1To investigate the making of space in painting is my goal. But we cannot make a start until we have considered a series of seemingly straightforward questions, not all of which have such straightforward answers: what is pictorial space? How does this space function in relation to its surround, be that quadrilateral frame, architectural surround, rock wall or streetscape? What, if anything, allows us to confer meaning on these spaces? How do these spaces construct a place for us, the spectators, from which we may view the artwork? Above all, what are the various means by which pictorial space is constructed and deconstructed, made and unmade, or otherwise brought into being for the purposes of art?

2I begin my exploration with one overriding assumption; that space is never neutral, never simply there, but is defined, demystified, valorized, challenged, given shape, created, through the operation of its frame, whether this frame be a material, ideological, or conceptual structure. It is therefore with this symbiotic relationship between frame as enclosure and frame as space-maker that we must begin. Yet there is one further factor to acknowledge: as with the frame, space in painting is a construction closely associated with the discourse on art; without discourse there is no frame and consequently no space. To borrow Henri Lefebvre’s formulation, the “production of space” is, at the same time, the production of meaning (Lefebvre 1992, 15).

3In a ground-breaking essay from 1969, Meyer Schapiro discusses the frame from the point of view of its dual limit as space and enclosure. He points out that there is nothing natural or inevitable about the quadrilateral frame around a painting, or for that matter the spaces of the pictorial field it defines (see fig. 1). Citing the example of the cave paintings of Paleolithic Europe, in which the images are drawn on rock walls, he comments that the natural irregularities of the surface militate against the production of a defined edge to the representation: “The artist worked then on a field with no set boundaries and thought so little of the surface as a distinct ground that he often painted his animal figure over a previously painted image without erasing the latter, as if it were invisible to the viewer” (Schapiro 1994, 1). Schapiro’s example reminds us that the framing of space has dimensions that concern both the width/height and the depth of the image, leading to the construction of a pictorial field that may be, as with cave painting, limited by the constraints of location, available pigments, and the difficulties of executing work in near-total darkness. But if Paleolithic peoples did not regard the representational field (the space of the painting) as being in any way defined by the expectation or presence of a frame around the image, that does not apply to us. As Schapiro points out: “we tend to take for granted the regular margin and frame as essential features of the image” (Schapiro 1994, 7). Indeed, from the time Jacques Derrida introduced discussion of these boundaries some forty years ago, the frame has served as the go-to example of the structural devices that “give rise to the work” (Derrida 1987, 9).

4This sounds a lot like the formation of pictorial space as envisaged by Filippo Brunelleschi, in which frame and space combine to allow for the representation of an apparently three-dimensional space on a two-dimensional surface (it is often overlooked that linear perspective cannot work without the operation of the frame to fix the distance point – the viewer’s position – and consequently position the orthogonal that recedes into pictorial space to meet at the vanishing/centric point). As Leon Battista Alberti explains in his treatise, On Painting: “On the surface on which I am going to paint, I draw a rectangle of whatever size I want, which I regard as an open window through which the subject of the painting is to be seen” (Alberti 1991, 54). Alberti’s intention, then, is to emphasize the ability of the spatial construction to serve as a setting for a narrative, which for humanists meant the representation of a dramatic action drawn from mythology, ancient history, or the Bible, and characterized by an emphasis on the human form as the vehicle for the narrative (Baxandall 1971, 121-39).

5Alberti had little interest in subjects that today we would consider legitimate topics for depiction, such as portrait, landscape, genre scenes or still life. Rather, he follows the lead of Aristotle in The Poetics in considering the historia – by definition a narrative based on human history – to be the basis of the artwork (Halliwell 1987, 37-41):

The first thing that gives pleasure in a “historia” is a plentiful variety [. . .] I would say a picture was richly varied if it contained a properly arranged mixture of old men, youths, boys, matrons, maidens, children, domestic animals, dogs, birds, horses, sheep, buildings and provinces; and I would praise any great variety, provided it is appropriate to what is going on in the picture [. . .]. (Alberti 1991, 75)

7While this staging of pictorial space serves the purpose of fifteenth-century picturing, Erwin Panofsky articulated doubts about both the naturalism and optical accuracy of what Alberti called a costruzione legittima (Grayson 1964, 14-17). Perceptively, Panofsky notes that pictorial space is a “systematic abstraction” and not a mirror held up to the world:

