITERARY

HISTORY

criticsfor labels, he suggestedthe term "Symbolistes"to replace the inappropriate"decadents."6In 1886 Moreas starteda reviewLe Symboliste which perished afterfour issues. On September18, 1886, he published a manifestoof "Symbolisme"in Figaro.7 Moreas, however, soon deserted his own brain-childand founded another school he called "6cole romane." On September 14, 1891, in another number of

Figaro Moreas blandlyannounced that "symbolisme"was dead.8 Thus "symbolisme"was an ephemeral name for a very small clique of French poets. The only name still rememberedbesides Moreas's is Gustave Kahn. It is easy to collect pronouncementsby the main contemporarypoets repudiatingthe termfor themselves.Verlaine, in particular, was vehementlyresentfulof this "Allemandisme" and wrote even a little poem beginning"A bas le symbolismemythe/et termite."9 In a way which would need detailed tracing,the term,however, caught on in the later 80o'sand early90's as a blanketname forrecent developmentsin French poetryand its anticipations.Before Moreas' manifesto,Anatole Baju, in Dicadent, April lo, 1886, spoke of Mallarm6 as "the masterwho was the firstto formulatethe symbolicdoctrine."10Two critics,Charles Morice, with La Litteraturede tout a l'heure (1889) and T6odor de Wyz6wa,born in Poland, firstin the essay "Le Symbolismede M. Mallarme" (1887), seemed to have been the main agents, though Morice spoke ratherof "synthese"than of symbol,and Wyzewa thoughtthat "symbol"was only a pretextand explained Mallarme's poetry purely by its analogy to music." As earlyas 1894 Saint Antoine (pseudonymforHenri Mazel) prophesied that "undoubtedly,symbolismwill be the label under which our period will be classed in the historyof French literature."12 Moreasin XIXe 6 Paul Bourdein Le Temps,6 August1885,was the aggressor, art . . . le pur dans leur avant tout cherchent "Les prdtendusd~cadents Sikcle, Conceptet l'6ternelSymbole."Quoted fromGuy Michaud,Messagepodtiquedu symbolisme(Paris,1947), II, 331. 7 Reprintedin Andr6Barre,Le Symbolisme(Paris,1911), p. 11o. 8 Quoted in M. D6caudin,La Crisedes valeurssymbolistes(Toulouse,1960), p. 22.

9 See Barre,pp. 16o-61.Verlaine'sversein Invectives(1896). lo Quoted fromMichaud, II, 335: "Le maitre qui a formul6le premierla doctrinesymbolique." 11 See Michaud,II, 355 ff,cf.427 ff.See also Wyz6wa,Nos Maitres (Paris,1895), du symbolisme: Charles pp. 115-29. On Morice,see Paul Delsemme,Un thdoricien Morice (Paris,1958). On Wyz6wa,Elga L. Duval, Tdodorde Wyzewa:Criticwithout a Country (Geneva, 1961).

12 Michel Decaudin,p. 15; quoted fromL'Ermitage,June,1894."Telle est sans doute l'6tiquettesous laquelle notre p6riode sera class6 dans l'histoirede la litteraturefrangaise."

THE TERM AND CONCEPT OF SYMBOLISM

253

It is still a matterof debate in French literaryhistorywhen this movementcame to an end. It was several timesrevivedexpressly,e.g. in 1905 around a review, Vers et prose. Its main critic,Robert de Souza, in a seriesof articles,"Oh nous en sommes" (also published separately,1906), ridiculed the many attemptsto bury symbolismas prematureand proudlyclaimed that Gustave Kahn, Verhaeren,Viel&Griffin,Maeterlinck and Regnier were then as active as ever.13 Valery professedso complete an allegiance to the ideals of Mallarme that it is difficult not to thinkof him as a continuatorof symbolism, in on though 1938, the occasion of the fiftieth anniversaryof the symbolist manifesto,Valery doubted the existence of symbolismand denied that there is a symbolistaesthetic.14Marcel Proust in the posthumouslypublished last volume of his great series, Le Temps retrouve (1926), formulatedan explicitlysymbolistaesthetic.But his own attitude to symbolistcontemporarieswas often ambiguous or negative.In 1896 Proust had writtenan essay condemningobscurity in poetry.15Proust admired Maeterlinck but disliked PWguyand Claudel. He even wrotea pasticheof Regnier,a mock-solemndescription of a headcold.16When Le Temps retrouve (1926) was published and when a fewyearslater (1933) ValeryLarbaud proclaimedProust a symbolist,symbolismhad, at least in French poetry,definitelybeen replaced by surrealism.17 Andre Barre's book on symbolism (1911) and particularlyGuy Michaud's Message podtique du symbolisme(1947) as well as many other books of French literaryscholarshiphave with the hindsightof literaryhistorians,traced the differentphases of a vast French symbolist movement:the precursorshipof Baudelaire who died in 1867, the second phase when Verlaine and Mallarme were at the heightof their power before the 1886 group, the third phase when the name became established,and then in the twentiethcenturywhat Michaud calls "Neo-symbolisme"representedby "La Jeune Parque" of Valery and L'Annonce faitei&Marie of Claudel, both dating from 1915.18 It seems a coherentand convincingconception which needs to be ex13 Vers et prose. Tome I, Mars - avril - Mai 1905, p. 79. "I1 me semble d'abord que 1'enterrementdu Symbolisme6tait un peu pr6matur6,Craignons les inhumations hitives." 14 "Existance du symbolisme" (1938) in Pleiade ed. (1957), I, 686-706. 15 "Contre l'obscurit6" in Revue blanche, 15 July 1896. Reprinted in Chroniques. 16 For details see Walter A. Strauss, Proust and Literature (Cambridge, Mass., 1957), PP. 191-93, 204. 17 Preface to Emeric Fiser, L'Esthitique de Marcel Proust (Paris, 1933). 18 See also Michaud's paper "Symbolique et symbolisme" in Cahiers de l'Association Internationale des AEtudesFranpaises,VI (1954), 75ff-

254

NEW LITERARY

HISTORY

tended to prose writersand dramatists:to HuysmansafterA Rebours (1884), to the early Gide, to Proust in part and among dramatists,at least to Maeterlinck,who, with his plays L'Intruse and Les Aveugles (1890) and Pelleas et Melisande

