Johns Hopkins University Press
  • Diagrammatic Formats from Page to Wall: Dante and the Strozzi Chapel, Revisited

In the 1350s, Nardo di Cione painted a fresco in Santa Maria Novella in Florence that blends narrative, allegorical, and diagrammatic modes of representation. Nardo’s fresco (fig. 1) is a giant vision of Hell, now badly faded, painted in the Strozzi Chapel in the mid-1350s as part of a larger pictorial program that included a Last Judgment, a vision of Paradise, and his brother Andrea’s famous “Strozzi Altarpiece” of Christ Enthroned.1 Nardo’s Hell is an example of a new type of painting that took shape over these years: a moralizing allegorical construction, devised in large scale for the edification of viewers, that placed narrativizing details in the service of a fundamentally diagrammatic painting. Nardo’s painting is a map-like representation of shallow fictive spaces that unfold across the painting’s surface rather than within an overarching pictorial space. Critically, Nardo made the decision to omit a clear protagonist, a figure that would have moved the painting in the direction of true pictorial narrative, choosing instead to focus on the place itself—its structure, the people and locations encountered along the way, and its moral and edifying message. As will be shown below, the actual content and message of the fresco have been well-established by scholars for decades. But most of the painting’s interpreters have overemphasized the narrative and directional qualities of the painting at the expense of its diagrammatic structure and rhetoric. If we understand the painting instead as a monumental diagram, how does that shift our understanding of how it may have communicated with and impacted its viewers? [End Page 24]

Figure 1. Nardo di Cione, Hell, Capella Strozzi di Mantova, Santa Maria Novella, 1350s. Photo: Wikimedia Commons.
Click for larger view
View full resolution
Figure 1.

Nardo di Cione, Hell, Capella Strozzi di Mantova, Santa Maria Novella, 1350s. Photo: Wikimedia Commons.

[End Page 25]

Despite its damaged and faded surface, Nardo’s Hell has long been well-known for a critical reason: it is the first surviving monumental Hell scene to copy, relatively faithfully, the structure of Hell invented by Dante in the Commedia about forty years earlier in 1310. This pictorial reference to the Inferno would have been easily legible to Florentine viewers at midcentury, when Dante’s text was widely read in the city. In his painting, Nardo seems to have carefully balanced two goals. He certainly aimed to design a representation that depicted the structure of Hell invented by Dante in Inferno, and was quite self-conscious about its correspondence to that text. But his painting also had to fulfill the function of a typical Hell scene within the larger program of the Last Judgment, and thus also followed and responded to earlier versions of such paintings. While Nardo’s fresco has been studied almost exclusively within the context of these two traditions (Dante “illustration” and the broader traditions of monumental Hell scenes), I’m interested here in the question of how Nardo’s Dantean vision of Hell intersected with other visual traditions at the time that similarly rendered allegorical and diagrammatic subjects.2

What exactly does the phrase “diagrammatic painting” imply? And how does describing Nardo’s painting as diagrammatic help us to see it differently? The thirteenth and fourteenth centuries witnessed a revival of monumental painting that had not been seen in Italy since the Roman Empire. But not all of these paintings were the Giottesque narrative frescoes and devotional panels so beloved by modern scholars and audiences (and by Renaissance writers like Vasari). In dozens of monumental examples across central and northern Italy, artists also created complex diagrammatic paintings whose content was conveyed not through proto-perspectival spaces but rather through circles, trees, hierarchical stemmata, and winding pathways.3 While there is a wealth of scholarship on some of the individual iconographic traditions, such as paintings of the Tree of Life, monumental diagrammatic painting has too rarely been studied as a sustained practice of image-making. Part of the reason for this gap is that numerous diagrammatic or otherwise non-narrative paintings by Giotto, the most famous artist of the period, have been lost, lending the impression that such paintings lay outside of the mainstream, as idiosyncratic “one-offs” or pictorial dead ends.4 In reality, diagrammatic painting emerged during the Trecento across a wide range of contexts as a fully formed pictorial mode with its own [End Page 26] unique visual and communicative strategies. In addition to narrative, devotional, and allegorical painting, with which it intersected, the diagrammatic mode was one of a number of options available to artists, patrons, and planners, favored especially when the goal was to present complex content to viewers.

Figure 2. Andrea di Bonaiuto, Triumph of Thomas, Spanish Chapel, Santa Maria Novella, Florence, 1365–67. Photo by the author.
Click for larger view
View full resolution
Figure 2.

Andrea di Bonaiuto, Triumph of Thomas, Spanish Chapel, Santa Maria Novella, Florence, 1365–67. Photo by the author.

A concrete example of these large painted diagrams can be found within Santa Maria Novella itself, off the adjacent Chiostro Verde and only steps from the Strozzi Chapel, in the complex’s chapter house now known as the Spanish Chapel. Here, a painting by Andrea di Bonaiuto da Firenze identified in modern scholarship by many different names, including “The Triumph of St. Thomas Aquinas,” “St. Thomas Enthroned,” “Allegory of Christian Doctrine,” and “Allegory of Christian Learning,” unfurls its content across a clearly diagrammatic surface (fig. 2).5 Thomas Aquinas sits enthroned at its center with a book opened for the viewer; above him fly personifications of the four cardinal virtues and three theological virtues: Justice, Fortitude, Prudence, and Temperance below, and Hope, Faith, and Charity above. Seated to either side of Thomas on a shallow bench are a collection of Biblical figures: moving from the viewer’s left are Job, David, Paul, [End Page 27] Mark, and John the Evangelist to one side, and Solomon, Isaiah, Moses, Luke, and Matthew on the other. On a small ledge that protrudes from beneath Thomas’s feet, disrupting the image’s flatness, crouch three defeated heretics: Sabellius, Ibn Rushd (Averroes), and Arius. Finally, in the bottom half of the painting sit twenty-eight figures in two rows: above are female personifications of the seven “Liberal Arts” and the seven “Theological Sciences,” and below sit historical practitioners or exempla of each. Taken together, the figures of the painting lay out the different kinds of wisdom and virtue that lead to the wisdom embodied by Thomas at the center, placing the Dominican order at the center of Christian learning. The meanings conveyed by the diagram’s painted and textual elements are complex, but its form is hierarchical and orderly. Such a vision of theological and philosophical order might at first seem better suited to the context of a manuscript than to the wall of a chapter house, but it is only one example of numerous such diagrammatic paintings, which created meaning not through narrative means but by harnessing the inherent authority that diagrammatic forms had long conveyed in the manuscript world, and by magnifying its impact through monumental scale and proximity to other holy images.

The diagrammatic mode of painting is constituted through certain common features, among them the inclusion of extensive painted texts, the conception of the picture as a surface as well as a space, and the conditional or relational arrangement of content. When we explore how Trecento viewers engaged with diagrams both miniature and monumental, we can see how these features began to shift practices of visual communication, eventually affecting the production and reception of images well beyond strictly diagrammatic examples. In this way the diagrammatic becomes an expansive and flexible category rather than a narrow one.6 It is in this expansion of “the diagrammatic” beyond a narrow set of categories like trees, circular diagrams (rotae), and stemmata that we will find a new way of looking at Nardo’s fresco of Hell. I return later in this essay to the paintings of the Spanish Chapel as contemporary and comparative examples through which to understand these new modes of painting and how they operate in Nardo’s Hell, but first the painting and its context must be more thoroughly introduced. [End Page 28]

The Strozzi Chapel

While documents regarding the construction and patronage of the Strozzi Chapel have not survived, it was already referred to as such by 1359, at which time the Strozzi family had likely already been involved as its patrons for several decades.7 The present architectural form of the chapel, an elevated space at the end of the west transept, was probably constructed in the 1330s and early 1340s on foundations of the Carboni Chapel in the Chiostro dei Morti, which lies below it (fig. 3).8 Frances Lee Pitts has concluded that its patrons were the Strozzi family sons of Rosso di Gerio, who gained burial rights to the space in the late 1340s; several members of the family’s next generation, most likely including Fra Pietro Strozzi, subsequently commissioned Nardo’s fresco program. Pitts discusses at length the connection between the Strozzi family and the use of Dante as a source, pointing out that such a use begs explanation, since there is little to suggest that Dante was popular among the Dominican clergy at Santa Maria Novella at that time.9 The clearest evidence on the matter comes from a document of 1335, about fifteen years before the fresco program. It was passed at a general meeting of the Dominican provincial chapter, which was held that year at Santa Maria Novella and presided over by Fra Pietro Strozzi, who was then head of the Provincia. Among other prohibitions, the document included a ruling forbidding the monks to read Dante. The language of the prohibition, however, suggests that Dante’s poetic works were not viewed as heretical or sinful but simply as distracting, leading monks away from the study of more proper “theology,” and, indeed, the only punishment for offenders was that the books would be taken away.10

While the popularity of the Commedia has long been noted among the laity and early humanists, the need for such a prohibition at all indicates the popularity of Dante among some of the monks, too. Pitts argues that it was almost certainly through the desires of the secular members of the Strozzi family that the Dantesque imagery entered the chapel. The Strozzi are documented in the eighteenth century as having numerous early illustrated copies of the Commedia in their library, including a lavish manuscript from the 1340s, possibly revealing their interest in the text in the years just before the fresco’s commission.11

Nardo’s Hell was painted on the chapel’s north wall, from the base of the dado at the bottom to the vaulted ceiling, a height of over thirty-three [End Page 29]

Figure 3. Capella Strozzi di Mantova, Santa Maria Novella, Florence. Photo: Wikimedia Commons.
Click for larger view
View full resolution
Figure 3.

