A Spacewalk With Joseph Frank’s “Spatial Form”

(For “The Novel as a Form of Thought,” a conference commemorating Joseph Frank, Committee on Social Thought, University of Chicago, May 15-16, 2014.)

Joseph Frank was fully developed before he came to the Committee on Social Thought. His essay “Spatial Form,” first published in the Sewanee Review in 1945, is an astonishing piece of synthesis when you consider the state of play at the time. Joe Frank, born in 1918, had already at age 27 the stylistic authority and the full deck of reference that we find in his older contemporary Clement Greenberg, for example; but Greenberg, who preceded Joe at Erasmus Hall High School, had a proper BA (Phi Beta Kappa) from Syracuse, whereas Joe cobbled together his few semesters of college education between bouts of paid work and was largely educated through talking with freelance or underemployed intellectuals. Later, joining the class of underemployed intellectuals as a copy writer for the Bureau of National Affairs in Washington D.C., Joe was again walking in Greenberg’s footsteps; Greenberg’s non-academic jobs were with various federal agencies until he became a full-time editor at Partisan Review and The Nation in the 1950s. And like Frank, Greenberg made a strong impression with an early essay, in his case “Avant-Garde and Kitsch” (1939), which showed him to be fully up to date with the conversation about art, ethics and politics then underway in what would later be known as the Frankfurt School.[1]

Joe’s biography is too big a subject for me today. My reason for enumerating these facts and making the parallel with Greenberg is this. For people who came of age in the 1930s and 40s, art mattered in a way that we can hardly recover, even as a theme for nostalgia. It was important to discern what made modern literature and art modern, because in the adequate description of those representational artifices lay, one thought, a diagnosis of the spirit of the age, and it was important to get that right. Part of the reason lay in the contending teleologies of the day, theories of history with vastly divergent political formations behind them, structures of intention that claimed the power to determine one’s day-to-day actions and options. How we got where we are today, in a much weakened state of mind, onlookers if not scroungers at the remote edges of a frightfully well-financed commercial culture, is a complicated story. You have heard the recurrent laments for the demise of the public intellectual, specifically the sub-species of public intellectual whose habitat was outside the universities. From a U.S. point of view, the relative but steady rise in living standards, the massification and commodification of university education, and the cultural assimilation of previously excluded groups must all have had something to do with it, as these generally good things both undermined the cause of the Left as people of the 1930s understood it and broke down the difference between high culture and mass culture. If “Spatial Form” had been published in 1965, it would have been an academic, formalist exercise, and it’s often been mistaken for one since.

Joe did not have an academic position, or the prospect of one, when he published that article. Allen Tate, who took it for the Sewanee Review, may have had ulterior motives that Joe would have quizzed (more on this later). First, let’s define the agenda. Joe offers his observations in order to “trace the evolution of form in modern poetry and, more particularly, in the novel. For modern literature… is moving in the direction of spatial form…. And since changes in aesthetic form always involve major changes in the sensibility of a particular cultural period, an effort will be made to outline the spiritual attitude that have led to the predominance of spatial form.”[2]  The essay proposes, then, to trace specific types of perceptual experience crystallized in works of art to a “spiritual attitude” and “sensibility.” In rhetorical terms: lexis is traced back to pathos and êthos. This is the sort of task that aesthetics was invented to perform, back in 1750 when Alexander Baumgarten defined that new discipline as “theoria liberalium artium” and “scientia cognitionis sensitivae.”

The artist and the critic together enable the age to see its own contours—the critic’s contribution being indispensable in an era of difficult art, art that reveals itself by hiding.