(1892), assured a limited penetration

of symbolismon the stage. Knowledge of the French movementand admirationfor it spread soon to the otherEuropean countries.We must,however,distinguish between reportingon French events and even admirationshown by translations,and a genuine transferand assimilationof the French movementin another literature.This processvaries fromcountryto countryconsiderably;and the variation has to be explained by the differenttraditions with which the French importation was confronted. In English, George Moore's Confessionsof a Young Man (1888) and his Impressionsand Opinions (1891) gave sketchyand often poorlyinformedaccountsof Verlaine,Mallarm6,Rimbaud and Laforgue. Mallarme's poetryis dismissedas "aberrationsof a refinedmind," and symbolismis oddly definedas "saying the opposite of what you mean." The three essayson Mallarm6 by Edmund Gosse, all dating from 1893, are hardlymore perceptive.Afterthe poet's death, Gosse turnedsharplyagainst him. "Now that he is no longerhere the truth must be said about Mallarm6. He was hardly a poet." Even Arthur Symons,whose book The SymbolistMovement in Literature (1899) made the decisive breakthroughfor England and Ireland, was very lukewarm at first.While praising Verlaine (in Academy, 1891) he referredto the "brain-sicklittleschool of Symbolistes"and "the noisy little school of Decadents" and even in later articleson Mallarm6 he complained of "jargon and meaningless riddles."19But then, he turned around and produced the entirelyfavorableSymbolistMovement. It should not, however,be overrated as literarycriticismor account of Nerval, Villiers history.It is a ratherlame impressionistic de l'Isle-Adam, Rimbaud, Verlaine, Laforgue, Mallarme, Huysmans and Maeterlinck,with emphasison Verlaine. There is no chapteron the book was dedicated to W. B. Bauderline.20But most importantly, Yeats proclaiminghim "the chiefrepresentativeof that movementin our country."Symonshad made his firsttripto Paris in 1889; he had visited Mallarme, met Huysmans and Maeterlinck,and a year later met Verlaine, who in 1893 became his guest on his ill-fatedvisit to London. Symonsknew Yeats vaguelysince 1891,but theybecame close 19 For referencessee Bruce Morrissette,"Early English and American Critics of French Symbolism," in Studies in Honor of Frederick W. Shipley (St. Louis, Missouri, 1942), pp. 159-80. A chapter on Baudelaire was added to the expanded edition in 1919. 20

THE TERM AND CONCEPT

OF SYMBOLISM

255

friendsin 1895 only afterYeats had completedhis studyof Blake and had elaborated his own systemof symbolsfromother sources: occultism,Blake, and Irish folklore.The edition of Blake Yeats had prepared with Edwin Ellis in 1893 was introduced by an essay on "The Necessityof Symbolism."In 1894 Yeats visited Paris in the companyof Symonsand saw therea performanceof Villiersde 1' IsleAdams's AxIl.21 The essay "The Symbolismof Poetry" (1900) is then Yeats' firstfull statementof his symbolistcreed.22Symons'sdedication to Yeats shows an awarenessof symbolismas an internationalmovement: "In Germany,"he says,exaggeratinggreatly,"it seems to be permeatingthe whole of literature,its spiritis that which is deepest in Ibsen, it has absorbed the one new forcein Italy, Gabriele D'Annunzio. I am told of a group of symbolistsin Russian literature,there is anotherin Dutch literature,in Portugal it has a littleschool of its own under Eugenio de Castro. I even saw some faint stirringsthat way in Spain." Symonsshould have added the United States.Or could he in 1899? There were intelligentand sympatheticreportsof the French movement veryearly. T. S. Perrywrote on "The Latest LiteraryFashion in France" in The Cosmopolitan (1892), T. Child on "LiteraryParis - The New Poetry"in Harper's (1896), and Aline Gorren on "The French Symbolists"in Scribner's (1893). The almost forgottenVance Thompson, who fresh from Paris, edited the oddly named review M'lle New York,wroteseveral perceptiveessays,mainlyon Mallarme in 1895 (reprinted in French Portraits, 1900oo) which convey some

accurate informationon his theoriesand attempteven some explication of his poetry with some success.23But only James Huneker became the main importerof recentFrenchliteratureinto the United States.In 1896 he defendedthe French symbolistsagainst the slurs in Max Nordau's silly Entartung and began to write a long series of articleson Maeterlinck,Laforgue and many others,not botheringto conceal his dependence on his French master,Remy de Gourmontto whom he dedicated his book of essays, Visionaries (1905).24 But the

See Richard Ellmann's Introduction to the 1958 New York reprint of The 21 SymbolistMovement. On Symonssee Roger Lhombreaud, Arthur Symons,A Critical Biography (London, 1963), and Ruth Zabriskie Temple, The Critic's Alchemy: A Study of the Introduction of French Symbolism into England (New Haven, 1953) 22 Reprinted in Ideas of Good and Evil (1903); since in Essays and Introductions (New York, 1961), pp. 153-64. See Morrissette'spaper quoted in note 19. 23 See Arnold T. Schwab, J. G. Huneker, Critic of the Seven Arts (Stanford, 24 1963) -

256

NEW LITERARY

HISTORY

actual impact of French symbolistpoetryon American writingwas greatly delayed. Rene Taupin in his L'Influence du symbolisme franpaissur la podsie amfricaine (1929) traced some echoes in forgotten Americanversifiers of the turnof the centurybut only two Americans living then in England, Ezra Pound around 1908 and T. S. Eliot around 1914, reflectthe French influencein significantpoetry. More recentlyand in retrospectone hears of a symbolistperiod in American literature:Hart Crane and Wallace Stevens are its main poets, Henry James, Faulkner and O'Neill, in very differentways and in different with its stagesof theircareer,show marked affinities teshniques and outlook. Edmund Wilson's Axel's Castle (1931) was apparentlythe veryfirstbook whichdefinitelyconceivedof symbolism as an internationalmovementand singled out Yeats, Joyce,Eliot, GertrudeStein, Valery,Proust, and Thomas Mann as examples of a movementwhich,he believed, had come to an end in the time of his writing.Here we find the conception formulatedwhich, very generally,is the thesisof this paper and the assumptionof many historians since Wilson's sketch.Wilson's sourceswere the writingsof Huneker whom he admiredgreatly,and the instructionin Frenchliterature he receivedat PrincetonfromChristianGauss.25But the insightinto the unity and continuityof the internationalmovement and the selectionof the great names was his own. We mightonly deplore the inclusion of Gertrude Stein. But I find it difficultto believe that Wilson's book could have had any influenceoutsidethe English-speaking world. In the United States,Wilson's reasonable and moderateplea foran internationalmovementwas soon displaced by attemptsto make the whole of the Americanliterarytraditionsymbolist.F. O. Matthiessen's The AmericanRenaissance (1941) is based on a distinctionintroduced by Goethe. Allegoryappears as inferiorto symbol: Hawthorneinferior to Melville. But in Charles Feidelson's Symbolismand American Literature (1956) the distinctionbetweenmodem symbolismand the use of symbolsby Romantic authors is completelyobliterated.Emerson, Hawthorne,Poe, Melville, and Whitman appear as pure symbolists avant la lettreand their ancestryis traced back to the Puritans who, paradoxically, appear as incomplete,frustratedsymbolists.It can be objected that the old Puritanswere sharplyinimical to images and symbolsand that thereis a gulf betweenthe religiousconception of signs of God's Providenceand the aestheticuse of symbolsin the 25 On Huneker see Classics and Commercials (New York, 1950), p. 114, and The Shores of Light (New York, 1952), p. 73- On Gauss the essay introducingthat volume.