Capella Strozzi di Mantova, Santa Maria Novella, Florence. Photo: Wikimedia Commons.

[End Page 30]

feet. Its surface is divided into sections using a series of rocky ledges, almost like shelves, each populated by small figures and vernacular inscriptions. Viewers familiar with the structure of Dante’s poem would quickly infer that the implied progression of the painting proceeds from top to bottom, with Limbo at the top of the painting and Satan at the bottom, but nothing in the painting itself expressly signals this idea of progression—there is no single path to be followed, and in the narrow space of the chapel the fresco’s lower half is much easier to see than its upper half. While in Dante’s poem Hell is divided into nine circles, neither the poem nor the painting gives equal weight to each circle, and numerous subdivisions in some of them make it difficult to grasp the fresco with complete clarity.

The painting is divided into two halves by the ring of fire at its center. The ring is emphasized by its arresting color, central placement, and the fact that it is the only one of the rings in the painting to extend unbroken from the left edge to the right. This is circle six, the place of the heretics, depicting figures encased in burning tombs, and the emphasis on it in the painting mirrors the Dominicans’ concern with heresy and the formation of their collective identity through fighting it. The bands above this central ring correspond to circles one through five: Limbo is at the top in the wall’s lunette, and the area below it is roughly divided into two concentric rows, each with a diagonal division across the center, creating zones for circles two through five, viewed from left to right and top to bottom: Lust, Gluttony, Avarice, and Wrath. Below the red zone of the heretics corresponding to level six, the painting’s structure changes again. A curving horizontal ring beneath the zone of fire is broken into three compartments, but rather than corresponding to separate circles of Hell (as was the case above) these compartments delineate the three subdivisions of the seventh circle: violence against self (suicides), violence against neighbor, and violence against God (blasphemers, sodomites, and usurers). Below this, the bottom third of the fresco contains the eighth and ninth circles: the “Malebolge,” where different kinds of fraud are punished in ten separate ditches, or bolge, and the ninth circle, a frozen lake with different categories of the treacherous and, at the very bottom, Satan. In Nardo’s painting, the sequence of the bolge alternates back and forth across the painting’s center, from the middle towards the lower corners. These zones are smaller than those above but would have been more easily legible because of their proximity to the viewer. [End Page 31]

Figure 4. Nardo di Cione, Hell (Detail of the Gluttunous in Circle 3), Capella Strozzi di Mantova, Santa Maria Novella, 1350s.
Click for larger view
View full resolution
Figure 4.

Nardo di Cione, Hell (Detail of the Gluttunous in Circle 3), Capella Strozzi di Mantova, Santa Maria Novella, 1350s.

Photo by Genevra Kornbluth, with permission.

Nardo’s vision of Hell is thus highly and very intentionally organized, but in interesting ways that sometimes correspond to the poem and sometimes do not. The poem, for example, describes Hell as funnel-shaped, with each ring getting smaller as you go down, and indeed this kind of shape is usually followed in strict “illustrations” of the Inferno found in manuscript copies of the text, such as the famous map of Dante’s Hell drawn by Botticelli around 1485. One gets the sense that Nardo rejected the use of this structure because he wanted to be able to use the whole surface of the wall for his painting, rather than losing part of it to the negative space that would have resulted from narrowing Hell as he moved down. Indeed, there is a great tension in the work as a whole between its schematic structure—which pushes viewers backwards in space both physically and metaphorically as they appreciate the its highly ordered arrangement—and the emotional intensity of some of its details. Both because of their small scale and their lack of bright color (largely owing to subsequent damage to the painting’s surface), the poignancy of these details is easy to overlook. But gazing, for example, at the representation of the third circle in the upper right, we see the [End Page 32] intense emotionality of the gluttonous (fig. 4). Tormented by Cerberus, the embodiment of their sin who tears at and devours their flesh, the sinners writhe and twist. To the right of Cerberus, other gluttons are denied nourishment at a table set with food and wine; a prominent female figure raises her arms in an extravagant gesture of grief, mimicking visual imagery of women weeping at Christ’s crucifixion. Like other diagrammatic paintings of the period such as Pacino di Bonaguida’s Tree of Life (fig. 5), Nardo’s painting counterbalances its elaborate diagrammatic structure with these smaller emotive details.12

Several authors have noted Nardo’s critical decision not to include Dante and Virgil in the painting. That omission, which I return to below, fundamentally changes the fresco, ensuring its status as an image of Hell inspired by Dante rather than a literal illustration of the poem itself. The painting both had to function as a Hell scene within the greater program of the chapel and had to be recognizable as such in relation to earlier famous paintings of Hell, from Giotto’s Arena Chapel to Buonamico Buffalmacco’s paintings in Pisa’s Camposanto and Orcagna’s largely lost Triumph of Death from Santa Croce, which included a Last Judgment. Visual representations of Hell had long been characterized in European medieval art by their disorganization, their chaos often heightened through a juxtaposition with the order and visual clarity of heaven. On the sculpted tympana of French twelfth-and thirteenth-century churches, for example, depictions of the Last Judgment such as the one at Conques contrasted orderly, upright rows of standing figures in heaven with contorted, twisting masses of bodies in Hell. Giotto’s Last Judgment in the Arena Chapel similarly contrasts rows of calm, monumental heavenly figures with the entangled smaller figures and demons in Hell, a model also seen in the mosaic of the Last Judgment in Florence’s Baptistery of Saint John. In the wake of Dante’s poem, however, depictions of Hell became increasingly organized, even when they did not exactly mirror the structure of the Inferno. The best example is Buffalmacco’s Hell (fig. 6) in the Camposanto, part of the larger series of paintings that include a Triumph of Death and Thebaid (stories of the Desert Fathers). The depiction, part of a larger Last Judgment now usually dated to around 1335–40 (formerly attributed to Francesco Traini, among others, and dated much later), similarly divides the space of Hell into a series of rocky compartments. The organization of the space into eight discrete zones surrounding the central devil-figure [End Page 33]

Figure 5. Pacino di Bonaguida, Tree of Life, Florence, Accademia, ca. 1310–15. Photo by the author.
Click for larger view
View full resolution
Figure 5.

Pacino di Bonaguida, Tree of Life, Florence, Accademia, ca. 1310–15. Photo by the author.

[End Page 34]

Figure 6. Buonamico Buffalmacco, Hell, Camposanto, Pisa, 1335–40. Photo by the author.
Click for larger view
View full resolution
Figure 6.

