“Imageability”: aligning bodies and imaging technologies
Neuroimaging: conceptions and issues
In 1918, the American neurosurgeon Walter Dandy said of the X-ray-based neuroimaging technique named ventriculography: “One or more of the cerebral ventricles can be sharply outlined in a roentgenogram if the ventricular fluid [is] withdrawn and replaced by an equal quantity of air”.1 Dandy suggested that the ventricles and sulci of the brain could be made visible if filled with a contrast medium that would create a shadow on radiographs—in this case air. He hoped that wherever the ventricles showed as distorted in size or shape, this would indicate the spread of a tumor inside the brain and reveal its morphology, thus allowing surgical intervention planning. In the realm of neurosurgery, imaging and intervention are closely intertwined; they cannot be separated into diagnostics and subsequent intervention. Before any imaging procedure can take place, such as Dandy’s ventriculography or, more precisely, pneumoencephalography, the patient has to be prepared by means of a procedure involving intervention—ventricular fluid must be drained away and replaced with air. In other words, pre-surgical imaging such as X-ray photography arranges and aligns the human body so that it becomes an object of diagnostically significant imaging and is accordingly made to “comply”.
This points to a broader media theoretical issue, relating, first, to the circumstances whereby media technologies themselves need to be installed and arranged in order to be functional; and, second, to the fact that technological infrastructures and apparatus are agential: they configure both their application contexts and the human actors involved. This issue has of course already arisen within visual theory with regard to aesthetic and epistemological constraints,2 as well as in science and technology studies in respect of the material and pragmatic conditions imposed by media technologies.3 However, the specific interrelation between these two levels has seldom been discussed. I will investigate, from the point of view of media theory, media technologies’ requirement that structural alignments be instructed before any imaging process can be initiated. I will then examine how this relates to aesthetic and epistemological characteristics of visualization, employing the historical example of the pneumoencephalography neuroimaging technique and the associated preparatory actions. This roentgenological technique was based on inflating the cavities in the brain with air as a contrast medium to deduce the relational morphology of malformations inside the brain. Using this example, I will attempt to demonstrate how the material and conceptual prerequisites of diagnostic imaging are inextricably linked. Taking historical subject-specific texts, photographs of application contexts, and radiograms as my basis, I will explore the ways in which the patient’s body as a three-dimensional, moving entity is aligned to the analogous imaging technique of pneumoencephalography on a conceptual, material, and aesthetic level.
This leads to the broader media theoretical issue of how media technologies determine their application context through the specific material and conceptual affordances they present to users, in this case patients and doctors. Drawing on the writings of Walter Benjamin, I will start by introducing the notion of imageability as a point of reference, with respect both to the affordances of image production and to the related implications for the aesthetics and functions of images to be produced.
Imageability—the affordances and aesthetics of imaging
With the arrival of the so-called practical turn, the contexts of the production and application of images became a major focus of analysis.4 Yet few studies to date have focused on the constraints imposed on objects, bodies, and processes in order to render them visible. Feminist science studies scholar Karen Barad has theorized on the interplay between objects and the technical processes of measuring and presenting them, receiving much attention.
Barad coined the term intra-action to describe the structural and functional interplay between objects and the applied technologies of measurement and observation, which act on both material and conceptual levels—“the inseparability of ‘objects’ and ‘agencies of observation’”.5 Barad’s notion of intra-action is intended to cover an epistemological as well as ontological understanding of materiality and agency, “to take account of the ways in which ‘matter comes to matter,’ including the active role of material constraints and conditions”.6 Barad’s idea that boundaries between technical systems, in this case X-ray photography devices, and objects of observation, in this case living humans, are constantly (re-)established in “intra-action” seems useful as a way of understanding the interconnection between and coaction of technical infrastructure and bodies in the realm of medical imaging, which aims to generate epistemologically as well as practically—that is, surgically—significant images.
A specific intra-action (involving a specific material configuration of the “apparatus of observation”) enacts an agential cut (in contrast to the Cartesian cut—an inherent distinction—between subject and object) effecting a separation between “subject” and “object”.... In other words, relata do not preexist relations; rather, relata within phenomena emerge through specific intra-actions.7
Barad’s notion of intra-action may be applied to the procedures that have to be undertaken simply to establish the conditions of possibility for imaging by enacting an “agential cut” between the different actors involved. This includes defining and constituting the relata that must be juxtaposed and deployed for a specific epistemological and practical purpose. Barad develops and applies the idea with regard to Niels Bohr’s physical experiments—which, unlike medical imaging, are not directed toward living objects and dynamic entities. Patients’ vitality is essential to medical imaging processes, whether these are diagnostic or preliminary to surgical intervention and whether these are analog such as early X-ray photography or digital such as more contemporary scanning devices like computer tomographs. Only when (aspects of) the vital processes inside the body can be made visible are diagnostic requests able to be properly addressed. At the same time, the vitality of the patient’s
Aligning bodies and imaging technologies 73 body is one of the major challenges when it comes to representing morphological and functional properties.8
Paradoxically, imaging the living body simultaneously demands that certain processes be arrested or “frozen” and others enhanced. For example, immobility and holding one’s breath during X-ray imaging is still a prerequisite for generating clinically valid images. Contrast media, such as air, or iodine- or barium-based media, may be applied to enhance the visibility of suspect structures since they establish an interplay—or intra-action—both with the patient’s body and with the opto-physical properties of the X-ray technique.