THE TERM AND CONCEPT

OF SYMBOLISM

257

novelsof Hawthorneand Melville and even in the Platonizingaesthetics of Emerson.26 The symbolistconception of American literatureis still prevalent today.It owes its dominance to the attemptto exalt the greatAmerican writersto myth-makers and providers of a substitutereligion. Ishmael in James Baird, (1956), puts it unabashedly,Melville is "the of artistic creatorengaged in the act of making the supremeexample new symbolsto replace the 'lost' symbolsof ProtestantChristianity."27 A very active trend in American criticismexpanded symbolistinterpretationto all typesand periods of literatureimposingit on writings whichhave no such meaningor have to be twistedto assume it. Harry Levin rightlycomplained in an address, "Symbolismand Fiction" (1956), that "everyhero may seem to have a thousand faces; every heroine may be a white goddess incognita; and every fishingtrip turnsout to be another quest for the Holy Grail."28The impact of ideas from the Cambridge anthropologistsand from Carl Jung is obvious. In the studyof medieval texts,a renewedinterestin the fourfold levels of meaning in Dante's "Letter to Can Grande" has persuaded a whole group of American scholars to interpretor misinterpret Chaucer, the Pearl poet, and Langland, in these terms.29They should bear in mind that Thomas Aquinas recognizedonly a literal sense in a work inventedby human industryand that he reservedthe other three senses for Scripture.30The symbolist interpretation reaches heights of ingenuityin the writingof Northrop Frye who began with a book on Blake and, in The Anatomy of Criticism (1957), conceived of the whole of literatureas a self-enclosedsystem of symbolsand myths,"existingin its own universe,no longera commentaryon life or reality,but containinglife and realityin a system of verbal relationships."In this grandiose conceptionall distinctions between periods and stylesare abolished: "the literaryuniverse is a universein which everythingis potentiallyidentical with everything Cf. Ursula Brumm,Die religibse Typologie im amerikanischenDenken (Leiden, 1963), e.g., p. 8ff. 27 Baltimore, 1956,p. xv. 28 Contextsof Criticism (Cambridge,Mass., 1957), p. 007. 26

29 I allude particularly to D. W. Robertson's A Preface to Chaucer (Princeton, 1963), and D. W. Robertson and B. F. Hupp6's Piers Plowman and Scriptural Tradition (Princeton,1951) . 30o Cf. Morton W. Bloomfield, "Symbolism in Medieval Literature" in Modern Philology, LVI (1958), pp. 73-81. He quotes Thomas Aquinas, Questiones quodlibetales, VII. a. 16. "Unde in nulla scientia, humana industria inventa, proprio loquendo, potest inveniri nisi litteralissensus; sed solum in ista Scriptura, cujus Spiritus sanctus est auctor, homo verum instrumentum."

258

NEW LITERARY HISTORY

else."31Hence the old distinctionsbetweenmyth,symboland allegory disappear. One of Frye's followers,Angus Fletcher,in his book on Allegory (1964), exalts allegoryto the centralprocedureof art, while Frye still holds fast to symbolism,recognizingthat "the criticsare often prejudiced against allegory without knowing the real reason, which is that continuousallegoryprescribesthe directionof his commentary,and so restrictshis freedom."32 The storyof the spread of symbolismis very differentin other countries. The effectin Italy was ostensiblyrather small. Soffici's pamphlet on Rimbaud, in 1911, is usually consideredthe beginning of the Frenchsymbolistinfluence,but therewas an earlypropagandist forMallarme,VittorioPica, who was heavilydependenton his French sources,particularlyT'odor de Wyz'wa. His articles in the Gazette letteraria (1885-6) on the French poets do not use the term; but in 1896 he replaced "decadent" and "Byzantine" by "symbolist."33 D'Annunzio, who knew and used some French symbolists,would be classed as "decadent" today, and the poets around Ungaretti and Montale as "hermetic." In a recent book by Mario Luzi, L'idea simbolista (1959), Pasoli, Dino Campana, and Arturo Onofri are called symbolistpoets,but Luzi uses the termso widelythathe begins his anthologyof symbolismwith H61derlin and Novalis, Coleridge and Wordsworth,and can include Poe, Browning,Patmore, Swinburne, Hopkins and Francis Thompson among its precursors.Still, his list of symbolistpoets,French,Russian, English,German,Spanish and Greekis, on the whole,reasonable.34Onofriwas certainlystrongly influencedby Mallarme and later by Rudolf Steiner;Pascoli, however, seems to me no symbolistin his poetry,though he gave extremely symbolistinterpretationsof Dante.35 It might be wiser to think of "ermetismo"as the Italian name forsymbolism:Montale and possibly Campana are genuine symbolists. While symbolismat least as a definiteschool or movementwas absent in Italy, it is central in the historyof Spanish poetry.The Nicaraguan poet, Ruben Dario initiatedit afterhis shortstayin Paris 31 Princeton,1957, PP. 122, 124. 32 Ibid., p. 9o. 33 See Olga Ragusa, "Vittorio Pica: First Champion of French Symbolism in Italy" in Italica, XXXV (1958), 255-61, and Luigi de Nardis, "Prospettive critiche per uno studio su VittorioPica e il decadentismofrancese"in Revista de letterature moderne e comparate, XIX (1966), 202-9. 34 Milano, 1959. Luzi lists besides the French Bryusov, Balmont, Ivanov, Blok, Yeats, Eliot; George, Hofmannsthal, Rilke, Benn; Pascoli, D'Annunzio, Onofri, Campana; Dario, Antonio Machado, Jimenez,and the Greek Chantzopoulos. 35 Pascoli, Minerva oscura (1898), Conferenzee studi dantesche (1921), etc.