Buonamico Buffalmacco, Hell, Camposanto, Pisa, 1335–40. Photo by the author.

creates a map-like view of Hell’s space, in contrast to the disorder of earlier representations. Still, while Buffalmacco’s act of division may be inspired by Dante’s description of Hell’s bolge, the overall organization of the painting, including how the torments are divided, lacks a systematic correspondence to Dante’s poem.13

Nardo’s Hell: Diagram, Narrative, Allegory

How did Nardo intend for a viewer to navigate his depiction of Hell? The painting operates at the nebulous intersection of narrative, allegory, and map, diagram, or route, each of which might prompt its own conventions for viewing. Yet the painting belongs to none of these categories completely. Modern viewers, Dante’s poem firmly in mind, are inclined to read the painting as narrative (and, indeed, there is much evidence to suggest that the reference to the Commedia would have been on the minds of many fourteenth-century viewers as well). Narrative [End Page 35] paintings are almost always described sequentially in art-historical writing or verbal description: we could imagine a description reading something like, “The viewer’s eye begins at the top, in Limbo, and works its way downward through the various levels, led by the diagonals of the rocky divisions.” But this kind of art-historical language, which ushers the viewer’s eye along a set path, ignores the insistent return to a macro-view that diagrammatic paintings provoke: the rocky ledges and the ostentatious visual organization of the scene are a core aspect of its meaning, not just a frame or setting. Without the presence of Dante and Virgil as surrogates advancing through a narrative journey, and without clear roads, ramps, pathways, or stairs depicted between the painting’s various levels, there is really no indication that such a progression is intended by the artist. The viewer, medieval or modern, brings with them this idea of progressive movement from their knowledge of Dante’s poem; in this way the painting serves almost as a table of contents. But the painting itself is much more of a map or diagram of Hell than it is a depiction of a journey through Hell’s spaces.

This does not detract from its ability to impact viewers, however. Diagrams in the fourteenth century were not just technical, emotionless abstractions, and numerous Italian artists of the period explored the potential effects of constructing large-scale diagrammatic paintings. For instance, a massive diagrammatic painting of the Creation in Pisa’s Camposanto, painted several decades after Nardo’s Hell, displays a typical cosmology of the period, with the earth at the center surrounded by concentric rings of the four elements, seven planetary spheres, the zone of the fixed stars showing the zodiac, the zone of the prime mover, and the hierarchies of angels (fig. 7). As I have argued elsewhere, such an image has a different status and effect than a similar cosmological diagram in a manuscript; in its monumental version, it is an imposing image of order, its harmonious geometric design likely intended to evoke awe and wonder.14 Nardo’s painting, too, creates a vision of order, yet one intended to provoke wonder, fear, and other emotional responses. Diagrams can function as demonstrations, in this case revealing that Hell, in Dante’s wake, could be depicted as a strictly regimented place, proof of the intention and design behind all parts of creation. Like the cosmic painting in Pisa, it proposes a visual order for something fundamentally unseeable. What makes Nardo’s painting so interesting, though, is how it oscillates between the macro-view that reveals order [End Page 36] and structure and the micro-view that is charged with highly emotional details.15 This strategy evokes works like Pacino’s Tree of Life panel in the Accademia, where the viewer could move between the tiny narratives within the fruits of the tree and the larger appreciation of the order and structure of the work, which uses the form of the tree to suggest the unfurling progression and fulfillment of salvation history. Ambrogio Lorenzetti’s government paintings in Siena’s Palazzo Pubblico are also structured around this oscillation between overarching diagrammatic [End Page 37]

Figure 7. Pietro di Puccio da Orvieto, Creation, Camposanto, Pisa, late 1380s. Photo by the author.
Click for larger view
View full resolution
Figure 7.

Pietro di Puccio da Orvieto, Creation, Camposanto, Pisa, late 1380s. Photo by the author.

compositions of governmental principles and emotionally charged or ambiguous details in the passage of the fresco that demonstrate the effects of those principles on city and countryside.

In Nardo’s painting, this contrast between micro and macro is taken to its limit. The painting teems with hundreds of tiny figures across the expansive wall, figures whose abject torments and highly emotional responses draw the viewer into their world. As the eye scans the wall, different figures stand out each time, competing in their grotesqueness for the viewer’s attention—a far cry from the kind of orderly, sequential viewing that is usually proposed for this kind of image (and which is fundamental to the structure of a text like Dante’s Commedia). Indeed, its lack of a clear narrative or sequence (even though scholars want to read one into it) seems to have contributed to numerous negative assessments of the painting; while scholars celebrate it for being the first monumental painting to recreate Dante’s Inferno, they also make repeated reference to its compositional failings, appreciating its iconography but ignoring it as a work of visual art. Similarly, the different modes of representation used across the chapel’s walls were characterized by many as a “stylistic disunity” rather than an intentional variety of pictorial strategies.16

Of course, some of the ambiguities of how to read Nardo’s fresco (as directional, sequential, or static, for example) are critical issues with the text of the Commedia itself. Scholars have long noted the way in which the poem’s structure is firmly anchored to the place that Dante has invented, and that the movement forward through the story is sometimes reminiscent of narrative progression but other times reads more as an unfolding description of a map or diagram, populated by figures who act as exempla. The text is self-conscious about this fact, most famously in Inferno 11 when Virgil suggests the formation of a mental map of the space the reader is moving through; as Theodore Cachey writes, “to interpret the structure requires a form of hermeneutic mapping that Virgil models, albeit imperfectly, for the reader of the poem.”17 Parker, making a similar point, writes that Inferno 11 “provides us with a moment in which Dante’s temporal perspective becomes spatial.”18 Nardo’s painting, then, is almost like a pictorialization of the mental map that Virgil invites the reader to create. But it is not only this—if it were, Nardo could have painted just the structure of Hell without its affective qualities. The painting, instead, combines the abstract vision of the structure that Virgil encourages readers to create in their minds [End Page 38] with the charged emotional details of the poem’s individual cantos (also inspired by the elaborate torments of earlier monumental Hell paintings by Giotto, Buffalmacco, and others). What the painting can’t do, however, is depict all of the personalized links to real people and places that Dante’s poem is so famous for. While some individuals from the poem could likely be read into the image by a viewer who was highly familiar with its text, Nardo tends to maintain a level of generality that would have made the fresco more accessible to all kinds of viewers. Rather than needing to know the exact stories of the personal encounters Dante describes in the poem, the painting’s vernacular inscriptions instead refer to the general torments found in each circle. The wrathful and slothful of the fifth circle, for example, are accompanied in the painting by the simple Italian inscription, “Qui sono puniti gli iracondi e accidiosi,” or “Here the wrathful and slothful are punished.” Such generality, while still corresponding to the canto in Dante and often including small scenes that reference particular characters, maintains the painting’s status as both a painting of Hell and a painting of Dante’s Hell, an interweaving of two traditions which speaks to the interests of the different groups involved in its commission and execution: the Dominican clergy of the church, certainly, but also the lay viewers whose family members were buried in the chapel.19

The painting may address and engage diverse audiences, but to what end? The painting harnesses a diagrammatic structure—strict compositional divisions, lack of narrative direction, and extensive texts and inscriptions—to create a moralizing argument. This idea of the argumentative image is fundamental to these kinds of allegory/narrative hybrids in fourteenth-century painting, differentiating them from diagrammatic images whose function is more straightforwardly mnemonic. Such moralizing is not completely novel, of course; both Hans Belting and Lina Bolzoni have pointed out the links between this new argumentative painting and traditions of rhetoric in preaching and theology, particularly of the Dominicans. Belting, for example, writes that in these narrative allegories, “narrative becomes instrumental, not an end unto itself . . . it tends to ‘argue’ with the beholder.”20 The details that Belting describes as “narrative” are placed in the service of the larger argument made by the allegory—they are “backdrop” or “evidence” for the work’s primary meaning, which he argues is a critical shift away from longstanding understandings of narrative in the Christian [End Page 39] tradition as an end unto itself. But Belting does not use the word “diagrammatic” to describe the overall structure of paintings like Nardo’s, and I think he overstates their narrative qualities. Without protagonists, there is truly very little about Nardo’s painting that is “narrative.” To my mind, the reception of Nardo’s work aligns it more closely with the way Bolzoni believes Buffalmacco’s Triumph of Death and Last Judgment in Pisa would have operated: focusing less on qualities of narrative and more on the tradition of argument through exempla.21 Bolzoni’s interpretation of Buffalmacco’s Camposanto fresco, which, though not directly inspired by Dante, still shares many features in common with Nardo’s fresco, including the extensive inscriptions and highly organized spatial divisions, calls particular attention to the textual portions of the painting, arguing that they reveal similar strategies to preaching (and, indeed, were likely designed to be viewed in tandem with the messages conveyed in sermons and devotional literature). In contrast to Pitts, who argues that the presence of the inscriptions is part of what suggests a narrative reading of the painting, Bolzoni uses their presence in Buffalmacco’s painting to suggest the exact opposite, showing that the language and rhetoric found in such inscriptions is “not one of narrating but of moralizing.”22 Her discussion is complex, and can be summarized only briefly here, but the key is that preaching, text, and image work together and work similarly, not just in terms of subject matter but in terms of structure: “they are linked by a precise rhetorical code . . . that is interested in acting upon the public, building up a set of mental images that will influence and form the soul’s faculties (intellect, memory, will). . . . At the same time, the code allowed for different levels of communication, selecting different messages that one single image could convey, according to the cultural and social level of the recipient.”23 The focus is thus less on creating concrete templates for memory, as diagrams and schemas are so often thought to do, but about communicating moralistic content that encourages reflection and the creation of additional mental images which act in dialogue with visual ones.24

In short, the details presented through inscriptions and within the various smaller areas of Nardo’s Hell are not stories but messages. The moralizing focus of these nuggets is particularly clear in a Last Judgment painting, which is fundamentally about influencing human behavior to guide a viewer towards salvation, but it is a feature found throughout these kinds of diagrammatic painted allegories. Bolzoni and Belting [End Page 40] observe the same phenomenon across a similar range of examples, but while Belting is more interested in looking retrospectively at the shifting function of narrative in these public paintings as compared to earlier periods, Bolzoni is concerned with looking more synchronically at how they worked together with things like preaching to communicate consistently structured messages across a range of media.