Homogeneous space is never given space, but space produced by construction [. . .], [it] is a systematic abstraction from the structure of the psychophysiological space [. . .]. It negates the differences between front and back, between right and left, between bodies and intervening space (“empty” space), so that the sum of all the parts of space and all its contents are absorbed into a single “quantum continuum”. (Panofsky 1997, 29-31)

8Panofsky’s recognition that space is constructed and not a mere copy of reality prompts him to identify perspective as a symbolic form, a concept he borrows from the Neo-Kantian philosopher Ernst Cassirer to explain how a system of representation such as perspective is not “the mere reproduction of a ready-made, given reality,” but a construction (Cassirer 1944, 143). Cassirer’s thought is not always easy to plumb, but the essence of his ideas is, as Panofsky realized, fundamental to the processes of pictorial construction. Against the popular notion that art succeeds the more it imitates reality (mimesis), Cassirer shows that art is the result of human consciousness shaping the world to a reality that we have evolved to inhabit.

9Panofsky’s insights tell us much about the nature of picturing. It is never a natural process, a straightforward matter of copying what is before one’s eyes. Rather, it is a process of exploration, examination, and simplification, of reconciling differences between seeing and knowing. Artists are required constantly to make choices, to evaluate and to select from among the visual data before them. Even then, as Panofsky remarks, “[i]t is not only the effect of perspectival construction, but indeed its intended purpose, to realize in the representation of space precisely that homogeneity and boundlessness foreign to the direct experience of that space” (Panofsky 1997, 30-31). We might paraphrase Panofsky by saying that it is the task of perspective to persuade us into accepting verisimilitude (the look of things) as the truth of their phenomenal reality, and that the apparently coherent spaces opened up by the picture plane really are there. But space is not an aporia between things, nor simply the neutral background to a subject; rather, it is as much the site of a dialogue between painting and beholder as any other part of the picture.

10That such an apparent paradox is not, in reality, a paradox at all is evidenced by the philosopher Martin Heidegger’s formulation in seeing in the “frame of art” neither a supplement nor completer of the artwork, but the means by which a work is given the space it needs to become art. A full discussion of Heidegger’s complex thought is beyond the scope of this paper, but his use of the neologism ge-stell [“enframing”], to indicate what he calls a “bounding outline (peras)”, is exemplified by the observation, in his essay “The Origin of the Work of Art,” that art is not art until it has a recognizable frame:

A building, a Greek temple, portrays nothing. It simply stands there in the middle of the rock-cleft valley. The building encloses the figure of the god, and in this concealment lets it stand out into the holy precinct through the open portico. By means of the temple, the god is present in the temple. (Heidegger 1992, 167)

11Heidegger’s example suggests the work of the frame is not only to provide a limitation to the space of the presentation, or the delimitation of an inside/outside to the artwork, and still less that of mere shelter or protective surround, but recognition of a mutual dependency between the act of enframing and the production of the spaces of art. It is this active attention to the notion of border or limit that allows Heidegger to argue that the artwork “sets in place” a world – a point he emphasizes in his essay “The Age of the World Picture,” where a world picture is not a picture of the world but the world framed as a picture, for which “man is prepared and which, correspondingly, he therefore intends to bring before himself” (Heidegger 1977, 129).

12To pursue this line of thinking in terms of “preparedness,” Michael Ann Holly, on a visit to the Kunsthistorisches Museum in Vienna, found herself caught up by how “backgrounds and foregrounds make each other come alive” in many of the paintings she contemplated. Borrowing Victor Stoichita’s phrase “self-aware images” (we might think, “prepared for viewing”) to describe pictures that effect “the splitting or inverting of space,” Holly asks rhetorically of one of the Kunsthistorisches’ many masterpieces – Peter Aertsen’s Vanitas Still Life entitled “Christ with Mary and Martha” (see fig. 2):

To what might the inside-outness of this painting be a testament, if not to its own ludic riddles? [. . .] This is a painting teasing us about the cut between real and fictive space, the sacred and the profane, text and image, foreground and background – in Immanuel Kant and Jacques Derrida’s terms, the split between ergon and parergon. (Holly 2019, 895-98)

13Aertsen’s still life could just as easily be entitled Christ in the House of Mary and Martha, as it includes a depiction of the story in Luke 10.38-42, in which Christ defends his follower Mary from her elder sister Martha’s objection that she should help in the kitchen instead of wasting her time listening to Christ’s teaching. Yet this reading does not depend exclusively on knowledge of the Biblical story, but on Aertsen’s inversion of the composition to make the foregrounded still life, ostensibly the principal subject (ergon), function as a pictorial and iconographical frame to the Biblical subject. Holly reminds us that “the cut between real and fictive spaces” is analogous to Jacques Derrida’s discussion of the parergon – that is, between the principal work or subject and the subaltern spaces that characterize the marginal elements of the composition.