Science studies scholar Michael Lynch, referring to Michel Foucault’s writings, used the phrase “docile object” to describe the behavior required of the object of an experiment in accordance with scientific programs and practices. Unlike Barad, Lynch stresses in his analysis of visual practices in neurobiology the agential and cooperative role of entities when they become “docile objects” in the process of preparation. A docile object, for him, “is an object that ‘behaves’ in accordance with a programme of normalization”.9 Lynch draws on practices of preparation and representation, but his reference to the docile object hints at the need for active behavior according to a certain epistemological idea or, as he calls it, a “programme of normalization”. In Lynch’s examples, however, it becomes clear that this cooperation consists also of moderation. The tension between active and passive cooperation is particularly relevant for medical contexts. Keeping still and holding one’s breath are only two examples of affordances that place an imposition on the patient’s body and behavior in connection with preparing the structural set-up of an imaging procedure.
To theoretically grasp the paradox inherent to the requirement, when making living entities visible, for material and conceptual affordances, I would like to introduce the notion of “imageability” (in German, Bildbarkeit). The neologism, composed of the noun image (Bild) and the suffix -ability (-barkeit), builds on the work of Walter Benjamin, who described the potentiality of certain actions by adding this same suffix to word stems, as in “reproducibility” (“Reproduzierbarkeit”) and “cognizability” or knowableness (Erkennbar&e/t) in the sense of “knowledge”.10 By extension, imageability or Bildbarkeit refers to the structural and conceptual strategies and formations that enable imaging processes. Semantically, the German noun Bild (image) also hints at the verb bilden, which translates as “to form” (or “to constitute”). The simultaneity of generating the structural potential for imaging by forming or shaping someone or something resonates in the notion of imageability.11
Whereas Barad’s “intra-action” describes the agential processes and interplay between objects and apparatuses of observation during a procedure, “imageability” shifts the focus to the structures and procedures that facilitate media-specific practices—in this case early imaging techniques—and must be installed or put into operation before any intelligible visualization can be attained. I now turn to the example of neuroimaging with pneumoencephalography in a historical context, where significant material interventions laid the groundwork for imageability for the purposes of diagnostics.
Intervention for imaging—pneumoencephalography
Neuroimaging and neurosurgery face a fundamental imaging-related challenge: the human brain is inevitably resistant to being comprehensively mapped by X-ray
Figure 5.1 Skull radiograph without contrast media, August 28,2014. Source: Wikipedia, author ANUG, CC BY-SA 4.0, October 10, 2016.
photography. Only dense structures can be differentiated by this technique: very homogeneous and soft tissues like the brain remain unseen. In the pre-digital era, X-ray imaging was the only method of neuroimaging, one example being radiographs of the human skull (Figure 5.1).
As this radiograph shows, only the bone structures of the skull are detectable, whereas soft tissue like gray matter is invisible. As X-ray photography reduces all of the spatial properties of an object to one photosensitive plane, morphological structures and spatial depth are overlaid in the image and must be differentiated cognitively.12 To overcome the visual and hence diagnostic limitations of X-ray imaging, several variants of X-ray techniques were developed for neuroradiological purposes.13 One of the procedures developed by neurosurgeon Walter Dandy both to diagnose malformations and to plan therapeutic procedures was ventriculography, and, more specifically, pneumoencephalography. For a more detailed and diagnostically significant X-ray of the skull, Dandy sketched
the possibility of filling the cerebral ventricles with a medium that [would] produce a shadow in the radiogram. If this could be done, an accurate outline of the cerebral ventricles could be photographed with x-rays, and since most neoplasms either directly or indirectly modify the size or shape of the ventricles, we should then possess an early and accurate aid to the localization of intracranial affections. In addition to its radiographic properties, any substance
Aligning bodies and imaging technologies 75 injected into the ventricles must satisfy two very rigid exactions: (1) it must be absolutely non-irritating and non-toxic; and (2) it must be readily absorbed and excreted.14
Dandy associated pneumoencephalographic imaging with consideration of the necessary intervention in patients’ bodies, such as the injection of a substance, i.e., a radiographic contrast medium. Only through the interplay of imaging and intervention did it seem possible to render the brain imageable and generate a radiogram that could determine the relative position of pathological neoplasms in the brain. The procedure of pneumoencephalography required most of the fluid inside the skull to be drained from around the brain and replaced with air. During the early phase of pneumoencephalography, Dandy drained the cerebrospinal fluid and injected the air directly into the skull through access points.