THE TERM AND CONCEPT OF SYMBOLISM

259

in 1892. He wrotepoems under the symbolistinfluenceand addressed, for instance,a ferventhymn to Verlaine.36The influenceof French symbolistpoetrychanged completelythe oratoricalor popular styleof Spanish lyrical poetry. The closeness of Guillhn to Mallarme and Valhry seems too obvious to deny and the Uruguayan poet Julio Herrera y Reissig (1873-1909) is clearly in the symbolisttradition, often of the obscurestmanner.37Still, the Spanish critics favor the term "Modernismo" which is used sometimesso inclusivelythat it covers all modern Spanish poetryand even the so-called "generation of 1898," the prose writersAzorin,Baroja and Unamuno, whose associations with symbolismwere quite tenuous.38"Symbolism"can apply only to one trendin modernSpanish literatureas the romanticpopular traditionwas therestrongerthan elsewhere.Garcia Lorca's poetry can serveas the best known example of the peculiar Spanish synthesis of the folksyand the symbolical,the gipsy song and myth.Still, the continuityfrom Dario to Jimenez,Antonio Machado, Alberti, and then to Guillen seems to me evident. Jorge Guillen in his Harvard lectures,Language and Poetry (1961), finds "no label convincing." "A period look," he argues,does not signifya "group style."In Spain therewere,he thinks,fewer"isms" than elsewhereand the break with the past was far less abrupt. He reflectsthat "any name seeking to give unity to a historicalperiod is the invention of posterity."But while eschewingthe term "symbolism,"he characterizeshimselfand his contemporarieswell enough by expounding theircommon creed: their belief in the marriageof Idea and Music, in short,their belief in the ideal of Mallarme.39Following a vague suggestionmade by Remy de Gourmont,the rediscoveryof G6ngora by Ortega y Gasset, Gerardo Diego, DdimasoAlonso, and Alfonso Reyes around 1927 fits into the picture: they couple G6ngora and Mallarme as the two poets who in the historyof all poetryhave gone furthestin the search for absolute poetry,for the quintessenceof the poetic.40 In Germany,the spread of symbolismwas far less complete than 36 "Verlaine: Responso" beginning "Padre y maestro migico, liriforo celeste." On Dario see E. K. Mapes, L'Influence frangaise dans l'oeuvre de Rubdn Dario (Paris, 1925). 37 Cf. Bernard Gicovate, Julio Herrera y Reissig (Berkeley, 1957). 38 See Gustav Siebenmann, Die moderne Lyrik in Spanien (Stuttgart,1965), esp., pp. 43ff.,and Guillermo Diaz-Plaja, Modernismo frentea Noventa y Ocho (Madrid, 1951). 39 Cambridge,Mass., 1961, p. 2144o Remy de Gourmont, Promenades littdraires,IVe s6rie (Paris, 1912). Ddmaso Alonso, G6ngora y la literatura contempordnea (Santander, 1932); also in Estudios y ensayos g6ngorinos (Madrid, 1955).

260

NEW LITERARY

HISTORY

Symonsassumed in 1899. Stefan George had come to Paris in 1889, had visited Mallarm6 and met many poets, but after his return to Germany he avoided, I assume deliberately,the term "symbolism" for himselfand his circle. He translateda selectionfromBaudelaire (1891) and smallersamples fromMallarm6,Verlaine and Regnier in Zeitgen*ssischeDichter (1905), but his own poetrydoes not, I think, show very close parallels to the French masters.Oddly enough, the seem to have left the most clearly discernible poems of Viel6-Griffin traceson George's own writings.41 As early as 1892 one of George's Carl adherents, August Klein, protested in George's periodical, die Kunst,against the view of George'sdependenceon the Bliitterfiir French. Wagner, Nietzsche,B6cklin and Klinger,he says,show that thereis an indigenousopposition to naturalismin Germanyas everywhere in the West.42George himselfspoke later of the French poets as his "formerallies" and in Gundolf'sauthoritativebook on George, the French influenceis minimizedif not completelydenied.43Among the theoristsof the George circleFriedrichGundolf had the strongest symbolistleanings: Shakespeare und der deutsche Geist (191i) and with Goethe (1916) are based on the distinctionof symbol-allegory symbol always the higher term.44Still, the term symbolismdid not catch on in Germanyas a name forany specificpoetic group, though Hofmannsthal,e.g. in "Das Gesprich iiber Gedichte" (1903), proclaimed the symbolthe one element necessaryin poetry.45Later, the influenceof Rimbaud - apparentlylargelyin German translationBut if we on Georg Trakl can be demonstratedwith certainty.46 examine German books on twentieth-century literature,symbolism seems rarelyused. I found a section so called in Willi Duwe's Die Dichtung des 2o. Jahrhunderts(1936) which includes Hofmannsthal, 41 See B. B6schenstein,"Wirkungen des franzbsischenSymbolismusauf die deutsche Lyrik der Jahrhundertwende,"in Euphorion, LVIII (1964), 375-95. Werner Vordtriede,"Direct Echoes of French Poetry in Stefan George's Works" in Modern Languages Notes, LX (1945), 461-68, lists trivial parallels to Baudelaire and Mallarm6. More in Claude David, Stefan George. Son oeuvre podtique (Paris, 1952). 42 Vol. I, No. 2, "Ober Stefan George, eine neue Kunst", reprintedin Die Sendung Stefan Georges (Berlin, 1935), pp. 69-70. 43 "Stern des Bundes," quoted in David, p. 285. Gundolf, George (Berlin, 1920o), PP. 50-51. 44 Shakespeare und der deutsche Geist (Berlin, 1914), pp. 1-2 for distinction of symbol-allegory;and Goethe (Berlin, 1916), pp. 16, 28, for classification of Goethe's works. 45 Prosa, II, 104. 46 See B6schenstein,quoted in note 41, and Herbert Lindenberger,"Georg Trakl and Rimbaud," in Comparative Literature,X (1958), 21-35. Trakl read the translation by K. L. Ammer (pseudonymof Karl Klammer) published in 1907.