Responses to the essay have often taken the form of quibbles about whether there is such a thing as “spatial form,” whether it is properly diagnostic of modernism, or whether it isn’t the form proper to a restricted subgroup of artists who might be marginal in some other account of modernism. By now there are several canons of modernism and hundreds of definitions of the phenomenon—but if you scratch them, I suspect most will have a large if unacknowledged debt to Joe Frank’s ideas. His main references in this essay are T.S. Eliot, Ezra Pound, Marcel Proust, James Joyce, Djuna Barnes—still eminent members of the modernist canon nearly seventy years after Joe based his argument on them, though many people today would affix to them the ambiguous label of “high modernism.” These references allow us to put Joe’s essay in a more specific temporal register. It is very much of the moment at which modernism became high and serious. As of the late 30s, the shock and iconoclasm of the new art of the 1920s had begun to be accounted for by aesthetic theories which, using the tools of traditional rhetoric and thematics, bridged their gaps and made tolerable their disharmonies. Works that had been printed on shocking pink paper in BLAST, or been responsible for the confiscation of numbers of The Little Review, would be made respectable by commentary in the Kenyon Review. This was, for the moment, paradoxical, as paradoxical as Ezra Pound receiving the Library of Congress’s Bollingen Prize from the mental hospital where he was confined instead of standing trial for treason; but the paradox settled into consensus, and as a result, by the 1970s and 80s, one finds The Waste Land, Ulysses, the Cantos, once disturbing and caustic, being described as “marmoreal,” “dull” and “classic.” But the young Joe Frank stood at the beginning of this process; in 1945 the transformation of modernist art into modern classics was still an exciting venture with the odds against it.

            “Modern literature,” says Joe, “is moving in the direction of spatial form.” Where does that remark, as we’d say today, come from? Where does it stand? With examples from Flaubert, Mallarmé, Hulme, Pound, Eliot, and the interwar novelists, Joe can certainly trace an increasing tendency to transform causal relations into atemporal patterns; but the words he uses, is moving in the direction of, are resolutely temporal, even teleological. They seem to indicate that his own description is not an example of spatial form, that it is not, therefore, part of the movement of modernist art. Is it an account of modernism that stands aside from the “spiritual attitudes” of its time? Where does Joseph Frank take his own bearings? We don’t get an answer from within the essay—perhaps because of the assurance of its voice. My hunch is that it is out of necessity and a sheer relativism of the will that the essay grants itself a position outside the moving universe of modernism, since if the writer were within it, he might be unable to detect that universe’s motion. He and it would be moving in the same direction and at the same speed.

Is moving in the direction of spatial form has a further quality of paradox, since that would be motion toward a state of motionlessness. What does this mean? Is it even possible? Obviously it is possible for things to slow down (it happens to me regularly), which would be an empirical sort of motion toward a state of rest. But Joe certainly isn’t saying that the experience of modernity has on the whole been a slowing down. No one would accept that as a description. Indeed Joe says: “If there is one theme that dominates the history of modern culture since the last quarter of the nineteenth century, it is precisely that of insecurity, instability, the feeling of loss of control over the meaning and purpose of life amidst the continuing triumphs of science and technics” (SF, 58). Note in this description the mention of “continuing triumphs.” For science and technology, progress is temporal and linear and contains this series of triumphs. Culture, then, is defined by its resistance to the forces that keep science and technics on their forward path. As people experienced a loss of control, they sought in art a compensating control, a means of arresting time. Spatial form is a sign of the refusal on the part of culture (or of avant-garde culture) to play along with the overall logic of progress. The part of culture that resists progress also presents itself as the newest, the most modern. If that is a paradox, it is just the defining paradox of the high modernists.

            Something very much like the “movement in the direction of spatial form” was articulated by R. P. Blackmur in his 1934 evaluation of Pound’s first forty Cantos. Perhaps because he is not committed to the language of time and space, Blackmur can help us see what Joe is doing with that particular vocabulary. “Mr. Pound put together the materials and roused the interest appropriate to a narrative, and then deliberately refused the materials a narrative form, without, however, destroying the interests that expected it. Whether intentionally or not, it is the presence of this defeated expectation which holds these Cantos together.”[3] The Cantos, that is, seem to promise a history because they are full of references to John Adams, Confucius, Sigismondo Malatesta, and so on. But the names and words of history are used to make an associative collage, and sooner or later the reader discovers, possibly in bitterness of heart, that there’s no story there, only a pattern, and an untidy one at that. It doesn’t go forward, it doesn’t conclude. “Defeated expectation” is stalled narrative.

