Almost every art historian, in coming face-to-face with an artwork they have previously seen only in reproduction, has had the startling realization that the physical piece differs significantly from what they had imagined. I distinctly remember having this experience when I first saw Planos em superfície modulada no. 1, a white monochromatic painting by Brazilian artist Lygia Clark from 1957. In 2001, during a visit to the São Paulo collector Adolpho Leirner's apartment, I noticed the painting on the wall, which I had only ever seen in reproduction (Figure 1). As I slowly approached, and the composition came into focus, I realized that the five black lines dissecting the white picture plane had not been drawn or painted, as I had previously assumed, but were negative space; they were hollow veins, mere millimeters wide and deep, between four wooden panels that had been glued to a larger rectangular board (Figures 2–4). Clark created the illusion of lines, not by making marks on the support, but by leaving space untouched. The artwork was something entirely different than what I had presumed, and it caused me to reevaluate what I thought I knew about Lygia Clark specifically, and geometric abstraction more broadly.
As art historians, we are trained to be critical of images; to decode their compositions, compare them to other works, and connect them to history. However, when we set out to do any of these by looking only at photographic reproductions, we willingly submit to being misled if we treat them as truthful representations of the work in question. Intellectually, we know information is missing in the photograph: for example, clues that reveal the scale of the work are absent, or the frame or pedestal has been cropped out.1 It is also possible that the color palette has become distorted due to the aging of the photograph, slide, or transparency, as well as by the calibration of color settings on our computers, projectors, and printers.2 Or, as I describe above, crevices, reliefs, and shadows may be interpreted as marks on a surface rather than signs of a third dimension. And though we would never presume that a single photograph could describe the totality of a sculpture or building, which need multiple images to reveal their faceted dimensions, I think we assume something different when we look at photographic reproductions of an artwork we deem two-dimensional. We more readily accept that the image on the page or projected on the screen is an acceptable substitute for the painting, the print, or the photograph. Joel Snyder rationalizes this visual leap writing that “photographs make a special claim upon our attention because they are supposed not only to look realistic … but also to derive from or be caused by the objects they represent. This ‘natural connection’ has been taken as a reinforcement and even as a guarantee of realistic depiction … and urges us to conclude that in certain important respects looking at a picture is equivalent to looking at what is pictured.”3 In the case of photographic reproductions of artworks, this pictorial melding is only reinforced by the tombstone information, which typically only consists of height and width, further underscoring the illusion of flatness.4 And though these variables may be subconsciously factored into our reading, our eyes and brains cannot account for what is simply not there.
In my field of research, modern art from Latin America, we have been especially reliant on reproductions because most of our canonical objects have belonged to private collectors and were only shown in temporary exhibitions.5 The irony is that although exhibition catalogs have been the source for the most groundbreaking scholarship in this field for the past three decades, the plate sections have always just recycled older reproductions. There are a few reasons for this. First, artworks are typically shipped to a museum just before they are installed for the exhibition. Catalogs, on the other hand, are customarily sent to press a full year ahead, making it impossible for museum photographers to shoot new images for the book. Instead, digital images are sent in advance by the owner of the work, thus resulting in the previously mentioned redundancies. Furthermore, shooting new photography is logistically complicated, is prohibitively expensive, and requires a significant amount of specialized labor. In addition to hiring a photographer who will shoot and edit all of the images, it requires arranging and paying for the packing and shipping of the works, renting studio space for the sessions, and hiring art handlers to unpack, install, de-install, and re-pack each work of art on-site, among many other steps in the process that are invisible in the final publication.6
Because of my many years studying geometric abstraction from Brazil, Argentina, and Venezuela, I know that the reproductions present their own set of interpretive challenges. For example, given that the original geometric paintings are not intended to reference the world around us, having rejected the convention of one-point perspective in favor of other formal strategies such as symmetry and repetition, it can be especially hard to discern a work's orientation looking only at a reproduction. This has resulted in compositions sometimes being flipped or rotated in publications, which then destabilizes the reliability of other books portraying the same work (Figures 5–7). The artist's signature, traditionally located along the bottom edge of a painting, cannot be relied on as a clue to the work's orientation, because many of the concrete artists in South America chose to sign their works only on the verso side, leaving the front free of any gestural marks (Figure 8).7 And even when there is a signature on the face, we cannot presume that it is always located along the bottom edge (Figure 9). Determining the perimeter of a work can also be complicated because many of these artists challenged the relationship between the work of art and the frame that contained it by experimenting with shaped canvases, or incorporating the frame into the composition, and, other times, rejecting the frame altogether in favor of wrapping the composition around the edges of the support (Figure 10). The proliferation of digital photography, which only became commonplace in the early 2000s, has resolved some of these ambiguities and is far more stable and consistent than working with physical transparencies, which could be unintentionally flipped or rotated in the book layout, and, because of their filmic nature, often faded, further confusing the artist's chromatic decisions. However, there is no limit to the number of alterations made to a digital file, so that the reproduction best approximates the work of art, and though the intention is to render the best visual approximation to the original artwork, as a viewer it is impossible to know which blemishes were touched up or corners sharpened, disguising the true physical status of the object.