THE TERM AND CONCEPT

OF SYMBOLISM

261

Dauthendey,Cale, Rilke and George,while E. H. Liith's Literaturals Geschichte (Deutsche Dichtung von 1885 bis 1947), publishedin 1947, treatsthe same poets under the label "Neuromantikund Impressionismus." Later, however,we find a section "Parasymbolismus"which deals with Musil and Broch. Hugo Friedrich,in his Strukturder modernenLyrik (1956), avoids the term and argues that the quick successionof moderniststyles:dadaism, surrealism,futurism,expressionism,unanimism,hermetism,etc. createsan optical illusion which hides the fact of a directcontinuitybetween Mallarmb,Valhry,Guil16n, Ungaretti and Eliot.47 The little anthologyin the back of the book adds St. John Perse,Jiminez,Garcia Lorca, Albertiand Montale to these names. Friedrich'slist seems to me the list of the main symbolist poets even though Friedrich objects to the name. Clearly, German literaryscholarship has not been converted to the term, though Wolfgang Kayser's article "Der europtiischeSymbolismus" (1953), had pleaded for a wide concept in which he included, in addition to the French poets,D'Annunzio, Yeats, Valery,Proust,Virginia Woolf and Faulkner.48 In Russia we findthe strongestsymbolistgroup of poets who called themselvesthat. The close links with Paris at that time may help to explain this,or possiblyalso the strongconsciousnessof a traditionof symbolismin the Russian Church and in some of the Orthodox thinkersof the immediatepast. VladmirirSolovev was thoughtof as a precursor.In 1892 Zinaida Vengerovawrote a sympatheticaccount of the French symbolistsfor VestnikEvropy49while in the following year Max Nordau's Entartung caused a sensation for its satirical account of recentFrenchpoetrywhichreverberatedas late as in Tolstoy's What is Art? (1898). Bryusovemergedas the leading symbolist poet: he translated Maeterlinck'sL'lntruse and wrote a poem "Iz Rimbaud" as earlyas 1892.50In 1894 he published two little volumes under the title Russkie simvolisty.That year Bryusovwrote poems with titlessuch as "In the spiritof the Frenchsymbolists"and "In the manner of Stephane Mallarme" (though thesewere not published till 1935) and brought out a translationof Verlaine's Romances sans paroles.51 Bryusov had later contacts with Rene Ghil, Mallarme's pupil, and derived fromhim the idea of "instrumentation"in poetry 47 Hamburg, 1956,p. io8. 48 In Die Vortragsreise(Bern, 1958), pp. 287-304. 49 IX (1892), 115-43. Reprinted in Literaturnye Kharakteristiki (St. Petersburg, 1897) . 50 cf. G. Donchin, The Influence of French Symbolism on Russian Poetry (The Hague, 1958), p. 2351 In Neizdannye stikhotvoreniya(Moscow, 1935), PP. 426, 428.

262

NEW LITERARY

HISTORY

which was to play a great role in the theoriesof the Russian formalhad, in 1893,published ists.52In the meantimeDimitri Merezhkovsky a manifesto:"On the causes of the decline and the new trendsof contemporary Russian literature," which recommended symbolism though Merezhkovskyappealed to the Germans: to Goethe and the Romanticsratherthan to the French.53Merezhkovsky's pamphletforeshadows the split in the Russian symbolistmovement.The younger men, Blok and VyacheslavIvanov as well as Bely distancedthemselves from Bryusov and Balmont. Blok in an early diary (1901-02) condemned Bryusovas decadent and opposed to his Parisan symbolism his own, Russian,rootedin the poetryof Tyutchev,Fet, Polonsky,and Soloviv.54VyacheslavIvanov in 1910o,shared Blok's view. The French influenceseemed to him "adolescentlyunreasonable and, in fact,not veryfertile,"while his own symbolismappealed to Russian nationalism and to the general mysticaltradition.55Later Bely was to add occultism, Rudolf Steiner and his "anthroposophy."The group of poets which called themselves"Acmeists" (Gumilev, Anna AkhmaThe tova, Osip Mandelshtam) was a directoutgrowthof Symbolism.56 mere factthat theyappealed to the earlysymbolistInnokentyAnnenskyshows the continuitywith Symbolismin spite of theirdistastefor the occult and their emphasis on what they thoughtof as classical clarity.SymbolismdominatesRussian poetrybetweenabout 1892 and 1914 when Futurismemergedas a slogan and the Russian formalists attacked the whole concept of poetryas imagery. If we glance at the other Slavic countrieswe are struckby the diversityof theirreactions.Poland was earlyinformedabout the French movement,and Polish poetrywas influencedby the French symbolist movementbut the term "Mltoda Polska" was preferred.In Wilhelm Feldmann's Wspdtczesnaliteraturapolska (1905) contemporarypoetryis discussedas "decadentism"but Wyspiaxiski(a symbolistif ever there was one) appears under the chapterheading: "On the heights All the historiesof Polish literatureI have seen of romanticism."57 and Ghil's Traitd du 52 See Lettres de Rend Ghil (Paris, 1935), pp. 13-16, 18-20o, verbe (Paris, 1886). 53 0 princhinakh upadka i o novykh techenyakhsovremennoyrusskoyliteratury (St. Petersburg,1893). 54 "Yunocheski dnevnik Aleksandra Bloka" (1901-2), in Literaturnoe Nasledstvo, XXVII-XXVIII (1937), 302. 55 "Zavety simvolizma," in Apollon, VIII (1910), 13, and in Borozdy i mezhi (Moscow, 1916), p. 133. 56 For a good discussion see Jurij Striedter,"Transparenz und Verfremdung:Zur Theorie des poetischen Bildes in der russichen Moderne" in Immanente Aesthetik: AesthetischeReflexion,ed. Wolfgang Iser (Munich, 1966), pp. 263-89. 57 In Vol. III: "Na wyiynach romantyzmu."