As many scholars have pointed out, one of the primary features of allegory in both painting and literature is that it shifts the burden of interpretation onto the viewer.25 In contrast to narrative painting, in which the viewer can function as a witness or bystander, allegory is fundamentally about inviting interpretation, speculation, and testing. In Nardo’s painting, then, what appears at first to be a map of Hell that one should read sequentially, as with Dante’s Inferno, is actually an allegorical prompt for the creation of a mental diagram that the viewer could then traverse and explore, consistently returning to its superstructures rather than following a single path. The painting offers little room for the viewer to enter—there is no space or concrete route in the image for the viewer to follow or occupy, and no images of Dante or Virgil to serve as avatars or guides.26 But as the diagram created by the painting is internalized in repeated, ongoing acts of viewing by those who visited the chapel regularly, and as this mental diagram intersected with the moralizing qualities of the individual scenes (likely informed, in turn, by memories of reading Dante’s text or hearing vernacular sermons), it could then be surveyed at the viewer’s leisure, with the viewer taking on the role of protagonist within an imaginative recreation that can be as narrative or sequential as desired. Manuscript diagrams, and their mental recreations, are often described as “machines for thought,” and these monumental allegorical paintings provide a similar experience.27 Such thought experiences are not completely open-ended, however. These paintings guide their viewers through recurring themes towards particular messages deemed appropriate for the audience. As Bolzoni explains, while these kinds of paintings appeared in a number of different contexts, they were consistently located in urban settings, where the moralizing messages of preachers and paintings were thought by the clergy to be most urgently needed.28

Nardo’s painting, then, has a very different purpose than the more direct “illustrations” of Dante’s text to which it is often compared. Manuscript illustrations of Dante’s text are necessarily bound to the [End Page 41] text’s sequential nature, even in images like that lay out Hell’s entire structure; they also typically include repeated depictions of the poem’s protagonists. In the ca. 1430 manuscript of the Inferno now in the Bibliothèque Nationale in Paris, the artist Bartolomeo di Fruosino depicted on the manuscript’s first folio a diagram of Hell clearly influenced by Nardo’s painting, but with the addition of Virgil and Dante over half a dozen times on the page.29 The two men are shown not only looking onto the scene, but interacting directly with characters across the page, engaged in animated discussion and responses with them and each other. Botticelli’s version is even more extreme: Dante and Virgil are repeated twenty-seven times in the same image, a pictorial device that indicates their movement through its space but is also awkward and repetitive. Parker argues that Botticelli’s map more clearly offers the simultaneous options of narrative reading and the quasi-cartographic totalizing viewpoint. But in Nardo’s painting, the viewer takes on the role of the protagonist (or even of Dante the author, designer of the totalizing view) and has to navigate between the emotive world of the torments and the structure of the whole system. While a reader of Dante’s poem can easily be lost in the drama and details of a particular encounter or description and lose sight of the overall structure, this is less possible in the painting, where Hell’s diagrammatic organization remains ever-present to the eye.

It is a further quality of monumental paintings like Nardo’s that, through scale, the organizational vision of Hell put forth by the artist comes itself to have an emotional impact. Like the cosmic painting in the Camposanto in Pisa, Nardo’s Hell contains a diagrammatic structure so massive that it nearly overpowers the viewer, especially in the small space of the Strozzi Chapel, where it is impossible to back up far away. In comparing this fresco to manuscript illuminations of Dante’s poem, as so many have done, we lose sight of the effect of the painting’s massive scale in shifting the image’s rhetoric. What in a miniature version can be fully internalized is more powerful in the monumental version: the painting’s diagrammatic structure is the visual field through which one encounters the torments of the sinners. It is no longer just an organizational device, but the ground (or surface) on which the content is conveyed. Thus, while it could prompt the creation of an internal image, a diagrammatic structure like Nardo’s can also emotionally impact [End Page 42] the viewer in the moment through its sheer scale and its alternation of between micro- and macro-structures.

Dominican Pathways: The Via Veritatis

Just ten years after the decoration of the Strozzi Chapel, another artist, Andrea di Bonaiuto da Firenze, was commissioned to create a fresco program for the entire interior of the recently constructed chapter house (sala capitolare) at Santa Maria Novella, just outside the west side of the nave on the north end of the Chiostro Verde. Now known as the Spanish Chapel, the space had been built to function as a chapter house for the Dominicans and also as a burial site for its patron, Buonamico Guidalotti, who died in 1355 when it was in the final stages of construction. The fresco program was overseen by the prior, Fra Zenobius Guasconi, who signed a contract with Andrea in 1365 which called for the frescoes to be completed by the end of 1367.

Much of the work on Andrea’s complex and beautifully preserved fresco program has focused on its connection to the multiple, intersecting functions of the room. As a chapter house, it was the site of the daily “chapter of faults” where the friars would hold themselves accountable to their monastic rule; as a funerary space, it perpetuated the memory of Guidalotti and his devotion to Christ and the Eucharist. Lastly, it served as a sacred space where masses were performed. The complexity of the fresco program speaks both to the space’s multiple functions and to the chapter’s elite, highly educated audience—the friars themselves. The chapter house was a place where the identity of the order was proclaimed and advertised. Most scholars agree about the nature of the paintings’ overall theme: they present a history of salvation through Christ and celebrate the role of the Dominican order in carrying out his mission on earth.30 The precise connections between the content of the frescoes and the theology, preaching, and activities of the Dominicans in the mid-fourteenth century have been explored by a series of major scholars, including Eve Borsook, Millard Meiss, Julian Gardner, Joseph Polzer, and Joanna Cannon, many of whom have proposed specific friars who may have been involved in its design, most notably Jacopo Passavanti (ca. 1302–1357).31 I have already briefly introduced the fresco on the chapel’s west wall, the Triumph of [End Page 43] Thomas Aquinas, as a classic example of diagrammatic painting in the Trecento. But the fresco on the chapel’s east wall, an allegorical painting variously known today as the “Path to Salvation,” “Church Militant and Triumphant,” “Glorification of the Dominican Order”—and most commonly as the “Via Veritatis”—makes a particularly interesting comparison with Nardo’s Hell in terms of the visual strategies they employ (fig. 8). Like Nardo’s Hell, I argue, the Via Veritatis is a largely non-narrative and diagrammatic painting, though neither is organized according to a strict diagrammatic schema like those paintings of the Creation, Tree of Life, and Triumph of Thomas Aquinas. While the paintings share certain similarities, it is their key differences that help illuminate the unique structure of the Hell. The Via Veritatis is a purely allegorical construction: a non-narrative fresco not reliant on a single text, populated both by stock figures and by recognizable people. Where Nardo’s Hell maintained a certain generality in its figures and setting despite its close correspondence to Dante, Andrea’s Via Veritatis is highly specific in certain aspects of its setting, including its clear directional path and the appearance of specific characters. Before we can explore how its visual rhetoric compares to Nardo’s Hell, its content must be briefly introduced.