            That sounds entirely negative (and many unhappy readers of the Cantos would be just fine with that as a diagnosis, I fear). Here is Joe’s recommendation for a happy way out of the jam.

Aesthetic form in modern poetry, then, is based on a space-logic that demands a complete reorientation in the reader’s attitude toward language. Since the primary reference of any word-group is to something inside the poem itself, language in modern poetry is really reflexive. The meaning-relationship is completed only by the simultaneous perception in space of word-groups that have no comprehensible relation to each other when read consecutively in time…. modern poetry asks its readers to suspend the process of individual reference temporarily until the entire pattern of internal references can be apprehended as a unity. (SF, 15)[4]

Modern poetry, and following poetry, the modern novel, turn the previously expected laws of sequence—logical implication, cause and effect, chronology, development—into patterns of elements that are perceived as similar or different, that is, they become the poles of reversible, non-temporal relations of which metaphor is the chief.  “By this juxtaposition of past and present, as Allen Tate realized, history becomes ahistorical…. Past and present are apprehended spatially,” accomplishing “the transformation of the historical imagination into myth” (SF, 63). Myth, of course, eternally recurs; as the last chorus of Goethe’s Faust, part 2, which is also the last chorus of Roman Jakobson’s “Linguistics and Poetics” talk, has it, “Alles Vergängliche ist nur ein Gleichnis,” or in the specialized translation provided under the banner of the Poetic Function, “anything sequent is a simile.”

“Spatial” form is, if we look closer, a somewhat bigger tent than Joe needed, because the space to which all modernist arts, in his account, aspire is not a three-dimensional one. Depth must be excluded, for “Depth… gives objects a time-value because it places them in the real world in which events occur. Now time is the very condition of that flux and change from which, as we have seen, man wishes to escape… hence nonnaturalistic styles shun the dimension of depth and prefer the plane” (SF, 60). Spatial form should properly be called planar form. It turns the objects of its representation into points arrayed on a single surface, meant to be apprehensible in a moment’s intuition.

Joe was led to this formulation by his response to a particular modern novel, Nightwood by Djuna Barnes, published in 1937 with an introduction by Eliot. In Nightwood, says Joe,

the naturalistic principle has lost its dominance…. Since the selection of detail in Nightwood is governed not by the logic of verisimilitude but by the demands of the décor necessary to enhance the symbolic significance of the characters, the novel has baffled even its most fascinated admirers. Let us attack the mystery by applying our method of reflexive reference, instead of approaching the book, as most of its readers have done, in terms of a coherent temporal pattern of narrative. (33)

Like Blackmur, but more cheerfully, Joe read Nightwood as the shell of a “defeated expectation.” It is one of those novels in which nothing much happens. If you are looking for a plot, you will be disappointed. But no one should read Nightwood in the privative mode, as a novel lacking something. What it has is poetic, as Eliot’s introduction had somewhat clumsily said. Many of its pages are taken up with the monologues of a single peripheral character, the doctor to whom another character brings an account of her sufferings as the frustrated lover of another woman who seems cold, absent, unpredictable. It occurs to me that her frustration in trying to have a regular human relation to this woman must be parallel to the frustration of the reader of a novel that is not wholly novel-like. The narrator of the novel says, describing the slippery second main character:

Sometimes one meets a woman who is beast turning human. Such a person’s every movement will reduce to an image of a forgotten experience: a mirage of an eternal wedding cast upon the racial memory; as insupportable a joy as would be the vision of an eland coming down an aisle of trees, chapleted with orange blossoms and bridal veil, a hoof raised in the economy of fear, stepping in the trepidation of flesh that will become a myth. (cited, SF, 35)