THE TERM AND CONCEPT OF SYMBOLISM

263

speak of "Modernism," "Decadentism," "Idealism," "Neo-romanticism" and occasionallycall a poet such as Miriam (Zenon Przesmycki) a symbolistbut theyneverseem to use the termas a general name for a period in Polish literature.58 In Czech literature the situation was more like that in Russia: Bfezina, Sova, and Hlavaicekwere called symbolistsand the idea of a school or at least a group of Czech symbolistpoets is firmly established. The term "Moderna" (possibly because of the periodical, Moderni Revue founded in 1894) is definitelyassociated with decadentism, fin de sidcle, a group representedby Arno't Prochizka. A hymnical, optimistic,even chiliasticpoet such as Bfezina cannot and could not be classedwith them.The greatcriticF. X. Salda wroteof the "school of symbolists"as earlyas 1891,calling Verlaine,Villiersand Mallarm6 its masters but denying that there is a school of symbolistswith His very firstimportant article dogmas, codices and manifestoes.59 "Synthetismin the new art" (1892) expounded the aesthetics of Morice and Hennequin forthe benefitof the Czechs,then still mainly dependenton German models.60 The unevennessof the penetrationboth of the influenceof the French movementand verystrikinglyof the acceptance of our term raises the question whetherwe can account for these differencesin causal terms.It sounds hereticalor obscurantistin this age of scientific explanation to ascribe much to chance, to casual contacts and personal predilections.Why was the term so immenselysuccessful in France, in the United States and in Russia, less so in England and Spain and hardlyat all in Italy and Germany?In Germanythere was even the traditionof the continuous debate about symbolsince Goethe and Schelling; before the French movementFriedrichTheodor Vischerdiscussedthe symbolelaboratelyand still the termdid not catch on.6xOne can thinkof all kinds of explanations: a deliberate 58 Zenon Presmycki had written an essay on Maeterlinck in 1891 (in ?wiat). More in Henryk Markiewicz,"Mtoda Polska i 'izmy'," in Iz Problem6w literatury polskiej XX wieku, Tome I (Warsaw, 1965), PP- 7-51, esp., p. 15; Teofil Wojefiski, Historia literaturypolskiej (Warsaw, 1946) has a chapter entitled "Symbolism i Neoromantyzmw Polsce"; Julian Krzyzanowski,Neoromantyzm Polski, z89o-1918 (Wroclaw- Warszawa, 1963), has a chapter, "Drama naturalistyczno-symboliczny," pp. 182ff. 59 "0 kole symbolistu" in Kritchkdprojevy (Prague, 1947), I, 185-86. Originally as "Zasldno" in Literarni listy, XIII (1891), 46-68, 65-66, 85-86. See J. Pistorius, Bibliografie dila F. X. Ialdy (Prague, 1948), p. 79. 60 "Synth6tismv nov6m um~ni," originally in Literarni listy (1891-2). A brief discussion in my "Modern Czech Criticism and Literary Scholarship," in Essays on Czech Literature (The Hague, 1963), pp. 179-80. 61 "Das Symbol" (1887) in Altes und Neues, Neue Folge, 1889.

264

NEW LITERARYHISTORY

decision by the poets to distancethemselvesfromthe French developments; or the success of the terms"Die Moderne" and "Neuromantik." Still, the verynumberof such explanationssuggeststhat the variables are so greatthatwe cannot account forthesedivergenciesin any systematicmanner. If we, at long last, turn to the centralquestion: what is the exact contentsof the term,we must obviouslydistinguishamong the four concentriccircles definingits scope. At its narrowest,"symbolism" refersto the French group which called itselfso in 1886. Its theory These poets mainlywanted poetryto be nonwas ratherrudimentary. for a break with the traditionof Hugo and asked i.e. rhetorical, they the Parnassiens.They wantedwordsnot merelyto statebut to suggest; they wanted to use metaphors,allegories and symbolsnot only as decorationsbut as organizingprinciplesof theirpoems; theywanted their verse to be "musical," in practice to stop using the oratorical cadences of the French alexandrines, and in some cases to break completely with rhyme. Free verse - whose invention is usually ascribed to Gustave Kahn - was possiblythe most enduringachievementwhich has survivedall vicissitudesof style.Kahn himselfin 1894 summed up the doctrine simply as "antinaturalism,antiprosaismin poetry,a search for freedomin the effortsin art, in reaction against the regimentationof the Parnasse and the naturalists."62 This sounds verymeager today: freedomfromrestrictionshas been afterall, the slogan of a great many movementsin art. It is betterto thinkof "symbolism"in a wider sense: as the broad movement in France from Nerval and Baudelaire to Claudel and Valkry.We can restatethe theoriespropoundedand will be confronted by an enormousvariety.We can characterizeit more concretelyand say,forexample, that in symbolistpoetrythe image becomes "thing." The relation of tenor and vehicle in the metaphoris reversed.The utterance is divorced, we may add, from the situation: time and place, historyand societyare played down. The innerworld,la durde, in the Bergsoniansense, is representedor oftenmerelyhinted at as "it," the thing or the person hidden. One could say that the grammatical predicate has become the subject. Clearly such poetrycan easily be justifiedby an occult view of the world. But this is not necessary:it might simplyimply a feelingfor analogy,for a web of correspondences,a rhetoricof metamorphosesin which everything reflectseverythingelse. Hence the great role of synaesthesia,which, 62 Decaudin, p. 15; quoted from La Socidtd nouvelle, avril, 1894. "Anti-naturalisme, anti-prosaismede la podsie, recherchede la libert6 dans des effortsdans l'art, en reaction contre 1'enr6gimentationparnassienne ou naturaliste."

THE TERM AND CONCEPT OF SYMBOLISM

265

thoughrooted in physiologicalfactsand found all over the historyof poetry,became at that time merelya stylisticdevice, a mannerism This characterizationcould be easily imitated and transmitted.63 elaborated considerablyif we bear in mind that styleand world-view go togetherand only togethercan definethe characterof a period or even of a single poet. Let me try to show, at least, how diverse and even incompatible were the theoriesof two such related poets as Baudelaire and MallarmL. Baudelaire's aestheticis mainly "romantic"; not in the sense of emotionalism,nature worship and exaltation of the ego, central in French romanticism,but ratherin the English and German tradition of a glorificationof creativeimagination,a rhetoricof metamorphoses and universalanalogy. Though thereare subsidiarystrandsin Baudelaire's aesthetics,at his finest,he graspsthe role of imagination, "constructiveimagination,"as he calls it in a termultimatelyderived fromColeridge.64It gives a metaphysicalmeaning,"a positiverelation with the infinite."65Art is another cosmos which transformsand hence humanizesnature. By his creation the artistabolishes the gulf between subject and object, man and nature. Art is "to create a suggestivemagic containingat one and the same time the object and the subject, the external world and the artisthimself."66 Mallarm6 says almost the opposite in spite of some superficial resemblancesand the common attachmentto Poe and Wagner. Mallarm6 was the firstpoet radically discontentwith the ordinarylanguage of communication;he attemptedto construean entirelyseparate language of poetryfar more consistentlythan older cultivatorsof "poetic diction" such as the practitionersof trobarclus, or G6ngora or Mallarme's contemporary,Gerard Manley Hopkins. His aim of transforminglanguage was no doubt in part negative: to exclude society,nature and the person of the poet himself.But it was also positive: language was again to become "real," language was to be magic,wordswere to become things.But thisis not, I think,sufficient reason to call Mallarm6 a mystic.Even the depersonalization he 63 See the many articles by Albert Wellek, e.g., "Das Doppelempfinden in der Geistesgeschichte,"in Zeitschriftfiir Aesthetik, XXIII (1929), 14-32; "Das Doppelempfinden im 18. Jahrhundert," in Deutsche Vierteljahrschriftfiir Geistesgeschichte und Literahurwissenchaft,XIV (1936), 75-102. 64 "Constructiveimagination" quoted in English fromMrs. Catherine Crowe, The Night Side of Nature, in Curiositdsesthdtiques,Conard ed. (Paris, 1923), p. 27965 Ibid., p. 275. "Elle est positivementapparent6e avec l'infini." 66 L'Art romantique, Conard ed. (Paris, 1925), p. 119: "C'est cr6er une magie suggestivecontenant A la fois l'objet et le sujet, le monde ext6rieur i l'artiste et

l'artistelui-mIme."