14This relation between subject and pictorial space is at the core of what French diplomat and author Blaise de Vigenère, writing in the late sixteenth century, calls “superfluous additions”:

Among their works painters paint perspectives, shrubby trees, little animals, old ruins, and collapsed buildings, mountains and valleys, together with such other accessories and incidents that serve to enrich and give grace to their task, and fill out that which otherwise would remain uselessly denuded and empty, in danger of being an eyesore. The Greeks call them παρεργα [parerga], or superfluous additions, besides what is needed. (De Vigenère 1999, 131)

15By damning with faint praise those “accessories” and “incidents” that serve to fill the otherwise “denuded” spaces of the composition, De Vigenère adheres to the tenets of history painting that require such details to contribute to the presentation of the main subject, and not detract from the main story.

16But while attention to the hierarchy of genres in Italy and particularly in France maintained an iron grip on the choice of subject matter through the early modern period, Christopher S. Wood argues that, in the North, artists like Albrecht Altdorfer had, as early as the sixteenth century, melded the genres, thereby “abolish[ing] the boundary between subject and setting, as well as the hierarchy, by banishing the subject and funneling setting toward the literal center of the image” (Wood 1993, 69). This is highly suggestive for a better understanding of parergonality. Wood’s comment reveals the structural displacement of the central subject to the margins of the painting, and the concomitant migration of the landscape, previously considered either secondary or incidental in relation to the narrative, to the center of the representation. This inversion of allows for the fusion of the center/periphery (ergon/parergon) relationship and the consequent depreciation, or loss, of the central historia. Such inessential details may draw attention away from the representation of the principal action to the point where there is no recognizable action, or where the subject is classed merely as a genre painting.

18As W. H. Auden writes in his poem, “Musée des Beaux-Arts”:

In Breughel’s Icarus, for instance: how everything turns away
Quite leisurely from the disaster; the ploughman may
Have heard the splash, the forsaken cry
But for him it was not an important failure […]. (Auden 1991, 179)

19Icarus had been imprisoned with his craftsman father, Daedalus, on the island of Crete. To escape, Daedalus fashioned wings that he attached to their arms with wax. But Icarus flies too close to the sun, and plunges into the ocean. Auden recognizes that Bruegel’s composition permits registration of separate actions – those of a ploughman, a fisherman, a shepherd, and protagonist Icarus – without immediately designating the principal narrative. At first glance, a newcomer to Bruegel’s painting might suppose it to be a genre subject. But on closer inspection our viewer would discover the painting to be more complex in both subject matter and composition than at first it appeared. The trigger for elucidating the various roles of the subordinate figures is the skyward gaze of the shepherd who watches, we suppose, the flight of an unseen Daedalus still high in the sky. By making strange a familiar narrative, complicating the reading, and introducing the perceptual conceit of an inverted order between ergon and parergon, we are obliged to “work through” the composition, somewhat in the manner of deciphering a painting by Cézanne, in order to arrive at the answer to the question, “What is the shepherd looking at?”

20So far, we have discussed action that corresponds with the main subject and confirms its centrality to the plot, whereas unrelated background elements serve little purpose other than fill up non-narrative space. This, at least, is the opinion of Edward Norgate, writing in Miniatura, or the Art of Limning (various editions between 1627/8 and 1648): “It doth appeare that the antients made any other Accompt or use of [landscape] but as a servant to their other peeces, to illustrate or sett their Historicall painting by filling up the empty Corners, or void places of Figures and story, with some fragment of Landscape” (Norgate 1919, 44). Norgate’s identification of landscape with absence of narrative alerts us to the assumption, from one perspective, that such spaces are merely there, while the dramatic action of a historia dominates through its centrality to the composition. These comments broadly echo his contemporary Thomas Blount, whose dictionary describes “landskip” as a “parergon,” countenancing its use in painting for its usefulness as a filler for background or non-narrative space: “As in the Table of our Saviours Passion, the picture of Christ on the Rood (which is the antient English word for Cross) the two Theeves, the blessed Virgin Mary, and St. John, are the Argument: But the City Jerusalem, the Country about, the clouds, and the like, are Landskip.”

