I.
Hornberg, Black Forest, 2 August 1935
Dear Herr Benjamin:
Today let me try to say something to you at long last about your draft essay, which I have studied very thoroughly and discussed with Felizitas1 again; she fully shares the views I express here. It seems to me to be in keeping with the importance of the subject – which, as you know, I rate extremely highly – if I speak with complete candour and proceed without preliminaries to the questions which I believe are equally central for both of us. But I shall preface my critical discussion by saying that even though your method of work means that a sketch and a ‘line of thought’ cannot convey an adequate representation, your draft seems to me full of the most important ideas. Of these I should like to emphasize only the magnificent passage about living as a leaving of traces, the conclusive sentences about the collector, and the liberation of things from the curse of being useful. The outline of the chapter on Baudelaire as an interpretation of the poet and the introduction of the category of nouveauté on p. 172 also seem to me entirely convincing.2 You will therefore guess what in any case you would hardly have expected to be otherwise: that I am still concerned with the complex which may be designated by the rubrics – prehistory of the 19th century, dialectical image, and configuration of myth and modernism. If I refrain from making a distinction between the ‘material’ and the ‘epistemological’ questions, this will still be in keeping, if not with the external organization of your draft, at all events with its philosophical core, whose movement is to make the antithesis between the two disappear (as in both the more recent traditional sketches of the dialectic).
Let me take as my point of departure the motto on p. 159, Chaque époque rêve la suivante [Every epoch dreams its successor]. This seems to me an important key in that all those motifs of the theory of the dialectical image which underlie my criticism, crystallize around it as an undialectical sentence whose elimination could lead to a clarification of the theory itself. For the sentence implies three things: a conception of the dialectical image as a content of consciousness, albeit a collective one; its direct – I would almost say: developmental – relatedness to the future as Utopia; and a notion of the ‘epoch’ as proper, and self-contained subject of this content of consciousness. It seems extremely significant to me that this version of the dialectical image, which can be called an immanent one, not only threatens the original force of the concept, which was theological in nature, introducing a simplification which attacks not so much its subjective nuance as its basic truth; it also fails to preserve that social movement within the contradiction, for the sake of which you sacrifice theology.
If you transpose the dialectical image into consciousness as a ‘dream’ you not only take the magic out of the concept and render it sociable, but you also deprive it of that objective liberating power which could legitimize it in materialistic terms. The fetish character of the commodity is not a fact of consciousness; rather, it is dialectical, in the eminent sense that it produces consciousness. This means, however, that consciousness or unconsciousness cannot simply depict it as a dream, but responds to it in equal measure with desire and fear. But it is precisely this dialectical power of the fetish character that is lost in the replica realism (sit venia verbo) of your present immanent version of the dialectical image. To return to the language of the glorious first draft of your Arcades project: if the dialectical image is nothing but the way in which the fetish character is perceived in a collective consciousness, the Saint Simonian conception of the commodity world may indeed reveal itself as Utopia, but not as its reverse – namely, a dialectical image of the 19th century as Hell. But only the latter could put the idea of a Golden Age into the right perspective, and precisely this dual sense could turn out to be highly appropriate for an interpretation of Offenbach – that is, the dual sense of Underworld and Arcadia; both are explicit categories of Offenbach and could be pursued down into details of his instrumentation. Thus the abandonment of the category of Hell in your draft, and particularly the elimination of the brilliant passage about the gambler (for which the passage about speculation and games of chance is no substitute), seems to me to be not only a loss of lustre but also of dialectical consistency. Now I am the last to be unaware of the relevance of the immanence of consciousness for the 19th century. But the concept of the dialectical image cannot be derived from it; rather, the immanence of consciousness itself is, as Intérieur, the dialectical image for the 19th century as alienation. There I shall also have to leave the stake of the second chapter of my Kierkegaard book in the new game.3 Accordingly, the dialectical image should not be transferred into consciousness as a dream, but in its dialectical construction the dream should be externalized and the immanence of consciousness itself be understood as a constellation of reality – the astronomical phase, as it were, in which Hell wanders through mankind. It seems to me that only the map of such a journey through the stars could offer a clear view of history as prehistory.
Let me try to formulate the same objection again from the diametrically opposite standpoint. In keeping with an immanent version of the dialectical image (with which, to use a positive term, I would contrast your earlier conception of a model) you construe the relationship between the oldest and the newest, which was already central to your first draft, as one of Utopian reference to a ‘classless society’. Thus the archaic becomes a complementary addition to the new, instead of being the ‘newest’ itself; it is de-dialecticized. However, at the same time, and equally undialectically, the image of classlessness is put back into mythology instead of becoming truly transparent as a phantasmagoria of Hell. Therefore the category in which the archaic coalesces into the modern seems to me far less a Golden Age than a catastrophe. I once noted that the recent past always presents itself as though it has been destroyed by catastrophes. Hie et nunc I would say that it thereby presents itself as prehistory. And at this point I know I am in agreement with the boldest passage in your book on tragedy [Der Ursprung desdeutschen Trauerspiels].4
If the disenchantment of the dialectical image as a ‘dream’ psychologizes it, by the same token it falls under the spell of bourgeois psychology. For who is the subject of the dream? In the 19th century it was surely only the individual; but in the individual’s dream no direct depiction of either the fetish character or its monuments may be found. Hence the collective consciousness is invoked, but I fear that in its present form it cannot be distinguished from Jung’s conception. It is open to criticism on both sides; from the vantage point of the social process, in that it hypostasizes archaic images where dialectical images are in fact generated by the commodity character, not in an archaic collective ego, but in alienated bourgeois individuals; and from the vantage point of psychology in that, as Horkheimer puts it, a mass ego exists only in earthquakes and catastrophes, while otherwise objective surplus value prevails precisely through individual subjects and against them. The notion of collective consciousness was invented only to divert attention from true objectivity and its correlate, alienated subjectivity. It is up to us to polarize and dissolve this ‘consciousness’ dialectically between society and singularities, and not to galvanize it as an imagistic correlate of the commodity character. It should be a clear and sufficient warning that in a dreaming collective no differences remain between classes.