266

NEW LITERARY

HISTORY

requires is not mystical.Impersonalityis rather objectivity,Truth. Art reaches for the Idea, which is ultimatelyinexpressible,because so abstractand general as to be devoid of any concretetraits.The term"flower"seemsto him poetic because it suggeststhe "one, absent fromall bouquets."67Art thus can only hint and suggest,not transformas it should in Baudelaire. The "symbol"is only one device to achieve this effect.The so-called "negative" aestheticsof Mallarme is thus nothing obscure. It had its psychologicalbasis in a feeling of sterility,impotence and final silence. He was a perfectionist,who proposed somethingimpossible of fulfillment: the book to end all books. "Everythingon earth existsto be containedin a book."68Like many poets before him, Mallarme wants to express the mysteryof the universe but feels that this mysteryis not only insoluble and immenselydark but also hollow, empty,silent, Nothingnessitself. There seemsno need to appeal to Buddhism,Hegel, Schopenhaueror Wagner to account for this.09The atmosphereof nineteenth-century pessimismand the general Neo-Platonic traditionin aestheticssuffice. Art searchesfor the Absolute but despairs of ever reaching it. The essence of the world is Nothingness,and the poet can only speak of this Nothingness.Art alone survives in the universe. Man's main vocation is to be an artist,a poet, who can save somethingfromthe general wreckage of time. The work or, in Mallarm6's terms,the Book, is suspended over the Void, the silentgodless Nothingness,Poetryis resolutelycut offfromconcretereality,fromthe expressionof the personalityof the poet, from any rhetoricor emotion, and becomes only a Sign, signifyingNothing.70In Baudelaire, on the other hand, poetrytransformsnature, extractsflowersfromevil, creates a new myth,reconcilesman and nature. But if we examine the actual verse of the symbolistsof this period we cannot be content with formulaseither of creative imagination, suggestion,pure or absolute poetry. 67 Oeuvres completes, Pl6iade ed. (Paris, 1949), p. 368: "une fleur . . l'absente de tous bouquets." 68 Ibid., p. 378: "Tout, au monde, existe pour aboutir Aun livre." 69 Jacques Scherer, L'Expression litteraire dans l'euvre de Mallarmd (Paris, 1947), PP- 155 ff,collects evidence for Mallarm"'s contacts with Platonism and occultism. Mallarme denied knowledge of Buddhism, Propos sur la podsie, ed. H. Mondor (Monaco, 1946), p. 59. Hasye Cooperman, The Aesthetic of Stiphane Mallarmd (New York, 1933), makes much of the influence of Wagner. Th-e only evidence of concern for Hegel is a letter of Villiers d'Isle-Adam to Mallarm6, quoted in Henri Mondor, Vie de Mallarmd (Paris, 1941), p. 222; "Quant A Hegel, je suis vraiment bien heureux que vous ayez accorde quelque attention A ce miraculeux genie." 70 See Guy Defel, L'Esthedtiquede Stdphane Mallarmd (Paris, 1951).

THE TERM AND CONCEPT OF SYMBOLISM

267

On the thirdwider circle of abstractionwe can apply the term to the whole period on an internationalscale. Every such termis arbitrary,but symbolismcan be defendedas rooted in the conceptsof the period,as distinctin meaningand as clearlysettingoffthe period from that precedingit: realism or naturalism.The differencefromromanticismmay be less certainlyimplied. Obviously there is a continuity with romanticism and particularly German romanticism,also in France as has been recentlyargued again by WernerVordtriedin his Novalis und die franz6sichenSymbolisten (1963) .71 The direct contact of the French with the German romanticscame late and should not be overrated. Jean Thorel in "Les Romantiques allemandes et les symbolistsfrangaises,"seems to have been the firstto point out the relation.72 Maeterlinck'sarticle on Novalis (1894) and his But Wagner of little anthology (1896) came late in the movement.73 course mediated between the symbolistsand German mythology though Mallarme's attitude,admiring toward the music, was tingled with ironyforWagner's subject matter.74 Early in the century,Heine, called a romantique defroqueas he himself,played the role of an intermediarywhich, to my mind, has been exaggeratedin Kurt Weinberg's study: Henri Heine: heraut du symbolismefrangais (1954).75 E. T. A. Hoffmann,we should not forget,was widely translatedinto Frenchand could supplyoccult motifs,a transcendentalview of music and the theoryand practice of synaesthesia. Possibly even more importantwere the indirect contacts through English writers: through Carlyle's chapter on symbolismin Sartor Resartus and his essay on Novalis; through Coleridge fromwhom, Mrs. Crowe,Baudelaire drewhis definithroughanotherintermediary, tion of creative imagination; and throughEmerson,who was translated by Edgar Quinet.76 Also French thinkersof the early nineteenthcenturyknew the theory of symbolism,at least, in the wide application to all the religions of the world made by Creuzer whose Symbolikwas translatedinto 71 Stuttgart,1963In Entretienspolitiques et litte'raires, September 1891. 73 In Nouvelle Revue, 1894,and Les Disciples & Sais, suivi de Fragments (Bruxelles, 1985). The article on Novalis is included in Le Trdsor des humbles (1896).

72

74 Cf. "Richard Wagner: Reverie d'un potte frangais" (1885) in PlMiade ed., pp. 541-45. 75 New Haven, 195476 A. G. Lehmann, The SymbolistAesthetic in France, 1885-1895 (Oxford, 1950), makes good suggestions.

