Paul Klee: The Visible and the Legible

By Annie Bourneuf

The fact that Paul Klee (1879–1940) consistently intertwined the visual and the verbal in his art has long fascinated commentators from Walter Benjamin to Michel Foucault. However, the questions it prompts have never been satisfactorily answered—until now. In Paul Klee, Annie Bourneuf offers the first full account of the interplay between the visible and the legible in Klee’s works from the 1910s and 1920s.

Bourneuf argues that Klee joined these elements to invite a manner of viewing that would unfold in time, a process analogous to reading. From his elaborate titles to the small scale he favored to his metaphoric play with materials, Klee created forms that hover between the pictorial and the written. Through his unique approach, he subverted forms of modernist painting that were generally seen to threaten slow, contemplative viewing. Tracing the fraught relations among seeing, reading, and imagining in the early twentieth century, Bourneuf shows how Klee reconceptualized abstraction at a key moment in its development.

• Language: English
• Publisher: University of Chicago Press
• Release date: Jul 20, 2015
• ISBN: 9780226233604

Book preview

Paul Klee

Paul Klee

The Visible and the Legible

Annie Bourneuf

The University of Chicago Press

Chicago and London

Annie Bourneuf is assistant professor of art history at the School of the Art Institute of Chicago.

The University of Chicago Press, Chicago 60637

The University of Chicago Press, Ltd., London

© 2015 by The University of Chicago

All rights reserved. Published 2015.

Printed in China

24 23 22 21 20 19 18 17 16 15 1 2 3 4 5

ISBN-13: 978-0-226-09118-1 (cloth)

ISBN-13: 978-0-226-23360-4 (e-book)

DOI: 10.7208/chicago/9780226233604.001.0001

This publication is made possible in part by a grant from the Barr Ferree Foundation Fund for Publications, Department of Art and Archaeology, Princeton University.

Library of Congress Cataloging-in-Publication Data

Bourneuf, Annie, author.

Paul Klee : the visible and the legible / Annie Bourneuf.

pages cm

Includes bibliographical references and index.

ISBN 978-0-226-09118-1 (hardcover : alk. paper) — ISBN 978-0-226-23360-4 (e-book) 1. Klee, Paul, 1879–1940. 2. Art, Modern. I. Title.

N6888.K55B68 2015

740.92—dc23

2014026786

This book has been printed on acid-free paper.

For Ben

Contents

Acknowledgments

A Note on Klee’s Sequential Numbering System

Introduction

1 The Painter-Draftsman

2 Seeing and Speculating

3 A Refuge for Script

Epilogue: Old Sound

Notes

Index

Acknowledgments

My first and greatest debt of gratitude I owe to Brigid Doherty for her support, her criticism, and her example. An incomparable mentor, her thinking has inspired and guided me from the first inklings I had of this project. Her questions and ideas have improved this book at every turn.

Hal Foster shaped the aims of this book, and his encouragement and wisdom have sustained it. The incisive and detailed comments of Yve-Alain Bois made me see more clearly the book’s weaknesses and strengths. Bridget Alsdorf helped me to understand how this project’s metamorphosis into a book might be accomplished. I owe special thanks to Charles W. Haxthausen, whose thinking on Klee has had a profound influence on this book and who could not have been a more welcoming and generous guide both to Klee and to Klee studies.

I am grateful to Carol Armstrong for her criticism in the first stage of this project, and to Michael W. Jennings for his encouragement and support. The book has benefited enormously from discussions over the years with Saul Anton, Gordon Hughes, and Joyce Tsai. And I am grateful as well to all who have read and criticized portions of my study, including Maureen Chun, Noam Elcott, Alex Kitnick, Jennifer King, and Michele Matteini. At the University of Chicago Press, I thank the two anonymous reviewers whose thoughtful and detailed comments much improved the book, and my editor Susan Bielstein, who steered the project to completion with the greatest address.