The painting is directional and even eschatological, allowing us to more responsibly describe it in sequential terms. The viewer is guided to begin at the painting’s lower left, where a crowd of figures is gathered in front of a large building, recognizable as Florence’s Cathedral, Santa Maria del Fiore.32 The Cathedral appears to symbolize the universal church, Ecclesia, and the crowd in front of it includes both contemporary individuals and historical ones (scholars differ on how individualized these figures are meant to be), including a pope (likely Benedict IX or Urban V), a cardinal (Niccolo Albertini or Gil de Albornoz), a bishop, artists and authors (perhaps including Arnolfo di Cambio, Giotto, Cimabue, Boccaccio, and Dante), a secular prince and emperor (Charles IV), friars, and laymen. At the feet of the pope are lambs (symbolic of the Christian people) guarded by black and white dogs (the Domini canes, symbolic of the Dominican order). Behind or above the church we see Paradise: St. Peter at the gates of heaven with rows of figures and saints behind him, including Dominican saints such as Thomas Aquinas, Dominic, and Peter of Verona (Peter Martyr). Thus, in the left half of the fresco we see what Polzer describes as an image of [End Page 44] the “universal Church,” both in the world and in heaven, populated by Dominican saints and protected by the Domini canes.

Figure 8. Andrea di Bonaiuto, Via Veritatis, Spanish Chapel, Santa Maria Novella, Florence, 1360s. Photo: Wikimedia Commons.
Click for larger view
View full resolution
Figure 8.

Andrea di Bonaiuto, Via Veritatis, Spanish Chapel, Santa Maria Novella, Florence, 1360s. Photo: Wikimedia Commons.

On the fresco’s right side is a much more diffuse set of images of the Dominican order in action, particularly in their role as combatants against heresy. At the bottom right, easily visible in their black and white garments, are the three Dominicans prominently celebrated throughout the whole chapel and observed to the left: Saints Dominic, Peter Martyr, and Thomas Aquinas. Dominic, closest to the crowd near the church, speaks directly to the black and white canes, urging them forward in their attack against a group of wolves (heretics). Next to this symbolic representation of the anti-heretical struggle stand Thomas and Peter, each before a large crowd of nonbelievers, many of whom are visually coded as non-Christian through skin tone and dress.33 Peter and Thomas each embody and enact the two primary strategies in the Dominican fight against heresy: preaching and theology. Peter speaks to a crowd as he ticks off arguments on his fingers, while Thomas stands in front of a crowd with an open book containing the opening lines of his Summa [End Page 45] contra gentiles, a treatise he composed between 1259 and 1265 to aid missionaries in explaining Christian doctrine and defending it against nonbelievers, specifically Muslims and Jews. The crowds’ reactions to both men are mixed, with some among them kneeling down, some looking away, some thinking carefully, some vehemently disagreeing, and one man in front of Thomas ripping apart the pages of his own “heretical” book.

The section of the painting above Thomas and Peter is the most enigmatic of the fresco. At the center we see Dominic guiding figures towards Saint Peter at the gates of heaven, and to the right is a Dominican preacher reaching out to touch the head of a kneeling elderly man, apparently absolving him of sin. This much makes sense: the Dominicans are depicted helping to prepare souls for salvation. But further to the right are three zones of figures whose role in the fresco’s argument is not at all clear. At the bottom, just above the groups of nonbelievers, are small figures dancing and playing music. Above them, seated on a ledge, are four larger figures: a woman playing a stringed instrument, a man holding a falcon, a woman playing with a dog on her lap, and an older man seemingly lost in thought. Above these four figures is a paradisiacal landscape of fruit trees, with small figures, probably children, climbing the trees and handing fruit down to people below. These three sets of figures have been interpreted in completely different ways by different scholars, who cannot even agree on whether the figures are intended to be positive or negative. Numerous authors comment on the similarity between the group of seated figures and the group in the Triumph of Death in the Camposanto in Pisa, which also includes a falconer, musician, and woman holding a dog; both groups take a lush garden as their backdrop (fig. 9). Though some have argued for a positive connotation for Andrea’s seated figures, in the context of the Pisan painting (and other images of the Triumph of Death from the Trecento) the figures in the garden are negative exemplars: these are figures of luxury, lost in the pleasures of the world (music, dancing, hunting, etc.) and failing to notice the figure of death that hovers just to the left with her scythe, having already cut down a pile of corpses below her.

Regarding the dancing figures in Florence, scholars have had difficulty interpreting the similar group of dancers in Ambrogio Lorenzetti’s government paintings in Siena, which may even have been intentionally multivalent. Other imagery of dancers can be either positive (signaling [End Page 46] harmony or joy in various poetic and mystical traditions) or negative (signifying luxury). The three dancing figures in the painted faux- relief below Giotto’s personification of justice in the dado of the Arena Chapel have been interpreted as positive figures, which together with imagery of hunters and merchants indicate the peace and prosperity brought about by the rule of justice.34 The small figures climbing the fruit trees of the Via Veritatis, too, have been interpreted variously as symbols of the fleeting pleasures of the world or as paradisical imagery rooted in the Song of Songs and its interpretive tradition.35 When these figures are interpreted in a positive light, as imagery that corresponds to the preparation of the soul for paradise, then the figures below would likely also hold positive connotations, a problem solved by Polzer by describing them as representatives of the stages of human life, a popular concept during the period.

Figure 9. Buonamico Buffalmacco, Triumph of Death, Camposanto, Pisa, 1335–40. Photo by the author.
Click for larger view
View full resolution
Figure 9.

Buonamico Buffalmacco, Triumph of Death, Camposanto, Pisa, 1335–40. Photo by the author.

Thus, while much of the painting follows an easily digested allegorical path—the Dominicans as representatives of the Church combating heresy and leading souls to salvation—the center-right of the painting disrupts the painting’s logical flow with poetic imagery that draws on both theological and courtly traditions but conveys no easy message. It is possible to try to thread the needle and argue that this strange and [End Page 47] unprecedented combination of images produces a single, intentional iconographic reading.36 But such arguments are premised on the idea that allegorical paintings such as the Via Veritatis (or Ambrogio’s government paintings or Buffalmacco’s Triumph of Death) must be coherent statements with stable iconographies. As Jean Campbell, Anne Dunlop, and others have argued, however, allegories in the Trecento often played with more ambiguous poetic strategies popular in the secular poetry of the period (and which in turn informed numerous theological traditions). Though everything about the context and imagery of the Spanish Chapel suggests a doctrinal, firmly Dominican program, the planners and artists of the paintings drew on poetic imagery and diagrammatic strategies from a variety of sources, destabilizing the easy legibility of the frescoes. The Via Veritatis notably eschews the extensive explanatory inscriptions found in so many of the other allegorical paintings of the period (including Nardo’s Hell, Buffalmacco’s Triumph of Death, and Ambrogio’s Allegory of Good Government); these could easily have been included in the fresco if transparent legibility had been desired. The elite theological audience in Santa Maria Novella’s chapter house, presumably, would have been prepared to interpret the fresco’s imagery as they saw fit, without being guided by vernacular inscriptions.

In some ways, the Via Veritatis is quite different from Nardo’s Hell. After all, while I am arguing that the Via Veritatis lacks a completely stable iconography, that is not the case with Nardo’s painting, which is quite clear in both its overall structure and the details of its imagery, explained through the inscriptions. But what I think the Via Veritatis shares with Hell, beyond the fact that both are doctrinal images whose forms and imagery are influenced by poetry, is that both have been misinterpreted as narratives. Just as I argued that the omission of Dante and Virgil in Nardo’s painting is a critical feature that denies the viewer an easy narrative pathway through the picture, so too the multivalent middle- right section of the Via Veritatis also disrupts the idea of the painting as a narrative allegory, already a tenuous idea to begin with. Many scholars have written engagingly about these kinds of mid- fourteenth-century allegorical paintings, but a constant feature of their interpretations is that the works are somehow fundamentally narrative. From Hans Belting to Joanna Cannon, critics take such paintings as marking a shift in narrative painting rather than as a different pictorial mode altogether (Bolzoni’s view, that they are argumentative but not narrative, is one exception). [End Page 48] The concept of the istoria, both before and after Alberti, is viewed as so fundamental to pictorial practice, particularly in the Trecento, that it has become difficult to recognize other modes, especially when their style and figures are so similar to narrative painting. To begin to challenge this reading of our two paintings, it is helpful to explain what features define a truly narrative painting in fourteenth-century Italy, so that we can see how these allegorical diagrammatic paintings are different. My argument is that Belting and others are right to call our attention to a seismic shift in the communicative strategies of these public allegorical paintings, but that it is not a shift within narrative painting, but a shift away from it.