About a year later, he tested a new procedure in which the spinal canal was used to replace fluid with air, also in different areas of the skull and brain.15 For this interventional procedure that should establish a different form of imageability and therefore of diagnostic knowing, the patient had to sit in front of an X-ray screen while the physician conducted the replacement procedure in the lower back (Figure 5.2).
Usually, it took four radiographs produced from different angles to gain a useful impression of the spatial properties. In contrast to the skull radiogram, which was produced without the use of a contrast medium (Figure 5.1), the following one shows a “photograph of a roentgenogram of the head after an intraspinous injection of air”.16 Even in the rather blurry illustration included in Dandy’s paper, the difference in detail of intracranial morphological structures is obvious (Figure 5.3).
Using air as a contrast medium for neuroradiological purposes significantly increased the precision of diagnostic imaging, since structures within the skull could be
Figure 5.2 Patient with pneumatic apparatus attached at lower back in front of an X-ray screen. Source: Otto Schiersmann, Einführung in die Enzephalographie: die Röntgenuntersuchung nach Luftfüllung der Liquorräume als diagnostisches Hilfsmittel in der Psychiatrie und Neurologie nebst einem Anhang über die Artériographie und die kombinierte Enzephal-Arteriographie (Leipzig: Thieme, 1942), 29.
Figure 5.3 Photograph of a roentgenogram of the head after an infraspinous injection of air. Source: Walter E. Dandy, “Roentgenography of the Brain after the Injection of Air into the Spinal Canal”, in Neurosurgical Classics, ed. Robert H. Wilkins (Park Ridge, IL: American Association of Neurological Surgeons, 1992), 253.
differentiated to a much greater degree. But what is also clear is the significant impact on the patient’s body of this procedure of diagnostic medical imaging—before the process of “visualization” can be initiated.
Exploring the affordances of imaging techniques prior to the actual visualization process offers a valuable analytical perspective on the material and conceptual conditions that constitute the aesthetics of an image. By thinking about the imageability of bodies in medical contexts, it becomes clear that images in this realm rely on a hybrid ensemble of apparatus, physical properties, somatic alignments, and diagnostic discourse. This must be configured and aligned before any image can be produced. Photographic, or more precisely roentgenographic, imaging demanded an ensemble of apparatuses, materialities, and processes to establish the relata for an intra-action that could finally lead to generating an image. Not only did the patient’s body have to be rendered imageable, e.g., by interventions such as draining spinal fluid, but a whole dispositive was needed to be mobilized just to establish the structural possibility for diagnostic imaging. In particular, instances of pneumoencephalography demonstrate the highly invasive preparatory procedures performed on the patient before X-ray
Aligning bodies and imaging technologies 77 imaging. Rendering the brain or the body visible implies not only the use of certain apparatus, but also knowledge of how to administer to patients before they can be therapeutically treated.
The procedures of establishing a certain form of imageability find their resonance in visual aesthetics of the radiograms produced. On this analytical level, a different form of “hybridity” becomes apparent. While the hybridity of the ensemble that establishes imageability concerns a rather material and pragmatic dimension, e.g., by physically interfacing apparatuses and bodies, hybridity in regard to the produced radiographs relates rather to epistemological dimensions. Certainly, both dimensions of hybridity are strongly connected in the example of pneumoencephalography, yet it seems analytically productive to separate them in order to also differentiate levels on which imageability becomes agential. On an epistemological level and in the medical context, the hybridity of visualizations entails intermedia! as well as procedural aspects. The X-rays not only allow to diagnostically visualize brain structures but as such provide a material basis for physicians to mark tumors and visually plan subsequent interventions. As artifacts they visually and temporally connect the prior, considerably ill condition of a patient with further treatment options. Hence, these kind of X-ray photographs become hybrid entanglements of time-space relations as well as different epistemological endeavors. As mentioned before, practically this dimension of hybridity can hardly be separated from the hybrid ensemble that establishes the imageability of a person. Nevertheless, in an analytical perspective, the differentiation allows to address the media theoretical issue of how media technologies (need to) install rather material and structural possibilities of imaging and, simultaneously, imply aesthetic and epistemological constraints that are agential in their very own respect.