21These comments remind us that issues of space cannot be fully understood outside the parameters imposed by the hierarchy of genres. To praise a painter for the uncritical depiction of external nature is to condemn that artist to irrelevance, at least in terms of the intelligent representation of subject matter. Where a significant subject is present, as in a Crucifixion, then the landscape could be what history painting always intended it to be – a parergonal (framing) support to the main subject. But if there is no main subject, no Aristotelian emphasis on dramatic action, then the painting would exhibit a lack for which no amount of skill in the depiction of the landscape could compensate.

22Another way of describing this emphasis on action is to recognize, along with Norgate and Blount, that space in history painting is meaningful when it is a support to the narrative, and an idle embellishment when it is not. That this interpretation is valid for all history paintings is evidenced by Nicolas Poussin’s Eliezer and Rebecca at the Well (1648), in which Poussin chose not to depict the train of camels the Biblical account states were present at the meeting at the well (see fig. 3). The question rests on the status of the camels in the promotion of the narrative. Are they essential to the understanding of the story, or would they merely fill up an “empty corner” that would otherwise lack interest? When, some thirty years later, the painting was the object of a lecture-debate in the Académie royale de peinture et de sculpture, the director Charles Le Brun offered cogent reasons why such “bizarre objects” – he meant the camels – would militate against the principal action and distract the spectator with inessential, and trivial, detail:

M. Poussin, seeking always to purify and clarify the subject of his works and to agreeably demonstrate the principal action he was addressing, rejected bizarre objects that could lead the eye of the spectator astray and amuse him only with trivia; that the field of representation is destined exclusively for the figures necessary to the subject and for those who are capable of an ingenious and agreeable expression; such that [Poussin] was correct not to concern himself with a train of camels that would have been as unprofitable for the work as embarrassing by their number [emphasis added]. (Mérot 1996, 136)

23Le Brun’s dogmatic defense of Poussin, the artist regarded as the epitome of history painting, is unsurprising. To have allowed any deviation from an adherence to the principles of la peinture d’histoire would open the floodgates to relativism. If camels, then why not gypsies, peasants, people of color or those afflicted with deformities? Not everyone present understood this. Philippe de Champaigne, a Jansonist, argued for the authority of the Bible, and one (unnamed) but well-meaning artist, presumably in a spirit of reconciliation between the disputing factions, suggested that Poussin should be allowed one camel to stand for the Biblical ten, but was silenced by Le Brun. The point here is that the framing of space within the pictorial field is not only a compositional matter of what to put where, but one touching on propriety, decorum, status, and value. As animals, camels fell into the category of genre, confined there due to their less-than-becoming appearance. In short, it is not possible, as Le Brun realized, to secure space designated as the site of sublime revelation (God’s choice of bride for Abraham’s son Isaac), from background detail without offense to the aims of history painting. In this case, the attention Poussin would have paid to irrelevant detail and unnecessary embellishment recalls Alberti’s comment that, “[i]n my opinion there will be no “historia” so rich in variety of things that the nine or ten men cannot worthily perform it” (Alberti 1991, 75).

24So far, we have considered the representation of space as a fundamental part of pictorial construction without focusing on space as the subject or theme of painting. It is now time to turn to three examples that will, I hope, move us towards an understanding of space as both an underpinning, and limit, to what we understand by the “framing” of space. The first of these examples is Joseph Wright of Derby’s A Philosopher giving that Lecture on the Orrery, in which a Lamp is put in the place of the Sun, first exhibited in London in 1766 (see fig. 6). In a darkened room, a man, resting his chin on his hand in contemplation of a mechanical model of the solar system, while listening to the discourse of a lecturer. Opposite, a lady, equally engaged, follows the movements of the machine into the vast distances of space, while an older man at right follows the words of the lecturer but looks off into the distance, as if contemplating the finiteness of human life in comparison to the boundless spaces of the universe. To the right of the lecturer, a youngish man takes notes, while a boy, seen from the back, looks down into the machinery that drives the model. Only the two young children, eagerly looking over the rim of the Orrery – for that is the name of this device – their faces illuminated by the glow of a concealed lamp, may be supposed to experience what Joseph Addison called the “pleasing Astonishment” of contemplating the Heavens, but less, we feel, as a result of the lecturer’s discourse, but rather through an understandable wonder at the workings of the marvellous machinery (Steele and Addison 1988, 399).