Theorists of narrative point to its ubiquity in human cultural production; Suzanne Lewis, in her essay summarizing over a century of scholarship, defines narrative as “the representation of an event or a series of events.”37 In the Trecento, pictorial narratives in religious art usually focus on stories from the Old and New Testament (and their elaboration in other patristic texts) or from the lives of the saints, and these usually fall into one of three categories. They can be long sequential stories of the life of a particular religious figure or series of figures (such as the Arena Chapel, which focuses on the life of Christ and his parents and grandparents). They can juxtapose events from the lives of several different holy figures (such as the Peruzzi Chapel in Santa Croce in Florence, which includes three narrative paintings of scenes from the life of John the Baptist and three from the life of John the Evangelist). Or they can be programs that combine single narrative excerpts from different stories to create meaningful contrasts or themes (such as the chapter house of San Francesco in Pistoia).38 In the first two categories, an essential quality of the narrative is that the viewer can follow a single character or protagonist (or several of them) through multiple scenes and different stories, creating narrative arcs or progressions. In cases where single narrative scenes are juxtaposed with others from different contexts, viewers are still able to recognize the narrative being told, comparing it in their minds against both various textual traditions that include it or its other appearances in visual art. The success of narrative painting depends on this legibility: of recognizing who the characters are and tracking them through actions, emotions, and transitions.

In addition to containing identifiable characters in action, usually tracked across multiple scenes, fourteenth-century narrative painting [End Page 49] featured clearly defined settings and spaces—outdoor settings defined by landscapes and urban exteriors and interiors delineated through space-boxes. Both Nardo’s Hell and Andrea’s Via Veritatis contain some of these features: in Nardo’s painting we see some identifiable figures (like Cerberus) and there are certainly figures engaging in actions with each other. In the Via Veritatis there are even more identifiable characters, such as Dominic or Thomas, and we are presented with several evocative settings that more closely resemble the visual environments of narrative paintings—the church, paradisiacal landscape, and heaven. But while both paintings include what we could view as narrative moments (though I would prefer to call them actions rather than narratives), neither painting as a whole tells a story. Instead, both make an argument. If, in the Via Veritatis, the point was to tell the story of the Dominicans shepherding an individual soul to heaven (such as the chapel’s patron), the same figure could have been painted multiple times across the different spaces of the painting as he or she was led through the stages on the route to salvation. But like in Nardo’s painting, the goal of the image is not a single narrative progression but rather the evocation of an entire system or setting—of the structure of the afterlife or of a moment in the history of the church where the Dominicans emerged as its defenders and pastoral guides. Belting acknowledges the argumentative quality of the fresco, but still calls it a “fictitious narrative . . . serv[ing] as a description or exposition of a program in narrative terms,” and Cannon writes of “the use of narrative in constructing visual arguments.”39 We would do better, I think, to leave narrative behind when describing paintings like Nardo’s Hell and Andrea’s Via Veritatis—or at least to expand our definition of it. Not only do they lack protagonists, conflict, and a true story, but they also seem less interested in creating the kind of eyewitness experience so critical to most narrative painting during this period.

Diagrammatic Painting in Santa Maria Novella

Understanding paintings like Nardo’s as diagrammatic rather than narrative helps release us from a reliance on iconography to “decode” the frescoes, and opens up new ways of understanding how they might have been viewed. Shifting from iconography and acts of making to reception and acts of viewing is helpful in contrasting visual allegories from [End Page 50] textual ones; viewing a work like Andrea’s Via Veritatis or Nardo’s Hell is a fundamentally different experience from reading an allegorical text that might convey similar content. Comparing the Trecento istoria with the allegory, Belting explains that the genres are distinct even though they often look similar; the allegorical paintings (and often texts) of the period harness narrative details in order to “develop their arguments vividly . . . [T]he detail serves to fill in the schema with the visual evidence of the sense world.”40 In regard to the Via Veritatis specifically, Belting writes that the structure of the wall as a whole is a fiction, metaphor, or argument, but the individual stations are portrayed in a lifelike manner, enhancing the impact of the allegory. This is largely true, though the sharp juxtapositions in scale between the figures, and inscrutability of some of the allegorical imagery (indeed, its alternation between absolute clarity and poetic ambiguity) undercuts a true sense of “the real,” even in the details. Belting and others have pointed out the connection between the kind of allegorical “path” portrayed in the Via Veritatis (and in works like Buffalmacco’s paintings in the Camposanto) and genres of literature from the period that similarly introduce a reader to an idea or subject by way of a fictional path with stations along the way, most famously in the Commedia and Brunetto Latini’s Tesoretto. But while the presence of this strategy in allegorical texts may be the source for a similar device in allegorical painting, it tells us little about the viewer’s experience of it in a pictorial version. The reader usually moves sequentially; though the overall fictional structure of the pathway can be held in the memory, for the most part a reader is absorbed in the momentary details of a particular place, event, interaction, or description.

Despite the incredible wealth of detail on display in both paintings, however, I don’t think they elicit the same kind of reception, where the viewer gets lost in the details. Instead, the viewer’s focus is firmly drawn to the path and structure itself. The individual actions, demonstrations, and details do more to clarify the route, purpose, and function of the overall path than to elicit immersive, affective viewing. What both paintings do is demand a constant awareness of their routes, structures, or layouts as a whole, not just as containers for the actions portrayed within them. Diagrammatic painting prompts a certain kind of viewing: in beholding an entire system, the viewer is elevated to a position of privilege, authority, and understanding. While reading an [End Page 51] allegory like Dante’s Commedia is immersive, viewing one is extractive, moving the audience outside of the system so that they can look at it in a different way. This view from the outside was likely appealing to the more educated and elite audiences of spaces like the Spanish Chapel and Strozzi Chapel, whose knowledge of the paintings’ references— theological, academic, and literary—could aid them in populating the systems on display with further meaningful details and characters. The spatial setup and patterns of use of the Spanish Chapel underscore this kind of viewing: during daily chapter meetings, the friars would sit on benches along the three walls (east, south, and west), giving those facing the Via Veritatis or Triumph of Thomas a viewing distance of over fifty feet from these paintings. These daily meetings provided the friars with endless hours to contemplate the paintings’ messages, but almost always from afar; they were not on their knees, holding candles, before an image like the Via Veritatis. Instead of a devotional figure or narrative moment, it was the path to salvation, its inexorable route towards a heaven whose access they controlled, that would have drawn their attention. Similarly, while direct devotional practice, connected to the Strozzi family tombs and also to Orcagna’s altarpiece, would have been one way of experiencing Nardo’s Hell, the scene was also highly visible from far away. Walking down the main nave of the church, or standing in the crossing or transept, viewers can see the Hell more easily than any of the other walls of the chapel. The painting offers its highly organized structure for appreciation long before anyone gets close enough to view its details.

These diagrammatic paintings in Santa Maria Novella are both prescriptive and generative. There is no denying the paintings’ moralizing quality; they present particular visions of the world and of salvation that are filtered through a Dominican viewpoint and filled with doctrinal instructions or perspectives. But their diagrammatic formats do more than just communicate doctrine; their organization and inevitability make a visual and intellectual impact. In the Via Veritatis, the imagery of the garden on the painting’s right side would have supported a multitude of meanings, based in the church’s myriad of visual and textual responses to this kind of paradisiacal imagery (imagery featuring prominently in both biblical texts and secular poetry). The figures in the garden call attention to the allegorical, poetic status of the painting, keeping it from being too linear, doctrinal, or limited; in short, making it more suitable [End Page 52] for long-term viewing. In incorporating these multivalent figures, the painting sustains its audience’s interest, even over repeated viewings. A friar preparing a moralizing sermon might focus on the seated figures’ evocations of luxury and pride, while one seeking an eschatological or visionary evocation of the heavenly sphere might appreciate the dancing figures for their suggestion of the ecstatic experience of the union with God. The route is fixed: it leads to heaven. But the actions along the way oscillate between clear explanations of Dominican priorities and activities and poetic suggestions of feelings or states. Similarly, while the map of Hell offered in Nardo’s painting is highly organized and purposeful, it also provides a mental scaffold for imaginative viewing, onto which readers of Dante’s poem or viewers of other scenes of Hell could insert their own images. Though they resemble and may have been inspired by traditions of allegorical literature and poetry that can more justly be called narrative, neither of the allegorical paintings in Santa Maria Novella tells a story. Instead, they create diagrammatic fictional structures or routes, populated by figures in action, that prompt viewers to consider the perfection and logic of a system, and to contemplate their own place within it.