This study could never have been written without the generous support of the Fulbright Program, the Dedalus Foundation, and the Department of Art and Archaeology at Princeton University. I am grateful for the Felix Gilbert Membership that allowed me to bring the book to completion at the School of Historical Studies at the Institute for Advanced Study—the ideal environment, both stimulating and serene. The Faculty Enrichment Grant I received from the School of the Art Institute of Chicago, as well as the Jane Faggen Prize from Princeton’s Department of Art and Archaeology, were also most helpful. The months I spent at the Zentrum Paul Klee in Bern were indispensable, and I am very grateful for the expert advice I received there from Christine Hopfengart, Fabienne Eggelhöfer, Patrizia Zeppetella, Myriam Weber, and Osamu Okuda, as well as for the kind help, during and after my visit, of Eva Wiederkehr Sladeczek, Edith Heinimann, and Heidi Frautschi. I thank Oskar Bätschmann for his hospitality.

I owe thanks as well to Claudine Metzger of the Kunstmuseum Bern; Christine Ramseyer of the Kunstmuseum Basel; Julia Friedrich of the Museum Ludwig, Cologne; John Prochilo of the Museum of Modern Art, New York; Mark Pascale of the Art Institute of Chicago; and the Galerie Jan Krugier, Geneva, for generously allowing me to spend time with works in their collections. In the search for illustrations, I thank Anthony Burton of the University of Chicago Press for his patience and helpfulness, as well as Jane Sykora of the Institute for Advanced Study, for her assistance. My translations benefited a great deal from the expert advice of Laura Frahm and Jens Klenner.

Portions of this book first appeared in altered form in Bauhaus Construct: Fashioning Identity, Discourse, and Modernism, ed. Robin Schuldenfrei and Jeffrey Saletnik (London: Routledge, 2009), 105–24, and in Walter Benjamin’s Media Tactics: Optics, Perception, and the Work of Art, ed. Michael W. Jennings and Tobias Wilke, special issue, Grey Room 39 (Spring 2010): 75–93. I would like to thank the editors of these publications.

I am grateful as well to the people and the institutions that allowed me to try out portions of this project in public—Charlotte Schoell-Glass and the Kunstgeschichtliche Seminar at the Universität Hamburg; Marsha Morton, Steven Mansbach, and the Historians of German and Central European Art and Architecture; Robin Schuldenfrei, Jeffrey Saletnik, and Peter Nisbet; Tobias Wilke and the Department of German at Princeton; and Beatriz Colomina and the Program in Media and Modernity at Princeton—and for the questions I received on these occasions. Megan Luke’s advice was invaluable. I would also like to thank Diane Schulte for so cheerfully smoothing out all manner of organizational complexities. I thank Maureen Chun, Lisa Lee, and Kate Nesin for their comradeship during the closing phases of writing, and Catharine Diehl for hers at the beginning.

I am thankful as well for my colleagues and my students at the School of the Art Institute of Chicago, who have aided this project in so many ways—with encouragement, advice, and tough questions of the best kind.

I thank my parents for their love and support these long years. I should probably never have written a book on Klee were it not for my father’s interest in him and the presence of Will Grohmann’s 1954 book in our home library. I would like to thank my father as well, now in his capacity as reference librarian, for crucial help in this project’s last stages. And I thank Ben Lytal for having worked on every paragraph that follows (as well as on all the book’s invisible reinforcements), for setting me an example by his own work, and for his sustaining faith in this project and in me—this book is dedicated to him.

A Note on Klee’s Sequential Numbering System

The captions for most of Paul Klee’s works reproduced in this book include a second number after the date; at times I use this number in the text as well. The second number is part of Klee’s own system for ordering his work: it corresponds to the sequence in which he entered the object into the oeuvre-catalogue he used to keep track of his production.

For instance, the caption for Klee’s Ship-star-festival (fig. 2.16) tells us that this was the sixty-second work that he entered into his oeuvre-catalogue in 1916. In this case, as in many others, Klee inscribed both the date and the second number on the work itself (here, in ink on the lower left corner of the cardboard mount).