268

NEW LITERARY

HISTORY

French in 1825.77 Pierre Leroux used the idea of "symbolicpoetry" There was Edgar Allan Poe who prominentlyin the early thirties.78 drew on Coleridge and A. W. Schlegeland seemedso closelyto anticipate Bauderlaire's views that Bauderlaire quoted him as if he were Poe himself,sometimesdropping all quotation marks.79 The enormousinfluenceof Poe on the French demonstrates,howbetweenromanticismand symbolism. ever,most clearlythe difference Poe is far frombeing a representativeof the romanticworld view or of the romanticaestheticin which imaginationis conceived as transformingnature. Poe has been aptly described as an "angel in a machine": he combines a faith in technique and even technology,a distrustof inspiration,a rationalisticeighteenth-century mind with a in of occult distrust belief The vague "supernal" beauty.80 inspiration, the enmityto nature is the crucial point which sets symbolismfrom romanticism.Baudelaire, Mallarme, Valkryall share it; while Rilke, a symbolistin many of his proceduresand views, appears as highly romanticin his relianceon momentsof inspiration.This is whyHugo Friedrichexcludes him fromhis book on the modernlyricand even disparageshim in a harsh passage.8' This is why the attemptto make Mallarmd a spiritualdescendantof Novalis, as Vordtriedetried,must fail. Mallarmd, one mightgrant,aims at transcendencebut it is an empty transcendencewhile Novalis rapturouslyadores the unity of the mysteriousuniverse.In short,the Romantics were Rousseauists, the symbolistsbeginningwith Baudelaire believe in the fall of man or if they do not use the religious phraseology,know that man is limited and is not, as Novalis believed, the Messiah of nature. The end of the romanticperiod is clearlymarked by the victoryof positivismand scientism,whichsoon led to disillusionmentand pessimism. Most symbolistswere non-Christiansand even atheists,even if they tried to finda new religionin occultismor flirtedwith Oriental religions. They were pessimistswho need not have read Schopenhauer and Eduard von Hartmann,as Laforguedid, to succumbto the mood 77 Friedrich Creuzer, Symbolik und Mythologie der alten peared as Religions de l'antiquitd considerdes dans leurs translatedby Guigniaut in 1825. 78 See "Du style symbolique" in Le Globe, 29 March and a series of articles in Revue Encyclopdedique,1831. See my Criticism,III, 27-28.

V6lker (181o) apformes symbolistes, 8 April, 1829, and History of Modern

79 In the essay on Gautier Baudelaire reproduces "The Poetic Principle." See also Marcel Frangon, "Poe et Baudelaire" in PMLA, LX (1945), 841-598o See my chapter in History of Modern Criticism,III, 152-63. 81 Strukturder modernenLyrik,p. 116.

THE TERM AND CONCEPT

OF SYMBOLISM

269

or the death of God of decadence, fin de sikcle, G6tterddimmerung, Nietzsche.82 prophesied by Symbolismis also clearly set offfrom the new avant-gardemovements after 1915: futurism,cubism, surrealism,expressionism,etc. There the faith in language has crumbled completelywhile in Mallarme and Valery language preservesit cognitive and even magic power; Val6ry'scollectionof poems is rightlycalled Charmes.Orpheus is the mythologicalhero of the poet: charmingthe animals, treesand even stones. With more recent art the view of analogy disappears: Kafka has nothingof it. Post-symbolist art is abstractand allegorical rather than symbolic.The image, in surrealism,has no beyond: it wells, at most,fromthe subconsciousof the individual. Finally, thereis the highestabstraction,the wide largestcircle; the use of "symbolism"in all literature,of all ages. But then the term, broken loose fromits historicalmoorings,lacks concretecontentand remains merelythe name for a phenomenon almost universal in all art. These reflectionsmustlead to what only can be a recommendation, to use the third sense of our term,to call the period of European literatureroughlybetween 1885 and 1914 "symbolism,"to see it as an internationalmovementwhich radiated originallyfrom France but produced great writersand great poetry also elsewhere. In Great Britain, Yeats and Eliot; in the United States,Wallace Stevens and Hart Crane; in Germany, George, Rilke and Hofmannsthal; in Russia, Blok, Ivanov and Bely; in Spain and South America,Dario, Machado and Guill6n. If we, as we should, extend the meaning of symbolismto prose,we can see it clearlyin the late Henry James,in Joyce, the later Thomas Mann, in Proust, in the early Gide, in Faulkner,and in D. H. Lawrence; and if we add the drama,we recognize it in the later stages of Ibsen, Strindberg,Hauptmann and in O'Neill. There is symbolistcriticismof distinction:an aestheticsin Mallarm6 and Val6ry,a looser creed in Remy de Gourmont,in Eliot and in Yeats and thereis a flourishingschool of symbolistinterpretation particularlyin the United States. Much of the French "new criticism"is franklysymbolist.Roland Barthes'snew pamphlet,Critique et vdritd(1966), pleads for a completelibertyof symbolistinterpretation. Still, we mustnot forgetour initial reminder.A period conceptcan never exhaust its meaning.It is not a class concept of which the individual worksare cases. It is a regulativeidea: it struggleswith preced82 See the review of Vordtriede's Novalis by Hans Robert Jauss in Romanische Forschungen,LXXVII (1965), 174-83.

270

NEW LITERARY

HISTORY

ing and followingideals of art. In the time under considerationthe strengthof the survivals was particularlygreat: Hauptmann's Die Weber was performedin the same year (1892) as Die Bliitterfifrdie Kunst began to appear; Blok's Poems on the Beautiful Lady were writtenin the same year (1901) as Gorky'sLower Depths. Within the same author and even within the same work of art the struggle was waged at times.Edmond Jaloux called Joyce"at the same time a The same is true of Proust and Mann. realist and a symbolist."'83 combines symbolismand naturalismas no other book of the Ulysses time into a synthesisof grand proportion and strong tension. In Trieste Joyce lectured on two English writersand on two English Defoe and Blake.84 writersalone: theywere characteristically As agreementon the main periods of European literaturegrows, so agreementto add the period term"symbolism"to the fiveperiods now accepted should increase. But even if a differentterm should be victorious (thoughnone I can thinkof seems to me even remotely preferable),we should always recognizethat such a termhas fulfilled its functionas a tool of historiographyif it has made us think not only about individual worksand authorsbut about schools,trendsand movementsand their internationalexpansion. Symbolismis at least a literaryterm which will help us to counteractthe dependence of much literaryhistoryon periodization derived from political and social history(such as the term"Imperialism"used in Marxistliterary histories which is perfectlymeaningless applied to poetry at that time). Symbolismis a term (and I am quoting the wordsI applied to Baroque in 1945) "which prepares for synthesis,draws our minds away fromthe mere accumulationof observationsand facts,and paves the way for a futurehistoryof literatureas a fine art."''85 YALE

UNIVERSITY

83 Quoted by Harry Levin, James Joyce (Norfolk,Conn., 1941), p. 19: "A la fois realiste et symboliste." 84 See Richard Ellmann, James Joyce (New York, 1959), PP- 329-30. The lectures in 1912 were called "Verismo ed idealismo nella letteraturainglese." 85 See my Concepts of Criticism,p. 114.