With these concerns in mind, I have looked for a new way to represent works of art that would help prevent others from having the experience I described when I first saw Clark's Planos em superficie modulada no 1, back in 2001, without succumbing to such a degree of skepticism that it is no longer practical to rely on them for research. A project that I initiated at the Getty Research Institute in 2014, which resulted in the 2017 exhibition Making Art Concrete: Works from Argentina and Brazil in the Colección de Patricia Phelps de Cisneros (J. Paul Getty Museum, Los Angeles) and a catalog of the same title, became the perfect opportunity to address these problems since two of the typical barriers for such an undertaking were already resolved; the works were on-site for two years prior to the show, and the Getty's photographers were willing to dedicate significant time and space to shoot every object from multiple perspectives. With my fellow editors, we decided that one way to mitigate misreadings was to publish a suite of images for each work of art so that its physical and material properties could be represented (Figure 11). We included shots of the front, back, and sides, as well as views from oblique angles and under different lighting conditions to reveal underpainting and surface texture. We also tried to include shadows whenever possible to demonstrate an object's physical presence.
The irony was that we had to rely on photography as our tool to ameliorate the original conundrum, which had been perpetuated by the very same medium. Despite our good intentions to represent the three-dimensional artworks as faithfully as possible, we were nevertheless confined by the static photograph, which meant that there were still aspects obscured by the limitations of the medium of photography and the format of the printed book. Similar to earlier photographs, these new perspectives are still mediated by photographers, book designers, curators, and editors, and therefore cannot be expected to represent the work neutrally or even comprehensively, as much as we might want the image to function as a surrogate for the artwork.
Much of this has to do with the conventions of art photography and the ethos of the museum photographer, whose objective is to shoot the (seemingly) flat painting straight on and neutralize her own authorship as much as possible. In other words, she has the impossible task of creating a document that purports to be a neutral and objective representation of a work of art, which she must manipulate so that her eye and hand are nearly undetectable. In a conversation with Getty photographer Stacey Rain Strickler, she explained that her objectives when shooting a work are to get “the contrast and tones right, then the colors as close as you can.”8 But she also said that she sought to capture “the impression the artist was trying to make. … It's all about the feeling that the original piece of art is giving you,” a statement that reveals her own subjectivity in the process.9 Later, in post-production, the photographer can adjust the light and contrasts, if she thinks it is required. Once the book designer incorporates the image into the volume's layout, she can still crop any undesirable elements from the periphery, change the background color, and even add in shadows, all with the magic of computer software.10
Nowhere in our catalog were these digital tools deployed to a greater extreme than with the image of Willys de Castro's Objeto ativo (cubo vermelho/branco) from 1962, which appeared on the front and back covers of the catalog, as well as on promotional materials for the exhibition all over Los Angeles. If you compare a photograph of the work taken during the exhibition with the photograph printed in the catalog, some stark differences are immediately visible, namely the shadows that the work casts, or doesn't (Figures 12, 13). I asked Strickler about this, wanting to understand how she approached the job. In the exhibition gallery, focused lighting overhead created dramatic shadows under the work, but where did Strickler position them to create such subtlety? How was it possible for the shadow below to dissolve so evenly on both sides of the cube? As it turned out, the process was more complex than that of many of the others.