This is narrative prose trying to escape the condition of novel or epic and setting its prow toward myth, and being explicit about it. Everything necessary is here: the suggestion of anamnesis, the “eternal wedding” uniting human and animal species, the moment of ritual stasis, a transgressive costuming, an “economy of fear,” the suggestion of imminent sacrifice, even an apparent pun on Bergson’s “élan vital.” If what Barnes calls “myth” lays out such a pattern, then it is no surprise that the characters of the novel fall into it. But there is a difference between her appeal to myth and the uses Eliot made of the Grail legend or Joyce made of the Odysseus epic. In Eliot and Joyce, there is a conferral of meanings from the mythic pattern to the tawdry present-day analogue—from the stone thrown by Polyphemus, for example, to the dinner pail thrown by the irascible Irish nationalist in Kiernan’s bar. In his review praising Ulysses Eliot had linked the “mythic method” with “order” and said that it was the closest thing to a classicism that the broken postwar world was going to get: “It is simply a way of controlling, of ordering, of giving a shape and a significance to the immense panorama of futility and anarchy which is contemporary history…. Instead of narrative method, we may now use the mythical method” (“Ulysses, Order, and Myth”). But for Djuna Barnes, the reference to archaic myth and ritual does not resolve anything, it does not serve the narrator’s ambition to create order, rather it is the expression of a character’s frantic need for an order that is delivered only as a delusion. Myth is the debunking of a ritual that is the fantasy generated by an impotence.[5]

One can think of many reasons that the co-opting of modernist literature by the academy would begin with Eliot’s Waste Land (reduced to an instance of organic form and a Christian parable by Cleanth Brooks in 1937) and Joyce’s Ulysses, rather than works like Nightwood, the Cantos, or Hart Crane’s The Bridge: works that resist a unitary reading, perhaps by their flaws as much as by their intent. Nightwood, weighing on Joe’s mind for some years already, was a curious, non-average choice to become his essay’s exemplary modernist work. In 1978 he seems to back off from his choice, musing that it was “not destined, as the passage of time has shown, to exercise a major influence on the course of the novel; quite the contrary, its metaphoric texture… has remained something of a sport, technically speaking” (SF, 109; but I’m glad to see he’d changed his mind again by 1991: SF, xii). Whether or not it is a “sport”—freak, anomaly or hybrid—Nightwood occupies the middle third of his essay and seems to me indispensable to making the strongest version of his argument about the modernist novel’s bent towards metaphor and planarity.

So does the essay’s correlation between aesthetic forms and forms of consciousness still convince? An argument through examples always invites other examples. The central claim, however, could, I think, be reformulated in a way that would include more kinds of modernist consciousness and aesthetic play. Joe begins his essay with a nod to Lessing’s Laocoon with its separation of sculpture as spatial art from poetry as a time art. Lessing does this in order to differentiate the media and urge us not to confuse their divergent excellences, not to judge a painting for the qualities that should distinguish a poem or vice versa. I will jump back a generation to Lessing’s predecessor as librarian at Wolfenbüttel, Leibniz, who defined time as “an order of successions” and space as “an order of simultaneities.” The brilliant thing about this definition is that it puts the two orders, space and time, on a basis of similarity. There is by these definitions no radical heterogeneity between space and time. Both are orders, by which is meant, sets of relationships. One set of relationships is described as an ongoing mutual exclusion (no two moments of time can be other than successive) and the other is a mutual compatibility. Both orders are entirely internally relative or self-positing; it is idle to imagine an absolute space or time. We can measure the interval between events, we can measure the distances and directions separating things, and these measurements are ways of stating the relations. Leibniz would just insist on our treating them indifferently, as both space and time are for him merely specious frames in which we experience an ontology that, from the monadological point of view, doesn’t divide up that way at all. When you need someone to explain “why there is anything at all, rather than nothing,” the Leibnizian will sketch out a world in which multiple sets of compossible elements combine to make virtual worlds, one and only one of which realizes the maximum of the features that a world ought to contain.