Introduction

Paul Klee’s Landscape with Gallows (Landschaft m. d. Galgen) of 1919 consists of a painted ground of red, yellow, and blue on gessoed gauze, traversed by thin pen lines and dotted with figures drawn in black paint (fig. 0.1). The title encourages us to pick out the form of the reversed and inverted L centered at the top as the gallows; just one corner of this gallows touches the pen outline of the hill, which is marked by a cross. Klee’s painting has been compared to Pieter Bruegel’s Magpie on the Gallows (1568; fig. 0.2)¹—and indeed, both pictures present us with a gallows hill with a cross at the center and a landscape around it, swarming with detail and incident, or at least their indications.

Figure 0.1 Paul Klee, Landscape with Gallows (Landschaft m. d. Galgen), 1919, 115. Oil and pen on primed gauze on cardboard, 36 × 46 cm. The Collection of David M. Solinger. © 2014 Artists Rights Society (ARS), New York.

But Klee’s painting seems to reverse the spatial order of Bruegel’s. In Bruegel’s, we ourselves seem to hover above the hilltop mound on which the gallows stand; two men, standing just a few steps below that mound, surveying the panoramic river landscape below them, figure our viewing within the painting. In Klee’s, the gallows hill appears far in the background—or, in any case, near the top of the picture.

But many of Klee’s other small schematic signs—these trees and buildings dotted here and there in the hilly Landscape—do seem to correspond spatially to the things of Bruegel’s landscape. The largest and most elaborate of Klee’s buildings, at the upper left, might be matched with Bruegel’s castle on a promontory in the corresponding position. So, too, might Klee’s boatlike hieroglyph near the center of his picture be matched with Bruegel’s boat navigating a bend in the river, and perhaps Klee’s spiral toward the bottom right with Bruegel’s water mill.

Figure 0.2 Pieter Bruegel the Elder, Magpie on the Gallows, 1568. Oil on panel, 45.9 cm × 50.8 cm. Hessisches Landesmuseum, Darmstadt, Germany.

We might even start to play this same game of matching with the letters that Klee places in his landscape like so many trees. Might that prominent E stand for Elster (magpie—the German title of Bruegel’s painting is Die Elster auf dem Galgen)? The dot or period that follows the letter would indicate that it is indeed an abbreviation. Perhaps the T at the left could stand for Tanz (dance), for the peasants’ dance at the left of Bruegel’s painting, another important element that nothing but a letter seems to represent in Klee’s.

Klee’s partial pattern of correspondences not only involves the viewer in this game of matching but raises questions of what this game is about. Does Klee’s painting seek to transcribe, as it were, the elements and arrangement of Bruegel’s painting, to translate them into another pictorial idiom that appears to be in part scriptural, ranging from what look like rudimentary pictographs, such as the trees, to the letters of the Latin script? And is painting even the right word for Klee’s transcription or translation? It’s made of paint, to be sure: Oil on stretched canvas primed with gesso and Krems white, noted Klee in the catalogue he kept of his work.² But the discontinuity between the linear black signs and letters, on the one hand, and the colorful field they inscribe, on the other, is so emphatic that one might prefer to call it a drawing on a painted ground.³ And why is it that Klee has positioned so many of his hieroglyphs and letters to parallel the placement of the ancillary incidents of Bruegel’s painting and yet inverted the position of the gallows hill, which, along with the title, invites the comparison in the first place? Are the rules governing how we are to understand the spatial positioning of the depicted things different in Klee’s Landscape? Whereas Bruegel’s painting unfurls up into the distance from the brambles and earth of the hilltop at the painting’s bottom edge, Klee’s placement of the gallows at the top might suggest that his picture is to be viewed instead from top to bottom, as one reads a page of text. The ladderlike paths Klee draws here and there seem to refer to the spatial order of Bruegel’s painting, the movement up the panel into the distance to examine this or that interesting feature of the landscape—but perhaps the task of referring to that upward recession into distance falls to them alone. And if Klee’s Landscape seems to bring writing and picture-making into close relation to each other through this translation, is this to be understood as responding to a closeness already obtaining in pictures in general? Klee did assert that writing and pictures are fundamentally one.⁴ Or perhaps in Bruegel’s painting in particular, which engages in its own form of play with the possibility of translating the visual into the textual in its rebus-like arrangement of vivid depictions of Flemish figures of speech (to dance under the gallows, to shit on the gallows) that seem bizarre and nonsensical unless they are translated back into words? Bruegel’s titular magpie is itself not only a figure of speech but also a figure for speech—for excessive gossip or chatter.⁵ What can it mean for Klee to transcribe this painted figure of speech as a laconic E.?