1. The best recent work on the Hell scene in the chapel has been done by Theresa Holler. See Holler, “L’aldilà della Cappella Strozzi: i Domenicani, l’esilio di Dante e il ritorno dell’Inferno,” in Images and Words in Exile: Avignon and Italy during the First Half of the 14th century, eds. Elisa Brilli, Gerhard Wolf, and Laura Fenelli (Florence: Galluzzo, 2015), 401–20; Holler, “Infernale Landschaften: Wie Dantes Commedia das Bild des Weltgerichts in Italian verändert,” Deutsches Dante-Jahrbuch, 93, no. 1 (2018): 74–103; and Holler, Jenseitsbilder: Dantes Commedia und ihr Weiterlebe im Weltgericht bis 1500 (Berlin: De Gruyter, 2020), 107–245.

2. On the painting, as discussed by earlier scholars in the context of Trecento and Quattrocento illustrations of Dante’s Commedia and in works on Hell or Last Judgment imagery, see especially Millard Meiss, “An Illuminated Inferno and Trecento Painting in Pisa,” The Art Bulletin 47, no. 1 (1965): 21–34; Peter Brieger, Millard Meiss, and Charles Singleton, Illuminated Manuscripts of the Divine Comedy (Princeton: Princeton University Press, 1969); Gert Kreytenberg, “L’Enfer d’Orcagna: la première peinture monumentale d’après les Chants de Dante,” Gazette des beaux-arts 114 (1989): 243–62; Christopher Kleinhenz, “Dante and the Tradition of the Visual Arts in the Middle Ages,” Thought 65 (1990): 17–26; Eugene Paul Nassar, “The Iconography of Hell: From the Baptistery Mosaic to the Michelangelo Fresco,” Dante Studies 111 (1993): 53–105; Nassar, Illustrations to Dante’s Inferno (Vancouver: Fairleigh Dickinson University Press, 1994), 17–26; Barbara Watts, “Sandro Botticelli’s Drawings for Dante’s Inferno: Narrative Structure, Typography, and Manuscript Design,” Artibus et Historiae 16, no. 32 (1995): 163–201; Leon Jacobowitz-Efron, “Dante in Pistoia: The Frescoes of the Cappella del Giudizio,” Quaderni storici, 47, no. 140 (2012): 443–68; and Deborah Parker, “Illuminating Botticelli’s Chart of Hell,” Modern Language Notes 128, no. 1 (2013): 84–102.

3. On this broader context of diagrammatic painting in the period, see Karl Whittington, Trecento Pictoriality: Diagrammatic Painting in Late Medieval Italy, forthcoming with Brepols. Other significant works that treat the monumental diagrammatic painting include Hans Belting and Dieter Blume, eds., Malerei und Stadtkultur in der Dantezeit (Munich: Hirmer, 1989); Dorothee Hansen, Das Bild des Ordenslehrers und die Allegorie des Wissens (Berlin: De Gruyter, 1995); Raphaèle Preisinger, Lignum Vitae: Zum Verhältnis materieller Bilder und metaler Bildpraxis im Mittelalter (Munich: Wilhelm Fink, 2014); Marius Hauknes, “The Image of the World in Thirteenth Century Rome” (PhD Diss., Princeton University, 2014); Iris Helffenstein, Wissentransfer in Bildprogrammen des Trecento: Allegorie, Imitation, und Medialität (Leiden: Brill, 2020); and Marius Hauknes, “Painting against Time: Spectatorship and Visual Entanglement in the Anagni Crypt,” The Art Bulletin 103, no. 1 (2021): 7–36.

4. Lost paintings by Giotto that were primarily diagrammatic or allegorical rather than narrative include two famous examples in Padua’s Palazzo della Ragione and Florence’s Palazzo della Signoria. In addition, other non-narrative works by Giotto included a “famous men” cycle painted in the sala grande of the Castel Nuovo in Naples for Robert of Anjou, an allegory of earthly glory in the Visconti Palace in Milan, and an allegory of the Christian faith in the Guelf Palace in Florence. See Creighton Gilbert, “The Fresco by Giotto in Milan,” Arte Lombarda, 47/48 (1977): 31–72; Eva Frojmovic, “Giotto’s Allegories of Justice and the Commune in the Palazzo della Ragione in Padua: A Reconstruction,” Journal of the Warburg and Courtauld Institutes 59 (1996): 24–47; and Daniela Parenti and Maria Monica Donato, Dal Giglio al David: Arte Civica a Firenze tra Medioevo e Rinascimento (Florence: Giunti, 2013).

5. On this painting, see Hansen, Das Bild des Ordenslehrers, 104–8, Hayden Maginnis, Painting in the Age of Giotto: A Historical Reevaluation (University Park: Penn State University Press, 2000), 164–91, Diana Norman, Siena, Florence, and Padua: Art, Society and Religion 1280–1400, vol. 2 (New Haven: Yale University Press, 1995), 217–42; Joanna Cannon, Religious Poverty, Visual Riches: Art in the Dominican Churches of Central Italy in the Thirteenth and Fourteenth Centuries (New Haven: Yale University Press, 2014); 186–99; and Andrea de Marchi, ed., Santa Maria Novella: La Basilica e il Convento (Florence: Mandragora, 2015), 227–30.

6. For broader ruminations on the “diagrammatic mode” of representation, concerned almost exclusively with manuscripts, see Michael Evans, “The Geometry of the Mind,” Architectural Association Quarterly 12 (1980): 32–55; Madeline Caviness, “Images of Divine Order and the Third Mode of Seeing,” Gesta, 22 (1983): 99–120; Steffen Bogen and Felix Thürlemann, “Jenseits der Opposition von Text und Bild: Uberlegungen zu einer Theorie des Diagramms,” in Die Bildwelt der Diagramme Joachims von Fiore, ed. Alexander Patschovsky (Ostfildern, Germany: Jan Thorbecke Verlag, 2003); Lina Bolzoni, The Web of Images: Vernacular Preaching from Its Origins to Saint Bernardino da Siena (London: Ashgate, 2004); Faith Wallis, “What a Medieval Diagram Shows,” Studies in Iconography 36 (2015): 1–40; Jeffrey Hamburger, Diagramming Devotion: Berthold of Nuremberg’s Transformation of Hrabanus Maurus’s Poems in Praise of the Cross (Chicago: University of Chicago Press, 2020); Marcia Kupfer, Adam Cohen, and J. H. Chajes, The Visualization of Knowledge in Medieval and Early Modern Europe (Turnhout: Brepols, 2020); and Ayelet Even-Ezra, Lines of Thought: Branching Diagrams and the Medieval Mind (Chicago: University of Chicago Press, 2020).

7. Two unpublished dissertations provide excellent background and references on the patronage, construction, and decoration of the chapel, as well as extensive discussions of its imagery: Kathleen Giles, “The Strozzi Chapel in Santa Maria Novella: Florentine Painting and Patronage, 1340–1355” (PhD Diss., New York University, 1977), and Frances Lee Pitts, “Nardo di Cione and the Strozzi Chapel Frescoes: Iconographic Problems in Mid-Trecento Florentine Painting,” (PhD Diss., Berkeley, 1982).

8. See especially Kathleen Giles Arthur, “The Strozzi Chapel: Notes on the Building History of Sta. Maria Novella,” The Art Bulletin 65, no. 3 (1983): 367–86.

9. See Pitts, “Nardo di Cione,” 29–38.

10. The status of Dante’s writings as theology or poetry was frequently discussed in the years after its publication. As Pitts notes, it would have indeed been strange for Dante’s theology to be refuted as such, given that it was in fact greatly influenced by the Dominican academic culture of Santa Maria Novella; Dante praised the famous Thomist Remigio de Girolami, a student of Aquinas who was the lector and prior at Santa Maria Novella off and on from the 1270s until his death in 1319. See Pitts, “Nardo di Cione,” 32.

11. Pitts, “Nardo di Cione,” 37. This early illustrated copy of the Commedia is Florence Biblioteca Laurenziana MS Strozziano 152.

12. On Pacino’s panel, see especially Raphaèle Preisinger, “Das Bild als Schwelle zum Jenseits: Zu Pacino di Bonaguidas Lignum Vitae, in der Accademia in Florenz,” in Topologien der Bilder, ed. Inge Hinterwaldner (Munich: Fink, 2008), 267–83, and Preisinger, Lignum Vitae.

13. Pitts, “Nardo di Cione,” 47–49. On the connection between the Hell scenes in Pisa and Florence, the best source is Holler, Jenseitsbilder.

14. See Whittington, Trecento Pictoriality, chapter 4. On the affective aspects of medieval diagrams, see Hamburger, Diagramming Devotion.