Lastly, moreover, the mythic-archaic category of the ‘Golden Age’ – and this is what seems socially decisive to me – has had fateful consequences for the commodity category itself. If the crucial ‘ambiguity’ of the Golden Age is suppressed (a concept which is itself greatly in need of a theory and should by no means be left unexamined), that is, its relationship to Hell, the commodity as the substance of the age becomes Hell pure and simple, yet negated in a way which would actually make the immediacy of the primal state appear as truth. Thus disenchantment of the dialectical image leads directly to purely mythical thinking, and here Klages appears as a danger,5 as Jung did earlier. Nowhere does your draft contain more remedies than at this point. Here would be the central place for the doctrine of the collector who liberates things from the curse of being useful. If I understand you correctly, this is also where Haussmann belongs; his class consciousness, precisely by a perfection of the commodity character in a Hegelian self-consciousness, inaugurates the explosion of its phantasmagoria. To understand the commodity as a dialectical image is also to see the latter as a motif of the decline and ‘supersession’ of the commodity, rather than as its mere regression to an older stage. The commodity is, on the one hand, an alienated object in which use-value perishes, and on the other, an alien survivor that outlives its own immediacy. We receive the promise of immortality in commodities and not for people. To develop the relationship between the Arcades project and the book on the Baroque, which you have rightly established, the fetish is a faithless final image, comparable only to a death’s-head. It seems to me that this is where the basic epistemological character of Kafka lies, particularly in Odradek, as a commodity that has survived to no purpose.6 In this fairy tale by Kafka surrealism may come to an end, as baroque drama did in Hamlet. But within society this means that the mere concept of use-value by no means suffices for a critique of the commodity character, but only leads back to a stage prior to the division of labour. This has always been my real reservation toward Brecht;7 his ‘collective’ and his unmediated concept of function have always been suspect to me, as themselves a ‘regression’. Perhaps you will see from these reflections, whose substance concerns precisely those categories in your draft which may conform to those of Brecht, that my opposition to them is not an insular attempt to rescue autonomous art or anything like that, but addresses itself solemnly to those motifs of our philosophical friendship which I regard as basic. If I were to close the circle of my critique with one bold grip, it would be bound to grasp the extremes. A restoration of theology, or better yet, a radicalization of the dialectic into the glowing centre of theology, would at the same time have to mean the utmost intensification of the social-dialectical, indeed economic, motifs. These, above all, must be viewed historically. The specific commodity character of the 19th century, in other words, the industrial production of commodities, would have to be worked out much more clearly and materially. After all, commodities and alienation have existed since the beginning of capitalism – i.e. the age of manufactures, which is also that of baroque art; while the ‘unity’ of the modern age has since then lain precisely in the commodity character. But the complete ‘prehistory’ and ontology of the 19th century could be established only by an exact definition of the industrial form of the commodity as one clearly distinguished historically from the older form. All references to the commodity form ‘as such’ give that prehistory a certain metaphorical character, which cannot be tolerated in this serious case. I would surmise that the greatest interpretative results will be achieved here if you unhesitatingly follow your own procedure, the blind processing of material. If, by contrast, my critique moves in a certain theoretical sphere of abstraction, that surely is a difficulty, but I know that you will not regard it as a mere problem of ‘outlook’ and thereby dismiss my reservations.
However, permit me to add a few specific remarks of a more concrete character, which will naturally be meaningful only against this theoretical background. As a title I should like to propose Paris, Capital of the Nineteenth Century, not The Capital – unless the Arcades title is revived along with Hell. The division into chapters according to men does not strike me as quite felicitous; it makes for a certain forced systemization which leaves me a little uneasy. Were there not once sections according to materials, like ‘plush’, ‘dust’, etc? The relationship between Fourier and the arcades is not very satisfactory either. Here I could imagine as a suitable pattern a constellation of the various urban and commodity materials, an arrangement later to be deciphered as both dialectical image and its theory.
In the motto on p. 157 the word portique very nicely supplies the motif of ‘antiquity’; in connection with the newest as the oldest, perhaps a morphology of the Empire should be given elementary treatment here (such as melancholy receives in the Baroque book). On p. 158, at any rate, the conception of the State in the Empire as an end in itself should be clearly shown to have been a mere ideology, which your subsequent remarks indicate that you had in mind. You have left the concept of construction completely unilluminated; as both alienation and mastery of material it is already eminently dialectical and should, in my opinion, forthwith be expounded dialectically (with a clear differentiation from the present concept of construction; the term engineer, which is very characteristic of the 19th century, probably provides a starting-point!) Incidentally, the introduction and exposition of the concept of the collective unconscious, on which I have already made some basic remarks, are not quite clear here. Regarding p. 158, I should like to ask whether cast iron really was the first artificial building material (bricks!); in general, I sometimes do not feel quite comfortable with the notion of ‘first’ in the text. Perhaps this formulation could be added: every epoch dreams that it has been destroyed by catastrophes. P. 159: The phrase ‘the new and the old are intermingled’ is highly dubious to me, given my critique of the dialectical image as regression. There is no reversion to the old, rather, the newest, as semblance and phantasmagoria, is itself the old. Here I may perhaps remind you, without being obtrusive, of some formulations, including certain remarks on ambiguity, in the Intérieur section of my work on Kierkegaard. By way of supplementing these: dialectical images are as models not social products, but objective constellations in which ‘the social’ situation represents itself. Consequently, no ideological or social ‘accomplishment’ can ever be expected of a dialectical image. My objection to your merely negative account of reification – the critique of the element of ‘Klages’ in your draft – is based primarily on the passage about machines on p. 159. An over-valuation of machine technology and machines as such has always been peculiar to bourgeois theories of retrospection; the relations of production are concealed by an abstract reference to the means of production.