Versions of the questions this picture raises about the relations among writing, drawing, and painting recur repeatedly in Klee’s art, and have often attracted commentators’ attention. Michel Foucault, for instance, asserted that Klee annulled one of the two defining principles of Western painting since the Renaissance—the separation between plastic representation and linguistic reference—by showing the juxtaposition of shapes and the syntax of lines in an uncertain reversible, floating space (simultaneously page and canvas, plane and volume, map and chronicle). Foucault continues to describe Klee’s art as follows: Boats, houses, persons are at the same time recognizable figures and elements of writing. They are placed and travel upon roads or canals that are also lines to be read. . . . The gaze encounters words as if they had strayed to the heart of things, words indicating the way to go and naming the landscape being crossed.⁶ In Klee’s work, writes Foucault, we see the intersection, within the same medium, of representation by resemblance and of representation by signs. Which presupposes that they meet in quite another space than that of painting.⁷

Clement Greenberg picks up on some of the same features of Klee’s work as Foucault does, but to tell a different story. In a fascinating essay of 1950, Greenberg argues that these crossings between the visible and the legible in Klee’s work are part of the struggle to save easel painting.⁸ What seems to interest Greenberg most is how the problematic relations between ‘literature’ and the ‘purely’ plastic come to a head in Klee’s work (54). He insists that Klee’s art results from a complex operation in which literary means are used to non-literary ends; Klee, according to Greenberg, required something resistant around which to deposit form, something more resistant—in Klee’s case—than just the medium, and the literary filled this requirement, as an impulse to formal invention and a source of it (55–57). The formal innovations of Klee’s art develop around his parodying of ‘literary’ and pictorial art in general:

No longer taking the pictorial for granted but seeing it as one cultural convention among others, Klee isolates its distinguishing properties in order to burlesque them. . . . The pictorial in Klee’s notion of it comprises every system for making marks on a surface that mankind has ever used for the purpose of communication: ideographs, diagrams, hieroglyphs, alphabets, handwriting, blueprints, musical notation, charts, maps, tables, etc., etc.  .  .  . the parody of the pictorial is but a core around which he wraps layer on layer of a parody that aims at all commonly held verities, all current sentiments, messages, attitudes, convictions, methods, procedures, formalities, etc., etc. (58–59)

This parody of the conventions of the pictorial, seen as including all other mark-making systems, is the main means of Klee’s defining irony—which is total, but not nihilistic, Greenberg remarks (59).

These striking features of Klee’s work that Foucault, Greenberg, and many others have picked up on are also the subject of my own investigation. How are we to understand Landscape with Gallows, for instance? Is it, as Foucault’s interpretation of Klee might suggest, an abrogation of a basic principle of painting since the Renaissance that displays its own historical awareness of what it is doing by referring to a sixteenth-century panel painting that itself plays with the boundaries between discourse and figures? Does it, like Klee in Greenberg’s reading, parody Bruegel and the pictorial as such as a spur to formal invention? According to Theodor W. Adorno, the manner in which Klee’s art resembles writing makes possible the recognition that all artworks are writing, not just those that are obviously such; they are hieroglyphs for which the code has been lost, a loss that plays into their content; Klee’s art is for him a beautiful example of how, more generally, some modern works can, in their own enigmaticalness, destroy the seeming comprehensibility of works sanctioned by tradition and public opinion. Might we see Klee’s Landscape as doing just this—revealing not only itself, but also Bruegel’s panel, as a picture puzzle with no solution?⁹ Or could it be, as Hubert Damisch has said of another work by Klee, the product of Klee’s laboring within the framework of signs and as close to them as possible, a project that opens onto a structural, and still more profoundly ‘scriptural’ problematic of abstraction in particular?¹⁰ Does it represent, as Rainer Crone’s interpretation of Klee would suggest, an investigation into how the Bruegel panel (or painting in general) signifies, parallel to the investigations of structuralist linguistics?¹¹ Is it a modernist revision of the Renaissance doctrine of ut pictura poesis (as is painting so is poetry), likening abstract painting to poetry as two kindred ways of arranging arbitrary signs to communicate ideas as well as appearances, as K. Porter Aichele claims?¹²