15. Pitts also discusses this oscillation between part and whole; see Pitts, “Nardo di Cione,” 52–53.

16. Pitts discusses the largely negative response to the painting in scholarship, particularly among earlier scholars. See Pitts, “Nardo di Cione,” 3.

17. Theodore Cachey, “Cartographic Dante,” Italica 87, no. 3 (2010): 325–54. For Cachey, this claim is part of a larger project that has involved reconciling multiple forms of mapping the poem, including the descriptions of the physical topography of Hell, the poem’s references of the “geo-physical reality” of the Mediterranean world as Dante knew it, and the idea of the route that the reader follows through the text; Cachey writes that, “mapping the intersection between textual space (that is, the space of representation) and represented space (both the physical space of Hell and the physical spaces outside the poem that are referenced) provides a more complex picture of the ethical terrain of the poem” (37–38).

18. Parker, “Illuminating Botticelli’s Chart of Hell,” 99.

19. On links between Dante and the Dominicans, see especially Robert Hollander, “Dante ‘Theologus-Poeta,” Dante Studies 94 (1976): 91–136; Giuseppe Ledda, “S. Domenico e l’Ordine dei Predicatori nella Commedia di Dante,” Domenicane Memories 39 (2008): 243–70; Vittorio Montemaggi and Matthew Treherne, ed, Dante’s Commedia: Theology as Poetry (Notre Dame, IN: Notre Dame Press, 2010), Christopher Ryan, Dante and Aquinas (London: Ubiquity Press, 2013); and Anna Pegoretti, “Un Dante ‘domenicano’: la Commedia Egerton 943 della British Library” in Dante visualizzato: carte ridenti, ed. Rossend Arqués Corominas and Marcello Ciccuto (Florence: Franco Cesati Editore, 2017), 127–42.

20. Hans Belting, “Historia and Allegory: The New Role of Narrative in Public Painting of the Trecento,” in Pictorial Narrative in Antiquity and the Middle Ages, eds. Herbert Kessler and Marianna Shreve Simpson (Washington, DC: National Gallery, 1985), 164.

21. The connection between exempla in preaching and painting is also treated by Laura Leeker; see Leeker, “Fragmented Narrative in the Chapter House of San Francesco in Pistoia,” in New Horizons in Trecento Italian Art, eds. Bryan Keene and Karl Whittington (Turnhout: Brepols, 2021) and Leeker, “Narrative and Experimentation in Fourteenth-Century Italian Chapter Houses” (PhD Diss., The Ohio State University, 2020).

22. Pitts, “Nardo di Cione,” 58, and Bolzoni, Web of Images, 29.

23. Bolzoni, Web of Images, 23–24.

24. Bolzoni discusses extensively this intersection of public allegorical and diagrammatic painting with the arts of memory in Web of Images, chapter 2. In general, I find the lens of memory somewhat less convincing for the interpretation of diagrammatic paintings such as Nardo’s than many other scholars. While memory work may have been part of the intent of the painting, it simply must be acknowledged that it would not have been terribly difficult for a medieval reader or viewer to remember the sequence of levels in Dante’s Hell. Mnemonics alone do not explain the painting.

25. See, for example, Anne Dunlop, “Allegory, Painting, and Petrarch,” Word and Image 24 (2008): 77–91.

26. The necessary involvement of the spectator as protagonist is noted by Gaia Ravalli in “L’egemonia degli Orcagna e un secolo di pittura a Santa Maria Novella,” in Santa Maria Novella: La Basilica e il Convento, ed. Andrea de Marchi (Florence: Mandragora, 2015), 157–246. Ravalli writes that “The absence of Dante and Virgil determines a strong involvement of the spectator who himself becomes a pilgrim to Hell, guided along the horrific journey through a variety of narrative devices” (184).

27. On diagrams as machines for thought, see Mary Carruthers, The Craft of Thought: Meditation, Rhetoric, and the Making of Images, 400–1200 (Cambridge: Cambridge University Press, 1998). See also Hamburger, Diagramming Devotion.

28. Bolzoni, Web of Images, 23. In a very different context, Gerald Guest makes a similar point about the moralizing qualities of urban imagery in Chartres; see Guest, “The Prodigal’s Journey: Ideologies of Self and City in the Gothic Cathedral,” Speculum 81, no. 1 (2006): 35–75.

29. On this manuscript, see Holler, Jenseitsbilder, 135 and plate 10.

30. Nirit Ben-Aryeh Debby, “Art and Sermons: Dominicans and Jews in Florence’s Santa Maria Novella,” Church History and Religious Culture 92 (2012): 171–200.

31. On the painting’s theological meanings, see Eve Borsook, The Mural Painters of Tuscany: From Cimabue to Andrea del Sarto (Oxford, UK: Clarendon, 1980), 51–55; Julian Gardner, “Andrea di Bonaiuto and the Chapterhouse Frescoes in Santa Maria Novella,” Art History 2 (1979): 107–38; Joseph Polzer, “Andrea di Bonaiuto’s Via Veritatis and Dominican Thought in Late Medieval Italy,” The Art Bulletin 77, no. 2 (1995): 262–89; Serena Romano, “Due affreschi del Cappellone degli Spagnuoli: Problemi iconografici,” Storia dell’Arte 28 (1976): 181–214; Millard Meiss, Painting in Florence and Siena after the Black Death (Princeton: Princeton University Press, 1951), 94–104, Daniel Russo, “Allégorie, analogie, paradigm. Étude sur la peinture de l’Eglise dominicaine par Andrea di Bonaiuto, à Florence, 1365/1367,” in L’allégorie dans l’art du Moyen Age, ed. Christian Heck (Turnhout: Brepols, 2011), 79–94; Daniel Russo, “Religion civique e art monumental a Florence au XIV siècle: La decoration peinte de la salle capitualre à Sainte-Marie-Nouvelle,” in La religion civique à l’époque médiévale et moderne, ed. André Vauchez (Rome: Ecole française, 1995), 279–96; and Joanna Cannon, Religious Poverty, Visual Riches, 186–96.

32. On the debates surrounding the identity of this building, and its relation to the actual fabric of the cathedral at the time when the painting was made, see especially Rahel Meier, “Wie kommt der Florentiner Dom in den Kapitelsaal der Dominikaner von Santa Maria Novella?,” Mitteilungen des Kunsthistorischen Institutes in Florenz 61, no. 1 (2019): 135–44. The strategy of placing a symbol of the universal Church within the fresco is part of how Polzer argues that “the mural strives to identify the Dominican order with the church as a whole” (Polzer, “Via Veritatis,” 266.

33. On the suggested racial/religious identities of these figures, see Debby, “Art and Sermons.”

34. For a convincing interpretation of this scene, and a recent review of the literature on it, see Péter Bokody, “Justice, Love and Rape in Giotto’s Allegories of Justice and Injustice in the Arena Chapel, Padua,” in The Iconology of Law and Order, eds. Anna Kérchy, György Szönyi, and Atilla Kiss (Szeged: University of Szeged Press, 2012), 50–61.

35. Polzer argues, for example, that “The garden is a topos for the medieval vision of Paradise. Andrea di Bonaiuto’s garden refers to the blessed afterlife, and the children consuming its fruit to spiritual cleansing.” The children, for Polzer, are part of a larger series of images in the painting which refer to Purgatory. Polzer, “Via Veritatis,” 271, 278.

36. This is the route taken by Polzer, who summarizes: “the right center of the Via Vertitatis allegorizes mortal existence from a Dominican perspective. The seated figures and kneeling man receiving absolution represent the normal course of human life, from youth to old age; the dancing maidens below them, together with the drummer and bagpiper, carnal pleasure; and the children climbing trees and picking and eating pomegranites and other fruit, spiritual cleansing. . . . [T]he programmer was influenced by Passavanti’s penitential Specchio, but what was terrifying and painful in it he excluded” (Polzer, “Via Veritatis,” 283).

37. Suzanne Lewis, “Narrative,” in A Companion to Medieval Art: Romanesque and Gothic in Northern Europe, ed. Conrad Rudolph (Oxford, UK: Blackwell, 2006), 86.

38. My understanding of this third category, combining narrative excerpts, has been greatly influenced by Laura Leeker’s dissertation; see Leeker, “Narrative and Experimentation.”

39. Cannon, Religious Poverty, Visual Riches, 194.

40. Belting and Blume, ed., Malerei und Stadtkultur in der Dantezeit, 31, 39.