The very important Hegelian concept of the second nature, which has since been taken up by Georg Lukács8 and others, belongs on p. 161f. Presumably the ‘Diable à Paris’ could lead to Hell. On p. 162, I would very much doubt that the worker appeared as a stage-extra etc, ‘for the last time’ outside his class. Incidentally, the idea of an early history of the feuilleton, about which so much is contained in your essay on Kraus, is most fascinating; this would be the place for Heine, too. In this connection an old journalistic term occurs to me: Schablonstil [cliché style], whose origin ought to be investigated. The term Lebensgefühl [attitude to life], used in cultural or intellectual history, is highly objectionable. It seems to me that your uncritical acceptance of the first appearance of technology is connected with your over-valuation of the archaic as such. I noted down this formulation: myth is not the classless longing of a true society, but the objective character of the alienated commodity itself. P. 163: Your conception of the history of painting in the 19th century as a flight from photography (to which there is an exact correspondence in the flight of music from ‘banality’) is formidable but undialectical, for the share of the forces of production not incorporated in commodity form in our store of paintings cannot be grasped concretely in this way but only in the negative of its trace (Manet is probably the source of this dialectic). This seems to be related to the mythologizing or archaizing tendency of your draft. Belonging to the past, the stock of paintings becomes, so to speak, fixed starry images in the philosophy of history, drained of their quota of productive force. The subjective side of the dialectic vanishes under an undialectically mythical glance, the glance of Medusa.
The Golden Age on p. 164 is perhaps the true transition to Hell. – I cannot see the relationship of the World Fairs to the workers; it sounds like conjecture and surely should be asserted only with extreme caution. Of course, a great definition and theory of phantasmagoria belong on p. 165f. The next page was a mene tekel [warning] to me. Felizitas and I remember the overwhelming impression which the Saturn quotation once made on us; the quotation has not survived a more sober inspection of it. The Saturn ring should not become a cast-iron balcony, but the balcony should become the real Saturn ring. Here I am happy not to offer you any abstract objections but to confront you with your own success: the incomparable moon chapter in your Kindheit whose philosophical content belongs here.9 At this point I remembered what you once said about your Arcades study: that it could be wrested away only from the realm of madness. That it has removed itself from this realm rather than subjugating it is proved by the interpretation of the Saturn quotation which bounced off it. This is the centre of my real objections … this is where I have to speak so brutally because of the enormous seriousness of the matter. As was probably your intention, the fetish conception of the commodity must be documented with the appropriate passages from the man who discovered it.
The concept of the organic, which also appears on p. 166 and points to a static anthropology, etc, is probably not tenable either, or only in the sense that it merely existed as such prior to the fetish and thus is itself historical, like the idea of ‘landscape’. The dialectical commodity motif of Odradek probably belongs on p. 166. The workers’ movement appears here somewhat like a deus ex machina again. To be sure, as with some other analogous forms, the abbreviated style of your draft may be to blame; this is a reservation that applies to many of my reservations.
A propos the passage about fashion, which seems to me very important, but in its construction should probably be detached from the concept of the organic and brought into relationship with the living, i.e. not to a superior ‘nature’: the idea of the changeant occurred to me – the shot fabric which seems to have had expressive significance for the 19th century and presumably was tied to industrial processes. Perhaps you will pursue this some day; Frau Hessel, whose [fashion] reports in the Frankfurter Zeitung we always read with great interest, will surely have some information on it. The passage where I have particular misgivings about the overly abstract use of the commodity category is to be found on p. 166; I doubt if it appeared as such ‘for the first time’ in the 19th century. (Incidentally, the same objection applies also to the Intérieur and the sociology of interiority in my Kierkegaard, and every criticism that I make of your draft also goes for my own earlier study.) I believe that the commodity category could be greatly concretized by the specifically modern categories of world trade and imperialism. Related to this is the arcade as a bazaar, also antique shops as world-trade markets for the temporal. The significance of ‘compressed distance’ lies perhaps in the problems of winning over aimless social strata and imperial conquest. I am only giving you suggestions; of course, you will be able to unearth incomparably more conclusive evidence from your material and define the specific shape of the world of things in the 19th century, perhaps viewing it from its seamy side – its refuse, remnants, debris.