25Examination of the painting suggests that Wright was acutely aware of the potential of his subject to raise questions of space, perception, and reality. Indeed, the painting positions itself between its audience’s understanding of contemporary science and our expectations of what the reactions of such an audience might be. As such, the Orrery is much less a straightforward depiction of a scientific lecture, or an exercise in the British fashion for “conversation pieces,” than an attempt at contemporary history painting (Hobson 1982). In any case it must be considered much more than an essay into lighting effects that aligns the painting with the amusements of the magic lantern, or the drawing room pastimes of contemporary society. Indeed, pointing to an altogether more serious purpose, Elizabeth E. Barker has remarked that the painting “appears deeply engaged in the discourses of its age involving representation, replication, and spectacle”. And referring to the painting’s “plausible fictions,” she notes its “breath-taking verisimilitude” is challenged by aporias of space and “disjointed perspective […] creating a mood of fractured intimacy [that] destabilizes any lingering illusions of reality” (Barker 2000, 34).

26This seems exactly right, but it may be that this is precisely the effect Wright aims to evoke. Denied the experience, as we all are, of estimating the span of the heavens when confronted with the real thing, Wright’s orrery serves as a model, however inaccurately, to measure the infinite expanse of the universe. In truth the model in no way prompts us to visualize planetary space; rather it reminds us that what we see on the tabletop model is just that – a schematic and highly unrealistic model that cannot represent the solar system without a concerted effort of the imagination. If we consider the painting from this point of view, it is less about representing an external reality than limiting the spectators’ opportunities to see beyond the legible spaces of the composition. In this way the Orrery exposes the difficulties of attempting to use a mechanical model to represent spaces which are literally unimaginable to the human mind. What Immanuel Kant has to say in the Critique of the Power of Judgment is pertinent here:

The infinite, however, is absolutely (not merely comparatively) great. Compared with this, everything else (of the same kind of magnitude) is small. But what is most important is that even being able to think of it as a whole indicates a faculty of the mind which surpasses every standard of sense. (Kant 2000, 138)

27What this might mean for Wright’s painting is not hard to guess. Whether encapsulated as a still-life or attempting, as in the Orrery, to point towards what we cannot see, the representation of things, as Peter Schwenger remarks in The Tears of Things, “always remains […] a longing towards what must always remain inaccessible” (Schwenger 2006, 40). In this sense, the orrery tries, and necessarily fails, in its attempt to represent the unrepresentable. Just as the model solar system carries its audience to the brink of understanding only to leave them in the dark about the actual dimensions of our solar system (itself an infinitesimally small part of the known universe), so too the Orrery teases us with its seemingly transparent representation of space, only to deny us the possibility of reconciling the boundlessness of extra-terrestrial space with the framed limitations of a pictorial representation.

28It is precisely the difficulty of resolving the differences between the motif in nature and its representation in painting that Paul Cézanne addresses in Woman with a Coffeepot (see fig. 7). In place of Wright’s model solar system, hedged around by the lecturer’s small audience, Cézanne posits a radical disjunction between the sitter and the means by which we enter the representation. Cézanne has posed his model slightly off-center, creating a destabilization that leaves the composition searching for balance. The woman, impassive and seemingly unengaged, occupies the greater part of the pictorial field. No gesture troubles the implacable bulk of her person, and despite her leftward bias, she remains as immovable as the Leaning Tower of Pisa. If, then, we consider the way the painting presents itself to our understanding, we note that, for all that Cézanne has arranged the composition in a conventional way (no change is necessary to the spatial structure for the painting to stand alongside any number of seated portraits), it denies a point of access to the viewer. In this sense Cézanne’s apparent indifference to the subject, her apparent inaccessibility, is deceptive. Ease of access has been sacrificed in the interests of pictorial presence.