Joe Frank’s account of how the twentieth century’s artworks rejected the leading temporal principle of organization in nineteenth-century artworks might benefit from a retelling that looked farther up and down the “aisle of trees,” or sideways into other disciplines. It is certain that the century or so preceding 1945 had been a hard one for space and a good one for time. Bergson, Husserl, Heidegger and many others took the intuition of the passage of time to be the thread on which an authentic subjectivity hung. To externalize thoughts or concepts into space was necessarily to debase them. Hegel mocked the empires of Asia as “realms of space” to which the touch of time had never come, “mechanical” assemblages of men and landmasses on which only an “ungeschichtliche Geschichte” (an unhistorical history) could take place. All moments there might as well be the same moment, since nothing changed. So to take the side of space was to defy not only progress, but le vécu, the dialectical unfolding of the concept and our relation to death. It is no accident that so many of the proponents of a spatialization of art were political reactionaries, starting from Maurras and Lasserre (SF, 59), and going on to the people who repeated their gospel as founding elements of Anglo-American modernism. Nor is it by chance that the main examples of spatialized narrative postdate World War I, that catastrophe for so many ideas of progress and purpose. The parti pris of space might also explain the twentieth-century rediscovery of Dante, the argument of whose poem is absolutely spatialized (hell, purgatory, heaven, each minutely subdivided) and set in a time outside all historical time—with the sole exception of the time it takes its narrator to experience it.

But the version of the space versus time duality asserted in Joe’s essay assumes that the artwork has to have its true residence in one or the other realm. Virtuality, which is what Leibniz was really interested in, doesn’t force you to choose in that way. The complex of atemporal relations, capable of spinning itself out in linear form but not obligated to do so, that Ferdinand de Saussure called “la langue” might have exemplified another way of standing aside from time, one that meant a great deal to many other twentieth-century modernists, particularly those with an interest in space as the form that culture takes. Thus the “open works” described by Umberto Eco, the generative poetics of the Oulipo group, the many poets who have taken up the hints of Mallarmé’s Coup de dés, the culture theory of the Prague School, the combinatory play of folklore or manuscript variance, John Cage’s composition through chance operations, Borges, Queneau, Cortázar, Calvino, all might suggest a canon of artworks structures by a non-time-bound logic and that are nonetheless not shut off from time. The tendency to read spatial form as a symptom of despair seems to me too narrow, in light of the broader affinities of the “order of coexistences”; but the tone of constraint and crisis probably could not have been avoided by any serious person setting pen to paper in 1944-45. And it’s no dismissive remark to say that “Spatial Form” is of its time.


[1] Clement Greenberg, “Avant-Garde and Kitsch,” The Collected Essays and Criticism, vol. 1: Perceptions and Judgments, 1939-1944, ed. John O’Brian (Chicago: University of Chicago Press, 1986), pp. 5-22.

[2] Joseph Frank, The Idea of Spatial Form (New Brunswick: Rutgers University Press, 1991), p. 10. Further references to this text will be in the form SF.

[3] R. P. Blackmur, “Masks of Ezra Pound,” Form and Value in Modern Poetry (New York: Anchor Books, 1952), p. 100.

[4] The oft-recurring word “pattern” suggests a parallel with Ruth Benedict’s anthropological relativism. As a good Boasian, she rejected one-track models of the evolution of human societies; her comparison emphasizes their “incommensurable” co-presence. See Benedict, Patterns of Culture (Boston: Houghton Mifflin, 1934), esp. p. 223; Susan Hegeman, Patterns for America: Modernism and the Concept of Culture (Princeton: Princeton University Press, 2000). “As if a magic lantern threw the nerves in patterns on a screen.”

[5] On myths that debunk rituals, see Wendy Doniger, On Hinduism (New York: Oxford University Press, 2014), p. 226-230.