The passage about the office, too, probably lacks historical exactitude. To me the office seems less a direct opposite of the home [intérieur] than a relic of older forms of rooms, probably baroque ones (cf. globes, maps on the walls, railings, and other kinds of material). Regarding the theory of Art Nouveau on p. 168: if I agree with you that it meant a decisive shattering of the interior, for me this excludes the idea that it ‘mobilizes all the reserve forces of interiority’. Rather, it seems to save and actualize them through ‘externalization’. (The theory of symbolism in particular belongs here, but above all Mallarmé’s interiors, which have exactly the opposite significance of Kierkegaard’s.) In place of interiority Art Nouveau put sex. It had recourse to sex precisely because only in sex could a private person encounter himself not as inward but as corporeal. This is true of all Art Nouveau from Ibsen to Maeterlinck and d’Annunzio. Its origin is Wagner and not the chamber music of Brahms. Concrete seems uncharacteristic of Art Nouveau; it presumably belongs in the strange vacuum around 1910. Incidentally, I think it is probable that the real Art Nouveau coincided with the great economic crisis around 1900. Concrete belongs to the pre-war boom. P. 168: Let me also draw your attention to the very remarkable interpretation of [Ibsen’s] The Master Builder in Wedekind’s posthumous works. I am not acquainted with any psychoanalytic literature about awakening, but I shall look into this. However, is not the dream-interpreting, awakening psychoanalysis which expressly and polemically dissociates itself from hypnotism (documentation in Freud’s lectures10) itself part of Art Nouveau, with which it coincides in time? This is probably a question of the first order and one that may be very far-reaching. As a corrective to my basic critique I should like to add the following here: if I reject the use of the notion of the collective consciousness, it is naturally not in order to leave the ‘bourgeois individual’ intact as the real substratum. The interior should be made transparent as a social function and its self-containedness should be revealed as an illusion – not vis-à-vis a hypostasized collective consciousness, but vis-à-vis the real social process itself. The ‘individual’ is a dialectical instrument of transition that must not be mythicized away, but can only be superseded. Once more I should like to emphasize most strongly the passage about the ‘liberation of things from the bondage of being useful’ as a brilliant turning-point for the dialectical salvation of the commodity. On p. 169 I should be pleased if the theory of the collector and of the interior as a casing were elaborated as fully as possible.
On p. 170 I should like to call your attention to Maupassant’s La Nuit, which seems to me the dialectical capstone to Poe’s Man of the Crowd as cornerstone. I find the passage about the crowd as a veil wonderful. P. 171 is the place for the critique of the dialectical image. You undoubtedly know better than I do that the theory given here does not yet do justice to the enormous demands of the subject. I should only like to say that ambiguity is not the translation of the dialectic into an image, but the ‘trace’ of that image which itself must first be dialecticized by theory. I seem to remember that there is a serviceable statement concerning this in the Interior chapter of my Kierkegaard book. Re p. 172, perhaps the last stanza of the great ‘Femmes Damnées’ from [Baudelaire’s] Pièces condamnées. In my view, the concept of false consciousness must be treated with the greatest caution and should in no case be used any longer without reference to its Hegelian(!) origin. ‘Snob’ was originally not an aesthetic concept but a social one; it was given currency by Thackeray. A very clear distinction should be made between snob and dandy; the history of the snob should be investigated, and Proust furnishes you the most splendid material for this. Your thesis on p. 172 about l’art pour l’art and the total work of art seems untenable to me in its present form. The total work of art and aestheticism in the precise sense of the word are not identical, but diametrically opposed attempts to escape from the commodity character. Thus Baudelaire’s relationship to Wagner is as dialectical as his association with a prostitute.
I am not at all satisfied with the theory of speculation on p. 174. For one thing, the theory of games of chance which was so magnificently included in the draft of the Arcades study is missing; another thing that is lacking is a real economic theory of the speculator. Speculation is the negative expression of the irrationality of capitalistic reason. Perhaps it would be possible to cope with this passage, too, by means of ‘extra polation to extremes’. An explicit theory of perspective would be indicated on p. 176; I believe there was something on that in the original draft. The stereoscope, which was invented between 1810 and 1820, is relevant here. The fine dialectical conception of the Haussmann chapter could perhaps be brought out more precisely in your study than it is in the draft, where one has to interpret it first.
I must ask you once more to excuse the carping form of these comments; but I believe I owe you at least a few specific examples of my basic criticism.
In true friendship, Yours
II.
London, 18 March 1936
Derr Herr Benjamin:
If today I prepare to convey to you some notes on your extraordinary study [‘The Work of Art in the Age of Mechanical Reproduction’], I certainly have no intention of offering you criticism or even an adequate response. The terrible pressure of work on me – the big book on logic,11 the completion of my contribution to the monograph on Berg,12 which is ready except for two analyses, and the study on jazz13 – makes any such endeavour hopeless. This is especially true of a work in the face of which I am very seriously aware of the inadequacy of written communication, for there is not a sentence which I would not wish to discuss with you in detail. I cling to the hope that this will be possible very soon, but on the other hand I do not want to wait so long before giving you some kind of response, however insufficient it may be.
Let me therefore confine myself to one main theme. My ardent interest and my complete approval attach to that aspect of your study which appears to me to carry out your original intention – the dialectical construction of the relationship between myth and history – within the intellectual field of the materialistic dialectic: namely, the dialectical self-dissolution of myth, which is here viewed as the disenchantment of art. You know that the subject of the ‘liquidation of art’ has for many years underlain my aesthetic studies and that my emphatic espousal of the primacy of technology, especially in music, must be understood strictly in this sense and in that of your second technique. It does not surprise me if we find common ground here; it does not surprise me, because in your book on the Baroque you accomplished the differentiation of the allegory from the symbol (in the new terminology, the ‘aural’ symbol) and in your Einbahnstrasse14 you differentiated the work of art from magical documentation. It is a splendid confirmation – I hope it does not sound immodest if I say: for both of us – that in an essay on Schönberg which appeared in a Festschrift two years ago15 and with which you are not familiar, I proposed formulations about technology and dialectics as well as the alteration of relationships to technology, which are in perfect accord with your own.