29This is not to argue that Cézanne has neglected the specificity, or humanity, of his model, or treated her as the visual equivalent of inert matter. As Joseph Rishel has noted, while nothing in the painting suggests an overtly “literary” complexion, Cézanne's focus on his model shows how obsessively close his analysis is (Cézanne 1996, 403). It is here that Cézanne forces us to acknowledge spatial contradictions that cannot be equated with distortion, yet are functionally contiguous with complicating the reading of the image. Specifically, Cézanne demands that Woman with a Coffeepot should not be seen as the meeting of subject and beholder at the site of the picture plane, but the sundering of the two into irreconcilable spaces. The painting is the starting point for an open-ended exploration of form, whereby the qualities of the medium are in perpetual combat with the dominant spatial paradigm. If we examine the woman's torso, we see that Cézanne had made her arms of different length, contrary both to what we may assume to be the case, and the expectations of portraiture. A more conventional artist (that is, one who employs the accepted norms of representation), would have “corrected” this mismatch; but Cézanne has avoided such conventions to leave his sitter – like the oddly elongated coffeepot and overlarge cup and spoon – contingent upon our reading of the painting in a manner that Roger Fry called its “sense of monumental gravity and resistance – of something that has found its center and can never be moved” (Fry 1927, 72).

30By “resistance” Fry has hit upon the most significant aspect of Cézanne’s pictorial practice. The subject of the painting is not the woman, or for that matter the coffeepot, but the process of painting. Significantly, when Fry seeks a comparison in the history of art with Cézanne’s method, it is Rembrandt he thinks of – the Rembrandt of the later paintings in which the surface at once brings into being the subject as an entity that must be studied as a tangible yet never-quite-graspable whole. Fry’s comment that Cézanne’s compositions demonstrate “something that has found its center and can never be moved” is echoed in Lionello Venturi’s remark that the Woman with a Coffeepot “gives the impression of a grandiose force of nature. She is firmly rooted like a powerful tower” (Venturi 1936, I.60). Cézanne’s radicalism lies less in these formal distortions that imply an anti-humanism that does not accord with what we know of his ardent and tempestuous character, than in his substitution of the complexity of the spatial relation between spectator and painting with an equally complex engagement with the problematics of surface.

31The most pertinent observation we can make about this painting in relation to the production of space, is that instead of the contiguity of space that the visual pyramid of perspective implies, Cézanne disconnects the Albertian window from the object of representation, replacing it with a problematized, opaque, screen. How is this accomplished? If we look at the still life of coffee pot, cup, saucer, and spoon, the first thing we notice is that the overlarge spoon, with handle proffered invitingly upwards, is after all, not so easy to grasp. This is because the spoon’s relationship to the coffee cup is itself ambiguous. Is the cup sitting securely upon its saucer, and if so, why is the saucer apparently tipping towards the viewer when the coffee pot is not? That such visual inconsistencies are paradigmatic of Cézanne’s later work goes without saying; his paintings abound with this kind of visual disjunction. But let us ask, from the point of view of Cézanne’s approach to the motif, why this might be so.

32In the first place the painter has refused himself, as he has refused us, a privileged position from which we might view the totality of the thing depicted. The work, despite the attention to the composition, remains fragmented, incoherent, a denial of the compositional norms dominant for the previous five hundred years, effectively returning to painting an autonomy – a rupture – between the referent and the beholder (this is perhaps what Anne Friedberg means when she comments, in relation to the cinematic image, on “our physically embodied and subjectively disembodied relation” to the representation before us, Friedberg 2006, 7). Her “subjectively disembodied” remark is significant; it fits Cézanne perfectly. The continuity of space between beholder and the motif is disrupted, forced into an acknowledgement of a perceptual break where the painting’s physical presence blocks the illusion of connection. In place of contiguity, Cézanne substitutes a troubled reading of surface, in which he presents us with an absence of narrative that ultimately leads to an unfathomable complexity of depiction, imbricating us, not in the “spaces” of a story, but in the labyrinth-like web of spatial clues that oppose the eye to the mind, and seeing to knowing.

33On the face of it this argument may appear to be a rather mechanistic one, an unproblematic development from the depiction of illusionistic space to a confrontation with the picture plane without engaging in contradiction along the way. Of course, this is not the case, and complexity and contradiction are, we might say, the hallmark of Cézanne’s painting. But as a way of considering what Serge Guilbaut has called modernism’s “critical/subversive stance, the negative side of a new culture which based its realizations on a coefficient of resistance to the prevailing system,” it suggests a link with the pictorial abstractionists of the mid-twentieth century (Guilbaut 1983, xi).