I will now step back from Landscape with Gallows and try to rephrase in more general terms the intertwined problems that this study investigates. The pursuit of analogies between writing and images, the solicitation of a reading-like mode of viewing, and the emphasis on often elaborately allusive titles are distinguishing features of much of Klee’s work. It seems that those relations between pictures and writing entailed by the visual appearance of the graphic inscription of language are central to Klee’s art. And it seems that relations between pictures and literature—in the sense that art criticism since the mid-nineteenth century has used this word to establish an opposition between the ‘purely’ plastic (to borrow Greenberg’s phrase), on the one hand, and elements of narrative, anecdote, or allegory, on the other—are central as well. One way or another, these are important issues for a great deal of modernist art and criticism, but they seem particularly pressing and pervasive for Klee. As Charles W. Haxthausen and Marcel Franciscono have written, Klee struggled in the first decade of the twentieth century with the opposition, as formulated in contemporary German art criticism, between poetic painting and what Klee termed, in his own skeptical quotation marks in a journal entry of 1905, ‘pure’ plastic art.¹³ During these early years of his career, Klee often wrote in his journal and letters about his often contradictory and unresolved attempts to come to terms with both the formalist aesthetics of ‘pure’ plastic art and what he saw as his own fundamentally poetic disposition.¹⁴ Later, he would place the relations of seeing to reading and of pictures to literature at the core of his art.¹⁵ This book seeks to explain how and why he did so.

In recent decades, much of the art-historical literature on Klee has been oriented so strongly around Otto Karl Werckmeister’s project of ideology critique (in part a reaction to the posthumous presentation of Klee in the Federal Republic of Germany in the 1950s as the exemplary prewar artist for the postwar period) that interpretive questions get short shrift.¹⁶ Werckmeister’s study The Making of Paul Klee’s Career, 1914–1920 is the most thorough and historically grounded to date, and I have drawn on it a great deal.¹⁷ However, in his monograph, which he presents as a critical challenge to the myth of [Klee’s] art, Werckmeister does not ask the questions I ask here and indeed could not do so within the framework of his premises. For Werckmeister constructs a stark alternative between two approaches to Klee’s art. On the one hand, Werckmeister writes, there is the acceptance of Klee’s art on his own terms, aiming at a scholarly ratification of Klee’s public self-presentation for posterity—the dominant approach of Klee scholars from Will Grohmann through Jürgen Glaesemer, anchored in biography rather than history.¹⁸ According to Werckmeister, the alternative to such art-historical affirmation of Klee’s self-presentation is an inquiry such as his own, which aims at a historical critique of culture as ideology and treats Klee’s works as not ends but functions of his career, the true object of Werckmeister’s study (9).