Photography for the Willys Cube implemented focus stacking, where images of the object at each focal plane are composited through a software program called “Helicon Focus.” It marries all the images into one image where every surface of the object is in focus. This is accomplished by taking a picture where the closest point of the object is in focus and another where the object at the farthest point is in focus, and all subsequent points in between, relying on depth of field to assist in covering for the entire object. Michael Smith [Manager of Imaging Services] took the digital images I shot for the exhibition [catalog] photo and added shadows to the background.11
In the end, the reproduction in the book is a composite of six individual shots. So, although these shadows help convey the three-dimensionality of the work, the photo is not a representation of the work in actual space, but in a manufactured digital environment. Even though it is a more conventional three-dimensional object, and not nearly as subtle as Clark's Planos em superfície modulada no. 1, the same issues are at stake—how to faithfully represent three dimensions with a single image printed on a page.
My intention here is not to suggest that there are nefarious deceptions taking place in our books and on our screens, but rather to alert all of us who utilize reproductions of artwork that we must turn our same critical art historical eyes on those reproductions we take for granted to avoid any presumption that they are always accurate and objective representations of our object of inquiry. I have concluded that one of the best ways to be reminded of the subjective nature of photographic reproductions is to consult many images of a single object, ideally taken under different conditions, and to include height, width, and depth whenever available and germane, as a means of conveying the physical nature of the artwork under consideration. Indeed, in Making Art Concrete, by including multiple images of the same object, our exhibition catalog underscores the incompleteness of the record it presents, calling attention to the fallacy that a single photograph can ever be a stand-in for a work of art.
The authors of the scholarly essays in this issue supply excellent examples of how one can attend to these considerations with a thoughtful selection of images and meticulous captions. For example, all of the authors found solutions to convey missing visual information about the scale and context of the works in question. In “Vitality Materialized: On the Piercing and Adornment of the Body in Mesoamerica,” Andrew Finegold illustrates entire manuscript pages indicating the specific scene in his captions, in order to show proportion within the larger visual program. Writing about the use of gold and mirrors in Quito's Jesuit Church, Isabel Oleas-Mogollón takes on a subject that has little photo documentation available, in large part because photography is not allowed inside the church. Making use of digital images on Wikimedia Commons, she expands the captions to provide contextual information about the paintings' location at the site, acknowledging that an image that included a viewer next to the painting would be more effective in giving a sense of scale and accessibility.12 For “Luis Jiménez's Mustang: Monumental Misreadings,” Marisa Lerer combats the photographic conventions of public sculpture, which often depicts monuments in pristine settings and perfect conditions. By including an archival photograph of the artist with the sculpture in his studio, she is able to show scale as well as materiality and process. The ability to publish multiple images and supplementary materials is a unique privilege available to authors in Latin American and Latinx Visual Culture. Because it is foremost a digital publication, many of the constraints placed on printed publications do not apply. Currently, there is no limit to the number of illustrations that can be included with a manuscript—the guidelines specify only the quality and format of the file, be it photograph, audio, or video—and everything can be published in color. Authors can also make supplementary materials available online, such as appendices, databases, digital models, or any other scholarly content that might originate in digital humanities. Even if the points I raise here seem trivial to the experts reading this, I am convinced that the small act of rewriting captions to include all three dimensions, or publishing multiple views of an object whenever possible, will deepen everyone's knowledge, and provide new scholars and students with a richer visual vocabulary to provoke new questions. Certainly, the conundrum of how to realistically depict an abstract composition, which I have discussed here, continues to generate theoretical and methodological puzzles.