It is this accord which for me constitutes the criterion for the differences that I must now state, with no other aim than to serve our ‘general line’, which is now so clearly discernible. In doing so, perhaps I can start out by following our old method of immanent criticism. In your earlier writings, of which your present essay is a continuation, you differentiated the idea of the work of art as a structure from the symbol of theology and from the taboo of magic. I now find it disquieting – and here I see a sublimated remnant of certain Brechtian motifs – that you now casually transfer the concept of magical aura to the ‘autonomous work of art’ and flatly assign to the latter a counter-revolutionary function. I need not assure you that I am fully aware of the magical element in the bourgeois work of art (particularly since I constantly attempt to expose the bourgeois philosophy of idealism, which is associated with the concept of aesthetic autonomy, as mythical in the fullest sense). However, it seems to me that the centre of the autonomous work of art does not itself belong on the side of myth – excuse my topic parlance – but is inherently dialectical; within itself it juxtaposes the magical and the mark of freedom. If I remember correctly, you once said something similar in connection with Mallarmé, and I cannot express to you my feeling about your entire essay more clearly than by telling you that I constantly found myself wishing for a study of Mallarmé as a counterpoint to your essay, a study which, in my estimation, you owe us as an important contribution to our knowledge. Dialectical though your essay may be, it is not so in the case of the autonomous work of art itself; it disregards an elementary experience which becomes more evident to me every day in my own musical experience – that precisely the uttermost consistency in the pursuit of the technical laws of autonomous art changes this art and instead of rendering it into a taboo or fetish, brings it close to the state of freedom, of something that can be consciously produced and made. I know of no better materialistic programme than that statement by Mallarmé in which he defines works of literature as something not inspired but made out of words; and the greatest figures of reaction, such as Valéry and Borchardt (the latter with his essay about villas16 which, despite an unspeakable comment about workers, could be taken over in a materialistic sense in its entirety), have this explosive power in their innermost cells. If you defend the kitsch film against the ‘quality’ film, no one can be more in agreement with you than I am; but l’art pour l’art is just as much in need of a defence, and the united front which exists against it and which to my knowledge extends from Brecht to the Youth Movement, would be encouragement enough to undertake a rescue.
[In your essay on The Elective Affinities]17 you speak of play and appearance as the elements of art; but I do not see why play should be dialectical, and appearance – the appearance which you have managed to preserve in Ottilie who, together with Mignon and Helena,18 now does not come off so well – should not. And at this point, to be sure, the debate turns political quickly enough. For if you render rightly technicization and alienation dialectical, but not in equal measure the world of objectified subjectivity, the political effect is to credit the proletariat (as the cinema’s subject) directly with an achievement which, according to Lenin, it can realize only through a theory introduced by intellectuals as dialectical subjects, who themselves belong to the sphere of works of art which you have consigned to Hell.
Understand me correctly. I would not want to claim the autonomy of the work of art as a prerogative, and I agree with you that the aural element of the work of art is declining – not only because of its technical reproducibility, incidentally, but above all because of the fulfilment of its own ‘autonomous’ formal laws (this is the subject of the theory of musical reproduction which Kolisch and I have been planning for years). But the autonomy of the work of art, and therefore its material form, is not identical with the magical element in it. The reification of a great work of art is not just loss, any more than the reification of the cinema is all loss. It would be bourgeois reaction to negate the reification of the cinema in the name of the ego, and it would border on anarchism to revoke the reification of a great work of art in the spirit of immediate use-values. ‘Les extrèmes me touchent’ [Gide], just as they touch you – but only if the dialectic of the lowest has the same value as the dialectic of the highest, rather than the latter simply decaying. Both bear the stigmata of capitalism, both contain elements of change (but never, of course, the middle-term between Schönberg and the American film). Both are torn halves of an integral freedom, to which however they do not add up. It would be romantic to sacrifice one to the other, either as the bourgeois romanticism of the conservation of personality and all that stuff, or as the anarchistic romanticism of blind confidence in the spontaneous power of the proletariat in the historical process – a proletariat which is itself a product of bourgeois society.
To a certain extent I must accuse your essay of this second romanticism. You have swept art out of the corners of its taboos – but it is as though you feared a consequent inrush of barbarism (who could share your fear more than I?) and protected yourself by raising what you fear to a kind of inverse taboo. The laughter of the audience at a cinema – I discussed this with Max, and he has probably told you about it already – is anything but good and revolutionary; instead, it is full of the worst bourgeois sadism. I very much doubt the expertise of the newspaper boys who discuss sports; and despite its shock-like seduction I do not find your theory of distraction convincing – if only for the simple reason that in a communist society work will be organized in such a way that people will no longer be so tired and so stultified that they need distraction. On the other hand, certain concepts of capitalist practice, like that of the test, seem to me almost ontologically congealed and taboo-like in function – whereas if anything does have an aural character, it is surely the film which possesses it to an extreme and highly suspect degree. To select only one more small item: the idea that a reactionary is turned into a member of the avant-garde by expert knowledge of Chaplin’s films strikes me as out-and-out romanticization. For I cannot count Kracauer’s19 favourite director, even after Modern Times, as an avant-garde artist (the reason will be perfectly clear from my article on jazz), nor do I believe that any of the decent elements in this work will attract attention. One need only have heard the laughter of the audience at the film to know what is actually happening.
Your dig at Werfel gave me great pleasure. But if you take Mickey Mouse instead, things are far more complicated, and the serious question arises as to whether the reproduction of every person really constitutes that a priori of the film which you claim it to be, or whether instead this reproduction belongs precisely to that ‘naïve realism’ whose bourgeois nature we so thoroughly agreed upon in Paris. After all, it is hardly an accident if that modern art which you counterpose to technical art as aural, is of such inherently dubious quality as Vlaminck20 and Rilke. The lower sphere, to be sure, can score an easy victory over this sort of art; but if instead there were the names of, let us say, Kafka and chönberg, the problem would be posed very differently. Certainly Schönberg’s music is not aural.