The aims of my study do not fit either side of Werckmeister’s false dilemma. A major part of my project is to comprehend the terms of Klee’s art in a conceptually and historically rigorous way: I aim to give an account of how the relations between pictures, writing, and literature signified in and for Klee’s art, as well as how and why these relations mattered to Klee and to his interpreters in the 1910s and 1920s. Such an account need not merely extend Klee’s self-presentation in his diary and publications. Take the notion Werckmeister calls one of the two guiding ideas of the Klee literature he criticizes as merely perpetuating Klee’s self-mythologization: the convergence of Klee’s protracted autodidactic formation as a painter with the systematic buildup of painting from its elements. This is the idea, as Werckmeister explains, underlying the often repeated developmental narrative of Klee’s life and art, in which he is said to have mastered first line, then tone, and then at last color, becoming a true painter with his 1914 trip to Tunisia (5). The argument of my study demonstrates that an interpretive investigation of the terms of Klee’s art need not merely repeat the mythic version of his artistic biography that the Klee literature of the fifties and sixties so often presented; in fact, such an investigation can complicate that story all the more decisively by keeping Klee’s art at the center of inquiry.

I have focused my study on Klee’s work of the later 1910s and the early 1920s, because it is during these years that these questions of the visible and the legible became most pressing.¹⁹ One of my own premises is that relations between verbal language and pictures are far from immutable, and may function and mean very differently in different historical moments. I do not think that we can assume that the terms of these relations hold constant across the long span of Klee’s career, from his work as an art student in turn-of-the-century Munich to his exile in Switzerland in the 1930s. Therefore I focus on Klee’s work beginning with its shift toward the literary at around the time of his conscription into the German army in 1916 and ending with his response to the 1923 reorientation of the Bauhaus, where he taught, because I see these relations as the crux of the new way of working that Klee arrives at in 1916, which provided the framework for his work in the following years.²⁰

Whereas, in 1912 and 1913, during the initial period of Klee’s induction into the Munich avant-garde, Robert Delaunay’s Windows series represented for him the most important paradigm for abstract painting and the model for his own work, I argue that Klee fundamentally rethought abstraction around 1916 by centering it not on painting but rather on graphic art (Graphik), understood not as the loose grouping of various processes and mediums usually implied by the English term graphic arts but rather as a singular art with its own proper essence, which, according to Klee, tends toward abstraction.²¹ The essence of graphic art, he writes, tempts easily and rightly to abstraction.²² The results of this reorganization, apparent in Klee’s art and writing, are peculiar. Klee’s graphic abstraction deploys eighteenth- and nineteenth-century theories of graphic art as near to writing and to literature (as has often been noted, a nearness emblematized etymologically by the Greek root of both the English word graphic and the German Graphik—graphein, to write or draw).²³ This graphic abstraction drew on, responded to, and pushed against a range of texts on the relations of art in general or of graphic art in particular to writing—above all Gotthold Ephraim Lessing’s Laocoon: An Essay on the Limits of Painting and Poetry (1766), but also texts by the Romantic writer E. T. A. Hoffmann, the poet and critic Charles Baudelaire, the artists Max Klinger and Wassily Kandinsky, the critic Julius Meier-Graefe, and the reform pedagogue Georg Kerschensteiner.

I argue that around 1916, Klee began to develop a theory and practice of a graphic abstraction that would be self-reflexive in its insistence on certain peculiarities of the graphic arts as written about in eighteenth- and nineteenth-century art theory: the close relation of picture to caption; the encouragement of ideas or speculation; the notions that the activity of looking at a work of graphic art is similar to that of reading a text, and that the demands such looking places on the viewer’s imagination make the graphic arts peculiarly appropriate for fantastic or invisible subjects. He began using the repertory of hieroglyphs (a word that occurs frequently in the titles of his works of the late 1910s)—rudimentary stars, crescents, trees, animals, buildings, and so on, often interspersed with letters—that seem to hover between signifying by resemblance and signifying by convention. He presented his Blätter (leaves, sheets, or pages, a word commonly used to refer both to artworks on paper and to the pages of a book)²⁴—not only works on paper, but also small pieces of gessoed and painted canvas that one would ordinarily think of as paintings—in a manner customary for prints or drawings, mounting them on pieces of cardboard and inscribing them with titles. In one case, he took one of his own abstract oil paintings of 1914, when Delaunay was his model, off its stretcher bars, to glue it to a mount inscribed with a title that invites speculation—Carpet of Memory.