34In the final paragraph of his 1948 essay, “The Sublime is Now,” Barnett Newman rejects “the obsolete props of [an] outmoded and antiquated legend” in favor of a “self-evident” reality that derives from the expression of “our own feelings” in which “memory, association, nostalgia, legend, myth” have no place (quoted in Harrison and Wood 2003, 581). From this point of view, we might conclude that Newman returns us to an earlier argument about the nature of visual representation and the deleterious effects of subjecting the spaces of art to an unreasonable restriction within the confines of traditional picturing. He later recalled: “In 1940, some of us woke up to find ourselves without hope – to find that painting did not really exist [. . .] It was that awakening that inspired the aspiration [. . .] to start from scratch, to paint as if painting never existed before” (Bois 1993, 187). Taken together, these comments suggest that Newman was searching for what, in his essay “Perceiving Newman,” Yve-Alain Bois calls, “a new beginning, for a new origin of painting,” that would emancipate him from the constraints of “myth and legend,” and the demands of a form of painting that, modernism’s advances notwithstanding, was still under the empire of the quadrilateral frame (187).

35Newman’s insistence on the need to avoid what he calls the “episodic” in painting denies the Albertian window its place as the “natural” presentation of pictorial space, or for its role in determining the position of the spectator, insisting instead on their “specific and separate embodiments of feeling, to be experienced, each picture for itself.” As if to drive the point home, Newman argues that he is an intuitive painter who seeks to express himself directly onto the canvas, free from the constraints of preparatory sketches, planning, or the working out of compositional problems ahead of the execution of the work (190). In a comment that strikes at the heart of the perspectival construction of pictorial space, he remarks: “It is precisely this death image, the grip of geometry that has to be confronted. In a world of geometry, geometry itself has become our moral crisis” (189-90).

37Newman’s brief viewing instructions do not quite succeed in resolving all the paradoxes of spatial construction that Vir heroicus sublimis raises. In a black and white photograph of the painting when it was exhibited at Bennington College, Vermont, in 1958, a man and a woman, seen from behind, stand in front of the painting. The woman has seemingly followed Newman’s advice and stands about eight feet from the canvas, slightly right of center, while the man, at left, is closer to the painting and views it obliquely, with a raking gaze. As Newman intended, both the closeness of the viewers and their choice of viewing positions militate against seeing the canvas as a window. As Yves-Alain Bois remarks, such a viewing strategy “force[s] [us] to relinquish our mastery over the visual field” (Foster 2006, 365). By this remark, I take it to mean that Bois considers the perspectival clues present in most pre-modern paintings to be radically absent in Newman’s case. If so, then Newman’s remark on his earlier Onement I (1948), a relatively small canvas with a vertical stripe of cadmium red dividing the painting into two equal halves, suggests a definitive break with traditional picturing: “Up until that moment I was able to remove myself from the act of painting, or from the painting itself. [Before], the painting was something that I was making, whereas somehow for the first time with this painting the painting had a life of its own.” In another interview Newman explains: “I felt that for the first time for myself there was no picture making. That stroke [of cadmium red] made the thing come to life for me” (Schreyach 2013, 372-73). It is as if Newman wanted to create paintings without the controlling presence or active input of the artist. As Tom Hess reminds us, Newman studied Onement I for some eight months before deciding he had achieved what he was looking for – a “conversion,” a step away from the norms of pictorial practice to search for an “origin” of painting (Hess 1971, 51).

38But what might that origin be? I take it to mean a search for an originary truth, something akin to the nineteenth-century desire to return to the painters who preceded Raphael in the interests of a purity of perception, free of the kind of the conventions that a system like perspective inevitably introduced. As Paul Crowther has argued, Newman’s anti-subjectivism should be read as a strategy of distanciation in which the material presence of the painting demands acknowledgement of its untranslatability with respect to “mundane reality and the pleasures of sense towards self-understanding acquired in the face of the unknown” (Crowther 1984, 53). In aiming for “wholeness or fulfilment,” Newman’s Vir Heroicus Sublimis consciously opposes the spatial assumptions of the western tradition, in which linear perspective and the totalizing frame open up a bounded viewpoint that situates the beholder in relation to the image field as in an implied three-dimensional space:

I was constantly concerned in doing a painting that would move in its totality as you see it. You look at it and you see it [. . .]. Otherwise, a painter is a kind of choreographer of space, and he creates a kind of dance of elements, and it becomes a narrative art instead of a visual art.” (Schreyach 2013, 373)

39Yet Newman is not necessarily looking to negate pictorial space as to re-invent it, making the apprehension of the pictorial a boundless, unframed, experience. Previously, the representation of space may be said to imprison the spectator, confining him/her to a fixed relationship to the picture plane, and to seeing in the illusion a solid reality, somewhat in the manner of Plato’s cave allegory, where the captives are forced to accept shadows cast by firelight as reality, and reality as shadows (Plato 2004, 208-12). In contrast, Newman wants to reveal a space that was “real” in the sense of free of the conventions of perspectival space. A clue to Newman’s intentions perhaps lies in the title, Vir heroicus sublimis. While “Man, Heroic and Sublime” is a literal translation of the Latin, Newman considered it to mean something rather more, glossing a conditional clause “that man can be or is sublime in his relation to his sense of being aware” (Schreyach 2013, 354; O’Neill 1990, 258). By this, Newman is referring to our awareness of the gap between perception and truth, leaving the displaced spectator adrift in a sea of ambiguity that can only be overcome by recognition of the unequal relationship between what we see and what is. Sublimity, then, is not a given, but depends on our sense of the thing itself, being something more than an imitation of the world. That space is crucial to this interpretation is clear in a comment Newman makes in 1962, “Anyone standing in front of my paintings must feel the vertical domelike vaults encompass him to awaken an awareness of his being alive in the sensation of complete space” (O’Neill 1990, 249-50). This suggests that, in contrast to the bounded space of traditional picturing, where an imagined window defines the space available to the observer, the spaces of the sublime are without borders, free of boundaries that would otherwise confine them to the limits of the pictorial.

40The question remains to be answered is whether Vir heroicus sublimis succeeds in its task of emancipating itself from the straightjacket of pictorial tradition to reveal a hitherto unexplored sense of space. Here we are back with Holly’s question, “What pictorial (decidedly nonhistorical) aspect might encourage [these paintings] into silent conversation?” (Holly 2019, 894-95) To formulate an answer, we might consider Kant’s assertion that “The beautiful in nature concerns the form of the object, which consists in [the object’s] being bounded. But the sublime can also be found in a formless object, insofar as we present unboundedness, either [as] in the object or because the object prompts us to present it, while yet we add to this unboundedness the thought of its totality” (Kant 1987, 98). Holly is not talking specifically of the sublime, but the sublime’s principal characteristic – to transcend boundaries that limit or contain our perception – seems well suited to her concern: “How do backgrounds, especially, summon images into presence? How does visibility offer clues to the invisible?” (Holly 2019, 895)

41It is this sundering of narrative art from visual representation that separates Newman and the abstractionists from earlier forms of picturing. As Allan Kaprow writes in 1963 about Vir heroicus sublimis, “the vast sea of intense red […] fills the space in front of the canvas, surrounding the viewer [. . .] with the relentless hue.” In similar vein, Lawrence Alloway speaks of “the irradiating surface” that expands “the paintings threshold into the area occupied by the spectator” (Schreyach 2013, 366). Instead of treating the picture plane as a window, Newman treats it as an opaque screen onto which images are projected, where light radiates back into the beholder’s space, both illuminating and occluding the truth of the relationship between spectator and image.

42I want to end by repeating what I implied earlier, but did not fully clarify – the spaces of art may be approached in an infinite number of ways, but two predominate in the post-Renaissance period. The first, Alberti’s “window onto the world” that mimics the appearance of things (so long as we accept this approach to space as merely a technical solution), offers a finite solution to the problem of representation. It solves the problem of how to represent space so that narrative action may be represented in all its complexity, borrowing its structure from Aristotle’s Poetics that, in history painting at least, allows for tragic action that looks back to the drama’s preliminaries and forward its consequences while focusing on the key moment (catharsis). The second approach is less literary, but more philosophical, seeing space as the place where art, as a category of experience, comes into being. In this sense artists like Newman are questioning the limits of painting, and consequently the extent to which the spectator’s role the construction of meaning in art must be taken into account. In this sense, whether we accept the premises on which one or the other of these systems is based matters less that the realization that what is a stake is not their putative verisimilitude or otherwise, but their role in the establishment of the boundaries of art, of which the framing of space is a foundational part.