- 1 Walter E. Dandy, “Ventriculography Following the Injection of Air into the Cerebral Ventricles,” Annals of Surgery 68 (1918): 5.
- 2 For a broader art historical and philosophical perspective, see Gottfried Boehm, “Zwischen Auge und Hand: Bilder als Instrumente der Erkenntnis,” in Mit dem Auge den-ken: Strategien der Sichtbarmachung in wissenschaftlichen und virtuellen Welten, eds. Bettina Heintz and Jorg Huber (Zurich: Edition Voldemeer, 2001), 43-61.
- 3 With regard to contexts of contemporary neuroimaging, see Morana Alac, Handling Digital Brains: A Laboratory Study of Multimodal Semiotic Interaction (Cambridge: MIT Press, 2011); and Anne Beaulieu, “Images Are Not the (Only) Truth: Brain Mapping, Visual Knowledge, and Iconoclasm,” Science, Technology, and Human Values 271, no. 1 (2002): 53-86.
- 4 For an overview see Theodore R. Schatzki, Karin Knorr Cetina and Eike von Savigny, eds., The Practice Turn in Contemporary Theory (New York: Routledge, 2006).
- 5 Karen Barad, “Getting Real: Technoscientific Practices and the Materialization of Reality,” Differences: A Journal of Feminist Cultural Studies 10, no. 2 (1998): 96.
- 6 Ibid., 89. Barad further defines: “...matter does not refer to a fixed substance; rather, matter is substance in its intra-active becoming—not a thing, but a doing, a congealing of agency. Matter is a stabilizing and destabilizing process of iterative intra-activity”. Karen Barad, “Posthumanist Performativity: Toward an Understanding of How Matter Comes to Matter,” Signs: Journal of Women in Culture and Society 28, no. 3 (2003): 822. (Italics in Org.).
- 7 Barad, “Posthumanist Performativity,” 815.
- 8 Scott Curtis, “Still/Moving: Digital Imaging and Medical Hermeneutics,” in Memory Bytes: History, Technology, and Culture, eds. Lauren Rabinovitz and Abraham Geil (Durham: Duke University Press, 2004), 218-254.
- 9 Michael Lynch, “Discipline and the Material Form of Images: An Analysis of Scientific Visibility,” Social Studies of Science 15, no. 1 (1985): 43.
- 10 Samuel Weber gives a detailed analysis of Benjamin’s extensive use of the suffix -ability: Nouns formed in this way refer to a possibility or a potentiality, to a capacity rather than to an actually existing reality. [...] Benjamin’s -abilities, then, refer to what Jacques Derrida [...] called “structural possibilities,” the necessity of which does not depend on actual fact or probable implementation. This emphasis on a possibility that is structurally necessary without being necessarily real disposes Benjamin to make use of such terms, even where they are rather uncommon. Samuel Weber, Benjamin’s-abilities (Cambridge, MA: Harvard University Press, 2008), 116.
- 11 As early as 1960, the American urban planner and architect Kevin Lynch used the notion of imageability to describe physical qualities which relate to the attributes of identity and structure in the mental image [of a physical environment]. This leads to the definition of what might be called imageability; that quality in a physical object which gives it a high probability of evoking a strong image in any given observer. Kevin Lynch, The Image of the City (Cambridge: MIT Press, 1960), 9; my thanks go to Matthias Bruhn for pointing this out to me. Lynch’s notion of imageability focuses on physical qualities of an urban space that evoke imaginative processes in an observer. It has a different connotation from the term image, namely, the evocation or creation of a mental image. Lynch hints at the performative character of built images and the relation between physical configurations and imaginative and epistemological processes, but does not include in his discussion the physical reconfigurations that precede or proceed from this relation.
- 12 See, for example, Bernike Pasveer, “Knowledge of Shadows: The Introduction of X-ray Images in Medicine,” Sociology of Health and Illness 11, no. 4 (1989): 360-381.
- 13 Matthew A. Kirkman, “The Role of Imaging in the Development of Neurosurgery,” Journal of Clinical Neuroscience 22 (2015): 55-61.
- 14 Dandy, “Ventriculography,” 5.
- 15 Walter E. Dandy, “Roentgenography of the Brain after the Injection of Air into the Spinal Canal,” in Neurosurgical Classics, ed. Robert H. Wilkins (Park Ridge: American Association of Neurological Surgeons, 1992), 251-256. For a description of the historical developments as well as the use of the terms ventriculography and pneumoencephalography, see Gergely Klinda, “Geschichte der Pneumoenzephalographie” (diss., Medizinische Fakultät der Charité, Universitätsmedizin Berlin, 2010).
- 16 Dandy, “Roentgenography,” 253.
6 Beyond retouching