Klee appears to have invented this graphic abstraction in order to work against dominant theories and practices of abstract painting as well as against the formalist aesthetics put forward by the art critics Meier-Graefe and Karl Scheffler, which he appears to have associated with Lessing’s Laocoon. I argue that Lessing’s treatise, which Klee read as a Gymnasium student, was decisive for him, and crucial for understanding the aims of his graphic abstraction of the 1910s: his invention of a graphic alternative to the painting of achieved tabular unity represented for him by Cézanne and Hans von Marées—a self-reflexive medium that includes those aspects of viewing that modernist painting, from impressionism through Kandinsky and Delaunay, excluded as literary, extrinsic to the concerns of the visual arts. Klee’s work of these years can be seen to push against conceptions of how pictures are to be looked at that emphasize instantaneousness—he deploys the idea of the graphic against these conceptions, redefining abstraction as he does so. What has made this difficult to see is Klee’s ambivalence toward the concepts, practices, and figures that he links together and works against through his graphic abstraction. They are of foundational importance for his art without being influences in the usual sense of positive models for identification (and yet they sometimes play that role, too).

This aspect of Klee’s work has been occluded both by his celebration as an exemplary modern painter in the 1950s and 1960s and by more recent scholarship that deals with Klee’s art as an epiphenomenon of the art market, although for very different reasons. Tracing it out requires attention to what Klee counters, what he works against, and to his frequent tactic of inverting the terms of art-critical discourse. For instance, as Jenny Anger has demonstrated, Klee took up the idea of the ornament in the 1910s, when Kandinsky and Franz Marc were at pains to distinguish their work from mere decoration.²⁵ My study shows that this operation of taking up and playing with the counterterm of an opposition is pervasive in Klee’s work of the later 1910s and early 1920s; this is an operation he performs not only on the expressionist opposition between abstract painting and decoration but on many other terms of art theory and criticism since the Enlightenment.²⁶

Besides turning to the texts that I argue served Klee as resources and impetuses of various kinds, I also draw on commentary on Klee’s art from this same period, and most of all on the writings of Wilhelm Hausenstein and Walter Benjamin, both of whom were deeply interested in Klee, to aid me in interpreting his work and particularly in thinking about the significance of the graphic during and after the First World War. Hausenstein, a prominent Munich-based art critic, became fascinated with Klee’s work as he became increasingly disillusioned with expressionism during the war and its aftermath. His 1921 monograph (the first substantial extended discussion of Klee’s art)²⁷ argues that the painter-draftsman Klee responds to the problem of painting in the crisis of representation brought about by modernity in general and modern warfare in particular by inventing a mode of painting-drawing that is adequate to what Hausenstein describes as the ruined condition of persons and things in a way that painting cannot be.

The first chapter of my study examines the shift in Klee’s work around the time of his 1916 conscription and the different explanations that have been offered for it, including Klee’s own reflections, the connections Hausenstein makes between the graphic mode of Klee’s art and conditions of social and technological modernity, and the explanations that art historians have offered more recently. Contrasting Klee’s 1913 translation of Delaunay’s essay on the aims of abstract painting, On Light, with the essay on graphic art that Klee drafted in 1918, which turns the logic of Delaunay’s text on its head, I argue that Klee shifts to a model of abstraction based on graphic art and sharply divergent from the theories of abstract painting developed by the leading avant-garde artists with whom he was in contact during the years right before the war. Furthermore, I show that this shift precipitated out of Klee’s protracted negotiations with ideas of ‘pure’ plastic art, as he termed it, arguing with and against figures ranging from Lessing to Delaunay.

The second chapter asks why Klee’s recentering of abstraction on graphic art had the consequences it did, and in particular how it allowed him to reconceive the activity of viewing a picture. Examining a little-known fragment written by Benjamin in 1918 or 1919, in which he speaks of the uncolorful picture as one that demands that the viewer describe it, as well as ideas of the graphic in older texts that served as resources for Klee, I show how graphic art was understood as inviting a mode of engagement on