Accordingly, what I would postulate is more dialectics. On the one hand, dialectical penetration of the ‘autonomous’ work of art which is transcended by its own technology into a planned work; on the other, an even stronger dialecticization of utilitarian art in its negativity, which you certainly do not fail to note but which you designate by relatively abstract categories like ‘film capital’, without tracking it down to its ultimate lair as immanent irrationality. When I spent a day in the studios of Neubabelsberg two years ago, what impressed me most was how little montage and all the advanced techniques that you emphasize are actually used; rather, reality is everywhere constructed with an infantile mimetism and then ‘photographed’. You under-estimate the technicality of autonomous art and over-estimate that of dependent art; this, in plain terms, would be my main objection. But this objection could only be given effect as a dialectic between extremes which you tear apart. In my estimation, this would involve nothing less than the complete liquidation of the Brechtian motifs which have already undergone an extensive transformation in your study – above all, the liquidation of any appeal to the immediacy of interconnected aesthetic effects, however fashioned, and to the actual consciousness of actual workers who have absolutely no advantage over the bourgeois except their interest in the revolution, but otherwise bear all the marks of mutilation of the typical bourgeois character. This prescribes our function for us clearly enough – which I certainly do not mean in the sense of an activist conception of ‘intellectuals’. But it cannot mean either that we may only escape the old taboos by entering into new ones – ‘tests’, so to speak. The goal of the revolution is the abolition of fear. Therefore we need have no fear of it, nor need we ontologize our fear. It is not bourgeois idealism if, in full knowledge and without mental prohibitions, we maintain our solidarity with the proletariat instead of making of our own necessity a virtue of the proletariat, as we are always tempted to do – the proletariat which itself experiences the same necessity and needs us for knowledge as much as we need the proletariat to make the revolution. I am convinced that the further development of the aesthetic debate which you have so magnificently inaugurated, depends essentially on a true accounting of the relationship of the intellectuals to the working-class.
Excuse the haste of these notes. All this could be seriously settled only on the basis of the details in which the Good Lord – possibly not magical after all – dwells.* Only the shortage of time leads me to use the large categories which you have taught me strictly to avoid. In order at least to indicate to you the concrete passages to which I refer, I have left my spontaneous pencilled annotations on the manuscript, though some of them may be too spontaneous to be communicated. I beg your indulgence for this as well as for the sketchy nature of my letter.
I am going to Germany on Sunday. It is possible that I shall be able to complete my jazz study there, something that I unfortunately did not have time to do in London. In that case I would send it to you without a covering letter and ask you to send it on to Max immediately after reading it (it probably will amount to no more than 25 printed pages). This is not certain, because I do not know whether I shall find the time or, especially, whether the nature of this study will permit me to send it from Germany without considerable danger. Max has probably told you that the idea of the clown is its focal point. I would be very pleased if it appeared together with your study. Its subject is a very modest one, but it probably converges with yours in its decisive points, and will attempt to express positively some of the things that I have formulated negatively today. It arrives at a complete verdict on jazz, in particular by revealing its ‘progressive’ elements (semblance of montage, collective work, primacy of reproduction over production) as façades of something that is in truth quite reactionary. I believe that I have succeeded in really decoding jazz and defining its social function. Max was quite taken with my study, and I could well imagine that you will be, too, Indeed I feel that our theoretical disagreement is not really a discord between us but rather, that it is my task to hold your arm steady until the sun of Brecht has once more sunk into exotic waters. Please understand my criticisms only in this spirit.
I cannot conclude, however, without telling you that your few sentences about the disintegration of the proletariat as ‘masses’ through revolution21 are among the profoundest and most powerful statements of political theory that I have encountered since I read State and Revolution.
Your old friend,
Teddie Wiesengrund*
I should also like to express my special agreement with your theory of Dadaism. It fits into the essay as nicely as the ‘bombast’ and the ‘horrors’ fit into your Baroque book.
III.
New York, 10 November 1938
Dear Walter:
The tardiness of this letter levels a menacing charge against me and all of us. But perhaps this accusation already contains a grain of defence. For it is almost self-evident that a full month’s delay in my response to your Baudelaire cannot be due to negligence.
The reasons are entirely objective in nature. They involve the attitude of all of us to the manuscript, and, considering my special interest in the question of the Arcades study, I can probably say without immodesty, my attitude in particular. I had been looking forward to the arrival of the Baudelaire with the greatest eagerness and literally devoured it. I am full of admiration for the fact that you were able to complete it by the appointed time, and it is this admiration which makes it particularly hard for me to speak of what has come between my passionate expectation and the text itself.
Your idea of providing in the Baudelaire a model for the Arcades study was something I took very seriously, and I approached the satanic scene much as Faust approached the phantasmagoria of the Brocken mountain when he thought that many a riddle would now be solved. May I be excused for having had to give myself Mephistopheles’ reply that many a riddle poses itself anew? Can you understand that reading your treatise, one of whose chapters is entitled The Flâneur and another Modernism, produced a certain disappointment in me?
The basic reason for this disappointment is that those parts of the study with which I am familiar do not constitute a model for the Arcades project so much as a prelude to it. Motifs are assembled but not elaborated. In your covering letter to Max [Horkheimer] you represented this as your express intention, and I am aware of the ascetic discipline which you impose on yourself to omit everywhere the conclusive theoretical answers to questions, and even make the questions themselves apparent only to initiates. But I wonder whether such an asceticism can be sustained in the face of such a subject and in a context which makes such powerful inner demands. As a faithful reader of your writings I know very well that in your work there is no lack of precedents for your procedure. I remember, for example, your essays on Proust and on Surrealism which appeared in Die literarische Welt. But can this method be applied to the complex of the Arcades? Panorama and ‘traces’, flâneur and arcades, modernism and the unchanging, without a theoretical interpretation – is this a ‘material’ which can patiently await interpretation without being consumed by its own aura? Rather, if the pragmatic content of these topics is isolated, does it not conspire in almost demonic fashion against the possibility of its own interpretation? In one of our unforgettable conversations in Königstein, you said that each idea in the Arcades had to be wrested away from a realm in which madness reigns. I wonder whether such ideas need to be as immured behind impenetrable layers of material as your ascetic discipline demands. In your present study the arcades are introduced with a reference to the narrowness of the pavements which impede the flâneur on the streets.22 This pragmatic introduction, it seems to me, prejudices the objectivity of phantasmagoria – something that I so stubbornly insisted upon even at the time of our Hornberg correspondence – as much as does the disposition of the first chapter to reduce phantasmagoria to types of behaviour of the literary bohème. You need not fear that I shall suggest that in your study phantasmagoria should survive unmediated or that the study itself should assume a phantasmagoric character. But the liquidation of phantasmagoria can only be accomplished with true profundity if they are treated as an objective historico-philosophical category and not as a ‘vision’ of social characters. It is precisely at this point that your conception differs from all other approaches to the 19th century. But the redemption of your postulate cannot be postponed for ever, or ‘prepared’ by a more harmless presentation of the matters in question. This is my objection. If in the third part, to use the old formulation, prehistory in the 19th century takes the place of the prehistory of the 19th century – most clearly in Péguy’s statement about Victor Hugo23 – this is only another way of stating the same point.
But it seems to me that my objection by no means concerns only the questionable procedure of ‘abstention’ in a subject which is transported by ascetic refusal of interpretation towards a realm to which asceticism is opposed: the realm where history and magic oscillate. Rather, I see a close connection between the points at which your essay falls behind its own a priori, and its relationship to dialectical materialism – and here in particular I speak not only for myself but equally for Max, with whom I have had an exhaustive discussion of this question. Let me express myself in as simple and Hegelian a manner as possible. Unless I am very much mistaken, your dialectic lacks one thing: mediation. Throughout your text there is a tendency to relate the pragmatic contents of Baudelaire’s work directly to adjacent features in the social history of his time, preferably economic features. I have in mind the passage about the duty on wine, certain statements about the barricades,24 or the above-mentioned passage about the arcades,25 which I find particularly problematic, for this is where the transition from a general theoretical discussion of physiologies to the ‘concrete’ representation of the flâneur is especially precarious.
I feel this artificiality wherever you put things in metaphorical rather than categorical terms. A case in point is the passage about the transformation of the city into an intérieur for the flâneur ;26 there one of the most powerful ideas in your study seems to me to be presented as a mere as-if. There is a very close connection between such materialistic excursions, in which one never quite loses the apprehension that one feels for a swimmer who, covered with goose pimples, plunges into cold water, and the appeal to concrete modes of behaviour like that of the flâneur, or the subsequent passage about the relationship between seeing and hearing in the city, which not entirely by accident uses a quotation from Simmel.27 I am not entirely happy with all this. You need not fear that I shall take this opportunity to mount my hobby-horse. I shall content myself with serving it, in passing, a lump of sugar, and for the rest I shall try to give you the theoretical grounds for my aversion to that particular type of concreteness and its behaviouristic overtones. The reason is that I regard it as methodologically unfortunate to give conspicuous individual features from the realm of the superstructure a ‘materialistic’ turn by relating them immediately and perhaps even causally to corresponding features of the infrastructure. Materialist determination of cultural traits is only possible if it is mediated through the total social process.
Even though Baudelaire’s wine poems may have been motivated by the wine duty and the town gates, the recurrence of these motifs in his work can only be explained by the overall social and economic tendency of the age – that is, in keeping with your formulation of the problem sensu strictissimo, by analysis of the commodity form in Baudelaire’s epoch. No one is more familiar with the difficulties this involves than I am; the phantasmagoria chapter in my Wagner28 certainly has not settled these problems as yet. Your Arcades study in its definitive form will not be able to shirk the same obligation. The direct inference from the duty on wine to L’Ame du Vin imputes to phenomena precisely that kind of spontaneity, palpability and density which they have lost in capitalism. In this sort of immediate – I would almost say again, anthropological – materialism, there is a profoundly romantic element, and the more crassly and roughly you confront the Baudelairean world of forms with the necessities of life, the more clearly I detect it. The ‘mediation’ which I miss and find obscured by materialistic-historiographic invocation, is nothing other than the theory which your study omits. The omission of the theory affects your empirical evidence itself. On the one hand, it lends it a deceptively epic character, and on the other it deprives the phenomena, which are experienced only subjectively, of their real historico-philosophical weight. To express it another way: the theological motif of calling things by their names tends to turn into a wide-eyed presentation of mere facts. If one wished to put it very drastically, one could say that your study is located at the crossroads of magic and positivism. That spot is bewitched. Only theory could break the spell – your own resolute, salutarily speculative theory. It is the claim of this theory alone that I am bringing against you.
Forgive me if this brings me to a subject which is bound to be of particular concern to me since my experiences with the Wagner study. I am referring to the ragpicker. It seems to me that his destiny as the figure of the lower limits of poverty is certainly not brought out by the way the word ragpicker appears in your study.29 It contains none of the dog-like cringing, nothing of the sack on his back or the voice which, for instance, in Charpentier’s Louise provides, as it were, the source of black light for an entire opera. There is nothing in it of the comet’s tail of jeering children behind the old man. If I may venture into the region of the arcades once more: in the figure of the ragpicker the retreat of cloaca and catacomb should have been decoded theoretically. But I wonder whether I exaggerate in assuming that your failure to do so is related to the fact that the capitalist function of the ragpicker – namely, to subject even rubbish to exchange value – is not articulated. At this point the asceticism of your study takes on features which would be worthy of Savonarola. For the return of the ragpicker in the Baudelaire quotation in the third section comes very close to this question.30 What it must have cost you not to close the gap completely!
This, I think, brings me to the centre of my criticism. The impression which your entire study conveys – and not only on me and my arcades orthodoxy – is that you have done violence to yourself. Your solidarity with the Institute [of Social Research], which pleases no one more than myself, has induced you to pay tributes to Marxism which are not really suited either to Marxism or to yourself. They are not suited to Marxism because the mediation through the total social process is missing, and you superstitiously attribute to material enumeration a power of illumination which is never kept for a pragmatic reference but only for theoretical construction. They do not suit your own individual nature because you have denied yourself your boldest and most fruitful ideas in a kind of pre-censorship according to materialist categories (which by no means coincide with the Marxist categories), even though it may be merely in the form of the above-mentioned postponement. I speak not only for myself, who am not qualified, but equally for Horkheimer and the others when I tell you that all of us are convinced that it would not only be beneficial to ‘your’ production if you elaborated your ideas without such considerations (in San Remo you raised counter-objections to this objection, and I am taking these very seriously), but that it would also be most helpful to the cause of dialectical materialism and the theoretical interests represented by the Institute, if you surrendered to your specific insights and conclusions without adding to them ingredients which you obviously find so distasteful to swallow that I cannot really regard them as beneficial. God knows, there is only one truth, and if your intelligence lays hold of this one truth in categories which on the basis of your idea of materialism may seem apocryphal to you, you will capture more of this one truth than if you use intellectual tools whose movements your hand resists at every turn. After all, there is more about this truth in Nietzsche’s Genealogy of Morals than in Bukharin’s ABC of Communism. I am confident that the thesis I am arguing cannot be suspected of laxity and eclecticism. Your study of Goethe’s Elective Affinities and your Baroque book are better Marxism than the wine duty and the deduction of phantasmagoria from the behaviour of the feuilletonists. You may be confident that we are ready to make the most extreme experiments of your theory our own. But we are equally confident that you will actually make these experiments. Gretel once said in jest that you are an inhabitant of the cave-like depths of your Arcades and that you shrink from finishing your study because you are afraid of having to leave what you have built. Let us encourage you to give us access to the holy of holies. I believe you need not be concerned with either the stability of the structure or its profanation.
As regards the fate of your study, a rather strange situation has developed, in which I have had to act much like the singer of the song ‘It is done to the sound of a muffled drum’.* Publication in the current issues of our periodical proved impossible because the weeks of discussion of your study would have caused an intolerable delay in our printing schedule. There was a plan to print the second chapter in extenso and the third in part; Leo Löwenthal urged that this be done. I myself am definitely opposed to it – not for editorial reasons, but for your own sake and for the sake of Baudelaire. This study does not represent you as it, of all your writings, must represent you. But since I am of the firm and unalterable conviction that it will be possible for you to produce a Baudelaire manuscript of full impact, I should like to entreat you to forgo the publication of the present version and to write that other version. Whether the latter would have to possess a new formal structure or could be essentially identical with the still unwritten final part of your book on Baudelaire, I cannot surmise. You alone can decide this. I should like to make it plain that this is a request on my part and not an editorial decision or a rejection.
…
Let me close with some epilegomena to the Baudelaire. First a stanza from the second Mazeppa poem of Victor Hugo (the man who is supposed to see all these things is Mazeppa, tied to the back of the horse):
Les six lunes d’Herschel, l’anneau du vieux Saturne,
Le pôle, arrondissant une aurore nocturne
Sur son front boréal,Il voit tout; et pour lui ton vol, que rien ne lasse,
De ce monde sans borne à chaque instant déplace
L’horizon idéal.
Also, the tendency toward ‘unqualified statements’ which you observe, citing Balzac and the description of the employees in ‘The Man of the Crowd’,31 applies, astonishingly enough, to Sade as well. One of the first tormentors of Justine, a banker, is described as follows: ‘Monsieur Dubourg, gros, court, et insolent comme tous les financiers’. The motif of the unknown beloved appears in rudimentary form in Hebbel’s poem about an unknown woman which contains these memorable lines: Und kann ich Form Dir und Gestalt nicht geben, So reisst auch keine Form Dich in die Gruft [And even if I cannot give you form and shape, no form will thrust you into the grave].
Finally, a few sentences from the Herbst-Blumine of Jean Paul which is a real trouvaille [find]: ‘The day received one single sun, but the night received a thousand suns, and the endless blue sea of the ether seems to be sinking down to us in a drizzle of light How many street lamps shimmer up and down the whole long Milky Way! These are lit, too, even though it is summer or the moon is shining. Meanwhile, the night does not merely adorn itself with the cloak full of stars which the ancients depicted it as wearing and which I shall more tastefully call its religious vestments rather than its ducal robe; it carries its beautification much farther and imitates the ladies of Spain. They replace the jewels in their head-dress with glow-worms in the darkness, and like them the night studs the lower part of its cloak, where there are no glittering stars, with such little animals, and often the children take them off.’ The following sentences from a quite different piece in the same collection seem to me to belong in the same context:
‘And more of the same; for I noticed not only that Italy was a moonlit Eden to us poor drift-ice people, because daily or nightly we encountered there the living fulfilment of the universal adolescent dream of nights spent wandering and singing, but I also asked why people merely walked around and sang in the streets at night like peevish nightwatchmen, instead of whole evening-star and morning-star parties assembling and in a colourful procession (for every soul was in love) roaming through the most magnificent leafy woods and the brightly moonlit flowery meadows, and adding two more phrases on the flute to the joyful harmony – namely, the double-ended extension of the brief night by a sunrise and a sunset plus the added dawn and dusk.’ The idea that the longing which draws one to Italy is a longing for a country where one does not need to sleep is profoundly related to the later image of the roofed-over city. But the light which rests equally on the two images is, I think, none other than the light of the gas lamp, with which Jean Paul was not acquainted.
Tout entier Yours